From 3c6af7fa787f21f8873a050568ed892312899eb5 Mon Sep 17 00:00:00 2001
From: Anton Altaparmakov <aia21@cantab.net>
Date: Thu, 24 Nov 2005 13:41:33 +0000
Subject: [PATCH 001/509] NTFS: Fix a potential overflow by casting (index + 1)
 to s64 before doing a       left shift using PAGE_CACHE_SHIFT in
 fs/ntfs/file.c.  Thanks to Andrew       Morton pointing this out to.

Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
---
 fs/ntfs/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index 727533891813..c73c8864cbad 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -248,7 +248,7 @@ do_non_resident_extend:
 		 * enough to make ntfs_writepage() work.
 		 */
 		write_lock_irqsave(&ni->size_lock, flags);
-		ni->initialized_size = (index + 1) << PAGE_CACHE_SHIFT;
+		ni->initialized_size = (s64)(index + 1) << PAGE_CACHE_SHIFT;
 		if (ni->initialized_size > new_init_size)
 			ni->initialized_size = new_init_size;
 		write_unlock_irqrestore(&ni->size_lock, flags);

From e4472cb3706ceea42797ae1dc79d624026986694 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Tue, 31 Jan 2006 15:53:55 -0800
Subject: [PATCH 002/509] [CPUFREQ] cpufreq_notify_transition cleanup.

Introduce caching of cpufreq_cpu_data[freqs->cpu], which allows us to
make the function a lot more readable, and as a nice side-effect, it
now fits in < 80 column displays again.

Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq.c | 47 +++++++++++++++++++++++----------------
 1 file changed, 28 insertions(+), 19 deletions(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 7a511479ae29..6bbe5825765a 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -229,44 +229,53 @@ static inline void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci) {
 
 
 /**
- * cpufreq_notify_transition - call notifier chain and adjust_jiffies on frequency transition
+ * cpufreq_notify_transition - call notifier chain and adjust_jiffies
+ * on frequency transition.
  *
- * This function calls the transition notifiers and the "adjust_jiffies" function. It is called
- * twice on all CPU frequency changes that have external effects. 
+ * This function calls the transition notifiers and the "adjust_jiffies"
+ * function. It is called twice on all CPU frequency changes that have
+ * external effects. 
  */
 void cpufreq_notify_transition(struct cpufreq_freqs *freqs, unsigned int state)
 {
+	struct cpufreq_policy *policy;
+
 	BUG_ON(irqs_disabled());
 
 	freqs->flags = cpufreq_driver->flags;
-	dprintk("notification %u of frequency transition to %u kHz\n", state, freqs->new);
+	dprintk("notification %u of frequency transition to %u kHz\n",
+		state, freqs->new);
 
 	down_read(&cpufreq_notifier_rwsem);
+
+	policy = cpufreq_cpu_data[freqs->cpu];
 	switch (state) {
+
 	case CPUFREQ_PRECHANGE:
-		/* detect if the driver reported a value as "old frequency" which
-		 * is not equal to what the cpufreq core thinks is "old frequency".
+		/* detect if the driver reported a value as "old frequency" 
+		 * which is not equal to what the cpufreq core thinks is
+		 * "old frequency".
 		 */
 		if (!(cpufreq_driver->flags & CPUFREQ_CONST_LOOPS)) {
-			if ((likely(cpufreq_cpu_data[freqs->cpu])) &&
-			    (likely(cpufreq_cpu_data[freqs->cpu]->cpu == freqs->cpu)) &&
-			    (likely(cpufreq_cpu_data[freqs->cpu]->cur)) &&
-			    (unlikely(freqs->old != cpufreq_cpu_data[freqs->cpu]->cur)))
-			{
-				dprintk(KERN_WARNING "Warning: CPU frequency is %u, "
-				       "cpufreq assumed %u kHz.\n", freqs->old, cpufreq_cpu_data[freqs->cpu]->cur);
-				freqs->old = cpufreq_cpu_data[freqs->cpu]->cur;
+			if ((policy) && (policy->cpu == freqs->cpu) &&
+			    (policy->cur) && (policy->cur != freqs->old)) {
+				dprintk(KERN_WARNING "Warning: CPU frequency is"
+					" %u, cpufreq assumed %u kHz.\n",
+					freqs->old, policy->cur);
+				freqs->old = policy->cur;
 			}
 		}
-		notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_PRECHANGE, freqs);
+		notifier_call_chain(&cpufreq_transition_notifier_list,
+					CPUFREQ_PRECHANGE, freqs);
 		adjust_jiffies(CPUFREQ_PRECHANGE, freqs);
 		break;
+
 	case CPUFREQ_POSTCHANGE:
 		adjust_jiffies(CPUFREQ_POSTCHANGE, freqs);
-		notifier_call_chain(&cpufreq_transition_notifier_list, CPUFREQ_POSTCHANGE, freqs);
-		if ((likely(cpufreq_cpu_data[freqs->cpu])) && 
-		    (likely(cpufreq_cpu_data[freqs->cpu]->cpu == freqs->cpu)))
-			cpufreq_cpu_data[freqs->cpu]->cur = freqs->new;
+		notifier_call_chain(&cpufreq_transition_notifier_list,
+					CPUFREQ_POSTCHANGE, freqs);
+		if (likely(policy) && likely(policy->cpu == freqs->cpu))
+			policy->cur = freqs->new;
 		break;
 	}
 	up_read(&cpufreq_notifier_rwsem);

From 9d2725bb815d915fc6c8531097d9e71b579a8763 Mon Sep 17 00:00:00 2001
From: Thomas Renninger <trenn@suse.de>
Date: Wed, 1 Feb 2006 11:38:37 +0100
Subject: [PATCH 003/509] [CPUFREQ] Check for not initialized freq on cpufreq
 changes

Test for old_freq equals 0 to insure not to divide by 0:
______________________________________________

Check for not initialized freq on cpufreq changes

Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 arch/i386/kernel/timers/timer_tsc.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c
index 47675bbbb316..591a642af884 100644
--- a/arch/i386/kernel/timers/timer_tsc.c
+++ b/arch/i386/kernel/timers/timer_tsc.c
@@ -272,6 +272,10 @@ time_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
 	if (val != CPUFREQ_RESUMECHANGE)
 		write_seqlock_irq(&xtime_lock);
 	if (!ref_freq) {
+		if (!freq->old){
+			ref_freq = freq->new;
+			goto end;
+		}
 		ref_freq = freq->old;
 		loops_per_jiffy_ref = cpu_data[freq->cpu].loops_per_jiffy;
 #ifndef CONFIG_SMP
@@ -297,6 +301,7 @@ time_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
 #endif
 	}
 
+end:
 	if (val != CPUFREQ_RESUMECHANGE)
 		write_sequnlock_irq(&xtime_lock);
 

From a85f7bd310dbc9010309bfe70b6b02432a11ef59 Mon Sep 17 00:00:00 2001
From: Thomas Renninger <trenn@suse.de>
Date: Wed, 1 Feb 2006 11:36:04 +0100
Subject: [PATCH 004/509] [CPUFREQ] Check whether driver init did not
 initialize current freq

Check whether driver init did not initialize current freq

Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 6bbe5825765a..35eafbcf5ad5 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -1442,8 +1442,13 @@ int cpufreq_update_policy(unsigned int cpu)
 	  -> ask driver for current freq and notify governors about a change */
 	if (cpufreq_driver->get) {
 		policy.cur = cpufreq_driver->get(cpu);
-		if (data->cur != policy.cur)
-			cpufreq_out_of_sync(cpu, data->cur, policy.cur);
+		if (!data->cur) {
+			dprintk("Driver did not initialize current freq");
+			data->cur = policy.cur;
+		} else {
+			if (data->cur != policy.cur)
+				cpufreq_out_of_sync(cpu, data->cur, policy.cur);
+		}
 	}
 
 	ret = __cpufreq_set_policy(data, &policy);

From 7d5e350fab47f1273bc8b52d5f133ed6e4baeb7f Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Thu, 2 Feb 2006 17:03:42 -0500
Subject: [PATCH 005/509] [CPUFREQ] Whitespace/CodingStyle cleanups

Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/cpufreq/cpufreq.c | 54 +++++++++++++++------------------------
 1 file changed, 21 insertions(+), 33 deletions(-)

diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index 35eafbcf5ad5..9582de1c9cad 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -35,8 +35,8 @@
  * level driver of CPUFreq support, and its spinlock. This lock
  * also protects the cpufreq_cpu_data array.
  */
-static struct cpufreq_driver   	*cpufreq_driver;
-static struct cpufreq_policy	*cpufreq_cpu_data[NR_CPUS];
+static struct cpufreq_driver *cpufreq_driver;
+static struct cpufreq_policy *cpufreq_cpu_data[NR_CPUS];
 static DEFINE_SPINLOCK(cpufreq_driver_lock);
 
 /* internal prototypes */
@@ -50,15 +50,15 @@ static void handle_update(void *data);
  * changes to devices when the CPU clock speed changes.
  * The mutex locks both lists.
  */
-static struct notifier_block    *cpufreq_policy_notifier_list;
-static struct notifier_block    *cpufreq_transition_notifier_list;
-static DECLARE_RWSEM		(cpufreq_notifier_rwsem);
+static struct notifier_block *cpufreq_policy_notifier_list;
+static struct notifier_block *cpufreq_transition_notifier_list;
+static DECLARE_RWSEM (cpufreq_notifier_rwsem);
 
 
 static LIST_HEAD(cpufreq_governor_list);
-static DEFINE_MUTEX		(cpufreq_governor_mutex);
+static DEFINE_MUTEX (cpufreq_governor_mutex);
 
-struct cpufreq_policy * cpufreq_cpu_get(unsigned int cpu)
+struct cpufreq_policy *cpufreq_cpu_get(unsigned int cpu)
 {
 	struct cpufreq_policy *data;
 	unsigned long flags;
@@ -85,20 +85,19 @@ struct cpufreq_policy * cpufreq_cpu_get(unsigned int cpu)
 	if (!kobject_get(&data->kobj))
 		goto err_out_put_module;
 
-
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
-
 	return data;
 
- err_out_put_module:
+err_out_put_module:
 	module_put(cpufreq_driver->owner);
- err_out_unlock:
+err_out_unlock:
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
- err_out:
+err_out:
 	return NULL;
 }
 EXPORT_SYMBOL_GPL(cpufreq_cpu_get);
 
+
 void cpufreq_cpu_put(struct cpufreq_policy *data)
 {
 	kobject_put(&data->kobj);
@@ -317,7 +316,7 @@ static int cpufreq_parse_governor (char *str_governor, unsigned int *policy,
 				return 0;
 			}
 		}
-	out:
+out:
 		mutex_unlock(&cpufreq_governor_mutex);
 	}
 	return -EINVAL;
@@ -424,7 +423,6 @@ static ssize_t store_scaling_governor (struct cpufreq_policy * policy,
 		return -EINVAL;
 
 	ret = cpufreq_set_policy(&new_policy);
-
 	return ret ? ret : count;
 }
 
@@ -455,7 +453,7 @@ static ssize_t show_scaling_available_governors (struct cpufreq_policy * policy,
 			goto out;
 		i += scnprintf(&buf[i], CPUFREQ_NAME_LEN, "%s ", t->name);
 	}
- out:
+out:
 	i += sprintf(&buf[i], "\n");
 	return i;
 }
@@ -798,7 +796,6 @@ static int cpufreq_remove_dev (struct sys_device * sys_dev)
 	kfree(data);
 
 	cpufreq_debug_enable_ratelimit();
-
 	return 0;
 }
 
@@ -879,8 +876,7 @@ unsigned int cpufreq_get(unsigned int cpu)
 
 	ret = cpufreq_driver->get(cpu);
 
-	if (ret && policy->cur && !(cpufreq_driver->flags & CPUFREQ_CONST_LOOPS)) 
-	{
+	if (ret && policy->cur && !(cpufreq_driver->flags & CPUFREQ_CONST_LOOPS)) {
 		/* verify no discrepancy between actual and saved value exists */
 		if (unlikely(ret != policy->cur)) {
 			cpufreq_out_of_sync(cpu, policy->cur, ret);
@@ -890,7 +886,7 @@ unsigned int cpufreq_get(unsigned int cpu)
 
 	mutex_unlock(&policy->lock);
 
- out:
+out:
 	cpufreq_cpu_put(policy);
 
 	return (ret);
@@ -971,7 +967,7 @@ static int cpufreq_suspend(struct sys_device * sysdev, pm_message_t pmsg)
 		cpu_policy->cur = cur_freq;
 	}
 
- out:
+out:
 	cpufreq_cpu_put(cpu_policy);
 	return 0;
 }
@@ -1178,7 +1174,6 @@ int cpufreq_driver_target(struct cpufreq_policy *policy,
 	mutex_unlock(&policy->lock);
 
 	cpufreq_cpu_put(policy);
-
 	return ret;
 }
 EXPORT_SYMBOL_GPL(cpufreq_driver_target);
@@ -1217,7 +1212,6 @@ int cpufreq_governor(unsigned int cpu, unsigned int event)
 	mutex_unlock(&policy->lock);
 
 	cpufreq_cpu_put(policy);
-
 	return ret;
 }
 EXPORT_SYMBOL_GPL(cpufreq_governor);
@@ -1241,7 +1235,6 @@ int cpufreq_register_governor(struct cpufreq_governor *governor)
 	list_add(&governor->governor_list, &cpufreq_governor_list);
 
  	mutex_unlock(&cpufreq_governor_mutex);
-
 	return 0;
 }
 EXPORT_SYMBOL_GPL(cpufreq_register_governor);
@@ -1286,7 +1279,6 @@ int cpufreq_get_policy(struct cpufreq_policy *policy, unsigned int cpu)
 	mutex_unlock(&cpu_policy->lock);
 
 	cpufreq_cpu_put(cpu_policy);
-
 	return 0;
 }
 EXPORT_SYMBOL(cpufreq_get_policy);
@@ -1300,9 +1292,7 @@ static int __cpufreq_set_policy(struct cpufreq_policy *data, struct cpufreq_poli
 	dprintk("setting new policy for CPU %u: %u - %u kHz\n", policy->cpu,
 		policy->min, policy->max);
 
-	memcpy(&policy->cpuinfo, 
-	       &data->cpuinfo, 
-	       sizeof(struct cpufreq_cpuinfo));
+	memcpy(&policy->cpuinfo, &data->cpuinfo, sizeof(struct cpufreq_cpuinfo));
 
 	/* verify the cpu speed can be set within this limit */
 	ret = cpufreq_driver->verify(policy);
@@ -1333,8 +1323,8 @@ static int __cpufreq_set_policy(struct cpufreq_policy *data, struct cpufreq_poli
 
 	up_read(&cpufreq_notifier_rwsem);
 
-	data->min    = policy->min;
-	data->max    = policy->max;
+	data->min = policy->min;
+	data->max = policy->max;
 
 	dprintk("new min and max freqs are %u - %u kHz\n", data->min, data->max);
 
@@ -1371,7 +1361,7 @@ static int __cpufreq_set_policy(struct cpufreq_policy *data, struct cpufreq_poli
 		__cpufreq_governor(data, CPUFREQ_GOV_LIMITS);
 	}
 
- error_out:
+error_out:
 	cpufreq_debug_enable_ratelimit();
 	return ret;
 }
@@ -1430,9 +1420,7 @@ int cpufreq_update_policy(unsigned int cpu)
 	mutex_lock(&data->lock);
 
 	dprintk("updating policy for CPU %u\n", cpu);
-	memcpy(&policy, 
-	       data,
-	       sizeof(struct cpufreq_policy));
+	memcpy(&policy, data, sizeof(struct cpufreq_policy));
 	policy.min = data->user_policy.min;
 	policy.max = data->user_policy.max;
 	policy.policy = data->user_policy.policy;

From 8cf3f04f45694db0699f608c0e3fb550c607cc88 Mon Sep 17 00:00:00 2001
From: Ralph Campbell <ralphc@pathscale.com>
Date: Fri, 3 Feb 2006 14:28:48 -0800
Subject: [PATCH 006/509] IB/mad: Handle DR SMPs with a LID routed part

Fix handling of directed route SMPs with a beginning or ending LID
routed part.

Signed-off-by: Ralph Campbell <ralphc@pathscale.com>
Signed-off-by: Hal Rosenstock <halr@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/mad.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
index d393b504bf26..c82f47a66e48 100644
--- a/drivers/infiniband/core/mad.c
+++ b/drivers/infiniband/core/mad.c
@@ -665,7 +665,15 @@ static int handle_outgoing_dr_smp(struct ib_mad_agent_private *mad_agent_priv,
 	struct ib_wc mad_wc;
 	struct ib_send_wr *send_wr = &mad_send_wr->send_wr;
 
-	if (!smi_handle_dr_smp_send(smp, device->node_type, port_num)) {
+	/*
+	 * Directed route handling starts if the initial LID routed part of
+	 * a request or the ending LID routed part of a response is empty.
+	 * If we are at the start of the LID routed part, don't update the
+	 * hop_ptr or hop_cnt.  See section 14.2.2, Vol 1 IB spec.
+	 */
+	if ((ib_get_smp_direction(smp) ? smp->dr_dlid : smp->dr_slid) ==
+	     IB_LID_PERMISSIVE &&
+	    !smi_handle_dr_smp_send(smp, device->node_type, port_num)) {
 		ret = -EINVAL;
 		printk(KERN_ERR PFX "Invalid directed route\n");
 		goto out;

From 04414013bbda644b65537e73f1dacb2821b36811 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 31 Jan 2006 16:04:51 -0800
Subject: [PATCH 007/509] [SCSI] qla2xxx: Add port-speed FC transport
 attribute.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_attr.c | 22 ++++++++++++++++++++++
 drivers/scsi/qla2xxx/qla_def.h  |  4 ++++
 drivers/scsi/qla2xxx/qla_isr.c  |  4 ++--
 drivers/scsi/qla2xxx/qla_os.c   |  1 +
 4 files changed, 29 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index b17ee62dd1a9..e03140fec179 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -425,6 +425,26 @@ qla2x00_get_host_port_id(struct Scsi_Host *shost)
 	    ha->d_id.b.area << 8 | ha->d_id.b.al_pa;
 }
 
+static void
+qla2x00_get_host_speed(struct Scsi_Host *shost)
+{
+	scsi_qla_host_t *ha = to_qla_host(shost);
+	uint32_t speed = 0;
+
+	switch (ha->link_data_rate) {
+	case LDR_1GB:
+		speed = 1;
+		break;
+	case LDR_2GB:
+		speed = 2;
+		break;
+	case LDR_4GB:
+		speed = 4;
+		break;
+	}
+	fc_host_speed(shost) = speed;
+}
+
 static void
 qla2x00_get_starget_node_name(struct scsi_target *starget)
 {
@@ -520,6 +540,8 @@ struct fc_function_template qla2xxx_transport_functions = {
 
 	.get_host_port_id = qla2x00_get_host_port_id,
 	.show_host_port_id = 1,
+	.get_host_speed = qla2x00_get_host_speed,
+	.show_host_speed = 1,
 
 	.dd_fcrport_size = sizeof(struct fc_port *),
 	.show_rport_supported_classes = 1,
diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h
index bad066e5772a..26af5319c219 100644
--- a/drivers/scsi/qla2xxx/qla_def.h
+++ b/drivers/scsi/qla2xxx/qla_def.h
@@ -2331,6 +2331,10 @@ typedef struct scsi_qla_host {
 	uint16_t	min_external_loopid;	/* First external loop Id */
 
 	uint16_t	link_data_rate;		/* F/W operating speed */
+#define LDR_1GB		0
+#define LDR_2GB		1
+#define LDR_4GB		3
+#define LDR_UNKNOWN	0xFFFF
 
 	uint8_t		current_topology;
 	uint8_t		prev_topology;
diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
index 71a46fcee8cc..42aa7a7c1a73 100644
--- a/drivers/scsi/qla2xxx/qla_isr.c
+++ b/drivers/scsi/qla2xxx/qla_isr.c
@@ -402,9 +402,9 @@ qla2x00_async_event(scsi_qla_host_t *ha, uint16_t *mb)
 		break;
 
 	case MBA_LOOP_UP:		/* Loop Up Event */
-		ha->link_data_rate = 0;
 		if (IS_QLA2100(ha) || IS_QLA2200(ha)) {
 			link_speed = link_speeds[0];
+			ha->link_data_rate = LDR_1GB;
 		} else {
 			link_speed = link_speeds[LS_UNKNOWN];
 			if (mb[1] < 5)
@@ -436,7 +436,7 @@ qla2x00_async_event(scsi_qla_host_t *ha, uint16_t *mb)
 		}
 
 		ha->flags.management_server_logged_in = 0;
-		ha->link_data_rate = 0;
+		ha->link_data_rate = LDR_UNKNOWN;
 		if (ql2xfdmienable)
 			set_bit(REGISTER_FDMI_NEEDED, &ha->dpc_flags);
 
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index 5866a7c706a8..6e133edb2016 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -1312,6 +1312,7 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 	ha->ports = MAX_BUSES;
 	ha->init_cb_size = sizeof(init_cb_t);
 	ha->mgmt_svr_loop_id = MANAGEMENT_SERVER;
+	ha->link_data_rate = LDR_UNKNOWN;
 
 	/* Assign ISP specific operations. */
 	ha->isp_ops.pci_config		= qla2100_pci_config;

From 8d067623adf119081b7a2683cdc6ee90eb8a70b2 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 31 Jan 2006 16:04:56 -0800
Subject: [PATCH 008/509] [SCSI] qla2xxx: Add host port-type FC transport
 attribute.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_attr.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index e03140fec179..55016328bd78 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -445,6 +445,29 @@ qla2x00_get_host_speed(struct Scsi_Host *shost)
 	fc_host_speed(shost) = speed;
 }
 
+static void
+qla2x00_get_host_port_type(struct Scsi_Host *shost)
+{
+	scsi_qla_host_t *ha = to_qla_host(shost);
+	uint32_t port_type = FC_PORTTYPE_UNKNOWN;
+
+	switch (ha->current_topology) {
+	case ISP_CFG_NL:
+		port_type = FC_PORTTYPE_LPORT;
+		break;
+	case ISP_CFG_FL:
+		port_type = FC_PORTTYPE_NLPORT;
+		break;
+	case ISP_CFG_N:
+		port_type = FC_PORTTYPE_PTP;
+		break;
+	case ISP_CFG_F:
+		port_type = FC_PORTTYPE_NPORT;
+		break;
+	}
+	fc_host_port_type(shost) = port_type;
+}
+
 static void
 qla2x00_get_starget_node_name(struct scsi_target *starget)
 {
@@ -542,6 +565,8 @@ struct fc_function_template qla2xxx_transport_functions = {
 	.show_host_port_id = 1,
 	.get_host_speed = qla2x00_get_host_speed,
 	.show_host_speed = 1,
+	.get_host_port_type = qla2x00_get_host_port_type,
+	.show_host_port_type = 1,
 
 	.dd_fcrport_size = sizeof(struct fc_port *),
 	.show_rport_supported_classes = 1,

From 392e2f651c8a83484116a407a9f121e534c22b5a Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 31 Jan 2006 16:05:02 -0800
Subject: [PATCH 009/509] [SCSI] qla2xxx: Add host-statistics FC transport
 attributes.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_attr.c |  37 ++++++++++-
 drivers/scsi/qla2xxx/qla_def.h  |   2 +
 drivers/scsi/qla2xxx/qla_gbl.h  |   7 +++
 drivers/scsi/qla2xxx/qla_init.c |   1 -
 drivers/scsi/qla2xxx/qla_mbx.c  | 105 +++++++++++++++++++++++++++++++-
 drivers/scsi/qla2xxx/qla_rscn.c |   2 -
 6 files changed, 147 insertions(+), 7 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index 55016328bd78..5a8d5c4c69ba 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -7,7 +7,6 @@
 #include "qla_def.h"
 
 #include <linux/vmalloc.h>
-#include <scsi/scsi_transport_fc.h>
 
 /* SYSFS attributes --------------------------------------------------------- */
 
@@ -555,6 +554,41 @@ qla2x00_issue_lip(struct Scsi_Host *shost)
 	return 0;
 }
 
+static struct fc_host_statistics *
+qla2x00_get_fc_host_stats(struct Scsi_Host *shost)
+{
+	scsi_qla_host_t *ha = to_qla_host(shost);
+	int rval;
+	uint16_t mb_stat[1];
+	link_stat_t stat_buf;
+	struct fc_host_statistics *pfc_host_stat;
+
+	pfc_host_stat = &ha->fc_host_stat;
+	memset(pfc_host_stat, -1, sizeof(struct fc_host_statistics));
+
+	if (IS_QLA24XX(ha) || IS_QLA25XX(ha)) {
+		rval = qla24xx_get_isp_stats(ha, (uint32_t *)&stat_buf,
+		    sizeof(stat_buf) / 4, mb_stat);
+	} else {
+		rval = qla2x00_get_link_status(ha, ha->loop_id, &stat_buf,
+		    mb_stat);
+	}
+	if (rval != 0) {
+		qla_printk(KERN_WARNING, ha,
+		    "Unable to retrieve host statistics (%d).\n", mb_stat[0]);
+		return pfc_host_stat;
+	}
+
+	pfc_host_stat->link_failure_count = stat_buf.link_fail_cnt;
+	pfc_host_stat->loss_of_sync_count = stat_buf.loss_sync_cnt;
+	pfc_host_stat->loss_of_signal_count = stat_buf.loss_sig_cnt;
+	pfc_host_stat->prim_seq_protocol_err_count = stat_buf.prim_seq_err_cnt;
+	pfc_host_stat->invalid_tx_word_count = stat_buf.inval_xmit_word_cnt;
+	pfc_host_stat->invalid_crc_count = stat_buf.inval_crc_cnt;
+
+	return pfc_host_stat;
+}
+
 struct fc_function_template qla2xxx_transport_functions = {
 
 	.show_host_node_name = 1,
@@ -583,6 +617,7 @@ struct fc_function_template qla2xxx_transport_functions = {
 	.show_rport_dev_loss_tmo = 1,
 
 	.issue_fc_host_lip = qla2x00_issue_lip,
+	.get_fc_host_stats = qla2x00_get_fc_host_stats,
 };
 
 void
diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h
index 26af5319c219..414580800dc0 100644
--- a/drivers/scsi/qla2xxx/qla_def.h
+++ b/drivers/scsi/qla2xxx/qla_def.h
@@ -29,6 +29,7 @@
 #include <scsi/scsi_host.h>
 #include <scsi/scsi_device.h>
 #include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_transport_fc.h>
 
 #if defined(CONFIG_SCSI_QLA2XXX_EMBEDDED_FIRMWARE)
 #if defined(CONFIG_SCSI_QLA21XX) || defined(CONFIG_SCSI_QLA21XX_MODULE)
@@ -2496,6 +2497,7 @@ typedef struct scsi_qla_host {
 
 	uint16_t	zio_mode;
 	uint16_t	zio_timer;
+	struct fc_host_statistics fc_host_stat;
 } scsi_qla_host_t;
 
 
diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h
index 35266bd5d538..f2f5454a05e9 100644
--- a/drivers/scsi/qla2xxx/qla_gbl.h
+++ b/drivers/scsi/qla2xxx/qla_gbl.h
@@ -185,6 +185,13 @@ qla2x00_get_resource_cnts(scsi_qla_host_t *, uint16_t *, uint16_t *, uint16_t *,
 extern int
 qla2x00_get_fcal_position_map(scsi_qla_host_t *ha, char *pos_map);
 
+extern int
+qla2x00_get_link_status(scsi_qla_host_t *, uint16_t, link_stat_t *,
+    uint16_t *);
+
+extern int
+qla24xx_get_isp_stats(scsi_qla_host_t *, uint32_t *, uint32_t, uint16_t *);
+
 extern int qla24xx_abort_command(scsi_qla_host_t *, srb_t *);
 extern int qla24xx_abort_target(fc_port_t *);
 
diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index e67bb0997818..634ee174bff2 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -8,7 +8,6 @@
 
 #include <linux/delay.h>
 #include <linux/vmalloc.h>
-#include <scsi/scsi_transport_fc.h>
 
 #include "qla_devtbl.h"
 
diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
index 3099b379de9d..49ce197876b4 100644
--- a/drivers/scsi/qla2xxx/qla_mbx.c
+++ b/drivers/scsi/qla2xxx/qla_mbx.c
@@ -7,7 +7,6 @@
 #include "qla_def.h"
 
 #include <linux/delay.h>
-#include <scsi/scsi_transport_fc.h>
 
 static void
 qla2x00_mbx_sem_timeout(unsigned long data)
@@ -2017,8 +2016,109 @@ qla2x00_get_fcal_position_map(scsi_qla_host_t *ha, char *pos_map)
 
 	return rval;
 }
+#endif
 
-uint8_t
+/*
+ * qla2x00_get_link_status
+ *
+ * Input:
+ *	ha = adapter block pointer.
+ *	loop_id = device loop ID.
+ *	ret_buf = pointer to link status return buffer.
+ *
+ * Returns:
+ *	0 = success.
+ *	BIT_0 = mem alloc error.
+ *	BIT_1 = mailbox error.
+ */
+int
+qla2x00_get_link_status(scsi_qla_host_t *ha, uint16_t loop_id,
+    link_stat_t *ret_buf, uint16_t *status)
+{
+	int rval;
+	mbx_cmd_t mc;
+	mbx_cmd_t *mcp = &mc;
+	link_stat_t *stat_buf;
+	dma_addr_t stat_buf_dma;
+
+	DEBUG11(printk("%s(%ld): entered.\n", __func__, ha->host_no);)
+
+	stat_buf = dma_pool_alloc(ha->s_dma_pool, GFP_ATOMIC, &stat_buf_dma);
+	if (stat_buf == NULL) {
+		DEBUG2_3_11(printk("%s(%ld): Failed to allocate memory.\n",
+		    __func__, ha->host_no));
+		return BIT_0;
+	}
+	memset(stat_buf, 0, sizeof(link_stat_t));
+
+	mcp->mb[0] = MBC_GET_LINK_STATUS;
+	mcp->mb[2] = MSW(stat_buf_dma);
+	mcp->mb[3] = LSW(stat_buf_dma);
+	mcp->mb[6] = MSW(MSD(stat_buf_dma));
+	mcp->mb[7] = LSW(MSD(stat_buf_dma));
+	mcp->out_mb = MBX_7|MBX_6|MBX_3|MBX_2|MBX_0;
+	mcp->in_mb = MBX_0;
+	if (IS_QLA24XX(ha) || IS_QLA25XX(ha)) {
+		mcp->mb[1] = loop_id;
+		mcp->mb[4] = 0;
+		mcp->mb[10] = 0;
+		mcp->out_mb |= MBX_10|MBX_4|MBX_1;
+		mcp->in_mb |= MBX_1;
+	} else if (HAS_EXTENDED_IDS(ha)) {
+		mcp->mb[1] = loop_id;
+		mcp->mb[10] = 0;
+		mcp->out_mb |= MBX_10|MBX_1;
+	} else {
+		mcp->mb[1] = loop_id << 8;
+		mcp->out_mb |= MBX_1;
+	}
+	mcp->tov = 30;
+	mcp->flags = IOCTL_CMD;
+	rval = qla2x00_mailbox_command(ha, mcp);
+
+	if (rval == QLA_SUCCESS) {
+		if (mcp->mb[0] != MBS_COMMAND_COMPLETE) {
+			DEBUG2_3_11(printk("%s(%ld): cmd failed. mbx0=%x.\n",
+			    __func__, ha->host_no, mcp->mb[0]);)
+			status[0] = mcp->mb[0];
+			rval = BIT_1;
+		} else {
+			/* copy over data -- firmware data is LE. */
+			ret_buf->link_fail_cnt =
+			    le32_to_cpu(stat_buf->link_fail_cnt);
+			ret_buf->loss_sync_cnt =
+			    le32_to_cpu(stat_buf->loss_sync_cnt);
+			ret_buf->loss_sig_cnt =
+			    le32_to_cpu(stat_buf->loss_sig_cnt);
+			ret_buf->prim_seq_err_cnt =
+			    le32_to_cpu(stat_buf->prim_seq_err_cnt);
+			ret_buf->inval_xmit_word_cnt =
+			    le32_to_cpu(stat_buf->inval_xmit_word_cnt);
+			ret_buf->inval_crc_cnt =
+			    le32_to_cpu(stat_buf->inval_crc_cnt);
+
+			DEBUG11(printk("%s(%ld): stat dump: fail_cnt=%d "
+			    "loss_sync=%d loss_sig=%d seq_err=%d "
+			    "inval_xmt_word=%d inval_crc=%d.\n", __func__,
+			    ha->host_no, stat_buf->link_fail_cnt,
+			    stat_buf->loss_sync_cnt, stat_buf->loss_sig_cnt,
+			    stat_buf->prim_seq_err_cnt,
+			    stat_buf->inval_xmit_word_cnt,
+			    stat_buf->inval_crc_cnt);)
+		}
+	} else {
+		/* Failed. */
+		DEBUG2_3_11(printk("%s(%ld): failed=%x.\n", __func__,
+		    ha->host_no, rval);)
+		rval = BIT_1;
+	}
+
+	dma_pool_free(ha->s_dma_pool, stat_buf, stat_buf_dma);
+
+	return rval;
+}
+
+int
 qla24xx_get_isp_stats(scsi_qla_host_t *ha, uint32_t *dwbuf, uint32_t dwords,
     uint16_t *status)
 {
@@ -2080,7 +2180,6 @@ qla24xx_get_isp_stats(scsi_qla_host_t *ha, uint32_t *dwbuf, uint32_t dwords,
 
 	return rval;
 }
-#endif
 
 int
 qla24xx_abort_command(scsi_qla_host_t *ha, srb_t *sp)
diff --git a/drivers/scsi/qla2xxx/qla_rscn.c b/drivers/scsi/qla2xxx/qla_rscn.c
index 2c3342108dd8..b70bebe18c01 100644
--- a/drivers/scsi/qla2xxx/qla_rscn.c
+++ b/drivers/scsi/qla2xxx/qla_rscn.c
@@ -6,8 +6,6 @@
  */
 #include "qla_def.h"
 
-#include <scsi/scsi_transport_fc.h>
-
 /**
  * IO descriptor handle definitions.
  *

From f6df144cca19cc60dda6dcce65d236b70cc46494 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 31 Jan 2006 16:05:07 -0800
Subject: [PATCH 010/509] [SCSI] qla2xxx: Add beacon support via class-device
 attribute.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_attr.c |  50 ++++++
 drivers/scsi/qla2xxx/qla_def.h  |  21 ++-
 drivers/scsi/qla2xxx/qla_gbl.h  |   9 +-
 drivers/scsi/qla2xxx/qla_os.c   |  16 ++
 drivers/scsi/qla2xxx/qla_sup.c  | 294 ++++++++++++++++++++++++++++++++
 5 files changed, 387 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index 5a8d5c4c69ba..049e5cf1af7f 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -196,6 +196,9 @@ qla2x00_free_sysfs_attr(scsi_qla_host_t *ha)
 
 	sysfs_remove_bin_file(&host->shost_gendev.kobj, &sysfs_fw_dump_attr);
 	sysfs_remove_bin_file(&host->shost_gendev.kobj, &sysfs_nvram_attr);
+
+	if (ha->beacon_blink_led == 1)
+		ha->isp_ops.beacon_off(ha);
 }
 
 /* Scsi_Host attributes. */
@@ -383,6 +386,50 @@ qla2x00_zio_timer_store(struct class_device *cdev, const char *buf,
 	return strlen(buf);
 }
 
+static ssize_t
+qla2x00_beacon_show(struct class_device *cdev, char *buf)
+{
+	scsi_qla_host_t *ha = to_qla_host(class_to_shost(cdev));
+	int len = 0;
+
+	if (ha->beacon_blink_led)
+		len += snprintf(buf + len, PAGE_SIZE-len, "Enabled\n");
+	else
+		len += snprintf(buf + len, PAGE_SIZE-len, "Disabled\n");
+	return len;
+}
+
+static ssize_t
+qla2x00_beacon_store(struct class_device *cdev, const char *buf,
+    size_t count)
+{
+	scsi_qla_host_t *ha = to_qla_host(class_to_shost(cdev));
+	int val = 0;
+	int rval;
+
+	if (IS_QLA2100(ha) || IS_QLA2200(ha))
+		return -EPERM;
+
+	if (test_bit(ABORT_ISP_ACTIVE, &ha->dpc_flags)) {
+		qla_printk(KERN_WARNING, ha,
+		    "Abort ISP active -- ignoring beacon request.\n");
+		return -EBUSY;
+	}
+
+	if (sscanf(buf, "%d", &val) != 1)
+		return -EINVAL;
+
+	if (val)
+		rval = ha->isp_ops.beacon_on(ha);
+	else
+		rval = ha->isp_ops.beacon_off(ha);
+
+	if (rval != QLA_SUCCESS)
+		count = 0;
+
+	return count;
+}
+
 static CLASS_DEVICE_ATTR(driver_version, S_IRUGO, qla2x00_drvr_version_show,
 	NULL);
 static CLASS_DEVICE_ATTR(fw_version, S_IRUGO, qla2x00_fw_version_show, NULL);
@@ -397,6 +444,8 @@ static CLASS_DEVICE_ATTR(zio, S_IRUGO | S_IWUSR, qla2x00_zio_show,
     qla2x00_zio_store);
 static CLASS_DEVICE_ATTR(zio_timer, S_IRUGO | S_IWUSR, qla2x00_zio_timer_show,
     qla2x00_zio_timer_store);
+static CLASS_DEVICE_ATTR(beacon, S_IRUGO | S_IWUSR, qla2x00_beacon_show,
+    qla2x00_beacon_store);
 
 struct class_device_attribute *qla2x00_host_attrs[] = {
 	&class_device_attr_driver_version,
@@ -410,6 +459,7 @@ struct class_device_attribute *qla2x00_host_attrs[] = {
 	&class_device_attr_state,
 	&class_device_attr_zio,
 	&class_device_attr_zio_timer,
+	&class_device_attr_beacon,
 	NULL,
 };
 
diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h
index 414580800dc0..7cb8b5a5e659 100644
--- a/drivers/scsi/qla2xxx/qla_def.h
+++ b/drivers/scsi/qla2xxx/qla_def.h
@@ -181,6 +181,13 @@
 #define WRT_REG_WORD(addr, data)	writew(data,addr)
 #define WRT_REG_DWORD(addr, data)	writel(data,addr)
 
+/*
+ * The ISP2312 v2 chip cannot access the FLASH/GPIO registers via MMIO in an
+ * 133Mhz slot.
+ */
+#define RD_REG_WORD_PIO(addr)		(inw((unsigned long)addr))
+#define WRT_REG_WORD_PIO(addr, data)	(outw(data,(unsigned long)addr))
+
 /*
  * Fibre Channel device definitions.
  */
@@ -433,6 +440,9 @@ struct device_reg_2xxx {
 #define GPIO_LED_GREEN_ON_AMBER_OFF	0x0040
 #define GPIO_LED_GREEN_OFF_AMBER_ON	0x0080
 #define GPIO_LED_GREEN_ON_AMBER_ON	0x00C0
+#define GPIO_LED_ALL_OFF		0x0000
+#define GPIO_LED_RED_ON_OTHER_OFF	0x0001	/* isp2322 */
+#define GPIO_LED_RGA_ON			0x00C1	/* isp2322: red green amber */
 
 	union {
 		struct {
@@ -2200,6 +2210,10 @@ struct isp_operations {
 
 	void (*fw_dump) (struct scsi_qla_host *, int);
 	void (*ascii_fw_dump) (struct scsi_qla_host *);
+
+	int (*beacon_on) (struct scsi_qla_host *);
+	int (*beacon_off) (struct scsi_qla_host *);
+	void (*beacon_blink) (struct scsi_qla_host *);
 };
 
 /*
@@ -2493,7 +2507,12 @@ typedef struct scsi_qla_host {
 
 	/* Needed for BEACON */
 	uint16_t	beacon_blink_led;
-	uint16_t	beacon_green_on;
+	uint8_t		beacon_color_state;
+#define QLA_LED_GRN_ON		0x01
+#define QLA_LED_YLW_ON		0x02
+#define QLA_LED_ABR_ON		0x04
+#define QLA_LED_ALL_ON		0x07	/* yellow, green, amber. */
+					/* ISP2322: red, green, amber. */
 
 	uint16_t	zio_mode;
 	uint16_t	zio_timer;
diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h
index f2f5454a05e9..eb35198bc8cf 100644
--- a/drivers/scsi/qla2xxx/qla_gbl.h
+++ b/drivers/scsi/qla2xxx/qla_gbl.h
@@ -75,8 +75,6 @@ extern void qla2x00_cmd_timeout(srb_t *);
 extern void qla2x00_mark_device_lost(scsi_qla_host_t *, fc_port_t *, int, int);
 extern void qla2x00_mark_all_devices_lost(scsi_qla_host_t *, int);
 
-extern void qla2x00_blink_led(scsi_qla_host_t *);
-
 extern int qla2x00_down_timeout(struct semaphore *, unsigned long);
 
 extern struct fw_blob *qla2x00_request_firmware(scsi_qla_host_t *);
@@ -235,6 +233,13 @@ extern int qla2x00_write_nvram_data(scsi_qla_host_t *, uint8_t *, uint32_t,
 extern int qla24xx_write_nvram_data(scsi_qla_host_t *, uint8_t *, uint32_t,
     uint32_t);
 
+extern int qla2x00_beacon_on(struct scsi_qla_host *);
+extern int qla2x00_beacon_off(struct scsi_qla_host *);
+extern void qla2x00_beacon_blink(struct scsi_qla_host *);
+extern int qla24xx_beacon_on(struct scsi_qla_host *);
+extern int qla24xx_beacon_off(struct scsi_qla_host *);
+extern void qla24xx_beacon_blink(struct scsi_qla_host *);
+
 /*
  * Global Function Prototypes in qla_dbg.c source file.
  */
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index 6e133edb2016..57179dabcccf 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -1365,6 +1365,9 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 		ha->isp_ops.intr_handler = qla2300_intr_handler;
 		ha->isp_ops.fw_dump = qla2300_fw_dump;
 		ha->isp_ops.ascii_fw_dump = qla2300_ascii_fw_dump;
+		ha->isp_ops.beacon_on = qla2x00_beacon_on;
+		ha->isp_ops.beacon_off = qla2x00_beacon_off;
+		ha->isp_ops.beacon_blink = qla2x00_beacon_blink;
 		ha->gid_list_info_size = 6;
 	} else if (IS_QLA24XX(ha) || IS_QLA25XX(ha)) {
 		host->max_id = MAX_TARGETS_2200;
@@ -1401,6 +1404,9 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 		ha->isp_ops.write_nvram = qla24xx_write_nvram_data;
 		ha->isp_ops.fw_dump = qla24xx_fw_dump;
 		ha->isp_ops.ascii_fw_dump = qla24xx_ascii_fw_dump;
+		ha->isp_ops.beacon_on = qla24xx_beacon_on;
+		ha->isp_ops.beacon_off = qla24xx_beacon_off;
+		ha->isp_ops.beacon_blink = qla24xx_beacon_blink;
 		ha->gid_list_info_size = 8;
 	}
 	host->can_queue = ha->request_q_length + 128;
@@ -2315,6 +2321,9 @@ qla2x00_do_dpc(void *data)
 		if (!ha->interrupts_on)
 			ha->isp_ops.enable_intrs(ha);
 
+		if (test_and_clear_bit(BEACON_BLINK_NEEDED, &ha->dpc_flags))
+			ha->isp_ops.beacon_blink(ha);
+
 		ha->dpc_active = 0;
 	} /* End of while(1) */
 
@@ -2492,6 +2501,12 @@ qla2x00_timer(scsi_qla_host_t *ha)
 		    atomic_read(&ha->loop_down_timer)));
 	}
 
+	/* Check if beacon LED needs to be blinked */
+	if (ha->beacon_blink_led == 1) {
+		set_bit(BEACON_BLINK_NEEDED, &ha->dpc_flags);
+		start_dpc++;
+	}
+
 	/* Schedule the DPC routine if needed */
 	if ((test_bit(ISP_ABORT_NEEDED, &ha->dpc_flags) ||
 	    test_bit(LOOP_RESYNC_NEEDED, &ha->dpc_flags) ||
@@ -2500,6 +2515,7 @@ qla2x00_timer(scsi_qla_host_t *ha)
 	    start_dpc ||
 	    test_bit(LOGIN_RETRY_NEEDED, &ha->dpc_flags) ||
 	    test_bit(RESET_MARKER_NEEDED, &ha->dpc_flags) ||
+	    test_bit(BEACON_BLINK_NEEDED, &ha->dpc_flags) ||
 	    test_bit(RELOGIN_NEEDED, &ha->dpc_flags)) &&
 	    ha->dpc_wait && !ha->dpc_active) {
 
diff --git a/drivers/scsi/qla2xxx/qla_sup.c b/drivers/scsi/qla2xxx/qla_sup.c
index f4d755a643e4..3105bb93cc19 100644
--- a/drivers/scsi/qla2xxx/qla_sup.c
+++ b/drivers/scsi/qla2xxx/qla_sup.c
@@ -695,3 +695,297 @@ qla24xx_write_nvram_data(scsi_qla_host_t *ha, uint8_t *buf, uint32_t naddr,
 
 	return ret;
 }
+
+
+static inline void
+qla2x00_flip_colors(scsi_qla_host_t *ha, uint16_t *pflags)
+{
+	if (IS_QLA2322(ha)) {
+		/* Flip all colors. */
+		if (ha->beacon_color_state == QLA_LED_ALL_ON) {
+			/* Turn off. */
+			ha->beacon_color_state = 0;
+			*pflags = GPIO_LED_ALL_OFF;
+		} else {
+			/* Turn on. */
+			ha->beacon_color_state = QLA_LED_ALL_ON;
+			*pflags = GPIO_LED_RGA_ON;
+		}
+	} else {
+		/* Flip green led only. */
+		if (ha->beacon_color_state == QLA_LED_GRN_ON) {
+			/* Turn off. */
+			ha->beacon_color_state = 0;
+			*pflags = GPIO_LED_GREEN_OFF_AMBER_OFF;
+		} else {
+			/* Turn on. */
+			ha->beacon_color_state = QLA_LED_GRN_ON;
+			*pflags = GPIO_LED_GREEN_ON_AMBER_OFF;
+		}
+	}
+}
+
+void
+qla2x00_beacon_blink(struct scsi_qla_host *ha)
+{
+	uint16_t gpio_enable;
+	uint16_t gpio_data;
+	uint16_t led_color = 0;
+	unsigned long flags;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	if (ha->pio_address)
+		reg = (struct device_reg_2xxx __iomem *)ha->pio_address;
+
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+
+	/* Save the Original GPIOE. */
+	if (ha->pio_address) {
+		gpio_enable = RD_REG_WORD_PIO(&reg->gpioe);
+		gpio_data = RD_REG_WORD_PIO(&reg->gpiod);
+	} else {
+		gpio_enable = RD_REG_WORD(&reg->gpioe);
+		gpio_data = RD_REG_WORD(&reg->gpiod);
+	}
+
+	/* Set the modified gpio_enable values */
+	gpio_enable |= GPIO_LED_MASK;
+
+	if (ha->pio_address) {
+		WRT_REG_WORD_PIO(&reg->gpioe, gpio_enable);
+	} else {
+		WRT_REG_WORD(&reg->gpioe, gpio_enable);
+		RD_REG_WORD(&reg->gpioe);
+	}
+
+	qla2x00_flip_colors(ha, &led_color);
+
+	/* Clear out any previously set LED color. */
+	gpio_data &= ~GPIO_LED_MASK;
+
+	/* Set the new input LED color to GPIOD. */
+	gpio_data |= led_color;
+
+	/* Set the modified gpio_data values */
+	if (ha->pio_address) {
+		WRT_REG_WORD_PIO(&reg->gpiod, gpio_data);
+	} else {
+		WRT_REG_WORD(&reg->gpiod, gpio_data);
+		RD_REG_WORD(&reg->gpiod);
+	}
+
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+}
+
+int
+qla2x00_beacon_on(struct scsi_qla_host *ha)
+{
+	uint16_t gpio_enable;
+	uint16_t gpio_data;
+	unsigned long flags;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	ha->fw_options[1] &= ~FO1_SET_EMPHASIS_SWING;
+	ha->fw_options[1] |= FO1_DISABLE_GPIO6_7;
+
+	if (qla2x00_set_fw_options(ha, ha->fw_options) != QLA_SUCCESS) {
+		qla_printk(KERN_WARNING, ha,
+		    "Unable to update fw options (beacon on).\n");
+		return QLA_FUNCTION_FAILED;
+	}
+
+	if (ha->pio_address)
+		reg = (struct device_reg_2xxx __iomem *)ha->pio_address;
+
+	/* Turn off LEDs. */
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+	if (ha->pio_address) {
+		gpio_enable = RD_REG_WORD_PIO(&reg->gpioe);
+		gpio_data = RD_REG_WORD_PIO(&reg->gpiod);
+	} else {
+		gpio_enable = RD_REG_WORD(&reg->gpioe);
+		gpio_data = RD_REG_WORD(&reg->gpiod);
+	}
+	gpio_enable |= GPIO_LED_MASK;
+
+	/* Set the modified gpio_enable values. */
+	if (ha->pio_address) {
+		WRT_REG_WORD_PIO(&reg->gpioe, gpio_enable);
+	} else {
+		WRT_REG_WORD(&reg->gpioe, gpio_enable);
+		RD_REG_WORD(&reg->gpioe);
+	}
+
+	/* Clear out previously set LED colour. */
+	gpio_data &= ~GPIO_LED_MASK;
+	if (ha->pio_address) {
+		WRT_REG_WORD_PIO(&reg->gpiod, gpio_data);
+	} else {
+		WRT_REG_WORD(&reg->gpiod, gpio_data);
+		RD_REG_WORD(&reg->gpiod);
+	}
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+
+	/*
+	 * Let the per HBA timer kick off the blinking process based on
+	 * the following flags. No need to do anything else now.
+	 */
+	ha->beacon_blink_led = 1;
+	ha->beacon_color_state = 0;
+
+	return QLA_SUCCESS;
+}
+
+int
+qla2x00_beacon_off(struct scsi_qla_host *ha)
+{
+	int rval = QLA_SUCCESS;
+
+	ha->beacon_blink_led = 0;
+
+	/* Set the on flag so when it gets flipped it will be off. */
+	if (IS_QLA2322(ha))
+		ha->beacon_color_state = QLA_LED_ALL_ON;
+	else
+		ha->beacon_color_state = QLA_LED_GRN_ON;
+
+	ha->isp_ops.beacon_blink(ha);	/* This turns green LED off */
+
+	ha->fw_options[1] &= ~FO1_SET_EMPHASIS_SWING;
+	ha->fw_options[1] &= ~FO1_DISABLE_GPIO6_7;
+
+	rval = qla2x00_set_fw_options(ha, ha->fw_options);
+	if (rval != QLA_SUCCESS)
+		qla_printk(KERN_WARNING, ha,
+		    "Unable to update fw options (beacon off).\n");
+	return rval;
+}
+
+
+static inline void
+qla24xx_flip_colors(scsi_qla_host_t *ha, uint16_t *pflags)
+{
+	/* Flip all colors. */
+	if (ha->beacon_color_state == QLA_LED_ALL_ON) {
+		/* Turn off. */
+		ha->beacon_color_state = 0;
+		*pflags = 0;
+	} else {
+		/* Turn on. */
+		ha->beacon_color_state = QLA_LED_ALL_ON;
+		*pflags = GPDX_LED_YELLOW_ON | GPDX_LED_AMBER_ON;
+	}
+}
+
+void
+qla24xx_beacon_blink(struct scsi_qla_host *ha)
+{
+	uint16_t led_color = 0;
+	uint32_t gpio_data;
+	unsigned long flags;
+	struct device_reg_24xx __iomem *reg = &ha->iobase->isp24;
+
+	/* Save the Original GPIOD. */
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+	gpio_data = RD_REG_DWORD(&reg->gpiod);
+
+	/* Enable the gpio_data reg for update. */
+	gpio_data |= GPDX_LED_UPDATE_MASK;
+
+	WRT_REG_DWORD(&reg->gpiod, gpio_data);
+	gpio_data = RD_REG_DWORD(&reg->gpiod);
+
+	/* Set the color bits. */
+	qla24xx_flip_colors(ha, &led_color);
+
+	/* Clear out any previously set LED color. */
+	gpio_data &= ~GPDX_LED_COLOR_MASK;
+
+	/* Set the new input LED color to GPIOD. */
+	gpio_data |= led_color;
+
+	/* Set the modified gpio_data values. */
+	WRT_REG_DWORD(&reg->gpiod, gpio_data);
+	gpio_data = RD_REG_DWORD(&reg->gpiod);
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+}
+
+int
+qla24xx_beacon_on(struct scsi_qla_host *ha)
+{
+	uint32_t gpio_data;
+	unsigned long flags;
+	struct device_reg_24xx __iomem *reg = &ha->iobase->isp24;
+
+	if (ha->beacon_blink_led == 0) {
+		/* Enable firmware for update */
+		ha->fw_options[1] |= ADD_FO1_DISABLE_GPIO_LED_CTRL;
+
+		if (qla2x00_set_fw_options(ha, ha->fw_options) != QLA_SUCCESS)
+			return QLA_FUNCTION_FAILED;
+
+		if (qla2x00_get_fw_options(ha, ha->fw_options) !=
+		    QLA_SUCCESS) {
+			qla_printk(KERN_WARNING, ha,
+			    "Unable to update fw options (beacon on).\n");
+			return QLA_FUNCTION_FAILED;
+		}
+
+		spin_lock_irqsave(&ha->hardware_lock, flags);
+		gpio_data = RD_REG_DWORD(&reg->gpiod);
+
+		/* Enable the gpio_data reg for update. */
+		gpio_data |= GPDX_LED_UPDATE_MASK;
+		WRT_REG_DWORD(&reg->gpiod, gpio_data);
+		RD_REG_DWORD(&reg->gpiod);
+
+		spin_unlock_irqrestore(&ha->hardware_lock, flags);
+	}
+
+	/* So all colors blink together. */
+	ha->beacon_color_state = 0;
+
+	/* Let the per HBA timer kick off the blinking process. */
+	ha->beacon_blink_led = 1;
+
+	return QLA_SUCCESS;
+}
+
+int
+qla24xx_beacon_off(struct scsi_qla_host *ha)
+{
+	uint32_t gpio_data;
+	unsigned long flags;
+	struct device_reg_24xx __iomem *reg = &ha->iobase->isp24;
+
+	ha->beacon_blink_led = 0;
+	ha->beacon_color_state = QLA_LED_ALL_ON;
+
+	ha->isp_ops.beacon_blink(ha);	/* Will flip to all off. */
+
+	/* Give control back to firmware. */
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+	gpio_data = RD_REG_DWORD(&reg->gpiod);
+
+	/* Disable the gpio_data reg for update. */
+	gpio_data &= ~GPDX_LED_UPDATE_MASK;
+	WRT_REG_DWORD(&reg->gpiod, gpio_data);
+	RD_REG_DWORD(&reg->gpiod);
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+
+	ha->fw_options[1] &= ~ADD_FO1_DISABLE_GPIO_LED_CTRL;
+
+	if (qla2x00_set_fw_options(ha, ha->fw_options) != QLA_SUCCESS) {
+		qla_printk(KERN_WARNING, ha,
+		    "Unable to update fw options (beacon off).\n");
+		return QLA_FUNCTION_FAILED;
+	}
+
+	if (qla2x00_get_fw_options(ha, ha->fw_options) != QLA_SUCCESS) {
+		qla_printk(KERN_WARNING, ha,
+		    "Unable to get fw options (beacon off).\n");
+		return QLA_FUNCTION_FAILED;
+	}
+
+	return QLA_SUCCESS;
+}

From 1b3f63659bd353ae460c35f5793a9fd46cc95014 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 31 Jan 2006 16:05:12 -0800
Subject: [PATCH 011/509] [SCSI] qla2xxx: Return correct data-len during NVRAM
 retrieval.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_attr.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index 049e5cf1af7f..73d10b35091e 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -113,7 +113,7 @@ qla2x00_sysfs_read_nvram(struct kobject *kobj, char *buf, loff_t off,
 	    struct device, kobj)));
 	unsigned long	flags;
 
-	if (!capable(CAP_SYS_ADMIN) || off != 0 || count != ha->nvram_size)
+	if (!capable(CAP_SYS_ADMIN) || off != 0)
 		return 0;
 
 	/* Read NVRAM. */
@@ -122,7 +122,7 @@ qla2x00_sysfs_read_nvram(struct kobject *kobj, char *buf, loff_t off,
 	    ha->nvram_size);
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
 
-	return (count);
+	return ha->nvram_size;
 }
 
 static ssize_t
@@ -174,7 +174,7 @@ static struct bin_attribute sysfs_nvram_attr = {
 		.mode = S_IRUSR | S_IWUSR,
 		.owner = THIS_MODULE,
 	},
-	.size = 0,
+	.size = 512,
 	.read = qla2x00_sysfs_read_nvram,
 	.write = qla2x00_sysfs_write_nvram,
 };
@@ -185,7 +185,6 @@ qla2x00_alloc_sysfs_attr(scsi_qla_host_t *ha)
 	struct Scsi_Host *host = ha->host;
 
 	sysfs_create_bin_file(&host->shost_gendev.kobj, &sysfs_fw_dump_attr);
-	sysfs_nvram_attr.size = ha->nvram_size;
 	sysfs_create_bin_file(&host->shost_gendev.kobj, &sysfs_nvram_attr);
 }
 

From 854165f4245c4a3b4a8cc363ba2050033151e196 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 31 Jan 2006 16:05:17 -0800
Subject: [PATCH 012/509] [SCSI] qla2xxx: Add support to retrieve/update HBA
 option-rom.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_attr.c | 135 +++++++
 drivers/scsi/qla2xxx/qla_def.h  |  17 +-
 drivers/scsi/qla2xxx/qla_gbl.h  |  11 +
 drivers/scsi/qla2xxx/qla_os.c   |  15 +-
 drivers/scsi/qla2xxx/qla_sup.c  | 669 ++++++++++++++++++++++++++++++++
 5 files changed, 845 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c
index 73d10b35091e..92b3e13e9061 100644
--- a/drivers/scsi/qla2xxx/qla_attr.c
+++ b/drivers/scsi/qla2xxx/qla_attr.c
@@ -179,6 +179,135 @@ static struct bin_attribute sysfs_nvram_attr = {
 	.write = qla2x00_sysfs_write_nvram,
 };
 
+static ssize_t
+qla2x00_sysfs_read_optrom(struct kobject *kobj, char *buf, loff_t off,
+    size_t count)
+{
+	struct scsi_qla_host *ha = to_qla_host(dev_to_shost(container_of(kobj,
+	    struct device, kobj)));
+
+	if (ha->optrom_state != QLA_SREADING)
+		return 0;
+	if (off > ha->optrom_size)
+		return 0;
+	if (off + count > ha->optrom_size)
+		count = ha->optrom_size - off;
+
+	memcpy(buf, &ha->optrom_buffer[off], count);
+
+	return count;
+}
+
+static ssize_t
+qla2x00_sysfs_write_optrom(struct kobject *kobj, char *buf, loff_t off,
+    size_t count)
+{
+	struct scsi_qla_host *ha = to_qla_host(dev_to_shost(container_of(kobj,
+	    struct device, kobj)));
+
+	if (ha->optrom_state != QLA_SWRITING)
+		return -EINVAL;
+	if (off > ha->optrom_size)
+		return -ERANGE;
+	if (off + count > ha->optrom_size)
+		count = ha->optrom_size - off;
+
+	memcpy(&ha->optrom_buffer[off], buf, count);
+
+	return count;
+}
+
+static struct bin_attribute sysfs_optrom_attr = {
+	.attr = {
+		.name = "optrom",
+		.mode = S_IRUSR | S_IWUSR,
+		.owner = THIS_MODULE,
+	},
+	.size = OPTROM_SIZE_24XX,
+	.read = qla2x00_sysfs_read_optrom,
+	.write = qla2x00_sysfs_write_optrom,
+};
+
+static ssize_t
+qla2x00_sysfs_write_optrom_ctl(struct kobject *kobj, char *buf, loff_t off,
+    size_t count)
+{
+	struct scsi_qla_host *ha = to_qla_host(dev_to_shost(container_of(kobj,
+	    struct device, kobj)));
+	int val;
+
+	if (off)
+		return 0;
+
+	if (sscanf(buf, "%d", &val) != 1)
+		return -EINVAL;
+
+	switch (val) {
+	case 0:
+		if (ha->optrom_state != QLA_SREADING &&
+		    ha->optrom_state != QLA_SWRITING)
+			break;
+
+		ha->optrom_state = QLA_SWAITING;
+		vfree(ha->optrom_buffer);
+		ha->optrom_buffer = NULL;
+		break;
+	case 1:
+		if (ha->optrom_state != QLA_SWAITING)
+			break;
+
+		ha->optrom_state = QLA_SREADING;
+		ha->optrom_buffer = (uint8_t *)vmalloc(ha->optrom_size);
+		if (ha->optrom_buffer == NULL) {
+			qla_printk(KERN_WARNING, ha,
+			    "Unable to allocate memory for optrom retrieval "
+			    "(%x).\n", ha->optrom_size);
+
+			ha->optrom_state = QLA_SWAITING;
+			return count;
+		}
+
+		memset(ha->optrom_buffer, 0, ha->optrom_size);
+		ha->isp_ops.read_optrom(ha, ha->optrom_buffer, 0,
+		    ha->optrom_size);
+		break;
+	case 2:
+		if (ha->optrom_state != QLA_SWAITING)
+			break;
+
+		ha->optrom_state = QLA_SWRITING;
+		ha->optrom_buffer = (uint8_t *)vmalloc(ha->optrom_size);
+		if (ha->optrom_buffer == NULL) {
+			qla_printk(KERN_WARNING, ha,
+			    "Unable to allocate memory for optrom update "
+			    "(%x).\n", ha->optrom_size);
+
+			ha->optrom_state = QLA_SWAITING;
+			return count;
+		}
+		memset(ha->optrom_buffer, 0, ha->optrom_size);
+		break;
+	case 3:
+		if (ha->optrom_state != QLA_SWRITING)
+			break;
+
+		ha->isp_ops.write_optrom(ha, ha->optrom_buffer, 0,
+		    ha->optrom_size);
+		break;
+	}
+	return count;
+}
+
+static struct bin_attribute sysfs_optrom_ctl_attr = {
+	.attr = {
+		.name = "optrom_ctl",
+		.mode = S_IWUSR,
+		.owner = THIS_MODULE,
+	},
+	.size = 0,
+	.write = qla2x00_sysfs_write_optrom_ctl,
+};
+
 void
 qla2x00_alloc_sysfs_attr(scsi_qla_host_t *ha)
 {
@@ -186,6 +315,9 @@ qla2x00_alloc_sysfs_attr(scsi_qla_host_t *ha)
 
 	sysfs_create_bin_file(&host->shost_gendev.kobj, &sysfs_fw_dump_attr);
 	sysfs_create_bin_file(&host->shost_gendev.kobj, &sysfs_nvram_attr);
+	sysfs_create_bin_file(&host->shost_gendev.kobj, &sysfs_optrom_attr);
+	sysfs_create_bin_file(&host->shost_gendev.kobj,
+	    &sysfs_optrom_ctl_attr);
 }
 
 void
@@ -195,6 +327,9 @@ qla2x00_free_sysfs_attr(scsi_qla_host_t *ha)
 
 	sysfs_remove_bin_file(&host->shost_gendev.kobj, &sysfs_fw_dump_attr);
 	sysfs_remove_bin_file(&host->shost_gendev.kobj, &sysfs_nvram_attr);
+	sysfs_remove_bin_file(&host->shost_gendev.kobj, &sysfs_optrom_attr);
+	sysfs_remove_bin_file(&host->shost_gendev.kobj,
+	    &sysfs_optrom_ctl_attr);
 
 	if (ha->beacon_blink_led == 1)
 		ha->isp_ops.beacon_off(ha);
diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h
index 7cb8b5a5e659..b31a03bbd14f 100644
--- a/drivers/scsi/qla2xxx/qla_def.h
+++ b/drivers/scsi/qla2xxx/qla_def.h
@@ -2214,6 +2214,11 @@ struct isp_operations {
 	int (*beacon_on) (struct scsi_qla_host *);
 	int (*beacon_off) (struct scsi_qla_host *);
 	void (*beacon_blink) (struct scsi_qla_host *);
+
+	uint8_t * (*read_optrom) (struct scsi_qla_host *, uint8_t *,
+		uint32_t, uint32_t);
+	int (*write_optrom) (struct scsi_qla_host *, uint8_t *, uint32_t,
+		uint32_t);
 };
 
 /*
@@ -2505,6 +2510,14 @@ typedef struct scsi_qla_host {
 	uint8_t		*port_name;
 	uint32_t    isp_abort_cnt;
 
+	/* Option ROM information. */
+	char		*optrom_buffer;
+	uint32_t	optrom_size;
+	int		optrom_state;
+#define QLA_SWAITING	0
+#define QLA_SREADING	1
+#define QLA_SWRITING	2
+
 	/* Needed for BEACON */
 	uint16_t	beacon_blink_led;
 	uint8_t		beacon_color_state;
@@ -2582,7 +2595,9 @@ struct _qla2x00stats  {
 /*
  * Flash support definitions
  */
-#define FLASH_IMAGE_SIZE	131072
+#define OPTROM_SIZE_2300	0x20000
+#define OPTROM_SIZE_2322	0x100000
+#define OPTROM_SIZE_24XX	0x100000
 
 #include "qla_gbl.h"
 #include "qla_dbg.h"
diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h
index eb35198bc8cf..ffdc2680f049 100644
--- a/drivers/scsi/qla2xxx/qla_gbl.h
+++ b/drivers/scsi/qla2xxx/qla_gbl.h
@@ -79,6 +79,8 @@ extern int qla2x00_down_timeout(struct semaphore *, unsigned long);
 
 extern struct fw_blob *qla2x00_request_firmware(scsi_qla_host_t *);
 
+extern int qla2x00_wait_for_hba_online(scsi_qla_host_t *);
+
 /*
  * Global Function Prototypes in qla_iocb.c source file.
  */
@@ -240,6 +242,15 @@ extern int qla24xx_beacon_on(struct scsi_qla_host *);
 extern int qla24xx_beacon_off(struct scsi_qla_host *);
 extern void qla24xx_beacon_blink(struct scsi_qla_host *);
 
+extern uint8_t *qla2x00_read_optrom_data(struct scsi_qla_host *, uint8_t *,
+    uint32_t, uint32_t);
+extern int qla2x00_write_optrom_data(struct scsi_qla_host *, uint8_t *,
+    uint32_t, uint32_t);
+extern uint8_t *qla24xx_read_optrom_data(struct scsi_qla_host *, uint8_t *,
+    uint32_t, uint32_t);
+extern int qla24xx_write_optrom_data(struct scsi_qla_host *, uint8_t *,
+    uint32_t, uint32_t);
+
 /*
  * Global Function Prototypes in qla_dbg.c source file.
  */
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index 57179dabcccf..495ccbc7f8cb 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -513,7 +513,7 @@ qla2x00_eh_wait_on_command(scsi_qla_host_t *ha, struct scsi_cmnd *cmd)
  *    Success (Adapter is online) : 0
  *    Failed  (Adapter is offline/disabled) : 1
  */
-static int
+int
 qla2x00_wait_for_hba_online(scsi_qla_host_t *ha)
 {
 	int		return_status;
@@ -1271,6 +1271,9 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 	fc_port_t *fcport;
 	struct scsi_host_template *sht;
 
+if (PCI_FUNC(pdev->devfn))
+	goto probe_out;
+
 	if (pci_enable_device(pdev))
 		goto probe_out;
 
@@ -1313,6 +1316,7 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 	ha->init_cb_size = sizeof(init_cb_t);
 	ha->mgmt_svr_loop_id = MANAGEMENT_SERVER;
 	ha->link_data_rate = LDR_UNKNOWN;
+	ha->optrom_size = OPTROM_SIZE_2300;
 
 	/* Assign ISP specific operations. */
 	ha->isp_ops.pci_config		= qla2100_pci_config;
@@ -1340,6 +1344,8 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 	ha->isp_ops.write_nvram		= qla2x00_write_nvram_data;
 	ha->isp_ops.fw_dump		= qla2100_fw_dump;
 	ha->isp_ops.ascii_fw_dump	= qla2100_ascii_fw_dump;
+	ha->isp_ops.read_optrom		= qla2x00_read_optrom_data;
+	ha->isp_ops.write_optrom	= qla2x00_write_optrom_data;
 	if (IS_QLA2100(ha)) {
 		host->max_id = MAX_TARGETS_2100;
 		ha->mbx_count = MAILBOX_REGISTER_COUNT_2100;
@@ -1369,6 +1375,8 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 		ha->isp_ops.beacon_off = qla2x00_beacon_off;
 		ha->isp_ops.beacon_blink = qla2x00_beacon_blink;
 		ha->gid_list_info_size = 6;
+		if (IS_QLA2322(ha) || IS_QLA6322(ha))
+			ha->optrom_size = OPTROM_SIZE_2322;
 	} else if (IS_QLA24XX(ha) || IS_QLA25XX(ha)) {
 		host->max_id = MAX_TARGETS_2200;
 		ha->mbx_count = MAILBOX_REGISTER_COUNT;
@@ -1404,10 +1412,13 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 		ha->isp_ops.write_nvram = qla24xx_write_nvram_data;
 		ha->isp_ops.fw_dump = qla24xx_fw_dump;
 		ha->isp_ops.ascii_fw_dump = qla24xx_ascii_fw_dump;
+		ha->isp_ops.read_optrom	= qla24xx_read_optrom_data;
+		ha->isp_ops.write_optrom = qla24xx_write_optrom_data;
 		ha->isp_ops.beacon_on = qla24xx_beacon_on;
 		ha->isp_ops.beacon_off = qla24xx_beacon_off;
 		ha->isp_ops.beacon_blink = qla24xx_beacon_blink;
 		ha->gid_list_info_size = 8;
+		ha->optrom_size = OPTROM_SIZE_24XX;
 	}
 	host->can_queue = ha->request_q_length + 128;
 
@@ -2073,6 +2084,8 @@ qla2x00_mem_free(scsi_qla_host_t *ha)
 	ha->fw_dumped = 0;
 	ha->fw_dump_reading = 0;
 	ha->fw_dump_buffer = NULL;
+
+	vfree(ha->optrom_buffer);
 }
 
 /*
diff --git a/drivers/scsi/qla2xxx/qla_sup.c b/drivers/scsi/qla2xxx/qla_sup.c
index 3105bb93cc19..3866a5760f15 100644
--- a/drivers/scsi/qla2xxx/qla_sup.c
+++ b/drivers/scsi/qla2xxx/qla_sup.c
@@ -989,3 +989,672 @@ qla24xx_beacon_off(struct scsi_qla_host *ha)
 
 	return QLA_SUCCESS;
 }
+
+
+/*
+ * Flash support routines
+ */
+
+/**
+ * qla2x00_flash_enable() - Setup flash for reading and writing.
+ * @ha: HA context
+ */
+static void
+qla2x00_flash_enable(scsi_qla_host_t *ha)
+{
+	uint16_t data;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	data = RD_REG_WORD(&reg->ctrl_status);
+	data |= CSR_FLASH_ENABLE;
+	WRT_REG_WORD(&reg->ctrl_status, data);
+	RD_REG_WORD(&reg->ctrl_status);		/* PCI Posting. */
+}
+
+/**
+ * qla2x00_flash_disable() - Disable flash and allow RISC to run.
+ * @ha: HA context
+ */
+static void
+qla2x00_flash_disable(scsi_qla_host_t *ha)
+{
+	uint16_t data;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	data = RD_REG_WORD(&reg->ctrl_status);
+	data &= ~(CSR_FLASH_ENABLE);
+	WRT_REG_WORD(&reg->ctrl_status, data);
+	RD_REG_WORD(&reg->ctrl_status);		/* PCI Posting. */
+}
+
+/**
+ * qla2x00_read_flash_byte() - Reads a byte from flash
+ * @ha: HA context
+ * @addr: Address in flash to read
+ *
+ * A word is read from the chip, but, only the lower byte is valid.
+ *
+ * Returns the byte read from flash @addr.
+ */
+static uint8_t
+qla2x00_read_flash_byte(scsi_qla_host_t *ha, uint32_t addr)
+{
+	uint16_t data;
+	uint16_t bank_select;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	bank_select = RD_REG_WORD(&reg->ctrl_status);
+
+	if (IS_QLA2322(ha) || IS_QLA6322(ha)) {
+		/* Specify 64K address range: */
+		/*  clear out Module Select and Flash Address bits [19:16]. */
+		bank_select &= ~0xf8;
+		bank_select |= addr >> 12 & 0xf0;
+		bank_select |= CSR_FLASH_64K_BANK;
+		WRT_REG_WORD(&reg->ctrl_status, bank_select);
+		RD_REG_WORD(&reg->ctrl_status);	/* PCI Posting. */
+
+		WRT_REG_WORD(&reg->flash_address, (uint16_t)addr);
+		data = RD_REG_WORD(&reg->flash_data);
+
+		return (uint8_t)data;
+	}
+
+	/* Setup bit 16 of flash address. */
+	if ((addr & BIT_16) && ((bank_select & CSR_FLASH_64K_BANK) == 0)) {
+		bank_select |= CSR_FLASH_64K_BANK;
+		WRT_REG_WORD(&reg->ctrl_status, bank_select);
+		RD_REG_WORD(&reg->ctrl_status);	/* PCI Posting. */
+	} else if (((addr & BIT_16) == 0) &&
+	    (bank_select & CSR_FLASH_64K_BANK)) {
+		bank_select &= ~(CSR_FLASH_64K_BANK);
+		WRT_REG_WORD(&reg->ctrl_status, bank_select);
+		RD_REG_WORD(&reg->ctrl_status);	/* PCI Posting. */
+	}
+
+	/* Always perform IO mapped accesses to the FLASH registers. */
+	if (ha->pio_address) {
+		uint16_t data2;
+
+		reg = (struct device_reg_2xxx __iomem *)ha->pio_address;
+		WRT_REG_WORD_PIO(&reg->flash_address, (uint16_t)addr);
+		do {
+			data = RD_REG_WORD_PIO(&reg->flash_data);
+			barrier();
+			cpu_relax();
+			data2 = RD_REG_WORD_PIO(&reg->flash_data);
+		} while (data != data2);
+	} else {
+		WRT_REG_WORD(&reg->flash_address, (uint16_t)addr);
+		data = qla2x00_debounce_register(&reg->flash_data);
+	}
+
+	return (uint8_t)data;
+}
+
+/**
+ * qla2x00_write_flash_byte() - Write a byte to flash
+ * @ha: HA context
+ * @addr: Address in flash to write
+ * @data: Data to write
+ */
+static void
+qla2x00_write_flash_byte(scsi_qla_host_t *ha, uint32_t addr, uint8_t data)
+{
+	uint16_t bank_select;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	bank_select = RD_REG_WORD(&reg->ctrl_status);
+	if (IS_QLA2322(ha) || IS_QLA6322(ha)) {
+		/* Specify 64K address range: */
+		/*  clear out Module Select and Flash Address bits [19:16]. */
+		bank_select &= ~0xf8;
+		bank_select |= addr >> 12 & 0xf0;
+		bank_select |= CSR_FLASH_64K_BANK;
+		WRT_REG_WORD(&reg->ctrl_status, bank_select);
+		RD_REG_WORD(&reg->ctrl_status);	/* PCI Posting. */
+
+		WRT_REG_WORD(&reg->flash_address, (uint16_t)addr);
+		RD_REG_WORD(&reg->ctrl_status);		/* PCI Posting. */
+		WRT_REG_WORD(&reg->flash_data, (uint16_t)data);
+		RD_REG_WORD(&reg->ctrl_status);		/* PCI Posting. */
+
+		return;
+	}
+
+	/* Setup bit 16 of flash address. */
+	if ((addr & BIT_16) && ((bank_select & CSR_FLASH_64K_BANK) == 0)) {
+		bank_select |= CSR_FLASH_64K_BANK;
+		WRT_REG_WORD(&reg->ctrl_status, bank_select);
+		RD_REG_WORD(&reg->ctrl_status);	/* PCI Posting. */
+	} else if (((addr & BIT_16) == 0) &&
+	    (bank_select & CSR_FLASH_64K_BANK)) {
+		bank_select &= ~(CSR_FLASH_64K_BANK);
+		WRT_REG_WORD(&reg->ctrl_status, bank_select);
+		RD_REG_WORD(&reg->ctrl_status);	/* PCI Posting. */
+	}
+
+	/* Always perform IO mapped accesses to the FLASH registers. */
+	if (ha->pio_address) {
+		reg = (struct device_reg_2xxx __iomem *)ha->pio_address;
+		WRT_REG_WORD_PIO(&reg->flash_address, (uint16_t)addr);
+		WRT_REG_WORD_PIO(&reg->flash_data, (uint16_t)data);
+	} else {
+		WRT_REG_WORD(&reg->flash_address, (uint16_t)addr);
+		RD_REG_WORD(&reg->ctrl_status);		/* PCI Posting. */
+		WRT_REG_WORD(&reg->flash_data, (uint16_t)data);
+		RD_REG_WORD(&reg->ctrl_status);		/* PCI Posting. */
+	}
+}
+
+/**
+ * qla2x00_poll_flash() - Polls flash for completion.
+ * @ha: HA context
+ * @addr: Address in flash to poll
+ * @poll_data: Data to be polled
+ * @man_id: Flash manufacturer ID
+ * @flash_id: Flash ID
+ *
+ * This function polls the device until bit 7 of what is read matches data
+ * bit 7 or until data bit 5 becomes a 1.  If that hapens, the flash ROM timed
+ * out (a fatal error).  The flash book recommeds reading bit 7 again after
+ * reading bit 5 as a 1.
+ *
+ * Returns 0 on success, else non-zero.
+ */
+static int
+qla2x00_poll_flash(scsi_qla_host_t *ha, uint32_t addr, uint8_t poll_data,
+    uint8_t man_id, uint8_t flash_id)
+{
+	int status;
+	uint8_t flash_data;
+	uint32_t cnt;
+
+	status = 1;
+
+	/* Wait for 30 seconds for command to finish. */
+	poll_data &= BIT_7;
+	for (cnt = 3000000; cnt; cnt--) {
+		flash_data = qla2x00_read_flash_byte(ha, addr);
+		if ((flash_data & BIT_7) == poll_data) {
+			status = 0;
+			break;
+		}
+
+		if (man_id != 0x40 && man_id != 0xda) {
+			if ((flash_data & BIT_5) && cnt > 2)
+				cnt = 2;
+		}
+		udelay(10);
+		barrier();
+	}
+	return status;
+}
+
+#define IS_OEM_001(ha) \
+	((ha)->pdev->device == PCI_DEVICE_ID_QLOGIC_ISP2322 && \
+	 (ha)->pdev->subsystem_vendor == 0x1028 && \
+	 (ha)->pdev->subsystem_device == 0x0170)
+
+/**
+ * qla2x00_program_flash_address() - Programs a flash address
+ * @ha: HA context
+ * @addr: Address in flash to program
+ * @data: Data to be written in flash
+ * @man_id: Flash manufacturer ID
+ * @flash_id: Flash ID
+ *
+ * Returns 0 on success, else non-zero.
+ */
+static int
+qla2x00_program_flash_address(scsi_qla_host_t *ha, uint32_t addr, uint8_t data,
+    uint8_t man_id, uint8_t flash_id)
+{
+	/* Write Program Command Sequence. */
+	if (IS_OEM_001(ha)) {
+		qla2x00_write_flash_byte(ha, 0xaaa, 0xaa);
+		qla2x00_write_flash_byte(ha, 0x555, 0x55);
+		qla2x00_write_flash_byte(ha, 0xaaa, 0xa0);
+		qla2x00_write_flash_byte(ha, addr, data);
+	} else {
+		if (man_id == 0xda && flash_id == 0xc1) {
+			qla2x00_write_flash_byte(ha, addr, data);
+			if (addr & 0x7e)
+				return 0;
+		} else {
+			qla2x00_write_flash_byte(ha, 0x5555, 0xaa);
+			qla2x00_write_flash_byte(ha, 0x2aaa, 0x55);
+			qla2x00_write_flash_byte(ha, 0x5555, 0xa0);
+			qla2x00_write_flash_byte(ha, addr, data);
+		}
+	}
+
+	udelay(150);
+
+	/* Wait for write to complete. */
+	return qla2x00_poll_flash(ha, addr, data, man_id, flash_id);
+}
+
+/**
+ * qla2x00_erase_flash() - Erase the flash.
+ * @ha: HA context
+ * @man_id: Flash manufacturer ID
+ * @flash_id: Flash ID
+ *
+ * Returns 0 on success, else non-zero.
+ */
+static int
+qla2x00_erase_flash(scsi_qla_host_t *ha, uint8_t man_id, uint8_t flash_id)
+{
+	/* Individual Sector Erase Command Sequence */
+	if (IS_OEM_001(ha)) {
+		qla2x00_write_flash_byte(ha, 0xaaa, 0xaa);
+		qla2x00_write_flash_byte(ha, 0x555, 0x55);
+		qla2x00_write_flash_byte(ha, 0xaaa, 0x80);
+		qla2x00_write_flash_byte(ha, 0xaaa, 0xaa);
+		qla2x00_write_flash_byte(ha, 0x555, 0x55);
+		qla2x00_write_flash_byte(ha, 0xaaa, 0x10);
+	} else {
+		qla2x00_write_flash_byte(ha, 0x5555, 0xaa);
+		qla2x00_write_flash_byte(ha, 0x2aaa, 0x55);
+		qla2x00_write_flash_byte(ha, 0x5555, 0x80);
+		qla2x00_write_flash_byte(ha, 0x5555, 0xaa);
+		qla2x00_write_flash_byte(ha, 0x2aaa, 0x55);
+		qla2x00_write_flash_byte(ha, 0x5555, 0x10);
+	}
+
+	udelay(150);
+
+	/* Wait for erase to complete. */
+	return qla2x00_poll_flash(ha, 0x00, 0x80, man_id, flash_id);
+}
+
+/**
+ * qla2x00_erase_flash_sector() - Erase a flash sector.
+ * @ha: HA context
+ * @addr: Flash sector to erase
+ * @sec_mask: Sector address mask
+ * @man_id: Flash manufacturer ID
+ * @flash_id: Flash ID
+ *
+ * Returns 0 on success, else non-zero.
+ */
+static int
+qla2x00_erase_flash_sector(scsi_qla_host_t *ha, uint32_t addr,
+    uint32_t sec_mask, uint8_t man_id, uint8_t flash_id)
+{
+	/* Individual Sector Erase Command Sequence */
+	qla2x00_write_flash_byte(ha, 0x5555, 0xaa);
+	qla2x00_write_flash_byte(ha, 0x2aaa, 0x55);
+	qla2x00_write_flash_byte(ha, 0x5555, 0x80);
+	qla2x00_write_flash_byte(ha, 0x5555, 0xaa);
+	qla2x00_write_flash_byte(ha, 0x2aaa, 0x55);
+	if (man_id == 0x1f && flash_id == 0x13)
+		qla2x00_write_flash_byte(ha, addr & sec_mask, 0x10);
+	else
+		qla2x00_write_flash_byte(ha, addr & sec_mask, 0x30);
+
+	udelay(150);
+
+	/* Wait for erase to complete. */
+	return qla2x00_poll_flash(ha, addr, 0x80, man_id, flash_id);
+}
+
+/**
+ * qla2x00_get_flash_manufacturer() - Read manufacturer ID from flash chip.
+ * @man_id: Flash manufacturer ID
+ * @flash_id: Flash ID
+ */
+static void
+qla2x00_get_flash_manufacturer(scsi_qla_host_t *ha, uint8_t *man_id,
+    uint8_t *flash_id)
+{
+	qla2x00_write_flash_byte(ha, 0x5555, 0xaa);
+	qla2x00_write_flash_byte(ha, 0x2aaa, 0x55);
+	qla2x00_write_flash_byte(ha, 0x5555, 0x90);
+	*man_id = qla2x00_read_flash_byte(ha, 0x0000);
+	*flash_id = qla2x00_read_flash_byte(ha, 0x0001);
+	qla2x00_write_flash_byte(ha, 0x5555, 0xaa);
+	qla2x00_write_flash_byte(ha, 0x2aaa, 0x55);
+	qla2x00_write_flash_byte(ha, 0x5555, 0xf0);
+}
+
+
+static inline void
+qla2x00_suspend_hba(struct scsi_qla_host *ha)
+{
+	int cnt;
+	unsigned long flags;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	/* Suspend HBA. */
+	scsi_block_requests(ha->host);
+	ha->isp_ops.disable_intrs(ha);
+	set_bit(MBX_UPDATE_FLASH_ACTIVE, &ha->mbx_cmd_flags);
+
+	/* Pause RISC. */
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+	WRT_REG_WORD(&reg->hccr, HCCR_PAUSE_RISC);
+	RD_REG_WORD(&reg->hccr);
+	if (IS_QLA2100(ha) || IS_QLA2200(ha) || IS_QLA2300(ha)) {
+		for (cnt = 0; cnt < 30000; cnt++) {
+			if ((RD_REG_WORD(&reg->hccr) & HCCR_RISC_PAUSE) != 0)
+				break;
+			udelay(100);
+		}
+	} else {
+		udelay(10);
+	}
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+}
+
+static inline void
+qla2x00_resume_hba(struct scsi_qla_host *ha)
+{
+	/* Resume HBA. */
+	clear_bit(MBX_UPDATE_FLASH_ACTIVE, &ha->mbx_cmd_flags);
+	set_bit(ISP_ABORT_NEEDED, &ha->dpc_flags);
+	up(ha->dpc_wait);
+	qla2x00_wait_for_hba_online(ha);
+	scsi_unblock_requests(ha->host);
+}
+
+uint8_t *
+qla2x00_read_optrom_data(struct scsi_qla_host *ha, uint8_t *buf,
+    uint32_t offset, uint32_t length)
+{
+	unsigned long flags;
+	uint32_t addr, midpoint;
+	uint8_t *data;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	/* Suspend HBA. */
+	qla2x00_suspend_hba(ha);
+
+	/* Go with read. */
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+	midpoint = ha->optrom_size / 2;
+
+	qla2x00_flash_enable(ha);
+	WRT_REG_WORD(&reg->nvram, 0);
+	RD_REG_WORD(&reg->nvram);		/* PCI Posting. */
+	for (addr = offset, data = buf; addr < length; addr++, data++) {
+		if (addr == midpoint) {
+			WRT_REG_WORD(&reg->nvram, NVR_SELECT);
+			RD_REG_WORD(&reg->nvram);	/* PCI Posting. */
+		}
+
+		*data = qla2x00_read_flash_byte(ha, addr);
+	}
+	qla2x00_flash_disable(ha);
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+
+	/* Resume HBA. */
+	qla2x00_resume_hba(ha);
+
+	return buf;
+}
+
+int
+qla2x00_write_optrom_data(struct scsi_qla_host *ha, uint8_t *buf,
+    uint32_t offset, uint32_t length)
+{
+
+	int rval;
+	unsigned long flags;
+	uint8_t man_id, flash_id, sec_number, data;
+	uint16_t wd;
+	uint32_t addr, liter, sec_mask, rest_addr;
+	struct device_reg_2xxx __iomem *reg = &ha->iobase->isp;
+
+	/* Suspend HBA. */
+	qla2x00_suspend_hba(ha);
+
+	rval = QLA_SUCCESS;
+	sec_number = 0;
+
+	/* Reset ISP chip. */
+	spin_lock_irqsave(&ha->hardware_lock, flags);
+	WRT_REG_WORD(&reg->ctrl_status, CSR_ISP_SOFT_RESET);
+	pci_read_config_word(ha->pdev, PCI_COMMAND, &wd);
+
+	/* Go with write. */
+	qla2x00_flash_enable(ha);
+	do {	/* Loop once to provide quick error exit */
+		/* Structure of flash memory based on manufacturer */
+		if (IS_OEM_001(ha)) {
+			/* OEM variant with special flash part. */
+			man_id = flash_id = 0;
+			rest_addr = 0xffff;
+			sec_mask   = 0x10000;
+			goto update_flash;
+		}
+		qla2x00_get_flash_manufacturer(ha, &man_id, &flash_id);
+		switch (man_id) {
+		case 0x20: /* ST flash. */
+			if (flash_id == 0xd2 || flash_id == 0xe3) {
+				/*
+				 * ST m29w008at part - 64kb sector size with
+				 * 32kb,8kb,8kb,16kb sectors at memory address
+				 * 0xf0000.
+				 */
+				rest_addr = 0xffff;
+				sec_mask = 0x10000;
+				break;   
+			}
+			/*
+			 * ST m29w010b part - 16kb sector size
+			 * Default to 16kb sectors
+			 */
+			rest_addr = 0x3fff;
+			sec_mask = 0x1c000;
+			break;
+		case 0x40: /* Mostel flash. */
+			/* Mostel v29c51001 part - 512 byte sector size. */
+			rest_addr = 0x1ff;
+			sec_mask = 0x1fe00;
+			break;
+		case 0xbf: /* SST flash. */
+			/* SST39sf10 part - 4kb sector size. */
+			rest_addr = 0xfff;
+			sec_mask = 0x1f000;
+			break;
+		case 0xda: /* Winbond flash. */
+			/* Winbond W29EE011 part - 256 byte sector size. */
+			rest_addr = 0x7f;
+			sec_mask = 0x1ff80;
+			break;
+		case 0xc2: /* Macronix flash. */
+			/* 64k sector size. */
+			if (flash_id == 0x38 || flash_id == 0x4f) {
+				rest_addr = 0xffff;
+				sec_mask = 0x10000;
+				break;
+			}
+			/* Fall through... */
+
+		case 0x1f: /* Atmel flash. */
+			/* 512k sector size. */
+			if (flash_id == 0x13) {
+				rest_addr = 0x7fffffff;
+				sec_mask =   0x80000000;
+				break;
+			}
+			/* Fall through... */
+
+		case 0x01: /* AMD flash. */
+			if (flash_id == 0x38 || flash_id == 0x40 ||
+			    flash_id == 0x4f) {
+				/* Am29LV081 part - 64kb sector size. */
+				/* Am29LV002BT part - 64kb sector size. */
+				rest_addr = 0xffff;
+				sec_mask = 0x10000;
+				break;
+			} else if (flash_id == 0x3e) {
+				/*
+				 * Am29LV008b part - 64kb sector size with
+				 * 32kb,8kb,8kb,16kb sector at memory address
+				 * h0xf0000.
+				 */
+				rest_addr = 0xffff;
+				sec_mask = 0x10000;
+				break;
+			} else if (flash_id == 0x20 || flash_id == 0x6e) {
+				/*
+				 * Am29LV010 part or AM29f010 - 16kb sector
+				 * size.
+				 */
+				rest_addr = 0x3fff;
+				sec_mask = 0x1c000;
+				break;
+			} else if (flash_id == 0x6d) {
+				/* Am29LV001 part - 8kb sector size. */
+				rest_addr = 0x1fff;
+				sec_mask = 0x1e000;
+				break;
+			}
+		default:
+			/* Default to 16 kb sector size. */
+			rest_addr = 0x3fff;
+			sec_mask = 0x1c000;
+			break;
+		}
+
+update_flash:
+		if (IS_QLA2322(ha) || IS_QLA6322(ha)) {
+			if (qla2x00_erase_flash(ha, man_id, flash_id)) {
+				rval = QLA_FUNCTION_FAILED;
+				break;
+			}
+		}
+
+		for (addr = offset, liter = 0; liter < length; liter++,
+		    addr++) {
+			data = buf[liter];
+			/* Are we at the beginning of a sector? */
+			if ((addr & rest_addr) == 0) {
+				if (IS_QLA2322(ha) || IS_QLA6322(ha)) {
+					if (addr >= 0x10000UL) {
+						if (((addr >> 12) & 0xf0) &&
+						    ((man_id == 0x01 &&
+							flash_id == 0x3e) ||
+						     (man_id == 0x20 &&
+							 flash_id == 0xd2))) {
+							sec_number++;
+							if (sec_number == 1) {
+								rest_addr =
+								    0x7fff;
+								sec_mask =
+								    0x18000;
+							} else if (
+							    sec_number == 2 ||
+							    sec_number == 3) {
+								rest_addr =
+								    0x1fff;
+								sec_mask =
+								    0x1e000;
+							} else if (
+							    sec_number == 4) {
+								rest_addr =
+								    0x3fff;
+								sec_mask =
+								    0x1c000;
+							}
+						}
+					}
+				} else if (addr == ha->optrom_size / 2) {
+					WRT_REG_WORD(&reg->nvram, NVR_SELECT);
+					RD_REG_WORD(&reg->nvram);
+				}
+
+				if (flash_id == 0xda && man_id == 0xc1) {
+					qla2x00_write_flash_byte(ha, 0x5555,
+					    0xaa);
+					qla2x00_write_flash_byte(ha, 0x2aaa,
+					    0x55);
+					qla2x00_write_flash_byte(ha, 0x5555,
+					    0xa0);
+				} else if (!IS_QLA2322(ha) && !IS_QLA6322(ha)) {
+					/* Then erase it */
+					if (qla2x00_erase_flash_sector(ha,
+					    addr, sec_mask, man_id,
+					    flash_id)) {
+						rval = QLA_FUNCTION_FAILED;
+						break;
+					}
+					if (man_id == 0x01 && flash_id == 0x6d)
+						sec_number++;
+				}
+			}
+
+			if (man_id == 0x01 && flash_id == 0x6d) {
+				if (sec_number == 1 &&
+				    addr == (rest_addr - 1)) {
+					rest_addr = 0x0fff;
+					sec_mask   = 0x1f000;
+				} else if (sec_number == 3 && (addr & 0x7ffe)) {
+					rest_addr = 0x3fff;
+					sec_mask   = 0x1c000;
+				}
+			}
+
+			if (qla2x00_program_flash_address(ha, addr, data,
+			    man_id, flash_id)) {
+				rval = QLA_FUNCTION_FAILED;
+				break;
+			}
+		}
+	} while (0);
+	qla2x00_flash_disable(ha);
+	spin_unlock_irqrestore(&ha->hardware_lock, flags);
+
+	/* Resume HBA. */
+	qla2x00_resume_hba(ha);
+
+	return rval;
+}
+
+uint8_t *
+qla24xx_read_optrom_data(struct scsi_qla_host *ha, uint8_t *buf,
+    uint32_t offset, uint32_t length)
+{
+	/* Suspend HBA. */
+	scsi_block_requests(ha->host);
+	ha->isp_ops.disable_intrs(ha);
+	set_bit(MBX_UPDATE_FLASH_ACTIVE, &ha->mbx_cmd_flags);
+
+	/* Go with read. */
+	qla24xx_read_flash_data(ha, (uint32_t *)buf, offset >> 2, length >> 2);
+
+	/* Resume HBA. */
+	clear_bit(MBX_UPDATE_FLASH_ACTIVE, &ha->mbx_cmd_flags);
+	ha->isp_ops.enable_intrs(ha);
+	scsi_unblock_requests(ha->host);
+
+	return buf;
+}
+
+int
+qla24xx_write_optrom_data(struct scsi_qla_host *ha, uint8_t *buf,
+    uint32_t offset, uint32_t length)
+{
+	int rval;
+
+	/* Suspend HBA. */
+	scsi_block_requests(ha->host);
+	ha->isp_ops.disable_intrs(ha);
+	set_bit(MBX_UPDATE_FLASH_ACTIVE, &ha->mbx_cmd_flags);
+
+	/* Go with write. */
+	rval = qla24xx_write_flash_data(ha, (uint32_t *)buf, offset >> 2,
+	    length >> 2);
+
+	/* Resume HBA -- RISC reset needed. */
+	clear_bit(MBX_UPDATE_FLASH_ACTIVE, &ha->mbx_cmd_flags);
+	set_bit(ISP_ABORT_NEEDED, &ha->dpc_flags);
+	up(ha->dpc_wait);
+	qla2x00_wait_for_hba_online(ha);
+	scsi_unblock_requests(ha->host);
+
+	return rval;
+}

From d8a571135aad527e5984c8094b7977c6914f2550 Mon Sep 17 00:00:00 2001
From: Mark Haverkamp <markh@osdl.org>
Date: Wed, 1 Feb 2006 09:30:31 -0800
Subject: [PATCH 013/509] [SCSI] aacraid: reduce device probe warnings

Received from Mark Salyzyn.

This patch sets up some device quirks surrounding arrays to inform the
scsi layer that various mode pages are not supported. This reduces the
severity of the complaints that show up in the logs as the array devices
are enumerated.

Signed-off-by: Mark Haverkamp <markh@osdl.org>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/aacraid/linit.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c
index 0bf5f9a943e8..45fc171e5f36 100644
--- a/drivers/scsi/aacraid/linit.c
+++ b/drivers/scsi/aacraid/linit.c
@@ -386,6 +386,10 @@ static int aac_biosparm(struct scsi_device *sdev, struct block_device *bdev,
 static int aac_slave_configure(struct scsi_device *sdev)
 {
 	struct Scsi_Host *host = sdev->host;
+	if (sdev_channel(sdev) == CONTAINER_CHANNEL) {
+		sdev->skip_ms_page_8 = 1;
+		sdev->skip_ms_page_3f = 1;
+	}
 
 	if (sdev->tagged_supported)
 		scsi_adjust_queue_depth(sdev, MSG_ORDERED_TAG, 128);

From bfb35aa85057da4336af56a7f26e08031f4e3468 Mon Sep 17 00:00:00 2001
From: Mark Haverkamp <markh@osdl.org>
Date: Wed, 1 Feb 2006 09:30:55 -0800
Subject: [PATCH 014/509] [SCSI] aacraid: Update global function names

Received from Mark Salyzyn,

Reduce the possibility of namespace collision.  Prefix with aac_.

Signed-off-by: Mark Haverkamp <markh@osdl.org>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/aacraid/aachba.c   | 164 ++++++++++++++++----------------
 drivers/scsi/aacraid/aacraid.h  |  18 ++--
 drivers/scsi/aacraid/commctrl.c |  22 ++---
 drivers/scsi/aacraid/comminit.c |  12 +--
 drivers/scsi/aacraid/commsup.c  |  50 +++++-----
 drivers/scsi/aacraid/dpcsup.c   |   2 +-
 drivers/scsi/aacraid/linit.c    |   4 +-
 7 files changed, 137 insertions(+), 135 deletions(-)

diff --git a/drivers/scsi/aacraid/aachba.c b/drivers/scsi/aacraid/aachba.c
index 7139659dd952..b0f314e415c9 100644
--- a/drivers/scsi/aacraid/aachba.c
+++ b/drivers/scsi/aacraid/aachba.c
@@ -173,10 +173,10 @@ int aac_get_config_status(struct aac_dev *dev)
 	int status = 0;
 	struct fib * fibptr;
 
-	if (!(fibptr = fib_alloc(dev)))
+	if (!(fibptr = aac_fib_alloc(dev)))
 		return -ENOMEM;
 
-	fib_init(fibptr);
+	aac_fib_init(fibptr);
 	{
 		struct aac_get_config_status *dinfo;
 		dinfo = (struct aac_get_config_status *) fib_data(fibptr);
@@ -186,7 +186,7 @@ int aac_get_config_status(struct aac_dev *dev)
 		dinfo->count = cpu_to_le32(sizeof(((struct aac_get_config_status_resp *)NULL)->data));
 	}
 
-	status = fib_send(ContainerCommand,
+	status = aac_fib_send(ContainerCommand,
 			    fibptr,
 			    sizeof (struct aac_get_config_status),
 			    FsaNormal,
@@ -209,30 +209,30 @@ int aac_get_config_status(struct aac_dev *dev)
 			status = -EINVAL;
 		}
 	}
-	fib_complete(fibptr);
+	aac_fib_complete(fibptr);
 	/* Send a CT_COMMIT_CONFIG to enable discovery of devices */
 	if (status >= 0) {
 		if (commit == 1) {
 			struct aac_commit_config * dinfo;
-			fib_init(fibptr);
+			aac_fib_init(fibptr);
 			dinfo = (struct aac_commit_config *) fib_data(fibptr);
 	
 			dinfo->command = cpu_to_le32(VM_ContainerConfig);
 			dinfo->type = cpu_to_le32(CT_COMMIT_CONFIG);
 	
-			status = fib_send(ContainerCommand,
+			status = aac_fib_send(ContainerCommand,
 				    fibptr,
 				    sizeof (struct aac_commit_config),
 				    FsaNormal,
 				    1, 1,
 				    NULL, NULL);
-			fib_complete(fibptr);
+			aac_fib_complete(fibptr);
 		} else if (commit == 0) {
 			printk(KERN_WARNING
 			  "aac_get_config_status: Foreign device configurations are being ignored\n");
 		}
 	}
-	fib_free(fibptr);
+	aac_fib_free(fibptr);
 	return status;
 }
 
@@ -255,15 +255,15 @@ int aac_get_containers(struct aac_dev *dev)
 
 	instance = dev->scsi_host_ptr->unique_id;
 
-	if (!(fibptr = fib_alloc(dev)))
+	if (!(fibptr = aac_fib_alloc(dev)))
 		return -ENOMEM;
 
-	fib_init(fibptr);
+	aac_fib_init(fibptr);
 	dinfo = (struct aac_get_container_count *) fib_data(fibptr);
 	dinfo->command = cpu_to_le32(VM_ContainerConfig);
 	dinfo->type = cpu_to_le32(CT_GET_CONTAINER_COUNT);
 
-	status = fib_send(ContainerCommand,
+	status = aac_fib_send(ContainerCommand,
 		    fibptr,
 		    sizeof (struct aac_get_container_count),
 		    FsaNormal,
@@ -272,7 +272,7 @@ int aac_get_containers(struct aac_dev *dev)
 	if (status >= 0) {
 		dresp = (struct aac_get_container_count_resp *)fib_data(fibptr);
 		maximum_num_containers = le32_to_cpu(dresp->ContainerSwitchEntries);
-		fib_complete(fibptr);
+		aac_fib_complete(fibptr);
 	}
 
 	if (maximum_num_containers < MAXIMUM_NUM_CONTAINERS)
@@ -280,7 +280,7 @@ int aac_get_containers(struct aac_dev *dev)
 	fsa_dev_ptr = (struct fsa_dev_info *) kmalloc(
 	  sizeof(*fsa_dev_ptr) * maximum_num_containers, GFP_KERNEL);
 	if (!fsa_dev_ptr) {
-		fib_free(fibptr);
+		aac_fib_free(fibptr);
 		return -ENOMEM;
 	}
 	memset(fsa_dev_ptr, 0, sizeof(*fsa_dev_ptr) * maximum_num_containers);
@@ -294,14 +294,14 @@ int aac_get_containers(struct aac_dev *dev)
 
 		fsa_dev_ptr[index].devname[0] = '\0';
 
-		fib_init(fibptr);
+		aac_fib_init(fibptr);
 		dinfo = (struct aac_query_mount *) fib_data(fibptr);
 
 		dinfo->command = cpu_to_le32(VM_NameServe);
 		dinfo->count = cpu_to_le32(index);
 		dinfo->type = cpu_to_le32(FT_FILESYS);
 
-		status = fib_send(ContainerCommand,
+		status = aac_fib_send(ContainerCommand,
 				    fibptr,
 				    sizeof (struct aac_query_mount),
 				    FsaNormal,
@@ -319,7 +319,7 @@ int aac_get_containers(struct aac_dev *dev)
 			dinfo->count = cpu_to_le32(index);
 			dinfo->type = cpu_to_le32(FT_FILESYS);
 
-			if (fib_send(ContainerCommand,
+			if (aac_fib_send(ContainerCommand,
 				    fibptr,
 				    sizeof(struct aac_query_mount),
 				    FsaNormal,
@@ -347,7 +347,7 @@ int aac_get_containers(struct aac_dev *dev)
 			if (le32_to_cpu(dresp->mnt[0].state) & FSCS_READONLY)
 				    fsa_dev_ptr[index].ro = 1;
 		}
-		fib_complete(fibptr);
+		aac_fib_complete(fibptr);
 		/*
 		 *	If there are no more containers, then stop asking.
 		 */
@@ -355,7 +355,7 @@ int aac_get_containers(struct aac_dev *dev)
 			break;
 		}
 	}
-	fib_free(fibptr);
+	aac_fib_free(fibptr);
 	return status;
 }
 
@@ -413,8 +413,8 @@ static void get_container_name_callback(void *context, struct fib * fibptr)
 
 	scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8 | SAM_STAT_GOOD;
 
-	fib_complete(fibptr);
-	fib_free(fibptr);
+	aac_fib_complete(fibptr);
+	aac_fib_free(fibptr);
 	scsicmd->scsi_done(scsicmd);
 }
 
@@ -430,10 +430,10 @@ static int aac_get_container_name(struct scsi_cmnd * scsicmd, int cid)
 
 	dev = (struct aac_dev *)scsicmd->device->host->hostdata;
 
-	if (!(cmd_fibcontext = fib_alloc(dev)))
+	if (!(cmd_fibcontext = aac_fib_alloc(dev)))
 		return -ENOMEM;
 
-	fib_init(cmd_fibcontext);
+	aac_fib_init(cmd_fibcontext);
 	dinfo = (struct aac_get_name *) fib_data(cmd_fibcontext);
 
 	dinfo->command = cpu_to_le32(VM_ContainerConfig);
@@ -441,7 +441,7 @@ static int aac_get_container_name(struct scsi_cmnd * scsicmd, int cid)
 	dinfo->cid = cpu_to_le32(cid);
 	dinfo->count = cpu_to_le32(sizeof(((struct aac_get_name_resp *)NULL)->data));
 
-	status = fib_send(ContainerCommand, 
+	status = aac_fib_send(ContainerCommand,
 		  cmd_fibcontext, 
 		  sizeof (struct aac_get_name),
 		  FsaNormal, 
@@ -455,14 +455,14 @@ static int aac_get_container_name(struct scsi_cmnd * scsicmd, int cid)
 	if (status == -EINPROGRESS) 
 		return 0;
 		
-	printk(KERN_WARNING "aac_get_container_name: fib_send failed with status: %d.\n", status);
-	fib_complete(cmd_fibcontext);
-	fib_free(cmd_fibcontext);
+	printk(KERN_WARNING "aac_get_container_name: aac_fib_send failed with status: %d.\n", status);
+	aac_fib_complete(cmd_fibcontext);
+	aac_fib_free(cmd_fibcontext);
 	return -1;
 }
 
 /**
- *	probe_container		-	query a logical volume
+ *	aac_probe_container		-	query a logical volume
  *	@dev: device to query
  *	@cid: container identifier
  *
@@ -470,7 +470,7 @@ static int aac_get_container_name(struct scsi_cmnd * scsicmd, int cid)
  *	is updated in the struct fsa_dev_info structure rather than returned.
  */
  
-int probe_container(struct aac_dev *dev, int cid)
+int aac_probe_container(struct aac_dev *dev, int cid)
 {
 	struct fsa_dev_info *fsa_dev_ptr;
 	int status;
@@ -482,10 +482,10 @@ int probe_container(struct aac_dev *dev, int cid)
 	fsa_dev_ptr = dev->fsa_dev;
 	instance = dev->scsi_host_ptr->unique_id;
 
-	if (!(fibptr = fib_alloc(dev)))
+	if (!(fibptr = aac_fib_alloc(dev)))
 		return -ENOMEM;
 
-	fib_init(fibptr);
+	aac_fib_init(fibptr);
 
 	dinfo = (struct aac_query_mount *)fib_data(fibptr);
 
@@ -493,14 +493,14 @@ int probe_container(struct aac_dev *dev, int cid)
 	dinfo->count = cpu_to_le32(cid);
 	dinfo->type = cpu_to_le32(FT_FILESYS);
 
-	status = fib_send(ContainerCommand,
+	status = aac_fib_send(ContainerCommand,
 			    fibptr,
 			    sizeof(struct aac_query_mount),
 			    FsaNormal,
 			    1, 1,
 			    NULL, NULL);
 	if (status < 0) {
-		printk(KERN_WARNING "aacraid: probe_container query failed.\n");
+		printk(KERN_WARNING "aacraid: aac_probe_container query failed.\n");
 		goto error;
 	}
 
@@ -512,7 +512,7 @@ int probe_container(struct aac_dev *dev, int cid)
 		dinfo->count = cpu_to_le32(cid);
 		dinfo->type = cpu_to_le32(FT_FILESYS);
 
-		if (fib_send(ContainerCommand,
+		if (aac_fib_send(ContainerCommand,
 			    fibptr,
 			    sizeof(struct aac_query_mount),
 			    FsaNormal,
@@ -535,8 +535,8 @@ int probe_container(struct aac_dev *dev, int cid)
 	}
 
 error:
-	fib_complete(fibptr);
-	fib_free(fibptr);
+	aac_fib_complete(fibptr);
+	aac_fib_free(fibptr);
 
 	return status;
 }
@@ -700,14 +700,14 @@ int aac_get_adapter_info(struct aac_dev* dev)
 	struct aac_bus_info *command;
 	struct aac_bus_info_response *bus_info;
 
-	if (!(fibptr = fib_alloc(dev)))
+	if (!(fibptr = aac_fib_alloc(dev)))
 		return -ENOMEM;
 
-	fib_init(fibptr);
+	aac_fib_init(fibptr);
 	info = (struct aac_adapter_info *) fib_data(fibptr);
 	memset(info,0,sizeof(*info));
 
-	rcode = fib_send(RequestAdapterInfo,
+	rcode = aac_fib_send(RequestAdapterInfo,
 			 fibptr, 
 			 sizeof(*info),
 			 FsaNormal, 
@@ -716,8 +716,8 @@ int aac_get_adapter_info(struct aac_dev* dev)
 			 NULL);
 
 	if (rcode < 0) {
-		fib_complete(fibptr);
-		fib_free(fibptr);
+		aac_fib_complete(fibptr);
+		aac_fib_free(fibptr);
 		return rcode;
 	}
 	memcpy(&dev->adapter_info, info, sizeof(*info));
@@ -725,13 +725,13 @@ int aac_get_adapter_info(struct aac_dev* dev)
 	if (dev->adapter_info.options & AAC_OPT_SUPPLEMENT_ADAPTER_INFO) {
 		struct aac_supplement_adapter_info * info;
 
-		fib_init(fibptr);
+		aac_fib_init(fibptr);
 
 		info = (struct aac_supplement_adapter_info *) fib_data(fibptr);
 
 		memset(info,0,sizeof(*info));
 
-		rcode = fib_send(RequestSupplementAdapterInfo,
+		rcode = aac_fib_send(RequestSupplementAdapterInfo,
 				 fibptr,
 				 sizeof(*info),
 				 FsaNormal,
@@ -748,7 +748,7 @@ int aac_get_adapter_info(struct aac_dev* dev)
 	 * GetBusInfo 
 	 */
 
-	fib_init(fibptr);
+	aac_fib_init(fibptr);
 
 	bus_info = (struct aac_bus_info_response *) fib_data(fibptr);
 
@@ -761,7 +761,7 @@ int aac_get_adapter_info(struct aac_dev* dev)
 	command->MethodId = cpu_to_le32(1);
 	command->CtlCmd = cpu_to_le32(GetBusInfo);
 
-	rcode = fib_send(ContainerCommand,
+	rcode = aac_fib_send(ContainerCommand,
 			 fibptr,
 			 sizeof (*bus_info),
 			 FsaNormal,
@@ -891,8 +891,8 @@ int aac_get_adapter_info(struct aac_dev* dev)
 		}
 	}
 
-	fib_complete(fibptr);
-	fib_free(fibptr);
+	aac_fib_complete(fibptr);
+	aac_fib_free(fibptr);
 
 	return rcode;
 }
@@ -976,8 +976,8 @@ static void io_callback(void *context, struct fib * fibptr)
 		    ? sizeof(scsicmd->sense_buffer)
 		    : sizeof(dev->fsa_dev[cid].sense_data));
 	}
-	fib_complete(fibptr);
-	fib_free(fibptr);
+	aac_fib_complete(fibptr);
+	aac_fib_free(fibptr);
 
 	scsicmd->scsi_done(scsicmd);
 }
@@ -1062,11 +1062,11 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid)
 	/*
 	 *	Alocate and initialize a Fib
 	 */
-	if (!(cmd_fibcontext = fib_alloc(dev))) {
+	if (!(cmd_fibcontext = aac_fib_alloc(dev))) {
 		return -1;
 	}
 
-	fib_init(cmd_fibcontext);
+	aac_fib_init(cmd_fibcontext);
 
 	if (dev->raw_io_interface) {
 		struct aac_raw_io *readcmd;
@@ -1086,7 +1086,7 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid)
 		/*
 		 *	Now send the Fib to the adapter
 		 */
-		status = fib_send(ContainerRawIo,
+		status = aac_fib_send(ContainerRawIo,
 			  cmd_fibcontext, 
 			  fibsize, 
 			  FsaNormal, 
@@ -1112,7 +1112,7 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid)
 		/*
 		 *	Now send the Fib to the adapter
 		 */
-		status = fib_send(ContainerCommand64, 
+		status = aac_fib_send(ContainerCommand64,
 			  cmd_fibcontext, 
 			  fibsize, 
 			  FsaNormal, 
@@ -1136,7 +1136,7 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid)
 		/*
 		 *	Now send the Fib to the adapter
 		 */
-		status = fib_send(ContainerCommand, 
+		status = aac_fib_send(ContainerCommand,
 			  cmd_fibcontext, 
 			  fibsize, 
 			  FsaNormal, 
@@ -1153,14 +1153,14 @@ static int aac_read(struct scsi_cmnd * scsicmd, int cid)
 	if (status == -EINPROGRESS) 
 		return 0;
 		
-	printk(KERN_WARNING "aac_read: fib_send failed with status: %d.\n", status);
+	printk(KERN_WARNING "aac_read: aac_fib_send failed with status: %d.\n", status);
 	/*
 	 *	For some reason, the Fib didn't queue, return QUEUE_FULL
 	 */
 	scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8 | SAM_STAT_TASK_SET_FULL;
 	scsicmd->scsi_done(scsicmd);
-	fib_complete(cmd_fibcontext);
-	fib_free(cmd_fibcontext);
+	aac_fib_complete(cmd_fibcontext);
+	aac_fib_free(cmd_fibcontext);
 	return 0;
 }
 
@@ -1228,12 +1228,12 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid)
 	/*
 	 *	Allocate and initialize a Fib then setup a BlockWrite command
 	 */
-	if (!(cmd_fibcontext = fib_alloc(dev))) {
+	if (!(cmd_fibcontext = aac_fib_alloc(dev))) {
 		scsicmd->result = DID_ERROR << 16;
 		scsicmd->scsi_done(scsicmd);
 		return 0;
 	}
-	fib_init(cmd_fibcontext);
+	aac_fib_init(cmd_fibcontext);
 
 	if (dev->raw_io_interface) {
 		struct aac_raw_io *writecmd;
@@ -1253,7 +1253,7 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid)
 		/*
 		 *	Now send the Fib to the adapter
 		 */
-		status = fib_send(ContainerRawIo,
+		status = aac_fib_send(ContainerRawIo,
 			  cmd_fibcontext, 
 			  fibsize, 
 			  FsaNormal, 
@@ -1279,7 +1279,7 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid)
 		/*
 		 *	Now send the Fib to the adapter
 		 */
-		status = fib_send(ContainerCommand64, 
+		status = aac_fib_send(ContainerCommand64,
 			  cmd_fibcontext, 
 			  fibsize, 
 			  FsaNormal, 
@@ -1305,7 +1305,7 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid)
 		/*
 		 *	Now send the Fib to the adapter
 		 */
-		status = fib_send(ContainerCommand, 
+		status = aac_fib_send(ContainerCommand,
 			  cmd_fibcontext, 
 			  fibsize, 
 			  FsaNormal, 
@@ -1322,15 +1322,15 @@ static int aac_write(struct scsi_cmnd * scsicmd, int cid)
 		return 0;
 	}
 
-	printk(KERN_WARNING "aac_write: fib_send failed with status: %d\n", status);
+	printk(KERN_WARNING "aac_write: aac_fib_send failed with status: %d\n", status);
 	/*
 	 *	For some reason, the Fib didn't queue, return QUEUE_FULL
 	 */
 	scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8 | SAM_STAT_TASK_SET_FULL;
 	scsicmd->scsi_done(scsicmd);
 
-	fib_complete(cmd_fibcontext);
-	fib_free(cmd_fibcontext);
+	aac_fib_complete(cmd_fibcontext);
+	aac_fib_free(cmd_fibcontext);
 	return 0;
 }
 
@@ -1369,8 +1369,8 @@ static void synchronize_callback(void *context, struct fib *fibptr)
 			  sizeof(cmd->sense_buffer)));
 	}
 
-	fib_complete(fibptr);
-	fib_free(fibptr);
+	aac_fib_complete(fibptr);
+	aac_fib_free(fibptr);
 	cmd->scsi_done(cmd);
 }
 
@@ -1407,10 +1407,10 @@ static int aac_synchronize(struct scsi_cmnd *scsicmd, int cid)
 	 *	Allocate and initialize a Fib
 	 */
 	if (!(cmd_fibcontext = 
-	    fib_alloc((struct aac_dev *)scsicmd->device->host->hostdata))) 
+	    aac_fib_alloc((struct aac_dev *)scsicmd->device->host->hostdata)))
 		return SCSI_MLQUEUE_HOST_BUSY;
 
-	fib_init(cmd_fibcontext);
+	aac_fib_init(cmd_fibcontext);
 
 	synchronizecmd = fib_data(cmd_fibcontext);
 	synchronizecmd->command = cpu_to_le32(VM_ContainerConfig);
@@ -1422,7 +1422,7 @@ static int aac_synchronize(struct scsi_cmnd *scsicmd, int cid)
 	/*
 	 *	Now send the Fib to the adapter
 	 */
-	status = fib_send(ContainerCommand,
+	status = aac_fib_send(ContainerCommand,
 		  cmd_fibcontext,
 		  sizeof(struct aac_synchronize),
 		  FsaNormal,
@@ -1437,9 +1437,9 @@ static int aac_synchronize(struct scsi_cmnd *scsicmd, int cid)
 		return 0;
 
 	printk(KERN_WARNING 
-		"aac_synchronize: fib_send failed with status: %d.\n", status);
-	fib_complete(cmd_fibcontext);
-	fib_free(cmd_fibcontext);
+		"aac_synchronize: aac_fib_send failed with status: %d.\n", status);
+	aac_fib_complete(cmd_fibcontext);
+	aac_fib_free(cmd_fibcontext);
 	return SCSI_MLQUEUE_HOST_BUSY;
 }
 
@@ -1488,7 +1488,7 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd)
 				case READ_CAPACITY:
 				case TEST_UNIT_READY:
 					spin_unlock_irq(host->host_lock);
-					probe_container(dev, cid);
+					aac_probe_container(dev, cid);
 					if ((fsa_dev_ptr[cid].valid & 1) == 0)
 						fsa_dev_ptr[cid].valid = 0;
 					spin_lock_irq(host->host_lock);
@@ -2089,8 +2089,8 @@ static void aac_srb_callback(void *context, struct fib * fibptr)
 	 */
 	scsicmd->result |= le32_to_cpu(srbreply->scsi_status);
 
-	fib_complete(fibptr);
-	fib_free(fibptr);
+	aac_fib_complete(fibptr);
+	aac_fib_free(fibptr);
 	scsicmd->scsi_done(scsicmd);
 }
 
@@ -2142,10 +2142,10 @@ static int aac_send_srb_fib(struct scsi_cmnd* scsicmd)
 	/*
 	 *	Allocate and initialize a Fib then setup a BlockWrite command
 	 */
-	if (!(cmd_fibcontext = fib_alloc(dev))) {
+	if (!(cmd_fibcontext = aac_fib_alloc(dev))) {
 		return -1;
 	}
-	fib_init(cmd_fibcontext);
+	aac_fib_init(cmd_fibcontext);
 
 	srbcmd = (struct aac_srb*) fib_data(cmd_fibcontext);
 	srbcmd->function = cpu_to_le32(SRBF_ExecuteScsi);
@@ -2179,7 +2179,7 @@ static int aac_send_srb_fib(struct scsi_cmnd* scsicmd)
 		/*
 		 *	Now send the Fib to the adapter
 		 */
-		status = fib_send(ScsiPortCommand64, cmd_fibcontext, 
+		status = aac_fib_send(ScsiPortCommand64, cmd_fibcontext,
 				fibsize, FsaNormal, 0, 1,
 				  (fib_callback) aac_srb_callback, 
 				  (void *) scsicmd);
@@ -2201,7 +2201,7 @@ static int aac_send_srb_fib(struct scsi_cmnd* scsicmd)
 		/*
 		 *	Now send the Fib to the adapter
 		 */
-		status = fib_send(ScsiPortCommand, cmd_fibcontext, fibsize, FsaNormal, 0, 1,
+		status = aac_fib_send(ScsiPortCommand, cmd_fibcontext, fibsize, FsaNormal, 0, 1,
 				  (fib_callback) aac_srb_callback, (void *) scsicmd);
 	}
 	/*
@@ -2211,9 +2211,9 @@ static int aac_send_srb_fib(struct scsi_cmnd* scsicmd)
 		return 0;
 	}
 
-	printk(KERN_WARNING "aac_srb: fib_send failed with status: %d\n", status);
-	fib_complete(cmd_fibcontext);
-	fib_free(cmd_fibcontext);
+	printk(KERN_WARNING "aac_srb: aac_fib_send failed with status: %d\n", status);
+	aac_fib_complete(cmd_fibcontext);
+	aac_fib_free(cmd_fibcontext);
 
 	return -1;
 }
diff --git a/drivers/scsi/aacraid/aacraid.h b/drivers/scsi/aacraid/aacraid.h
index 66dbb6d2c506..2d430b7e8cf4 100644
--- a/drivers/scsi/aacraid/aacraid.h
+++ b/drivers/scsi/aacraid/aacraid.h
@@ -1774,16 +1774,16 @@ static inline u32 cap_to_cyls(sector_t capacity, u32 divisor)
 struct scsi_cmnd;
 
 const char *aac_driverinfo(struct Scsi_Host *);
-struct fib *fib_alloc(struct aac_dev *dev);
-int fib_setup(struct aac_dev *dev);
-void fib_map_free(struct aac_dev *dev);
-void fib_free(struct fib * context);
-void fib_init(struct fib * context);
+struct fib *aac_fib_alloc(struct aac_dev *dev);
+int aac_fib_setup(struct aac_dev *dev);
+void aac_fib_map_free(struct aac_dev *dev);
+void aac_fib_free(struct fib * context);
+void aac_fib_init(struct fib * context);
 void aac_printf(struct aac_dev *dev, u32 val);
-int fib_send(u16 command, struct fib * context, unsigned long size, int priority, int wait, int reply, fib_callback callback, void *ctxt);
+int aac_fib_send(u16 command, struct fib * context, unsigned long size, int priority, int wait, int reply, fib_callback callback, void *ctxt);
 int aac_consumer_get(struct aac_dev * dev, struct aac_queue * q, struct aac_entry **entry);
 void aac_consumer_free(struct aac_dev * dev, struct aac_queue * q, u32 qnum);
-int fib_complete(struct fib * context);
+int aac_fib_complete(struct fib * context);
 #define fib_data(fibctx) ((void *)(fibctx)->hw_fib->data)
 struct aac_dev *aac_init_adapter(struct aac_dev *dev);
 int aac_get_config_status(struct aac_dev *dev);
@@ -1799,11 +1799,11 @@ unsigned int aac_command_normal(struct aac_queue * q);
 unsigned int aac_intr_normal(struct aac_dev * dev, u32 Index);
 int aac_command_thread(struct aac_dev * dev);
 int aac_close_fib_context(struct aac_dev * dev, struct aac_fib_context *fibctx);
-int fib_adapter_complete(struct fib * fibptr, unsigned short size);
+int aac_fib_adapter_complete(struct fib * fibptr, unsigned short size);
 struct aac_driver_ident* aac_get_driver_ident(int devtype);
 int aac_get_adapter_info(struct aac_dev* dev);
 int aac_send_shutdown(struct aac_dev *dev);
-int probe_container(struct aac_dev *dev, int cid);
+int aac_probe_container(struct aac_dev *dev, int cid);
 extern int numacb;
 extern int acbsize;
 extern char aac_driver_version[];
diff --git a/drivers/scsi/aacraid/commctrl.c b/drivers/scsi/aacraid/commctrl.c
index 4fe79cd7c957..47fefca72695 100644
--- a/drivers/scsi/aacraid/commctrl.c
+++ b/drivers/scsi/aacraid/commctrl.c
@@ -63,7 +63,7 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 	unsigned size;
 	int retval;
 
-	fibptr = fib_alloc(dev);
+	fibptr = aac_fib_alloc(dev);
 	if(fibptr == NULL) {
 		return -ENOMEM;
 	}
@@ -73,7 +73,7 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 	 *	First copy in the header so that we can check the size field.
 	 */
 	if (copy_from_user((void *)kfib, arg, sizeof(struct aac_fibhdr))) {
-		fib_free(fibptr);
+		aac_fib_free(fibptr);
 		return -EFAULT;
 	}
 	/*
@@ -110,13 +110,13 @@ static int ioctl_send_fib(struct aac_dev * dev, void __user *arg)
 		 */
 		kfib->header.XferState = 0;
 	} else {
-		retval = fib_send(le16_to_cpu(kfib->header.Command), fibptr,
+		retval = aac_fib_send(le16_to_cpu(kfib->header.Command), fibptr,
 				le16_to_cpu(kfib->header.Size) , FsaNormal,
 				1, 1, NULL, NULL);
 		if (retval) {
 			goto cleanup;
 		}
-		if (fib_complete(fibptr) != 0) {
+		if (aac_fib_complete(fibptr) != 0) {
 			retval = -EINVAL;
 			goto cleanup;
 		}
@@ -138,7 +138,7 @@ cleanup:
 		fibptr->hw_fib_pa = hw_fib_pa;
 		fibptr->hw_fib = hw_fib;
 	}
-	fib_free(fibptr);
+	aac_fib_free(fibptr);
 	return retval;
 }
 
@@ -464,10 +464,10 @@ static int aac_send_raw_srb(struct aac_dev* dev, void __user * arg)
 	/*
 	 *	Allocate and initialize a Fib then setup a BlockWrite command
 	 */
-	if (!(srbfib = fib_alloc(dev))) {
+	if (!(srbfib = aac_fib_alloc(dev))) {
 		return -ENOMEM;
 	}
-	fib_init(srbfib);
+	aac_fib_init(srbfib);
 
 	srbcmd = (struct aac_srb*) fib_data(srbfib);
 
@@ -601,7 +601,7 @@ static int aac_send_raw_srb(struct aac_dev* dev, void __user * arg)
 
 		srbcmd->count = cpu_to_le32(byte_count);
 		psg->count = cpu_to_le32(sg_indx+1);
-		status = fib_send(ScsiPortCommand64, srbfib, actual_fibsize, FsaNormal, 1, 1,NULL,NULL);
+		status = aac_fib_send(ScsiPortCommand64, srbfib, actual_fibsize, FsaNormal, 1, 1,NULL,NULL);
 	} else {
 		struct user_sgmap* upsg = &user_srbcmd->sg;
 		struct sgmap* psg = &srbcmd->sg;
@@ -649,7 +649,7 @@ static int aac_send_raw_srb(struct aac_dev* dev, void __user * arg)
 		}
 		srbcmd->count = cpu_to_le32(byte_count);
 		psg->count = cpu_to_le32(sg_indx+1);
-		status = fib_send(ScsiPortCommand, srbfib, actual_fibsize, FsaNormal, 1, 1, NULL, NULL);
+		status = aac_fib_send(ScsiPortCommand, srbfib, actual_fibsize, FsaNormal, 1, 1, NULL, NULL);
 	}
 
 	if (status != 0){
@@ -684,8 +684,8 @@ cleanup:
 	for(i=0; i <= sg_indx; i++){
 		kfree(sg_list[i]);
 	}
-	fib_complete(srbfib);
-	fib_free(srbfib);
+	aac_fib_complete(srbfib);
+	aac_fib_free(srbfib);
 
 	return rcode;
 }
diff --git a/drivers/scsi/aacraid/comminit.c b/drivers/scsi/aacraid/comminit.c
index 82821d331c07..1628d094943d 100644
--- a/drivers/scsi/aacraid/comminit.c
+++ b/drivers/scsi/aacraid/comminit.c
@@ -185,17 +185,17 @@ int aac_send_shutdown(struct aac_dev * dev)
 	struct aac_close *cmd;
 	int status;
 
-	fibctx = fib_alloc(dev);
+	fibctx = aac_fib_alloc(dev);
 	if (!fibctx)
 		return -ENOMEM;
-	fib_init(fibctx);
+	aac_fib_init(fibctx);
 
 	cmd = (struct aac_close *) fib_data(fibctx);
 
 	cmd->command = cpu_to_le32(VM_CloseAll);
 	cmd->cid = cpu_to_le32(0xffffffff);
 
-	status = fib_send(ContainerCommand,
+	status = aac_fib_send(ContainerCommand,
 			  fibctx,
 			  sizeof(struct aac_close),
 			  FsaNormal,
@@ -203,8 +203,8 @@ int aac_send_shutdown(struct aac_dev * dev)
 			  NULL, NULL);
 
 	if (status == 0)
-		fib_complete(fibctx);
-	fib_free(fibctx);
+		aac_fib_complete(fibctx);
+	aac_fib_free(fibctx);
 	return status;
 }
 
@@ -427,7 +427,7 @@ struct aac_dev *aac_init_adapter(struct aac_dev *dev)
 	/*
 	 *	Initialize the list of fibs
 	 */
-	if(fib_setup(dev)<0){
+	if (aac_fib_setup(dev) < 0) {
 		kfree(dev->queues);
 		return NULL;
 	}
diff --git a/drivers/scsi/aacraid/commsup.c b/drivers/scsi/aacraid/commsup.c
index 38d6d00fb0fc..97354ba7b185 100644
--- a/drivers/scsi/aacraid/commsup.c
+++ b/drivers/scsi/aacraid/commsup.c
@@ -67,27 +67,27 @@ static int fib_map_alloc(struct aac_dev *dev)
 }
 
 /**
- *	fib_map_free		-	free the fib objects
+ *	aac_fib_map_free		-	free the fib objects
  *	@dev: Adapter to free
  *
  *	Free the PCI mappings and the memory allocated for FIB blocks
  *	on this adapter.
  */
 
-void fib_map_free(struct aac_dev *dev)
+void aac_fib_map_free(struct aac_dev *dev)
 {
 	pci_free_consistent(dev->pdev, dev->max_fib_size * (dev->scsi_host_ptr->can_queue + AAC_NUM_MGT_FIB), dev->hw_fib_va, dev->hw_fib_pa);
 }
 
 /**
- *	fib_setup	-	setup the fibs
+ *	aac_fib_setup	-	setup the fibs
  *	@dev: Adapter to set up
  *
  *	Allocate the PCI space for the fibs, map it and then intialise the
  *	fib area, the unmapped fib data and also the free list
  */
 
-int fib_setup(struct aac_dev * dev)
+int aac_fib_setup(struct aac_dev * dev)
 {
 	struct fib *fibptr;
 	struct hw_fib *hw_fib_va;
@@ -134,14 +134,14 @@ int fib_setup(struct aac_dev * dev)
 }
 
 /**
- *	fib_alloc	-	allocate a fib
+ *	aac_fib_alloc	-	allocate a fib
  *	@dev: Adapter to allocate the fib for
  *
  *	Allocate a fib from the adapter fib pool. If the pool is empty we
  *	return NULL.
  */
  
-struct fib * fib_alloc(struct aac_dev *dev)
+struct fib *aac_fib_alloc(struct aac_dev *dev)
 {
 	struct fib * fibptr;
 	unsigned long flags;
@@ -170,14 +170,14 @@ struct fib * fib_alloc(struct aac_dev *dev)
 }
 
 /**
- *	fib_free	-	free a fib
+ *	aac_fib_free	-	free a fib
  *	@fibptr: fib to free up
  *
  *	Frees up a fib and places it on the appropriate queue
  *	(either free or timed out)
  */
  
-void fib_free(struct fib * fibptr)
+void aac_fib_free(struct fib *fibptr)
 {
 	unsigned long flags;
 
@@ -188,7 +188,7 @@ void fib_free(struct fib * fibptr)
 		fibptr->dev->timeout_fib = fibptr;
 	} else {
 		if (fibptr->hw_fib->header.XferState != 0) {
-			printk(KERN_WARNING "fib_free, XferState != 0, fibptr = 0x%p, XferState = 0x%x\n", 
+			printk(KERN_WARNING "aac_fib_free, XferState != 0, fibptr = 0x%p, XferState = 0x%x\n",
 				 (void*)fibptr, 
 				 le32_to_cpu(fibptr->hw_fib->header.XferState));
 		}
@@ -199,13 +199,13 @@ void fib_free(struct fib * fibptr)
 }
 
 /**
- *	fib_init	-	initialise a fib
+ *	aac_fib_init	-	initialise a fib
  *	@fibptr: The fib to initialize
  *	
  *	Set up the generic fib fields ready for use
  */
  
-void fib_init(struct fib *fibptr)
+void aac_fib_init(struct fib *fibptr)
 {
 	struct hw_fib *hw_fib = fibptr->hw_fib;
 
@@ -362,7 +362,7 @@ static int aac_queue_get(struct aac_dev * dev, u32 * index, u32 qid, struct hw_f
  */
 
 /**
- *	fib_send	-	send a fib to the adapter
+ *	aac_fib_send	-	send a fib to the adapter
  *	@command: Command to send
  *	@fibptr: The fib
  *	@size: Size of fib data area
@@ -378,7 +378,9 @@ static int aac_queue_get(struct aac_dev * dev, u32 * index, u32 qid, struct hw_f
  *	response FIB is received from the adapter.
  */
  
-int fib_send(u16 command, struct fib * fibptr, unsigned long size,  int priority, int wait, int reply, fib_callback callback, void * callback_data)
+int aac_fib_send(u16 command, struct fib *fibptr, unsigned long size,
+		int priority, int wait, int reply, fib_callback callback,
+		void *callback_data)
 {
 	struct aac_dev * dev = fibptr->dev;
 	struct hw_fib * hw_fib = fibptr->hw_fib;
@@ -493,7 +495,7 @@ int fib_send(u16 command, struct fib * fibptr, unsigned long size,  int priority
 		q->numpending++;
 		*(q->headers.producer) = cpu_to_le32(index + 1);
 		spin_unlock_irqrestore(q->lock, qflags);
-		dprintk((KERN_DEBUG "fib_send: inserting a queue entry at index %d.\n",index));
+		dprintk((KERN_DEBUG "aac_fib_send: inserting a queue entry at index %d.\n",index));
 		if (!(nointr & aac_config.irq_mod))
 			aac_adapter_notify(dev, AdapNormCmdQueue);
 	}
@@ -520,7 +522,7 @@ int fib_send(u16 command, struct fib * fibptr, unsigned long size,  int priority
 					list_del(&fibptr->queue);
 					spin_unlock_irqrestore(q->lock, qflags);
 					if (wait == -1) {
-	        				printk(KERN_ERR "aacraid: fib_send: first asynchronous command timed out.\n"
+	        				printk(KERN_ERR "aacraid: aac_fib_send: first asynchronous command timed out.\n"
 						  "Usually a result of a PCI interrupt routing problem;\n"
 						  "update mother board BIOS or consider utilizing one of\n"
 						  "the SAFE mode kernel options (acpi, apic etc)\n");
@@ -624,7 +626,7 @@ void aac_consumer_free(struct aac_dev * dev, struct aac_queue *q, u32 qid)
 }        
 
 /**
- *	fib_adapter_complete	-	complete adapter issued fib
+ *	aac_fib_adapter_complete	-	complete adapter issued fib
  *	@fibptr: fib to complete
  *	@size: size of fib
  *
@@ -632,7 +634,7 @@ void aac_consumer_free(struct aac_dev * dev, struct aac_queue *q, u32 qid)
  *	the adapter.
  */
 
-int fib_adapter_complete(struct fib * fibptr, unsigned short size)
+int aac_fib_adapter_complete(struct fib *fibptr, unsigned short size)
 {
 	struct hw_fib * hw_fib = fibptr->hw_fib;
 	struct aac_dev * dev = fibptr->dev;
@@ -683,20 +685,20 @@ int fib_adapter_complete(struct fib * fibptr, unsigned short size)
 	}
 	else 
 	{
-        	printk(KERN_WARNING "fib_adapter_complete: Unknown xferstate detected.\n");
+        	printk(KERN_WARNING "aac_fib_adapter_complete: Unknown xferstate detected.\n");
         	BUG();
 	}   
 	return 0;
 }
 
 /**
- *	fib_complete	-	fib completion handler
+ *	aac_fib_complete	-	fib completion handler
  *	@fib: FIB to complete
  *
  *	Will do all necessary work to complete a FIB.
  */
  
-int fib_complete(struct fib * fibptr)
+int aac_fib_complete(struct fib *fibptr)
 {
 	struct hw_fib * hw_fib = fibptr->hw_fib;
 
@@ -995,14 +997,14 @@ static void aac_handle_aif(struct aac_dev * dev, struct fib * fibptr)
 	if (!dev || !dev->scsi_host_ptr)
 		return;
 	/*
-	 * force reload of disk info via probe_container
+	 * force reload of disk info via aac_probe_container
 	 */
 	if ((device_config_needed == CHANGE)
 	 && (dev->fsa_dev[container].valid == 1))
 		dev->fsa_dev[container].valid = 2;
 	if ((device_config_needed == CHANGE) ||
 			(device_config_needed == ADD))
-		probe_container(dev, container);
+		aac_probe_container(dev, container);
 	device = scsi_device_lookup(dev->scsi_host_ptr, 
 		CONTAINER_TO_CHANNEL(container), 
 		CONTAINER_TO_ID(container), 
@@ -1104,7 +1106,7 @@ int aac_command_thread(struct aac_dev * dev)
 				/* Handle Driver Notify Events */
 				aac_handle_aif(dev, fib);
 				*(__le32 *)hw_fib->data = cpu_to_le32(ST_OK);
-				fib_adapter_complete(fib, (u16)sizeof(u32));
+				aac_fib_adapter_complete(fib, (u16)sizeof(u32));
 			} else {
 				struct list_head *entry;
 				/* The u32 here is important and intended. We are using
@@ -1241,7 +1243,7 @@ int aac_command_thread(struct aac_dev * dev)
 				 *	Set the status of this FIB
 				 */
 				*(__le32 *)hw_fib->data = cpu_to_le32(ST_OK);
-				fib_adapter_complete(fib, sizeof(u32));
+				aac_fib_adapter_complete(fib, sizeof(u32));
 				spin_unlock_irqrestore(&dev->fib_lock, flagv);
 				/* Free up the remaining resources */
 				hw_fib_p = hw_fib_pool;
diff --git a/drivers/scsi/aacraid/dpcsup.c b/drivers/scsi/aacraid/dpcsup.c
index 439948ef8251..f6bcb9486f85 100644
--- a/drivers/scsi/aacraid/dpcsup.c
+++ b/drivers/scsi/aacraid/dpcsup.c
@@ -206,7 +206,7 @@ unsigned int aac_command_normal(struct aac_queue *q)
 			 *	Set the status of this FIB
 			 */
 			*(__le32 *)hw_fib->data = cpu_to_le32(ST_OK);
-			fib_adapter_complete(fib, sizeof(u32));
+			aac_fib_adapter_complete(fib, sizeof(u32));
 			spin_lock_irqsave(q->lock, flags);
 		}		
 	}
diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c
index 45fc171e5f36..9defee03b823 100644
--- a/drivers/scsi/aacraid/linit.c
+++ b/drivers/scsi/aacraid/linit.c
@@ -917,7 +917,7 @@ static int __devinit aac_probe_one(struct pci_dev *pdev,
 	aac_adapter_disable_int(aac);
 	free_irq(pdev->irq, aac);
  out_unmap:
-	fib_map_free(aac);
+	aac_fib_map_free(aac);
 	pci_free_consistent(aac->pdev, aac->comm_size, aac->comm_addr, aac->comm_phys);
 	kfree(aac->queues);
 	iounmap(aac->regs.sa);
@@ -951,7 +951,7 @@ static void __devexit aac_remove_one(struct pci_dev *pdev)
 
 	aac_send_shutdown(aac);
 	aac_adapter_disable_int(aac);
-	fib_map_free(aac);
+	aac_fib_map_free(aac);
 	pci_free_consistent(aac->pdev, aac->comm_size, aac->comm_addr,
 			aac->comm_phys);
 	kfree(aac->queues);

From bb08f92ebd75704e07d69bb9d8ee234d1a500b98 Mon Sep 17 00:00:00 2001
From: Mark Haverkamp <markh@osdl.org>
Date: Wed, 1 Feb 2006 09:30:44 -0800
Subject: [PATCH 015/509] [SCSI] aacraid: use no_uld_attach flag

Received From Mark Salyzyn.

In order to support user tools accessing the array components (SMART,
Mode Page information, Cache page adjustments, WWN determination,
Firmware updates etc), we take advantage of the no_uld_attach flag and
deprecate the code that filters Inquiries to block the requests to array
components. The quirk prevents the sd layer from attaching to the
components.

We also took the opportunity to balance the queue depths based on the
total adapter queue depth to the array devices to reduce the chances of
starvation.

Signed-off-by: Mark Haverkamp <markh@osdl.org>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/aacraid/aachba.c | 53 ++---------------------------------
 drivers/scsi/aacraid/linit.c  | 42 +++++++++++++++++++++------
 2 files changed, 36 insertions(+), 59 deletions(-)

diff --git a/drivers/scsi/aacraid/aachba.c b/drivers/scsi/aacraid/aachba.c
index b0f314e415c9..a16f8ded8f1d 100644
--- a/drivers/scsi/aacraid/aachba.c
+++ b/drivers/scsi/aacraid/aachba.c
@@ -1465,7 +1465,7 @@ int aac_scsi_cmd(struct scsi_cmnd * scsicmd)
 	 *	itself.
 	 */
 	if (scmd_id(scsicmd) != host->this_id) {
-		if ((scsicmd->device->channel == 0) ){
+		if ((scsicmd->device->channel == CONTAINER_CHANNEL)) {
 			if( (scsicmd->device->id >= dev->maximum_num_containers) || (scsicmd->device->lun != 0)){ 
 				scsicmd->result = DID_NO_CONNECT << 16;
 				scsicmd->scsi_done(scsicmd);
@@ -1935,33 +1935,7 @@ static void aac_srb_callback(void *context, struct fib * fibptr)
 	case SRB_STATUS_ERROR_RECOVERY:
 	case SRB_STATUS_PENDING:
 	case SRB_STATUS_SUCCESS:
-		if(scsicmd->cmnd[0] == INQUIRY ){
-			u8 b;
-			u8 b1;
-			/* We can't expose disk devices because we can't tell whether they
-			 * are the raw container drives or stand alone drives.  If they have
-			 * the removable bit set then we should expose them though.
-			 */
-			b = (*(u8*)scsicmd->buffer)&0x1f;
-			b1 = ((u8*)scsicmd->buffer)[1];
-			if( b==TYPE_TAPE || b==TYPE_WORM || b==TYPE_ROM || b==TYPE_MOD|| b==TYPE_MEDIUM_CHANGER 
-					|| (b==TYPE_DISK && (b1&0x80)) ){
-				scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8;
-			/*
-			 * We will allow disk devices if in RAID/SCSI mode and
-			 * the channel is 2
-			 */
-			} else if ((dev->raid_scsi_mode) &&
-					(scmd_channel(scsicmd) == 2)) {
-				scsicmd->result = DID_OK << 16 | 
-						COMMAND_COMPLETE << 8;
-			} else {
-				scsicmd->result = DID_NO_CONNECT << 16 | 
-						COMMAND_COMPLETE << 8;
-			}
-		} else {
-			scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8;
-		}
+		scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8;
 		break;
 	case SRB_STATUS_DATA_OVERRUN:
 		switch(scsicmd->cmnd[0]){
@@ -1981,28 +1955,7 @@ static void aac_srb_callback(void *context, struct fib * fibptr)
 			scsicmd->result = DID_ERROR << 16 | COMMAND_COMPLETE << 8;
 			break;
 		case INQUIRY: {
-			u8 b;
-			u8 b1;
-			/* We can't expose disk devices because we can't tell whether they
-			* are the raw container drives or stand alone drives
-			*/
-			b = (*(u8*)scsicmd->buffer)&0x0f;
-			b1 = ((u8*)scsicmd->buffer)[1];
-			if( b==TYPE_TAPE || b==TYPE_WORM || b==TYPE_ROM || b==TYPE_MOD|| b==TYPE_MEDIUM_CHANGER
-					|| (b==TYPE_DISK && (b1&0x80)) ){
-				scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8;
-			/*
-			 * We will allow disk devices if in RAID/SCSI mode and
-			 * the channel is 2
-			 */
-			} else if ((dev->raid_scsi_mode) &&
-					(scmd_channel(scsicmd) == 2)) {
-				scsicmd->result = DID_OK << 16 | 
-						COMMAND_COMPLETE << 8;
-			} else {
-				scsicmd->result = DID_NO_CONNECT << 16 | 
-						COMMAND_COMPLETE << 8;
-			}
+			scsicmd->result = DID_OK << 16 | COMMAND_COMPLETE << 8;
 			break;
 		}
 		default:
diff --git a/drivers/scsi/aacraid/linit.c b/drivers/scsi/aacraid/linit.c
index 9defee03b823..271617890562 100644
--- a/drivers/scsi/aacraid/linit.c
+++ b/drivers/scsi/aacraid/linit.c
@@ -385,21 +385,45 @@ static int aac_biosparm(struct scsi_device *sdev, struct block_device *bdev,
 
 static int aac_slave_configure(struct scsi_device *sdev)
 {
-	struct Scsi_Host *host = sdev->host;
 	if (sdev_channel(sdev) == CONTAINER_CHANNEL) {
 		sdev->skip_ms_page_8 = 1;
 		sdev->skip_ms_page_3f = 1;
 	}
+	if ((sdev->type == TYPE_DISK) &&
+			(sdev_channel(sdev) != CONTAINER_CHANNEL)) {
+		struct aac_dev *aac = (struct aac_dev *)sdev->host->hostdata;
+		if (!aac->raid_scsi_mode || (sdev_channel(sdev) != 2))
+			sdev->no_uld_attach = 1;
+	}
+	if (sdev->tagged_supported && (sdev->type == TYPE_DISK) &&
+			(sdev_channel(sdev) == CONTAINER_CHANNEL)) {
+		struct scsi_device * dev;
+		struct Scsi_Host *host = sdev->host;
+		unsigned num_lsu = 0;
+		unsigned num_one = 0;
+		unsigned depth;
 
-	if (sdev->tagged_supported)
-		scsi_adjust_queue_depth(sdev, MSG_ORDERED_TAG, 128);
-	else
+		__shost_for_each_device(dev, host) {
+			if (dev->tagged_supported && (dev->type == TYPE_DISK) &&
+				(sdev_channel(dev) == CONTAINER_CHANNEL))
+				++num_lsu;
+			else
+				++num_one;
+		}
+		if (num_lsu == 0)
+			++num_lsu;
+		depth = (host->can_queue - num_one) / num_lsu;
+		if (depth > 256)
+			depth = 256;
+		else if (depth < 2)
+			depth = 2;
+		scsi_adjust_queue_depth(sdev, MSG_ORDERED_TAG, depth);
+		if (!(((struct aac_dev *)host->hostdata)->adapter_info.options &
+				AAC_OPT_NEW_COMM))
+			blk_queue_max_segment_size(sdev->request_queue, 65536);
+	} else
 		scsi_adjust_queue_depth(sdev, 0, 1);
 
-	if (!(((struct aac_dev *)host->hostdata)->adapter_info.options
-	  & AAC_OPT_NEW_COMM))
-		blk_queue_max_segment_size(sdev->request_queue, 65536);
-
 	return 0;
 }
 
@@ -874,7 +898,7 @@ static int __devinit aac_probe_one(struct pci_dev *pdev,
  
  	/*
 	 * max channel will be the physical channels plus 1 virtual channel
-	 * all containers are on the virtual channel 0
+	 * all containers are on the virtual channel 0 (CONTAINER_CHANNEL)
 	 * physical channels are address by their actual physical number+1
 	 */
 	if (aac->nondasd_support == 1)

From 7b7232f3fb5ecd7c30cb52df368070cc5f5ca614 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:06:49 -0600
Subject: [PATCH 016/509] [SCSI] iscsi update: cleanup iscsi class interface

From:
michaelc@cs.wisc.edu
fujita.tomonori@lab.ntt.co.jp
da-x@monatomic.org

and err path fixup from:
ogerlitz@voltaire.com

This patch cleans up that interface by having the lld and class
pass a iscsi_cls_session or iscsi_cls_conn between each other when
the function is used by HW and SW iscsi llds. This way the lld
does not have to remember if it has to send a handle or pointer
and a handle or pointer to connection, session or host.

This also has the class verify the session handle that gets passed from
userspace instead of using the pointer passed into the kernel directly.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/iscsi_tcp.c            |  70 +++++----
 drivers/scsi/iscsi_tcp.h            |   3 +
 drivers/scsi/scsi_transport_iscsi.c | 221 ++++++++++++++--------------
 include/scsi/iscsi_if.h             |   3 -
 include/scsi/scsi_transport_iscsi.h |  34 +++--
 5 files changed, 172 insertions(+), 159 deletions(-)

diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 780bfcc67096..d07d309ac026 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -146,7 +146,7 @@ iscsi_conn_failure(struct iscsi_conn *conn, enum iscsi_err err)
 	spin_unlock_irqrestore(&session->lock, flags);
 	set_bit(SUSPEND_BIT, &conn->suspend_tx);
 	set_bit(SUSPEND_BIT, &conn->suspend_rx);
-	iscsi_conn_error(iscsi_handle(conn), err);
+	iscsi_conn_error(conn->cls_conn, err);
 }
 
 static inline int
@@ -689,7 +689,7 @@ iscsi_hdr_recv(struct iscsi_conn *conn)
 				break;
 
 			if (!conn->in.datalen) {
-				rc = iscsi_recv_pdu(iscsi_handle(conn), hdr,
+				rc = iscsi_recv_pdu(conn->cls_conn, hdr,
 						    NULL, 0);
 				if (conn->login_mtask != mtask) {
 					spin_lock(&session->lock);
@@ -737,7 +737,7 @@ iscsi_hdr_recv(struct iscsi_conn *conn)
 			if (!conn->in.datalen) {
 				struct iscsi_mgmt_task *mtask;
 
-				rc = iscsi_recv_pdu(iscsi_handle(conn), hdr,
+				rc = iscsi_recv_pdu(conn->cls_conn, hdr,
 						    NULL, 0);
 				mtask = (struct iscsi_mgmt_task *)
 					session->mgmt_cmds[conn->in.itt -
@@ -761,7 +761,7 @@ iscsi_hdr_recv(struct iscsi_conn *conn)
 				rc = iscsi_check_assign_cmdsn(session,
 						 (struct iscsi_nopin*)hdr);
 				if (!rc && hdr->ttt != ISCSI_RESERVED_TAG)
-					rc = iscsi_recv_pdu(iscsi_handle(conn),
+					rc = iscsi_recv_pdu(conn->cls_conn,
 							    hdr, NULL, 0);
 			} else
 				rc = ISCSI_ERR_PROTO;
@@ -1044,7 +1044,7 @@ iscsi_data_recv(struct iscsi_conn *conn)
 			goto exit;
 		}
 
-		rc = iscsi_recv_pdu(iscsi_handle(conn), conn->in.hdr,
+		rc = iscsi_recv_pdu(conn->cls_conn, conn->in.hdr,
 				    conn->data, conn->in.datalen);
 
 		if (!rc && conn->datadgst_en &&
@@ -2428,19 +2428,20 @@ iscsi_pool_free(struct iscsi_queue *q, void **items)
 }
 
 static struct iscsi_cls_conn *
-iscsi_conn_create(struct Scsi_Host *shost, uint32_t conn_idx)
+iscsi_conn_create(struct iscsi_cls_session *cls_session, uint32_t conn_idx)
 {
+	struct Scsi_Host *shost = iscsi_session_to_shost(cls_session);
 	struct iscsi_session *session = iscsi_hostdata(shost->hostdata);
 	struct iscsi_conn *conn;
 	struct iscsi_cls_conn *cls_conn;
 
-	cls_conn = iscsi_create_conn(hostdata_session(shost->hostdata),
-				     conn_idx);
+	cls_conn = iscsi_create_conn(cls_session, conn_idx);
 	if (!cls_conn)
 		return NULL;
 	conn = cls_conn->dd_data;
+	memset(conn, 0, sizeof(*conn));
 
-	memset(conn, 0, sizeof(struct iscsi_conn));
+	conn->cls_conn = cls_conn;
 	conn->c_stage = ISCSI_CONN_INITIAL_STAGE;
 	conn->in_progress = IN_PROGRESS_WAIT_HEADER;
 	conn->id = conn_idx;
@@ -2625,11 +2626,13 @@ iscsi_conn_destroy(struct iscsi_cls_conn *cls_conn)
 }
 
 static int
-iscsi_conn_bind(iscsi_sessionh_t sessionh, iscsi_connh_t connh,
-		uint32_t transport_fd, int is_leading)
+iscsi_conn_bind(struct iscsi_cls_session *cls_session,
+		struct iscsi_cls_conn *cls_conn, uint32_t transport_fd,
+		int is_leading)
 {
-	struct iscsi_session *session = iscsi_ptr(sessionh);
-	struct iscsi_conn *tmp = ERR_PTR(-EEXIST), *conn = iscsi_ptr(connh);
+	struct Scsi_Host *shost = iscsi_session_to_shost(cls_session);
+	struct iscsi_session *session = iscsi_hostdata(shost->hostdata);
+	struct iscsi_conn *tmp = ERR_PTR(-EEXIST), *conn = cls_conn->dd_data;
 	struct sock *sk;
 	struct socket *sock;
 	int err;
@@ -2703,9 +2706,9 @@ iscsi_conn_bind(iscsi_sessionh_t sessionh, iscsi_connh_t connh,
 }
 
 static int
-iscsi_conn_start(iscsi_connh_t connh)
+iscsi_conn_start(struct iscsi_cls_conn *cls_conn)
 {
-	struct iscsi_conn *conn = iscsi_ptr(connh);
+	struct iscsi_conn *conn = cls_conn->dd_data;
 	struct iscsi_session *session = conn->session;
 	struct sock *sk;
 
@@ -2754,9 +2757,9 @@ iscsi_conn_start(iscsi_connh_t connh)
 }
 
 static void
-iscsi_conn_stop(iscsi_connh_t connh, int flag)
+iscsi_conn_stop(struct iscsi_cls_conn *cls_conn, int flag)
 {
-	struct iscsi_conn *conn = iscsi_ptr(connh);
+	struct iscsi_conn *conn = cls_conn->dd_data;
 	struct iscsi_session *session = conn->session;
 	struct sock *sk;
 	unsigned long flags;
@@ -3253,9 +3256,9 @@ static struct scsi_host_template iscsi_sht = {
 
 static struct iscsi_transport iscsi_tcp_transport;
 
-static struct Scsi_Host *
+static struct iscsi_cls_session *
 iscsi_session_create(struct scsi_transport_template *scsit,
-		     uint32_t initial_cmdsn)
+		     uint32_t initial_cmdsn, uint32_t *sid)
 {
 	struct Scsi_Host *shost;
 	struct iscsi_session *session;
@@ -3275,6 +3278,7 @@ iscsi_session_create(struct scsi_transport_template *scsit,
 	session->exp_cmdsn = initial_cmdsn + 1;
 	session->max_cmdsn = initial_cmdsn + 1;
 	session->max_r2t = 1;
+	*sid = shost->host_no;
 
 	/* initialize SCSI PDU commands pool */
 	if (iscsi_pool_init(&session->cmdpool, session->cmds_max,
@@ -3311,7 +3315,7 @@ iscsi_session_create(struct scsi_transport_template *scsit,
 	if (iscsi_r2tpool_alloc(session))
 		goto r2tpool_alloc_fail;
 
-	return shost;
+	return hostdata_session(shost->hostdata);
 
 r2tpool_alloc_fail:
 	for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++)
@@ -3321,12 +3325,14 @@ immdata_alloc_fail:
 mgmtpool_alloc_fail:
 	iscsi_pool_free(&session->cmdpool, (void**)session->cmds);
 cmdpool_alloc_fail:
+	iscsi_transport_destroy_session(shost);
 	return NULL;
 }
 
 static void
-iscsi_session_destroy(struct Scsi_Host *shost)
+iscsi_session_destroy(struct iscsi_cls_session *cls_session)
 {
+	struct Scsi_Host *shost = iscsi_session_to_shost(cls_session);
 	struct iscsi_session *session = iscsi_hostdata(shost->hostdata);
 	int cmd_i;
 	struct iscsi_data_task *dtask, *n;
@@ -3350,10 +3356,10 @@ iscsi_session_destroy(struct Scsi_Host *shost)
 }
 
 static int
-iscsi_conn_set_param(iscsi_connh_t connh, enum iscsi_param param,
+iscsi_conn_set_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param,
 		     uint32_t value)
 {
-	struct iscsi_conn *conn = iscsi_ptr(connh);
+	struct iscsi_conn *conn = cls_conn->dd_data;
 	struct iscsi_session *session = conn->session;
 
 	spin_lock_bh(&session->lock);
@@ -3495,9 +3501,10 @@ iscsi_conn_set_param(iscsi_connh_t connh, enum iscsi_param param,
 }
 
 static int
-iscsi_session_get_param(struct Scsi_Host *shost,
+iscsi_session_get_param(struct iscsi_cls_session *cls_session,
 			enum iscsi_param param, uint32_t *value)
 {
+	struct Scsi_Host *shost = iscsi_session_to_shost(cls_session);
 	struct iscsi_session *session = iscsi_hostdata(shost->hostdata);
 
 	switch(param) {
@@ -3539,9 +3546,10 @@ iscsi_session_get_param(struct Scsi_Host *shost,
 }
 
 static int
-iscsi_conn_get_param(void *data, enum iscsi_param param, uint32_t *value)
+iscsi_conn_get_param(struct iscsi_cls_conn *cls_conn,
+		     enum iscsi_param param, uint32_t *value)
 {
-	struct iscsi_conn *conn = data;
+	struct iscsi_conn *conn = cls_conn->dd_data;
 
 	switch(param) {
 	case ISCSI_PARAM_MAX_RECV_DLENGTH:
@@ -3564,9 +3572,9 @@ iscsi_conn_get_param(void *data, enum iscsi_param param, uint32_t *value)
 }
 
 static void
-iscsi_conn_get_stats(iscsi_connh_t connh, struct iscsi_stats *stats)
+iscsi_conn_get_stats(struct iscsi_cls_conn *cls_conn, struct iscsi_stats *stats)
 {
-	struct iscsi_conn *conn = iscsi_ptr(connh);
+	struct iscsi_conn *conn = cls_conn->dd_data;
 
 	stats->txdata_octets = conn->txdata_octets;
 	stats->rxdata_octets = conn->rxdata_octets;
@@ -3587,10 +3595,10 @@ iscsi_conn_get_stats(iscsi_connh_t connh, struct iscsi_stats *stats)
 }
 
 static int
-iscsi_conn_send_pdu(iscsi_connh_t connh, struct iscsi_hdr *hdr, char *data,
-		    uint32_t data_size)
+iscsi_conn_send_pdu(struct iscsi_cls_conn *cls_conn, struct iscsi_hdr *hdr,
+		    char *data, uint32_t data_size)
 {
-	struct iscsi_conn *conn = iscsi_ptr(connh);
+	struct iscsi_conn *conn = cls_conn->dd_data;
 	int rc;
 
 	mutex_lock(&conn->xmitmutex);
diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h
index f95e61b76f70..6766b817db2d 100644
--- a/drivers/scsi/iscsi_tcp.h
+++ b/drivers/scsi/iscsi_tcp.h
@@ -113,7 +113,10 @@ struct iscsi_tcp_recv {
 	int			datadgst;
 };
 
+struct iscsi_cls_conn;
+
 struct iscsi_conn {
+	struct iscsi_cls_conn	*cls_conn;	/* ptr to class connection */
 	struct iscsi_hdr	hdr;		/* header placeholder */
 	char			hdrext[4*sizeof(__u16) +
 				    sizeof(__u32)];
diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 59a1c9d9d3bd..b61868587dca 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -164,9 +164,43 @@ static struct mempool_zone *z_reply;
 #define Z_MAX_ERROR	16
 #define Z_HIWAT_ERROR	12
 
+static LIST_HEAD(sesslist);
+static DEFINE_SPINLOCK(sesslock);
 static LIST_HEAD(connlist);
 static DEFINE_SPINLOCK(connlock);
 
+static struct iscsi_cls_session *iscsi_session_lookup(uint64_t handle)
+{
+	unsigned long flags;
+	struct iscsi_cls_session *sess;
+
+	spin_lock_irqsave(&sesslock, flags);
+	list_for_each_entry(sess, &sesslist, sess_list) {
+		if (sess == iscsi_ptr(handle)) {
+			spin_unlock_irqrestore(&sesslock, flags);
+			return sess;
+		}
+	}
+	spin_unlock_irqrestore(&sesslock, flags);
+	return NULL;
+}
+
+static struct iscsi_cls_conn *iscsi_conn_lookup(uint64_t handle)
+{
+	unsigned long flags;
+	struct iscsi_cls_conn *conn;
+
+	spin_lock_irqsave(&connlock, flags);
+	list_for_each_entry(conn, &connlist, conn_list) {
+		if (conn == iscsi_ptr(handle)) {
+			spin_unlock_irqrestore(&connlock, flags);
+			return conn;
+		}
+	}
+	spin_unlock_irqrestore(&connlock, flags);
+	return NULL;
+}
+
 /*
  * The following functions can be used by LLDs that allocate
  * their own scsi_hosts or by software iscsi LLDs
@@ -365,6 +399,7 @@ iscsi_transport_create_session(struct scsi_transport_template *scsit,
 {
 	struct iscsi_cls_session *session;
 	struct Scsi_Host *shost;
+	unsigned long flags;
 
 	shost = scsi_host_alloc(transport->host_template,
 				hostdata_privsize(transport));
@@ -389,6 +424,9 @@ iscsi_transport_create_session(struct scsi_transport_template *scsit,
 		goto remove_host;
 
 	*(unsigned long*)shost->hostdata = (unsigned long)session;
+	spin_lock_irqsave(&sesslock, flags);
+	list_add(&session->sess_list, &sesslist);
+	spin_unlock_irqrestore(&sesslock, flags);
 	return shost;
 
 remove_host:
@@ -410,9 +448,13 @@ EXPORT_SYMBOL_GPL(iscsi_transport_create_session);
 int iscsi_transport_destroy_session(struct Scsi_Host *shost)
 {
 	struct iscsi_cls_session *session;
+	unsigned long flags;
 
 	scsi_remove_host(shost);
 	session = hostdata_session(shost->hostdata);
+	spin_lock_irqsave(&sesslock, flags);
+	list_del(&session->sess_list);
+	spin_unlock_irqrestore(&sesslock, flags);
 	iscsi_destroy_session(session);
 	/* ref from host alloc */
 	scsi_host_put(shost);
@@ -424,22 +466,6 @@ EXPORT_SYMBOL_GPL(iscsi_transport_destroy_session);
 /*
  * iscsi interface functions
  */
-static struct iscsi_cls_conn*
-iscsi_if_find_conn(uint64_t key)
-{
-	unsigned long flags;
-	struct iscsi_cls_conn *conn;
-
-	spin_lock_irqsave(&connlock, flags);
-	list_for_each_entry(conn, &connlist, conn_list)
-		if (conn->connh == key) {
-			spin_unlock_irqrestore(&connlock, flags);
-			return conn;
-		}
-	spin_unlock_irqrestore(&connlock, flags);
-	return NULL;
-}
-
 static struct iscsi_internal *
 iscsi_if_transport_lookup(struct iscsi_transport *tt)
 {
@@ -559,25 +585,21 @@ iscsi_unicast_skb(struct mempool_zone *zone, struct sk_buff *skb)
 	return 0;
 }
 
-int iscsi_recv_pdu(iscsi_connh_t connh, struct iscsi_hdr *hdr,
+int iscsi_recv_pdu(struct iscsi_cls_conn *conn, struct iscsi_hdr *hdr,
 		   char *data, uint32_t data_size)
 {
 	struct nlmsghdr	*nlh;
 	struct sk_buff *skb;
 	struct iscsi_uevent *ev;
-	struct iscsi_cls_conn *conn;
 	char *pdu;
 	int len = NLMSG_SPACE(sizeof(*ev) + sizeof(struct iscsi_hdr) +
 			      data_size);
 
-	conn = iscsi_if_find_conn(connh);
-	BUG_ON(!conn);
-
 	mempool_zone_complete(conn->z_pdu);
 
 	skb = mempool_zone_get_skb(conn->z_pdu);
 	if (!skb) {
-		iscsi_conn_error(connh, ISCSI_ERR_CONN_FAILED);
+		iscsi_conn_error(conn, ISCSI_ERR_CONN_FAILED);
 		dev_printk(KERN_ERR, &conn->dev, "iscsi: can not deliver "
 			   "control PDU: OOM\n");
 		return -ENOMEM;
@@ -590,7 +612,7 @@ int iscsi_recv_pdu(iscsi_connh_t connh, struct iscsi_hdr *hdr,
 	ev->type = ISCSI_KEVENT_RECV_PDU;
 	if (atomic_read(&conn->z_pdu->allocated) >= conn->z_pdu->hiwat)
 		ev->iferror = -ENOMEM;
-	ev->r.recv_req.conn_handle = connh;
+	ev->r.recv_req.conn_handle = iscsi_handle(conn);
 	pdu = (char*)ev + sizeof(*ev);
 	memcpy(pdu, hdr, sizeof(struct iscsi_hdr));
 	memcpy(pdu + sizeof(struct iscsi_hdr), data, data_size);
@@ -599,17 +621,13 @@ int iscsi_recv_pdu(iscsi_connh_t connh, struct iscsi_hdr *hdr,
 }
 EXPORT_SYMBOL_GPL(iscsi_recv_pdu);
 
-void iscsi_conn_error(iscsi_connh_t connh, enum iscsi_err error)
+void iscsi_conn_error(struct iscsi_cls_conn *conn, enum iscsi_err error)
 {
 	struct nlmsghdr	*nlh;
 	struct sk_buff	*skb;
 	struct iscsi_uevent *ev;
-	struct iscsi_cls_conn *conn;
 	int len = NLMSG_SPACE(sizeof(*ev));
 
-	conn = iscsi_if_find_conn(connh);
-	BUG_ON(!conn);
-
 	mempool_zone_complete(conn->z_error);
 
 	skb = mempool_zone_get_skb(conn->z_error);
@@ -626,7 +644,7 @@ void iscsi_conn_error(iscsi_connh_t connh, enum iscsi_err error)
 	if (atomic_read(&conn->z_error->allocated) >= conn->z_error->hiwat)
 		ev->iferror = -ENOMEM;
 	ev->r.connerror.error = error;
-	ev->r.connerror.conn_handle = connh;
+	ev->r.connerror.conn_handle = iscsi_handle(conn);
 
 	iscsi_unicast_skb(conn->z_error, skb);
 
@@ -677,7 +695,7 @@ iscsi_if_get_stats(struct iscsi_transport *transport, struct sk_buff *skb,
 			      ISCSI_STATS_CUSTOM_MAX);
 	int err = 0;
 
-	conn = iscsi_if_find_conn(ev->u.get_stats.conn_handle);
+	conn = iscsi_conn_lookup(ev->u.get_stats.conn_handle);
 	if (!conn)
 		return -EEXIST;
 
@@ -707,7 +725,7 @@ iscsi_if_get_stats(struct iscsi_transport *transport, struct sk_buff *skb,
 			((char*)evstat + sizeof(*evstat));
 		memset(stats, 0, sizeof(*stats));
 
-		transport->get_stats(ev->u.get_stats.conn_handle, stats);
+		transport->get_stats(conn, stats);
 		actual_size = NLMSG_SPACE(sizeof(struct iscsi_uevent) +
 					  sizeof(struct iscsi_stats) +
 					  sizeof(struct iscsi_stats_custom) *
@@ -727,58 +745,34 @@ static int
 iscsi_if_create_session(struct iscsi_internal *priv, struct iscsi_uevent *ev)
 {
 	struct iscsi_transport *transport = priv->iscsi_transport;
-	struct Scsi_Host *shost;
+	struct iscsi_cls_session *session;
+	uint32_t sid;
 
-	if (!transport->create_session)
-		return -EINVAL;
-
-	shost = transport->create_session(&priv->t,
-					  ev->u.c_session.initial_cmdsn);
-	if (!shost)
+	session = transport->create_session(&priv->t,
+					    ev->u.c_session.initial_cmdsn,
+					    &sid);
+	if (!session)
 		return -ENOMEM;
 
-	ev->r.c_session_ret.session_handle = iscsi_handle(iscsi_hostdata(shost->hostdata));
-	ev->r.c_session_ret.sid = shost->host_no;
+	ev->r.c_session_ret.session_handle = iscsi_handle(session);
+	ev->r.c_session_ret.sid = sid;
 	return 0;
 }
 
 static int
-iscsi_if_destroy_session(struct iscsi_internal *priv, struct iscsi_uevent *ev)
+iscsi_if_create_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev)
 {
-	struct iscsi_transport *transport = priv->iscsi_transport;
-
-	struct Scsi_Host *shost;
-
-	if (!transport->destroy_session)
-		return -EINVAL;
-
-	shost = scsi_host_lookup(ev->u.d_session.sid);
-	if (shost == ERR_PTR(-ENXIO))
-		return -EEXIST;
-
-	if (transport->destroy_session)
-		transport->destroy_session(shost);
-        /* ref from host lookup */
-        scsi_host_put(shost);
-	return 0;
-}
-
-static int
-iscsi_if_create_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev){
-	struct Scsi_Host *shost;
 	struct iscsi_cls_conn *conn;
+	struct iscsi_cls_session *session;
 	unsigned long flags;
 
-	if (!transport->create_conn)
+	session = iscsi_session_lookup(ev->u.c_conn.session_handle);
+	if (!session)
 		return -EINVAL;
 
-	shost = scsi_host_lookup(ev->u.c_conn.sid);
-	if (shost == ERR_PTR(-ENXIO))
-		return -EEXIST;
-
-	conn = transport->create_conn(shost, ev->u.c_conn.cid);
+	conn = transport->create_conn(session, ev->u.c_conn.cid);
 	if (!conn)
-		goto release_ref;
+		return -ENOMEM;
 
 	conn->z_pdu = mempool_zone_init(Z_MAX_PDU,
 			NLMSG_SPACE(sizeof(struct iscsi_uevent) +
@@ -800,14 +794,13 @@ iscsi_if_create_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev)
 		goto free_pdu_pool;
 	}
 
-	ev->r.handle = conn->connh = iscsi_handle(conn->dd_data);
+	ev->r.handle = iscsi_handle(conn);
 
 	spin_lock_irqsave(&connlock, flags);
 	list_add(&conn->conn_list, &connlist);
 	conn->active = 1;
 	spin_unlock_irqrestore(&connlock, flags);
 
-	scsi_host_put(shost);
 	return 0;
 
 free_pdu_pool:
@@ -815,8 +808,6 @@ free_pdu_pool:
 destroy_conn:
 	if (transport->destroy_conn)
 		transport->destroy_conn(conn->dd_data);
-release_ref:
-	scsi_host_put(shost);
 	return -ENOMEM;
 }
 
@@ -827,13 +818,9 @@ iscsi_if_destroy_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev
 	struct iscsi_cls_conn *conn;
 	struct mempool_zone *z_error, *z_pdu;
 
-	conn = iscsi_if_find_conn(ev->u.d_conn.conn_handle);
+	conn = iscsi_conn_lookup(ev->u.d_conn.conn_handle);
 	if (!conn)
-		return -EEXIST;
-
-	if (!transport->destroy_conn)
 		return -EINVAL;
-
 	spin_lock_irqsave(&connlock, flags);
 	conn->active = 0;
 	list_del(&conn->conn_list);
@@ -858,6 +845,8 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 	struct iscsi_uevent *ev = NLMSG_DATA(nlh);
 	struct iscsi_transport *transport = NULL;
 	struct iscsi_internal *priv;
+	struct iscsi_cls_session *session;
+	struct iscsi_cls_conn *conn;
 
 	if (NETLINK_CREDS(skb)->uid)
 		return -EPERM;
@@ -867,6 +856,9 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		return -EINVAL;
 	transport = priv->iscsi_transport;
 
+	if (!try_module_get(transport->owner))
+		return -EINVAL;
+
 	daemon_pid = NETLINK_CREDS(skb)->pid;
 
 	switch (nlh->nlmsg_type) {
@@ -874,7 +866,11 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		err = iscsi_if_create_session(priv, ev);
 		break;
 	case ISCSI_UEVENT_DESTROY_SESSION:
-		err = iscsi_if_destroy_session(priv, ev);
+		session = iscsi_session_lookup(ev->u.d_session.session_handle);
+		if (session)
+			transport->destroy_session(session);
+		else
+			err = -EINVAL;
 		break;
 	case ISCSI_UEVENT_CREATE_CONN:
 		err = iscsi_if_create_conn(transport, ev);
@@ -883,41 +879,48 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		err = iscsi_if_destroy_conn(transport, ev);
 		break;
 	case ISCSI_UEVENT_BIND_CONN:
-		if (!iscsi_if_find_conn(ev->u.b_conn.conn_handle))
-			return -EEXIST;
-		ev->r.retcode = transport->bind_conn(
-			ev->u.b_conn.session_handle,
-			ev->u.b_conn.conn_handle,
-			ev->u.b_conn.transport_fd,
-			ev->u.b_conn.is_leading);
+		session = iscsi_session_lookup(ev->u.b_conn.session_handle);
+		conn = iscsi_conn_lookup(ev->u.b_conn.conn_handle);
+
+		if (session && conn)
+			ev->r.retcode =	transport->bind_conn(session, conn,
+					ev->u.b_conn.transport_fd,
+					ev->u.b_conn.is_leading);
+		else
+			err = -EINVAL;
 		break;
 	case ISCSI_UEVENT_SET_PARAM:
-		if (!iscsi_if_find_conn(ev->u.set_param.conn_handle))
-			return -EEXIST;
-		ev->r.retcode = transport->set_param(
-			ev->u.set_param.conn_handle,
-			ev->u.set_param.param, ev->u.set_param.value);
+		conn = iscsi_conn_lookup(ev->u.set_param.conn_handle);
+		if (conn)
+			ev->r.retcode =	transport->set_param(conn,
+				ev->u.set_param.param, ev->u.set_param.value);
+		else
+			err = -EINVAL;
 		break;
 	case ISCSI_UEVENT_START_CONN:
-		if (!iscsi_if_find_conn(ev->u.start_conn.conn_handle))
-			return -EEXIST;
-		ev->r.retcode = transport->start_conn(
-			ev->u.start_conn.conn_handle);
+		conn = iscsi_conn_lookup(ev->u.start_conn.conn_handle);
+		if (conn)
+			ev->r.retcode = transport->start_conn(conn);
+		else
+			err = -EINVAL;
+
 		break;
 	case ISCSI_UEVENT_STOP_CONN:
-		if (!iscsi_if_find_conn(ev->u.stop_conn.conn_handle))
-			return -EEXIST;
-		transport->stop_conn(ev->u.stop_conn.conn_handle,
-			ev->u.stop_conn.flag);
+		conn = iscsi_conn_lookup(ev->u.stop_conn.conn_handle);
+		if (conn)
+			transport->stop_conn(conn, ev->u.stop_conn.flag);
+		else
+			err = -EINVAL;
 		break;
 	case ISCSI_UEVENT_SEND_PDU:
-		if (!iscsi_if_find_conn(ev->u.send_pdu.conn_handle))
-			return -EEXIST;
-		ev->r.retcode = transport->send_pdu(
-		       ev->u.send_pdu.conn_handle,
-		       (struct iscsi_hdr*)((char*)ev + sizeof(*ev)),
-		       (char*)ev + sizeof(*ev) + ev->u.send_pdu.hdr_size,
-			ev->u.send_pdu.data_size);
+		conn = iscsi_conn_lookup(ev->u.send_pdu.conn_handle);
+		if (conn)
+			ev->r.retcode =	transport->send_pdu(conn,
+				(struct iscsi_hdr*)((char*)ev + sizeof(*ev)),
+				(char*)ev + sizeof(*ev) + ev->u.send_pdu.hdr_size,
+				ev->u.send_pdu.data_size);
+		else
+			err = -EINVAL;
 		break;
 	case ISCSI_UEVENT_GET_STATS:
 		err = iscsi_if_get_stats(transport, skb, nlh);
@@ -927,6 +930,7 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		break;
 	}
 
+	module_put(transport->owner);
 	return err;
 }
 
@@ -997,7 +1001,7 @@ show_conn_int_param_##param(struct class_device *cdev, char *buf)	\
 	struct iscsi_cls_conn *conn = iscsi_cdev_to_conn(cdev);		\
 	struct iscsi_transport *t = conn->transport;			\
 									\
-	t->get_conn_param(conn->dd_data, param, &value);		\
+	t->get_conn_param(conn, param, &value);				\
 	return snprintf(buf, 20, format"\n", value);			\
 }
 
@@ -1024,10 +1028,9 @@ show_session_int_param_##param(struct class_device *cdev, char *buf)	\
 {									\
 	uint32_t value = 0;						\
 	struct iscsi_cls_session *session = iscsi_cdev_to_session(cdev);	\
-	struct Scsi_Host *shost = iscsi_session_to_shost(session);	\
 	struct iscsi_transport *t = session->transport;			\
 									\
-	t->get_session_param(shost, param, &value);			\
+	t->get_session_param(session, param, &value);			\
 	return snprintf(buf, 20, format"\n", value);			\
 }
 
diff --git a/include/scsi/iscsi_if.h b/include/scsi/iscsi_if.h
index 3e5cb5ab2d34..e5618b90996e 100644
--- a/include/scsi/iscsi_if.h
+++ b/include/scsi/iscsi_if.h
@@ -163,9 +163,6 @@ enum iscsi_param {
 };
 #define ISCSI_PARAM_MAX			14
 
-typedef uint64_t iscsi_sessionh_t;	/* iSCSI Data-Path session handle */
-typedef uint64_t iscsi_connh_t;		/* iSCSI Data-Path connection handle */
-
 #define iscsi_ptr(_handle) ((void*)(unsigned long)_handle)
 #define iscsi_handle(_ptr) ((uint64_t)(unsigned long)_ptr)
 #define hostdata_session(_hostdata) (iscsi_ptr(*(unsigned long *)_hostdata))
diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h
index 16602a547a63..b41cf077e54b 100644
--- a/include/scsi/scsi_transport_iscsi.h
+++ b/include/scsi/scsi_transport_iscsi.h
@@ -63,25 +63,28 @@ struct iscsi_transport {
 	int max_lun;
 	unsigned int max_conn;
 	unsigned int max_cmd_len;
-	struct Scsi_Host *(*create_session) (struct scsi_transport_template *t,
-					     uint32_t initial_cmdsn);
-	void (*destroy_session) (struct Scsi_Host *shost);
-	struct iscsi_cls_conn *(*create_conn) (struct Scsi_Host *shost,
+	struct iscsi_cls_session *(*create_session)
+		(struct scsi_transport_template *t, uint32_t sn, uint32_t *sid);
+	void (*destroy_session) (struct iscsi_cls_session *session);
+	struct iscsi_cls_conn *(*create_conn) (struct iscsi_cls_session *sess,
 				uint32_t cid);
-	int (*bind_conn) (iscsi_sessionh_t session, iscsi_connh_t conn,
+	int (*bind_conn) (struct iscsi_cls_session *session,
+			  struct iscsi_cls_conn *cls_conn,
 			  uint32_t transport_fd, int is_leading);
-	int (*start_conn) (iscsi_connh_t conn);
-	void (*stop_conn) (iscsi_connh_t conn, int flag);
+	int (*start_conn) (struct iscsi_cls_conn *conn);
+	void (*stop_conn) (struct iscsi_cls_conn *conn, int flag);
 	void (*destroy_conn) (struct iscsi_cls_conn *conn);
-	int (*set_param) (iscsi_connh_t conn, enum iscsi_param param,
+	int (*set_param) (struct iscsi_cls_conn *conn, enum iscsi_param param,
 			  uint32_t value);
-	int (*get_conn_param) (void *conndata, enum iscsi_param param,
+	int (*get_conn_param) (struct iscsi_cls_conn *conn,
+			       enum iscsi_param param,
 			       uint32_t *value);
-	int (*get_session_param) (struct Scsi_Host *shost,
+	int (*get_session_param) (struct iscsi_cls_session *session,
 				  enum iscsi_param param, uint32_t *value);
-	int (*send_pdu) (iscsi_connh_t conn, struct iscsi_hdr *hdr,
+	int (*send_pdu) (struct iscsi_cls_conn *conn, struct iscsi_hdr *hdr,
 			 char *data, uint32_t data_size);
-	void (*get_stats) (iscsi_connh_t conn, struct iscsi_stats *stats);
+	void (*get_stats) (struct iscsi_cls_conn *conn,
+			   struct iscsi_stats *stats);
 };
 
 /*
@@ -93,15 +96,14 @@ extern int iscsi_unregister_transport(struct iscsi_transport *tt);
 /*
  * control plane upcalls
  */
-extern void iscsi_conn_error(iscsi_connh_t conn, enum iscsi_err error);
-extern int iscsi_recv_pdu(iscsi_connh_t conn, struct iscsi_hdr *hdr,
+extern void iscsi_conn_error(struct iscsi_cls_conn *conn, enum iscsi_err error);
+extern int iscsi_recv_pdu(struct iscsi_cls_conn *conn, struct iscsi_hdr *hdr,
 			  char *data, uint32_t data_size);
 
 struct iscsi_cls_conn {
 	struct list_head conn_list;	/* item in connlist */
 	void *dd_data;			/* LLD private data */
 	struct iscsi_transport *transport;
-	iscsi_connh_t connh;
 	int active;			/* must be accessed with the connlock */
 	struct device dev;		/* sysfs transport/container device */
 	struct mempool_zone *z_error;
@@ -113,7 +115,7 @@ struct iscsi_cls_conn {
 	container_of(_dev, struct iscsi_cls_conn, dev)
 
 struct iscsi_cls_session {
-	struct list_head list;	/* item in session_list */
+	struct list_head sess_list;		/* item in session_list */
 	struct iscsi_transport *transport;
 	struct device dev;	/* sysfs transport/container device */
 };

From 5b940adf5b341b12dbb94e7cbdb416b35f52017b Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:06:56 -0600
Subject: [PATCH 017/509] [SCSI] iscsi update: pass correct skb to skb_trim

>From da-x@monatomic.org:

Wrong skb is passed to skb_trim in iscsi_if_get_stats.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index b61868587dca..79ca29ee1aee 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -680,8 +680,7 @@ iscsi_if_send_reply(int pid, int seq, int type, int done, int multi,
 }
 
 static int
-iscsi_if_get_stats(struct iscsi_transport *transport, struct sk_buff *skb,
-		   struct nlmsghdr *nlh)
+iscsi_if_get_stats(struct iscsi_transport *transport, struct nlmsghdr *nlh)
 {
 	struct iscsi_uevent *ev = NLMSG_DATA(nlh);
 	struct iscsi_stats *stats;
@@ -732,7 +731,7 @@ iscsi_if_get_stats(struct iscsi_transport *transport, struct sk_buff *skb,
 					  stats->custom_length);
 		actual_size -= sizeof(*nlhstat);
 		actual_size = NLMSG_LENGTH(actual_size);
-		skb_trim(skb, NLMSG_ALIGN(actual_size));
+		skb_trim(skbstat, NLMSG_ALIGN(actual_size));
 		nlhstat->nlmsg_len = actual_size;
 
 		err = iscsi_unicast_skb(conn->z_pdu, skbstat);
@@ -923,7 +922,7 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 			err = -EINVAL;
 		break;
 	case ISCSI_UEVENT_GET_STATS:
-		err = iscsi_if_get_stats(transport, skb, nlh);
+		err = iscsi_if_get_stats(transport, nlh);
 		break;
 	default:
 		err = -EINVAL;

From 142e301fc818de9b116706835cd9fc864e73f203 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:06:58 -0600
Subject: [PATCH 018/509] [SCSI] iscsi update: setup pool before using

>From andmike@us.ibm.com:

 Ensure that pool data is setup prior to calling mempool_create as it will
 call the the alloc function during create.

Signed-off-by: Mike Anderson <andmike@us.ibm.com>
Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 79ca29ee1aee..448fd78777f9 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -530,6 +530,12 @@ mempool_zone_init(unsigned max, unsigned size, unsigned hiwat)
 	if (!zp)
 		return NULL;
 
+	zp->size = size;
+	zp->hiwat = hiwat;
+	INIT_LIST_HEAD(&zp->freequeue);
+	spin_lock_init(&zp->freelock);
+	atomic_set(&zp->allocated, 0);
+
 	zp->pool = mempool_create(max, mempool_zone_alloc_skb,
 				  mempool_zone_free_skb, zp);
 	if (!zp->pool) {
@@ -537,13 +543,6 @@ mempool_zone_init(unsigned max, unsigned size, unsigned hiwat)
 		return NULL;
 	}
 
-	zp->size = size;
-	zp->hiwat = hiwat;
-
-	INIT_LIST_HEAD(&zp->freequeue);
-	spin_lock_init(&zp->freelock);
-	atomic_set(&zp->allocated, 0);
-
 	return zp;
 }
 

From ee7f8e405342722e42c15fe8e841a679f8951eea Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:07:01 -0600
Subject: [PATCH 019/509] [SCSI] iscsi update: set deamon pid earlier

>From michaelc@cs.wisc.edu:

If the transport lookup fails we set the daemon pid too late.
This can cause us deadlock since the netlink code will think we
meant to call back into our iscsi_if_rx function.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 448fd78777f9..7fb69183c72d 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -846,9 +846,6 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 	struct iscsi_cls_session *session;
 	struct iscsi_cls_conn *conn;
 
-	if (NETLINK_CREDS(skb)->uid)
-		return -EPERM;
-
 	priv = iscsi_if_transport_lookup(iscsi_ptr(ev->transport_handle));
 	if (!priv)
 		return -EINVAL;
@@ -857,8 +854,6 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 	if (!try_module_get(transport->owner))
 		return -EINVAL;
 
-	daemon_pid = NETLINK_CREDS(skb)->pid;
-
 	switch (nlh->nlmsg_type) {
 	case ISCSI_UEVENT_CREATE_SESSION:
 		err = iscsi_if_create_session(priv, ev);
@@ -934,7 +929,7 @@ iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 
 /* Get message from skb (based on rtnetlink_rcv_skb).  Each message is
  * processed by iscsi_if_recv_msg.  Malformed skbs with wrong length are
- * discarded silently.  */
+ * or invalid creds discarded silently.  */
 static void
 iscsi_if_rx(struct sock *sk, int len)
 {
@@ -942,6 +937,12 @@ iscsi_if_rx(struct sock *sk, int len)
 
 	mutex_lock(&rx_queue_mutex);
 	while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) {
+		if (NETLINK_CREDS(skb)->uid) {
+			skb_pull(skb, skb->len);
+			goto free_skb;
+		}
+		daemon_pid = NETLINK_CREDS(skb)->pid;
+
 		while (skb->len >= NLMSG_SPACE(0)) {
 			int err;
 			uint32_t rlen;
@@ -953,10 +954,12 @@ iscsi_if_rx(struct sock *sk, int len)
 			    skb->len < nlh->nlmsg_len) {
 				break;
 			}
+
 			ev = NLMSG_DATA(nlh);
 			rlen = NLMSG_ALIGN(nlh->nlmsg_len);
 			if (rlen > skb->len)
 				rlen = skb->len;
+
 			err = iscsi_if_recv_msg(skb, nlh);
 			if (err) {
 				ev->type = ISCSI_KEVENT_IF_ERROR;
@@ -980,6 +983,7 @@ iscsi_if_rx(struct sock *sk, int len)
 			} while (err < 0 && err != -ECONNREFUSED);
 			skb_pull(skb, rlen);
 		}
+free_skb:
 		kfree_skb(skb);
 	}
 	mutex_unlock(&rx_queue_mutex);

From 1fd459e367657f595ddf192b9a46298e18d4fc13 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:07:03 -0600
Subject: [PATCH 020/509] [SCSI] iscsi update: rm conn lock

>From erezz@voltaire.com:

rm conn->lock since it is not used anymore. The dataqueue is protected
by the session lock and xmitmutex.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/iscsi_tcp.c | 4 ----
 drivers/scsi/iscsi_tcp.h | 1 -
 2 files changed, 5 deletions(-)

diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index d07d309ac026..0cd78b1d1aaa 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -244,12 +244,10 @@ iscsi_ctask_cleanup(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask)
 	if (sc->sc_data_direction == DMA_TO_DEVICE) {
 		struct iscsi_data_task *dtask, *n;
 		/* WRITE: cleanup Data-Out's if any */
-		spin_lock(&conn->lock);
 		list_for_each_entry_safe(dtask, n, &ctask->dataqueue, item) {
 			list_del(&dtask->item);
 			mempool_free(dtask, ctask->datapool);
 		}
-		spin_unlock(&conn->lock);
 	}
 	ctask->xmstate = XMSTATE_IDLE;
 	ctask->r2t = NULL;
@@ -2453,8 +2451,6 @@ iscsi_conn_create(struct iscsi_cls_session *cls_session, uint32_t conn_idx)
 	conn->data_size = DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH;
 	conn->max_recv_dlength = DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH;
 
-	spin_lock_init(&conn->lock);
-
 	/* initialize general xmit PDU commands queue */
 	conn->xmitqueue = kfifo_alloc(session->cmds_max * sizeof(void*),
 					GFP_KERNEL, NULL);
diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h
index 6766b817db2d..ba26741ac154 100644
--- a/drivers/scsi/iscsi_tcp.h
+++ b/drivers/scsi/iscsi_tcp.h
@@ -146,7 +146,6 @@ struct iscsi_conn {
 	struct iscsi_mgmt_task	*login_mtask;	/* mtask used for login/text */
 	struct iscsi_mgmt_task	*mtask;		/* xmit mtask in progress */
 	struct iscsi_cmd_task	*ctask;		/* xmit ctask in progress */
-	spinlock_t		lock;		/* FIXME: to be removed */
 
 	/* old values for socket callbacks */
 	void			(*old_data_ready)(struct sock *, int);

From 351f739e68e97a0316136a5bda145b952d99a989 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:07:06 -0600
Subject: [PATCH 021/509] [SCSI] iscsi update: set correct state at creation
 time

>From erezz@voltaire.com:

We are still in ISCSI_STATE_FREE state at create time. The addition
of the first connection puts us in ISCSI_STATE_LOGGED_IN.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/iscsi_tcp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 0cd78b1d1aaa..579eecfd186e 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -3267,7 +3267,7 @@ iscsi_session_create(struct scsi_transport_template *scsit,
 	session = iscsi_hostdata(shost->hostdata);
 	memset(session, 0, sizeof(struct iscsi_session));
 	session->host = shost;
-	session->state = ISCSI_STATE_LOGGED_IN;
+	session->state = ISCSI_STATE_FREE;
 	session->mgmtpool_max = ISCSI_MGMT_CMDS_MAX;
 	session->cmds_max = ISCSI_XMIT_CMDS_MAX;
 	session->cmdsn = initial_cmdsn;

From b36ae07cb7757bd3eabd13c79844083dccac2f77 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:07:09 -0600
Subject: [PATCH 022/509] [SCSI] iscsi update: fix mgmt pool err path release

>From ogerlitz@voltaire.com:

mgmtpool shoild be frees in immdata_alloc_fail label.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/iscsi_tcp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c
index 579eecfd186e..ff79e68b347c 100644
--- a/drivers/scsi/iscsi_tcp.c
+++ b/drivers/scsi/iscsi_tcp.c
@@ -3316,8 +3316,8 @@ iscsi_session_create(struct scsi_transport_template *scsit,
 r2tpool_alloc_fail:
 	for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++)
 		kfree(session->mgmt_cmds[cmd_i]->data);
-	iscsi_pool_free(&session->mgmtpool, (void**)session->mgmt_cmds);
 immdata_alloc_fail:
+	iscsi_pool_free(&session->mgmtpool, (void**)session->mgmt_cmds);
 mgmtpool_alloc_fail:
 	iscsi_pool_free(&session->cmdpool, (void**)session->cmds);
 cmdpool_alloc_fail:

From 28e5554df63085be3b8bd2aee6ddbc479f0d136e Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:07:11 -0600
Subject: [PATCH 023/509] [SCSI] iscsi update: use gfp_t

Use gfp_t. I accidentally removed this in our last update.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 7fb69183c72d..55860d26f999 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -489,7 +489,7 @@ static inline struct list_head *skb_to_lh(struct sk_buff *skb)
 }
 
 static void*
-mempool_zone_alloc_skb(unsigned int gfp_mask, void *pool_data)
+mempool_zone_alloc_skb(gfp_t gfp_mask, void *pool_data)
 {
 	struct mempool_zone *zone = pool_data;
 

From b5b81016538cf84a10c80438b7aa750dd375ba93 Mon Sep 17 00:00:00 2001
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 1 Feb 2006 21:07:14 -0600
Subject: [PATCH 024/509] [SCSI] iscsi update: rm unused sessions list

rm unused sessions list.

This patch is last becuase I was not sure if this patchset was
going to be applied over the kmalloc2kzalloc one by JesS. If it
is then this patch will not apply and can be dropped for now. I will
resend later when things setttle down.

Signed-off-by: Mike Christie <michaelc@cs.wisc.edu>
Signed-off-by: Alex Aizman <itn780@yahoo.com>
Signed-off-by: Dmitry Yusupov <dmitry_yus@yahoo.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_transport_iscsi.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c
index 55860d26f999..71e54a64adca 100644
--- a/drivers/scsi/scsi_transport_iscsi.c
+++ b/drivers/scsi/scsi_transport_iscsi.c
@@ -38,10 +38,6 @@ struct iscsi_internal {
 	struct scsi_transport_template t;
 	struct iscsi_transport *iscsi_transport;
 	struct list_head list;
-	/*
-	 * List of sessions for this transport
-	 */
-	struct list_head sessions;
 	/*
 	 * based on transport capabilities, at register time we set these
 	 * bits to tell the transport class it wants attributes displayed
@@ -1126,7 +1122,6 @@ iscsi_register_transport(struct iscsi_transport *tt)
 		return NULL;
 	memset(priv, 0, sizeof(*priv));
 	INIT_LIST_HEAD(&priv->list);
-	INIT_LIST_HEAD(&priv->sessions);
 	priv->iscsi_transport = tt;
 
 	priv->cdev.class = &iscsi_transport_class;

From 096f7a2a094af3007937d6fd21560e28dca0994d Mon Sep 17 00:00:00 2001
From: "Moore, Eric" <Eric.Moore@lsil.com>
Date: Thu, 2 Feb 2006 17:19:30 -0700
Subject: [PATCH 025/509] [SCSI] fusion - mptctl - MPTCOMMAND - adding function
 types.

This adds support for new function types in
the existing MPTCOMMAND ioctl.

Signed-off-by: Eric Moore <Eric.Moore@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptctl.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index bdf709987982..70a812a2514d 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -1817,6 +1817,8 @@ mptctl_do_mpt_command (struct mpt_ioctl_command karg, void __user *mfPtr)
 	case MPI_FUNCTION_SCSI_ENCLOSURE_PROCESSOR:
 	case MPI_FUNCTION_FW_DOWNLOAD:
 	case MPI_FUNCTION_FC_PRIMITIVE_SEND:
+	case MPI_FUNCTION_TOOLBOX:
+	case MPI_FUNCTION_SAS_IO_UNIT_CONTROL:
 		break;
 
 	case MPI_FUNCTION_SCSI_IO_REQUEST:
@@ -1888,6 +1890,25 @@ mptctl_do_mpt_command (struct mpt_ioctl_command karg, void __user *mfPtr)
 		}
 		break;
 
+	case MPI_FUNCTION_SMP_PASSTHROUGH:
+		/* Check mf->PassthruFlags to determine if
+		 * transfer is ImmediateMode or not.
+		 * Immediate mode returns data in the ReplyFrame.
+		 * Else, we are sending request and response data
+		 * in two SGLs at the end of the mf.
+		 */
+		break;
+
+	case MPI_FUNCTION_SATA_PASSTHROUGH:
+		if (!ioc->sh) {
+			printk(KERN_ERR "%s@%d::mptctl_do_mpt_command - "
+				"SCSI driver is not loaded. \n",
+					__FILE__, __LINE__);
+			rc = -EFAULT;
+			goto done_free_mem;
+		}
+		break;
+
 	case MPI_FUNCTION_RAID_ACTION:
 		/* Just add a SGE
 		 */

From 9cc1cfbc67d77164f5b612fcf833460eca4d81e9 Mon Sep 17 00:00:00 2001
From: "Moore, Eric" <Eric.Moore@lsil.com>
Date: Thu, 2 Feb 2006 17:19:33 -0700
Subject: [PATCH 026/509] [SCSI] fusion - mptctl - adding support for
 bus_type=SAS

Add bus_type recognization in ioctl path for SAS.

Signed-off-by: Eric Moore <Eric.Moore@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptctl.c | 8 +++++---
 drivers/message/fusion/mptctl.h | 4 +++-
 2 files changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index 70a812a2514d..1a1bc66c8e9c 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -1145,7 +1145,9 @@ mptctl_getiocinfo (unsigned long arg, unsigned int data_size)
 	/* Fill in the data and return the structure to the calling
 	 * program
 	 */
-	if (ioc->bus_type == FC)
+	if (ioc->bus_type == SAS)
+		karg->adapterType = MPT_IOCTL_INTERFACE_SAS;
+	else if (ioc->bus_type == FC)
 		karg->adapterType = MPT_IOCTL_INTERFACE_FC;
 	else
 		karg->adapterType = MPT_IOCTL_INTERFACE_SCSI;
@@ -2391,7 +2393,7 @@ mptctl_hp_hostinfo(unsigned long arg, unsigned int data_size)
 
 	karg.base_io_addr = pci_resource_start(pdev, 0);
 
-	if (ioc->bus_type == FC)
+	if ((ioc->bus_type == SAS) || (ioc->bus_type == FC))
 		karg.bus_phys_width = HP_BUS_WIDTH_UNK;
 	else
 		karg.bus_phys_width = HP_BUS_WIDTH_16;
@@ -2480,7 +2482,7 @@ mptctl_hp_targetinfo(unsigned long arg)
 
 	/*  There is nothing to do for FCP parts.
 	 */
-	if (ioc->bus_type == FC)
+	if ((ioc->bus_type == SAS) || (ioc->bus_type == FC))
 		return 0;
 
 	if ((ioc->spi_data.sdp0length == 0) || (ioc->sh == NULL))
diff --git a/drivers/message/fusion/mptctl.h b/drivers/message/fusion/mptctl.h
index 518996e03481..a2f8a97992e6 100644
--- a/drivers/message/fusion/mptctl.h
+++ b/drivers/message/fusion/mptctl.h
@@ -169,8 +169,10 @@ struct mpt_ioctl_pci_info2 {
  *  Read only.
  *  Data starts at offset 0xC
  */
-#define MPT_IOCTL_INTERFACE_FC		(0x01)
 #define MPT_IOCTL_INTERFACE_SCSI	(0x00)
+#define MPT_IOCTL_INTERFACE_FC		(0x01)
+#define MPT_IOCTL_INTERFACE_FC_IP	(0x02)
+#define MPT_IOCTL_INTERFACE_SAS		(0x03)
 #define MPT_IOCTL_VERSION_LENGTH	(32)
 
 struct mpt_ioctl_iocinfo {

From 86a7dcaae9c67a344e51190734b98684072d181c Mon Sep 17 00:00:00 2001
From: "Moore, Eric" <Eric.Moore@lsil.com>
Date: Thu, 2 Feb 2006 17:19:37 -0700
Subject: [PATCH 027/509] [SCSI] fusion - mtctl - change to wait_event_timeout

Change from using wait_event_interruptible_timeout to
wait_event_timeout.  Also delete white space and duplicate
line of code.

Signed-off-by: Eric Moore <Eric.Moore@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptctl.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index 1a1bc66c8e9c..8e2369ff3322 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -385,18 +385,18 @@ static int mptctl_bus_reset(MPT_IOCTL *ioctl)
 	}
 
 	/* Now wait for the command to complete */
-	ii = wait_event_interruptible_timeout(mptctl_wait,
+	ii = wait_event_timeout(mptctl_wait,
 	     ioctl->wait_done == 1,
 	     HZ*5 /* 5 second timeout */);
 
 	if(ii <=0 && (ioctl->wait_done != 1 ))  {
+		mpt_free_msg_frame(hd->ioc, mf);
 		ioctl->wait_done = 0;
 		retval = -1; /* return failure */
 	}
 
 mptctl_bus_reset_done:
 
-	mpt_free_msg_frame(hd->ioc, mf);
 	mptctl_free_tm_flags(ioctl->ioc);
 	return retval;
 }
@@ -807,7 +807,7 @@ mptctl_do_fw_download(int ioc, char __user *ufwbuf, size_t fwlen)
 	mpt_put_msg_frame(mptctl_id, iocp, mf);
 
 	/* Now wait for the command to complete */
-	ret = wait_event_interruptible_timeout(mptctl_wait,
+	ret = wait_event_timeout(mptctl_wait,
 	     iocp->ioctl->wait_done == 1,
 	     HZ*60);
 
@@ -1172,12 +1172,11 @@ mptctl_getiocinfo (unsigned long arg, unsigned int data_size)
 		karg->pciInfo.u.bits.deviceNumber = PCI_SLOT( pdev->devfn );
 		karg->pciInfo.u.bits.functionNumber = PCI_FUNC( pdev->devfn );
 	} else if (cim_rev == 2) {
-		/* Get the PCI bus, device, function and segment ID numbers 
+		/* Get the PCI bus, device, function and segment ID numbers
 		   for the IOC */
 		karg->pciInfo.u.bits.busNumber = pdev->bus->number;
 		karg->pciInfo.u.bits.deviceNumber = PCI_SLOT( pdev->devfn );
 		karg->pciInfo.u.bits.functionNumber = PCI_FUNC( pdev->devfn );
-		karg->pciInfo.u.bits.functionNumber = PCI_FUNC( pdev->devfn );
 		karg->pciInfo.segmentID = pci_domain_nr(pdev->bus);
 	}
 
@@ -2153,7 +2152,7 @@ mptctl_do_mpt_command (struct mpt_ioctl_command karg, void __user *mfPtr)
 
 	/* Now wait for the command to complete */
 	timeout = (karg.timeout > 0) ? karg.timeout : MPT_IOCTL_DEFAULT_TIMEOUT;
-	timeout = wait_event_interruptible_timeout(mptctl_wait,
+	timeout = wait_event_timeout(mptctl_wait,
 	     ioc->ioctl->wait_done == 1,
 	     HZ*timeout);
 

From 5b5ef4f617f1706a23b0433d89a9c02ceb0980a5 Mon Sep 17 00:00:00 2001
From: "Moore, Eric" <Eric.Moore@lsil.com>
Date: Thu, 2 Feb 2006 17:19:40 -0700
Subject: [PATCH 028/509] [SCSI] fusion - mptctl - Event Log Fix

Use the hard coded value MPTCTL_EVENT_LOG_SIZE to fix
bug where in certain cases, the ioc->eventLogSize was
initialized.

Signed-off-by: Eric Moore <Eric.Moore@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptbase.c  | 2 +-
 drivers/message/fusion/mptctl.c   | 5 ++---
 drivers/message/fusion/mptscsih.c | 2 +-
 3 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c
index 9a2c7605d49c..a3751d86216e 100644
--- a/drivers/message/fusion/mptbase.c
+++ b/drivers/message/fusion/mptbase.c
@@ -6142,7 +6142,7 @@ ProcessEventNotification(MPT_ADAPTER *ioc, EventNotificationReply_t *pEventReply
 	if (ioc->events && (ioc->eventTypes & ( 1 << event))) {
 		int idx;
 
-		idx = ioc->eventContext % ioc->eventLogSize;
+		idx = ioc->eventContext % MPTCTL_EVENT_LOG_SIZE;
 
 		ioc->events[idx].event = event;
 		ioc->events[idx].eventContext = ioc->eventContext;
diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index 8e2369ff3322..b603fb8e604b 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -1501,7 +1501,7 @@ mptctl_eventquery (unsigned long arg)
 		return -ENODEV;
 	}
 
-	karg.eventEntries = ioc->eventLogSize;
+	karg.eventEntries = MPTCTL_EVENT_LOG_SIZE;
 	karg.eventTypes = ioc->eventTypes;
 
 	/* Copy the data from kernel memory to user memory
@@ -1551,7 +1551,6 @@ mptctl_eventenable (unsigned long arg)
 		memset(ioc->events, 0, sz);
 		ioc->alloc_total += sz;
 
-		ioc->eventLogSize = MPTCTL_EVENT_LOG_SIZE;
 		ioc->eventContext = 0;
         }
 
@@ -1591,7 +1590,7 @@ mptctl_eventreport (unsigned long arg)
 	maxEvents = numBytes/sizeof(MPT_IOCTL_EVENTS);
 
 
-	max = ioc->eventLogSize < maxEvents ? ioc->eventLogSize : maxEvents;
+	max = MPTCTL_EVENT_LOG_SIZE < maxEvents ? MPTCTL_EVENT_LOG_SIZE : maxEvents;
 
 	/* If fewer than 1 event is requested, there must have
 	 * been some type of error.
diff --git a/drivers/message/fusion/mptscsih.c b/drivers/message/fusion/mptscsih.c
index 05789e505464..4fee6befc93d 100644
--- a/drivers/message/fusion/mptscsih.c
+++ b/drivers/message/fusion/mptscsih.c
@@ -2489,7 +2489,7 @@ mptscsih_copy_sense_data(struct scsi_cmnd *sc, MPT_SCSI_HOST *hd, MPT_FRAME_HDR
 				int idx;
 				MPT_ADAPTER *ioc = hd->ioc;
 
-				idx = ioc->eventContext % ioc->eventLogSize;
+				idx = ioc->eventContext % MPTCTL_EVENT_LOG_SIZE;
 				ioc->events[idx].event = MPI_EVENT_SCSI_DEVICE_STATUS_CHANGE;
 				ioc->events[idx].eventContext = ioc->eventContext;
 

From 5f07e2499d629045f7f8a60a5b442792f08732cb Mon Sep 17 00:00:00 2001
From: "Moore, Eric" <Eric.Moore@lsil.com>
Date: Thu, 2 Feb 2006 17:19:44 -0700
Subject: [PATCH 029/509] [SCSI] fusion - mptctl -sense width fix

Bug fix for correctly setting sense width
for the MPTCOMMAND ioctl.

Signed-off-by: Eric Moore <Eric.Moore@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptctl.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index b603fb8e604b..be5fcd8db63b 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -1839,7 +1839,9 @@ mptctl_do_mpt_command (struct mpt_ioctl_command karg, void __user *mfPtr)
 				goto done_free_mem;
 			}
 
-			pScsiReq->MsgFlags = mpt_msg_flags();
+			pScsiReq->MsgFlags &= ~MPI_SCSIIO_MSGFLGS_SENSE_WIDTH;
+			pScsiReq->MsgFlags |= mpt_msg_flags();
+
 
 			/* verify that app has not requested
 			 *	more sense data than driver
@@ -1921,7 +1923,9 @@ mptctl_do_mpt_command (struct mpt_ioctl_command karg, void __user *mfPtr)
 			int scsidir = MPI_SCSIIO_CONTROL_READ;
 			int dataSize;
 
-			pScsiReq->MsgFlags = mpt_msg_flags();
+			pScsiReq->MsgFlags &= ~MPI_SCSIIO_MSGFLGS_SENSE_WIDTH;
+			pScsiReq->MsgFlags |= mpt_msg_flags();
+
 
 			/* verify that app has not requested
 			 *	more sense data than driver

From 592f9c2fc9725b922ba8c4b1d67318ea4a301b59 Mon Sep 17 00:00:00 2001
From: "Moore, Eric" <Eric.Moore@lsil.com>
Date: Thu, 2 Feb 2006 17:19:47 -0700
Subject: [PATCH 030/509] [SCSI] fusion - mptctl - backplane istwi fix

Moving the toolbox call from mptbase.c, over to
mptctl.c, and using the mptctl infastructure to issue
the call.  The existing code is hanging on certain HP platforms
when this ioctl is issued, and this patch fix's that.

Signed-off-by: Eric Moore <Eric.Moore@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptbase.c | 113 +------------------------------
 drivers/message/fusion/mptbase.h |   1 -
 drivers/message/fusion/mptctl.c  |  76 +++++++++++++++++----
 3 files changed, 64 insertions(+), 126 deletions(-)

diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c
index a3751d86216e..642a61b6d0a4 100644
--- a/drivers/message/fusion/mptbase.c
+++ b/drivers/message/fusion/mptbase.c
@@ -452,8 +452,7 @@ mpt_base_reply(MPT_ADAPTER *ioc, MPT_FRAME_HDR *mf, MPT_FRAME_HDR *reply)
 	} else if (func == MPI_FUNCTION_EVENT_ACK) {
 		dprintk((MYIOC_s_INFO_FMT "mpt_base_reply, EventAck reply received\n",
 				ioc->name));
-	} else if (func == MPI_FUNCTION_CONFIG ||
-		   func == MPI_FUNCTION_TOOLBOX) {
+	} else if (func == MPI_FUNCTION_CONFIG) {
 		CONFIGPARMS *pCfg;
 		unsigned long flags;
 
@@ -5326,115 +5325,6 @@ mpt_config(MPT_ADAPTER *ioc, CONFIGPARMS *pCfg)
 	return rc;
 }
 
-/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/
-/**
- *	mpt_toolbox - Generic function to issue toolbox message
- *	@ioc - Pointer to an adapter structure
- *	@cfg - Pointer to a toolbox structure. Struct contains
- *		action, page address, direction, physical address
- *		and pointer to a configuration page header
- *		Page header is updated.
- *
- *	Returns 0 for success
- *	-EPERM if not allowed due to ISR context
- *	-EAGAIN if no msg frames currently available
- *	-EFAULT for non-successful reply or no reply (timeout)
- */
-int
-mpt_toolbox(MPT_ADAPTER *ioc, CONFIGPARMS *pCfg)
-{
-	ToolboxIstwiReadWriteRequest_t	*pReq;
-	MPT_FRAME_HDR	*mf;
-	struct pci_dev	*pdev;
-	unsigned long	 flags;
-	int		 rc;
-	u32		 flagsLength;
-	int		 in_isr;
-
-	/*	Prevent calling wait_event() (below), if caller happens
-	 *	to be in ISR context, because that is fatal!
-	 */
-	in_isr = in_interrupt();
-	if (in_isr) {
-		dcprintk((MYIOC_s_WARN_FMT "toobox request not allowed in ISR context!\n",
-				ioc->name));
-		return -EPERM;
-	}
-
-	/* Get and Populate a free Frame
-	 */
-	if ((mf = mpt_get_msg_frame(mpt_base_index, ioc)) == NULL) {
-		dcprintk((MYIOC_s_WARN_FMT "mpt_toolbox: no msg frames!\n",
-				ioc->name));
-		return -EAGAIN;
-	}
-	pReq = (ToolboxIstwiReadWriteRequest_t	*)mf;
-	pReq->Tool = pCfg->action;
-	pReq->Reserved = 0;
-	pReq->ChainOffset = 0;
-	pReq->Function = MPI_FUNCTION_TOOLBOX;
-	pReq->Reserved1 = 0;
-	pReq->Reserved2 = 0;
-	pReq->MsgFlags = 0;
-	pReq->Flags = pCfg->dir;
-	pReq->BusNum = 0;
-	pReq->Reserved3 = 0;
-	pReq->NumAddressBytes = 0x01;
-	pReq->Reserved4 = 0;
-	pReq->DataLength = cpu_to_le16(0x04);
-	pdev = ioc->pcidev;
-	if (pdev->devfn & 1)
-		pReq->DeviceAddr = 0xB2;
-	else
-		pReq->DeviceAddr = 0xB0;
-	pReq->Addr1 = 0;
-	pReq->Addr2 = 0;
-	pReq->Addr3 = 0;
-	pReq->Reserved5 = 0;
-
-	/* Add a SGE to the config request.
-	 */
-
-	flagsLength = MPT_SGE_FLAGS_SSIMPLE_READ | 4;
-
-	mpt_add_sge((char *)&pReq->SGL, flagsLength, pCfg->physAddr);
-
-	dcprintk((MYIOC_s_INFO_FMT "Sending Toolbox request, Tool=%x\n",
-		ioc->name, pReq->Tool));
-
-	/* Append pCfg pointer to end of mf
-	 */
-	*((void **) (((u8 *) mf) + (ioc->req_sz - sizeof(void *)))) =  (void *) pCfg;
-
-	/* Initalize the timer
-	 */
-	init_timer(&pCfg->timer);
-	pCfg->timer.data = (unsigned long) ioc;
-	pCfg->timer.function = mpt_timer_expired;
-	pCfg->wait_done = 0;
-
-	/* Set the timer; ensure 10 second minimum */
-	if (pCfg->timeout < 10)
-		pCfg->timer.expires = jiffies + HZ*10;
-	else
-		pCfg->timer.expires = jiffies + HZ*pCfg->timeout;
-
-	/* Add to end of Q, set timer and then issue this command */
-	spin_lock_irqsave(&ioc->FreeQlock, flags);
-	list_add_tail(&pCfg->linkage, &ioc->configQ);
-	spin_unlock_irqrestore(&ioc->FreeQlock, flags);
-
-	add_timer(&pCfg->timer);
-	mpt_put_msg_frame(mpt_base_index, ioc, mf);
-	wait_event(mpt_waitq, pCfg->wait_done);
-
-	/* mf has been freed - do not access */
-
-	rc = pCfg->status;
-
-	return rc;
-}
-
 /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/
 /*
  *	mpt_timer_expired - Call back for timer process.
@@ -6540,7 +6430,6 @@ EXPORT_SYMBOL(mpt_lan_index);
 EXPORT_SYMBOL(mpt_stm_index);
 EXPORT_SYMBOL(mpt_HardResetHandler);
 EXPORT_SYMBOL(mpt_config);
-EXPORT_SYMBOL(mpt_toolbox);
 EXPORT_SYMBOL(mpt_findImVolumes);
 EXPORT_SYMBOL(mpt_read_ioc_pg_3);
 EXPORT_SYMBOL(mpt_alloc_fw_memory);
diff --git a/drivers/message/fusion/mptbase.h b/drivers/message/fusion/mptbase.h
index ea2649ecad1f..2e5377309cea 100644
--- a/drivers/message/fusion/mptbase.h
+++ b/drivers/message/fusion/mptbase.h
@@ -1026,7 +1026,6 @@ extern u32	 mpt_GetIocState(MPT_ADAPTER *ioc, int cooked);
 extern void	 mpt_print_ioc_summary(MPT_ADAPTER *ioc, char *buf, int *size, int len, int showlan);
 extern int	 mpt_HardResetHandler(MPT_ADAPTER *ioc, int sleepFlag);
 extern int	 mpt_config(MPT_ADAPTER *ioc, CONFIGPARMS *cfg);
-extern int	 mpt_toolbox(MPT_ADAPTER *ioc, CONFIGPARMS *cfg);
 extern void	 mpt_alloc_fw_memory(MPT_ADAPTER *ioc, int size);
 extern void	 mpt_free_fw_memory(MPT_ADAPTER *ioc);
 extern int	 mpt_findImVolumes(MPT_ADAPTER *ioc);
diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index be5fcd8db63b..2df3b8756545 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -2271,13 +2271,16 @@ mptctl_hp_hostinfo(unsigned long arg, unsigned int data_size)
 	hp_host_info_t	__user *uarg = (void __user *) arg;
 	MPT_ADAPTER		*ioc;
 	struct pci_dev		*pdev;
-	char			*pbuf;
+	char                    *pbuf=NULL;
 	dma_addr_t		buf_dma;
 	hp_host_info_t		karg;
 	CONFIGPARMS		cfg;
 	ConfigPageHeader_t	hdr;
 	int			iocnum;
 	int			rc, cim_rev;
+	ToolboxIstwiReadWriteRequest_t	*IstwiRWRequest;
+	MPT_FRAME_HDR		*mf = NULL;
+	MPIHeader_t		*mpi_hdr;
 
 	dctlprintk((": mptctl_hp_hostinfo called.\n"));
 	/* Reset long to int. Should affect IA64 and SPARC only
@@ -2413,20 +2416,67 @@ mptctl_hp_hostinfo(unsigned long arg, unsigned int data_size)
 		}
 	}
 
-	cfg.pageAddr = 0;
-	cfg.action = MPI_TOOLBOX_ISTWI_READ_WRITE_TOOL;
-	cfg.dir = MPI_TB_ISTWI_FLAGS_READ;
-	cfg.timeout = 10;
-	pbuf = pci_alloc_consistent(ioc->pcidev, 4, &buf_dma);
-	if (pbuf) {
-		cfg.physAddr = buf_dma;
-		if ((mpt_toolbox(ioc, &cfg)) == 0) {
-			karg.rsvd = *(u32 *)pbuf;
-		}
-		pci_free_consistent(ioc->pcidev, 4, pbuf, buf_dma);
-		pbuf = NULL;
+	/* 
+	 * Gather ISTWI(Industry Standard Two Wire Interface) Data
+	 */
+	if ((mf = mpt_get_msg_frame(mptctl_id, ioc)) == NULL) {
+		dfailprintk((MYIOC_s_WARN_FMT "%s, no msg frames!!\n",
+		    ioc->name,__FUNCTION__));
+		goto out;
 	}
 
+	IstwiRWRequest = (ToolboxIstwiReadWriteRequest_t *)mf;
+	mpi_hdr = (MPIHeader_t *) mf;
+	memset(IstwiRWRequest,0,sizeof(ToolboxIstwiReadWriteRequest_t));
+	IstwiRWRequest->Function = MPI_FUNCTION_TOOLBOX;
+	IstwiRWRequest->Tool = MPI_TOOLBOX_ISTWI_READ_WRITE_TOOL;
+	IstwiRWRequest->MsgContext = mpi_hdr->MsgContext;
+	IstwiRWRequest->Flags = MPI_TB_ISTWI_FLAGS_READ;
+	IstwiRWRequest->NumAddressBytes = 0x01;
+	IstwiRWRequest->DataLength = cpu_to_le16(0x04);
+	if (pdev->devfn & 1)
+		IstwiRWRequest->DeviceAddr = 0xB2;
+	else
+		IstwiRWRequest->DeviceAddr = 0xB0;
+
+	pbuf = pci_alloc_consistent(ioc->pcidev, 4, &buf_dma);
+	if (!pbuf)
+		goto out;
+	mpt_add_sge((char *)&IstwiRWRequest->SGL,
+	    (MPT_SGE_FLAGS_SSIMPLE_READ|4), buf_dma);
+
+	ioc->ioctl->wait_done = 0;
+	mpt_put_msg_frame(mptctl_id, ioc, mf);
+
+	rc = wait_event_timeout(mptctl_wait,
+	     ioc->ioctl->wait_done == 1,
+	     HZ*MPT_IOCTL_DEFAULT_TIMEOUT /* 10 sec */);
+
+	if(rc <=0 && (ioc->ioctl->wait_done != 1 )) {
+		/* 
+		 * Now we need to reset the board
+		 */
+		mpt_free_msg_frame(ioc, mf);
+		mptctl_timeout_expired(ioc->ioctl);
+		goto out;
+	}
+
+	/* 
+	 *ISTWI Data Definition
+	 * pbuf[0] = FW_VERSION = 0x4
+	 * pbuf[1] = Bay Count = 6 or 4 or 2, depending on
+	 *  the config, you should be seeing one out of these three values
+	 * pbuf[2] = Drive Installed Map = bit pattern depend on which
+	 *   bays have drives in them
+	 * pbuf[3] = Checksum (0x100 = (byte0 + byte2 + byte3)
+	 */
+	if (ioc->ioctl->status & MPT_IOCTL_STATUS_RF_VALID)
+		karg.rsvd = *(u32 *)pbuf;
+
+ out:
+	if (pbuf)
+		pci_free_consistent(ioc->pcidev, 4, pbuf, buf_dma);
+
 	/* Copy the data from kernel memory to user memory
 	 */
 	if (copy_to_user((char __user *)arg, &karg, sizeof(hp_host_info_t))) {

From 946cbf040adb9db05bb895a4b629537fd2d03b0e Mon Sep 17 00:00:00 2001
From: "Moore, Eric" <Eric.Moore@lsil.com>
Date: Thu, 2 Feb 2006 17:19:50 -0700
Subject: [PATCH 031/509] [SCSI] fusion - mptctl -firmware download fix

Fix's firmware download ioctl to work with SAS.

Signed-off-by: Eric Moore <Eric.Moore@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptctl.c | 35 +++++++++++++++++++--------------
 1 file changed, 20 insertions(+), 15 deletions(-)

diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index 2df3b8756545..fe10cc0fe18e 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -674,22 +674,23 @@ mptctl_do_fw_download(int ioc, char __user *ufwbuf, size_t fwlen)
 	u16			 iocstat;
 	pFWDownloadReply_t	 ReplyMsg = NULL;
 
-	dctlprintk((KERN_INFO "mptctl_do_fwdl called. mptctl_id = %xh.\n", mptctl_id));
+	dctlprintk(("mptctl_do_fwdl called. mptctl_id = %xh.\n", mptctl_id));
 
-	dctlprintk((KERN_INFO "DbG: kfwdl.bufp  = %p\n", ufwbuf));
-	dctlprintk((KERN_INFO "DbG: kfwdl.fwlen = %d\n", (int)fwlen));
-	dctlprintk((KERN_INFO "DbG: kfwdl.ioc   = %04xh\n", ioc));
+	dctlprintk(("DbG: kfwdl.bufp  = %p\n", ufwbuf));
+	dctlprintk(("DbG: kfwdl.fwlen = %d\n", (int)fwlen));
+	dctlprintk(("DbG: kfwdl.ioc   = %04xh\n", ioc));
 
-	if ((ioc = mpt_verify_adapter(ioc, &iocp)) < 0) {
-		dctlprintk(("%s@%d::_ioctl_fwdl - ioc%d not found!\n",
-				__FILE__, __LINE__, ioc));
+	if (mpt_verify_adapter(ioc, &iocp) < 0) {
+		dctlprintk(("ioctl_fwdl - ioc%d not found!\n",
+				 ioc));
 		return -ENODEV; /* (-6) No such device or address */
-	}
+	} else {
 
-	/*  Valid device. Get a message frame and construct the FW download message.
-	 */
-	if ((mf = mpt_get_msg_frame(mptctl_id, iocp)) == NULL)
-		return -EAGAIN;
+		/*  Valid device. Get a message frame and construct the FW download message.
+	 	*/
+		if ((mf = mpt_get_msg_frame(mptctl_id, iocp)) == NULL)
+			return -EAGAIN;
+	}
 	dlmsg = (FWDownload_t*) mf;
 	ptsge = (FWDownloadTCSGE_t *) &dlmsg->SGL;
 	sgOut = (char *) (ptsge + 1);
@@ -702,7 +703,11 @@ mptctl_do_fw_download(int ioc, char __user *ufwbuf, size_t fwlen)
 	dlmsg->ChainOffset = 0;
 	dlmsg->Function = MPI_FUNCTION_FW_DOWNLOAD;
 	dlmsg->Reserved1[0] = dlmsg->Reserved1[1] = dlmsg->Reserved1[2] = 0;
-	dlmsg->MsgFlags = 0;
+	if (iocp->facts.MsgVersion >= MPI_VERSION_01_05)
+		dlmsg->MsgFlags = MPI_FW_DOWNLOAD_MSGFLGS_LAST_SEGMENT;
+	else
+		dlmsg->MsgFlags = 0;
+
 
 	/* Set up the Transaction SGE.
 	 */
@@ -754,7 +759,7 @@ mptctl_do_fw_download(int ioc, char __user *ufwbuf, size_t fwlen)
 		goto fwdl_out;
 	}
 
-	dctlprintk((KERN_INFO "DbG: sgl buffer  = %p, sgfrags = %d\n", sgl, numfrags));
+	dctlprintk(("DbG: sgl buffer  = %p, sgfrags = %d\n", sgl, numfrags));
 
 	/*
 	 * Parse SG list, copying sgl itself,
@@ -803,7 +808,7 @@ mptctl_do_fw_download(int ioc, char __user *ufwbuf, size_t fwlen)
 	/*
 	 * Finally, perform firmware download.
 	 */
-	iocp->ioctl->wait_done = 0;
+	ReplyMsg = NULL;
 	mpt_put_msg_frame(mptctl_id, iocp, mf);
 
 	/* Now wait for the command to complete */

From ea5a7a82f9d2d8a81f8fa541c34a12b43d390f61 Mon Sep 17 00:00:00 2001
From: "Moore, Eric" <Eric.Moore@lsil.com>
Date: Thu, 2 Feb 2006 17:20:01 -0700
Subject: [PATCH 032/509] [SCSI] fusion - mptctl -adding asyn event
 notification support

Adding aen support.

Signed-off-by: Eric Moore <Eric.Moore@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/message/fusion/mptbase.h |  1 +
 drivers/message/fusion/mptctl.c  | 79 ++++++++++++++++++++++++++++++++
 2 files changed, 80 insertions(+)

diff --git a/drivers/message/fusion/mptbase.h b/drivers/message/fusion/mptbase.h
index 2e5377309cea..723d54300953 100644
--- a/drivers/message/fusion/mptbase.h
+++ b/drivers/message/fusion/mptbase.h
@@ -616,6 +616,7 @@ typedef struct _MPT_ADAPTER
 	 * increments by 32 bytes
 	 */
 	int			 errata_flag_1064;
+	int			 aen_event_read_flag; /* flag to indicate event log was read*/
 	u8			 FirstWhoInit;
 	u8			 upload_fw;	/* If set, do a fw upload */
 	u8			 reload_fw;	/* Force a FW Reload on next reset */
diff --git a/drivers/message/fusion/mptctl.c b/drivers/message/fusion/mptctl.c
index fe10cc0fe18e..9b64e07400da 100644
--- a/drivers/message/fusion/mptctl.c
+++ b/drivers/message/fusion/mptctl.c
@@ -136,6 +136,12 @@ static void mptctl_free_tm_flags(MPT_ADAPTER *ioc);
  */
 static int  mptctl_ioc_reset(MPT_ADAPTER *ioc, int reset_phase);
 
+/*
+ * Event Handler function
+ */
+static int mptctl_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *pEvReply);
+struct fasync_struct *async_queue=NULL;
+
 /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/
 /*
  * Scatter gather list (SGL) sizes and limits...
@@ -471,6 +477,69 @@ mptctl_ioc_reset(MPT_ADAPTER *ioc, int reset_phase)
 	return 1;
 }
 
+/*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/
+/* ASYNC Event Notification Support */
+static int
+mptctl_event_process(MPT_ADAPTER *ioc, EventNotificationReply_t *pEvReply)
+{
+	u8 event;
+
+	event = le32_to_cpu(pEvReply->Event) & 0xFF;
+
+	dctlprintk(("%s() called\n", __FUNCTION__));
+	if(async_queue == NULL)
+		return 1;
+
+	/* Raise SIGIO for persistent events.
+	 * TODO - this define is not in MPI spec yet,
+	 * but they plan to set it to 0x21
+	 */
+	 if (event == 0x21 ) {
+		ioc->aen_event_read_flag=1;
+		dctlprintk(("Raised SIGIO to application\n"));
+		devtprintk(("Raised SIGIO to application\n"));
+		kill_fasync(&async_queue, SIGIO, POLL_IN);
+		return 1;
+	 }
+
+	/* This flag is set after SIGIO was raised, and
+	 * remains set until the application has read
+	 * the event log via ioctl=MPTEVENTREPORT
+	 */
+	if(ioc->aen_event_read_flag)
+		return 1;
+
+	/* Signal only for the events that are
+	 * requested for by the application
+	 */
+	if (ioc->events && (ioc->eventTypes & ( 1 << event))) {
+		ioc->aen_event_read_flag=1;
+		dctlprintk(("Raised SIGIO to application\n"));
+		devtprintk(("Raised SIGIO to application\n"));
+		kill_fasync(&async_queue, SIGIO, POLL_IN);
+	}
+	return 1;
+}
+
+static int
+mptctl_fasync(int fd, struct file *filep, int mode)
+{
+	MPT_ADAPTER	*ioc;
+
+	list_for_each_entry(ioc, &ioc_list, list)
+		ioc->aen_event_read_flag=0;
+
+	dctlprintk(("%s() called\n", __FUNCTION__));
+	return fasync_helper(fd, filep, mode, &async_queue);
+}
+
+static int
+mptctl_release(struct inode *inode, struct file *filep)
+{
+	dctlprintk(("%s() called\n", __FUNCTION__));
+	return fasync_helper(-1, filep, 0, &async_queue);
+}
+
 /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/
 /*
  *  MPT ioctl handler
@@ -1603,6 +1672,9 @@ mptctl_eventreport (unsigned long arg)
 	if ((max < 1) || !ioc->events)
 		return -ENODATA;
 
+	/* reset this flag so SIGIO can restart */
+	ioc->aen_event_read_flag=0;
+
 	/* Copy the data from kernel memory to user memory
 	 */
 	numBytes = max * sizeof(MPT_IOCTL_EVENTS);
@@ -2649,6 +2721,8 @@ mptctl_hp_targetinfo(unsigned long arg)
 static struct file_operations mptctl_fops = {
 	.owner =	THIS_MODULE,
 	.llseek =	no_llseek,
+	.release =	mptctl_release,
+	.fasync = 	mptctl_fasync,
 	.unlocked_ioctl = mptctl_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl = compat_mpctl_ioctl,
@@ -2893,6 +2967,11 @@ static int __init mptctl_init(void)
 		/* FIXME! */
 	}
 
+	if (mpt_event_register(mptctl_id, mptctl_event_process) == 0) {
+		devtprintk((KERN_INFO MYNAM
+		  ": Registered for IOC event notifications\n"));
+	}
+
 	return 0;
 
 out_fail:

From 122da30223c06cee181044af6d32e88b256d10df Mon Sep 17 00:00:00 2001
From: Joshua Giles <joshua_giles@dell.com>
Date: Fri, 3 Feb 2006 15:34:17 -0800
Subject: [PATCH 033/509] [SCSI] megaraid_sas: register 16 byte CDB capability

This patch properly registers the 16 byte command length capability of
the megaraid_sas controlled hardware with the scsi midlayer.  All
megaraid_sas hardware supports 16 byte CDB's.

Signed-off-by: Joshua Giles <joshua_giles@dell.com>
Signed-off-by: Sumant Patro <Sumant.Patro@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 Documentation/scsi/ChangeLog.megaraid_sas | 12 ++++++++++++
 drivers/scsi/megaraid/megaraid_sas.c      |  3 ++-
 drivers/scsi/megaraid/megaraid_sas.h      |  6 +++---
 3 files changed, 17 insertions(+), 4 deletions(-)

diff --git a/Documentation/scsi/ChangeLog.megaraid_sas b/Documentation/scsi/ChangeLog.megaraid_sas
index f8c16cbf56ba..9e8085bd2e42 100644
--- a/Documentation/scsi/ChangeLog.megaraid_sas
+++ b/Documentation/scsi/ChangeLog.megaraid_sas
@@ -1,3 +1,15 @@
+1 Release Date    : Fri Feb 03 14:16:25 PST 2006 - Sumant Patro 
+							<Sumant.Patro@lsil.com>
+2 Current Version : 00.00.02.04
+3 Older Version   : 00.00.02.02 
+i.	Register 16 byte CDB capability with scsi midlayer 
+
+	"Ths patch properly registers the 16 byte command length capability of the 
+	megaraid_sas controlled hardware with the scsi midlayer. All megaraid_sas 
+	hardware supports 16 byte CDB's."
+
+		-Joshua Giles <joshua_giles@dell.com> 
+
 1 Release Date    : Mon Jan 23 14:09:01 PST 2006 - Sumant Patro <Sumant.Patro@lsil.com>
 2 Current Version : 00.00.02.02
 3 Older Version   : 00.00.02.01 
diff --git a/drivers/scsi/megaraid/megaraid_sas.c b/drivers/scsi/megaraid/megaraid_sas.c
index a487f414960e..0b738243782e 100644
--- a/drivers/scsi/megaraid/megaraid_sas.c
+++ b/drivers/scsi/megaraid/megaraid_sas.c
@@ -10,7 +10,7 @@
  *	   2 of the License, or (at your option) any later version.
  *
  * FILE		: megaraid_sas.c
- * Version	: v00.00.02.02
+ * Version	: v00.00.02.04
  *
  * Authors:
  * 	Sreenivas Bagalkote	<Sreenivas.Bagalkote@lsil.com>
@@ -1983,6 +1983,7 @@ static int megasas_io_attach(struct megasas_instance *instance)
 	host->max_channel = MEGASAS_MAX_CHANNELS - 1;
 	host->max_id = MEGASAS_MAX_DEV_PER_CHANNEL;
 	host->max_lun = MEGASAS_MAX_LUN;
+	host->max_cmd_len = 16;
 
 	/*
 	 * Notify the mid-layer about the new controller
diff --git a/drivers/scsi/megaraid/megaraid_sas.h b/drivers/scsi/megaraid/megaraid_sas.h
index d6d166c0663f..917326f3770e 100644
--- a/drivers/scsi/megaraid/megaraid_sas.h
+++ b/drivers/scsi/megaraid/megaraid_sas.h
@@ -18,9 +18,9 @@
 /**
  * MegaRAID SAS Driver meta data
  */
-#define MEGASAS_VERSION				"00.00.02.02"
-#define MEGASAS_RELDATE				"Jan 23, 2006"
-#define MEGASAS_EXT_VERSION			"Mon Jan 23 14:09:01 PST 2006"
+#define MEGASAS_VERSION				"00.00.02.04"
+#define MEGASAS_RELDATE				"Feb 03, 2006"
+#define MEGASAS_EXT_VERSION			"Fri Feb 03 14:16:25 PST 2006"
 /*
  * =====================================
  * MegaRAID SAS MFI firmware definitions

From f9876f0b67c3f0b04ee2167602df54e7ae139ad7 Mon Sep 17 00:00:00 2001
From: Sumant Patro <sumantp@lsil.com>
Date: Fri, 3 Feb 2006 15:34:35 -0800
Subject: [PATCH 034/509] [SCSI] megaraid_sas: support for 1078 type controller
 added

This patch adds support for 1078 type controller (device id : 0x60).

Signed-off-by: Sumant Patro <Sumant.Patro@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 Documentation/scsi/ChangeLog.megaraid_sas | 11 +++
 drivers/scsi/megaraid/megaraid_sas.c      | 98 ++++++++++++++++++++++-
 drivers/scsi/megaraid/megaraid_sas.h      | 49 ++++++++----
 3 files changed, 140 insertions(+), 18 deletions(-)

diff --git a/Documentation/scsi/ChangeLog.megaraid_sas b/Documentation/scsi/ChangeLog.megaraid_sas
index 9e8085bd2e42..2dafa63bd370 100644
--- a/Documentation/scsi/ChangeLog.megaraid_sas
+++ b/Documentation/scsi/ChangeLog.megaraid_sas
@@ -1,3 +1,14 @@
+1 Release Date    : Wed Feb 03 14:31:44 PST 2006 - Sumant Patro <Sumant.Patro@lsil.com>
+2 Current Version : 00.00.02.04
+3 Older Version   : 00.00.02.04 
+
+i.	Support for 1078 type (ppc IOP) controller, device id : 0x60 added.
+	During initialization, depending on the device id, the template members 
+	are initialized with function pointers specific to the ppc or 
+	xscale controllers.  
+
+		-Sumant Patro <Sumant.Patro@lsil.com>
+		
 1 Release Date    : Fri Feb 03 14:16:25 PST 2006 - Sumant Patro 
 							<Sumant.Patro@lsil.com>
 2 Current Version : 00.00.02.04
diff --git a/drivers/scsi/megaraid/megaraid_sas.c b/drivers/scsi/megaraid/megaraid_sas.c
index 0b738243782e..7de267e14458 100644
--- a/drivers/scsi/megaraid/megaraid_sas.c
+++ b/drivers/scsi/megaraid/megaraid_sas.c
@@ -59,6 +59,12 @@ static struct pci_device_id megasas_pci_table[] = {
 	 PCI_ANY_ID,
 	 PCI_ANY_ID,
 	 },
+	{
+	 PCI_VENDOR_ID_LSI_LOGIC,
+	 PCI_DEVICE_ID_LSI_SAS1078R, // ppc IOP
+	 PCI_ANY_ID,
+	 PCI_ANY_ID,
+	},
 	{
 	 PCI_VENDOR_ID_DELL,
 	 PCI_DEVICE_ID_DELL_PERC5, // xscale IOP
@@ -198,6 +204,86 @@ static struct megasas_instance_template megasas_instance_template_xscale = {
 *	to xscale (deviceid : 1064R, PERC5) controllers
 */
 
+/**
+*	The following functions are defined for ppc (deviceid : 0x60) 
+* 	controllers
+*/
+
+/**
+ * megasas_enable_intr_ppc -	Enables interrupts
+ * @regs:			MFI register set
+ */
+static inline void
+megasas_enable_intr_ppc(struct megasas_register_set __iomem * regs)
+{
+	writel(0xFFFFFFFF, &(regs)->outbound_doorbell_clear);
+    
+	writel(~0x80000004, &(regs)->outbound_intr_mask);
+
+	/* Dummy readl to force pci flush */
+	readl(&regs->outbound_intr_mask);
+}
+
+/**
+ * megasas_read_fw_status_reg_ppc - returns the current FW status value
+ * @regs:			MFI register set
+ */
+static u32
+megasas_read_fw_status_reg_ppc(struct megasas_register_set __iomem * regs)
+{
+	return readl(&(regs)->outbound_scratch_pad);
+}
+
+/**
+ * megasas_clear_interrupt_ppc -	Check & clear interrupt
+ * @regs:				MFI register set
+ */
+static int 
+megasas_clear_intr_ppc(struct megasas_register_set __iomem * regs)
+{
+	u32 status;
+	/*
+	 * Check if it is our interrupt
+	 */
+	status = readl(&regs->outbound_intr_status);
+
+	if (!(status & MFI_REPLY_1078_MESSAGE_INTERRUPT)) {
+		return 1;
+	}
+
+	/*
+	 * Clear the interrupt by writing back the same value
+	 */
+	writel(status, &regs->outbound_doorbell_clear);
+
+	return 0;
+}
+/**
+ * megasas_fire_cmd_ppc -	Sends command to the FW
+ * @frame_phys_addr :		Physical address of cmd
+ * @frame_count :		Number of frames for the command
+ * @regs :			MFI register set
+ */
+static inline void 
+megasas_fire_cmd_ppc(dma_addr_t frame_phys_addr, u32 frame_count, struct megasas_register_set __iomem *regs)
+{
+	writel((frame_phys_addr | (frame_count<<1))|1, 
+			&(regs)->inbound_queue_port);
+}
+
+static struct megasas_instance_template megasas_instance_template_ppc = {
+	
+	.fire_cmd = megasas_fire_cmd_ppc,
+	.enable_intr = megasas_enable_intr_ppc,
+	.clear_intr = megasas_clear_intr_ppc,
+	.read_fw_status_reg = megasas_read_fw_status_reg_ppc,
+};
+
+/**
+*	This is the end of set of functions & definitions
+* 	specific to ppc (deviceid : 0x60) controllers
+*/
+
 /**
  * megasas_disable_intr -	Disables interrupts
  * @regs:			MFI register set
@@ -1607,7 +1693,17 @@ static int megasas_init_mfi(struct megasas_instance *instance)
 
 	reg_set = instance->reg_set;
 
-	instance->instancet = &megasas_instance_template_xscale;
+	switch(instance->pdev->device)
+	{
+		case PCI_DEVICE_ID_LSI_SAS1078R:	
+			instance->instancet = &megasas_instance_template_ppc;
+			break;
+		case PCI_DEVICE_ID_LSI_SAS1064R:
+		case PCI_DEVICE_ID_DELL_PERC5:
+		default:
+			instance->instancet = &megasas_instance_template_xscale;
+			break;
+	}
 
 	/*
 	 * We expect the FW state to be READY
diff --git a/drivers/scsi/megaraid/megaraid_sas.h b/drivers/scsi/megaraid/megaraid_sas.h
index 917326f3770e..89639f0c38ef 100644
--- a/drivers/scsi/megaraid/megaraid_sas.h
+++ b/drivers/scsi/megaraid/megaraid_sas.h
@@ -20,7 +20,7 @@
  */
 #define MEGASAS_VERSION				"00.00.02.04"
 #define MEGASAS_RELDATE				"Feb 03, 2006"
-#define MEGASAS_EXT_VERSION			"Fri Feb 03 14:16:25 PST 2006"
+#define MEGASAS_EXT_VERSION			"Fri Feb 03 14:31:44 PST 2006"
 /*
  * =====================================
  * MegaRAID SAS MFI firmware definitions
@@ -553,31 +553,46 @@ struct megasas_ctrl_info {
 #define MFI_OB_INTR_STATUS_MASK			0x00000002
 #define MFI_POLL_TIMEOUT_SECS			10
 
+#define MFI_REPLY_1078_MESSAGE_INTERRUPT	0x80000000
+#define PCI_DEVICE_ID_LSI_SAS1078R		0x00000060
+ 
 struct megasas_register_set {
+	u32 	reserved_0[4];			/*0000h*/
 
-	u32 reserved_0[4];	/*0000h */
+	u32 	inbound_msg_0;			/*0010h*/
+	u32 	inbound_msg_1;			/*0014h*/
+	u32 	outbound_msg_0;			/*0018h*/
+	u32 	outbound_msg_1;			/*001Ch*/
 
-	u32 inbound_msg_0;	/*0010h */
-	u32 inbound_msg_1;	/*0014h */
-	u32 outbound_msg_0;	/*0018h */
-	u32 outbound_msg_1;	/*001Ch */
+	u32 	inbound_doorbell;		/*0020h*/
+	u32 	inbound_intr_status;		/*0024h*/
+	u32 	inbound_intr_mask;		/*0028h*/
 
-	u32 inbound_doorbell;	/*0020h */
-	u32 inbound_intr_status;	/*0024h */
-	u32 inbound_intr_mask;	/*0028h */
+	u32 	outbound_doorbell;		/*002Ch*/
+	u32 	outbound_intr_status;		/*0030h*/
+	u32 	outbound_intr_mask;		/*0034h*/
 
-	u32 outbound_doorbell;	/*002Ch */
-	u32 outbound_intr_status;	/*0030h */
-	u32 outbound_intr_mask;	/*0034h */
+	u32 	reserved_1[2];			/*0038h*/
 
-	u32 reserved_1[2];	/*0038h */
+	u32 	inbound_queue_port;		/*0040h*/
+	u32 	outbound_queue_port;		/*0044h*/
 
-	u32 inbound_queue_port;	/*0040h */
-	u32 outbound_queue_port;	/*0044h */
+	u32 	reserved_2[22];			/*0048h*/
 
-	u32 reserved_2;		/*004Ch */
+	u32 	outbound_doorbell_clear;	/*00A0h*/
 
-	u32 index_registers[1004];	/*0050h */
+	u32 	reserved_3[3];			/*00A4h*/
+
+	u32 	outbound_scratch_pad ;		/*00B0h*/
+
+	u32 	reserved_4[3];			/*00B4h*/
+
+	u32 	inbound_low_queue_port ;	/*00C0h*/
+
+	u32 	inbound_high_queue_port ;	/*00C4h*/
+
+	u32 	reserved_5;			/*00C8h*/
+	u32 	index_registers[820];		/*00CCh*/
 
 } __attribute__ ((packed));
 

From 40cdc840dc337cb17d81bcf028b40834e78c1038 Mon Sep 17 00:00:00 2001
From: Jenx Axboe <axboe@suse.de>
Date: Sun, 5 Feb 2006 16:36:23 +0100
Subject: [PATCH 035/509] [SCSI] gdth: don't map zero-length requests

Don't map zero-length requests in gdth, zome architectures don't like
that in their dma mapping routines.

[ I'm pretty sure Jens posted this before, but for some reason it got
  forgotten  --hch ]

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/gdth.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/gdth.c b/drivers/scsi/gdth.c
index bd3ffdf6c800..62e3cda859af 100644
--- a/drivers/scsi/gdth.c
+++ b/drivers/scsi/gdth.c
@@ -2816,7 +2816,7 @@ static int gdth_fill_cache_cmd(int hanum,Scsi_Cmnd *scp,ushort hdrive)
             }
 #endif
 
-        } else {
+        } else if (scp->request_bufflen) {
             scp->SCp.Status = GDTH_MAP_SINGLE;
             scp->SCp.Message = (read_write == 1 ? 
                 PCI_DMA_TODEVICE : PCI_DMA_FROMDEVICE);

From 64419d93a5906600af5817ad0cae3c6ecf7fb389 Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Sun, 5 Feb 2006 21:43:57 +0000
Subject: [PATCH 036/509] NTFS: We have struct kmem_cache now so use it instead
 of the typedef.

Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
---
 fs/ntfs/ntfs.h  | 10 +++++-----
 fs/ntfs/super.c | 12 ++++++------
 2 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/fs/ntfs/ntfs.h b/fs/ntfs/ntfs.h
index 446b5014115c..653d2a5c4899 100644
--- a/fs/ntfs/ntfs.h
+++ b/fs/ntfs/ntfs.h
@@ -50,11 +50,11 @@ typedef enum {
 /* Global variables. */
 
 /* Slab caches (from super.c). */
-extern kmem_cache_t *ntfs_name_cache;
-extern kmem_cache_t *ntfs_inode_cache;
-extern kmem_cache_t *ntfs_big_inode_cache;
-extern kmem_cache_t *ntfs_attr_ctx_cache;
-extern kmem_cache_t *ntfs_index_ctx_cache;
+extern struct kmem_cache *ntfs_name_cache;
+extern struct kmem_cache *ntfs_inode_cache;
+extern struct kmem_cache *ntfs_big_inode_cache;
+extern struct kmem_cache *ntfs_attr_ctx_cache;
+extern struct kmem_cache *ntfs_index_ctx_cache;
 
 /* The various operations structs defined throughout the driver files. */
 extern struct address_space_operations ntfs_aops;
diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c
index c3a3f1a8310b..e9c0d80dfab1 100644
--- a/fs/ntfs/super.c
+++ b/fs/ntfs/super.c
@@ -2987,14 +2987,14 @@ err_out_now:
  * strings of the maximum length allowed by NTFS, which is NTFS_MAX_NAME_LEN
  * (255) Unicode characters + a terminating NULL Unicode character.
  */
-kmem_cache_t *ntfs_name_cache;
+struct kmem_cache *ntfs_name_cache;
 
 /* Slab caches for efficient allocation/deallocation of inodes. */
-kmem_cache_t *ntfs_inode_cache;
-kmem_cache_t *ntfs_big_inode_cache;
+struct kmem_cache *ntfs_inode_cache;
+struct kmem_cache *ntfs_big_inode_cache;
 
 /* Init once constructor for the inode slab cache. */
-static void ntfs_big_inode_init_once(void *foo, kmem_cache_t *cachep,
+static void ntfs_big_inode_init_once(void *foo, struct kmem_cache *cachep,
 		unsigned long flags)
 {
 	ntfs_inode *ni = (ntfs_inode *)foo;
@@ -3008,8 +3008,8 @@ static void ntfs_big_inode_init_once(void *foo, kmem_cache_t *cachep,
  * Slab caches to optimize allocations and deallocations of attribute search
  * contexts and index contexts, respectively.
  */
-kmem_cache_t *ntfs_attr_ctx_cache;
-kmem_cache_t *ntfs_index_ctx_cache;
+struct kmem_cache *ntfs_attr_ctx_cache;
+struct kmem_cache *ntfs_index_ctx_cache;
 
 /* Driver wide semaphore. */
 DECLARE_MUTEX(ntfs_lock);

From 62288f105b3cad0b8643526d2a41b5503d0a1476 Mon Sep 17 00:00:00 2001
From: adam radford <aradford@gmail.com>
Date: Sun, 5 Feb 2006 14:51:43 -0800
Subject: [PATCH 037/509] [SCSI] 3ware 9000 driver >4GB memory fix

The attached patch fixes a bug in the 3ware 9000 series driver:

- Fix use_sg == 0 mapping on systems with 4GB or higher.

  This fixes REPORT_LUNS (0xa0) failing with 3ware 9000 controllers on systems
  with lots of ram, mentioned in bugzilla # 6009:

      http://bugzilla.kernel.org/show_bug.cgi?id=6009

Signed-off-by: Adam Radford <linuxraid@amcc.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/3w-9xxx.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/3w-9xxx.c b/drivers/scsi/3w-9xxx.c
index 31c497542272..d9152d02088c 100644
--- a/drivers/scsi/3w-9xxx.c
+++ b/drivers/scsi/3w-9xxx.c
@@ -61,6 +61,7 @@
                  Add support for embedded firmware error strings.
    2.26.02.003 - Correctly handle single sgl's with use_sg=1.
    2.26.02.004 - Add support for 9550SX controllers.
+   2.26.02.005 - Fix use_sg == 0 mapping on systems with 4GB or higher.
 */
 
 #include <linux/module.h>
@@ -84,7 +85,7 @@
 #include "3w-9xxx.h"
 
 /* Globals */
-#define TW_DRIVER_VERSION "2.26.02.004"
+#define TW_DRIVER_VERSION "2.26.02.005"
 static TW_Device_Extension *twa_device_extension_list[TW_MAX_SLOT];
 static unsigned int twa_device_extension_count;
 static int twa_major = -1;
@@ -1408,7 +1409,7 @@ static dma_addr_t twa_map_scsi_single_data(TW_Device_Extension *tw_dev, int requ
 	dma_addr_t mapping;
 	struct scsi_cmnd *cmd = tw_dev->srb[request_id];
 	struct pci_dev *pdev = tw_dev->tw_pci_dev;
-	int retval = 0;
+	dma_addr_t retval = 0;
 
 	if (cmd->request_bufflen == 0) {
 		retval = 0;
@@ -1798,7 +1799,7 @@ static int twa_scsiop_execute_scsi(TW_Device_Extension *tw_dev, int request_id,
 	int i, sg_count;
 	struct scsi_cmnd *srb = NULL;
 	struct scatterlist *sglist = NULL;
-	u32 buffaddr = 0x0;
+	dma_addr_t buffaddr = 0x0;
 	int retval = 1;
 
 	if (tw_dev->srb[request_id]) {

From 387f96b4d9391bf3ce6928fb9cd90c9c7df37291 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 7 Feb 2006 08:45:45 -0800
Subject: [PATCH 038/509] [PATCH] qla2xxx: Close window on race between rport
 removal and fcport transition.

Fcport visibility is recognized during interrupt time, but,
rport removal can only occur during a process
(sleeping)-context.  Return a DID_IMM_RETRY status for
commands submitted within this window to insure I/Os do not
prematurely run-out of retries.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_os.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index 495ccbc7f8cb..9b5a10ac3e25 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -366,6 +366,12 @@ qla2x00_queuecommand(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *))
 		goto qc_fail_command;
 	}
 
+	/* Close window on fcport/rport state-transitioning. */
+	if (!*(fc_port_t **)rport->dd_data) {
+		cmd->result = DID_IMM_RETRY << 16;
+		goto qc_fail_command;
+	}
+
 	if (atomic_read(&fcport->state) != FCS_ONLINE) {
 		if (atomic_read(&fcport->state) == FCS_DEVICE_DEAD ||
 		    atomic_read(&ha->loop_state) == LOOP_DEAD) {
@@ -421,6 +427,12 @@ qla24xx_queuecommand(struct scsi_cmnd *cmd, void (*done)(struct scsi_cmnd *))
 		goto qc24_fail_command;
 	}
 
+	/* Close window on fcport/rport state-transitioning. */
+	if (!*(fc_port_t **)rport->dd_data) {
+		cmd->result = DID_IMM_RETRY << 16;
+		goto qc24_fail_command;
+	}
+
 	if (atomic_read(&fcport->state) != FCS_ONLINE) {
 		if (atomic_read(&fcport->state) == FCS_DEVICE_DEAD ||
 		    atomic_read(&ha->loop_state) == LOOP_DEAD) {
@@ -1675,11 +1687,13 @@ qla2x00_schedule_rport_del(struct scsi_qla_host *ha, fc_port_t *fcport,
 		spin_lock_irqsave(&fcport->rport_lock, flags);
 		fcport->drport = rport;
 		fcport->rport = NULL;
+		*(fc_port_t **)rport->dd_data = NULL;
 		spin_unlock_irqrestore(&fcport->rport_lock, flags);
 		set_bit(FCPORT_UPDATE_NEEDED, &ha->dpc_flags);
 	} else {
 		spin_lock_irqsave(&fcport->rport_lock, flags);
 		fcport->rport = NULL;
+		*(fc_port_t **)rport->dd_data = NULL;
 		spin_unlock_irqrestore(&fcport->rport_lock, flags);
 		fc_remote_port_delete(rport);
 	}

From f7757a5f0fa94a0ce4d6ea55f440845bb3faf712 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 7 Feb 2006 08:45:30 -0800
Subject: [PATCH 039/509] [PATCH] qla2xxx: Remove bogus debug-code.

Commit 854165f4245c4a3b4a8cc363ba2050033151e196
inadvertently added some code meant only for testing -- the
driver was ignoring the non-zero function numbers of a
multi-port HBA.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_os.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index 9b5a10ac3e25..9f91f1a20542 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -1283,9 +1283,6 @@ int qla2x00_probe_one(struct pci_dev *pdev, struct qla_board_info *brd_info)
 	fc_port_t *fcport;
 	struct scsi_host_template *sht;
 
-if (PCI_FUNC(pdev->devfn))
-	goto probe_out;
-
 	if (pci_enable_device(pdev))
 		goto probe_out;
 

From 247ec457ce108000987be517c83868f6361d9f76 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 7 Feb 2006 08:45:40 -0800
Subject: [PATCH 040/509] [PATCH] qla2xxx: Pass input-buffer length to
 Get-ID-List mailbox command.

Recent ISP24xx firmwares require that mailbox register 8 be
set to the maximum number of bytes to transfer during DMA
copying of the list.  We safely set this value to zero
(infinite), since the call is *only* made in FCAL
topologies.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_mbx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
index 49ce197876b4..363dfdd042b0 100644
--- a/drivers/scsi/qla2xxx/qla_mbx.c
+++ b/drivers/scsi/qla2xxx/qla_mbx.c
@@ -1873,7 +1873,8 @@ qla2x00_get_id_list(scsi_qla_host_t *ha, void *id_list, dma_addr_t id_list_dma,
 		mcp->mb[3] = LSW(id_list_dma);
 		mcp->mb[6] = MSW(MSD(id_list_dma));
 		mcp->mb[7] = LSW(MSD(id_list_dma));
-		mcp->out_mb |= MBX_7|MBX_6|MBX_3|MBX_2;
+		mcp->mb[8] = 0;
+		mcp->out_mb |= MBX_8|MBX_7|MBX_6|MBX_3|MBX_2;
 	} else {
 		mcp->mb[1] = MSW(id_list_dma);
 		mcp->mb[2] = LSW(id_list_dma);

From 0d4be1240b2668b6a3ffadb15eb660baf52f8377 Mon Sep 17 00:00:00 2001
From: "andrew.vasquez@qlogic.com" <andrew.vasquez@qlogic.com>
Date: Tue, 7 Feb 2006 08:45:35 -0800
Subject: [PATCH 041/509] [PATCH] qla2xxx: Correct lun assignment during IOCB
 submission.

4gb products require an IOCB's FCP-LUN to be formatted in
wire-format prior to submission.

Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/qla2xxx/qla_iocb.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/scsi/qla2xxx/qla_iocb.c b/drivers/scsi/qla2xxx/qla_iocb.c
index 7ec0b8d6f07b..6544b6d0891d 100644
--- a/drivers/scsi/qla2xxx/qla_iocb.c
+++ b/drivers/scsi/qla2xxx/qla_iocb.c
@@ -814,6 +814,7 @@ qla24xx_start_scsi(srb_t *sp)
 	cmd_pkt->port_id[2] = sp->fcport->d_id.b.domain;
 
 	int_to_scsilun(sp->cmd->device->lun, &cmd_pkt->lun);
+	host_to_fcp_swap((uint8_t *)&cmd_pkt->lun, sizeof(cmd_pkt->lun));
 
 	/* Load SCSI command packet. */
 	memcpy(cmd_pkt->fcp_cdb, cmd->cmnd, cmd->cmd_len);

From 479a079663bd4c5f3d2714643b1b8c406aaba3e0 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@mellanox.co.il>
Date: Tue, 7 Feb 2006 16:37:08 -0800
Subject: [PATCH 042/509] IPoIB: Don't start send-only joins while multicast
 thread is stopped

Fix the following race scenario:
  - Device is up.
  - Port event or set mcast list triggers ipoib_mcast_stop_thread,
    this cancels the query and waits on mcast "done" completion.
  - Completion is called and "done" is set.
  - Meanwhile, ipoib_mcast_send arrives and starts a new query,
    re-initializing "done".

Fix this by adding a "multicast started" bit and checking it before
starting a send-only join.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/ulp/ipoib/ipoib.h           |  1 +
 drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 15 +++++++++++++++
 2 files changed, 16 insertions(+)

diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h
index e0a5412b7e68..2f85a9a831b1 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -78,6 +78,7 @@ enum {
 	IPOIB_FLAG_SUBINTERFACE   = 4,
 	IPOIB_MCAST_RUN 	  = 5,
 	IPOIB_STOP_REAPER         = 6,
+	IPOIB_MCAST_STARTED       = 7,
 
 	IPOIB_MAX_BACKOFF_SECONDS = 16,
 
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index ccaa0c387076..1c7148208ce8 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -601,6 +601,10 @@ int ipoib_mcast_start_thread(struct net_device *dev)
 		queue_work(ipoib_workqueue, &priv->mcast_task);
 	mutex_unlock(&mcast_mutex);
 
+	spin_lock_irq(&priv->lock);
+	set_bit(IPOIB_MCAST_STARTED, &priv->flags);
+	spin_unlock_irq(&priv->lock);
+
 	return 0;
 }
 
@@ -611,6 +615,10 @@ int ipoib_mcast_stop_thread(struct net_device *dev, int flush)
 
 	ipoib_dbg_mcast(priv, "stopping multicast thread\n");
 
+	spin_lock_irq(&priv->lock);
+	clear_bit(IPOIB_MCAST_STARTED, &priv->flags);
+	spin_unlock_irq(&priv->lock);
+
 	mutex_lock(&mcast_mutex);
 	clear_bit(IPOIB_MCAST_RUN, &priv->flags);
 	cancel_delayed_work(&priv->mcast_task);
@@ -693,6 +701,12 @@ void ipoib_mcast_send(struct net_device *dev, union ib_gid *mgid,
 	 */
 	spin_lock(&priv->lock);
 
+	if (!test_bit(IPOIB_MCAST_STARTED, &priv->flags)) {
+		++priv->stats.tx_dropped;
+		dev_kfree_skb_any(skb);
+		goto unlock;
+	}
+
 	mcast = __ipoib_mcast_find(dev, mgid);
 	if (!mcast) {
 		/* Let's create a new send only group now */
@@ -754,6 +768,7 @@ out:
 		ipoib_send(dev, skb, mcast->ah, IB_MULTICAST_QPN);
 	}
 
+unlock:
 	spin_unlock(&priv->lock);
 }
 

From 7bcb974ef6a0ae903888272c92c66ea779388c01 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@mellanox.co.il>
Date: Tue, 7 Feb 2006 16:39:26 -0800
Subject: [PATCH 043/509] IPoIB: Fix another send-only join race

Further, there's an additional issue that I saw in testing:
ipoib_mcast_send may get called when priv->broadcast is NULL (e.g. if
the device was downed and then upped internally because of a port
event).

If this happends and the send-only join request gets completed before
priv->broadcast is set, we get an oops.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 1c7148208ce8..932bf139a65d 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -701,7 +701,7 @@ void ipoib_mcast_send(struct net_device *dev, union ib_gid *mgid,
 	 */
 	spin_lock(&priv->lock);
 
-	if (!test_bit(IPOIB_MCAST_STARTED, &priv->flags)) {
+	if (!test_bit(IPOIB_MCAST_STARTED, &priv->flags) || !priv->broadcast) {
 		++priv->stats.tx_dropped;
 		dev_kfree_skb_any(skb);
 		goto unlock;

From f823bcae2b9f194cfc164b8cbb87d71695dec563 Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@parisc-linux.org>
Date: Sun, 5 Feb 2006 20:37:53 -0700
Subject: [PATCH 044/509] [PARISC] Convert ccio-dma.c to use seq_file

Gut ccio-dma.c of the ugly proc append and snprintf cruft and
just use seq_printf instead. Tested on a K-class.

Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
---
 drivers/parisc/ccio-dma.c | 160 ++++++++++++++++++--------------------
 1 file changed, 74 insertions(+), 86 deletions(-)

diff --git a/drivers/parisc/ccio-dma.c b/drivers/parisc/ccio-dma.c
index f46e8438e0d2..93f8a8fa8890 100644
--- a/drivers/parisc/ccio-dma.c
+++ b/drivers/parisc/ccio-dma.c
@@ -40,6 +40,8 @@
 #include <linux/string.h>
 #include <linux/pci.h>
 #include <linux/reboot.h>
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
 
 #include <asm/byteorder.h>
 #include <asm/cache.h>		/* for L1_CACHE_BYTES */
@@ -1019,62 +1021,33 @@ static struct hppa_dma_ops ccio_ops = {
 };
 
 #ifdef CONFIG_PROC_FS
-static int proc_append(char *src, int len, char **dst, off_t *offset, int *max)
+static int ccio_proc_info(struct seq_file *m, void *p)
 {
-	if (len < *offset) {
-		*offset -= len;
-		return 0;
-	}
-	if (*offset > 0) {
-		src += *offset;
-		len -= *offset;
-		*offset = 0;
-	}
-	if (len > *max) {
-		len = *max;
-	}
-	memcpy(*dst, src, len);
-	*dst += len;
-	*max -= len;
-	return (*max == 0);
-}
-
-static int ccio_proc_info(char *buf, char **start, off_t offset, int count,
-			  int *eof, void *data)
-{
-	int max = count;
-	char tmp[80]; /* width of an ANSI-standard terminal */
+	int len = 0;
 	struct ioc *ioc = ioc_list;
 
 	while (ioc != NULL) {
 		unsigned int total_pages = ioc->res_size << 3;
 		unsigned long avg = 0, min, max;
-		int j, len;
+		int j;
 
-		len = sprintf(tmp, "%s\n", ioc->name);
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
+		len += seq_printf(m, "%s\n", ioc->name);
 		
-		len = sprintf(tmp, "Cujo 2.0 bug    : %s\n",
-			      (ioc->cujo20_bug ? "yes" : "no"));
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
+		len += seq_printf(m, "Cujo 2.0 bug    : %s\n",
+				  (ioc->cujo20_bug ? "yes" : "no"));
 		
-		len = sprintf(tmp, "IO PDIR size    : %d bytes (%d entries)\n",
-			      total_pages * 8, total_pages);
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
+		len += seq_printf(m, "IO PDIR size    : %d bytes (%d entries)\n",
+			       total_pages * 8, total_pages);
+
 #ifdef CCIO_MAP_STATS
-		len = sprintf(tmp, "IO PDIR entries : %ld free  %ld used (%d%%)\n",
-			      total_pages - ioc->used_pages, ioc->used_pages,
-			      (int)(ioc->used_pages * 100 / total_pages));
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
+		len += seq_printf(m, "IO PDIR entries : %ld free  %ld used (%d%%)\n",
+				  total_pages - ioc->used_pages, ioc->used_pages,
+				  (int)(ioc->used_pages * 100 / total_pages));
 #endif
-		len = sprintf(tmp, "Resource bitmap : %d bytes (%d pages)\n", 
-			ioc->res_size, total_pages);
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
+
+		len += seq_printf(m, "Resource bitmap : %d bytes (%d pages)\n", 
+				  ioc->res_size, total_pages);
+
 #ifdef CCIO_SEARCH_TIME
 		min = max = ioc->avg_search[0];
 		for(j = 0; j < CCIO_SEARCH_SAMPLE; ++j) {
@@ -1085,70 +1058,83 @@ static int ccio_proc_info(char *buf, char **start, off_t offset, int count,
 				min = ioc->avg_search[j];
 		}
 		avg /= CCIO_SEARCH_SAMPLE;
-		len = sprintf(tmp, "  Bitmap search : %ld/%ld/%ld (min/avg/max CPU Cycles)\n",
-			      min, avg, max);
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
+		len += seq_printf(m, "  Bitmap search : %ld/%ld/%ld (min/avg/max CPU Cycles)\n",
+				  min, avg, max);
 #endif
 #ifdef CCIO_MAP_STATS
-		len = sprintf(tmp, "pci_map_single(): %8ld calls  %8ld pages (avg %d/1000)\n",
-			      ioc->msingle_calls, ioc->msingle_pages,
-			      (int)((ioc->msingle_pages * 1000)/ioc->msingle_calls));
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
-		
+		len += seq_printf(m, "pci_map_single(): %8ld calls  %8ld pages (avg %d/1000)\n",
+				  ioc->msingle_calls, ioc->msingle_pages,
+				  (int)((ioc->msingle_pages * 1000)/ioc->msingle_calls));
 
 		/* KLUGE - unmap_sg calls unmap_single for each mapped page */
 		min = ioc->usingle_calls - ioc->usg_calls;
 		max = ioc->usingle_pages - ioc->usg_pages;
-		len = sprintf(tmp, "pci_unmap_single: %8ld calls  %8ld pages (avg %d/1000)\n",
-			      min, max, (int)((max * 1000)/min));
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
+		len += seq_printf(m, "pci_unmap_single: %8ld calls  %8ld pages (avg %d/1000)\n",
+				  min, max, (int)((max * 1000)/min));
  
-		len = sprintf(tmp, "pci_map_sg()    : %8ld calls  %8ld pages (avg %d/1000)\n",
-			      ioc->msg_calls, ioc->msg_pages,
-			      (int)((ioc->msg_pages * 1000)/ioc->msg_calls));
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
-		len = sprintf(tmp, "pci_unmap_sg()  : %8ld calls  %8ld pages (avg %d/1000)\n\n\n",
-			      ioc->usg_calls, ioc->usg_pages,
-			      (int)((ioc->usg_pages * 1000)/ioc->usg_calls));
-		if (proc_append(tmp, len, &buf, &offset, &count))
-			break;
+		len += seq_printf(m, "pci_map_sg()    : %8ld calls  %8ld pages (avg %d/1000)\n",
+				  ioc->msg_calls, ioc->msg_pages,
+				  (int)((ioc->msg_pages * 1000)/ioc->msg_calls));
+
+		len += seq_printf(m, "pci_unmap_sg()  : %8ld calls  %8ld pages (avg %d/1000)\n\n\n",
+				  ioc->usg_calls, ioc->usg_pages,
+				  (int)((ioc->usg_pages * 1000)/ioc->usg_calls));
 #endif	/* CCIO_MAP_STATS */
+
 		ioc = ioc->next;
 	}
 
-	if (count == 0) {
-		*eof = 1;
-	}
-	return (max - count);
+	return 0;
 }
 
-static int ccio_resource_map(char *buf, char **start, off_t offset, int len,
-			     int *eof, void *data)
+static int ccio_proc_info_open(struct inode *inode, struct file *file)
 {
+	return single_open(file, &ccio_proc_info, NULL);
+}
+
+static struct file_operations ccio_proc_info_fops = {
+	.owner = THIS_MODULE,
+	.open = ccio_proc_info_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
+static int ccio_proc_bitmap_info(struct seq_file *m, void *p)
+{
+	int len = 0;
 	struct ioc *ioc = ioc_list;
 
-	buf[0] = '\0';
 	while (ioc != NULL) {
 		u32 *res_ptr = (u32 *)ioc->res_map;
 		int j;
 
 		for (j = 0; j < (ioc->res_size / sizeof(u32)); j++) {
 			if ((j & 7) == 0)
-				strcat(buf,"\n   ");
-			sprintf(buf, "%s %08x", buf, *res_ptr);
+				len += seq_puts(m, "\n   ");
+			len += seq_printf(m, "%08x", *res_ptr);
 			res_ptr++;
 		}
-		strcat(buf, "\n\n");
+		len += seq_puts(m, "\n\n");
 		ioc = ioc->next;
 		break; /* XXX - remove me */
 	}
 
-	return strlen(buf);
+	return 0;
 }
+
+static int ccio_proc_bitmap_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, &ccio_proc_bitmap_info, NULL);
+}
+
+static struct file_operations ccio_proc_bitmap_fops = {
+	.owner = THIS_MODULE,
+	.open = ccio_proc_bitmap_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
 #endif
 
 /**
@@ -1556,6 +1542,7 @@ static int ccio_probe(struct parisc_device *dev)
 {
 	int i;
 	struct ioc *ioc, **ioc_p = &ioc_list;
+	struct proc_dir_entry *info_entry, *bitmap_entry;
 	
 	ioc = kzalloc(sizeof(struct ioc), GFP_KERNEL);
 	if (ioc == NULL) {
@@ -1583,13 +1570,14 @@ static int ccio_probe(struct parisc_device *dev)
 	BUG_ON(dev->dev.platform_data == NULL);
 	HBA_DATA(dev->dev.platform_data)->iommu = ioc;
 	
-
 	if (ioc_count == 0) {
-		/* FIXME: Create separate entries for each ioc */
-		create_proc_read_entry(MODULE_NAME, S_IRWXU, proc_runway_root,
-				       ccio_proc_info, NULL);
-		create_proc_read_entry(MODULE_NAME"-bitmap", S_IRWXU,
-				       proc_runway_root, ccio_resource_map, NULL);
+		info_entry = create_proc_entry(MODULE_NAME, 0, proc_runway_root);
+		if (info_entry)
+			info_entry->proc_fops = &ccio_proc_info_fops;
+
+		bitmap_entry = create_proc_entry(MODULE_NAME"-bitmap", 0, proc_runway_root);
+		if (bitmap_entry)
+			bitmap_entry->proc_fops = &ccio_proc_bitmap_fops;
 	}
 
 	ioc_count++;

From 7ec14e49b72da20d7212c707f226271525aee4ae Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@parisc-linux.org>
Date: Mon, 6 Feb 2006 10:10:15 -0700
Subject: [PATCH 045/509] [PARISC] Convert sba_iommu.c to use seq_file

Use seq_file interface for proc output in sba_iommu. Also
clean up the bus root assignment, and give the proc files
a more logical name. Tested on my J6000.

Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
---
 drivers/parisc/sba_iommu.c | 125 +++++++++++++++++++++++--------------
 1 file changed, 78 insertions(+), 47 deletions(-)

diff --git a/drivers/parisc/sba_iommu.c b/drivers/parisc/sba_iommu.c
index 52f265e97729..5d47c5965c51 100644
--- a/drivers/parisc/sba_iommu.c
+++ b/drivers/parisc/sba_iommu.c
@@ -37,6 +37,8 @@
 #include <asm/hardware.h>	/* for register_parisc_driver() stuff */
 
 #include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+
 #include <asm/runway.h>		/* for proc_runway_root */
 #include <asm/pdc.h>		/* for PDC_MODEL_* */
 #include <asm/pdcpat.h>		/* for is_pdc_pat() */
@@ -1892,46 +1894,43 @@ sba_common_init(struct sba_device *sba_dev)
 }
 
 #ifdef CONFIG_PROC_FS
-static int sba_proc_info(char *buf, char **start, off_t offset, int len)
+static int sba_proc_info(struct seq_file *m, void *p)
 {
 	struct sba_device *sba_dev = sba_list;
 	struct ioc *ioc = &sba_dev->ioc[0];	/* FIXME: Multi-IOC support! */
 	int total_pages = (int) (ioc->res_size << 3); /* 8 bits per byte */
-	unsigned long i;
 #ifdef SBA_COLLECT_STATS
 	unsigned long avg = 0, min, max;
 #endif
+	int i, len = 0;
 
-	sprintf(buf, "%s rev %d.%d\n",
+	len += seq_printf(m, "%s rev %d.%d\n",
 		sba_dev->name,
 		(sba_dev->hw_rev & 0x7) + 1,
 		(sba_dev->hw_rev & 0x18) >> 3
 		);
-	sprintf(buf, "%sIO PDIR size    : %d bytes (%d entries)\n",
-		buf,
+	len += seq_printf(m, "IO PDIR size    : %d bytes (%d entries)\n",
 		(int) ((ioc->res_size << 3) * sizeof(u64)), /* 8 bits/byte */
 		total_pages);
 
-	sprintf(buf, "%sResource bitmap : %d bytes (%d pages)\n", 
-		buf, ioc->res_size, ioc->res_size << 3);   /* 8 bits per byte */
+	len += seq_printf(m, "Resource bitmap : %d bytes (%d pages)\n", 
+		ioc->res_size, ioc->res_size << 3);   /* 8 bits per byte */
 
-	sprintf(buf, "%sLMMIO_BASE/MASK/ROUTE %08x %08x %08x\n",
-		buf,
+	len += seq_printf(m, "LMMIO_BASE/MASK/ROUTE %08x %08x %08x\n",
 		READ_REG32(sba_dev->sba_hpa + LMMIO_DIST_BASE),
 		READ_REG32(sba_dev->sba_hpa + LMMIO_DIST_MASK),
 		READ_REG32(sba_dev->sba_hpa + LMMIO_DIST_ROUTE)
 		);
 
 	for (i=0; i<4; i++)
-		sprintf(buf, "%sDIR%ld_BASE/MASK/ROUTE %08x %08x %08x\n",
-			buf, i,
+		len += seq_printf(m, "DIR%d_BASE/MASK/ROUTE %08x %08x %08x\n", i,
 			READ_REG32(sba_dev->sba_hpa + LMMIO_DIRECT0_BASE  + i*0x18),
 			READ_REG32(sba_dev->sba_hpa + LMMIO_DIRECT0_MASK  + i*0x18),
 			READ_REG32(sba_dev->sba_hpa + LMMIO_DIRECT0_ROUTE + i*0x18)
 		);
 
 #ifdef SBA_COLLECT_STATS
-	sprintf(buf, "%sIO PDIR entries : %ld free  %ld used (%d%%)\n", buf,
+	len += seq_printf(m, "IO PDIR entries : %ld free  %ld used (%d%%)\n",
 		total_pages - ioc->used_pages, ioc->used_pages,
 		(int) (ioc->used_pages * 100 / total_pages));
 
@@ -1942,53 +1941,76 @@ static int sba_proc_info(char *buf, char **start, off_t offset, int len)
 		if (ioc->avg_search[i] < min) min = ioc->avg_search[i];
 	}
 	avg /= SBA_SEARCH_SAMPLE;
-	sprintf(buf, "%s  Bitmap search : %ld/%ld/%ld (min/avg/max CPU Cycles)\n",
-		buf, min, avg, max);
+	len += seq_printf(m, "  Bitmap search : %ld/%ld/%ld (min/avg/max CPU Cycles)\n",
+		min, avg, max);
 
-	sprintf(buf, "%spci_map_single(): %12ld calls  %12ld pages (avg %d/1000)\n",
-		buf, ioc->msingle_calls, ioc->msingle_pages,
+	len += seq_printf(m, "pci_map_single(): %12ld calls  %12ld pages (avg %d/1000)\n",
+		ioc->msingle_calls, ioc->msingle_pages,
 		(int) ((ioc->msingle_pages * 1000)/ioc->msingle_calls));
 
 	/* KLUGE - unmap_sg calls unmap_single for each mapped page */
 	min = ioc->usingle_calls;
 	max = ioc->usingle_pages - ioc->usg_pages;
-	sprintf(buf, "%spci_unmap_single: %12ld calls  %12ld pages (avg %d/1000)\n",
-		buf, min, max,
-		(int) ((max * 1000)/min));
+	len += seq_printf(m, "pci_unmap_single: %12ld calls  %12ld pages (avg %d/1000)\n",
+		min, max, (int) ((max * 1000)/min));
 
-	sprintf(buf, "%spci_map_sg()    : %12ld calls  %12ld pages (avg %d/1000)\n",
-		buf, ioc->msg_calls, ioc->msg_pages,
+	len += seq_printf(m, "pci_map_sg()    : %12ld calls  %12ld pages (avg %d/1000)\n",
+		ioc->msg_calls, ioc->msg_pages, 
 		(int) ((ioc->msg_pages * 1000)/ioc->msg_calls));
 
-	sprintf(buf, "%spci_unmap_sg()  : %12ld calls  %12ld pages (avg %d/1000)\n",
-		buf, ioc->usg_calls, ioc->usg_pages,
+	len += seq_printf(m, "pci_unmap_sg()  : %12ld calls  %12ld pages (avg %d/1000)\n",
+		ioc->usg_calls, ioc->usg_pages,
 		(int) ((ioc->usg_pages * 1000)/ioc->usg_calls));
 #endif
 
-	return strlen(buf);
+	return 0;
 }
 
-#if 0
-/* XXX too much output - exceeds 4k limit and needs to be re-written */
 static int
-sba_resource_map(char *buf, char **start, off_t offset, int len)
+sba_proc_open(struct inode *i, struct file *f)
+{
+	return single_open(f, &sba_proc_info, NULL);
+}
+
+static struct file_operations sba_proc_fops = {
+	.owner = THIS_MODULE,
+	.open = sba_proc_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
+
+static int
+sba_proc_bitmap_info(struct seq_file *m, void *p)
 {
 	struct sba_device *sba_dev = sba_list;
-	struct ioc *ioc = &sba_dev->ioc[0];	/* FIXME: Mutli-IOC suppoer! */
+	struct ioc *ioc = &sba_dev->ioc[0];	/* FIXME: Multi-IOC support! */
 	unsigned int *res_ptr = (unsigned int *)ioc->res_map;
-	int i;
+	int i, len = 0;
 
-	buf[0] = '\0';
-	for(i = 0; i < (ioc->res_size / sizeof(unsigned int)); ++i, ++res_ptr) {
+	for (i = 0; i < (ioc->res_size/sizeof(unsigned int)); ++i, ++res_ptr) {
 		if ((i & 7) == 0)
-		    strcat(buf,"\n   ");
-		sprintf(buf, "%s %08x", buf, *res_ptr);
+			len += seq_printf(m, "\n   ");
+		len += seq_printf(m, " %08x", *res_ptr);
 	}
-	strcat(buf, "\n");
+	len += seq_printf(m, "\n");
 
-	return strlen(buf);
+	return 0;
 }
-#endif /* 0 */
+
+static int
+sba_proc_bitmap_open(struct inode *i, struct file *f)
+{
+	return single_open(f, &sba_proc_bitmap_info, NULL);
+}
+
+static struct file_operations sba_proc_bitmap_fops = {
+	.owner = THIS_MODULE,
+	.open = sba_proc_bitmap_open,
+	.read = seq_read,
+	.llseek = seq_lseek,
+	.release = single_release,
+};
 #endif /* CONFIG_PROC_FS */
 
 static struct parisc_device_id sba_tbl[] = {
@@ -2021,6 +2043,7 @@ sba_driver_callback(struct parisc_device *dev)
 	int i;
 	char *version;
 	void __iomem *sba_addr = ioremap(dev->hpa.start, SBA_FUNC_SIZE);
+	struct proc_dir_entry *info_entry, *bitmap_entry, *root;
 
 	sba_dump_ranges(sba_addr);
 
@@ -2088,19 +2111,27 @@ sba_driver_callback(struct parisc_device *dev)
 	hppa_dma_ops = &sba_ops;
 
 #ifdef CONFIG_PROC_FS
-	if (IS_ASTRO(&dev->id)) {
-		create_proc_info_entry("Astro", 0, proc_runway_root, sba_proc_info);
-	} else if (IS_IKE(&dev->id)) {
-		create_proc_info_entry("Ike", 0, proc_runway_root, sba_proc_info);
-	} else if (IS_PLUTO(&dev->id)) {
-		create_proc_info_entry("Pluto", 0, proc_mckinley_root, sba_proc_info);
-	} else {
-		create_proc_info_entry("Reo", 0, proc_runway_root, sba_proc_info);
+	switch (dev->id.hversion) {
+	case PLUTO_MCKINLEY_PORT:
+		root = proc_mckinley_root;
+		break;
+	case ASTRO_RUNWAY_PORT:
+	case IKE_MERCED_PORT:
+	default:
+		root = proc_runway_root;
+		break;
 	}
-#if 0
-	create_proc_info_entry("bitmap", 0, proc_runway_root, sba_resource_map);
-#endif
+
+	info_entry = create_proc_entry("sba_iommu", 0, root);
+	bitmap_entry = create_proc_entry("sba_iommu-bitmap", 0, root);
+
+	if (info_entry)
+		info_entry->proc_fops = &sba_proc_fops;
+
+	if (bitmap_entry)
+		bitmap_entry->proc_fops = &sba_proc_bitmap_fops;
 #endif
+
 	parisc_vmerge_boundary = IOVP_SIZE;
 	parisc_vmerge_max_size = IOVP_SIZE * BITS_PER_LONG;
 	parisc_has_iommu();

From 28b2425ac76b0f043f2efd34521f2a3c7ec8ccc5 Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@parisc-linux.org>
Date: Wed, 8 Feb 2006 23:33:15 -0500
Subject: [PATCH 046/509] [PARISC] Stub out pselect6/ppoll until
 TIF_RESTORE_SIGMASK is done

Swap out pselect6/ppoll for ni_syscall for now. We also have to switch
the macro to ENTRY_SAME since compat_sys_ni_syscall does not exist.

Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
---
 arch/parisc/kernel/syscall_table.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S
index 51d2480627d1..66224f710e85 100644
--- a/arch/parisc/kernel/syscall_table.S
+++ b/arch/parisc/kernel/syscall_table.S
@@ -377,8 +377,8 @@
 	ENTRY_SAME(inotify_init)
 	ENTRY_SAME(inotify_add_watch)	/* 270 */
 	ENTRY_SAME(inotify_rm_watch)
-	ENTRY_COMP(pselect6)
-	ENTRY_COMP(ppoll)
+	ENTRY_SAME(ni_syscall)		/* 271 ENTRY_COMP(pselect6) */
+	ENTRY_SAME(ni_syscall)		/* 272 ENTRY_COMP(ppoll) */
 	ENTRY_SAME(migrate_pages)
 	ENTRY_COMP(openat)		/* 275 */
 	ENTRY_SAME(mkdirat)

From b514d3192736563dee3f4ba4b659558b3cadc7f7 Mon Sep 17 00:00:00 2001
From: Lucas Correia Villa Real <lucasvr@gobolinux.org>
Date: Thu, 9 Feb 2006 16:47:58 +0000
Subject: [PATCH 047/509] [ARM] 3314/1: S3C2400 - adds s3c2400.h

Patch from Lucas Correia Villa Real

This patch adds s3c2400.h, fixing the build for the 2410/2440
platforms.

Signed-off-by: Lucas Correia Villa Real <lucasvr@gobolinux.org>
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2410/s3c2400.h | 31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)
 create mode 100644 arch/arm/mach-s3c2410/s3c2400.h

diff --git a/arch/arm/mach-s3c2410/s3c2400.h b/arch/arm/mach-s3c2410/s3c2400.h
new file mode 100644
index 000000000000..8b2394e1ed40
--- /dev/null
+++ b/arch/arm/mach-s3c2410/s3c2400.h
@@ -0,0 +1,31 @@
+/* arch/arm/mach-s3c2410/s3c2400.h
+ *
+ * Copyright (c) 2004 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * Header file for S3C2400 cpu support
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * Modifications:
+ *     09-Fev-2006 LCVR  First version, based on s3c2410.h
+*/
+
+#ifdef CONFIG_CPU_S3C2400
+
+extern  int s3c2400_init(void);
+
+extern void s3c2400_map_io(struct map_desc *mach_desc, int size);
+
+extern void s3c2400_init_uarts(struct s3c2410_uartcfg *cfg, int no);
+
+extern void s3c2400_init_clocks(int xtal);
+
+#else
+#define s3c2400_init_clocks NULL
+#define s3c2400_init_uarts NULL
+#define s3c2400_map_io NULL
+#define s3c2400_init NULL
+#endif

From a70ea994a0d83fd0151a070be72b87d014ef0a7e Mon Sep 17 00:00:00 2001
From: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Date: Thu, 9 Feb 2006 16:40:11 -0800
Subject: [PATCH 048/509] [NETLINK]: Fix a severe bug

netlink overrun was broken while improvement of netlink.
Destination socket is used in the place where it was meant to be source socket,
so that now overrun is never sent to user netlink sockets, when it should be,
and it even can be set on kernel socket, which results in complete deadlock
of rtnetlink.

Suggested fix is to restore status quo passing source socket as additional
argument to netlink_attachskb().

A little explanation: overrun is set on a socket, when it failed
to receive some message and sender of this messages does not or even
have no way to handle this error. This happens in two cases:
1. when kernel sends something. Kernel never retransmits and cannot
   wait for buffer space.
2. when user sends a broadcast and the message was not delivered
   to some recipients.

Signed-off-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netlink.h  | 3 ++-
 ipc/mqueue.c             | 3 ++-
 net/netlink/af_netlink.c | 7 ++++---
 3 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index 6a2ccf78a356..c256ebe2a7b4 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -160,7 +160,8 @@ extern int netlink_unregister_notifier(struct notifier_block *nb);
 
 /* finegrained unicast helpers: */
 struct sock *netlink_getsockbyfilp(struct file *filp);
-int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock, long timeo);
+int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock,
+		long timeo, struct sock *ssk);
 void netlink_detachskb(struct sock *sk, struct sk_buff *skb);
 int netlink_sendskb(struct sock *sk, struct sk_buff *skb, int protocol);
 
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 59302fc3643b..fd2e26b6f966 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -1018,7 +1018,8 @@ retry:
 				goto out;
 			}
 
-			ret = netlink_attachskb(sock, nc, 0, MAX_SCHEDULE_TIMEOUT);
+			ret = netlink_attachskb(sock, nc, 0,
+					MAX_SCHEDULE_TIMEOUT, NULL);
 			if (ret == 1)
 		       		goto retry;
 			if (ret) {
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 2101b45d2ec6..6b9772d95872 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -702,7 +702,8 @@ struct sock *netlink_getsockbyfilp(struct file *filp)
  * 0: continue
  * 1: repeat lookup - reference dropped while waiting for socket memory.
  */
-int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock, long timeo)
+int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock,
+		long timeo, struct sock *ssk)
 {
 	struct netlink_sock *nlk;
 
@@ -712,7 +713,7 @@ int netlink_attachskb(struct sock *sk, struct sk_buff *skb, int nonblock, long t
 	    test_bit(0, &nlk->state)) {
 		DECLARE_WAITQUEUE(wait, current);
 		if (!timeo) {
-			if (!nlk->pid)
+			if (!ssk || nlk_sk(ssk)->pid == 0)
 				netlink_overrun(sk);
 			sock_put(sk);
 			kfree_skb(skb);
@@ -797,7 +798,7 @@ retry:
 		kfree_skb(skb);
 		return PTR_ERR(sk);
 	}
-	err = netlink_attachskb(sk, skb, nonblock, timeo);
+	err = netlink_attachskb(sk, skb, nonblock, timeo, ssk);
 	if (err == 1)
 		goto retry;
 	if (err)

From 28633514afd68afa77ed2fa34fa53626837bf2d5 Mon Sep 17 00:00:00 2001
From: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Date: Thu, 9 Feb 2006 16:40:58 -0800
Subject: [PATCH 049/509] [NETLINK]: illegal use of pid in rtnetlink

When a netlink message is not related to a netlink socket,
it is issued by kernel socket with pid 0. Netlink "pid" has nothing
to do with current->pid. I called it incorrectly, if it was named "port",
the confusion would be avoided.

Signed-off-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/rtnetlink.c     | 2 +-
 net/ipv4/devinet.c       | 2 +-
 net/ipv4/fib_semantics.c | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 8700379685e0..eca2976abb25 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -455,7 +455,7 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change)
 	if (!skb)
 		return;
 
-	if (rtnetlink_fill_ifinfo(skb, dev, type, current->pid, 0, change, 0) < 0) {
+	if (rtnetlink_fill_ifinfo(skb, dev, type, 0, 0, change, 0) < 0) {
 		kfree_skb(skb);
 		return;
 	}
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 95b9d81ac488..3ffa60dadc0c 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1135,7 +1135,7 @@ static void rtmsg_ifa(int event, struct in_ifaddr* ifa)
 
 	if (!skb)
 		netlink_set_err(rtnl, 0, RTNLGRP_IPV4_IFADDR, ENOBUFS);
-	else if (inet_fill_ifaddr(skb, ifa, current->pid, 0, event, 0) < 0) {
+	else if (inet_fill_ifaddr(skb, ifa, 0, 0, event, 0) < 0) {
 		kfree_skb(skb);
 		netlink_set_err(rtnl, 0, RTNLGRP_IPV4_IFADDR, EINVAL);
 	} else {
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index ef4724de7350..0f4145babb14 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -1045,7 +1045,7 @@ fib_convert_rtentry(int cmd, struct nlmsghdr *nl, struct rtmsg *rtm,
 	}
 
 	nl->nlmsg_flags = NLM_F_REQUEST;
-	nl->nlmsg_pid = current->pid;
+	nl->nlmsg_pid = 0;
 	nl->nlmsg_seq = 0;
 	nl->nlmsg_len = NLMSG_LENGTH(sizeof(*rtm));
 	if (cmd == SIOCDELRT) {

From d93077fb0e7cb9d4f4094a649501d840c55fdc8b Mon Sep 17 00:00:00 2001
From: Samuel Ortiz <samuel.ortiz@nokia.com>
Date: Thu, 9 Feb 2006 16:58:46 -0800
Subject: [PATCH 050/509] [IRDA]: Set proper IrLAP device address length

This patch set IrDA's addr_len properly, i.e to 4 bytes, the size of the
IrLAP device address.

Signed-off-by: Samuel Ortiz <samuel.ortiz@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/irda/irlap.h | 3 +++
 net/irda/irda_device.c   | 4 ++--
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/include/net/irda/irlap.h b/include/net/irda/irlap.h
index f55e86e75030..2127cae1e0a6 100644
--- a/include/net/irda/irlap.h
+++ b/include/net/irda/irlap.h
@@ -50,6 +50,9 @@
 /* May be different when we get VFIR */
 #define LAP_MAX_HEADER (LAP_ADDR_HEADER + LAP_CTRL_HEADER)
 
+/* Each IrDA device gets a random 32 bits IRLAP device address */
+#define LAP_ALEN 4
+
 #define BROADCAST  0xffffffff /* Broadcast device address */
 #define CBROADCAST 0xfe       /* Connection broadcast address */
 #define XID_FORMAT 0x01       /* Discovery XID format */
diff --git a/net/irda/irda_device.c b/net/irda/irda_device.c
index 890bac0d4a56..e3debbdb67f5 100644
--- a/net/irda/irda_device.c
+++ b/net/irda/irda_device.c
@@ -343,12 +343,12 @@ static void irda_task_timer_expired(void *data)
 static void irda_device_setup(struct net_device *dev)
 {
         dev->hard_header_len = 0;
-        dev->addr_len        = 0;
+        dev->addr_len        = LAP_ALEN;
 
         dev->type            = ARPHRD_IRDA;
         dev->tx_queue_len    = 8; /* Window size + 1 s-frame */
 
-	memset(dev->broadcast, 0xff, 4);
+	memset(dev->broadcast, 0xff, LAP_ALEN);
 
 	dev->mtu = 2048;
 	dev->flags = IFF_NOARP;

From 80ba250e59ced808a8c9b79560938bbe4509c0a7 Mon Sep 17 00:00:00 2001
From: David Binderman <dcb314@hotmail.com>
Date: Thu, 9 Feb 2006 16:59:48 -0800
Subject: [PATCH 051/509] [IRDA]: out of range array access

This patch fixes an out of range array access in irnet_irda.c.

Author: David Binderman <dcb314@hotmail.com>
Signed-off-by: Samuel Ortiz <samuel.ortiz@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/irda/irnet/irnet_irda.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/irda/irnet/irnet_irda.c b/net/irda/irnet/irnet_irda.c
index 07ec326c71f5..f65c7a83bc5c 100644
--- a/net/irda/irnet/irnet_irda.c
+++ b/net/irda/irnet/irnet_irda.c
@@ -696,7 +696,7 @@ irnet_daddr_to_dname(irnet_socket *	self)
 	{
 	  /* Yes !!! Get it.. */
 	  strlcpy(self->rname, discoveries[i].info, sizeof(self->rname));
-	  self->rname[NICKNAME_MAX_LEN + 1] = '\0';
+	  self->rname[sizeof(self->rname) - 1] = '\0';
 	  DEBUG(IRDA_SERV_INFO, "Device 0x%08x is in fact ``%s''.\n",
 		self->daddr, self->rname);
 	  kfree(discoveries);

From 6fcf9412de64056238a6295f21db7aa9c37a532e Mon Sep 17 00:00:00 2001
From: John Heffner <jheffner@psc.edu>
Date: Thu, 9 Feb 2006 17:06:57 -0800
Subject: [PATCH 052/509] [TCP]: rcvbuf lock when tcp_moderate_rcvbuf enabled

The rcvbuf lock should probably be honored here.

Signed-off-by: John Heffner <jheffner@psc.edu>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_input.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index a97ed5416c28..e9a54ae7d690 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -456,7 +456,8 @@ void tcp_rcv_space_adjust(struct sock *sk)
 
 		tp->rcvq_space.space = space;
 
-		if (sysctl_tcp_moderate_rcvbuf) {
+		if (sysctl_tcp_moderate_rcvbuf &&
+		    !(sk->sk_userlocks & SOCK_RCVBUF_LOCK)) {
 			int new_clamp = space;
 
 			/* Receive space grows, normalize in order to

From b3f1be4b5412e34647764457bec901e06b03e624 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 9 Feb 2006 17:08:52 -0800
Subject: [PATCH 053/509] [BRIDGE]: fix for RCU and deadlock on device removal

Change Bridge receive path to correctly handle RCU removal of device
from bridge.  Also fixes deadlock between carrier_check and del_nbp.
This replaces the previous deleted flag fix.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_if.c       | 21 +++++++++++----------
 net/bridge/br_input.c    | 21 +++++++++++++--------
 net/bridge/br_private.h  |  1 -
 net/bridge/br_stp_bpdu.c | 30 ++++++++++++++++++------------
 4 files changed, 42 insertions(+), 31 deletions(-)

diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index da687c8dc6ff..70b7ef917234 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -79,9 +79,14 @@ static int port_cost(struct net_device *dev)
  */
 static void port_carrier_check(void *arg)
 {
-	struct net_bridge_port *p = arg;
+	struct net_device *dev = arg;
+	struct net_bridge_port *p;
 
 	rtnl_lock();
+	p = dev->br_port;
+	if (!p)
+		goto done;
+
 	if (netif_carrier_ok(p->dev)) {
 		u32 cost = port_cost(p->dev);
 
@@ -97,6 +102,7 @@ static void port_carrier_check(void *arg)
 			br_stp_disable_port(p);
 		spin_unlock_bh(&p->br->lock);
 	}
+done:
 	rtnl_unlock();
 }
 
@@ -104,7 +110,6 @@ static void destroy_nbp(struct net_bridge_port *p)
 {
 	struct net_device *dev = p->dev;
 
-	dev->br_port = NULL;
 	p->br = NULL;
 	p->dev = NULL;
 	dev_put(dev);
@@ -133,24 +138,20 @@ static void del_nbp(struct net_bridge_port *p)
 	struct net_bridge *br = p->br;
 	struct net_device *dev = p->dev;
 
-	/* Race between RTNL notify and RCU callback */
-	if (p->deleted)
-		return;
-
 	dev_set_promiscuity(dev, -1);
 
 	cancel_delayed_work(&p->carrier_check);
-	flush_scheduled_work();
 
 	spin_lock_bh(&br->lock);
 	br_stp_disable_port(p);
-	p->deleted = 1;
 	spin_unlock_bh(&br->lock);
 
 	br_fdb_delete_by_port(br, p);
 
 	list_del_rcu(&p->list);
 
+	rcu_assign_pointer(dev->br_port, NULL);
+
 	call_rcu(&p->rcu, destroy_nbp_rcu);
 }
 
@@ -254,11 +255,10 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br,
 	p->dev = dev;
 	p->path_cost = port_cost(dev);
  	p->priority = 0x8000 >> BR_PORT_BITS;
-	dev->br_port = p;
 	p->port_no = index;
 	br_init_port(p);
 	p->state = BR_STATE_DISABLED;
-	INIT_WORK(&p->carrier_check, port_carrier_check, p);
+	INIT_WORK(&p->carrier_check, port_carrier_check, dev);
 	kobject_init(&p->kobj);
 
 	return p;
@@ -397,6 +397,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
 	else if ((err = br_sysfs_addif(p)))
 		del_nbp(p);
 	else {
+		rcu_assign_pointer(dev->br_port, p);
 		dev_set_promiscuity(dev, 1);
 
 		list_add_rcu(&p->list, &br->port_list);
diff --git a/net/bridge/br_input.c b/net/bridge/br_input.c
index e3a73cead6b6..4eef83755315 100644
--- a/net/bridge/br_input.c
+++ b/net/bridge/br_input.c
@@ -45,18 +45,20 @@ static void br_pass_frame_up(struct net_bridge *br, struct sk_buff *skb)
 int br_handle_frame_finish(struct sk_buff *skb)
 {
 	const unsigned char *dest = eth_hdr(skb)->h_dest;
-	struct net_bridge_port *p = skb->dev->br_port;
-	struct net_bridge *br = p->br;
+	struct net_bridge_port *p = rcu_dereference(skb->dev->br_port);
+	struct net_bridge *br;
 	struct net_bridge_fdb_entry *dst;
 	int passedup = 0;
 
-	/* insert into forwarding database after filtering to avoid spoofing */
-	br_fdb_update(p->br, p, eth_hdr(skb)->h_source);
+	if (!p || p->state == BR_STATE_DISABLED)
+		goto drop;
 
-	if (p->state == BR_STATE_LEARNING) {
-		kfree_skb(skb);
-		goto out;
-	}
+	/* insert into forwarding database after filtering to avoid spoofing */
+	br = p->br;
+	br_fdb_update(br, p, eth_hdr(skb)->h_source);
+
+	if (p->state == BR_STATE_LEARNING)
+		goto drop;
 
 	if (br->dev->flags & IFF_PROMISC) {
 		struct sk_buff *skb2;
@@ -93,6 +95,9 @@ int br_handle_frame_finish(struct sk_buff *skb)
 
 out:
 	return 0;
+drop:
+	kfree_skb(skb);
+	goto out;
 }
 
 /*
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index e330b17b6d81..c5bd631ffcd5 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -68,7 +68,6 @@ struct net_bridge_port
 	/* STP */
 	u8				priority;
 	u8				state;
-	u8				deleted;
 	u16				port_no;
 	unsigned char			topology_change_ack;
 	unsigned char			config_pending;
diff --git a/net/bridge/br_stp_bpdu.c b/net/bridge/br_stp_bpdu.c
index d071f1c9ad0b..296f6a487c52 100644
--- a/net/bridge/br_stp_bpdu.c
+++ b/net/bridge/br_stp_bpdu.c
@@ -133,29 +133,35 @@ void br_send_tcn_bpdu(struct net_bridge_port *p)
 
 static const unsigned char header[6] = {0x42, 0x42, 0x03, 0x00, 0x00, 0x00};
 
-/* NO locks */
+/* NO locks, but rcu_read_lock (preempt_disabled)  */
 int br_stp_handle_bpdu(struct sk_buff *skb)
 {
-	struct net_bridge_port *p = skb->dev->br_port;
-	struct net_bridge *br = p->br;
+	struct net_bridge_port *p = rcu_dereference(skb->dev->br_port);
+	struct net_bridge *br;
 	unsigned char *buf;
 
+	if (!p)
+		goto err;
+
+	br = p->br;
+	spin_lock(&br->lock);
+
+	if (p->state == BR_STATE_DISABLED || !(br->dev->flags & IFF_UP))
+		goto out;
+
 	/* insert into forwarding database after filtering to avoid spoofing */
-	br_fdb_update(p->br, p, eth_hdr(skb)->h_source);
+	br_fdb_update(br, p, eth_hdr(skb)->h_source);
+
+	if (!br->stp_enabled)
+		goto out;
 
 	/* need at least the 802 and STP headers */
 	if (!pskb_may_pull(skb, sizeof(header)+1) ||
 	    memcmp(skb->data, header, sizeof(header)))
-		goto err;
+		goto out;
 
 	buf = skb_pull(skb, sizeof(header));
 
-	spin_lock_bh(&br->lock);
-	if (p->state == BR_STATE_DISABLED 
-	    || !(br->dev->flags & IFF_UP)
-	    || !br->stp_enabled)
-		goto out;
-
 	if (buf[0] == BPDU_TYPE_CONFIG) {
 		struct br_config_bpdu bpdu;
 
@@ -201,7 +207,7 @@ int br_stp_handle_bpdu(struct sk_buff *skb)
 		br_received_tcn_bpdu(p);
 	}
  out:
-	spin_unlock_bh(&br->lock);
+	spin_unlock(&br->lock);
  err:
 	kfree_skb(skb);
 	return 0;

From 5dce971acf2ae20c80d5e9d1f6bbf17376870911 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 9 Feb 2006 17:09:38 -0800
Subject: [PATCH 054/509] [BRIDGE]: netfilter handle RCU during removal

Bridge netfilter code needs to handle the case where device is
removed from bridge while packet in process. In these cases the
bridge_parent can become null while processing.

This should fix: http://bugzilla.kernel.org/show_bug.cgi?id=5803

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_netfilter.c | 53 ++++++++++++++++++++++++++++-----------
 1 file changed, 38 insertions(+), 15 deletions(-)

diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index 7cac3fb9f809..b5018166b0e5 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -51,9 +51,6 @@
 #define store_orig_dstaddr(skb)	 (skb_origaddr(skb) = (skb)->nh.iph->daddr)
 #define dnat_took_place(skb)	 (skb_origaddr(skb) != (skb)->nh.iph->daddr)
 
-#define has_bridge_parent(device)	((device)->br_port != NULL)
-#define bridge_parent(device)		((device)->br_port->br->dev)
-
 #ifdef CONFIG_SYSCTL
 static struct ctl_table_header *brnf_sysctl_header;
 static int brnf_call_iptables = 1;
@@ -98,6 +95,12 @@ static struct rtable __fake_rtable = {
 	.rt_flags	= 0,
 };
 
+static inline struct net_device *bridge_parent(const struct net_device *dev)
+{
+	struct net_bridge_port *port = rcu_dereference(dev->br_port);
+
+	return port ? port->br->dev : NULL;
+}
 
 /* PF_BRIDGE/PRE_ROUTING *********************************************/
 /* Undo the changes made for ip6tables PREROUTING and continue the
@@ -189,11 +192,15 @@ static int br_nf_pre_routing_finish_bridge(struct sk_buff *skb)
 	skb->nf_bridge->mask ^= BRNF_NF_BRIDGE_PREROUTING;
 
 	skb->dev = bridge_parent(skb->dev);
-	if (skb->protocol == __constant_htons(ETH_P_8021Q)) {
-		skb_pull(skb, VLAN_HLEN);
-		skb->nh.raw += VLAN_HLEN;
+	if (!skb->dev)
+		kfree_skb(skb);
+	else {
+		if (skb->protocol == __constant_htons(ETH_P_8021Q)) {
+			skb_pull(skb, VLAN_HLEN);
+			skb->nh.raw += VLAN_HLEN;
+		}
+		skb->dst->output(skb);
 	}
-	skb->dst->output(skb);
 	return 0;
 }
 
@@ -270,7 +277,7 @@ bridged_dnat:
 }
 
 /* Some common code for IPv4/IPv6 */
-static void setup_pre_routing(struct sk_buff *skb)
+static struct net_device *setup_pre_routing(struct sk_buff *skb)
 {
 	struct nf_bridge_info *nf_bridge = skb->nf_bridge;
 
@@ -282,6 +289,8 @@ static void setup_pre_routing(struct sk_buff *skb)
 	nf_bridge->mask |= BRNF_NF_BRIDGE_PREROUTING;
 	nf_bridge->physindev = skb->dev;
 	skb->dev = bridge_parent(skb->dev);
+
+	return skb->dev;
 }
 
 /* We only check the length. A bridge shouldn't do any hop-by-hop stuff anyway */
@@ -376,7 +385,8 @@ static unsigned int br_nf_pre_routing_ipv6(unsigned int hook,
  	nf_bridge_put(skb->nf_bridge);
 	if ((nf_bridge = nf_bridge_alloc(skb)) == NULL)
 		return NF_DROP;
-	setup_pre_routing(skb);
+	if (!setup_pre_routing(skb))
+		return NF_DROP;
 
 	NF_HOOK(PF_INET6, NF_IP6_PRE_ROUTING, skb, skb->dev, NULL,
 		br_nf_pre_routing_finish_ipv6);
@@ -465,7 +475,8 @@ static unsigned int br_nf_pre_routing(unsigned int hook, struct sk_buff **pskb,
  	nf_bridge_put(skb->nf_bridge);
 	if ((nf_bridge = nf_bridge_alloc(skb)) == NULL)
 		return NF_DROP;
-	setup_pre_routing(skb);
+	if (!setup_pre_routing(skb))
+		return NF_DROP;
 	store_orig_dstaddr(skb);
 
 	NF_HOOK(PF_INET, NF_IP_PRE_ROUTING, skb, skb->dev, NULL,
@@ -539,11 +550,16 @@ static unsigned int br_nf_forward_ip(unsigned int hook, struct sk_buff **pskb,
 	struct sk_buff *skb = *pskb;
 	struct nf_bridge_info *nf_bridge;
 	struct vlan_ethhdr *hdr = vlan_eth_hdr(skb);
+	struct net_device *parent;
 	int pf;
 
 	if (!skb->nf_bridge)
 		return NF_ACCEPT;
 
+	parent = bridge_parent(out);
+	if (!parent)
+		return NF_DROP;
+
 	if (skb->protocol == __constant_htons(ETH_P_IP) || IS_VLAN_IP)
 		pf = PF_INET;
 	else
@@ -564,8 +580,8 @@ static unsigned int br_nf_forward_ip(unsigned int hook, struct sk_buff **pskb,
 	nf_bridge->mask |= BRNF_BRIDGED;
 	nf_bridge->physoutdev = skb->dev;
 
-	NF_HOOK(pf, NF_IP_FORWARD, skb, bridge_parent(in),
-		bridge_parent(out), br_nf_forward_finish);
+	NF_HOOK(pf, NF_IP_FORWARD, skb, bridge_parent(in), parent,
+		br_nf_forward_finish);
 
 	return NF_STOLEN;
 }
@@ -688,6 +704,8 @@ static unsigned int br_nf_local_out(unsigned int hook, struct sk_buff **pskb,
 		goto out;
 	}
 	realoutdev = bridge_parent(skb->dev);
+	if (!realoutdev)
+		return NF_DROP;
 
 #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
 	/* iptables should match -o br0.x */
@@ -701,9 +719,11 @@ static unsigned int br_nf_local_out(unsigned int hook, struct sk_buff **pskb,
 	/* IP forwarded traffic has a physindev, locally
 	 * generated traffic hasn't. */
 	if (realindev != NULL) {
-		if (!(nf_bridge->mask & BRNF_DONT_TAKE_PARENT) &&
-		    has_bridge_parent(realindev))
-			realindev = bridge_parent(realindev);
+		if (!(nf_bridge->mask & BRNF_DONT_TAKE_PARENT) ) {
+			struct net_device *parent = bridge_parent(realindev);
+			if (parent)
+				realindev = parent;
+		}
 
 		NF_HOOK_THRESH(pf, NF_IP_FORWARD, skb, realindev,
 			       realoutdev, br_nf_local_out_finish,
@@ -743,6 +763,9 @@ static unsigned int br_nf_post_routing(unsigned int hook, struct sk_buff **pskb,
 	if (!nf_bridge)
 		return NF_ACCEPT;
 
+	if (!realoutdev)
+		return NF_DROP;
+
 	if (skb->protocol == __constant_htons(ETH_P_IP) || IS_VLAN_IP)
 		pf = PF_INET;
 	else

From bab1deea308afcf9200837d6ac20aefe92972efb Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Thu, 9 Feb 2006 17:10:12 -0800
Subject: [PATCH 055/509] [BRIDGE]: fix error handling for add interface to
 bridge

Refactor how the bridge code interacts with kobject system.
It should still use kobjects even if not using sysfs.
Fix the error unwind handling in br_add_if.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_if.c       | 77 ++++++++++++++++++++++++++++------------
 net/bridge/br_private.h  |  5 +--
 net/bridge/br_sysfs_if.c | 50 ++------------------------
 3 files changed, 59 insertions(+), 73 deletions(-)

diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 70b7ef917234..7fa3a5a9971f 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -106,6 +106,20 @@ done:
 	rtnl_unlock();
 }
 
+static void release_nbp(struct kobject *kobj)
+{
+	struct net_bridge_port *p
+		= container_of(kobj, struct net_bridge_port, kobj);
+	kfree(p);
+}
+
+static struct kobj_type brport_ktype = {
+#ifdef CONFIG_SYSFS
+	.sysfs_ops = &brport_sysfs_ops,
+#endif
+	.release = release_nbp,
+};
+
 static void destroy_nbp(struct net_bridge_port *p)
 {
 	struct net_device *dev = p->dev;
@@ -114,7 +128,7 @@ static void destroy_nbp(struct net_bridge_port *p)
 	p->dev = NULL;
 	dev_put(dev);
 
-	br_sysfs_freeif(p);
+	kobject_put(&p->kobj);
 }
 
 static void destroy_nbp_rcu(struct rcu_head *head)
@@ -138,6 +152,8 @@ static void del_nbp(struct net_bridge_port *p)
 	struct net_bridge *br = p->br;
 	struct net_device *dev = p->dev;
 
+	sysfs_remove_link(&br->ifobj, dev->name);
+
 	dev_set_promiscuity(dev, -1);
 
 	cancel_delayed_work(&p->carrier_check);
@@ -152,6 +168,8 @@ static void del_nbp(struct net_bridge_port *p)
 
 	rcu_assign_pointer(dev->br_port, NULL);
 
+	kobject_del(&p->kobj);
+
 	call_rcu(&p->rcu, destroy_nbp_rcu);
 }
 
@@ -161,7 +179,6 @@ static void del_br(struct net_bridge *br)
 	struct net_bridge_port *p, *n;
 
 	list_for_each_entry_safe(p, n, &br->port_list, list) {
-		br_sysfs_removeif(p);
 		del_nbp(p);
 	}
 
@@ -261,6 +278,11 @@ static struct net_bridge_port *new_nbp(struct net_bridge *br,
 	INIT_WORK(&p->carrier_check, port_carrier_check, dev);
 	kobject_init(&p->kobj);
 
+	kobject_set_name(&p->kobj, SYSFS_BRIDGE_PORT_ATTR);
+	p->kobj.ktype = &brport_ktype;
+	p->kobj.parent = &(dev->class_dev.kobj);
+	p->kobj.kset = NULL;
+
 	return p;
 }
 
@@ -388,31 +410,43 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
 	if (dev->br_port != NULL)
 		return -EBUSY;
 
-	if (IS_ERR(p = new_nbp(br, dev)))
+	p = new_nbp(br, dev);
+	if (IS_ERR(p))
 		return PTR_ERR(p);
 
- 	if ((err = br_fdb_insert(br, p, dev->dev_addr)))
-		destroy_nbp(p);
- 
-	else if ((err = br_sysfs_addif(p)))
-		del_nbp(p);
-	else {
-		rcu_assign_pointer(dev->br_port, p);
-		dev_set_promiscuity(dev, 1);
+	err = kobject_add(&p->kobj);
+	if (err)
+		goto err0;
 
-		list_add_rcu(&p->list, &br->port_list);
+ 	err = br_fdb_insert(br, p, dev->dev_addr);
+	if (err)
+		goto err1;
 
-		spin_lock_bh(&br->lock);
-		br_stp_recalculate_bridge_id(br);
-		br_features_recompute(br);
-		if ((br->dev->flags & IFF_UP) 
-		    && (dev->flags & IFF_UP) && netif_carrier_ok(dev))
-			br_stp_enable_port(p);
-		spin_unlock_bh(&br->lock);
+	err = br_sysfs_addif(p);
+	if (err)
+		goto err2;
 
-		dev_set_mtu(br->dev, br_min_mtu(br));
-	}
+	rcu_assign_pointer(dev->br_port, p);
+	dev_set_promiscuity(dev, 1);
 
+	list_add_rcu(&p->list, &br->port_list);
+
+	spin_lock_bh(&br->lock);
+	br_stp_recalculate_bridge_id(br);
+	br_features_recompute(br);
+	schedule_delayed_work(&p->carrier_check, BR_PORT_DEBOUNCE);
+	spin_unlock_bh(&br->lock);
+
+	dev_set_mtu(br->dev, br_min_mtu(br));
+	kobject_uevent(&p->kobj, KOBJ_ADD);
+
+	return 0;
+err2:
+	br_fdb_delete_by_port(br, p);
+err1:
+	kobject_del(&p->kobj);
+err0:
+	kobject_put(&p->kobj);
 	return err;
 }
 
@@ -424,7 +458,6 @@ int br_del_if(struct net_bridge *br, struct net_device *dev)
 	if (!p || p->br != br) 
 		return -EINVAL;
 
-	br_sysfs_removeif(p);
 	del_nbp(p);
 
 	spin_lock_bh(&br->lock);
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index c5bd631ffcd5..8f10e09f251b 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -232,9 +232,8 @@ extern void (*br_fdb_put_hook)(struct net_bridge_fdb_entry *ent);
 
 #ifdef CONFIG_SYSFS
 /* br_sysfs_if.c */
+extern struct sysfs_ops brport_sysfs_ops;
 extern int br_sysfs_addif(struct net_bridge_port *p);
-extern void br_sysfs_removeif(struct net_bridge_port *p);
-extern void br_sysfs_freeif(struct net_bridge_port *p);
 
 /* br_sysfs_br.c */
 extern int br_sysfs_addbr(struct net_device *dev);
@@ -243,8 +242,6 @@ extern void br_sysfs_delbr(struct net_device *dev);
 #else
 
 #define br_sysfs_addif(p)	(0)
-#define br_sysfs_removeif(p)	do { } while(0)
-#define br_sysfs_freeif(p)	kfree(p)
 #define br_sysfs_addbr(dev)	(0)
 #define br_sysfs_delbr(dev)	do { } while(0)
 #endif /* CONFIG_SYSFS */
diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c
index 0ac0355d16dd..c51c9e42aeb3 100644
--- a/net/bridge/br_sysfs_if.c
+++ b/net/bridge/br_sysfs_if.c
@@ -195,23 +195,11 @@ static ssize_t brport_store(struct kobject * kobj,
 	return ret;
 }
 
-/* called from kobject_put when port ref count goes to zero. */
-static void brport_release(struct kobject *kobj)
-{
-	kfree(container_of(kobj, struct net_bridge_port, kobj));
-}
-
-static struct sysfs_ops brport_sysfs_ops = {
+struct sysfs_ops brport_sysfs_ops = {
 	.show = brport_show,
 	.store = brport_store,
 };
 
-static struct kobj_type brport_ktype = {
-	.sysfs_ops = &brport_sysfs_ops,
-	.release = brport_release,
-};
-
-
 /*
  * Add sysfs entries to ethernet device added to a bridge.
  * Creates a brport subdirectory with bridge attributes.
@@ -223,17 +211,6 @@ int br_sysfs_addif(struct net_bridge_port *p)
 	struct brport_attribute **a;
 	int err;
 
-	ASSERT_RTNL();
-
-	kobject_set_name(&p->kobj, SYSFS_BRIDGE_PORT_ATTR);
-	p->kobj.ktype = &brport_ktype;
-	p->kobj.parent = &(p->dev->class_dev.kobj);
-	p->kobj.kset = NULL;
-
-	err = kobject_add(&p->kobj);
-	if(err)
-		goto out1;
-
 	err = sysfs_create_link(&p->kobj, &br->dev->class_dev.kobj, 
 				SYSFS_BRIDGE_PORT_LINK);
 	if (err)
@@ -245,28 +222,7 @@ int br_sysfs_addif(struct net_bridge_port *p)
 			goto out2;
 	}
 
-	err = sysfs_create_link(&br->ifobj, &p->kobj, p->dev->name);
-	if (err)
-		goto out2;
-
-	kobject_uevent(&p->kobj, KOBJ_ADD);
-	return 0;
- out2:
-	kobject_del(&p->kobj);
- out1:
+	err= sysfs_create_link(&br->ifobj, &p->kobj, p->dev->name);
+out2:
 	return err;
 }
-
-void br_sysfs_removeif(struct net_bridge_port *p)
-{
-	pr_debug("br_sysfs_removeif\n");
-	sysfs_remove_link(&p->br->ifobj, p->dev->name);
-	kobject_uevent(&p->kobj, KOBJ_REMOVE);
-	kobject_del(&p->kobj);
-}
-
-void br_sysfs_freeif(struct net_bridge_port *p)
-{
-	pr_debug("br_sysfs_freeif\n");
-	kobject_put(&p->kobj);
-}

From 00adbf62bd16f6527e046b422349a54d783a3d86 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@gate.crashing.org>
Date: Mon, 16 Jan 2006 10:53:22 -0600
Subject: [PATCH 056/509] [PATCH] powerpc: Add CONFIG_DEFAULT_UIMAGE for
 embedded boards

Embedded boards that u-boot require a kernel image in the uImage format.
This allows a given board to specify it wants a uImage built by default.

This also fixes a warning at config time, as this symbol is referred
to in arch/powerpc/platforms/83xx/Kconfig.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/Kconfig  | 6 ++++++
 arch/powerpc/Makefile | 1 +
 2 files changed, 7 insertions(+)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index df338c5cc910..80d114a3a837 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -83,6 +83,12 @@ config GENERIC_TBSYNC
 	default y if PPC32 && SMP
 	default n
 
+config DEFAULT_UIMAGE
+	bool
+	help
+	  Used to allow a board to specify it wants a uImage built by default
+	default n
+
 menu "Processor support"
 choice
 	prompt "Processor Type"
diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile
index 44dd82b791d1..15fc3e98ac5c 100644
--- a/arch/powerpc/Makefile
+++ b/arch/powerpc/Makefile
@@ -142,6 +142,7 @@ drivers-$(CONFIG_OPROFILE)	+= arch/powerpc/oprofile/
 # Default to zImage, override when needed
 defaultimage-y			:= zImage
 defaultimage-$(CONFIG_PPC_ISERIES) := vmlinux
+defaultimage-$(CONFIG_DEFAULT_UIMAGE) := uImage
 KBUILD_IMAGE := $(defaultimage-y)
 all: $(KBUILD_IMAGE)
 

From a2000572ad511f5f43091ed7bd2cc3b913104a1e Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 10 Feb 2006 01:51:02 -0800
Subject: [PATCH 057/509] [PATCH] sched: remove smpnice

I don't think the code is quite ready, which is why I asked for Peter's
additions to also be merged before I acked it (although it turned out that
it still isn't quite ready with his additions either).

Basically I have had similar observations to Suresh in that it does not
play nicely with the rest of the balancing infrastructure (and raised
similar concerns in my review).

The samples (group of 4) I got for "maximum recorded imbalance" on a 2x2
SMP+HT Xeon are as follows:

            | Following boot | hackbench 20        | hackbench 40
 -----------+----------------+---------------------+---------------------
 2.6.16-rc2 | 30,37,100,112  | 5600,5530,6020,6090 | 6390,7090,8760,8470
 +nosmpnice |  3, 2,  4,  2  |   28, 150, 294, 132 |  348, 348, 294, 347

Hackbench raw performance is down around 15% with smpnice (but that in
itself isn't a huge deal because it is just a benchmark).  However, the
samples show that the imbalance passed into move_tasks is increased by
about a factor of 10-30.  I think this would also go some way to explaining
latency blips turning up in the balancing code (though I haven't actually
measured that).

We'll probably have to revert this in the SUSE kernel.

Cc: "Siddha, Suresh B" <suresh.b.siddha@intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Peter Williams <pwil3058@bigpond.net.au>
Cc: "Martin J. Bligh" <mbligh@aracnet.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/sched.c | 137 ++++++++-----------------------------------------
 1 file changed, 22 insertions(+), 115 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index bc38804e40dd..87d93be336a1 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -215,7 +215,6 @@ struct runqueue {
 	 */
 	unsigned long nr_running;
 #ifdef CONFIG_SMP
-	unsigned long prio_bias;
 	unsigned long cpu_load[3];
 #endif
 	unsigned long long nr_switches;
@@ -669,68 +668,13 @@ static int effective_prio(task_t *p)
 	return prio;
 }
 
-#ifdef CONFIG_SMP
-static inline void inc_prio_bias(runqueue_t *rq, int prio)
-{
-	rq->prio_bias += MAX_PRIO - prio;
-}
-
-static inline void dec_prio_bias(runqueue_t *rq, int prio)
-{
-	rq->prio_bias -= MAX_PRIO - prio;
-}
-
-static inline void inc_nr_running(task_t *p, runqueue_t *rq)
-{
-	rq->nr_running++;
-	if (rt_task(p)) {
-		if (p != rq->migration_thread)
-			/*
-			 * The migration thread does the actual balancing. Do
-			 * not bias by its priority as the ultra high priority
-			 * will skew balancing adversely.
-			 */
-			inc_prio_bias(rq, p->prio);
-	} else
-		inc_prio_bias(rq, p->static_prio);
-}
-
-static inline void dec_nr_running(task_t *p, runqueue_t *rq)
-{
-	rq->nr_running--;
-	if (rt_task(p)) {
-		if (p != rq->migration_thread)
-			dec_prio_bias(rq, p->prio);
-	} else
-		dec_prio_bias(rq, p->static_prio);
-}
-#else
-static inline void inc_prio_bias(runqueue_t *rq, int prio)
-{
-}
-
-static inline void dec_prio_bias(runqueue_t *rq, int prio)
-{
-}
-
-static inline void inc_nr_running(task_t *p, runqueue_t *rq)
-{
-	rq->nr_running++;
-}
-
-static inline void dec_nr_running(task_t *p, runqueue_t *rq)
-{
-	rq->nr_running--;
-}
-#endif
-
 /*
  * __activate_task - move a task to the runqueue.
  */
 static inline void __activate_task(task_t *p, runqueue_t *rq)
 {
 	enqueue_task(p, rq->active);
-	inc_nr_running(p, rq);
+	rq->nr_running++;
 }
 
 /*
@@ -739,7 +683,7 @@ static inline void __activate_task(task_t *p, runqueue_t *rq)
 static inline void __activate_idle_task(task_t *p, runqueue_t *rq)
 {
 	enqueue_task_head(p, rq->active);
-	inc_nr_running(p, rq);
+	rq->nr_running++;
 }
 
 static int recalc_task_prio(task_t *p, unsigned long long now)
@@ -863,7 +807,7 @@ static void activate_task(task_t *p, runqueue_t *rq, int local)
  */
 static void deactivate_task(struct task_struct *p, runqueue_t *rq)
 {
-	dec_nr_running(p, rq);
+	rq->nr_running--;
 	dequeue_task(p, p->array);
 	p->array = NULL;
 }
@@ -1007,61 +951,27 @@ void kick_process(task_t *p)
  * We want to under-estimate the load of migration sources, to
  * balance conservatively.
  */
-static unsigned long __source_load(int cpu, int type, enum idle_type idle)
-{
-	runqueue_t *rq = cpu_rq(cpu);
-	unsigned long running = rq->nr_running;
-	unsigned long source_load, cpu_load = rq->cpu_load[type-1],
-		load_now = running * SCHED_LOAD_SCALE;
-
-	if (type == 0)
-		source_load = load_now;
-	else
-		source_load = min(cpu_load, load_now);
-
-	if (running > 1 || (idle == NOT_IDLE && running))
-		/*
-		 * If we are busy rebalancing the load is biased by
-		 * priority to create 'nice' support across cpus. When
-		 * idle rebalancing we should only bias the source_load if
-		 * there is more than one task running on that queue to
-		 * prevent idle rebalance from trying to pull tasks from a
-		 * queue with only one running task.
-		 */
-		source_load = source_load * rq->prio_bias / running;
-
-	return source_load;
-}
-
 static inline unsigned long source_load(int cpu, int type)
 {
-	return __source_load(cpu, type, NOT_IDLE);
+	runqueue_t *rq = cpu_rq(cpu);
+	unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE;
+	if (type == 0)
+		return load_now;
+
+	return min(rq->cpu_load[type-1], load_now);
 }
 
 /*
  * Return a high guess at the load of a migration-target cpu
  */
-static inline unsigned long __target_load(int cpu, int type, enum idle_type idle)
-{
-	runqueue_t *rq = cpu_rq(cpu);
-	unsigned long running = rq->nr_running;
-	unsigned long target_load, cpu_load = rq->cpu_load[type-1],
-		load_now = running * SCHED_LOAD_SCALE;
-
-	if (type == 0)
-		target_load = load_now;
-	else
-		target_load = max(cpu_load, load_now);
-
-	if (running > 1 || (idle == NOT_IDLE && running))
-		target_load = target_load * rq->prio_bias / running;
-
-	return target_load;
-}
-
 static inline unsigned long target_load(int cpu, int type)
 {
-	return __target_load(cpu, type, NOT_IDLE);
+	runqueue_t *rq = cpu_rq(cpu);
+	unsigned long load_now = rq->nr_running * SCHED_LOAD_SCALE;
+	if (type == 0)
+		return load_now;
+
+	return max(rq->cpu_load[type-1], load_now);
 }
 
 /*
@@ -1530,7 +1440,7 @@ void fastcall wake_up_new_task(task_t *p, unsigned long clone_flags)
 				list_add_tail(&p->run_list, &current->run_list);
 				p->array = current->array;
 				p->array->nr_active++;
-				inc_nr_running(p, rq);
+				rq->nr_running++;
 			}
 			set_need_resched();
 		} else
@@ -1875,9 +1785,9 @@ void pull_task(runqueue_t *src_rq, prio_array_t *src_array, task_t *p,
 	       runqueue_t *this_rq, prio_array_t *this_array, int this_cpu)
 {
 	dequeue_task(p, src_array);
-	dec_nr_running(p, src_rq);
+	src_rq->nr_running--;
 	set_task_cpu(p, this_cpu);
-	inc_nr_running(p, this_rq);
+	this_rq->nr_running++;
 	enqueue_task(p, this_array);
 	p->timestamp = (p->timestamp - src_rq->timestamp_last_tick)
 				+ this_rq->timestamp_last_tick;
@@ -2056,9 +1966,9 @@ find_busiest_group(struct sched_domain *sd, int this_cpu,
 
 			/* Bias balancing toward cpus of our domain */
 			if (local_group)
-				load = __target_load(i, load_idx, idle);
+				load = target_load(i, load_idx);
 			else
-				load = __source_load(i, load_idx, idle);
+				load = source_load(i, load_idx);
 
 			avg_load += load;
 		}
@@ -2171,7 +2081,7 @@ static runqueue_t *find_busiest_queue(struct sched_group *group,
 	int i;
 
 	for_each_cpu_mask(i, group->cpumask) {
-		load = __source_load(i, 0, idle);
+		load = source_load(i, 0);
 
 		if (load > max_load) {
 			max_load = load;
@@ -3571,10 +3481,8 @@ void set_user_nice(task_t *p, long nice)
 		goto out_unlock;
 	}
 	array = p->array;
-	if (array) {
+	if (array)
 		dequeue_task(p, array);
-		dec_prio_bias(rq, p->static_prio);
-	}
 
 	old_prio = p->prio;
 	new_prio = NICE_TO_PRIO(nice);
@@ -3584,7 +3492,6 @@ void set_user_nice(task_t *p, long nice)
 
 	if (array) {
 		enqueue_task(p, array);
-		inc_prio_bias(rq, p->static_prio);
 		/*
 		 * If the task increased its priority or is running and
 		 * lowered its priority, then reschedule its CPU:

From afcd024183d8a6eae7e489ce50b2485c5ae4f662 Mon Sep 17 00:00:00 2001
From: Jesper Juhl <jesper.juhl@gmail.com>
Date: Fri, 10 Feb 2006 01:51:03 -0800
Subject: [PATCH 058/509] [PATCH] wrong firmware location in IPW2100 Kconfig
 entry

Firmware should go into /lib/firmware, not /etc/firmware.

Found by Alejandro Bonilla.

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Zhu Yi <yi.zhu@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/net/wireless/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/wireless/Kconfig b/drivers/net/wireless/Kconfig
index 233a4f608084..ef85d76575a2 100644
--- a/drivers/net/wireless/Kconfig
+++ b/drivers/net/wireless/Kconfig
@@ -148,7 +148,7 @@ config IPW2100
 	  In order to use this driver, you will need a firmware image for it.
           You can obtain the firmware from
 	  <http://ipw2100.sf.net/>.  Once you have the firmware image, you 
-	  will need to place it in /etc/firmware.
+	  will need to place it in /lib/firmware.
 
           You will also very likely need the Wireless Tools in order to
           configure your card:

From 9c15e852a524d55ab768cf48c97f5c684f876af2 Mon Sep 17 00:00:00 2001
From: Haren Myneni <haren@us.ibm.com>
Date: Fri, 10 Feb 2006 01:51:05 -0800
Subject: [PATCH 059/509] [PATCH] kexec: fix in free initrd when overlapped
 with crashkernel region

It is possible that the reserved crashkernel region can be overlapped with
initrd since the bootloader sets the initrd location.  When the initrd
region is freed, the second kernel memory will not be contiguous.  The
Kexec_load can cause an oops since there is no contiguous memory to write
the second kernel or this memory could be used in the first kernel itself
and may not be part of the dump.  For example, on powerpc, the initrd is
located at 36MB and the crashkernel starts at 32MB.  The kexec_load caused
panic since writing into non-allocated memory (after 36MB).  We could see
the similar issue even on other archs.

One possibility is to move the initrd outside of crashkernel region.  But,
the initrd region will be freed anyway before the system is up.  This patch
fixes this issue and frees only regions that are not part of crashkernel
memory in case overlaps.

Signed-off-by: Haren Myneni <haren@us.ibm.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Vivek Goyal <vgoyal@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/kexec.h |  1 +
 init/initramfs.c      | 24 +++++++++++++++++++++++-
 2 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/include/linux/kexec.h b/include/linux/kexec.h
index a311f58c8a7c..cfb3410e32b1 100644
--- a/include/linux/kexec.h
+++ b/include/linux/kexec.h
@@ -6,6 +6,7 @@
 #include <linux/list.h>
 #include <linux/linkage.h>
 #include <linux/compat.h>
+#include <linux/ioport.h>
 #include <asm/kexec.h>
 
 /* Verify architecture specific macros are defined */
diff --git a/init/initramfs.c b/init/initramfs.c
index 0c5d9a3f951b..637344b05981 100644
--- a/init/initramfs.c
+++ b/init/initramfs.c
@@ -466,10 +466,32 @@ static char * __init unpack_to_rootfs(char *buf, unsigned len, int check_only)
 extern char __initramfs_start[], __initramfs_end[];
 #ifdef CONFIG_BLK_DEV_INITRD
 #include <linux/initrd.h>
+#include <linux/kexec.h>
 
 static void __init free_initrd(void)
 {
-	free_initrd_mem(initrd_start, initrd_end);
+#ifdef CONFIG_KEXEC
+	unsigned long crashk_start = (unsigned long)__va(crashk_res.start);
+	unsigned long crashk_end   = (unsigned long)__va(crashk_res.end);
+
+	/*
+	 * If the initrd region is overlapped with crashkernel reserved region,
+	 * free only memory that is not part of crashkernel region.
+	 */
+	if (initrd_start < crashk_end && initrd_end > crashk_start) {
+		/*
+		 * Initialize initrd memory region since the kexec boot does
+		 * not do.
+		 */
+		memset((void *)initrd_start, 0, initrd_end - initrd_start);
+		if (initrd_start < crashk_start)
+			free_initrd_mem(initrd_start, crashk_start);
+		if (initrd_end > crashk_end)
+			free_initrd_mem(crashk_end, initrd_end);
+	} else
+#endif
+		free_initrd_mem(initrd_start, initrd_end);
+
 	initrd_start = 0;
 	initrd_end = 0;
 }

From 02df360bf38ca2acb78ddee9fd28262e9474153c Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 10 Feb 2006 01:51:06 -0800
Subject: [PATCH 060/509] [PATCH] remove bogus comment from init/main.c

Remove bogus comment from init function which could lead to the assumption
that cpu_possible_map is setup in smp_prepare_cpus().

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 init/main.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/init/main.c b/init/main.c
index 7c79da57d3a2..4c194c47395f 100644
--- a/init/main.c
+++ b/init/main.c
@@ -668,7 +668,6 @@ static int init(void * unused)
 	 */
 	child_reaper = current;
 
-	/* Sets up cpus_possible() */
 	smp_prepare_cpus(max_cpus);
 
 	do_pre_smp_initcalls();

From 705672935f8a9e661264e34cd8c409e8cddcc7db Mon Sep 17 00:00:00 2001
From: Andreas Gruenbacher <agruen@suse.de>
Date: Fri, 10 Feb 2006 01:51:07 -0800
Subject: [PATCH 061/509] [PATCH] Fix building external modules on ppc32

We are setting up sources for building external modules like this:

  /usr/src/linux-obj> # create a .config file
  /usr/src/linux-obj> make -C /usr/src/linux O=$PWD oldconfig
  /usr/src/linux-obj> make -C /usr/src/linux O=$PWD prepare
  /usr/src/linux-obj> make -C /usr/src/linux O=$PWD scripts
  /usr/src/linux-obj> make -C /usr/src/linux O=$PWD clean

After that, external modules can be built with:

  /usr/src/module> make -C /usr/src/linux-obj M=$PWD

This fails for ppc32 because the `make clean' removes the
arch/powerpc/include directory.  This should be done in archmrproper
instead of in archclean.

Signed-off-by: Andreas Gruenbacher <agruen@suse.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/Makefile | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile
index 44dd82b791d1..efcad7601a7a 100644
--- a/arch/powerpc/Makefile
+++ b/arch/powerpc/Makefile
@@ -167,6 +167,8 @@ endef
 
 archclean:
 	$(Q)$(MAKE) $(clean)=$(boot)
+
+archmrproper:
 	$(Q)rm -rf arch/$(ARCH)/include
 
 archprepare: checkbin

From 7a8ef1cb774e5438d292365626f9b96616283706 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Fri, 10 Feb 2006 01:51:08 -0800
Subject: [PATCH 062/509] [PATCH] x86: don't initialise cpu_possible_map to all
 ones

Initialising cpu_possible_map to all-ones with CONFIG_HOTPLUG_CPU means that

a) All for_each_cpu() loops will iterate across all NR_CPUS CPUs, rather
   than over possible ones.  That can be quite expensive.

b) Soon we'll be allocating per-cpu areas only for possible CPUs.  So with
   CPU_MASK_ALL, we'll be wasting memory.

I also switched voyager over to not use CPU_MASK_ALL in the non-CPU-hotplug
case.  Should be OK..

I note that parisc is also using CPU_MASK_ALL.  Suggest that it stop doing
that.

Cc: James Bottomley <James.Bottomley@steeleye.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Paul Jackson <pj@sgi.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Cc: Zwane Mwaikambo <zwane@linuxpower.ca>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/smpboot.c           | 4 ----
 arch/i386/mach-voyager/voyager_smp.c | 2 +-
 include/linux/cpumask.h              | 2 +-
 3 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 255adb498268..fb00ab7b7612 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -87,11 +87,7 @@ EXPORT_SYMBOL(cpu_online_map);
 cpumask_t cpu_callin_map;
 cpumask_t cpu_callout_map;
 EXPORT_SYMBOL(cpu_callout_map);
-#ifdef CONFIG_HOTPLUG_CPU
-cpumask_t cpu_possible_map = CPU_MASK_ALL;
-#else
 cpumask_t cpu_possible_map;
-#endif
 EXPORT_SYMBOL(cpu_possible_map);
 static cpumask_t smp_commenced_mask;
 
diff --git a/arch/i386/mach-voyager/voyager_smp.c b/arch/i386/mach-voyager/voyager_smp.c
index 72a1b9cae2e4..6e4c3baef6cc 100644
--- a/arch/i386/mach-voyager/voyager_smp.c
+++ b/arch/i386/mach-voyager/voyager_smp.c
@@ -240,7 +240,7 @@ static cpumask_t smp_commenced_mask = CPU_MASK_NONE;
 cpumask_t cpu_callin_map = CPU_MASK_NONE;
 cpumask_t cpu_callout_map = CPU_MASK_NONE;
 EXPORT_SYMBOL(cpu_callout_map);
-cpumask_t cpu_possible_map = CPU_MASK_ALL;
+cpumask_t cpu_possible_map = CPU_MASK_NONE;
 EXPORT_SYMBOL(cpu_possible_map);
 
 /* The per processor IRQ masks (these are usually kept in sync) */
diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
index 20b446f26ecd..60e56c6e03dd 100644
--- a/include/linux/cpumask.h
+++ b/include/linux/cpumask.h
@@ -328,7 +328,7 @@ static inline void __cpus_remap(cpumask_t *dstp, const cpumask_t *srcp,
  * bitmap of size NR_CPUS.
  *
  *  #ifdef CONFIG_HOTPLUG_CPU
- *     cpu_possible_map - all NR_CPUS bits set
+ *     cpu_possible_map - has bit 'cpu' set iff cpu is populatable
  *     cpu_present_map  - has bit 'cpu' set iff cpu is populated
  *     cpu_online_map   - has bit 'cpu' set iff cpu available to scheduler
  *  #else

From c22db9412736204b25aeba19d18e5ea922f7d632 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Fri, 10 Feb 2006 01:51:11 -0800
Subject: [PATCH 063/509] [PATCH] prevent recursive panic from softlockup
 watchdog

When panic_timeout is zero, suppress triggering a nested panic due to soft
lockup detection.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/panic.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/panic.c b/kernel/panic.c
index c5c4ab255834..126dc43f1c74 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -130,6 +130,7 @@ NORET_TYPE void panic(const char * fmt, ...)
 #endif
 	local_irq_enable();
 	for (i = 0;;) {
+		touch_softlockup_watchdog();
 		i += panic_blink(i);
 		mdelay(1);
 		i++;

From 8e36709d8cea48a4d341294ce2b46678a2e77159 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 10 Feb 2006 01:51:12 -0800
Subject: [PATCH 064/509] [PATCH] shmdt cannot detach not-alined shm segment
 cleanly.

sys_shmdt() can manage shm segments which are covered by multiple vmas.  (This
can happen when a user uses mprotect() after shmat().)

This works well if shm is aligned to PAGE_SIZE, but if not, the last
segment cannot be detached.  It is because a comparison in sys_shmdt()

	(vma->vm_end - addr) < size
		addr == return address of shmat()
		size == shmsize, argments to shmget()

size should be aligned to PAGE_SIZE before being compared with vma->vm_end,
which is aligned.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Acked-by: Hugh Dickins <hugh@veritas.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 ipc/shm.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/ipc/shm.c b/ipc/shm.c
index 4c28d2d8e305..9162123a7b23 100644
--- a/ipc/shm.c
+++ b/ipc/shm.c
@@ -870,6 +870,7 @@ asmlinkage long sys_shmdt(char __user *shmaddr)
 	 * could possibly have landed at. Also cast things to loff_t to
 	 * prevent overflows and make comparisions vs. equal-width types.
 	 */
+	size = PAGE_ALIGN(size);
 	while (vma && (loff_t)(vma->vm_end - addr) <= size) {
 		next = vma->vm_next;
 

From f0188f47482efdbd2e005103bb4f0224a835dfad Mon Sep 17 00:00:00 2001
From: Ravikiran G Thirumalai <kiran@scalex86.org>
Date: Fri, 10 Feb 2006 01:51:13 -0800
Subject: [PATCH 065/509] [PATCH] slab: Avoid deadlock at
 kmem_cache_create/kmem_cache_destroy

Prevents deadlock situation between
kmem_cache_create()/kmem_cache_destory(), and kmem_cache_create() /cpu
hotplug.  The locking order probably got moved over time.

Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/slab.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/mm/slab.c b/mm/slab.c
index d66c2b0d9715..add05d808a4a 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -1717,6 +1717,12 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 		BUG();
 	}
 
+	/*
+	 * Prevent CPUs from coming and going.
+	 * lock_cpu_hotplug() nests outside cache_chain_mutex
+	 */
+	lock_cpu_hotplug();
+
 	mutex_lock(&cache_chain_mutex);
 
 	list_for_each(p, &cache_chain) {
@@ -1918,8 +1924,6 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 	cachep->dtor = dtor;
 	cachep->name = name;
 
-	/* Don't let CPUs to come and go */
-	lock_cpu_hotplug();
 
 	if (g_cpucache_up == FULL) {
 		enable_cpucache(cachep);
@@ -1978,12 +1982,12 @@ kmem_cache_create (const char *name, size_t size, size_t align,
 
 	/* cache setup completed, link it into the list */
 	list_add(&cachep->next, &cache_chain);
-	unlock_cpu_hotplug();
       oops:
 	if (!cachep && (flags & SLAB_PANIC))
 		panic("kmem_cache_create(): failed to create slab `%s'\n",
 		      name);
 	mutex_unlock(&cache_chain_mutex);
+	unlock_cpu_hotplug();
 	return cachep;
 }
 EXPORT_SYMBOL(kmem_cache_create);

From 8977d929e49021d9a6e031310aab01fa72f849c2 Mon Sep 17 00:00:00 2001
From: Paul Fulghum <paulkf@microgate.com>
Date: Fri, 10 Feb 2006 01:51:14 -0800
Subject: [PATCH 066/509] [PATCH] tty buffering stall fix

Prevent stalled processing of received data when a driver allocates tty
buffer space but does not immediately follow the allocation with more data
and a call to schedule receive tty processing.  (example: hvc_console) This
bug was introduced by the first locking patch for the new tty buffering.

Signed-off-by: Paul Fulghum <paulkf@microgate.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/tty_io.c    | 30 ++++++++++++++++++++++--------
 include/linux/kbd_kern.h |  4 +++-
 include/linux/tty.h      |  2 ++
 include/linux/tty_flip.h |  4 +++-
 4 files changed, 30 insertions(+), 10 deletions(-)

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 076e07c1da38..a23816d3e9a1 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -268,6 +268,8 @@ static struct tty_buffer *tty_buffer_alloc(size_t size)
 	p->size = size;
 	p->next = NULL;
 	p->active = 0;
+	p->commit = 0;
+	p->read = 0;
 	p->char_buf_ptr = (char *)(p->data);
 	p->flag_buf_ptr = (unsigned char *)p->char_buf_ptr + size;
 /* 	printk("Flip create %p\n", p); */
@@ -298,6 +300,8 @@ static struct tty_buffer *tty_buffer_find(struct tty_struct *tty, size_t size)
 			*tbh = t->next;
 			t->next = NULL;
 			t->used = 0;
+			t->commit = 0;
+			t->read = 0;
 			/* DEBUG ONLY */
 			memset(t->data, '*', size);
 /* 			printk("Flip recycle %p\n", t); */
@@ -335,6 +339,7 @@ int tty_buffer_request_room(struct tty_struct *tty, size_t size)
 			if (b != NULL) {
 				b->next = n;
 				b->active = 0;
+				b->commit = b->used;
 			} else
 				tty->buf.head = n;
 			tty->buf.tail = n;
@@ -2752,6 +2757,9 @@ static void flush_to_ldisc(void *private_)
 	unsigned long 	flags;
 	struct tty_ldisc *disc;
 	struct tty_buffer *tbuf;
+	int count;
+	char *char_buf;
+	unsigned char *flag_buf;
 
 	disc = tty_ldisc_ref(tty);
 	if (disc == NULL)	/*  !TTY_LDISC */
@@ -2765,16 +2773,20 @@ static void flush_to_ldisc(void *private_)
 		goto out;
 	}
 	spin_lock_irqsave(&tty->buf.lock, flags);
-	while((tbuf = tty->buf.head) != NULL && !tbuf->active) {
+	while((tbuf = tty->buf.head) != NULL) {
+		while ((count = tbuf->commit - tbuf->read) != 0) {
+			char_buf = tbuf->char_buf_ptr + tbuf->read;
+			flag_buf = tbuf->flag_buf_ptr + tbuf->read;
+			tbuf->read += count;
+			spin_unlock_irqrestore(&tty->buf.lock, flags);
+			disc->receive_buf(tty, char_buf, flag_buf, count);
+			spin_lock_irqsave(&tty->buf.lock, flags);
+		}
+		if (tbuf->active)
+			break;
 		tty->buf.head = tbuf->next;
 		if (tty->buf.head == NULL)
 			tty->buf.tail = NULL;
-		spin_unlock_irqrestore(&tty->buf.lock, flags);
-		/* printk("Process buffer %p for %d\n", tbuf, tbuf->used); */
-		disc->receive_buf(tty, tbuf->char_buf_ptr,
-				       tbuf->flag_buf_ptr,
-				       tbuf->used);
-		spin_lock_irqsave(&tty->buf.lock, flags);
 		tty_buffer_free(tty, tbuf);
 	}
 	spin_unlock_irqrestore(&tty->buf.lock, flags);
@@ -2871,8 +2883,10 @@ void tty_flip_buffer_push(struct tty_struct *tty)
 {
 	unsigned long flags;
 	spin_lock_irqsave(&tty->buf.lock, flags);
-	if (tty->buf.tail != NULL)
+	if (tty->buf.tail != NULL) {
 		tty->buf.tail->active = 0;
+		tty->buf.tail->commit = tty->buf.tail->used;
+	}
 	spin_unlock_irqrestore(&tty->buf.lock, flags);
 
 	if (tty->low_latency)
diff --git a/include/linux/kbd_kern.h b/include/linux/kbd_kern.h
index 3aed37314ab8..e87c32a5c86a 100644
--- a/include/linux/kbd_kern.h
+++ b/include/linux/kbd_kern.h
@@ -153,8 +153,10 @@ static inline void con_schedule_flip(struct tty_struct *t)
 {
 	unsigned long flags;
 	spin_lock_irqsave(&t->buf.lock, flags);
-	if (t->buf.tail != NULL)
+	if (t->buf.tail != NULL) {
 		t->buf.tail->active = 0;
+		t->buf.tail->commit = t->buf.tail->used;
+	}
 	spin_unlock_irqrestore(&t->buf.lock, flags);
 	schedule_work(&t->buf.work);
 }
diff --git a/include/linux/tty.h b/include/linux/tty.h
index a7bd3b4558d2..f45cd74e6f24 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -58,6 +58,8 @@ struct tty_buffer {
 	int used;
 	int size;
 	int active;
+	int commit;
+	int read;
 	/* Data points here */
 	unsigned long data[0];
 };
diff --git a/include/linux/tty_flip.h b/include/linux/tty_flip.h
index 82961eb19888..222faf97d5f9 100644
--- a/include/linux/tty_flip.h
+++ b/include/linux/tty_flip.h
@@ -29,8 +29,10 @@ _INLINE_ void tty_schedule_flip(struct tty_struct *tty)
 {
 	unsigned long flags;
 	spin_lock_irqsave(&tty->buf.lock, flags);
-	if (tty->buf.tail != NULL)
+	if (tty->buf.tail != NULL) {
 		tty->buf.tail->active = 0;
+		tty->buf.tail->commit = tty->buf.tail->used;
+	}
 	spin_unlock_irqrestore(&tty->buf.lock, flags);
 	schedule_delayed_work(&tty->buf.work, 1);
 }

From 418aade459f03318defd18ef0b11981a63bd81b0 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Fri, 10 Feb 2006 01:51:15 -0800
Subject: [PATCH 067/509] [PATCH] Updates for page migration

This adds some additional comments in order to help others figure out how
exactly the code works.  And fix a variable name.

Also swap_page does need to ignore all reference bits when unmapping a
page.  Otherwise we may have to repeatedly unmap a frequently touched page.
So change the try_to_unmap parameter to 1.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 25 ++++++++++++++++++++-----
 1 file changed, 20 insertions(+), 5 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5a610804cd06..5db32fdfaf39 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -632,7 +632,7 @@ static int swap_page(struct page *page)
 	struct address_space *mapping = page_mapping(page);
 
 	if (page_mapped(page) && mapping)
-		if (try_to_unmap(page, 0) != SWAP_SUCCESS)
+		if (try_to_unmap(page, 1) != SWAP_SUCCESS)
 			goto unlock_retry;
 
 	if (PageDirty(page)) {
@@ -839,7 +839,7 @@ EXPORT_SYMBOL(migrate_page);
  * pages are swapped out.
  *
  * The function returns after 10 attempts or if no pages
- * are movable anymore because t has become empty
+ * are movable anymore because to has become empty
  * or no retryable pages exist anymore.
  *
  * Return: Number of pages not migrated when "to" ran empty.
@@ -928,12 +928,21 @@ redo:
 			goto unlock_both;
 
 		if (mapping->a_ops->migratepage) {
+			/*
+			 * Most pages have a mapping and most filesystems
+			 * should provide a migration function. Anonymous
+			 * pages are part of swap space which also has its
+			 * own migration function. This is the most common
+			 * path for page migration.
+			 */
 			rc = mapping->a_ops->migratepage(newpage, page);
 			goto unlock_both;
                 }
 
 		/*
-		 * Trigger writeout if page is dirty
+		 * Default handling if a filesystem does not provide
+		 * a migration function. We can only migrate clean
+		 * pages so try to write out any dirty pages first.
 		 */
 		if (PageDirty(page)) {
 			switch (pageout(page, mapping)) {
@@ -949,9 +958,10 @@ redo:
 				; /* try to migrate the page below */
 			}
                 }
+
 		/*
-		 * If we have no buffer or can release the buffer
-		 * then do a simple migration.
+		 * Buffers are managed in a filesystem specific way.
+		 * We must have no buffers or drop them.
 		 */
 		if (!page_has_buffers(page) ||
 		    try_to_release_page(page, GFP_KERNEL)) {
@@ -966,6 +976,11 @@ redo:
 		 * swap them out.
 		 */
 		if (pass > 4) {
+			/*
+			 * Persistently unable to drop buffers..... As a
+			 * measure of last resort we fall back to
+			 * swap_page().
+			 */
 			unlock_page(newpage);
 			newpage = NULL;
 			rc = swap_page(page);

From 1311c24fad3b2acad5c6545d2107c226774f02e8 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Fri, 10 Feb 2006 11:56:30 +0100
Subject: [PATCH 068/509] [AGPGART] help text updates

This patch contains help text updates including the following:
- XFree86 * -> X
- there is no need for repeating part of the help text of the AGP
  option and having "If unsure, say Y/N." in the chip specific
  options.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/char/agp/Kconfig | 55 +++++++++++++---------------------------
 1 file changed, 17 insertions(+), 38 deletions(-)

diff --git a/drivers/char/agp/Kconfig b/drivers/char/agp/Kconfig
index 486ed8a11b59..a4d425d2dce2 100644
--- a/drivers/char/agp/Kconfig
+++ b/drivers/char/agp/Kconfig
@@ -15,22 +15,23 @@ config AGP
 	  due to kernel allocation issues), you could use PCI accesses
 	  and have up to a couple gigs of texture space.
 
-	  Note that this is the only means to have XFree4/GLX use
+	  Note that this is the only means to have X/GLX use
 	  write-combining with MTRR support on the AGP bus. Without it, OpenGL
 	  direct rendering will be a lot slower but still faster than PIO.
 
-	  You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-	  use GLX or DRI.  If unsure, say N.
-
 	  To compile this driver as a module, choose M here: the
 	  module will be called agpgart.
 
+	  You should say Y here if you want to use GLX or DRI.
+
+	  If unsure, say N.
+
 config AGP_ALI
 	tristate "ALI chipset support"
 	depends on AGP && X86_32
 	---help---
 	  This option gives you AGP support for the GLX component of
-	  XFree86 4.x on the following ALi chipsets.  The supported chipsets
+	  X on the following ALi chipsets.  The supported chipsets
 	  include M1541, M1621, M1631, M1632, M1641,M1647,and M1651.
 	  For the ALi-chipset question, ALi suggests you refer to
 	  <http://www.ali.com.tw/eng/support/index.shtml>.
@@ -40,28 +41,19 @@ config AGP_ALI
 	  timing issues, this chipset cannot do AGP 2x with the G200.
 	  This is a hardware limitation. AGP 1x seems to be fine, though.
 
-	  You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-	  use GLX or DRI.  If unsure, say N.
-
 config AGP_ATI
 	tristate "ATI chipset support"
 	depends on AGP && X86_32
 	---help---
-      This option gives you AGP support for the GLX component of
-      XFree86 4.x on the ATI RadeonIGP family of chipsets.
-
-      You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-      use GLX or DRI.  If unsure, say N.
+	  This option gives you AGP support for the GLX component of
+	  X on the ATI RadeonIGP family of chipsets.
 
 config AGP_AMD
 	tristate "AMD Irongate, 761, and 762 chipset support"
 	depends on AGP && X86_32
 	help
 	  This option gives you AGP support for the GLX component of
-	  XFree86 4.x on AMD Irongate, 761, and 762 chipsets.
-
-	  You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-	  use GLX or DRI.  If unsure, say N.
+	  X on AMD Irongate, 761, and 762 chipsets.
 
 config AGP_AMD64
 	tristate "AMD Opteron/Athlon64 on-CPU GART support" if !GART_IOMMU
@@ -69,45 +61,38 @@ config AGP_AMD64
 	default y if GART_IOMMU
 	help
 	  This option gives you AGP support for the GLX component of
-	  XFree86 4.x using the on-CPU northbridge of the AMD Athlon64/Opteron CPUs.
+	  X using the on-CPU northbridge of the AMD Athlon64/Opteron CPUs.
 	  You still need an external AGP bridge like the AMD 8151, VIA
           K8T400M, SiS755. It may also support other AGP bridges when loaded
 	  with agp_try_unsupported=1.
-	  You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-	  use GLX or DRI.  If unsure, say Y
 
 config AGP_INTEL
 	tristate "Intel 440LX/BX/GX, I8xx and E7x05 chipset support"
 	depends on AGP && X86
 	help
-	  This option gives you AGP support for the GLX component of XFree86 4.x
+	  This option gives you AGP support for the GLX component of X
 	  on Intel 440LX/BX/GX, 815, 820, 830, 840, 845, 850, 860, 875,
-	  E7205 and E7505 chipsets and full support for the 810, 815, 830M, 845G,
-	  852GM, 855GM, 865G and I915 integrated graphics chipsets.
+	  E7205 and E7505 chipsets and full support for the 810, 815, 830M,
+	  845G, 852GM, 855GM, 865G and I915 integrated graphics chipsets.
+
 
-	  You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-	  use GLX or DRI, or if you have any Intel integrated graphics
-	  chipsets.  If unsure, say Y.
 
 config AGP_NVIDIA
 	tristate "NVIDIA nForce/nForce2 chipset support"
 	depends on AGP && X86_32
 	help
 	  This option gives you AGP support for the GLX component of
-	  XFree86 4.x on the following NVIDIA chipsets.  The supported chipsets
-	  include nForce and nForce2
+	  X on NVIDIA chipsets including nForce and nForce2
 
 config AGP_SIS
 	tristate "SiS chipset support"
 	depends on AGP && X86_32
 	help
 	  This option gives you AGP support for the GLX component of
-	  XFree86 4.x on Silicon Integrated Systems [SiS] chipsets.
+	  X on Silicon Integrated Systems [SiS] chipsets.
 
 	  Note that 5591/5592 AGP chipsets are NOT supported.
 
-	  You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-	  use GLX or DRI.  If unsure, say N.
 
 config AGP_SWORKS
 	tristate "Serverworks LE/HE chipset support"
@@ -121,10 +106,7 @@ config AGP_VIA
 	depends on AGP && X86_32
 	help
 	  This option gives you AGP support for the GLX component of
-	  XFree86 4.x on VIA MVP3/Apollo Pro chipsets.
-
-	  You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-	  use GLX or DRI.  If unsure, say N.
+	  X on VIA MVP3/Apollo Pro chipsets.
 
 config AGP_I460
 	tristate "Intel 460GX chipset support"
@@ -159,9 +141,6 @@ config AGP_EFFICEON
 	  This option gives you AGP support for the Transmeta Efficeon
 	  series processors with integrated northbridges.
 
-	  You should say Y here if you use XFree86 3.3.6 or 4.x and want to
-	  use GLX or DRI.  If unsure, say Y.
-
 config AGP_SGI_TIOCA
         tristate "SGI TIO chipset AGP support"
         depends on AGP && (IA64_SGI_SN2 || IA64_GENERIC)

From 21b4da78c941f292f6daf87abb562d6285216e51 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Fri, 10 Feb 2006 16:27:11 -0500
Subject: [PATCH 069/509] [PATCH] Fix s390 build failure.

arch/s390/kernel/compat_signal.c:199: error: conflicting types for 'do_sigaction'
include/linux/sched.h:1115: error: previous declaration of 'do_sigaction' was here

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/compat_signal.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c
index ef706694a0c1..5291b5f8788d 100644
--- a/arch/s390/kernel/compat_signal.c
+++ b/arch/s390/kernel/compat_signal.c
@@ -195,9 +195,6 @@ sys32_sigaction(int sig, const struct old_sigaction32 __user *act,
 	return ret;
 }
 
-int
-do_sigaction(int sig, const struct k_sigaction *act, struct k_sigaction *oact);
-
 asmlinkage long
 sys32_rt_sigaction(int sig, const struct sigaction32 __user *act,
 	   struct sigaction32 __user *oact,  size_t sigsetsize)

From fd401aee6273e869e2711de498e28f5208184797 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Fri, 10 Feb 2006 21:50:43 +0000
Subject: [PATCH 070/509] [SERIAL] Remove incorrect code from ioc4 serial
 driver

Serial drivers in general should not write uart_info->flags - they're
private to serial_core.  Serial drivers have no need to fiddle with
tty->alt_speed, nor manipulate TTY_IO_ERROR in tty->flags.  Fix the
ioc4 serial driver for both these points by simply removing the
offending code.

Acked-by: pfg@sgi.com
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/serial/ioc4_serial.c | 14 --------------
 1 file changed, 14 deletions(-)

diff --git a/drivers/serial/ioc4_serial.c b/drivers/serial/ioc4_serial.c
index 1d85533d46d2..f3763d2ccb86 100644
--- a/drivers/serial/ioc4_serial.c
+++ b/drivers/serial/ioc4_serial.c
@@ -1717,11 +1717,9 @@ ioc4_change_speed(struct uart_port *the_port,
 	}
 
 	if (cflag & CRTSCTS) {
-		info->flags |= ASYNC_CTS_FLOW;
 		port->ip_sscr |= IOC4_SSCR_HFC_EN;
 	}
 	else {
-		info->flags &= ~ASYNC_CTS_FLOW;
 		port->ip_sscr &= ~IOC4_SSCR_HFC_EN;
 	}
 	writel(port->ip_sscr, &port->ip_serial_regs->sscr);
@@ -1760,18 +1758,6 @@ static inline int ic4_startup_local(struct uart_port *the_port)
 
 	info = the_port->info;
 
-	if (info->tty) {
-		set_bit(TTY_IO_ERROR, &info->tty->flags);
-		clear_bit(TTY_IO_ERROR, &info->tty->flags);
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_HI)
-			info->tty->alt_speed = 57600;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_VHI)
-			info->tty->alt_speed = 115200;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_SHI)
-			info->tty->alt_speed = 230400;
-		if ((info->flags & ASYNC_SPD_MASK) == ASYNC_SPD_WARP)
-			info->tty->alt_speed = 460800;
-	}
 	local_open(port);
 
 	/* set the speed of the serial port */

From e19816808346cc1619733532a267a11dce8f8a12 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 10 Feb 2006 22:40:51 +0000
Subject: [PATCH 071/509] [ARM] 3326/1: H1940 - Control latches

Patch from Ben Dooks

Define the bits for the two board control latches
that control various items on the H1940 iPAQ.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Arnaud Patard <arnaud.patard@rtp-net.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-s3c2410/mach-h1940.c         | 29 +++++++++-
 include/asm-arm/arch-s3c2410/h1940-latch.h | 64 ++++++++++++++++++++++
 2 files changed, 91 insertions(+), 2 deletions(-)
 create mode 100644 include/asm-arm/arch-s3c2410/h1940-latch.h

diff --git a/arch/arm/mach-s3c2410/mach-h1940.c b/arch/arm/mach-s3c2410/mach-h1940.c
index 1c316f14ed94..646a3a5d33a5 100644
--- a/arch/arm/mach-s3c2410/mach-h1940.c
+++ b/arch/arm/mach-s3c2410/mach-h1940.c
@@ -46,10 +46,11 @@
 #include <asm/irq.h>
 #include <asm/mach-types.h>
 
-//#include <asm/debug-ll.h>
+
 #include <asm/arch/regs-serial.h>
 #include <asm/arch/regs-lcd.h>
 
+#include <asm/arch/h1940-latch.h>
 #include <asm/arch/fb.h>
 
 #include <linux/serial_core.h>
@@ -59,7 +60,12 @@
 #include "cpu.h"
 
 static struct map_desc h1940_iodesc[] __initdata = {
-	/* nothing here yet */
+	[0] = {
+		.virtual	= (unsigned long)H1940_LATCH,
+		.pfn		= __phys_to_pfn(H1940_PA_LATCH),
+		.length		= SZ_16K,
+		.type		= MT_DEVICE
+	},
 };
 
 #define UCON S3C2410_UCON_DEFAULT | S3C2410_UCON_UCLK
@@ -92,6 +98,25 @@ static struct s3c2410_uartcfg h1940_uartcfgs[] = {
 	}
 };
 
+/* Board control latch control */
+
+static unsigned int latch_state = H1940_LATCH_DEFAULT;
+
+void h1940_latch_control(unsigned int clear, unsigned int set)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+
+	latch_state &= ~clear;
+	latch_state |= set;
+
+	__raw_writel(latch_state, H1940_LATCH);
+
+	local_irq_restore(flags);
+}
+
+EXPORT_SYMBOL_GPL(h1940_latch_control);
 
 
 /**
diff --git a/include/asm-arm/arch-s3c2410/h1940-latch.h b/include/asm-arm/arch-s3c2410/h1940-latch.h
new file mode 100644
index 000000000000..c5802411f43d
--- /dev/null
+++ b/include/asm-arm/arch-s3c2410/h1940-latch.h
@@ -0,0 +1,64 @@
+/* linux/include/asm-arm/arch-s3c2410/h1940-latch.h
+ *
+ * (c) 2005 Simtec Electronics
+ *	http://armlinux.simtec.co.uk/
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ *  iPAQ H1940 series - latch definitions
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#ifndef __ASM_ARCH_H1940_LATCH_H
+#define __ASM_ARCH_H1940_LATCH_H
+
+
+#ifndef __ASSEMBLY__
+#define H1940_LATCH		((void __iomem *)0xF8000000)
+#else
+#define H1940_LATCH		0xF8000000
+#endif
+
+#define H1940_PA_LATCH		(S3C2410_CS2)
+
+/* SD layer latch */
+
+#define H1940_LATCH_SDQ1		(1<<16)
+#define H1940_LATCH_LCD_P1		(1<<17)
+#define H1940_LATCH_LCD_P2		(1<<18)
+#define H1940_LATCH_LCD_P3		(1<<19)
+#define H1940_LATCH_MAX1698_nSHUTDOWN	(1<<20)		/* LCD backlight */
+#define H1940_LATCH_LED_RED		(1<<21)
+#define H1940_LATCH_SDQ7		(1<<22)
+#define H1940_LATCH_USB_DP		(1<<23)
+
+/* CPU layer latch */
+
+#define H1940_LATCH_UDA_POWER		(1<<24)
+#define H1940_LATCH_AUDIO_POWER		(1<<25)
+#define H1940_LATCH_SM803_ENABLE	(1<<26)
+#define H1940_LATCH_LCD_P4		(1<<27)
+#define H1940_LATCH_CPUQ5		(1<<28)		/* untraced */
+#define H1940_LATCH_BLUETOOTH_POWER	(1<<29)		/* active high */
+#define H1940_LATCH_LED_GREEN		(1<<30)
+#define H1940_LATCH_LED_FLASH		(1<<31)
+
+/* default settings */
+
+#define H1940_LATCH_DEFAULT		\
+	H1940_LATCH_LCD_P4		| \
+	H1940_LATCH_SM803_ENABLE	| \
+	H1940_LATCH_SDQ1		| \
+	H1940_LATCH_LCD_P1		| \
+	H1940_LATCH_LCD_P2		| \
+	H1940_LATCH_LCD_P3		| \
+	H1940_LATCH_MAX1698_nSHUTDOWN   | \
+	H1940_LATCH_CPUQ5
+
+/* control functions */
+
+extern void h1940_latch_control(unsigned int clear, unsigned int set);
+
+#endif /* __ASM_ARCH_H1940_LATCH_H */

From f295c79b6766b25fe8c1aad88211c54d1caa7e0b Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Fri, 10 Feb 2006 18:02:44 -0800
Subject: [PATCH 072/509] IB/mthca: Don't print debugging info until we have
 all values

When debugging is enabled, the mthca_QUERY_DEV_LIM() firmware command
function prints out some of the device limits that it queries.
However the debugging prints happen before all of the fields are
extracted from the firmware response, so some of the values that get
printed are uninitialized junk.  Move the prints to the end of the
function to fix this.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mthca/mthca_cmd.c | 38 ++++++++++++-------------
 1 file changed, 19 insertions(+), 19 deletions(-)

diff --git a/drivers/infiniband/hw/mthca/mthca_cmd.c b/drivers/infiniband/hw/mthca/mthca_cmd.c
index f9b9b93dc501..2825615ce81c 100644
--- a/drivers/infiniband/hw/mthca/mthca_cmd.c
+++ b/drivers/infiniband/hw/mthca/mthca_cmd.c
@@ -1029,25 +1029,6 @@ int mthca_QUERY_DEV_LIM(struct mthca_dev *dev,
 	MTHCA_GET(size, outbox, QUERY_DEV_LIM_UAR_ENTRY_SZ_OFFSET);
 	dev_lim->uar_scratch_entry_sz = size;
 
-	mthca_dbg(dev, "Max QPs: %d, reserved QPs: %d, entry size: %d\n",
-		  dev_lim->max_qps, dev_lim->reserved_qps, dev_lim->qpc_entry_sz);
-	mthca_dbg(dev, "Max SRQs: %d, reserved SRQs: %d, entry size: %d\n",
-		  dev_lim->max_srqs, dev_lim->reserved_srqs, dev_lim->srq_entry_sz);
-	mthca_dbg(dev, "Max CQs: %d, reserved CQs: %d, entry size: %d\n",
-		  dev_lim->max_cqs, dev_lim->reserved_cqs, dev_lim->cqc_entry_sz);
-	mthca_dbg(dev, "Max EQs: %d, reserved EQs: %d, entry size: %d\n",
-		  dev_lim->max_eqs, dev_lim->reserved_eqs, dev_lim->eqc_entry_sz);
-	mthca_dbg(dev, "reserved MPTs: %d, reserved MTTs: %d\n",
-		  dev_lim->reserved_mrws, dev_lim->reserved_mtts);
-	mthca_dbg(dev, "Max PDs: %d, reserved PDs: %d, reserved UARs: %d\n",
-		  dev_lim->max_pds, dev_lim->reserved_pds, dev_lim->reserved_uars);
-	mthca_dbg(dev, "Max QP/MCG: %d, reserved MGMs: %d\n",
-		  dev_lim->max_pds, dev_lim->reserved_mgms);
-	mthca_dbg(dev, "Max CQEs: %d, max WQEs: %d, max SRQ WQEs: %d\n",
-		  dev_lim->max_cq_sz, dev_lim->max_qp_sz, dev_lim->max_srq_sz);
-
-	mthca_dbg(dev, "Flags: %08x\n", dev_lim->flags);
-
 	if (mthca_is_memfree(dev)) {
 		MTHCA_GET(field, outbox, QUERY_DEV_LIM_MAX_SRQ_SZ_OFFSET);
 		dev_lim->max_srq_sz = 1 << field;
@@ -1093,6 +1074,25 @@ int mthca_QUERY_DEV_LIM(struct mthca_dev *dev,
 		dev_lim->mpt_entry_sz = MTHCA_MPT_ENTRY_SIZE;
 	}
 
+	mthca_dbg(dev, "Max QPs: %d, reserved QPs: %d, entry size: %d\n",
+		  dev_lim->max_qps, dev_lim->reserved_qps, dev_lim->qpc_entry_sz);
+	mthca_dbg(dev, "Max SRQs: %d, reserved SRQs: %d, entry size: %d\n",
+		  dev_lim->max_srqs, dev_lim->reserved_srqs, dev_lim->srq_entry_sz);
+	mthca_dbg(dev, "Max CQs: %d, reserved CQs: %d, entry size: %d\n",
+		  dev_lim->max_cqs, dev_lim->reserved_cqs, dev_lim->cqc_entry_sz);
+	mthca_dbg(dev, "Max EQs: %d, reserved EQs: %d, entry size: %d\n",
+		  dev_lim->max_eqs, dev_lim->reserved_eqs, dev_lim->eqc_entry_sz);
+	mthca_dbg(dev, "reserved MPTs: %d, reserved MTTs: %d\n",
+		  dev_lim->reserved_mrws, dev_lim->reserved_mtts);
+	mthca_dbg(dev, "Max PDs: %d, reserved PDs: %d, reserved UARs: %d\n",
+		  dev_lim->max_pds, dev_lim->reserved_pds, dev_lim->reserved_uars);
+	mthca_dbg(dev, "Max QP/MCG: %d, reserved MGMs: %d\n",
+		  dev_lim->max_pds, dev_lim->reserved_mgms);
+	mthca_dbg(dev, "Max CQEs: %d, max WQEs: %d, max SRQ WQEs: %d\n",
+		  dev_lim->max_cq_sz, dev_lim->max_qp_sz, dev_lim->max_srq_sz);
+
+	mthca_dbg(dev, "Flags: %08x\n", dev_lim->flags);
+
 out:
 	mthca_free_mailbox(dev, mailbox);
 	return err;

From 20b83382d1c5d4d1a73fc5671261db5239d1dbb3 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Sat, 11 Feb 2006 12:22:12 -0800
Subject: [PATCH 073/509] IPoIB: Yet another fix for send-only joins

Even after the last fix, it's still possible for a send-only join to
start before the join for the broadcast group has finished.  This
could cause us to create a multicast group using attributes from the
broadcast group that haven't been initialized yet, so we would use
garbage for the Q_Key, etc.  Fix this by waiting until the broadcast
group's attached flag is set before starting send-only joins.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/ulp/ipoib/ipoib_multicast.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 932bf139a65d..a2408d7ec598 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -533,8 +533,10 @@ void ipoib_mcast_join_task(void *dev_ptr)
 	}
 
 	if (!priv->broadcast) {
-		priv->broadcast = ipoib_mcast_alloc(dev, 1);
-		if (!priv->broadcast) {
+		struct ipoib_mcast *broadcast;
+
+		broadcast = ipoib_mcast_alloc(dev, 1);
+		if (!broadcast) {
 			ipoib_warn(priv, "failed to allocate broadcast group\n");
 			mutex_lock(&mcast_mutex);
 			if (test_bit(IPOIB_MCAST_RUN, &priv->flags))
@@ -544,10 +546,11 @@ void ipoib_mcast_join_task(void *dev_ptr)
 			return;
 		}
 
-		memcpy(priv->broadcast->mcmember.mgid.raw, priv->dev->broadcast + 4,
-		       sizeof (union ib_gid));
-
 		spin_lock_irq(&priv->lock);
+		memcpy(broadcast->mcmember.mgid.raw, priv->dev->broadcast + 4,
+		       sizeof (union ib_gid));
+		priv->broadcast = broadcast;
+
 		__ipoib_mcast_add(dev, priv->broadcast);
 		spin_unlock_irq(&priv->lock);
 	}
@@ -701,7 +704,9 @@ void ipoib_mcast_send(struct net_device *dev, union ib_gid *mgid,
 	 */
 	spin_lock(&priv->lock);
 
-	if (!test_bit(IPOIB_MCAST_STARTED, &priv->flags) || !priv->broadcast) {
+	if (!test_bit(IPOIB_MCAST_STARTED, &priv->flags)	||
+	    !priv->broadcast					||
+	    !test_bit(IPOIB_MCAST_FLAG_ATTACHED, &priv->broadcast->flags)) {
 		++priv->stats.tx_dropped;
 		dev_kfree_skb_any(skb);
 		goto unlock;

From cff2b760096d1e6feaa31948e7af4abbefe47822 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Sat, 11 Feb 2006 17:55:47 -0800
Subject: [PATCH 074/509] [PATCH] fstatat64 support

The *at patches introduced fstatat and, due to inusfficient research, I
used the newfstat functions generally as the guideline.  The result is that
on 32-bit platforms we don't have all the information needed to implement
fstatat64.

This patch modifies the code to pass up 64-bit information if
__ARCH_WANT_STAT64 is defined.  I renamed the syscall entry point to make
this clear.  Other archs will continue to use the existing code.  On x86-64
the compat code is implemented using a new sys32_ function.  this is what
is done for the other stat syscalls as well.

This patch might break some other archs (those which define
__ARCH_WANT_STAT64 and which already wired up the syscall).  Yet others
might need changes to accomodate the compatibility mode.  I really don't
want to do that work because all this stat handling is a mess (more so in
glibc, but the kernel is also affected).  It should be done by the arch
maintainers.  I'll provide some stand-alone test shortly.  Those who are
eager could compile glibc and run 'make check' (no installation needed).

The patch below has been tested on x86 and x86-64.

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/syscall_table.S |  2 +-
 arch/x86_64/ia32/ia32entry.S     |  2 +-
 arch/x86_64/ia32/sys_ia32.c      | 22 ++++++++++++++++++++++
 fs/stat.c                        | 22 ++++++++++++++++++++++
 include/asm-i386/unistd.h        |  2 +-
 include/asm-x86_64/ia32_unistd.h |  2 +-
 include/linux/syscalls.h         |  2 ++
 7 files changed, 50 insertions(+), 4 deletions(-)

diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S
index 5a8b3fb6d27b..ac687d00a1ce 100644
--- a/arch/i386/kernel/syscall_table.S
+++ b/arch/i386/kernel/syscall_table.S
@@ -299,7 +299,7 @@ ENTRY(sys_call_table)
 	.long sys_mknodat
 	.long sys_fchownat
 	.long sys_futimesat
-	.long sys_newfstatat		/* 300 */
+	.long sys_fstatat64		/* 300 */
 	.long sys_unlinkat
 	.long sys_renameat
 	.long sys_linkat
diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S
index ada4535d0161..00dee176c08e 100644
--- a/arch/x86_64/ia32/ia32entry.S
+++ b/arch/x86_64/ia32/ia32entry.S
@@ -677,7 +677,7 @@ ia32_sys_call_table:
 	.quad sys_mknodat
 	.quad sys_fchownat
 	.quad compat_sys_futimesat
-	.quad compat_sys_newfstatat	/* 300 */
+	.quad sys32_fstatat		/* 300 */
 	.quad sys_unlinkat
 	.quad sys_renameat
 	.quad sys_linkat
diff --git a/arch/x86_64/ia32/sys_ia32.c b/arch/x86_64/ia32/sys_ia32.c
index 54481af5344a..2bc55af95419 100644
--- a/arch/x86_64/ia32/sys_ia32.c
+++ b/arch/x86_64/ia32/sys_ia32.c
@@ -180,6 +180,28 @@ sys32_fstat64(unsigned int fd, struct stat64 __user *statbuf)
 	return ret;
 }
 
+asmlinkage long
+sys32_fstatat(unsigned int dfd, char __user *filename,
+	      struct stat64 __user* statbuf, int flag)
+{
+	struct kstat stat;
+	int error = -EINVAL;
+
+	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
+		goto out;
+
+	if (flag & AT_SYMLINK_NOFOLLOW)
+		error = vfs_lstat_fd(dfd, filename, &stat);
+	else
+		error = vfs_stat_fd(dfd, filename, &stat);
+
+	if (!error)
+		error = cp_stat64(statbuf, &stat);
+
+out:
+	return error;
+}
+
 /*
  * Linux/i386 didn't use to be able to handle more than
  * 4 system call parameters, so these system calls used a memory
diff --git a/fs/stat.c b/fs/stat.c
index 24211b030f39..9948cc1685a4 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -261,6 +261,7 @@ asmlinkage long sys_newlstat(char __user *filename, struct stat __user *statbuf)
 	return error;
 }
 
+#ifndef __ARCH_WANT_STAT64
 asmlinkage long sys_newfstatat(int dfd, char __user *filename,
 				struct stat __user *statbuf, int flag)
 {
@@ -281,6 +282,7 @@ asmlinkage long sys_newfstatat(int dfd, char __user *filename,
 out:
 	return error;
 }
+#endif
 
 asmlinkage long sys_newfstat(unsigned int fd, struct stat __user *statbuf)
 {
@@ -395,6 +397,26 @@ asmlinkage long sys_fstat64(unsigned long fd, struct stat64 __user * statbuf)
 	return error;
 }
 
+asmlinkage long sys_fstatat64(int dfd, char __user *filename,
+			       struct stat64 __user *statbuf, int flag)
+{
+	struct kstat stat;
+	int error = -EINVAL;
+
+	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
+		goto out;
+
+	if (flag & AT_SYMLINK_NOFOLLOW)
+		error = vfs_lstat_fd(dfd, filename, &stat);
+	else
+		error = vfs_stat_fd(dfd, filename, &stat);
+
+	if (!error)
+		error = cp_new_stat64(&stat, statbuf);
+
+out:
+	return error;
+}
 #endif /* __ARCH_WANT_STAT64 */
 
 void inode_add_bytes(struct inode *inode, loff_t bytes)
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index cf6f2cd9c514..dc81a55dd94d 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -305,7 +305,7 @@
 #define __NR_mknodat		297
 #define __NR_fchownat		298
 #define __NR_futimesat		299
-#define __NR_newfstatat		300
+#define __NR_fstatat64		300
 #define __NR_unlinkat		301
 #define __NR_renameat		302
 #define __NR_linkat		303
diff --git a/include/asm-x86_64/ia32_unistd.h b/include/asm-x86_64/ia32_unistd.h
index 20468983d453..eeb2bcd635de 100644
--- a/include/asm-x86_64/ia32_unistd.h
+++ b/include/asm-x86_64/ia32_unistd.h
@@ -305,7 +305,7 @@
 #define __NR_ia32_mknodat		297
 #define __NR_ia32_fchownat		298
 #define __NR_ia32_futimesat		299
-#define __NR_ia32_newfstatat		300
+#define __NR_ia32_fstatat64		300
 #define __NR_ia32_unlinkat		301
 #define __NR_ia32_renameat		302
 #define __NR_ia32_linkat		303
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 3877209d23c3..d73501ba7e44 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -557,6 +557,8 @@ asmlinkage long sys_openat(int dfd, const char __user *filename, int flags,
 			   int mode);
 asmlinkage long sys_newfstatat(int dfd, char __user *filename,
 			       struct stat __user *statbuf, int flag);
+asmlinkage long sys_fstatat64(int dfd, char __user *filename,
+			       struct stat64 __user *statbuf, int flag);
 asmlinkage long sys_readlinkat(int dfd, const char __user *path, char __user *buf,
 			       int bufsiz);
 asmlinkage long compat_sys_futimesat(unsigned int dfd, char __user *filename,

From e00d82d07fb112446586d225763d3572e64b7abf Mon Sep 17 00:00:00 2001
From: Matt Waddel <Matt.Waddel@freescale.com>
Date: Sat, 11 Feb 2006 17:55:48 -0800
Subject: [PATCH 075/509] [PATCH] Add wording to m68k .S files to help clarify
 license info

Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Matt Waddel <Matt.Waddel@freescale.com>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/m68k/fpsp040/bindec.S     | 5 ++---
 arch/m68k/fpsp040/binstr.S     | 5 ++---
 arch/m68k/fpsp040/bugfix.S     | 5 ++---
 arch/m68k/fpsp040/decbin.S     | 5 ++---
 arch/m68k/fpsp040/do_func.S    | 5 ++---
 arch/m68k/fpsp040/fpsp.h       | 5 ++---
 arch/m68k/fpsp040/gen_except.S | 5 ++---
 arch/m68k/fpsp040/get_op.S     | 5 ++---
 arch/m68k/fpsp040/kernel_ex.S  | 5 ++---
 arch/m68k/fpsp040/res_func.S   | 5 ++---
 arch/m68k/fpsp040/round.S      | 5 ++---
 arch/m68k/fpsp040/sacos.S      | 5 ++---
 arch/m68k/fpsp040/sasin.S      | 5 ++---
 arch/m68k/fpsp040/satan.S      | 5 ++---
 arch/m68k/fpsp040/satanh.S     | 5 ++---
 arch/m68k/fpsp040/scale.S      | 5 ++---
 arch/m68k/fpsp040/scosh.S      | 5 ++---
 arch/m68k/fpsp040/setox.S      | 5 ++---
 arch/m68k/fpsp040/sgetem.S     | 5 ++---
 arch/m68k/fpsp040/sint.S       | 5 ++---
 arch/m68k/fpsp040/skeleton.S   | 5 ++---
 arch/m68k/fpsp040/slog2.S      | 5 ++---
 arch/m68k/fpsp040/slogn.S      | 5 ++---
 arch/m68k/fpsp040/smovecr.S    | 5 ++---
 arch/m68k/fpsp040/srem_mod.S   | 5 ++---
 arch/m68k/fpsp040/ssin.S       | 5 ++---
 arch/m68k/fpsp040/ssinh.S      | 5 ++---
 arch/m68k/fpsp040/stan.S       | 5 ++---
 arch/m68k/fpsp040/stanh.S      | 5 ++---
 arch/m68k/fpsp040/sto_res.S    | 5 ++---
 arch/m68k/fpsp040/stwotox.S    | 5 ++---
 arch/m68k/fpsp040/tbldo.S      | 5 ++---
 arch/m68k/fpsp040/util.S       | 5 ++---
 arch/m68k/fpsp040/x_bsun.S     | 5 ++---
 arch/m68k/fpsp040/x_fline.S    | 5 ++---
 arch/m68k/fpsp040/x_operr.S    | 5 ++---
 arch/m68k/fpsp040/x_ovfl.S     | 5 ++---
 arch/m68k/fpsp040/x_snan.S     | 5 ++---
 arch/m68k/fpsp040/x_store.S    | 5 ++---
 arch/m68k/fpsp040/x_unfl.S     | 5 ++---
 arch/m68k/fpsp040/x_unimp.S    | 5 ++---
 arch/m68k/fpsp040/x_unsupp.S   | 5 ++---
 42 files changed, 84 insertions(+), 126 deletions(-)

diff --git a/arch/m68k/fpsp040/bindec.S b/arch/m68k/fpsp040/bindec.S
index 3ba446a99a12..72f1159cb804 100644
--- a/arch/m68k/fpsp040/bindec.S
+++ b/arch/m68k/fpsp040/bindec.S
@@ -131,9 +131,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |BINDEC    idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/binstr.S b/arch/m68k/fpsp040/binstr.S
index d53555c0a2b6..8a05ba92a8a0 100644
--- a/arch/m68k/fpsp040/binstr.S
+++ b/arch/m68k/fpsp040/binstr.S
@@ -60,9 +60,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |BINSTR    idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/bugfix.S b/arch/m68k/fpsp040/bugfix.S
index 942c4f6f4fd1..3bb9c84bb058 100644
--- a/arch/m68k/fpsp040/bugfix.S
+++ b/arch/m68k/fpsp040/bugfix.S
@@ -152,9 +152,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |BUGFIX    idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/decbin.S b/arch/m68k/fpsp040/decbin.S
index 2160609e328d..16ed796bad87 100644
--- a/arch/m68k/fpsp040/decbin.S
+++ b/arch/m68k/fpsp040/decbin.S
@@ -69,9 +69,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |DECBIN    idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/do_func.S b/arch/m68k/fpsp040/do_func.S
index 81f6a9856dce..3eff99a80413 100644
--- a/arch/m68k/fpsp040/do_func.S
+++ b/arch/m68k/fpsp040/do_func.S
@@ -22,9 +22,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 DO_FUNC:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/fpsp.h b/arch/m68k/fpsp040/fpsp.h
index 984a4eb8010a..5df4cd772934 100644
--- a/arch/m68k/fpsp040/fpsp.h
+++ b/arch/m68k/fpsp040/fpsp.h
@@ -5,9 +5,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |	fpsp.h --- stack frame offsets during FPSP exception handling
 |
diff --git a/arch/m68k/fpsp040/gen_except.S b/arch/m68k/fpsp040/gen_except.S
index 401d06f39f73..3642cb7e3641 100644
--- a/arch/m68k/fpsp040/gen_except.S
+++ b/arch/m68k/fpsp040/gen_except.S
@@ -29,9 +29,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 GEN_EXCEPT:    |idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/get_op.S b/arch/m68k/fpsp040/get_op.S
index c7c2f3727425..64c36d79ef83 100644
--- a/arch/m68k/fpsp040/get_op.S
+++ b/arch/m68k/fpsp040/get_op.S
@@ -54,9 +54,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 GET_OP:    |idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/kernel_ex.S b/arch/m68k/fpsp040/kernel_ex.S
index 476b711967ce..45bcf3455d34 100644
--- a/arch/m68k/fpsp040/kernel_ex.S
+++ b/arch/m68k/fpsp040/kernel_ex.S
@@ -12,9 +12,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 KERNEL_EX:    |idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/res_func.S b/arch/m68k/fpsp040/res_func.S
index 8f6b95217865..d9cdf4383545 100644
--- a/arch/m68k/fpsp040/res_func.S
+++ b/arch/m68k/fpsp040/res_func.S
@@ -16,9 +16,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 RES_FUNC:    |idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/round.S b/arch/m68k/fpsp040/round.S
index 00f98068783f..f84ae0dd4358 100644
--- a/arch/m68k/fpsp040/round.S
+++ b/arch/m68k/fpsp040/round.S
@@ -8,9 +8,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |ROUND	idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/sacos.S b/arch/m68k/fpsp040/sacos.S
index 83b00ab1c48f..513c7cca7318 100644
--- a/arch/m68k/fpsp040/sacos.S
+++ b/arch/m68k/fpsp040/sacos.S
@@ -38,9 +38,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SACOS	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/sasin.S b/arch/m68k/fpsp040/sasin.S
index 5647a6043903..2a269a58ceaa 100644
--- a/arch/m68k/fpsp040/sasin.S
+++ b/arch/m68k/fpsp040/sasin.S
@@ -38,9 +38,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SASIN	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/satan.S b/arch/m68k/fpsp040/satan.S
index 20dae222d51e..c8a664998f92 100644
--- a/arch/m68k/fpsp040/satan.S
+++ b/arch/m68k/fpsp040/satan.S
@@ -43,9 +43,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |satan	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/satanh.S b/arch/m68k/fpsp040/satanh.S
index 20f07810bcda..ba91f77a7571 100644
--- a/arch/m68k/fpsp040/satanh.S
+++ b/arch/m68k/fpsp040/satanh.S
@@ -45,9 +45,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |satanh	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/scale.S b/arch/m68k/fpsp040/scale.S
index 5c9b805265f2..04829dd4f1f4 100644
--- a/arch/m68k/fpsp040/scale.S
+++ b/arch/m68k/fpsp040/scale.S
@@ -21,9 +21,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SCALE    idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/scosh.S b/arch/m68k/fpsp040/scosh.S
index e81edbb87642..07d3a4d7c86d 100644
--- a/arch/m68k/fpsp040/scosh.S
+++ b/arch/m68k/fpsp040/scosh.S
@@ -49,9 +49,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SCOSH	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/setox.S b/arch/m68k/fpsp040/setox.S
index 0aa75f9bf7d1..145af5447581 100644
--- a/arch/m68k/fpsp040/setox.S
+++ b/arch/m68k/fpsp040/setox.S
@@ -331,9 +331,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |setox	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/sgetem.S b/arch/m68k/fpsp040/sgetem.S
index 0fcbd045ba75..d9234f4aed57 100644
--- a/arch/m68k/fpsp040/sgetem.S
+++ b/arch/m68k/fpsp040/sgetem.S
@@ -24,9 +24,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SGETEM	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/sint.S b/arch/m68k/fpsp040/sint.S
index 0f9bd28e55a0..0e92d4e5d231 100644
--- a/arch/m68k/fpsp040/sint.S
+++ b/arch/m68k/fpsp040/sint.S
@@ -51,9 +51,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SINT    idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/skeleton.S b/arch/m68k/fpsp040/skeleton.S
index a1629194e3fd..a8f41615d94a 100644
--- a/arch/m68k/fpsp040/skeleton.S
+++ b/arch/m68k/fpsp040/skeleton.S
@@ -30,9 +30,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |
 |	Modified for Linux-1.3.x by Jes Sorensen (jds@kom.auc.dk)
diff --git a/arch/m68k/fpsp040/slog2.S b/arch/m68k/fpsp040/slog2.S
index 517fa4563246..fac2c738382e 100644
--- a/arch/m68k/fpsp040/slog2.S
+++ b/arch/m68k/fpsp040/slog2.S
@@ -96,9 +96,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SLOG2    idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/slogn.S b/arch/m68k/fpsp040/slogn.S
index 2aaa0725c035..d98eaf641ec4 100644
--- a/arch/m68k/fpsp040/slogn.S
+++ b/arch/m68k/fpsp040/slogn.S
@@ -63,9 +63,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |slogn	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/smovecr.S b/arch/m68k/fpsp040/smovecr.S
index a0127fa55e9c..73c36512081b 100644
--- a/arch/m68k/fpsp040/smovecr.S
+++ b/arch/m68k/fpsp040/smovecr.S
@@ -15,9 +15,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SMOVECR	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/srem_mod.S b/arch/m68k/fpsp040/srem_mod.S
index 8c8d7f50cc68..a27e70c9a0eb 100644
--- a/arch/m68k/fpsp040/srem_mod.S
+++ b/arch/m68k/fpsp040/srem_mod.S
@@ -66,9 +66,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 SREM_MOD:    |idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/ssin.S b/arch/m68k/fpsp040/ssin.S
index 043c91cdd657..a1ef8e01bf06 100644
--- a/arch/m68k/fpsp040/ssin.S
+++ b/arch/m68k/fpsp040/ssin.S
@@ -83,9 +83,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SSIN	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/ssinh.S b/arch/m68k/fpsp040/ssinh.S
index c8b3308bb143..8a560edc7653 100644
--- a/arch/m68k/fpsp040/ssinh.S
+++ b/arch/m68k/fpsp040/ssinh.S
@@ -49,9 +49,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |SSINH	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/stan.S b/arch/m68k/fpsp040/stan.S
index b5c2a196e617..f8553aaececb 100644
--- a/arch/m68k/fpsp040/stan.S
+++ b/arch/m68k/fpsp040/stan.S
@@ -50,9 +50,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |STAN	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/stanh.S b/arch/m68k/fpsp040/stanh.S
index 33b009802243..7e12e59ee8c7 100644
--- a/arch/m68k/fpsp040/stanh.S
+++ b/arch/m68k/fpsp040/stanh.S
@@ -49,9 +49,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |STANH	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/sto_res.S b/arch/m68k/fpsp040/sto_res.S
index 0cdca3b060ad..484b47d4eaad 100644
--- a/arch/m68k/fpsp040/sto_res.S
+++ b/arch/m68k/fpsp040/sto_res.S
@@ -19,9 +19,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 STO_RES:	|idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/stwotox.S b/arch/m68k/fpsp040/stwotox.S
index 4e3c1407d3df..0d5e6a1436a6 100644
--- a/arch/m68k/fpsp040/stwotox.S
+++ b/arch/m68k/fpsp040/stwotox.S
@@ -76,9 +76,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |STWOTOX	idnt	2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/tbldo.S b/arch/m68k/fpsp040/tbldo.S
index fe60cf4d20d7..fd5c37a5a2b9 100644
--- a/arch/m68k/fpsp040/tbldo.S
+++ b/arch/m68k/fpsp040/tbldo.S
@@ -17,9 +17,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |TBLDO	idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/util.S b/arch/m68k/fpsp040/util.S
index 452f3d65857b..65b26fa88c60 100644
--- a/arch/m68k/fpsp040/util.S
+++ b/arch/m68k/fpsp040/util.S
@@ -16,9 +16,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 |UTIL	idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_bsun.S b/arch/m68k/fpsp040/x_bsun.S
index 039247b09c8b..d5a576bfac79 100644
--- a/arch/m68k/fpsp040/x_bsun.S
+++ b/arch/m68k/fpsp040/x_bsun.S
@@ -13,9 +13,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_BSUN:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_fline.S b/arch/m68k/fpsp040/x_fline.S
index 3917710b0fde..264e126d1db7 100644
--- a/arch/m68k/fpsp040/x_fline.S
+++ b/arch/m68k/fpsp040/x_fline.S
@@ -13,9 +13,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_FLINE:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_operr.S b/arch/m68k/fpsp040/x_operr.S
index b0f54bcb49a7..e2c371c3a45d 100644
--- a/arch/m68k/fpsp040/x_operr.S
+++ b/arch/m68k/fpsp040/x_operr.S
@@ -43,9 +43,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_OPERR:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_ovfl.S b/arch/m68k/fpsp040/x_ovfl.S
index 22cb8b42c7b6..6fe4989ee31f 100644
--- a/arch/m68k/fpsp040/x_ovfl.S
+++ b/arch/m68k/fpsp040/x_ovfl.S
@@ -35,9 +35,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_OVFL:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_snan.S b/arch/m68k/fpsp040/x_snan.S
index 039af573312e..4ed766416378 100644
--- a/arch/m68k/fpsp040/x_snan.S
+++ b/arch/m68k/fpsp040/x_snan.S
@@ -22,9 +22,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_SNAN:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_store.S b/arch/m68k/fpsp040/x_store.S
index 4282fa67d449..402dc0c0ebc0 100644
--- a/arch/m68k/fpsp040/x_store.S
+++ b/arch/m68k/fpsp040/x_store.S
@@ -11,9 +11,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_STORE:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_unfl.S b/arch/m68k/fpsp040/x_unfl.S
index 077fcc230fcc..eb772ff3b812 100644
--- a/arch/m68k/fpsp040/x_unfl.S
+++ b/arch/m68k/fpsp040/x_unfl.S
@@ -21,9 +21,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_UNFL:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_unimp.S b/arch/m68k/fpsp040/x_unimp.S
index 920cb9410e9e..6f382b21228b 100644
--- a/arch/m68k/fpsp040/x_unimp.S
+++ b/arch/m68k/fpsp040/x_unimp.S
@@ -22,9 +22,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_UNIMP:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 
diff --git a/arch/m68k/fpsp040/x_unsupp.S b/arch/m68k/fpsp040/x_unsupp.S
index 4ec57285b683..d7cf46208c62 100644
--- a/arch/m68k/fpsp040/x_unsupp.S
+++ b/arch/m68k/fpsp040/x_unsupp.S
@@ -23,9 +23,8 @@
 |		Copyright (C) Motorola, Inc. 1990
 |			All Rights Reserved
 |
-|	THIS IS UNPUBLISHED PROPRIETARY SOURCE CODE OF MOTOROLA
-|	The copyright notice above does not evidence any
-|	actual or intended publication of such source code.
+|       For details on the license for this file, please see the
+|       file, README, in this same directory.
 
 X_UNSUPP:	|idnt    2,1 | Motorola 040 Floating Point Software Package
 

From 33042a9ff4d126ba944b9dc3076665a2029e0a34 Mon Sep 17 00:00:00 2001
From: Chris McDermott <lcm@us.ibm.com>
Date: Sat, 11 Feb 2006 17:55:50 -0800
Subject: [PATCH 076/509] [PATCH] x86-64: Fix HPET timer on x460

[description from AK]

The IBM Summit 3 chipset doesn't implement the HPET timer replacement
option.  Since the current Linux code relies on it use a mixed mode with
both PIT for the interrupt and HPET counters for the time keeping.  That
was already implemented, but didn't work properly because it was still
using the last interrupt offset in HPET.  This resulted in x460 not
booting.  Fix this up by using the free running HPET counter.

Shouldn't affect any other machine because they either use full HPET mode
or no HPET at all.

TBD needs a similar 32bit fix.

Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Pallipadi, Venkatesh" <venkatesh.pallipadi@intel.com>
Cc: Bob Picco <bob.picco@hp.com>
Cc: Bjorn Helgaas <bjorn.helgaas@hp.com>
Cc: john stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/apic.c |  2 +-
 arch/x86_64/kernel/time.c | 12 +++++++++---
 include/asm-x86_64/hpet.h |  2 ++
 3 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 6147770b4347..7a0a3e8d5d72 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -708,7 +708,7 @@ static void setup_APIC_timer(unsigned int clocks)
 	local_irq_save(flags);
 
 	/* wait for irq slice */
- 	if (vxtime.hpet_address) {
+ 	if (vxtime.hpet_address && hpet_use_timer) {
  		int trigger = hpet_readl(HPET_T0_CMP);
  		while (hpet_readl(HPET_COUNTER) >= trigger)
  			/* do nothing */ ;
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index dba7237be5c1..3c58c30506a1 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -59,7 +59,7 @@ static int notsc __initdata = 0;
 unsigned int cpu_khz;					/* TSC clocks / usec, not used here */
 static unsigned long hpet_period;			/* fsecs / HPET clock */
 unsigned long hpet_tick;				/* HPET clocks / interrupt */
-static int hpet_use_timer;				/* Use counter of hpet for time keeping, otherwise PIT */
+int hpet_use_timer;				/* Use counter of hpet for time keeping, otherwise PIT */
 unsigned long vxtime_hz = PIT_TICK_RATE;
 int report_lost_ticks;				/* command line option */
 unsigned long long monotonic_base;
@@ -326,7 +326,10 @@ static noinline void handle_lost_ticks(int lost, struct pt_regs *regs)
 	    print_symbol("rip %s\n", regs->rip);
 	    if (vxtime.mode == VXTIME_TSC && vxtime.hpet_address) {
 		    printk(KERN_WARNING "Falling back to HPET\n");
-		    vxtime.last = hpet_readl(HPET_T0_CMP) - hpet_tick;
+		    if (hpet_use_timer)
+		    	vxtime.last = hpet_readl(HPET_T0_CMP) - hpet_tick;
+		    else
+		    	vxtime.last = hpet_readl(HPET_COUNTER);
 		    vxtime.mode = VXTIME_HPET;
 		    do_gettimeoffset = do_gettimeoffset_hpet;
 	    }
@@ -988,7 +991,10 @@ void __init time_init_gtod(void)
 		notsc = 1;
 	if (vxtime.hpet_address && notsc) {
 		timetype = hpet_use_timer ? "HPET" : "PIT/HPET";
-		vxtime.last = hpet_readl(HPET_T0_CMP) - hpet_tick;
+		if (hpet_use_timer)
+			vxtime.last = hpet_readl(HPET_T0_CMP) - hpet_tick;
+		else
+			vxtime.last = hpet_readl(HPET_COUNTER);
 		vxtime.mode = VXTIME_HPET;
 		do_gettimeoffset = do_gettimeoffset_hpet;
 #ifdef CONFIG_X86_PM_TIMER
diff --git a/include/asm-x86_64/hpet.h b/include/asm-x86_64/hpet.h
index c20c28f5c7a0..08b75c15269a 100644
--- a/include/asm-x86_64/hpet.h
+++ b/include/asm-x86_64/hpet.h
@@ -55,6 +55,8 @@ extern int is_hpet_enabled(void);
 extern int hpet_rtc_timer_init(void);
 extern int oem_force_hpet_timer(void);
 
+extern int hpet_use_timer;
+
 #ifdef CONFIG_HPET_EMULATE_RTC
 extern int hpet_mask_rtc_irq_bit(unsigned long bit_mask);
 extern int hpet_set_rtc_irq_bit(unsigned long bit_mask);

From 643a654540579b0dcc7a206a4a7475276a41aff0 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Sat, 11 Feb 2006 17:55:52 -0800
Subject: [PATCH 077/509] [PATCH] select: fix returned timeval

With David Woodhouse <dwmw2@infradead.org>

select() presently has a habit of increasing the value of the user's
`timeout' argument on return.

We were writing back a timeout larger than the original.  We _deliberately_
round up, since we know we must wait at _least_ as long as the caller asks
us to.

The patch adds a couple of helper functions for magnitude comparison of
timespecs and of timevals, and uses them to prevent the various poll and
select functions from returning a timeout which is larger than the one which
was passed in.

The patch also fixes a bug in compat_sys_pselect7(): it was adding the new
timeout value to the old one and was returning that.  It should just return
the new timeout value.

(We have various handy timespec/timeval-to-from-nsec conversion functions in
time.h.  But this code open-codes it all).

Cc: "David S. Miller" <davem@davemloft.net>
Cc: Andi Kleen <ak@muc.de>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: george anzinger <george@mvista.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/compat.c            | 37 +++++++++++++++++++++++++------------
 fs/select.c            | 32 +++++++++++++++++++++++---------
 include/linux/compat.h | 20 ++++++++++++++++++++
 include/linux/time.h   | 25 ++++++++++++++++++++++++-
 4 files changed, 92 insertions(+), 22 deletions(-)

diff --git a/fs/compat.c b/fs/compat.c
index 70c5af4cc270..a2ba78bdf7f7 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -1751,11 +1751,15 @@ asmlinkage long compat_sys_select(int n, compat_ulong_t __user *inp,
 	ret = compat_core_sys_select(n, inp, outp, exp, &timeout);
 
 	if (tvp) {
+		struct compat_timeval rtv;
+
 		if (current->personality & STICKY_TIMEOUTS)
 			goto sticky;
-		tv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ));
-		tv.tv_sec = timeout;
-		if (copy_to_user(tvp, &tv, sizeof(tv))) {
+		rtv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ));
+		rtv.tv_sec = timeout;
+		if (compat_timeval_compare(&rtv, &tv) < 0)
+			rtv = tv;
+		if (copy_to_user(tvp, &rtv, sizeof(rtv))) {
 sticky:
 			/*
 			 * If an application puts its timeval in read-only
@@ -1822,13 +1826,17 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp,
 	} while (!ret && !timeout && tsp && (ts.tv_sec || ts.tv_nsec));
 
 	if (tsp && !(current->personality & STICKY_TIMEOUTS)) {
-		ts.tv_sec += timeout / HZ;
-		ts.tv_nsec += (timeout % HZ) * (1000000000/HZ);
-		if (ts.tv_nsec >= 1000000000) {
-			ts.tv_sec++;
-			ts.tv_nsec -= 1000000000;
+		struct compat_timespec rts;
+
+		rts.tv_sec = timeout / HZ;
+		rts.tv_nsec = (timeout % HZ) * (NSEC_PER_SEC/HZ);
+		if (rts.tv_nsec >= NSEC_PER_SEC) {
+			rts.tv_sec++;
+			rts.tv_nsec -= NSEC_PER_SEC;
 		}
-		(void)copy_to_user(tsp, &ts, sizeof(ts));
+		if (compat_timespec_compare(&rts, &ts) < 0)
+			rts = ts;
+		copy_to_user(tsp, &rts, sizeof(rts));
 	}
 
 	if (ret == -ERESTARTNOHAND) {
@@ -1918,12 +1926,17 @@ asmlinkage long compat_sys_ppoll(struct pollfd __user *ufds,
 		sigprocmask(SIG_SETMASK, &sigsaved, NULL);
 
 	if (tsp && timeout >= 0) {
+		struct compat_timespec rts;
+
 		if (current->personality & STICKY_TIMEOUTS)
 			goto sticky;
 		/* Yes, we know it's actually an s64, but it's also positive. */
-		ts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) * 1000;
-		ts.tv_sec = timeout;
-		if (copy_to_user(tsp, &ts, sizeof(ts))) {
+		rts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) *
+					1000;
+		rts.tv_sec = timeout;
+		if (compat_timespec_compare(&rts, &ts) < 0)
+			rts = ts;
+		if (copy_to_user(tsp, &rts, sizeof(rts))) {
 sticky:
 			/*
 			 * If an application puts its timeval in read-only
diff --git a/fs/select.c b/fs/select.c
index bc60a3e14ef3..6ce68a9c8976 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -398,11 +398,15 @@ asmlinkage long sys_select(int n, fd_set __user *inp, fd_set __user *outp,
 	ret = core_sys_select(n, inp, outp, exp, &timeout);
 
 	if (tvp) {
+		struct timeval rtv;
+
 		if (current->personality & STICKY_TIMEOUTS)
 			goto sticky;
-		tv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ));
-		tv.tv_sec = timeout;
-		if (copy_to_user(tvp, &tv, sizeof(tv))) {
+		rtv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ));
+		rtv.tv_sec = timeout;
+		if (timeval_compare(&rtv, &tv) < 0)
+			rtv = tv;
+		if (copy_to_user(tvp, &rtv, sizeof(rtv))) {
 sticky:
 			/*
 			 * If an application puts its timeval in read-only
@@ -460,11 +464,16 @@ asmlinkage long sys_pselect7(int n, fd_set __user *inp, fd_set __user *outp,
 	ret = core_sys_select(n, inp, outp, exp, &timeout);
 
 	if (tsp) {
+		struct timespec rts;
+
 		if (current->personality & STICKY_TIMEOUTS)
 			goto sticky;
-		ts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) * 1000;
-		ts.tv_sec = timeout;
-		if (copy_to_user(tsp, &ts, sizeof(ts))) {
+		rts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) *
+						1000;
+		rts.tv_sec = timeout;
+		if (timespec_compare(&rts, &ts) < 0)
+			rts = ts;
+		if (copy_to_user(tsp, &rts, sizeof(rts))) {
 sticky:
 			/*
 			 * If an application puts its timeval in read-only
@@ -758,12 +767,17 @@ asmlinkage long sys_ppoll(struct pollfd __user *ufds, unsigned int nfds,
 		sigprocmask(SIG_SETMASK, &sigsaved, NULL);
 
 	if (tsp && timeout >= 0) {
+		struct timespec rts;
+
 		if (current->personality & STICKY_TIMEOUTS)
 			goto sticky;
 		/* Yes, we know it's actually an s64, but it's also positive. */
-		ts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) * 1000;
-		ts.tv_sec = timeout;
-		if (copy_to_user(tsp, &ts, sizeof(ts))) {
+		rts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) *
+						1000;
+		rts.tv_sec = timeout;
+		if (timespec_compare(&rts, &ts) < 0)
+			rts = ts;
+		if (copy_to_user(tsp, &rts, sizeof(rts))) {
 		sticky:
 			/*
 			 * If an application puts its timeval in read-only
diff --git a/include/linux/compat.h b/include/linux/compat.h
index f9ca534787e2..c9ab2a26348c 100644
--- a/include/linux/compat.h
+++ b/include/linux/compat.h
@@ -161,5 +161,25 @@ int copy_siginfo_to_user32(struct compat_siginfo __user *to, siginfo_t *from);
 int get_compat_sigevent(struct sigevent *event,
 		const struct compat_sigevent __user *u_event);
 
+static inline int compat_timeval_compare(struct compat_timeval *lhs,
+					struct compat_timeval *rhs)
+{
+	if (lhs->tv_sec < rhs->tv_sec)
+		return -1;
+	if (lhs->tv_sec > rhs->tv_sec)
+		return 1;
+	return lhs->tv_usec - rhs->tv_usec;
+}
+
+static inline int compat_timespec_compare(struct compat_timespec *lhs,
+					struct compat_timespec *rhs)
+{
+	if (lhs->tv_sec < rhs->tv_sec)
+		return -1;
+	if (lhs->tv_sec > rhs->tv_sec)
+		return 1;
+	return lhs->tv_nsec - rhs->tv_nsec;
+}
+
 #endif /* CONFIG_COMPAT */
 #endif /* _LINUX_COMPAT_H */
diff --git a/include/linux/time.h b/include/linux/time.h
index 7b4dc36532bb..d9cdba54b789 100644
--- a/include/linux/time.h
+++ b/include/linux/time.h
@@ -33,11 +33,34 @@ struct timezone {
 #define NSEC_PER_SEC		1000000000L
 #define NSEC_PER_USEC		1000L
 
-static __inline__ int timespec_equal(struct timespec *a, struct timespec *b)
+static inline int timespec_equal(struct timespec *a, struct timespec *b)
 {
 	return (a->tv_sec == b->tv_sec) && (a->tv_nsec == b->tv_nsec);
 }
 
+/*
+ * lhs < rhs:  return <0
+ * lhs == rhs: return 0
+ * lhs > rhs:  return >0
+ */
+static inline int timespec_compare(struct timespec *lhs, struct timespec *rhs)
+{
+	if (lhs->tv_sec < rhs->tv_sec)
+		return -1;
+	if (lhs->tv_sec > rhs->tv_sec)
+		return 1;
+	return lhs->tv_nsec - rhs->tv_nsec;
+}
+
+static inline int timeval_compare(struct timeval *lhs, struct timeval *rhs)
+{
+	if (lhs->tv_sec < rhs->tv_sec)
+		return -1;
+	if (lhs->tv_sec > rhs->tv_sec)
+		return 1;
+	return lhs->tv_usec - rhs->tv_usec;
+}
+
 extern unsigned long mktime(const unsigned int year, const unsigned int mon,
 			    const unsigned int day, const unsigned int hour,
 			    const unsigned int min, const unsigned int sec);

From 80e4342601abfafacb5f20571e40b56d73d10819 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Sat, 11 Feb 2006 17:55:53 -0800
Subject: [PATCH 078/509] [PATCH] zone reclaim: do not check references to a
 page during zone reclaim

shrink_list() and refill_inactive() check all ptes pointing to a page for
reference bits in order to decide if the page should be put on the active
list.  This is not necessary for zone_reclaim since we are only interested
in removing unmapped pages.  Skip the checks in both functions.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5db32fdfaf39..e1c64230ffdd 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -443,6 +443,10 @@ static int shrink_list(struct list_head *page_list, struct scan_control *sc)
 		BUG_ON(PageActive(page));
 
 		sc->nr_scanned++;
+
+		if (!sc->may_swap && page_mapped(page))
+			goto keep_locked;
+
 		/* Double the slab pressure for mapped and swapcache pages */
 		if (page_mapped(page) || PageSwapCache(page))
 			sc->nr_scanned++;
@@ -1231,7 +1235,7 @@ refill_inactive_zone(struct zone *zone, struct scan_control *sc)
 	 * Now use this metric to decide whether to start moving mapped memory
 	 * onto the inactive list.
 	 */
-	if (swap_tendency >= 100)
+	if (swap_tendency >= 100 && sc->may_swap)
 		reclaim_mapped = 1;
 
 	while (!list_empty(&l_hold)) {

From 072eaa5d9cc3e63f567ffd9ad87b36194fdd8010 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Sat, 11 Feb 2006 17:55:54 -0800
Subject: [PATCH 079/509] [PATCH] vmscan: remove duplicate increment of
 reclaim_in_progress

shrink_zone() already increments reclaim_in_progress.  No need to do it in
balance_pgdat.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index e1c64230ffdd..58ed5125b1a7 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1614,9 +1614,7 @@ scan:
 			sc.nr_reclaimed = 0;
 			sc.priority = priority;
 			sc.swap_cluster_max = nr_pages? nr_pages : SWAP_CLUSTER_MAX;
-			atomic_inc(&zone->reclaim_in_progress);
 			shrink_zone(zone, &sc);
-			atomic_dec(&zone->reclaim_in_progress);
 			reclaim_state->reclaimed_slab = 0;
 			nr_slab = shrink_slab(sc.nr_scanned, GFP_KERNEL,
 						lru_pages);

From 2903fb1694dcb08a3c1d9d823cfae7ba30e66cd3 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Sat, 11 Feb 2006 17:55:55 -0800
Subject: [PATCH 080/509] [PATCH] vmscan: skip reclaim_mapped determination if
 we do not swap

This puts the variables and the way to get to reclaim_mapped in one block.
And allows zone_reclaim or other things to skip the determination (maybe
this whole block of code does not belong into refill_inactive_zone()?)

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 75 +++++++++++++++++++++++++++++------------------------
 1 file changed, 41 insertions(+), 34 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 58ed5125b1a7..1838c15ca4fd 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1195,9 +1195,47 @@ refill_inactive_zone(struct zone *zone, struct scan_control *sc)
 	struct page *page;
 	struct pagevec pvec;
 	int reclaim_mapped = 0;
-	long mapped_ratio;
-	long distress;
-	long swap_tendency;
+
+	if (unlikely(sc->may_swap)) {
+		long mapped_ratio;
+		long distress;
+		long swap_tendency;
+
+		/*
+		 * `distress' is a measure of how much trouble we're having
+		 * reclaiming pages.  0 -> no problems.  100 -> great trouble.
+		 */
+		distress = 100 >> zone->prev_priority;
+
+		/*
+		 * The point of this algorithm is to decide when to start
+		 * reclaiming mapped memory instead of just pagecache.  Work out
+		 * how much memory
+		 * is mapped.
+		 */
+		mapped_ratio = (sc->nr_mapped * 100) / total_memory;
+
+		/*
+		 * Now decide how much we really want to unmap some pages.  The
+		 * mapped ratio is downgraded - just because there's a lot of
+		 * mapped memory doesn't necessarily mean that page reclaim
+		 * isn't succeeding.
+		 *
+		 * The distress ratio is important - we don't want to start
+		 * going oom.
+		 *
+		 * A 100% value of vm_swappiness overrides this algorithm
+		 * altogether.
+		 */
+		swap_tendency = mapped_ratio / 2 + distress + vm_swappiness;
+
+		/*
+		 * Now use this metric to decide whether to start moving mapped
+		 * memory onto the inactive list.
+		 */
+		if (swap_tendency >= 100)
+			reclaim_mapped = 1;
+	}
 
 	lru_add_drain();
 	spin_lock_irq(&zone->lru_lock);
@@ -1207,37 +1245,6 @@ refill_inactive_zone(struct zone *zone, struct scan_control *sc)
 	zone->nr_active -= pgmoved;
 	spin_unlock_irq(&zone->lru_lock);
 
-	/*
-	 * `distress' is a measure of how much trouble we're having reclaiming
-	 * pages.  0 -> no problems.  100 -> great trouble.
-	 */
-	distress = 100 >> zone->prev_priority;
-
-	/*
-	 * The point of this algorithm is to decide when to start reclaiming
-	 * mapped memory instead of just pagecache.  Work out how much memory
-	 * is mapped.
-	 */
-	mapped_ratio = (sc->nr_mapped * 100) / total_memory;
-
-	/*
-	 * Now decide how much we really want to unmap some pages.  The mapped
-	 * ratio is downgraded - just because there's a lot of mapped memory
-	 * doesn't necessarily mean that page reclaim isn't succeeding.
-	 *
-	 * The distress ratio is important - we don't want to start going oom.
-	 *
-	 * A 100% value of vm_swappiness overrides this algorithm altogether.
-	 */
-	swap_tendency = mapped_ratio / 2 + distress + vm_swappiness;
-
-	/*
-	 * Now use this metric to decide whether to start moving mapped memory
-	 * onto the inactive list.
-	 */
-	if (swap_tendency >= 100 && sc->may_swap)
-		reclaim_mapped = 1;
-
 	while (!list_empty(&l_hold)) {
 		cond_resched();
 		page = lru_to_page(&l_hold);

From c0cdf1935cf328730fa068e0f39a22e6149555aa Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 11 Feb 2006 17:55:56 -0800
Subject: [PATCH 081/509] [PATCH] x86: print out early faults via
 early_printk()

Lost a few hours debugging an early-bootup fault within printk itself,
which manifested itself as a hard to debug early hang.

This patch makes it much easier by printing out early faults via
early_printk(), which function is a lot simpler than a full printk, and
hence more likely to succeed in emergencies.  (We do not recover from early
faults anyway, so there's no loss from not having these messages in the
normal printk buffer.)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/head.S | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index 5884469f6bfe..2bee6499edd9 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -398,7 +398,11 @@ ignore_int:
 	pushl 32(%esp)
 	pushl 40(%esp)
 	pushl $int_msg
+#ifdef CONFIG_EARLY_PRINTK
+	call early_printk
+#else
 	call printk
+#endif
 	addl $(5*4),%esp
 	popl %ds
 	popl %es

From c48d865c50e8626372a52094385fb1f5a2d2a7fd Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Sat, 11 Feb 2006 17:55:57 -0800
Subject: [PATCH 082/509] [PATCH] s390: fix locking in __chp_add() and
 s390_subchannel_remove_chpid()

Fix locking in __chp_add() and s390_subchannel_remove_chpid(): Need to
disable/enable because they are always called from a thread (and not
directly from a machine check...)

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/s390/cio/chsc.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/s390/cio/chsc.c b/drivers/s390/cio/chsc.c
index 92be75d99a56..8cf9905d484b 100644
--- a/drivers/s390/cio/chsc.c
+++ b/drivers/s390/cio/chsc.c
@@ -232,7 +232,7 @@ s390_subchannel_remove_chpid(struct device *dev, void *data)
 		return 0;
 
 	mask = 0x80 >> j;
-	spin_lock(&sch->lock);
+	spin_lock_irq(&sch->lock);
 
 	stsch(sch->schid, &schib);
 	if (!schib.pmcw.dnv)
@@ -281,10 +281,10 @@ s390_subchannel_remove_chpid(struct device *dev, void *data)
 	if (sch->driver && sch->driver->verify)
 		sch->driver->verify(&sch->dev);
 out_unlock:
-	spin_unlock(&sch->lock);
+	spin_unlock_irq(&sch->lock);
 	return 0;
 out_unreg:
-	spin_unlock(&sch->lock);
+	spin_unlock_irq(&sch->lock);
 	sch->lpm = 0;
 	if (css_enqueue_subchannel_slow(sch->schid)) {
 		css_clear_subchannel_slow_list();
@@ -652,7 +652,7 @@ __chp_add(struct subchannel_id schid, void *data)
 	if (!sch)
 		/* Check if the subchannel is now available. */
 		return __chp_add_new_sch(schid);
-	spin_lock(&sch->lock);
+	spin_lock_irq(&sch->lock);
 	for (i=0; i<8; i++)
 		if (sch->schib.pmcw.chpid[i] == chp->id) {
 			if (stsch(sch->schid, &sch->schib) != 0) {
@@ -674,7 +674,7 @@ __chp_add(struct subchannel_id schid, void *data)
 	if (sch->driver && sch->driver->verify)
 		sch->driver->verify(&sch->dev);
 
-	spin_unlock(&sch->lock);
+	spin_unlock_irq(&sch->lock);
 	put_device(&sch->dev);
 	return 0;
 }

From e6f3601a7275216c48c2635f46b388d970901bb9 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 11 Feb 2006 17:55:58 -0800
Subject: [PATCH 083/509] [PATCH] s390: update default configuration

Switch on CONFIG_DEBUG_FS again.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/defconfig | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/arch/s390/defconfig b/arch/s390/defconfig
index 3525c91204d4..f8d0cd540a06 100644
--- a/arch/s390/defconfig
+++ b/arch/s390/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.16-rc1
-# Thu Jan 19 10:58:53 2006
+# Linux kernel version: 2.6.16-rc2
+# Wed Feb  8 10:44:39 2006
 #
 CONFIG_MMU=y
 CONFIG_RWSEM_XCHGADD_ALGORITHM=y
@@ -12,7 +12,6 @@ CONFIG_S390=y
 # Code maturity level options
 #
 CONFIG_EXPERIMENTAL=y
-CONFIG_CLEAN_COMPILE=y
 CONFIG_LOCK_KERNEL=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
 
@@ -154,6 +153,7 @@ CONFIG_NET=y
 #
 # Networking options
 #
+# CONFIG_NETDEBUG is not set
 CONFIG_PACKET=y
 # CONFIG_PACKET_MMAP is not set
 CONFIG_UNIX=y
@@ -607,6 +607,7 @@ CONFIG_MSDOS_PARTITION=y
 # Instrumentation Support
 #
 # CONFIG_PROFILING is not set
+# CONFIG_STATISTICS is not set
 
 #
 # Kernel hacking
@@ -624,7 +625,7 @@ CONFIG_DEBUG_MUTEXES=y
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
-# CONFIG_DEBUG_FS is not set
+CONFIG_DEBUG_FS=y
 # CONFIG_DEBUG_VM is not set
 CONFIG_FORCED_INLINING=y
 # CONFIG_RCU_TORTURE_TEST is not set

From 25fab9ebac445d57b656f5faabac5a195bed2f82 Mon Sep 17 00:00:00 2001
From: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Date: Sat, 11 Feb 2006 17:55:59 -0800
Subject: [PATCH 084/509] [PATCH] s390: fix sclp memory corruption in tty pages
 list

When the sclp interface takes very long to serve a request, the sclp core
driver will report a failed request to the sclp tty driver even though the
request is still being processed by the sclp interface.  Eventually the sclp
interface completes the request and updates some fields in the request buffer
which leads to a corrupted tty pages list.  The next time function
sclp_tty_write_room is called, the corrupted list will be traversed, resulting
in an oops.

To avoid this remove the busy retry limit and increase retry intervals.

Signed-off-by: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/s390/char/sclp.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/s390/char/sclp.c b/drivers/s390/char/sclp.c
index ceb0e474fde4..4138564402b8 100644
--- a/drivers/s390/char/sclp.c
+++ b/drivers/s390/char/sclp.c
@@ -85,11 +85,10 @@ static volatile enum sclp_mask_state_t {
 /* Maximum retry counts */
 #define SCLP_INIT_RETRY		3
 #define SCLP_MASK_RETRY		3
-#define SCLP_REQUEST_RETRY	3
 
 /* Timeout intervals in seconds.*/
-#define SCLP_BUSY_INTERVAL	2
-#define SCLP_RETRY_INTERVAL	5
+#define SCLP_BUSY_INTERVAL	10
+#define SCLP_RETRY_INTERVAL	15
 
 static void sclp_process_queue(void);
 static int sclp_init_mask(int calculate);
@@ -153,11 +152,9 @@ __sclp_start_request(struct sclp_req *req)
 	if (sclp_running_state != sclp_running_state_idle)
 		return 0;
 	del_timer(&sclp_request_timer);
-	if (req->start_count <= SCLP_REQUEST_RETRY) {
-		rc = service_call(req->command, req->sccb);
-		req->start_count++;
-	} else
-		rc = -EIO;
+	rc = service_call(req->command, req->sccb);
+	req->start_count++;
+
 	if (rc == 0) {
 		/* Sucessfully started request */
 		req->status = SCLP_REQ_RUNNING;

From 9733e2407ad2237867cb13c04e7d619397fa3090 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 11 Feb 2006 17:56:00 -0800
Subject: [PATCH 085/509] [PATCH] s390: earlier initialization of
 cpu_possible_map

Initiliazing of cpu_possible_map was done in smp_prepare_cpus which is way too
late.  Therefore assign a static value to cpu_possible_map, since we don't
have access to max_cpus in setup_arch.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/smp.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index cbfcfd02a43a..0d1ad5dbe2b1 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -52,7 +52,7 @@ extern volatile int __cpu_logical_map[];
 struct _lowcore *lowcore_ptr[NR_CPUS];
 
 cpumask_t cpu_online_map;
-cpumask_t cpu_possible_map;
+cpumask_t cpu_possible_map = CPU_MASK_ALL;
 
 static struct task_struct *current_set[NR_CPUS];
 
@@ -514,9 +514,6 @@ __init smp_check_cpus(unsigned int max_cpus)
 		num_cpus++;
 	}
 
-	for (cpu = 1; cpu < max_cpus; cpu++)
-		cpu_set(cpu, cpu_possible_map);
-
 	printk("Detected %d CPU's\n",(int) num_cpus);
 	printk("Boot cpu address %2X\n", boot_cpu_addr);
 }
@@ -810,7 +807,6 @@ void __devinit smp_prepare_boot_cpu(void)
 
 	cpu_set(0, cpu_online_map);
 	cpu_set(0, cpu_present_map);
-	cpu_set(0, cpu_possible_map);
 	S390_lowcore.percpu_offset = __per_cpu_offset[0];
 	current_set[0] = current;
 }

From 5238da45f345898a8bfcd14e53b0431fcee36a04 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 11 Feb 2006 17:56:01 -0800
Subject: [PATCH 086/509] [PATCH] s390: update maintainers file

Update URL for s390 and add maintainers for s390 networking and zfcp driver.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 MAINTAINERS | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index b22db521cec1..9c592aa0280c 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2232,7 +2232,23 @@ P:	Martin Schwidefsky
 M:	schwidefsky@de.ibm.com
 M:	linux390@de.ibm.com
 L:	linux-390@vm.marist.edu
-W:	http://oss.software.ibm.com/developerworks/opensource/linux390
+W:	http://www.ibm.com/developerworks/linux/linux390/
+S:	Supported
+
+S390 NETWORK DRIVERS
+P:	Frank Pavlic
+M:	fpavlic@de.ibm.com
+M:	linux390@de.ibm.com
+L:	linux-390@vm.marist.edu
+W:	http://www.ibm.com/developerworks/linux/linux390/
+S:	Supported
+
+S390 ZFCP DRIVER
+P:	Andreas Herrmann
+M:	aherrman@de.ibm.com
+M:	linux390@de.ibm.com
+L:	linux-390@vm.marist.edu
+W:	http://www.ibm.com/developerworks/linux/linux390/
 S:	Supported
 
 SAA7146 VIDEO4LINUX-2 DRIVER

From a386fba2516b5404864647906219ced57bf2f2b7 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 11 Feb 2006 17:56:01 -0800
Subject: [PATCH 087/509] [PATCH] s390: fix non smp build of kexec

Add missing smp_cpu_not_running define to avoid build warnings in the non smp
case.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/machine_kexec.c | 5 +++--
 include/asm-s390/smp.h           | 1 +
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/s390/kernel/machine_kexec.c b/arch/s390/kernel/machine_kexec.c
index f0ed5c642c74..bad81b5832db 100644
--- a/arch/s390/kernel/machine_kexec.c
+++ b/arch/s390/kernel/machine_kexec.c
@@ -12,15 +12,16 @@
  * on the S390 architecture.
  */
 
-#include <asm/cio.h>
-#include <asm/setup.h>
 #include <linux/device.h>
 #include <linux/mm.h>
 #include <linux/kexec.h>
 #include <linux/delay.h>
+#include <asm/cio.h>
+#include <asm/setup.h>
 #include <asm/pgtable.h>
 #include <asm/pgalloc.h>
 #include <asm/system.h>
+#include <asm/smp.h>
 
 static void kexec_halt_all_cpus(void *);
 
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index a2ae7628bbaa..9c6e9c300eb9 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -101,6 +101,7 @@ smp_call_function_on(void (*func) (void *info), void *info,
 	func(info);
 	return 0;
 }
+#define smp_cpu_not_running(cpu)	1
 #define smp_get_cpu(cpu) ({ 0; })
 #define smp_put_cpu(cpu) ({ 0; })
 #endif

From e7684277f6882a23cfb734cb7450c3a469e6e8b1 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 11 Feb 2006 17:56:02 -0800
Subject: [PATCH 088/509] [PATCH] s390: add support for unshare system call

Add support for unshare system call.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/compat_wrapper.S | 5 +++++
 arch/s390/kernel/syscalls.S       | 1 +
 include/asm-s390/unistd.h         | 3 ++-
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index 83b33fe1923c..38a6ef5ec624 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -1602,3 +1602,8 @@ compat_sys_ppoll_wrapper:
 	llgtr	%r5,%r5			# const sigset_t *
 	llgfr	%r6,%r6			# size_t
 	jg	compat_sys_ppoll
+
+	.globl sys_unshare_wrapper
+sys_unshare_wrapper:
+	llgfr	%r2,%r2			# unsigned long
+	jg	sys_unshare
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 3280345efacd..e86a4debb160 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -311,3 +311,4 @@ SYSCALL(sys_fchmodat,sys_fchmodat,sys_fchmodat_wrapper)
 SYSCALL(sys_faccessat,sys_faccessat,sys_faccessat_wrapper)	/* 300 */
 SYSCALL(sys_pselect6,sys_pselect6,compat_sys_pselect6_wrapper)
 SYSCALL(sys_ppoll,sys_ppoll,compat_sys_ppoll_wrapper)
+SYSCALL(sys_unshare,sys_unshare,sys_unshare_wrapper)
diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h
index 29a9f357eb9e..0a2f6664c4cf 100644
--- a/include/asm-s390/unistd.h
+++ b/include/asm-s390/unistd.h
@@ -295,8 +295,9 @@
 #define __NR_faccessat		300
 #define __NR_pselect6		301
 #define __NR_ppoll		302
+#define __NR_unshare		303
 
-#define NR_syscalls 303
+#define NR_syscalls 304
 
 /* 
  * There are some system calls that are not present on 64 bit, some

From 0defa3c19e7792001df09d6fa5ab461d3599ff6d Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 11 Feb 2006 17:56:03 -0800
Subject: [PATCH 089/509] [PATCH] s390: add #ifdef __KERNEL__ to
 asm-s390/setup.h

Based on a patch from Maximilian Attems <maks@sternwelten.at> .  Nothing in
asm-s390/setup.h is of interest for user space.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-s390/setup.h | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/include/asm-s390/setup.h b/include/asm-s390/setup.h
index 348a88137445..da3fd4a7bb32 100644
--- a/include/asm-s390/setup.h
+++ b/include/asm-s390/setup.h
@@ -8,6 +8,8 @@
 #ifndef _ASM_S390_SETUP_H
 #define _ASM_S390_SETUP_H
 
+#ifdef __KERNEL__
+
 #include <asm/types.h>
 
 #define PARMAREA		0x10400
@@ -114,7 +116,7 @@ extern u16 ipl_devno;
 				 IPL_PARMBLOCK_ORIGIN)
 #define IPL_PARMBLOCK_SIZE	(IPL_PARMBLOCK_START->hdr.length)
 
-#else 
+#else /* __ASSEMBLY__ */
 
 #ifndef __s390x__
 #define IPL_DEVICE        0x10404
@@ -127,6 +129,6 @@ extern u16 ipl_devno;
 #endif /* __s390x__ */
 #define COMMAND_LINE      0x10480
 
-#endif
-
-#endif
+#endif /* __ASSEMBLY__ */
+#endif /* __KERNEL__ */
+#endif /* _ASM_S390_SETUP_H */

From ef1bea9e2a5a72d2c3362522e0a09099406732ff Mon Sep 17 00:00:00 2001
From: Eric Paris <eparis@redhat.com>
Date: Sat, 11 Feb 2006 17:56:04 -0800
Subject: [PATCH 090/509] [PATCH] s390: remove one set of brackets in
 __constant_test_bit()

Right now in __constant_test_bit for the s390 there is an extra set of ()
surrounding the calculation.  This patch simply removes one set of () that is
surrounding the whole clause.

Signed-off-by: Eric Paris <eparis@redhat.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-s390/bitops.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/asm-s390/bitops.h b/include/asm-s390/bitops.h
index 61232760cc3b..3628899f48bb 100644
--- a/include/asm-s390/bitops.h
+++ b/include/asm-s390/bitops.h
@@ -518,8 +518,8 @@ static inline int __test_bit(unsigned long nr, const volatile unsigned long *ptr
 
 static inline int 
 __constant_test_bit(unsigned long nr, const volatile unsigned long *addr) {
-    return ((((volatile char *) addr)
-	    [(nr^(__BITOPS_WORDSIZE-8))>>3] & (1<<(nr&7)))) != 0;
+    return (((volatile char *) addr)
+	    [(nr^(__BITOPS_WORDSIZE-8))>>3] & (1<<(nr&7))) != 0;
 }
 
 #define test_bit(nr,addr) \

From 1d30883942cfe8a1e3f88f8b7f4c292aeba3db5a Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Sat, 11 Feb 2006 17:56:05 -0800
Subject: [PATCH 091/509] [PATCH] tipar fixes

- tipar_open(): fix unsigned comparison

- tipar_open(): don't permit NULL pardevice (probably unneeded given the
  above fix).

- tipar_init_module(): handle the situation where parport_register_driver()
  failed to register any devices (parport_register_driver() drops the ->attach
  return value on the floor).

  This probably makes fixes #1 and #2 unneeded.

- tipar_init_module(): fix various error-path resource leaks.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/tipar.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/char/tipar.c b/drivers/char/tipar.c
index 41a94bc79f67..eb2eb3e12d6a 100644
--- a/drivers/char/tipar.c
+++ b/drivers/char/tipar.c
@@ -250,12 +250,17 @@ tipar_open(struct inode *inode, struct file *file)
 {
 	unsigned int minor = iminor(inode) - TIPAR_MINOR;
 
-	if (minor > tp_count - 1)
+	if (tp_count == 0 || minor > tp_count - 1)
 		return -ENXIO;
 
 	if (test_and_set_bit(minor, &opened))
 		return -EBUSY;
 
+	if (!table[minor].dev) {
+		printk(KERN_ERR "%s: NULL device for minor %u\n",
+				__FUNCTION__, minor);
+		return -ENXIO;
+	}
 	parport_claim_or_block(table[minor].dev);
 	init_ti_parallel(minor);
 	parport_release(table[minor].dev);
@@ -510,16 +515,20 @@ tipar_init_module(void)
 		err = PTR_ERR(tipar_class);
 		goto out_chrdev;
 	}
-	if (parport_register_driver(&tipar_driver)) {
+	if (parport_register_driver(&tipar_driver) || tp_count == 0) {
 		printk(KERN_ERR "tipar: unable to register with parport\n");
 		err = -EIO;
-		goto out;
+		goto out_class;
 	}
 
 	err = 0;
 	goto out;
 
+out_class:
+	class_destroy(tipar_class);
+
 out_chrdev:
+	devfs_remove("ticables/par");
 	unregister_chrdev(TIPAR_MAJOR, "tipar");
 out:
 	return err;	

From 891e5e5edaf13216f9f4c2710aebd066b1d98583 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Sat, 11 Feb 2006 17:56:05 -0800
Subject: [PATCH 092/509] [PATCH] drivers/video/Kconfig: remove unused BUS_I2C
 option

The BUS_I2C option is neither available (since there is no VISWS option in
the kernel) nor does it have any effect - so why not remove it?

Based on a report by Jean-Luc Leger <reiga@dspnet.fr.eu.org>.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/Kconfig | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 3e153d313bb0..e64ed16bd42f 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -525,11 +525,6 @@ config FB_GBE_MEM
 	  This is the amount of memory reserved for the framebuffer,
 	  which can be any value between 1MB and 8MB.
 
-config BUS_I2C
-	bool
-	depends on (FB = y) && VISWS
-	default y
-
 config FB_SUN3
 	bool "Sun3 framebuffer support"
 	depends on (FB = y) && (SUN3 || SUN3X) && BROKEN

From bc7fc0601b3eb2254f080492f3fd69e319ed32d0 Mon Sep 17 00:00:00 2001
From: "Antonino A. Daplas" <adaplas@gmail.com>
Date: Sat, 11 Feb 2006 17:56:07 -0800
Subject: [PATCH 093/509] [PATCH] nvidiafb: Add support for Geforce4 MX 4000

Add support for Geforce4 MX 4000 (0x185)

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/nvidia/nvidia.c | 2 ++
 include/linux/pci_ids.h       | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/video/nvidia/nvidia.c b/drivers/video/nvidia/nvidia.c
index dbcb8962e57d..a7c4e5e8ead6 100644
--- a/drivers/video/nvidia/nvidia.c
+++ b/drivers/video/nvidia/nvidia.c
@@ -138,6 +138,8 @@ static struct pci_device_id nvidiafb_pci_tbl[] = {
 	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
 	{PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_GEFORCE4_MX_420_8X,
 	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_GEFORCE4_MX_4000,
+	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
 	{PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_GEFORCE4_448_GO,
 	 PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
 	{PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_GEFORCE4_488_GO,
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 7a61ccdcbc4b..82b83da25d77 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1087,6 +1087,7 @@
 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_MX_440_8X 0x0181
 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_MX_440SE_8X 0x0182
 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_MX_420_8X 0x0183
+#define PCI_DEVICE_ID_NVIDIA_GEFORCE4_MX_4000   0x0185
 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_448_GO    0x0186
 #define PCI_DEVICE_ID_NVIDIA_GEFORCE4_488_GO    0x0187
 #define PCI_DEVICE_ID_NVIDIA_QUADRO4_580_XGL    0x0188

From bc6d7fdf460ec5292d66bb551dbfa49ca682bebf Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Sat, 11 Feb 2006 17:56:08 -0800
Subject: [PATCH 094/509] [PATCH] fbdev: video_setup() warning fix

drivers/video/fbmem.c:1567: warning: 'video_setup' defined but not used

Acked-by: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/fbmem.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index d2dede6ed3e5..996c7b58564e 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -1550,6 +1550,7 @@ int fb_get_options(char *name, char **option)
 	return retval;
 }
 
+#ifndef MODULE
 /**
  *	video_setup - process command line options
  *	@options: string of options
@@ -1593,6 +1594,7 @@ static int __init video_setup(char *options)
 	return 0;
 }
 __setup("video=", video_setup);
+#endif
 
     /*
      *  Visible symbols for modules

From 3a69e5791379a7c7d23c531a7679428300bb5072 Mon Sep 17 00:00:00 2001
From: Ian Campbell <icampbell@arcom.com>
Date: Mon, 7 Nov 2005 10:21:24 +0000
Subject: [PATCH 095/509] [WATCHDOG] sa1100_wdt.c sparse clean (2)

The following makes drivers/char/watchdog/sa1100_wdt.c sparse clean.
(similar to the other watchdog drivers)

Signed-off-by: Ian Campbell <icampbell@arcom.com>
Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/char/watchdog/sa1100_wdt.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/char/watchdog/sa1100_wdt.c b/drivers/char/watchdog/sa1100_wdt.c
index b474ea52d6e8..522a9370db94 100644
--- a/drivers/char/watchdog/sa1100_wdt.c
+++ b/drivers/char/watchdog/sa1100_wdt.c
@@ -93,23 +93,25 @@ static int sa1100dog_ioctl(struct inode *inode, struct file *file,
 {
 	int ret = -ENOIOCTLCMD;
 	int time;
+	void __user *argp = (void __user *)arg;
+	int __user *p = argp;
 
 	switch (cmd) {
 	case WDIOC_GETSUPPORT:
-		ret = copy_to_user((struct watchdog_info __user *)arg, &ident,
+		ret = copy_to_user(argp, &ident,
 				   sizeof(ident)) ? -EFAULT : 0;
 		break;
 
 	case WDIOC_GETSTATUS:
-		ret = put_user(0, (int __user *)arg);
+		ret = put_user(0, p);
 		break;
 
 	case WDIOC_GETBOOTSTATUS:
-		ret = put_user(boot_status, (int __user *)arg);
+		ret = put_user(boot_status, p);
 		break;
 
 	case WDIOC_SETTIMEOUT:
-		ret = get_user(time, (int __user *)arg);
+		ret = get_user(time, p);
 		if (ret)
 			break;
 
@@ -123,7 +125,7 @@ static int sa1100dog_ioctl(struct inode *inode, struct file *file,
 		/*fall through*/
 
 	case WDIOC_GETTIMEOUT:
-		ret = put_user(pre_margin / OSCR_FREQ, (int __user *)arg);
+		ret = put_user(pre_margin / OSCR_FREQ, p);
 		break;
 
 	case WDIOC_KEEPALIVE:

From fd41fa616f21efc36eb80696475ceb33ea047a6a Mon Sep 17 00:00:00 2001
From: Wim Van Sebroeck <wim@iguana.be>
Date: Sat, 10 Dec 2005 14:22:37 +0100
Subject: [PATCH 096/509] [WATCHDOG] pcwd.c add comments + tabs

add extra comments for the include files
changes spaces by tabs where it is appropriate.

Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/char/watchdog/pcwd.c | 71 +++++++++++++++++++-----------------
 1 file changed, 37 insertions(+), 34 deletions(-)

diff --git a/drivers/char/watchdog/pcwd.c b/drivers/char/watchdog/pcwd.c
index 37c9e13ad3ac..85c94b205df1 100644
--- a/drivers/char/watchdog/pcwd.c
+++ b/drivers/char/watchdog/pcwd.c
@@ -49,27 +49,30 @@
  *	More info available at http://www.berkprod.com/ or http://www.pcwatchdog.com/
  */
 
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/types.h>
-#include <linux/timer.h>
-#include <linux/jiffies.h>
-#include <linux/config.h>
-#include <linux/wait.h>
-#include <linux/slab.h>
-#include <linux/ioport.h>
-#include <linux/delay.h>
-#include <linux/fs.h>
-#include <linux/miscdevice.h>
-#include <linux/watchdog.h>
-#include <linux/notifier.h>
-#include <linux/init.h>
-#include <linux/spinlock.h>
-#include <linux/reboot.h>
+#include <linux/config.h>	/* For CONFIG_WATCHDOG_NOWAYOUT/... */
+#include <linux/module.h>	/* For module specific items */
+#include <linux/moduleparam.h>	/* For new moduleparam's */
+#include <linux/types.h>	/* For standard types (like size_t) */
+#include <linux/errno.h>	/* For the -ENODEV/... values */
+#include <linux/kernel.h>	/* For printk/panic/... */
+#include <linux/delay.h>	/* For mdelay function */
+#include <linux/timer.h>	/* For timer related operations */
+#include <linux/jiffies.h>	/* For jiffies stuff */
+#include <linux/miscdevice.h>	/* For MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR) */
+#include <linux/watchdog.h>	/* For the watchdog specific items */
+#include <linux/notifier.h>	/* For notifier support */
+#include <linux/reboot.h>	/* For reboot_notifier stuff */
+#include <linux/init.h>		/* For __init/__exit/... */
+#include <linux/fs.h>		/* For file operations */
+#include <linux/ioport.h>	/* For io-port access */
+#include <linux/spinlock.h>	/* For spin_lock/spin_unlock/... */
 #include <linux/sched.h>	/* TASK_INTERRUPTIBLE, set_current_state() and friends */
-#include <asm/uaccess.h>
-#include <asm/io.h>
+#include <linux/slab.h>		/* For kmalloc */
 
+#include <asm/uaccess.h>	/* For copy_to_user/put_user/... */
+#include <asm/io.h>		/* For inb/outb/... */
+
+/* Module and version information */
 #define WD_VER                  "1.16 (06/12/2004)"
 #define PFX			"pcwd: "
 
@@ -84,7 +87,7 @@
 #define	PCWD_REVISION_C		2
 
 /*
- * These are the defines that describe the control status bits for the
+ * These are the defines that describe the control status #1 bits for the
  * PC Watchdog card, revision A.
  */
 #define WD_WDRST                0x01	/* Previously reset state */
@@ -94,7 +97,7 @@
 #define WD_SRLY2                0x80	/* Software external relay triggered */
 
 /*
- * These are the defines that describe the control status bits for the
+ * These are the defines that describe the control status #1 bits for the
  * PC Watchdog card, revision C.
  */
 #define WD_REVC_WTRP            0x01	/* Watchdog Trip status */
@@ -106,15 +109,15 @@
 #define ISA_COMMAND_TIMEOUT     1000
 
 /* Watchdog's internal commands */
-#define CMD_ISA_IDLE                    0x00
-#define CMD_ISA_VERSION_INTEGER         0x01
-#define CMD_ISA_VERSION_TENTH           0x02
-#define CMD_ISA_VERSION_HUNDRETH        0x03
-#define CMD_ISA_VERSION_MINOR           0x04
-#define CMD_ISA_SWITCH_SETTINGS         0x05
-#define CMD_ISA_DELAY_TIME_2SECS        0x0A
-#define CMD_ISA_DELAY_TIME_4SECS        0x0B
-#define CMD_ISA_DELAY_TIME_8SECS        0x0C
+#define CMD_ISA_IDLE			0x00
+#define CMD_ISA_VERSION_INTEGER		0x01
+#define CMD_ISA_VERSION_TENTH		0x02
+#define CMD_ISA_VERSION_HUNDRETH	0x03
+#define CMD_ISA_VERSION_MINOR		0x04
+#define CMD_ISA_SWITCH_SETTINGS		0x05
+#define CMD_ISA_DELAY_TIME_2SECS	0x0A
+#define CMD_ISA_DELAY_TIME_4SECS	0x0B
+#define CMD_ISA_DELAY_TIME_8SECS	0x0C
 
 /*
  * We are using an kernel timer to do the pinging of the watchdog
@@ -767,10 +770,10 @@ static int __devinit pcwatchdog_init(int base_addr)
 		printk(KERN_INFO PFX "No previous trip detected - Cold boot or reset\n");
 
 	/* Check that the heartbeat value is within it's range ; if not reset to the default */
-        if (pcwd_set_heartbeat(heartbeat)) {
-                pcwd_set_heartbeat(WATCHDOG_HEARTBEAT);
-                printk(KERN_INFO PFX "heartbeat value must be 2<=heartbeat<=7200, using %d\n",
-                        WATCHDOG_HEARTBEAT);
+	if (pcwd_set_heartbeat(heartbeat)) {
+		pcwd_set_heartbeat(WATCHDOG_HEARTBEAT);
+		printk(KERN_INFO PFX "heartbeat value must be 2<=heartbeat<=7200, using %d\n",
+			WATCHDOG_HEARTBEAT);
 	}
 
 	ret = register_reboot_notifier(&pcwd_notifier);

From f1c3a0567aa5086e755e58385740f9ece911c06e Mon Sep 17 00:00:00 2001
From: Wim Van Sebroeck <wim@iguana.be>
Date: Sat, 10 Dec 2005 14:36:24 +0100
Subject: [PATCH 097/509] [WATCHDOG] pcwd.c card_found-- fix.

When doing a __devexit from a card we should also
decrement the cards_found counter.

Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/char/watchdog/pcwd.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/char/watchdog/pcwd.c b/drivers/char/watchdog/pcwd.c
index 85c94b205df1..3329cbffdb34 100644
--- a/drivers/char/watchdog/pcwd.c
+++ b/drivers/char/watchdog/pcwd.c
@@ -828,6 +828,7 @@ static void __devexit pcwatchdog_exit(void)
 	unregister_reboot_notifier(&pcwd_notifier);
 	release_region(current_readport, (revision == PCWD_REVISION_A) ? 2 : 4);
 	current_readport = 0x0000;
+	cards_found--;
 }
 
 /*

From a2be8786006ec0d21dcb1d322fc480b85ea82c66 Mon Sep 17 00:00:00 2001
From: Wim Van Sebroeck <wim@iguana.be>
Date: Mon, 9 Jan 2006 21:53:33 +0100
Subject: [PATCH 098/509] [WATCHDOG] pcwd.c private data struct patch

more private data of the card to one struct.

Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/char/watchdog/pcwd.c | 206 ++++++++++++++++++-----------------
 1 file changed, 104 insertions(+), 102 deletions(-)

diff --git a/drivers/char/watchdog/pcwd.c b/drivers/char/watchdog/pcwd.c
index 3329cbffdb34..1112ec8e61f3 100644
--- a/drivers/char/watchdog/pcwd.c
+++ b/drivers/char/watchdog/pcwd.c
@@ -73,7 +73,7 @@
 #include <asm/io.h>		/* For inb/outb/... */
 
 /* Module and version information */
-#define WD_VER                  "1.16 (06/12/2004)"
+#define WD_VER                  "1.16 (03/01/2006)"
 #define PFX			"pcwd: "
 
 /*
@@ -133,15 +133,17 @@ static int cards_found;
 /* internal variables */
 static atomic_t open_allowed = ATOMIC_INIT(1);
 static char expect_close;
-static struct timer_list timer;
-static unsigned long next_heartbeat;
 static int temp_panic;
-static int revision;			/* The card's revision */
-static int supports_temp;		/* Wether or not the card has a temperature device */
-static int command_mode;		/* Wether or not the card is in command mode */
-static int initial_status;		/* The card's boot status */
-static int current_readport;		/* The cards I/O address */
-static spinlock_t io_lock;
+static struct {				/* this is private data for each ISA-PC watchdog card */
+	int revision;			/* The card's revision */
+	int supports_temp;		/* Wether or not the card has a temperature device */
+	int command_mode;		/* Wether or not the card is in command mode */
+	int boot_status;		/* The card's boot status */
+	int io_addr;			/* The cards I/O address */
+	spinlock_t io_lock;		/* the lock for io operations */
+	struct timer_list timer;	/* The timer that pings the watchdog */
+	unsigned long next_heartbeat;	/* the next_heartbeat for the timer */
+} pcwd_private;
 
 /* module parameters */
 #define WATCHDOG_HEARTBEAT 60		/* 60 sec default heartbeat */
@@ -165,13 +167,13 @@ static int send_isa_command(int cmd)
 
 	/* The WCMD bit must be 1 and the command is only 4 bits in size */
 	control_status = (cmd & 0x0F) | 0x80;
-	outb_p(control_status, current_readport + 2);
+	outb_p(control_status, pcwd_private.io_addr + 2);
 	udelay(ISA_COMMAND_TIMEOUT);
 
-	port0 = inb_p(current_readport);
+	port0 = inb_p(pcwd_private.io_addr);
 	for (i = 0; i < 25; ++i) {
 		last_port0 = port0;
-		port0 = inb_p(current_readport);
+		port0 = inb_p(pcwd_private.io_addr);
 
 		if (port0 == last_port0)
 			break;	/* Data is stable */
@@ -187,7 +189,7 @@ static int set_command_mode(void)
 	int i, found=0, count=0;
 
 	/* Set the card into command mode */
-	spin_lock(&io_lock);
+	spin_lock(&pcwd_private.io_lock);
 	while ((!found) && (count < 3)) {
 		i = send_isa_command(CMD_ISA_IDLE);
 
@@ -195,15 +197,15 @@ static int set_command_mode(void)
 			found = 1;
 		else if (i == 0xF3) {
 			/* Card does not like what we've done to it */
-			outb_p(0x00, current_readport + 2);
+			outb_p(0x00, pcwd_private.io_addr + 2);
 			udelay(1200);	/* Spec says wait 1ms */
-			outb_p(0x00, current_readport + 2);
+			outb_p(0x00, pcwd_private.io_addr + 2);
 			udelay(ISA_COMMAND_TIMEOUT);
 		}
 		count++;
 	}
-	spin_unlock(&io_lock);
-	command_mode = found;
+	spin_unlock(&pcwd_private.io_lock);
+	pcwd_private.command_mode = found;
 
 	return(found);
 }
@@ -211,12 +213,12 @@ static int set_command_mode(void)
 static void unset_command_mode(void)
 {
 	/* Set the card into normal mode */
-	spin_lock(&io_lock);
-	outb_p(0x00, current_readport + 2);
+	spin_lock(&pcwd_private.io_lock);
+	outb_p(0x00, pcwd_private.io_addr + 2);
 	udelay(ISA_COMMAND_TIMEOUT);
-	spin_unlock(&io_lock);
+	spin_unlock(&pcwd_private.io_lock);
 
-	command_mode = 0;
+	pcwd_private.command_mode = 0;
 }
 
 static void pcwd_timer_ping(unsigned long data)
@@ -225,25 +227,25 @@ static void pcwd_timer_ping(unsigned long data)
 
 	/* If we got a heartbeat pulse within the WDT_INTERVAL
 	 * we agree to ping the WDT */
-	if(time_before(jiffies, next_heartbeat)) {
+	if(time_before(jiffies, pcwd_private.next_heartbeat)) {
 		/* Ping the watchdog */
-		spin_lock(&io_lock);
-		if (revision == PCWD_REVISION_A) {
+		spin_lock(&pcwd_private.io_lock);
+		if (pcwd_private.revision == PCWD_REVISION_A) {
 			/*  Rev A cards are reset by setting the WD_WDRST bit in register 1 */
-			wdrst_stat = inb_p(current_readport);
+			wdrst_stat = inb_p(pcwd_private.io_addr);
 			wdrst_stat &= 0x0F;
 			wdrst_stat |= WD_WDRST;
 
-			outb_p(wdrst_stat, current_readport + 1);
+			outb_p(wdrst_stat, pcwd_private.io_addr + 1);
 		} else {
 			/* Re-trigger watchdog by writing to port 0 */
-			outb_p(0x00, current_readport);
+			outb_p(0x00, pcwd_private.io_addr);
 		}
 
 		/* Re-set the timer interval */
-		mod_timer(&timer, jiffies + WDT_INTERVAL);
+		mod_timer(&pcwd_private.timer, jiffies + WDT_INTERVAL);
 
-		spin_unlock(&io_lock);
+		spin_unlock(&pcwd_private.io_lock);
 	} else {
 		printk(KERN_WARNING PFX "Heartbeat lost! Will not ping the watchdog\n");
 	}
@@ -253,18 +255,18 @@ static int pcwd_start(void)
 {
 	int stat_reg;
 
-	next_heartbeat = jiffies + (heartbeat * HZ);
+	pcwd_private.next_heartbeat = jiffies + (heartbeat * HZ);
 
 	/* Start the timer */
-	mod_timer(&timer, jiffies + WDT_INTERVAL);
+	mod_timer(&pcwd_private.timer, jiffies + WDT_INTERVAL);
 
 	/* Enable the port */
-	if (revision == PCWD_REVISION_C) {
-		spin_lock(&io_lock);
-		outb_p(0x00, current_readport + 3);
+	if (pcwd_private.revision == PCWD_REVISION_C) {
+		spin_lock(&pcwd_private.io_lock);
+		outb_p(0x00, pcwd_private.io_addr + 3);
 		udelay(ISA_COMMAND_TIMEOUT);
-		stat_reg = inb_p(current_readport + 2);
-		spin_unlock(&io_lock);
+		stat_reg = inb_p(pcwd_private.io_addr + 2);
+		spin_unlock(&pcwd_private.io_lock);
 		if (stat_reg & 0x10) {
 			printk(KERN_INFO PFX "Could not start watchdog\n");
 			return -EIO;
@@ -278,17 +280,17 @@ static int pcwd_stop(void)
 	int stat_reg;
 
 	/* Stop the timer */
-	del_timer(&timer);
+	del_timer(&pcwd_private.timer);
 
 	/*  Disable the board  */
-	if (revision == PCWD_REVISION_C) {
-		spin_lock(&io_lock);
-		outb_p(0xA5, current_readport + 3);
+	if (pcwd_private.revision == PCWD_REVISION_C) {
+		spin_lock(&pcwd_private.io_lock);
+		outb_p(0xA5, pcwd_private.io_addr + 3);
 		udelay(ISA_COMMAND_TIMEOUT);
-		outb_p(0xA5, current_readport + 3);
+		outb_p(0xA5, pcwd_private.io_addr + 3);
 		udelay(ISA_COMMAND_TIMEOUT);
-		stat_reg = inb_p(current_readport + 2);
-		spin_unlock(&io_lock);
+		stat_reg = inb_p(pcwd_private.io_addr + 2);
+		spin_unlock(&pcwd_private.io_lock);
 		if ((stat_reg & 0x10) == 0) {
 			printk(KERN_INFO PFX "Could not stop watchdog\n");
 			return -EIO;
@@ -300,7 +302,7 @@ static int pcwd_stop(void)
 static int pcwd_keepalive(void)
 {
 	/* user land ping */
-	next_heartbeat = jiffies + (heartbeat * HZ);
+	pcwd_private.next_heartbeat = jiffies + (heartbeat * HZ);
 	return 0;
 }
 
@@ -318,23 +320,23 @@ static int pcwd_get_status(int *status)
 	int card_status;
 
 	*status=0;
-	spin_lock(&io_lock);
-	if (revision == PCWD_REVISION_A)
+	spin_lock(&pcwd_private.io_lock);
+	if (pcwd_private.revision == PCWD_REVISION_A)
 		/* Rev A cards return status information from
 		 * the base register, which is used for the
 		 * temperature in other cards. */
-		card_status = inb(current_readport);
+		card_status = inb(pcwd_private.io_addr);
 	else {
 		/* Rev C cards return card status in the base
 		 * address + 1 register. And use different bits
 		 * to indicate a card initiated reset, and an
 		 * over-temperature condition. And the reboot
 		 * status can be reset. */
-		card_status = inb(current_readport + 1);
+		card_status = inb(pcwd_private.io_addr + 1);
 	}
-	spin_unlock(&io_lock);
+	spin_unlock(&pcwd_private.io_lock);
 
-	if (revision == PCWD_REVISION_A) {
+	if (pcwd_private.revision == PCWD_REVISION_A) {
 		if (card_status & WD_WDRST)
 			*status |= WDIOF_CARDRESET;
 
@@ -363,10 +365,10 @@ static int pcwd_get_status(int *status)
 
 static int pcwd_clear_status(void)
 {
-	if (revision == PCWD_REVISION_C) {
-		spin_lock(&io_lock);
-		outb_p(0x00, current_readport + 1); /* clear reset status */
-		spin_unlock(&io_lock);
+	if (pcwd_private.revision == PCWD_REVISION_C) {
+		spin_lock(&pcwd_private.io_lock);
+		outb_p(0x00, pcwd_private.io_addr + 1); /* clear reset status */
+		spin_unlock(&pcwd_private.io_lock);
 	}
 	return 0;
 }
@@ -374,20 +376,20 @@ static int pcwd_clear_status(void)
 static int pcwd_get_temperature(int *temperature)
 {
 	/* check that port 0 gives temperature info and no command results */
-	if (command_mode)
+	if (pcwd_private.command_mode)
 		return -1;
 
 	*temperature = 0;
-	if (!supports_temp)
+	if (!pcwd_private.supports_temp)
 		return -ENODEV;
 
 	/*
 	 * Convert celsius to fahrenheit, since this was
 	 * the decided 'standard' for this return value.
 	 */
-	spin_lock(&io_lock);
-	*temperature = ((inb(current_readport)) * 9 / 5) + 32;
-	spin_unlock(&io_lock);
+	spin_lock(&pcwd_private.io_lock);
+	*temperature = ((inb(pcwd_private.io_addr)) * 9 / 5) + 32;
+	spin_unlock(&pcwd_private.io_lock);
 
 	return 0;
 }
@@ -428,7 +430,7 @@ static int pcwd_ioctl(struct inode *inode, struct file *file,
 		return put_user(status, argp);
 
 	case WDIOC_GETBOOTSTATUS:
-		return put_user(initial_status, argp);
+		return put_user(pcwd_private.boot_status, argp);
 
 	case WDIOC_GETTEMP:
 		if (pcwd_get_temperature(&temperature))
@@ -437,7 +439,7 @@ static int pcwd_ioctl(struct inode *inode, struct file *file,
 		return put_user(temperature, argp);
 
 	case WDIOC_SETOPTIONS:
-		if (revision == PCWD_REVISION_C)
+		if (pcwd_private.revision == PCWD_REVISION_C)
 		{
 			if(copy_from_user(&rv, argp, sizeof(int)))
 				return -EFAULT;
@@ -553,7 +555,7 @@ static ssize_t pcwd_temp_read(struct file *file, char __user *buf, size_t count,
 
 static int pcwd_temp_open(struct inode *inode, struct file *file)
 {
-	if (!supports_temp)
+	if (!pcwd_private.supports_temp)
 		return -ENODEV;
 
 	return nonseekable_open(inode, file);
@@ -621,21 +623,21 @@ static struct notifier_block pcwd_notifier = {
 
 static inline void get_support(void)
 {
-	if (inb(current_readport) != 0xF0)
-		supports_temp = 1;
+	if (inb(pcwd_private.io_addr) != 0xF0)
+		pcwd_private.supports_temp = 1;
 }
 
 static inline int get_revision(void)
 {
 	int r = PCWD_REVISION_C;
 
-	spin_lock(&io_lock);
+	spin_lock(&pcwd_private.io_lock);
 	/* REV A cards use only 2 io ports; test
 	 * presumes a floating bus reads as 0xff. */
-	if ((inb(current_readport + 2) == 0xFF) ||
-	    (inb(current_readport + 3) == 0xFF))
+	if ((inb(pcwd_private.io_addr + 2) == 0xFF) ||
+	    (inb(pcwd_private.io_addr + 3) == 0xFF))
 		r=PCWD_REVISION_A;
-	spin_unlock(&io_lock);
+	spin_unlock(&pcwd_private.io_lock);
 
 	return r;
 }
@@ -695,32 +697,32 @@ static int __devinit pcwatchdog_init(int base_addr)
 		printk(KERN_ERR PFX "No I/O-Address for card detected\n");
 		return -ENODEV;
 	}
-	current_readport = base_addr;
+	pcwd_private.io_addr = base_addr;
 
 	/* Check card's revision */
-	revision = get_revision();
+	pcwd_private.revision = get_revision();
 
-	if (!request_region(current_readport, (revision == PCWD_REVISION_A) ? 2 : 4, "PCWD")) {
+	if (!request_region(pcwd_private.io_addr, (pcwd_private.revision == PCWD_REVISION_A) ? 2 : 4, "PCWD")) {
 		printk(KERN_ERR PFX "I/O address 0x%04x already in use\n",
-			current_readport);
-		current_readport = 0x0000;
+			pcwd_private.io_addr);
+		pcwd_private.io_addr = 0x0000;
 		return -EIO;
 	}
 
 	/* Initial variables */
-	supports_temp = 0;
+	pcwd_private.supports_temp = 0;
 	temp_panic = 0;
-	initial_status = 0x0000;
+	pcwd_private.boot_status = 0x0000;
 
 	/* get the boot_status */
-	pcwd_get_status(&initial_status);
+	pcwd_get_status(&pcwd_private.boot_status);
 
 	/* clear the "card caused reboot" flag */
 	pcwd_clear_status();
 
-	init_timer(&timer);
-	timer.function = pcwd_timer_ping;
-	timer.data = 0;
+	init_timer(&pcwd_private.timer);
+	pcwd_private.timer.function = pcwd_timer_ping;
+	pcwd_private.timer.data = 0;
 
 	/*  Disable the board  */
 	pcwd_stop();
@@ -729,12 +731,12 @@ static int __devinit pcwatchdog_init(int base_addr)
 	get_support();
 
 	/* Get some extra info from the hardware (in command/debug/diag mode) */
-	if (revision == PCWD_REVISION_A)
-		printk(KERN_INFO PFX "ISA-PC Watchdog (REV.A) detected at port 0x%04x\n", current_readport);
-	else if (revision == PCWD_REVISION_C) {
+	if (pcwd_private.revision == PCWD_REVISION_A)
+		printk(KERN_INFO PFX "ISA-PC Watchdog (REV.A) detected at port 0x%04x\n", pcwd_private.io_addr);
+	else if (pcwd_private.revision == PCWD_REVISION_C) {
 		firmware = get_firmware();
 		printk(KERN_INFO PFX "ISA-PC Watchdog (REV.C) detected at port 0x%04x (Firmware version: %s)\n",
-			current_readport, firmware);
+			pcwd_private.io_addr, firmware);
 		kfree(firmware);
 		option_switches = get_option_switches();
 		printk(KERN_INFO PFX "Option switches (0x%02x): Temperature Reset Enable=%s, Power On Delay=%s\n",
@@ -750,23 +752,23 @@ static int __devinit pcwatchdog_init(int base_addr)
 	} else {
 		/* Should NEVER happen, unless get_revision() fails. */
 		printk(KERN_INFO PFX "Unable to get revision\n");
-		release_region(current_readport, (revision == PCWD_REVISION_A) ? 2 : 4);
-		current_readport = 0x0000;
+		release_region(pcwd_private.io_addr, (pcwd_private.revision == PCWD_REVISION_A) ? 2 : 4);
+		pcwd_private.io_addr = 0x0000;
 		return -1;
 	}
 
-	if (supports_temp)
+	if (pcwd_private.supports_temp)
 		printk(KERN_INFO PFX "Temperature Option Detected\n");
 
-	if (initial_status & WDIOF_CARDRESET)
+	if (pcwd_private.boot_status & WDIOF_CARDRESET)
 		printk(KERN_INFO PFX "Previous reboot was caused by the card\n");
 
-	if (initial_status & WDIOF_OVERHEAT) {
+	if (pcwd_private.boot_status & WDIOF_OVERHEAT) {
 		printk(KERN_EMERG PFX "Card senses a CPU Overheat. Panicking!\n");
 		printk(KERN_EMERG PFX "CPU Overheat\n");
 	}
 
-	if (initial_status == 0)
+	if (pcwd_private.boot_status == 0)
 		printk(KERN_INFO PFX "No previous trip detected - Cold boot or reset\n");
 
 	/* Check that the heartbeat value is within it's range ; if not reset to the default */
@@ -780,19 +782,19 @@ static int __devinit pcwatchdog_init(int base_addr)
 	if (ret) {
 		printk(KERN_ERR PFX "cannot register reboot notifier (err=%d)\n",
 			ret);
-		release_region(current_readport, (revision == PCWD_REVISION_A) ? 2 : 4);
-		current_readport = 0x0000;
+		release_region(pcwd_private.io_addr, (pcwd_private.revision == PCWD_REVISION_A) ? 2 : 4);
+		pcwd_private.io_addr = 0x0000;
 		return ret;
 	}
 
-	if (supports_temp) {
+	if (pcwd_private.supports_temp) {
 		ret = misc_register(&temp_miscdev);
 		if (ret) {
 			printk(KERN_ERR PFX "cannot register miscdev on minor=%d (err=%d)\n",
 				TEMP_MINOR, ret);
 			unregister_reboot_notifier(&pcwd_notifier);
-			release_region(current_readport, (revision == PCWD_REVISION_A) ? 2 : 4);
-			current_readport = 0x0000;
+			release_region(pcwd_private.io_addr, (pcwd_private.revision == PCWD_REVISION_A) ? 2 : 4);
+			pcwd_private.io_addr = 0x0000;
 			return ret;
 		}
 	}
@@ -801,11 +803,11 @@ static int __devinit pcwatchdog_init(int base_addr)
 	if (ret) {
 		printk(KERN_ERR PFX "cannot register miscdev on minor=%d (err=%d)\n",
 			WATCHDOG_MINOR, ret);
-		if (supports_temp)
+		if (pcwd_private.supports_temp)
 			misc_deregister(&temp_miscdev);
 		unregister_reboot_notifier(&pcwd_notifier);
-		release_region(current_readport, (revision == PCWD_REVISION_A) ? 2 : 4);
-		current_readport = 0x0000;
+		release_region(pcwd_private.io_addr, (pcwd_private.revision == PCWD_REVISION_A) ? 2 : 4);
+		pcwd_private.io_addr = 0x0000;
 		return ret;
 	}
 
@@ -823,11 +825,11 @@ static void __devexit pcwatchdog_exit(void)
 
 	/* Deregister */
 	misc_deregister(&pcwd_miscdev);
-	if (supports_temp)
+	if (pcwd_private.supports_temp)
 		misc_deregister(&temp_miscdev);
 	unregister_reboot_notifier(&pcwd_notifier);
-	release_region(current_readport, (revision == PCWD_REVISION_A) ? 2 : 4);
-	current_readport = 0x0000;
+	release_region(pcwd_private.io_addr, (pcwd_private.revision == PCWD_REVISION_A) ? 2 : 4);
+	pcwd_private.io_addr = 0x0000;
 	cards_found--;
 }
 
@@ -891,7 +893,7 @@ static int __init pcwd_init_module(void)
 {
 	int i, found = 0;
 
-	spin_lock_init(&io_lock);
+	spin_lock_init(&pcwd_private.io_lock);
 
 	for (i = 0; pcwd_ioports[i] != 0; i++) {
 		if (pcwd_checkcard(pcwd_ioports[i])) {
@@ -910,7 +912,7 @@ static int __init pcwd_init_module(void)
 
 static void __exit pcwd_cleanup_module(void)
 {
-	if (current_readport)
+	if (pcwd_private.io_addr)
 		pcwatchdog_exit();
 	return;
 }

From 8f0235dccc3f7bffc32abcef2aec3d1b15c61927 Mon Sep 17 00:00:00 2001
From: Wim Van Sebroeck <wim@iguana.be>
Date: Mon, 9 Jan 2006 21:56:09 +0100
Subject: [PATCH 099/509] [WATCHDOG] pcwd.c Control Status #2 patch

Add Control Status #2 bits (with defines)

Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/char/watchdog/pcwd.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)

diff --git a/drivers/char/watchdog/pcwd.c b/drivers/char/watchdog/pcwd.c
index 1112ec8e61f3..0635cd724365 100644
--- a/drivers/char/watchdog/pcwd.c
+++ b/drivers/char/watchdog/pcwd.c
@@ -87,22 +87,24 @@
 #define	PCWD_REVISION_C		2
 
 /*
- * These are the defines that describe the control status #1 bits for the
- * PC Watchdog card, revision A.
- */
+ * These are the defines that describe the control status bits for the
+ * PCI-PC Watchdog card.
+*/
+/* Port 1 : Control Status #1 for the PC Watchdog card, revision A. */
 #define WD_WDRST                0x01	/* Previously reset state */
 #define WD_T110                 0x02	/* Temperature overheat sense */
 #define WD_HRTBT                0x04	/* Heartbeat sense */
 #define WD_RLY2                 0x08	/* External relay triggered */
 #define WD_SRLY2                0x80	/* Software external relay triggered */
-
-/*
- * These are the defines that describe the control status #1 bits for the
- * PC Watchdog card, revision C.
- */
+/* Port 1 : Control Status #1 for the PC Watchdog card, revision C. */
 #define WD_REVC_WTRP            0x01	/* Watchdog Trip status */
 #define WD_REVC_HRBT            0x02	/* Watchdog Heartbeat */
 #define WD_REVC_TTRP            0x04	/* Temperature Trip status */
+/* Port 2 : Control Status #2 */
+#define WD_WDIS			0x10	/* Watchdog Disabled */
+#define WD_ENTP			0x20	/* Watchdog Enable Temperature Trip */
+#define WD_SSEL			0x40	/* Watchdog Switch Select (1:SW1 <-> 0:SW2) */
+#define WD_WCMD			0x80	/* Watchdog Command Mode */
 
 /* max. time we give an ISA watchdog card to process a command */
 /* 500ms for each 4 bit response (according to spec.) */
@@ -166,7 +168,7 @@ static int send_isa_command(int cmd)
 	int port0, last_port0;	/* Double read for stabilising */
 
 	/* The WCMD bit must be 1 and the command is only 4 bits in size */
-	control_status = (cmd & 0x0F) | 0x80;
+	control_status = (cmd & 0x0F) | WD_WCMD;
 	outb_p(control_status, pcwd_private.io_addr + 2);
 	udelay(ISA_COMMAND_TIMEOUT);
 
@@ -267,7 +269,7 @@ static int pcwd_start(void)
 		udelay(ISA_COMMAND_TIMEOUT);
 		stat_reg = inb_p(pcwd_private.io_addr + 2);
 		spin_unlock(&pcwd_private.io_lock);
-		if (stat_reg & 0x10) {
+		if (stat_reg & WD_WDIS) {
 			printk(KERN_INFO PFX "Could not start watchdog\n");
 			return -EIO;
 		}
@@ -291,7 +293,7 @@ static int pcwd_stop(void)
 		udelay(ISA_COMMAND_TIMEOUT);
 		stat_reg = inb_p(pcwd_private.io_addr + 2);
 		spin_unlock(&pcwd_private.io_lock);
-		if ((stat_reg & 0x10) == 0) {
+		if ((stat_reg & WD_WDIS) == 0) {
 			printk(KERN_INFO PFX "Could not stop watchdog\n");
 			return -EIO;
 		}

From 85875211acc94ecb76fe04fbebc6aca12b6da60d Mon Sep 17 00:00:00 2001
From: Wim Van Sebroeck <wim@iguana.be>
Date: Mon, 9 Jan 2006 21:59:39 +0100
Subject: [PATCH 100/509] [WATCHDOG] pcwd.c move get_support to
 pcwd_check_temperature_support

Rename get_support function to pcwd_check_temperature_support
so that it is clearer what the function does.

Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/char/watchdog/pcwd.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/char/watchdog/pcwd.c b/drivers/char/watchdog/pcwd.c
index 0635cd724365..0549b2e89966 100644
--- a/drivers/char/watchdog/pcwd.c
+++ b/drivers/char/watchdog/pcwd.c
@@ -223,6 +223,12 @@ static void unset_command_mode(void)
 	pcwd_private.command_mode = 0;
 }
 
+static inline void pcwd_check_temperature_support(void)
+{
+	if (inb(pcwd_private.io_addr) != 0xF0)
+		pcwd_private.supports_temp = 1;
+}
+
 static void pcwd_timer_ping(unsigned long data)
 {
 	int wdrst_stat;
@@ -623,12 +629,6 @@ static struct notifier_block pcwd_notifier = {
  *	Init & exit routines
  */
 
-static inline void get_support(void)
-{
-	if (inb(pcwd_private.io_addr) != 0xF0)
-		pcwd_private.supports_temp = 1;
-}
-
 static inline int get_revision(void)
 {
 	int r = PCWD_REVISION_C;
@@ -730,7 +730,7 @@ static int __devinit pcwatchdog_init(int base_addr)
 	pcwd_stop();
 
 	/*  Check whether or not the card supports the temperature device */
-	get_support();
+	pcwd_check_temperature_support();
 
 	/* Get some extra info from the hardware (in command/debug/diag mode) */
 	if (pcwd_private.revision == PCWD_REVISION_A)

From af3b38d99d7d52340cf59a06ff90d90e0fa25b6d Mon Sep 17 00:00:00 2001
From: Wim Van Sebroeck <wim@iguana.be>
Date: Mon, 9 Jan 2006 22:03:41 +0100
Subject: [PATCH 101/509] [WATCHDOG] pcwd.c show card info patch

Put all code for showing the card's boot info in
one sub-routine.

Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/char/watchdog/pcwd.c | 157 ++++++++++++++++++-----------------
 1 file changed, 79 insertions(+), 78 deletions(-)

diff --git a/drivers/char/watchdog/pcwd.c b/drivers/char/watchdog/pcwd.c
index 0549b2e89966..f8f80d52d4de 100644
--- a/drivers/char/watchdog/pcwd.c
+++ b/drivers/char/watchdog/pcwd.c
@@ -229,6 +229,83 @@ static inline void pcwd_check_temperature_support(void)
 		pcwd_private.supports_temp = 1;
 }
 
+static inline char *get_firmware(void)
+{
+	int one, ten, hund, minor;
+	char *ret;
+
+	ret = kmalloc(6, GFP_KERNEL);
+	if(ret == NULL)
+		return NULL;
+
+	if (set_command_mode()) {
+		one = send_isa_command(CMD_ISA_VERSION_INTEGER);
+		ten = send_isa_command(CMD_ISA_VERSION_TENTH);
+		hund = send_isa_command(CMD_ISA_VERSION_HUNDRETH);
+		minor = send_isa_command(CMD_ISA_VERSION_MINOR);
+		sprintf(ret, "%c.%c%c%c", one, ten, hund, minor);
+	}
+	else
+		sprintf(ret, "ERROR");
+
+	unset_command_mode();
+	return(ret);
+}
+
+static inline int pcwd_get_option_switches(void)
+{
+	int option_switches=0;
+
+	if (set_command_mode()) {
+		/* Get switch settings */
+		option_switches = send_isa_command(CMD_ISA_SWITCH_SETTINGS);
+	}
+
+	unset_command_mode();
+	return(option_switches);
+}
+
+static void pcwd_show_card_info(void)
+{
+	char *firmware;
+	int option_switches;
+
+	/* Get some extra info from the hardware (in command/debug/diag mode) */
+	if (pcwd_private.revision == PCWD_REVISION_A)
+		printk(KERN_INFO PFX "ISA-PC Watchdog (REV.A) detected at port 0x%04x\n", pcwd_private.io_addr);
+	else if (pcwd_private.revision == PCWD_REVISION_C) {
+		firmware = get_firmware();
+		printk(KERN_INFO PFX "ISA-PC Watchdog (REV.C) detected at port 0x%04x (Firmware version: %s)\n",
+			pcwd_private.io_addr, firmware);
+		kfree(firmware);
+		option_switches = pcwd_get_option_switches();
+		printk(KERN_INFO PFX "Option switches (0x%02x): Temperature Reset Enable=%s, Power On Delay=%s\n",
+			option_switches,
+			((option_switches & 0x10) ? "ON" : "OFF"),
+			((option_switches & 0x08) ? "ON" : "OFF"));
+
+		/* Reprogram internal heartbeat to 2 seconds */
+		if (set_command_mode()) {
+			send_isa_command(CMD_ISA_DELAY_TIME_2SECS);
+			unset_command_mode();
+		}
+	}
+
+	if (pcwd_private.supports_temp)
+		printk(KERN_INFO PFX "Temperature Option Detected\n");
+
+	if (pcwd_private.boot_status & WDIOF_CARDRESET)
+		printk(KERN_INFO PFX "Previous reboot was caused by the card\n");
+
+	if (pcwd_private.boot_status & WDIOF_OVERHEAT) {
+		printk(KERN_EMERG PFX "Card senses a CPU Overheat. Panicking!\n");
+		printk(KERN_EMERG PFX "CPU Overheat\n");
+	}
+
+	if (pcwd_private.boot_status == 0)
+		printk(KERN_INFO PFX "No previous trip detected - Cold boot or reset\n");
+}
+
 static void pcwd_timer_ping(unsigned long data)
 {
 	int wdrst_stat;
@@ -644,47 +721,9 @@ static inline int get_revision(void)
 	return r;
 }
 
-static inline char *get_firmware(void)
-{
-	int one, ten, hund, minor;
-	char *ret;
-
-	ret = kmalloc(6, GFP_KERNEL);
-	if(ret == NULL)
-		return NULL;
-
-	if (set_command_mode()) {
-		one = send_isa_command(CMD_ISA_VERSION_INTEGER);
-		ten = send_isa_command(CMD_ISA_VERSION_TENTH);
-		hund = send_isa_command(CMD_ISA_VERSION_HUNDRETH);
-		minor = send_isa_command(CMD_ISA_VERSION_MINOR);
-		sprintf(ret, "%c.%c%c%c", one, ten, hund, minor);
-	}
-	else
-		sprintf(ret, "ERROR");
-
-	unset_command_mode();
-	return(ret);
-}
-
-static inline int get_option_switches(void)
-{
-	int rv=0;
-
-	if (set_command_mode()) {
-		/* Get switch settings */
-		rv = send_isa_command(CMD_ISA_SWITCH_SETTINGS);
-	}
-
-	unset_command_mode();
-	return(rv);
-}
-
 static int __devinit pcwatchdog_init(int base_addr)
 {
 	int ret;
-	char *firmware;
-	int option_switches;
 
 	cards_found++;
 	if (cards_found == 1)
@@ -732,46 +771,8 @@ static int __devinit pcwatchdog_init(int base_addr)
 	/*  Check whether or not the card supports the temperature device */
 	pcwd_check_temperature_support();
 
-	/* Get some extra info from the hardware (in command/debug/diag mode) */
-	if (pcwd_private.revision == PCWD_REVISION_A)
-		printk(KERN_INFO PFX "ISA-PC Watchdog (REV.A) detected at port 0x%04x\n", pcwd_private.io_addr);
-	else if (pcwd_private.revision == PCWD_REVISION_C) {
-		firmware = get_firmware();
-		printk(KERN_INFO PFX "ISA-PC Watchdog (REV.C) detected at port 0x%04x (Firmware version: %s)\n",
-			pcwd_private.io_addr, firmware);
-		kfree(firmware);
-		option_switches = get_option_switches();
-		printk(KERN_INFO PFX "Option switches (0x%02x): Temperature Reset Enable=%s, Power On Delay=%s\n",
-			option_switches,
-			((option_switches & 0x10) ? "ON" : "OFF"),
-			((option_switches & 0x08) ? "ON" : "OFF"));
-
-		/* Reprogram internal heartbeat to 2 seconds */
-		if (set_command_mode()) {
-			send_isa_command(CMD_ISA_DELAY_TIME_2SECS);
-			unset_command_mode();
-		}
-	} else {
-		/* Should NEVER happen, unless get_revision() fails. */
-		printk(KERN_INFO PFX "Unable to get revision\n");
-		release_region(pcwd_private.io_addr, (pcwd_private.revision == PCWD_REVISION_A) ? 2 : 4);
-		pcwd_private.io_addr = 0x0000;
-		return -1;
-	}
-
-	if (pcwd_private.supports_temp)
-		printk(KERN_INFO PFX "Temperature Option Detected\n");
-
-	if (pcwd_private.boot_status & WDIOF_CARDRESET)
-		printk(KERN_INFO PFX "Previous reboot was caused by the card\n");
-
-	if (pcwd_private.boot_status & WDIOF_OVERHEAT) {
-		printk(KERN_EMERG PFX "Card senses a CPU Overheat. Panicking!\n");
-		printk(KERN_EMERG PFX "CPU Overheat\n");
-	}
-
-	if (pcwd_private.boot_status == 0)
-		printk(KERN_INFO PFX "No previous trip detected - Cold boot or reset\n");
+	/* Show info about the card itself */
+	pcwd_show_card_info();
 
 	/* Check that the heartbeat value is within it's range ; if not reset to the default */
 	if (pcwd_set_heartbeat(heartbeat)) {

From a7122f916978a6cd58b765949cb315aabcddf151 Mon Sep 17 00:00:00 2001
From: Wim Van Sebroeck <wim@iguana.be>
Date: Mon, 9 Jan 2006 22:07:22 +0100
Subject: [PATCH 102/509] [WATCHDOG] pcwd.c - update module version info

Update the module version defines.

Signed-off-by: Wim Van Sebroeck <wim@iguana.be>
---
 drivers/char/watchdog/pcwd.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/char/watchdog/pcwd.c b/drivers/char/watchdog/pcwd.c
index f8f80d52d4de..8d6b249ad66b 100644
--- a/drivers/char/watchdog/pcwd.c
+++ b/drivers/char/watchdog/pcwd.c
@@ -73,8 +73,13 @@
 #include <asm/io.h>		/* For inb/outb/... */
 
 /* Module and version information */
-#define WD_VER                  "1.16 (03/01/2006)"
-#define PFX			"pcwd: "
+#define WATCHDOG_VERSION "1.16"
+#define WATCHDOG_DATE "03 Jan 2006"
+#define WATCHDOG_DRIVER_NAME "ISA-PC Watchdog"
+#define WATCHDOG_NAME "pcwd"
+#define PFX WATCHDOG_NAME ": "
+#define DRIVER_VERSION WATCHDOG_DRIVER_NAME " driver, v" WATCHDOG_VERSION " (" WATCHDOG_DATE ")\n"
+#define WD_VER WATCHDOG_VERSION " (" WATCHDOG_DATE ")"
 
 /*
  * It should be noted that PCWD_REVISION_B was removed because A and B

From 4733804c9f62fbc17ba69e8654a5fdf465f5bc41 Mon Sep 17 00:00:00 2001
From: Brian King <brking@us.ibm.com>
Date: Wed, 8 Feb 2006 20:57:42 -0600
Subject: [PATCH 103/509] [SCSI] ipr: Fix adapter initialization failure

Since scsi core is always sending scatterlists now, remove
some code which was written with the bad assumption that
a small transfer would not be sent down in a scatterlist.
Without this fix, the ipr driver ends up sending garbage
data to the adapter following a reset, causing it to
fail the reset and take the adapter offline.

Signed-off-by: Brian King <brking@us.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/ipr.c | 49 +++++-----------------------------------------
 drivers/scsi/ipr.h |  5 ++---
 2 files changed, 7 insertions(+), 47 deletions(-)

diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 27acf78cf8d8..2bba5e55d7bc 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -4235,35 +4235,6 @@ static void ipr_scsi_done(struct ipr_cmnd *ipr_cmd)
 		ipr_erp_start(ioa_cfg, ipr_cmd);
 }
 
-/**
- * ipr_save_ioafp_mode_select - Save adapters mode select data
- * @ioa_cfg:	ioa config struct
- * @scsi_cmd:	scsi command struct
- *
- * This function saves mode select data for the adapter to
- * use following an adapter reset.
- *
- * Return value:
- *	0 on success / SCSI_MLQUEUE_HOST_BUSY on failure
- **/
-static int ipr_save_ioafp_mode_select(struct ipr_ioa_cfg *ioa_cfg,
-				       struct scsi_cmnd *scsi_cmd)
-{
-	if (!ioa_cfg->saved_mode_pages) {
-		ioa_cfg->saved_mode_pages  = kmalloc(sizeof(struct ipr_mode_pages),
-						     GFP_ATOMIC);
-		if (!ioa_cfg->saved_mode_pages) {
-			dev_err(&ioa_cfg->pdev->dev,
-				"IOA mode select buffer allocation failed\n");
-			return SCSI_MLQUEUE_HOST_BUSY;
-		}
-	}
-
-	memcpy(ioa_cfg->saved_mode_pages, scsi_cmd->buffer, scsi_cmd->cmnd[4]);
-	ioa_cfg->saved_mode_page_len = scsi_cmd->cmnd[4];
-	return 0;
-}
-
 /**
  * ipr_queuecommand - Queue a mid-layer request
  * @scsi_cmd:	scsi command struct
@@ -4338,9 +4309,6 @@ static int ipr_queuecommand(struct scsi_cmnd *scsi_cmd,
 	    (!ipr_is_gscsi(res) || scsi_cmd->cmnd[0] == IPR_QUERY_RSRC_STATE))
 		ioarcb->cmd_pkt.request_type = IPR_RQTYPE_IOACMD;
 
-	if (ipr_is_ioa_resource(res) && scsi_cmd->cmnd[0] == MODE_SELECT)
-		rc = ipr_save_ioafp_mode_select(ioa_cfg, scsi_cmd);
-
 	if (likely(rc == 0))
 		rc = ipr_build_ioadl(ioa_cfg, ipr_cmd);
 
@@ -4829,17 +4797,11 @@ static int ipr_ioafp_mode_select_page28(struct ipr_cmnd *ipr_cmd)
 	int length;
 
 	ENTER;
-	if (ioa_cfg->saved_mode_pages) {
-		memcpy(mode_pages, ioa_cfg->saved_mode_pages,
-		       ioa_cfg->saved_mode_page_len);
-		length = ioa_cfg->saved_mode_page_len;
-	} else {
-		ipr_scsi_bus_speed_limit(ioa_cfg);
-		ipr_check_term_power(ioa_cfg, mode_pages);
-		ipr_modify_ioafp_mode_page_28(ioa_cfg, mode_pages);
-		length = mode_pages->hdr.length + 1;
-		mode_pages->hdr.length = 0;
-	}
+	ipr_scsi_bus_speed_limit(ioa_cfg);
+	ipr_check_term_power(ioa_cfg, mode_pages);
+	ipr_modify_ioafp_mode_page_28(ioa_cfg, mode_pages);
+	length = mode_pages->hdr.length + 1;
+	mode_pages->hdr.length = 0;
 
 	ipr_build_mode_select(ipr_cmd, cpu_to_be32(IPR_IOA_RES_HANDLE), 0x11,
 			      ioa_cfg->vpd_cbs_dma + offsetof(struct ipr_misc_cbs, mode_pages),
@@ -5969,7 +5931,6 @@ static void ipr_free_mem(struct ipr_ioa_cfg *ioa_cfg)
 	}
 
 	ipr_free_dump(ioa_cfg);
-	kfree(ioa_cfg->saved_mode_pages);
 	kfree(ioa_cfg->trace);
 }
 
diff --git a/drivers/scsi/ipr.h b/drivers/scsi/ipr.h
index b639332131f1..fd360bfe56dd 100644
--- a/drivers/scsi/ipr.h
+++ b/drivers/scsi/ipr.h
@@ -36,8 +36,8 @@
 /*
  * Literals
  */
-#define IPR_DRIVER_VERSION "2.1.1"
-#define IPR_DRIVER_DATE "(November 15, 2005)"
+#define IPR_DRIVER_VERSION "2.1.2"
+#define IPR_DRIVER_DATE "(February 8, 2006)"
 
 /*
  * IPR_MAX_CMD_PER_LUN: This defines the maximum number of outstanding
@@ -1000,7 +1000,6 @@ struct ipr_ioa_cfg {
 	struct Scsi_Host *host;
 	struct pci_dev *pdev;
 	struct ipr_sglist *ucode_sglist;
-	struct ipr_mode_pages *saved_mode_pages;
 	u8 saved_mode_page_len;
 
 	struct work_struct work_q;

From 3542adcb354fea4fca792d36b91cb44d0da147e5 Mon Sep 17 00:00:00 2001
From: "Ju, Seokmann" <Seokmann.Ju@lsil.com>
Date: Thu, 9 Feb 2006 12:32:59 -0700
Subject: [PATCH 104/509] [SCSI] megaraid_legacy: kobject_register failure

Attached patch fixes problem that cause kobject_register failure
during loading.  Kobject_register would fail when there are more than
1 module with same module name.  This patch will change module name of
megaraid_legacy from 'megaraid' to 'megaraid_legacy'.

Signed-Off-by: Seokmann Ju <seokmann.ju@lsil.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/megaraid.c | 2 +-
 drivers/scsi/megaraid.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/megaraid.c b/drivers/scsi/megaraid.c
index d101a8a6f4e8..7144674bc8e6 100644
--- a/drivers/scsi/megaraid.c
+++ b/drivers/scsi/megaraid.c
@@ -5049,7 +5049,7 @@ static struct pci_device_id megaraid_pci_tbl[] = {
 MODULE_DEVICE_TABLE(pci, megaraid_pci_tbl);
 
 static struct pci_driver megaraid_pci_driver = {
-	.name		= "megaraid",
+	.name		= "megaraid_legacy",
 	.id_table	= megaraid_pci_tbl,
 	.probe		= megaraid_probe_one,
 	.remove		= __devexit_p(megaraid_remove_one),
diff --git a/drivers/scsi/megaraid.h b/drivers/scsi/megaraid.h
index 4b3e0d6e5afa..4b75fe619d9c 100644
--- a/drivers/scsi/megaraid.h
+++ b/drivers/scsi/megaraid.h
@@ -5,7 +5,7 @@
 #include <linux/mutex.h>
 
 #define MEGARAID_VERSION	\
-	"v2.00.3 (Release Date: Wed Feb 19 08:51:30 EST 2003)\n"
+	"v2.00.4 (Release Date: Thu Feb 9 08:51:30 EST 2006)\n"
 
 /*
  * Driver features - change the values to enable or disable features in the

From c8024eb549f0c701e6d1c46c32e997f06f05d76d Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <aherrman@de.ibm.com>
Date: Sat, 11 Feb 2006 01:40:11 +0100
Subject: [PATCH 105/509] [SCSI] zfcp: get rid of physical_wwpn and
 physical_s_id

Remove all remainders of obsolete zfcp adapter attributes physical_wwpn and
physical_s_id.

Signed-off-by: Andreas Herrmann <aherrman@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/s390/scsi/zfcp_def.h           | 2 --
 drivers/s390/scsi/zfcp_sysfs_adapter.c | 4 ----
 2 files changed, 6 deletions(-)

diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h
index e260d19fa717..7dac56c49ade 100644
--- a/drivers/s390/scsi/zfcp_def.h
+++ b/drivers/s390/scsi/zfcp_def.h
@@ -915,8 +915,6 @@ struct zfcp_adapter {
 	wwn_t			peer_wwnn;	   /* P2P peer WWNN */
 	wwn_t			peer_wwpn;	   /* P2P peer WWPN */
 	u32			peer_d_id;	   /* P2P peer D_ID */
-	wwn_t			physical_wwpn;     /* WWPN of physical port */
-	u32			physical_s_id;     /* local FC port ID */
 	struct ccw_device       *ccw_device;	   /* S/390 ccw device */
 	u8			fc_service_class;
 	u32			hydra_version;	   /* Hydra version */
diff --git a/drivers/s390/scsi/zfcp_sysfs_adapter.c b/drivers/s390/scsi/zfcp_sysfs_adapter.c
index dfc07370f412..b29ac25e07f3 100644
--- a/drivers/s390/scsi/zfcp_sysfs_adapter.c
+++ b/drivers/s390/scsi/zfcp_sysfs_adapter.c
@@ -55,8 +55,6 @@ ZFCP_DEFINE_ADAPTER_ATTR(status, "0x%08x\n", atomic_read(&adapter->status));
 ZFCP_DEFINE_ADAPTER_ATTR(peer_wwnn, "0x%016llx\n", adapter->peer_wwnn);
 ZFCP_DEFINE_ADAPTER_ATTR(peer_wwpn, "0x%016llx\n", adapter->peer_wwpn);
 ZFCP_DEFINE_ADAPTER_ATTR(peer_d_id, "0x%06x\n", adapter->peer_d_id);
-ZFCP_DEFINE_ADAPTER_ATTR(physical_wwpn, "0x%016llx\n", adapter->physical_wwpn);
-ZFCP_DEFINE_ADAPTER_ATTR(physical_s_id, "0x%06x\n", adapter->physical_s_id);
 ZFCP_DEFINE_ADAPTER_ATTR(card_version, "0x%04x\n", adapter->hydra_version);
 ZFCP_DEFINE_ADAPTER_ATTR(lic_version, "0x%08x\n", adapter->fsf_lic_version);
 ZFCP_DEFINE_ADAPTER_ATTR(hardware_version, "0x%08x\n",
@@ -241,8 +239,6 @@ static struct attribute *zfcp_adapter_attrs[] = {
 	&dev_attr_peer_wwnn.attr,
 	&dev_attr_peer_wwpn.attr,
 	&dev_attr_peer_d_id.attr,
-	&dev_attr_physical_wwpn.attr,
-	&dev_attr_physical_s_id.attr,
 	&dev_attr_card_version.attr,
 	&dev_attr_lic_version.attr,
 	&dev_attr_status.attr,

From 2f8f3ed5fc566700cf45d422f4cf1624bd123d93 Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <aherrman@de.ibm.com>
Date: Sat, 11 Feb 2006 01:41:50 +0100
Subject: [PATCH 106/509] [SCSI] zfcp: fix adapter erp when link is unplugged

Remove endless polling for replug of the local link. Just wait for
link up notification.

Signed-off-by: Andreas Herrmann <aherrman@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/s390/scsi/zfcp_def.h |  5 ---
 drivers/s390/scsi/zfcp_erp.c | 80 +++++++++++++-----------------------
 drivers/s390/scsi/zfcp_fsf.c | 74 ++++++++++++++++++---------------
 3 files changed, 69 insertions(+), 90 deletions(-)

diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h
index 7dac56c49ade..f031199c7414 100644
--- a/drivers/s390/scsi/zfcp_def.h
+++ b/drivers/s390/scsi/zfcp_def.h
@@ -152,11 +152,6 @@ typedef u32 scsi_lun_t;
 #define ZFCP_EXCHANGE_CONFIG_DATA_FIRST_SLEEP	100
 #define ZFCP_EXCHANGE_CONFIG_DATA_RETRIES	7
 
-/* Retry 5 times every 2 second, then every minute */
-#define ZFCP_EXCHANGE_PORT_DATA_SHORT_RETRIES	5
-#define ZFCP_EXCHANGE_PORT_DATA_SHORT_SLEEP	200
-#define ZFCP_EXCHANGE_PORT_DATA_LONG_SLEEP	6000
-
 /* timeout value for "default timer" for fsf requests */
 #define ZFCP_FSF_REQUEST_TIMEOUT (60*HZ);
 
diff --git a/drivers/s390/scsi/zfcp_erp.c b/drivers/s390/scsi/zfcp_erp.c
index da947e662031..8ed6fcb41653 100644
--- a/drivers/s390/scsi/zfcp_erp.c
+++ b/drivers/s390/scsi/zfcp_erp.c
@@ -2246,15 +2246,6 @@ zfcp_erp_adapter_strategy_open_fsf(struct zfcp_erp_action *erp_action)
 {
 	int retval;
 
-	if ((atomic_test_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED,
-			      &erp_action->adapter->status)) &&
-	    (erp_action->adapter->adapter_features &
-	     FSF_FEATURE_HBAAPI_MANAGEMENT)) {
-		zfcp_erp_adapter_strategy_open_fsf_xport(erp_action);
-		atomic_set(&erp_action->adapter->erp_counter, 0);
-		return ZFCP_ERP_FAILED;
-	}
-
 	retval = zfcp_erp_adapter_strategy_open_fsf_xconfig(erp_action);
 	if (retval == ZFCP_ERP_FAILED)
 		return ZFCP_ERP_FAILED;
@@ -2266,13 +2257,6 @@ zfcp_erp_adapter_strategy_open_fsf(struct zfcp_erp_action *erp_action)
 	return zfcp_erp_adapter_strategy_open_fsf_statusread(erp_action);
 }
 
-/*
- * function:	
- *
- * purpose:	
- *
- * returns:
- */
 static int
 zfcp_erp_adapter_strategy_open_fsf_xconfig(struct zfcp_erp_action *erp_action)
 {
@@ -2350,48 +2334,40 @@ static int
 zfcp_erp_adapter_strategy_open_fsf_xport(struct zfcp_erp_action *erp_action)
 {
 	int ret;
-	int retries;
-	int sleep;
-	struct zfcp_adapter *adapter = erp_action->adapter;
+	struct zfcp_adapter *adapter;
 
+	adapter = erp_action->adapter;
 	atomic_clear_mask(ZFCP_STATUS_ADAPTER_XPORT_OK, &adapter->status);
 
-	retries = 0;
-	do {
-		write_lock(&adapter->erp_lock);
-		zfcp_erp_action_to_running(erp_action);
-		write_unlock(&adapter->erp_lock);
-		zfcp_erp_timeout_init(erp_action);
-		ret = zfcp_fsf_exchange_port_data(erp_action, adapter, NULL);
-		if (ret == -EOPNOTSUPP) {
-			debug_text_event(adapter->erp_dbf, 3, "a_xport_notsupp");
-			return ZFCP_ERP_SUCCEEDED;
-		} else if (ret) {
-			debug_text_event(adapter->erp_dbf, 3, "a_xport_failed");
-			return ZFCP_ERP_FAILED;
-		}
-		debug_text_event(adapter->erp_dbf, 6, "a_xport_ok");
+	write_lock(&adapter->erp_lock);
+	zfcp_erp_action_to_running(erp_action);
+	write_unlock(&adapter->erp_lock);
 
-		down(&adapter->erp_ready_sem);
-		if (erp_action->status & ZFCP_STATUS_ERP_TIMEDOUT) {
-			ZFCP_LOG_INFO("error: exchange of port data "
-				      "for adapter %s timed out\n",
-				      zfcp_get_busid_by_adapter(adapter));
-			break;
-		}
-		if (!atomic_test_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED,
-				      &adapter->status))
-			break;
+	zfcp_erp_timeout_init(erp_action);
+	ret = zfcp_fsf_exchange_port_data(erp_action, adapter, NULL);
+	if (ret == -EOPNOTSUPP) {
+		debug_text_event(adapter->erp_dbf, 3, "a_xport_notsupp");
+		return ZFCP_ERP_SUCCEEDED;
+	} else if (ret) {
+		debug_text_event(adapter->erp_dbf, 3, "a_xport_failed");
+		return ZFCP_ERP_FAILED;
+	}
+	debug_text_event(adapter->erp_dbf, 6, "a_xport_ok");
 
-		if (retries < ZFCP_EXCHANGE_PORT_DATA_SHORT_RETRIES) {
-			sleep = ZFCP_EXCHANGE_PORT_DATA_SHORT_SLEEP;
-			retries++;
-		} else
-			sleep = ZFCP_EXCHANGE_PORT_DATA_LONG_SLEEP;
-		schedule_timeout(sleep);
-	} while (1);
+	ret = ZFCP_ERP_SUCCEEDED;
+	down(&adapter->erp_ready_sem);
+	if (erp_action->status & ZFCP_STATUS_ERP_TIMEDOUT) {
+		ZFCP_LOG_INFO("error: exchange port data timed out (adapter "
+			      "%s)\n", zfcp_get_busid_by_adapter(adapter));
+		ret = ZFCP_ERP_FAILED;
+	}
+	if (!atomic_test_mask(ZFCP_STATUS_ADAPTER_XPORT_OK, &adapter->status)) {
+		ZFCP_LOG_INFO("error: exchange port data failed (adapter "
+			      "%s\n", zfcp_get_busid_by_adapter(adapter));
+		ret = ZFCP_ERP_FAILED;
+	}
 
-	return ZFCP_ERP_SUCCEEDED;
+	return ret;
 }
 
 /*
diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c
index 9f0cb3d820c0..bd8cd4d46134 100644
--- a/drivers/s390/scsi/zfcp_fsf.c
+++ b/drivers/s390/scsi/zfcp_fsf.c
@@ -388,6 +388,7 @@ zfcp_fsf_protstatus_eval(struct zfcp_fsf_req *fsf_req)
 	case FSF_PROT_LINK_DOWN:
 		zfcp_fsf_link_down_info_eval(adapter,
 					     &prot_status_qual->link_down_info);
+		zfcp_erp_adapter_reopen(adapter, 0);
 		fsf_req->status |= ZFCP_STATUS_FSFREQ_ERROR;
 		break;
 
@@ -558,10 +559,8 @@ zfcp_fsf_link_down_info_eval(struct zfcp_adapter *adapter,
 
 	atomic_set_mask(ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED, &adapter->status);
 
-	if (link_down == NULL) {
-		zfcp_erp_adapter_reopen(adapter, 0);
-		return;
-	}
+	if (link_down == NULL)
+		goto out;
 
 	switch (link_down->error_code) {
 	case FSF_PSQ_LINK_NO_LIGHT:
@@ -643,16 +642,8 @@ zfcp_fsf_link_down_info_eval(struct zfcp_adapter *adapter,
 				link_down->explanation_code,
 				link_down->vendor_specific_code);
 
-	switch (link_down->error_code) {
-	case FSF_PSQ_LINK_NO_LIGHT:
-	case FSF_PSQ_LINK_WRAP_PLUG:
-	case FSF_PSQ_LINK_NO_FCP:
-	case FSF_PSQ_LINK_FIRMWARE_UPDATE:
-		zfcp_erp_adapter_reopen(adapter, 0);
-		break;
-	default:
-		zfcp_erp_adapter_failed(adapter);
-	}
+ out:
+	zfcp_erp_adapter_failed(adapter);
 }
 
 /*
@@ -2304,6 +2295,35 @@ zfcp_fsf_exchange_port_data(struct zfcp_erp_action *erp_action,
 	return retval;
 }
 
+/**
+ * zfcp_fsf_exchange_port_evaluate
+ * @fsf_req: fsf_req which belongs to xchg port data request
+ * @xchg_ok: specifies if xchg port data was incomplete or complete (0/1)
+ */
+static void
+zfcp_fsf_exchange_port_evaluate(struct zfcp_fsf_req *fsf_req, int xchg_ok)
+{
+	struct zfcp_adapter *adapter;
+	struct fsf_qtcb *qtcb;
+	struct fsf_qtcb_bottom_port *bottom, *data;
+	struct Scsi_Host *shost;
+
+	adapter = fsf_req->adapter;
+	qtcb = fsf_req->qtcb;
+	bottom = &qtcb->bottom.port;
+	shost = adapter->scsi_host;
+
+	data = (struct fsf_qtcb_bottom_port*) fsf_req->data;
+	if (data)
+		memcpy(data, bottom, sizeof(struct fsf_qtcb_bottom_port));
+
+	if (adapter->connection_features & FSF_FEATURE_NPIV_MODE)
+		fc_host_permanent_port_name(shost) = bottom->wwpn;
+	else
+		fc_host_permanent_port_name(shost) = fc_host_port_name(shost);
+	fc_host_maxframe_size(shost) = bottom->maximum_frame_size;
+	fc_host_supported_speeds(shost) = bottom->supported_speed;
+}
 
 /**
  * zfcp_fsf_exchange_port_data_handler - handler for exchange_port_data request
@@ -2312,38 +2332,26 @@ zfcp_fsf_exchange_port_data(struct zfcp_erp_action *erp_action,
 static void
 zfcp_fsf_exchange_port_data_handler(struct zfcp_fsf_req *fsf_req)
 {
-	struct zfcp_adapter *adapter = fsf_req->adapter;
-	struct Scsi_Host *shost = adapter->scsi_host;
-	struct fsf_qtcb *qtcb = fsf_req->qtcb;
-	struct fsf_qtcb_bottom_port *bottom, *data;
+	struct zfcp_adapter *adapter;
+	struct fsf_qtcb *qtcb;
+
+	adapter = fsf_req->adapter;
+	qtcb = fsf_req->qtcb;
 
 	if (fsf_req->status & ZFCP_STATUS_FSFREQ_ERROR)
 		return;
 
 	switch (qtcb->header.fsf_status) {
         case FSF_GOOD:
+		zfcp_fsf_exchange_port_evaluate(fsf_req, 1);
 		atomic_set_mask(ZFCP_STATUS_ADAPTER_XPORT_OK, &adapter->status);
-
-		bottom = &qtcb->bottom.port;
-		data = (struct fsf_qtcb_bottom_port*) fsf_req->data;
-		if (data)
-			memcpy(data, bottom, sizeof(struct fsf_qtcb_bottom_port));
-		if (adapter->connection_features & FSF_FEATURE_NPIV_MODE)
-			fc_host_permanent_port_name(shost) = bottom->wwpn;
-		else
-			fc_host_permanent_port_name(shost) =
-				fc_host_port_name(shost);
-		fc_host_maxframe_size(shost) = bottom->maximum_frame_size;
-		fc_host_supported_speeds(shost) = bottom->supported_speed;
 		break;
-
 	case FSF_EXCHANGE_CONFIG_DATA_INCOMPLETE:
+		zfcp_fsf_exchange_port_evaluate(fsf_req, 0);
 		atomic_set_mask(ZFCP_STATUS_ADAPTER_XPORT_OK, &adapter->status);
-
 		zfcp_fsf_link_down_info_eval(adapter,
 			&qtcb->header.fsf_status_qual.link_down_info);
                 break;
-
         default:
 		debug_text_event(adapter->erp_dbf, 0, "xchg-port-ng");
 		debug_event(adapter->erp_dbf, 0,

From ed829ad607a9c334cea490d3a8c0f874153fb42d Mon Sep 17 00:00:00 2001
From: Maxim Shchetynin <maxim@de.ibm.com>
Date: Sat, 11 Feb 2006 01:42:58 +0100
Subject: [PATCH 107/509] [SCSI] zfcp: fix logging during device reset

Avoid access to old fsf_requests if device reset is logged.

Signed-off-by: Maxim Shchetynin <maxim@de.ibm.com>
Signed-off-by: Andreas Herrmann <aherrman@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/s390/scsi/zfcp_dbf.c  | 76 ++++++++++++++---------------------
 drivers/s390/scsi/zfcp_def.h  |  6 +--
 drivers/s390/scsi/zfcp_ext.h  |  5 ++-
 drivers/s390/scsi/zfcp_fsf.c  |  6 +--
 drivers/s390/scsi/zfcp_scsi.c | 15 ++++---
 5 files changed, 47 insertions(+), 61 deletions(-)

diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c
index 4d7d47cf2394..a5f2ba9a8fdb 100644
--- a/drivers/s390/scsi/zfcp_dbf.c
+++ b/drivers/s390/scsi/zfcp_dbf.c
@@ -710,10 +710,9 @@ static inline void
 _zfcp_scsi_dbf_event_common(const char *tag, const char *tag2, int level,
 			    struct zfcp_adapter *adapter,
 			    struct scsi_cmnd *scsi_cmnd,
-			    struct zfcp_fsf_req *new_fsf_req)
+			    struct zfcp_fsf_req *fsf_req,
+			    struct zfcp_fsf_req *old_fsf_req)
 {
-	struct zfcp_fsf_req *fsf_req =
-	    (struct zfcp_fsf_req *)scsi_cmnd->host_scribble;
 	struct zfcp_scsi_dbf_record *rec = &adapter->scsi_dbf_buf;
 	struct zfcp_dbf_dump *dump = (struct zfcp_dbf_dump *)rec;
 	unsigned long flags;
@@ -727,19 +726,20 @@ _zfcp_scsi_dbf_event_common(const char *tag, const char *tag2, int level,
 		if (offset == 0) {
 			strncpy(rec->tag, tag, ZFCP_DBF_TAG_SIZE);
 			strncpy(rec->tag2, tag2, ZFCP_DBF_TAG_SIZE);
-			if (scsi_cmnd->device) {
-				rec->scsi_id = scsi_cmnd->device->id;
-				rec->scsi_lun = scsi_cmnd->device->lun;
+			if (scsi_cmnd != NULL) {
+				if (scsi_cmnd->device) {
+					rec->scsi_id = scsi_cmnd->device->id;
+					rec->scsi_lun = scsi_cmnd->device->lun;
+				}
+				rec->scsi_result = scsi_cmnd->result;
+				rec->scsi_cmnd = (unsigned long)scsi_cmnd;
+				rec->scsi_serial = scsi_cmnd->serial_number;
+				memcpy(rec->scsi_opcode, &scsi_cmnd->cmnd,
+					min((int)scsi_cmnd->cmd_len,
+						ZFCP_DBF_SCSI_OPCODE));
+				rec->scsi_retries = scsi_cmnd->retries;
+				rec->scsi_allowed = scsi_cmnd->allowed;
 			}
-			rec->scsi_result = scsi_cmnd->result;
-			rec->scsi_cmnd = (unsigned long)scsi_cmnd;
-			rec->scsi_serial = scsi_cmnd->serial_number;
-			memcpy(rec->scsi_opcode,
-			       &scsi_cmnd->cmnd,
-			       min((int)scsi_cmnd->cmd_len,
-				   ZFCP_DBF_SCSI_OPCODE));
-			rec->scsi_retries = scsi_cmnd->retries;
-			rec->scsi_allowed = scsi_cmnd->allowed;
 			if (fsf_req != NULL) {
 				fcp_rsp = (struct fcp_rsp_iu *)
 				    &(fsf_req->qtcb->bottom.io.fcp_rsp);
@@ -772,15 +772,8 @@ _zfcp_scsi_dbf_event_common(const char *tag, const char *tag2, int level,
 				rec->fsf_seqno = fsf_req->seq_no;
 				rec->fsf_issued = fsf_req->issued;
 			}
-			if (new_fsf_req != NULL) {
-				rec->type.new_fsf_req.fsf_reqid =
-				    (unsigned long)
-				    new_fsf_req;
-				rec->type.new_fsf_req.fsf_seqno =
-				    new_fsf_req->seq_no;
-				rec->type.new_fsf_req.fsf_issued =
-				    new_fsf_req->issued;
-			}
+			rec->type.old_fsf_reqid =
+				    (unsigned long) old_fsf_req;
 		} else {
 			strncpy(dump->tag, "dump", ZFCP_DBF_TAG_SIZE);
 			dump->total_size = buflen;
@@ -801,19 +794,21 @@ _zfcp_scsi_dbf_event_common(const char *tag, const char *tag2, int level,
 inline void
 zfcp_scsi_dbf_event_result(const char *tag, int level,
 			   struct zfcp_adapter *adapter,
-			   struct scsi_cmnd *scsi_cmnd)
+			   struct scsi_cmnd *scsi_cmnd,
+			   struct zfcp_fsf_req *fsf_req)
 {
-	_zfcp_scsi_dbf_event_common("rslt",
-				    tag, level, adapter, scsi_cmnd, NULL);
+	_zfcp_scsi_dbf_event_common("rslt", tag, level,
+			adapter, scsi_cmnd, fsf_req, NULL);
 }
 
 inline void
 zfcp_scsi_dbf_event_abort(const char *tag, struct zfcp_adapter *adapter,
 			  struct scsi_cmnd *scsi_cmnd,
-			  struct zfcp_fsf_req *new_fsf_req)
+			  struct zfcp_fsf_req *new_fsf_req,
+			  struct zfcp_fsf_req *old_fsf_req)
 {
-	_zfcp_scsi_dbf_event_common("abrt",
-				    tag, 1, adapter, scsi_cmnd, new_fsf_req);
+	_zfcp_scsi_dbf_event_common("abrt", tag, 1,
+			adapter, scsi_cmnd, new_fsf_req, old_fsf_req);
 }
 
 inline void
@@ -823,7 +818,7 @@ zfcp_scsi_dbf_event_devreset(const char *tag, u8 flag, struct zfcp_unit *unit,
 	struct zfcp_adapter *adapter = unit->port->adapter;
 
 	_zfcp_scsi_dbf_event_common(flag == FCP_TARGET_RESET ? "trst" : "lrst",
-				    tag, 1, adapter, scsi_cmnd, NULL);
+			tag, 1, adapter, scsi_cmnd, NULL, NULL);
 }
 
 static int
@@ -856,6 +851,10 @@ zfcp_scsi_dbf_view_format(debug_info_t * id, struct debug_view *view,
 			     rec->scsi_retries);
 	len += zfcp_dbf_view(out_buf + len, "scsi_allowed", "0x%02x",
 			     rec->scsi_allowed);
+	if (strncmp(rec->tag, "abrt", ZFCP_DBF_TAG_SIZE) == 0) {
+		len += zfcp_dbf_view(out_buf + len, "old_fsf_reqid", "0x%0Lx",
+				     rec->type.old_fsf_reqid);
+	}
 	len += zfcp_dbf_view(out_buf + len, "fsf_reqid", "0x%0Lx",
 			     rec->fsf_reqid);
 	len += zfcp_dbf_view(out_buf + len, "fsf_seqno", "0x%08x",
@@ -883,21 +882,6 @@ zfcp_scsi_dbf_view_format(debug_info_t * id, struct debug_view *view,
 				       min((int)rec->type.fcp.sns_info_len,
 					   ZFCP_DBF_SCSI_FCP_SNS_INFO), 0,
 				       rec->type.fcp.sns_info_len);
-	} else if (strncmp(rec->tag, "abrt", ZFCP_DBF_TAG_SIZE) == 0) {
-		len += zfcp_dbf_view(out_buf + len, "fsf_reqid_abort", "0x%0Lx",
-				     rec->type.new_fsf_req.fsf_reqid);
-		len += zfcp_dbf_view(out_buf + len, "fsf_seqno_abort", "0x%08x",
-				     rec->type.new_fsf_req.fsf_seqno);
-		len += zfcp_dbf_stck(out_buf + len, "fsf_issued",
-				     rec->type.new_fsf_req.fsf_issued);
-	} else if ((strncmp(rec->tag, "trst", ZFCP_DBF_TAG_SIZE) == 0) ||
-		   (strncmp(rec->tag, "lrst", ZFCP_DBF_TAG_SIZE) == 0)) {
-		len += zfcp_dbf_view(out_buf + len, "fsf_reqid_reset", "0x%0Lx",
-				     rec->type.new_fsf_req.fsf_reqid);
-		len += zfcp_dbf_view(out_buf + len, "fsf_seqno_reset", "0x%08x",
-				     rec->type.new_fsf_req.fsf_seqno);
-		len += zfcp_dbf_stck(out_buf + len, "fsf_issued",
-				     rec->type.new_fsf_req.fsf_issued);
 	}
 
 	len += sprintf(out_buf + len, "\n");
diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h
index f031199c7414..7f551d66f47f 100644
--- a/drivers/s390/scsi/zfcp_def.h
+++ b/drivers/s390/scsi/zfcp_def.h
@@ -424,11 +424,7 @@ struct zfcp_scsi_dbf_record {
 	u32 fsf_seqno;
 	u64 fsf_issued;
 	union {
-		struct {
-			u64 fsf_reqid;
-			u32 fsf_seqno;
-			u64 fsf_issued;
-		} new_fsf_req;
+		u64 old_fsf_reqid;
 		struct {
 			u8 rsp_validity;
 			u8 rsp_scsi_status;
diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h
index c1ba7cf1b496..700f5402a978 100644
--- a/drivers/s390/scsi/zfcp_ext.h
+++ b/drivers/s390/scsi/zfcp_ext.h
@@ -194,9 +194,10 @@ extern void zfcp_san_dbf_event_els_response(struct zfcp_fsf_req *);
 extern void zfcp_san_dbf_event_incoming_els(struct zfcp_fsf_req *);
 
 extern void zfcp_scsi_dbf_event_result(const char *, int, struct zfcp_adapter *,
-				       struct scsi_cmnd *);
+				       struct scsi_cmnd *,
+				       struct zfcp_fsf_req *);
 extern void zfcp_scsi_dbf_event_abort(const char *, struct zfcp_adapter *,
-				      struct scsi_cmnd *,
+				      struct scsi_cmnd *, struct zfcp_fsf_req *,
 				      struct zfcp_fsf_req *);
 extern void zfcp_scsi_dbf_event_devreset(const char *, u8, struct zfcp_unit *,
 					 struct scsi_cmnd *);
diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c
index bd8cd4d46134..662ec571d73b 100644
--- a/drivers/s390/scsi/zfcp_fsf.c
+++ b/drivers/s390/scsi/zfcp_fsf.c
@@ -4211,11 +4211,11 @@ zfcp_fsf_send_fcp_command_task_handler(struct zfcp_fsf_req *fsf_req)
 	ZFCP_LOG_DEBUG("scpnt->result =0x%x\n", scpnt->result);
 
 	if (scpnt->result != 0)
-		zfcp_scsi_dbf_event_result("erro", 3, fsf_req->adapter, scpnt);
+		zfcp_scsi_dbf_event_result("erro", 3, fsf_req->adapter, scpnt, fsf_req);
 	else if (scpnt->retries > 0)
-		zfcp_scsi_dbf_event_result("retr", 4, fsf_req->adapter, scpnt);
+		zfcp_scsi_dbf_event_result("retr", 4, fsf_req->adapter, scpnt, fsf_req);
 	else
-		zfcp_scsi_dbf_event_result("norm", 6, fsf_req->adapter, scpnt);
+		zfcp_scsi_dbf_event_result("norm", 6, fsf_req->adapter, scpnt, fsf_req);
 
 	/* cleanup pointer (need this especially for abort) */
 	scpnt->host_scribble = NULL;
diff --git a/drivers/s390/scsi/zfcp_scsi.c b/drivers/s390/scsi/zfcp_scsi.c
index e0803757c0fa..9f6b4d7a46f3 100644
--- a/drivers/s390/scsi/zfcp_scsi.c
+++ b/drivers/s390/scsi/zfcp_scsi.c
@@ -242,7 +242,7 @@ zfcp_scsi_command_fail(struct scsi_cmnd *scpnt, int result)
 	if ((scpnt->device != NULL) && (scpnt->device->host != NULL))
 		zfcp_scsi_dbf_event_result("fail", 4,
 			(struct zfcp_adapter*) scpnt->device->host->hostdata[0],
-			scpnt);
+			scpnt, NULL);
 	/* return directly */
 	scpnt->scsi_done(scpnt);
 }
@@ -446,7 +446,7 @@ zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt)
 	old_fsf_req = (struct zfcp_fsf_req *) scpnt->host_scribble;
 	if (!old_fsf_req) {
 		write_unlock_irqrestore(&adapter->abort_lock, flags);
-		zfcp_scsi_dbf_event_abort("lte1", adapter, scpnt, new_fsf_req);
+		zfcp_scsi_dbf_event_abort("lte1", adapter, scpnt, NULL, NULL);
 		retval = SUCCESS;
 		goto out;
 	}
@@ -460,6 +460,8 @@ zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt)
 						 adapter, unit, 0);
 	if (!new_fsf_req) {
 		ZFCP_LOG_INFO("error: initiation of Abort FCP Cmnd failed\n");
+		zfcp_scsi_dbf_event_abort("nres", adapter, scpnt, NULL,
+					  old_fsf_req);
 		retval = FAILED;
 		goto out;
 	}
@@ -470,13 +472,16 @@ zfcp_scsi_eh_abort_handler(struct scsi_cmnd *scpnt)
 
 	/* status should be valid since signals were not permitted */
 	if (new_fsf_req->status & ZFCP_STATUS_FSFREQ_ABORTSUCCEEDED) {
-		zfcp_scsi_dbf_event_abort("okay", adapter, scpnt, new_fsf_req);
+		zfcp_scsi_dbf_event_abort("okay", adapter, scpnt, new_fsf_req,
+					  NULL);
 		retval = SUCCESS;
 	} else if (new_fsf_req->status & ZFCP_STATUS_FSFREQ_ABORTNOTNEEDED) {
-		zfcp_scsi_dbf_event_abort("lte2", adapter, scpnt, new_fsf_req);
+		zfcp_scsi_dbf_event_abort("lte2", adapter, scpnt, new_fsf_req,
+					  NULL);
 		retval = SUCCESS;
 	} else {
-		zfcp_scsi_dbf_event_abort("fail", adapter, scpnt, new_fsf_req);
+		zfcp_scsi_dbf_event_abort("fail", adapter, scpnt, new_fsf_req,
+					  NULL);
 		retval = FAILED;
 	}
 	zfcp_fsf_req_free(new_fsf_req);

From 61c41823c50302ca6cd455c48a1395f944c61f8f Mon Sep 17 00:00:00 2001
From: Andreas Herrmann <aherrman@de.ibm.com>
Date: Sat, 11 Feb 2006 01:43:55 +0100
Subject: [PATCH 108/509] [SCSI] zfcp: fix: avoid race between
 fc_remote_port_add and scsi_add_device

Flush workqueue of a scsi host after a remote port for that host
is registered at the fc transport class. Otherwise immediate
registration of a scsi device on that host is racy.

Signed-off-by: Andreas Herrmann <aherrman@de.ibm.com>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/s390/scsi/zfcp_erp.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/s390/scsi/zfcp_erp.c b/drivers/s390/scsi/zfcp_erp.c
index 8ed6fcb41653..e3c4bdd29a60 100644
--- a/drivers/s390/scsi/zfcp_erp.c
+++ b/drivers/s390/scsi/zfcp_erp.c
@@ -3415,6 +3415,8 @@ zfcp_erp_action_cleanup(int action, struct zfcp_adapter *adapter,
 						"(adapter %s, wwpn=0x%016Lx)\n",
 						zfcp_get_busid_by_port(port),
 						port->wwpn);
+			else
+				scsi_flush_work(adapter->scsi_host);
 		}
 		zfcp_port_put(port);
 		break;

From e2230eac17486e2ee07091d54d898eb40bcd0fdd Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Sun, 12 Feb 2006 09:28:19 -0700
Subject: [PATCH 109/509] [SCSI] sym2: Mask off opcode from RBC

pm->sg.size is set from the Residual Byte Count register.  However,
the upper byte of the RBC is the opcode of the instruction that was
executing, so we need to mask it off.  This fixes some spurious rejects
of IGNORE WIDE RESIDUE messages.

Signed-off-by: Matthew Wilcox <matthew@wil.cx>
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/sym53c8xx_2/sym_hipd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/scsi/sym53c8xx_2/sym_hipd.c b/drivers/scsi/sym53c8xx_2/sym_hipd.c
index 8260f040d39c..f4854c33f48d 100644
--- a/drivers/scsi/sym53c8xx_2/sym_hipd.c
+++ b/drivers/scsi/sym53c8xx_2/sym_hipd.c
@@ -3588,7 +3588,7 @@ static int sym_evaluate_dp(struct sym_hcb *np, struct sym_ccb *cp, u32 scr, int
 
 	if (pm) {
 		dp_scr  = scr_to_cpu(pm->ret);
-		dp_ofs -= scr_to_cpu(pm->sg.size);
+		dp_ofs -= scr_to_cpu(pm->sg.size) & 0x00ffffff;
 	}
 
 	/*

From 19bf9cbf6b313ae79a0c7278ccaa9c72c86931bd Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sun, 12 Feb 2006 12:35:03 +0100
Subject: [PATCH 110/509] [PATCH] s390: fstatat64 support

Add fstatat64 support to s390 in order to follow changes with
commit cff2b760096d1e6feaa31948e7af4abbefe47822 .
Also fixes compilation for 31 bit.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/compat_linux.c   | 20 ++++++++++++++++++++
 arch/s390/kernel/compat_wrapper.S |  8 ++++----
 arch/s390/kernel/syscalls.S       |  2 +-
 include/asm-s390/unistd.h         |  4 +++-
 4 files changed, 28 insertions(+), 6 deletions(-)

diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c
index cc20f0e3a7d3..2d021626c1a6 100644
--- a/arch/s390/kernel/compat_linux.c
+++ b/arch/s390/kernel/compat_linux.c
@@ -905,6 +905,26 @@ asmlinkage long sys32_fstat64(unsigned long fd, struct stat64_emu31 __user * sta
 	return ret;
 }
 
+asmlinkage long sys32_fstatat(unsigned int dfd, char __user *filename,
+			      struct stat64_emu31 __user* statbuf, int flag)
+{
+	struct kstat stat;
+	int error = -EINVAL;
+
+	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
+		goto out;
+
+	if (flag & AT_SYMLINK_NOFOLLOW)
+		error = vfs_lstat_fd(dfd, filename, &stat);
+	else
+		error = vfs_stat_fd(dfd, filename, &stat);
+
+	if (!error)
+		error = cp_stat64(statbuf, &stat);
+out:
+	return error;
+}
+
 /*
  * Linux/i386 didn't use to be able to handle more than
  * 4 system call parameters, so these system calls used a memory
diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index 38a6ef5ec624..dd2d6c3e8df8 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -1523,13 +1523,13 @@ compat_sys_futimesat_wrapper:
 	llgtr	%r4,%r4			# struct timeval *
 	jg	compat_sys_futimesat
 
-	.globl compat_sys_newfstatat_wrapper
-compat_sys_newfstatat_wrapper:
+	.globl sys32_fstatat_wrapper
+sys32_fstatat_wrapper:
 	llgfr	%r2,%r2			# unsigned int
 	llgtr	%r3,%r3			# char *
-	llgtr	%r4,%r4			# struct stat *
+	llgtr	%r4,%r4			# struct stat64 *
 	lgfr	%r5,%r5			# int
-	jg	compat_sys_newfstatat
+	jg	sys32_fstatat
 
 	.globl sys_unlinkat_wrapper
 sys_unlinkat_wrapper:
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index e86a4debb160..84921fe8d266 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -301,7 +301,7 @@ SYSCALL(sys_mkdirat,sys_mkdirat,sys_mkdirat_wrapper)
 SYSCALL(sys_mknodat,sys_mknodat,sys_mknodat_wrapper)	/* 290 */
 SYSCALL(sys_fchownat,sys_fchownat,sys_fchownat_wrapper)
 SYSCALL(sys_futimesat,sys_futimesat,compat_sys_futimesat_wrapper)
-SYSCALL(sys_newfstatat,sys_newfstatat,compat_sys_newfstatat_wrapper)
+SYSCALL(sys_fstatat64,sys_newfstatat,sys32_fstatat_wrapper)
 SYSCALL(sys_unlinkat,sys_unlinkat,sys_unlinkat_wrapper)
 SYSCALL(sys_renameat,sys_renameat,sys_renameat_wrapper)	/* 295 */
 SYSCALL(sys_linkat,sys_linkat,sys_linkat_wrapper)
diff --git a/include/asm-s390/unistd.h b/include/asm-s390/unistd.h
index 0a2f6664c4cf..657d582e8149 100644
--- a/include/asm-s390/unistd.h
+++ b/include/asm-s390/unistd.h
@@ -285,7 +285,7 @@
 #define __NR_mknodat		290
 #define __NR_fchownat		291
 #define __NR_futimesat		292
-#define __NR_newfstatat		293
+#define __NR_fstatat64		293
 #define __NR_unlinkat		294
 #define __NR_renameat		295
 #define __NR_linkat		296
@@ -359,6 +359,7 @@
 #undef  __NR_fcntl64
 #undef  __NR_sendfile64
 #undef  __NR_fadvise64_64
+#undef  __NR_fstatat64
 
 #define __NR_select		142
 #define __NR_getrlimit		191	/* SuS compliant getrlimit */
@@ -381,6 +382,7 @@
 #define __NR_setgid  		214
 #define __NR_setfsuid  		215
 #define __NR_setfsgid  		216
+#define __NR_newfstatat		293
 
 #endif
 

From 3c791925da0e6108cda15e3c2c7bfaebcd9ab9cf Mon Sep 17 00:00:00 2001
From: Jesper Juhl <jesper.juhl@gmail.com>
Date: Sun, 12 Feb 2006 14:34:53 -0800
Subject: [PATCH 111/509] [PATCH] netfilter: fix build error due to missing
 has_bridge_parent macro

net/bridge/br_netfilter.c: In function `br_nf_post_routing':
net/bridge/br_netfilter.c:808: warning: implicit declaration of function `has_bridge_parent'

Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com>
Cc: Harald Welte <laforge@netfilter.org>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 net/bridge/br_netfilter.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index b5018166b0e5..c06cb0983530 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -805,7 +805,7 @@ static unsigned int br_nf_post_routing(unsigned int hook, struct sk_buff **pskb,
 print_error:
 	if (skb->dev != NULL) {
 		printk("[%s]", skb->dev->name);
-		if (has_bridge_parent(skb->dev))
+		if (bridge_parent(skb->dev))
 			printk("[%s]", bridge_parent(skb->dev)->name);
 	}
 	printk(" head:%p, raw:%p, data:%p\n", skb->head, skb->mac.raw,

From 89edc3d2b429136a0e25f40275fd82dc58f147fd Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Sun, 12 Feb 2006 14:34:55 -0800
Subject: [PATCH 112/509] [PATCH] reiserfs: disable automatic enabling of
 reiserfs inode attributes

Unfortunately, the reiserfs_attrs_cleared bit in the superblock flag can
lie.  File systems have been observed with the bit set, yet still contain
garbage in the stat data field, causing unpredictable results.

This patch backs out the enable-by-default behavior.

It eliminates the changes from: d50a5cd860ce721dbeac6a4f3c6e42abcde68cd8,
and ef5e5414e7a83eb9b4295bbaba5464410b11e030.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/super.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index ef5e5414e7a8..d63da756eb49 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -1124,8 +1124,6 @@ static void handle_attrs(struct super_block *s)
 					 "reiserfs: cannot support attributes until flag is set in super-block");
 			REISERFS_SB(s)->s_mount_opt &= ~(1 << REISERFS_ATTRS);
 		}
-	} else if (le32_to_cpu(rs->s_flags) & reiserfs_attrs_cleared) {
-		REISERFS_SB(s)->s_mount_opt |= (1 << REISERFS_ATTRS);
 	}
 }
 

From 8f6da52aeff1fd7272ff5082552a39c050565b57 Mon Sep 17 00:00:00 2001
From: Jesse Allen <the3dfxdude@gmail.com>
Date: Sun, 12 Feb 2006 14:34:56 -0800
Subject: [PATCH 113/509] [PATCH] orinoco: support smc2532w

The orinoco wireless driver can support the SMC 2532W-B PC Card, so add the
id for it.

Signed-off-by: Jesse Allen <the3dfxdude@gmail.com>
Cc: Pavel Roskin <proski@gnu.org>
Cc: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/net/wireless/orinoco_cs.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/wireless/orinoco_cs.c b/drivers/net/wireless/orinoco_cs.c
index 3c128b692bce..ec6f2a48895b 100644
--- a/drivers/net/wireless/orinoco_cs.c
+++ b/drivers/net/wireless/orinoco_cs.c
@@ -590,6 +590,7 @@ static struct pcmcia_device_id orinoco_cs_ids[] = {
 	PCMCIA_DEVICE_PROD_ID12("PROXIM", "LAN PC CARD HARMONY 80211B", 0xc6536a5e, 0x090c3cd9),
 	PCMCIA_DEVICE_PROD_ID12("PROXIM", "LAN PCI CARD HARMONY 80211B", 0xc6536a5e, 0x9f494e26),
 	PCMCIA_DEVICE_PROD_ID12("SAMSUNG", "11Mbps WLAN Card", 0x43d74cb4, 0x579bd91b),
+	PCMCIA_DEVICE_PROD_ID12("SMC", "SMC2532W-B EliteConnect Wireless Adapter", 0xc4f8b18b, 0x196bd757),
 	PCMCIA_DEVICE_PROD_ID12("SMC", "SMC2632W", 0xc4f8b18b, 0x474a1f2a),
 	PCMCIA_DEVICE_PROD_ID12("Symbol Technologies", "LA4111 Spectrum24 Wireless LAN PC Card", 0x3f02b4d6, 0x3663cb0e),
 	PCMCIA_DEVICE_PROD_ID123("The Linksys Group, Inc.", "Instant Wireless Network PC Card", "ISL37300P", 0xa5f472c2, 0x590eb502, 0xc9049a39),

From a65d17c9d27a85782cfe1bbc36c747ffa1f81814 Mon Sep 17 00:00:00 2001
From: John Blackwood <john.blackwood@ccur.com>
Date: Sun, 12 Feb 2006 14:34:58 -0800
Subject: [PATCH 114/509] [PATCH] arch/x86_64/kernel/traps.c PTRACE_SINGLESTEP
 oops

We found a problem with x86_64 kernels with preemption enabled, where
having multiple tasks doing ptrace singlesteps around the same time will
cause the system to 'oops'.  The problem seems that a task can get
preempted out of the do_debug() processing while it is running on the
DEBUG_STACK stack.  If another task on that same cpu then enters do_debug()
and uses the same per-cpu DEBUG_STACK stack, the previous preempted tasks's
stack contents can be corrupted, and the system will oops when the
preempted task is context switched back in again.

The typical oops looks like the following:

  Unable to handle kernel paging request at ffffffffffffffae RIP: <ffffffff805452a1>{thread_return+34}
  PGD 103027 PUD 102429067 PMD 0
  Oops: 0002 [1] PREEMPT SMP
  CPU 0
  Modules linked in:
  Pid: 3786, comm: ssdd Not tainted 2.6.15.2 #1
  RIP: 0010:[<ffffffff805452a1>] <ffffffff805452a1>{thread_return+34}
  RSP: 0018:ffffffff80824058  EFLAGS: 000136c2
  RAX: ffff81017e12cea0 RBX: 0000000000000000 RCX: 00000000c0000100
  RDX: 0000000000000000 RSI: ffff8100f7856e20 RDI: ffff81017e12cea0
  RBP: 0000000000000046 R08: ffff8100f68a6000 R09: 0000000000000000
  R10: 0000000000000000 R11: ffff81017e12cea0 R12: ffff81000c2d53e8
  R13: ffff81017f5b3be8 R14: ffff81000c0036e0 R15: 000001056cbfc899
  FS:  00002aaaaaad9b00(0000) GS:ffffffff80883800(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
  CR2: ffffffffffffffae CR3: 00000000f6fcf000 CR4: 00000000000006e0
  Process ssdd (pid: 3786, threadinfo ffff8100f68a6000, task ffff8100f7856e20)
  Stack: ffffffff808240d8 ffffffff8012a84a ffff8100055f6c00 0000000000000020
         0000000000000001 ffff81000c0036e0 ffffffff808240b8 0000000000000000
         0000000000000000 0000000000000000
  Call Trace: <#DB>
	<ffffffff8012a84a>{try_to_wake_up+985}
	<ffffffff8012c0d3>{kick_process+87}
        <ffffffff8013b262>{signal_wake_up+48}
	<ffffffff8013b5ce>{specific_send_sig_info+179}
        <ffffffff80546abc>{_spin_unlock_irqrestore+27}
	<ffffffff8013b67c>{force_sig_info+159}
        <ffffffff801103a0>{do_debug+289} <ffffffff80110278>{sync_regs+103}
        <ffffffff8010ed9a>{paranoid_userspace+35}
  Unable to handle kernel paging request at 00007fffffb7d000 RIP: <ffffffff8010f2e4>{show_trace+465}
  PGD f6f25067 PUD f6fcc067 PMD f6957067 PTE 0
  Oops: 0000 [2] PREEMPT SMP

This patch disables preemptions for the task upon entry to do_debug(), before
interrupts are reenabled, and then disables preemption before exiting
do_debug(), after disabling interrupts.  I've noticed that the task can be
preempted either at the end of an interrupt, or on the call to
force_sig_info() on the spin_unlock_irqrestore() processing.  It might be
better to attempt to code a fix in entry.S around the code that calls
do_debug().

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/traps.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c
index ee1b2da9e5e7..28d50dc540e8 100644
--- a/arch/x86_64/kernel/traps.c
+++ b/arch/x86_64/kernel/traps.c
@@ -90,6 +90,20 @@ static inline void conditional_sti(struct pt_regs *regs)
 		local_irq_enable();
 }
 
+static inline void preempt_conditional_sti(struct pt_regs *regs)
+{
+	preempt_disable();
+	if (regs->eflags & X86_EFLAGS_IF)
+		local_irq_enable();
+}
+
+static inline void preempt_conditional_cli(struct pt_regs *regs)
+{
+	if (regs->eflags & X86_EFLAGS_IF)
+		local_irq_disable();
+	preempt_enable_no_resched();
+}
+
 static int kstack_depth_to_print = 10;
 
 #ifdef CONFIG_KALLSYMS
@@ -693,7 +707,7 @@ asmlinkage void __kprobes do_debug(struct pt_regs * regs,
 						SIGTRAP) == NOTIFY_STOP)
 		return;
 
-	conditional_sti(regs);
+	preempt_conditional_sti(regs);
 
 	/* Mask out spurious debug traps due to lazy DR7 setting */
 	if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
@@ -738,11 +752,13 @@ asmlinkage void __kprobes do_debug(struct pt_regs * regs,
 
 clear_dr7:
 	set_debugreg(0UL, 7);
+	preempt_conditional_cli(regs);
 	return;
 
 clear_TF_reenable:
 	set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
 	regs->eflags &= ~TF_MASK;
+	preempt_conditional_cli(regs);
 }
 
 static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr)

From 0d541064e8f58858e11cd34d81b6e83617f6eb4a Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sun, 12 Feb 2006 14:34:59 -0800
Subject: [PATCH 115/509] [PATCH] x86_64: GART DMA merging fix

Don't touch the non DMA members in the sg list in dma_map_sg in the IOMMU

Some drivers (in particular ST) ran into problems because they reused the sg
lists after passing them to pci_map_sg().  The merging procedure in the K8
GART IOMMU corrupted the state.  This patch changes it to only touch the dma*
entries during merging, but not the other fields.  Approach suggested by Dave
Miller.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/pci-gart.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/arch/x86_64/kernel/pci-gart.c b/arch/x86_64/kernel/pci-gart.c
index 2fe23a6c361b..dd0718dc178b 100644
--- a/arch/x86_64/kernel/pci-gart.c
+++ b/arch/x86_64/kernel/pci-gart.c
@@ -310,7 +310,7 @@ void gart_unmap_sg(struct device *dev, struct scatterlist *sg, int nents, int di
 
 	for (i = 0; i < nents; i++) {
 		struct scatterlist *s = &sg[i];
-		if (!s->dma_length || !s->length)
+		if (!s->dma_length)
 			break;
 		dma_unmap_single(dev, s->dma_address, s->dma_length, dir);
 	}
@@ -364,7 +364,6 @@ static int __dma_map_cont(struct scatterlist *sg, int start, int stopat,
 		
 		BUG_ON(i > start && s->offset);
 		if (i == start) {
-			*sout = *s; 
 			sout->dma_address = iommu_bus_base;
 			sout->dma_address += iommu_page*PAGE_SIZE + s->offset;
 			sout->dma_length = s->length;
@@ -379,7 +378,7 @@ static int __dma_map_cont(struct scatterlist *sg, int start, int stopat,
 			SET_LEAK(iommu_page);
 			addr += PAGE_SIZE;
 			iommu_page++;
-	} 
+		}
 	} 
 	BUG_ON(iommu_page - iommu_start != pages);	
 	return 0;
@@ -391,7 +390,6 @@ static inline int dma_map_cont(struct scatterlist *sg, int start, int stopat,
 {
 	if (!need) { 
 		BUG_ON(stopat - start != 1);
-		*sout = sg[start]; 
 		sout->dma_length = sg[start].length; 
 		return 0;
 	} 

From e9bb4c9929a63b23dcc637fae312b36b038bdc61 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@g5.osdl.org>
Date: Sun, 12 Feb 2006 16:27:25 -0800
Subject: [PATCH 116/509] Linux v2.6.16-rc3

---
 Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Makefile b/Makefile
index a1158d1c051e..74d67b2c35d9 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 16
-EXTRAVERSION =-rc2
+EXTRAVERSION =-rc3
 NAME=Sliding Snow Leopard
 
 # *DOCUMENTATION*

From cf5e40221bc509e13e22dc83c77c0c115eab531f Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Sun, 12 Feb 2006 21:05:32 -0500
Subject: [PATCH 117/509] [AGPGART] Improve the error message shown when we
 detect a ServerWorks CNB20HE

This chipset is unsupported, and likely to remain that way.

Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/char/agp/sworks-agp.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/char/agp/sworks-agp.c b/drivers/char/agp/sworks-agp.c
index 268f78d926d3..efef9999f1cf 100644
--- a/drivers/char/agp/sworks-agp.c
+++ b/drivers/char/agp/sworks-agp.c
@@ -468,9 +468,7 @@ static int __devinit agp_serverworks_probe(struct pci_dev *pdev,
 
 	switch (pdev->device) {
 	case 0x0006:
-		/* ServerWorks CNB20HE
-		Fail silently.*/
-		printk (KERN_ERR PFX "Detected ServerWorks CNB20HE chipset: No AGP present.\n");
+		printk (KERN_ERR PFX "ServerWorks CNB20HE is unsupported due to lack of documentation.\n");
 		return -ENODEV;
 
 	case PCI_DEVICE_ID_SERVERWORKS_HE:

From 40ad7a6afc53217ad95b5ae2221e42d7655e057b Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Sun, 12 Feb 2006 23:30:11 -0800
Subject: [PATCH 118/509] [SPARC]: sys_newfstatat --> sys_fstatat64

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/kernel/systbls.S       |  2 +-
 arch/sparc64/kernel/sys_sparc32.c | 21 +++++++++++++++++++++
 arch/sparc64/kernel/systbls.S     |  4 ++--
 include/asm-sparc/unistd.h        |  2 +-
 include/asm-sparc64/unistd.h      |  2 +-
 5 files changed, 26 insertions(+), 5 deletions(-)

diff --git a/arch/sparc/kernel/systbls.S b/arch/sparc/kernel/systbls.S
index c0314705d73a..768de64b371f 100644
--- a/arch/sparc/kernel/systbls.S
+++ b/arch/sparc/kernel/systbls.S
@@ -76,7 +76,7 @@ sys_call_table:
 /*270*/	.long sys_io_submit, sys_io_cancel, sys_io_getevents, sys_mq_open, sys_mq_unlink
 /*275*/	.long sys_mq_timedsend, sys_mq_timedreceive, sys_mq_notify, sys_mq_getsetattr, sys_waitid
 /*280*/	.long sys_ni_syscall, sys_add_key, sys_request_key, sys_keyctl, sys_openat
-/*285*/	.long sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_newfstatat
+/*285*/	.long sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_fstatat64
 /*290*/	.long sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat
 /*295*/	.long sys_fchmodat, sys_faccessat, sys_pselect6, sys_ppoll, sys_unshare
 
diff --git a/arch/sparc64/kernel/sys_sparc32.c b/arch/sparc64/kernel/sys_sparc32.c
index 9264ccbaaafa..417727bd87ba 100644
--- a/arch/sparc64/kernel/sys_sparc32.c
+++ b/arch/sparc64/kernel/sys_sparc32.c
@@ -428,6 +428,27 @@ asmlinkage long compat_sys_fstat64(unsigned int fd,
 	return error;
 }
 
+asmlinkage long compat_sys_fstatat64(unsigned int dfd, char __user *filename,
+		struct compat_stat64 __user * statbuf, int flag)
+{
+	struct kstat stat;
+	int error = -EINVAL;
+
+	if ((flag & ~AT_SYMLINK_NOFOLLOW) != 0)
+		goto out;
+
+	if (flag & AT_SYMLINK_NOFOLLOW)
+		error = vfs_lstat_fd(dfd, filename, &stat);
+	else
+		error = vfs_stat_fd(dfd, filename, &stat);
+
+	if (!error)
+		error = cp_compat_stat64(&stat, statbuf);
+
+out:
+	return error;
+}
+
 asmlinkage long compat_sys_sysfs(int option, u32 arg1, u32 arg2)
 {
 	return sys_sysfs(option, arg1, arg2);
diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S
index a19168510be2..c3adb7ac167d 100644
--- a/arch/sparc64/kernel/systbls.S
+++ b/arch/sparc64/kernel/systbls.S
@@ -77,7 +77,7 @@ sys_call_table32:
 /*270*/	.word sys32_io_submit, sys_io_cancel, compat_sys_io_getevents, sys32_mq_open, sys_mq_unlink
 	.word compat_sys_mq_timedsend, compat_sys_mq_timedreceive, compat_sys_mq_notify, compat_sys_mq_getsetattr, compat_sys_waitid
 /*280*/	.word sys_ni_syscall, sys_add_key, sys_request_key, sys_keyctl, compat_sys_openat
-	.word sys_mkdirat, sys_mknodat, sys_fchownat, compat_sys_futimesat, compat_sys_newfstatat
+	.word sys_mkdirat, sys_mknodat, sys_fchownat, compat_sys_futimesat, compat_sys_fstatat64
 /*285*/	.word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat
 	.word sys_fchmodat, sys_faccessat, compat_sys_pselect6, compat_sys_ppoll, sys_unshare
 
@@ -146,7 +146,7 @@ sys_call_table:
 /*270*/	.word sys_io_submit, sys_io_cancel, sys_io_getevents, sys_mq_open, sys_mq_unlink
 	.word sys_mq_timedsend, sys_mq_timedreceive, sys_mq_notify, sys_mq_getsetattr, sys_waitid
 /*280*/	.word sys_nis_syscall, sys_add_key, sys_request_key, sys_keyctl, sys_openat
-	.word sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_newfstatat
+	.word sys_mkdirat, sys_mknodat, sys_fchownat, sys_futimesat, sys_fstatat64
 /*285*/	.word sys_unlinkat, sys_renameat, sys_linkat, sys_symlinkat, sys_readlinkat
 	.word sys_fchmodat, sys_faccessat, sys_pselect6, sys_ppoll, sys_unshare
 
diff --git a/include/asm-sparc/unistd.h b/include/asm-sparc/unistd.h
index 0615d601a7c6..64ec640a40ee 100644
--- a/include/asm-sparc/unistd.h
+++ b/include/asm-sparc/unistd.h
@@ -305,7 +305,7 @@
 #define __NR_mknodat		286
 #define __NR_fchownat		287
 #define __NR_futimesat		288
-#define __NR_newfstatat		289
+#define __NR_fstatat64		289
 #define __NR_unlinkat		290
 #define __NR_renameat		291
 #define __NR_linkat		292
diff --git a/include/asm-sparc64/unistd.h b/include/asm-sparc64/unistd.h
index c58ba8a096cf..a284986b1541 100644
--- a/include/asm-sparc64/unistd.h
+++ b/include/asm-sparc64/unistd.h
@@ -307,7 +307,7 @@
 #define __NR_mknodat		286
 #define __NR_fchownat		287
 #define __NR_futimesat		288
-#define __NR_newfstatat		289
+#define __NR_fstatat64		289
 #define __NR_unlinkat		290
 #define __NR_renameat		291
 #define __NR_linkat		292

From 56f3a40a5e7586043260669cc794e56fa58339e1 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 13 Feb 2006 11:39:57 +0100
Subject: [PATCH 119/509] [Bluetooth] Reduce L2CAP MTU for RFCOMM connections

This patch reduces the default L2CAP MTU for all RFCOMM connections
from 1024 to 1013 to improve the interoperability with some broken
RFCOMM implementations. To make this more flexible the L2CAP MTU
becomes also a module parameter and so it can changed at runtime.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 include/net/bluetooth/rfcomm.h |  2 +-
 net/bluetooth/rfcomm/core.c    | 13 +++++++++----
 2 files changed, 10 insertions(+), 5 deletions(-)

diff --git a/include/net/bluetooth/rfcomm.h b/include/net/bluetooth/rfcomm.h
index bbfac86734ec..89d743cfdfdf 100644
--- a/include/net/bluetooth/rfcomm.h
+++ b/include/net/bluetooth/rfcomm.h
@@ -33,7 +33,7 @@
 #define RFCOMM_DEFAULT_MTU	127
 #define RFCOMM_DEFAULT_CREDITS	7
 
-#define RFCOMM_MAX_L2CAP_MTU	1024
+#define RFCOMM_MAX_L2CAP_MTU	1013
 #define RFCOMM_MAX_CREDITS	40
 
 #define RFCOMM_SKB_HEAD_RESERVE	8
diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c
index 0d89d6434136..5b4253c61f62 100644
--- a/net/bluetooth/rfcomm/core.c
+++ b/net/bluetooth/rfcomm/core.c
@@ -46,13 +46,15 @@
 #include <net/bluetooth/l2cap.h>
 #include <net/bluetooth/rfcomm.h>
 
-#define VERSION "1.6"
-
 #ifndef CONFIG_BT_RFCOMM_DEBUG
 #undef  BT_DBG
 #define BT_DBG(D...)
 #endif
 
+#define VERSION "1.7"
+
+static unsigned int l2cap_mtu = RFCOMM_MAX_L2CAP_MTU;
+
 static struct task_struct *rfcomm_thread;
 
 static DECLARE_MUTEX(rfcomm_sem);
@@ -623,7 +625,7 @@ static struct rfcomm_session *rfcomm_session_create(bdaddr_t *src, bdaddr_t *dst
 	/* Set L2CAP options */
 	sk = sock->sk;
 	lock_sock(sk);
-	l2cap_pi(sk)->imtu = RFCOMM_MAX_L2CAP_MTU;
+	l2cap_pi(sk)->imtu = l2cap_mtu;
 	release_sock(sk);
 
 	s = rfcomm_session_add(sock, BT_BOUND);
@@ -1868,7 +1870,7 @@ static int rfcomm_add_listener(bdaddr_t *ba)
 	/* Set L2CAP options */
 	sk = sock->sk;
 	lock_sock(sk);
-	l2cap_pi(sk)->imtu = RFCOMM_MAX_L2CAP_MTU;
+	l2cap_pi(sk)->imtu = l2cap_mtu;
 	release_sock(sk);
 
 	/* Start listening on the socket */
@@ -2070,6 +2072,9 @@ static void __exit rfcomm_exit(void)
 module_init(rfcomm_init);
 module_exit(rfcomm_exit);
 
+module_param(l2cap_mtu, uint, 0644);
+MODULE_PARM_DESC(l2cap_mtu, "Default MTU for the L2CAP connection");
+
 MODULE_AUTHOR("Maxim Krasnyansky <maxk@qualcomm.com>, Marcel Holtmann <marcel@holtmann.org>");
 MODULE_DESCRIPTION("Bluetooth RFCOMM ver " VERSION);
 MODULE_VERSION(VERSION);

From 7b005bd34c895ebeefd1c62f90a329730b88946b Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 13 Feb 2006 11:40:03 +0100
Subject: [PATCH 120/509] [Bluetooth] Fix NULL pointer dereferences of the HCI
 socket

This patch fixes the two NULL pointer dereferences found by the sfuzz
tool from Ilja van Sprundel. The first one was a call of getsockname()
for an unbound socket and the second was calling accept() while this
operation isn't implemented for the HCI socket interface.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 net/bluetooth/hci_sock.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index bdb6458c6bd5..97bdec73d17e 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -143,13 +143,15 @@ void hci_send_to_sock(struct hci_dev *hdev, struct sk_buff *skb)
 static int hci_sock_release(struct socket *sock)
 {
 	struct sock *sk = sock->sk;
-	struct hci_dev *hdev = hci_pi(sk)->hdev;
+	struct hci_dev *hdev;
 
 	BT_DBG("sock %p sk %p", sock, sk);
 
 	if (!sk)
 		return 0;
 
+	hdev = hci_pi(sk)->hdev;
+
 	bt_sock_unlink(&hci_sk_list, sk);
 
 	if (hdev) {
@@ -311,14 +313,18 @@ static int hci_sock_getname(struct socket *sock, struct sockaddr *addr, int *add
 {
 	struct sockaddr_hci *haddr = (struct sockaddr_hci *) addr;
 	struct sock *sk = sock->sk;
+	struct hci_dev *hdev = hci_pi(sk)->hdev;
 
 	BT_DBG("sock %p sk %p", sock, sk);
 
+	if (!hdev)
+		return -EBADFD;
+
 	lock_sock(sk);
 
 	*addr_len = sizeof(*haddr);
 	haddr->hci_family = AF_BLUETOOTH;
-	haddr->hci_dev    = hci_pi(sk)->hdev->id;
+	haddr->hci_dev    = hdev->id;
 
 	release_sock(sk);
 	return 0;

From 9225806386e398eeba46958a7befa017bda73f58 Mon Sep 17 00:00:00 2001
From: Marcel Holtmann <marcel@holtmann.org>
Date: Mon, 13 Feb 2006 11:40:07 +0100
Subject: [PATCH 121/509] [Bluetooth] Fix firmware loading problem of BT3C
 driver

Before the PCMCIA subsystem was fully integrated into the device and
driver model, the BT3C driver had to workaround this when loading the
firmware. This workaround is broken and makes the driver oops when
loading the firmware. This patch removes this workaround and uses now
the provided device structure from the PCMCIA subsystem.

Signed-off-by: Marcel Holtmann <marcel@holtmann.org>
---
 drivers/bluetooth/bt3c_cs.c | 17 ++++-------------
 1 file changed, 4 insertions(+), 13 deletions(-)

diff --git a/drivers/bluetooth/bt3c_cs.c b/drivers/bluetooth/bt3c_cs.c
index e522d19ad886..7e21b1ff27c4 100644
--- a/drivers/bluetooth/bt3c_cs.c
+++ b/drivers/bluetooth/bt3c_cs.c
@@ -474,18 +474,6 @@ static int bt3c_hci_ioctl(struct hci_dev *hdev, unsigned int cmd, unsigned long
 /* ======================== Card services HCI interaction ======================== */
 
 
-static struct device *bt3c_device(void)
-{
-	static struct device dev = {
-		.bus_id = "pcmcia",
-	};
-	kobject_set_name(&dev.kobj, "bt3c");
-	kobject_init(&dev.kobj);
-
-	return &dev;
-}
-
-
 static int bt3c_load_firmware(bt3c_info_t *info, unsigned char *firmware, int count)
 {
 	char *ptr = (char *) firmware;
@@ -574,6 +562,7 @@ static int bt3c_open(bt3c_info_t *info)
 {
 	const struct firmware *firmware;
 	struct hci_dev *hdev;
+	client_handle_t handle;
 	int err;
 
 	spin_lock_init(&(info->lock));
@@ -605,8 +594,10 @@ static int bt3c_open(bt3c_info_t *info)
 
 	hdev->owner = THIS_MODULE;
 
+	handle = info->link.handle;
+
 	/* Load firmware */
-	err = request_firmware(&firmware, "BT3CPCC.bin", bt3c_device());
+	err = request_firmware(&firmware, "BT3CPCC.bin", &handle_to_dev(handle));
 	if (err < 0) {
 		BT_ERR("Firmware request failed");
 		goto error;

From 326a625748535c4cdb1c632b1dcb07030989a393 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Mon, 13 Feb 2006 18:07:30 +0900
Subject: [PATCH 122/509] [PATCH] MIPS 32bit machines need fstatat64 support.

As noted by Jan Dittmer <jdi@l4x.org>

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/mips/kernel/scall32-o32.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index d7c4a38ed5ae..d83e033dbc87 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -623,7 +623,7 @@ einval:	li	v0, -EINVAL
 	sys	sys_mknodat		4	/* 4290 */
 	sys	sys_fchownat		5
 	sys	sys_futimesat		3
-	sys	sys_newfstatat		4
+	sys	sys_fstatat64		4
 	sys	sys_unlinkat		3
 	sys	sys_renameat		4	/* 4295 */
 	sys	sys_linkat		4

From 90947ef26fa689a3252aa8282a01f60648e70fdb Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Mon, 13 Feb 2006 11:12:36 -0500
Subject: [PATCH 123/509] [PATCH] reiserfs: fix potential (unlikely) oops in
 reiserfs_get_acl

This fixes a potential oops if there is an error reported by
posix_acl_from_disk().  This is mostly theoretical due to the use of
magics and checksums in xattrs, but is still possible.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/reiserfs/xattr_acl.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c
index 43de3ba83332..ab8894c3b9e5 100644
--- a/fs/reiserfs/xattr_acl.c
+++ b/fs/reiserfs/xattr_acl.c
@@ -228,7 +228,8 @@ struct posix_acl *reiserfs_get_acl(struct inode *inode, int type)
 		acl = ERR_PTR(retval);
 	} else {
 		acl = posix_acl_from_disk(value, retval);
-		*p_acl = posix_acl_dup(acl);
+		if (!IS_ERR(acl))
+			*p_acl = posix_acl_dup(acl);
 	}
 
 	kfree(value);

From 75c0141ca2fdae7c332d8f17412fbe0939dd005f Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Mon, 13 Feb 2006 12:46:58 -0500
Subject: [PATCH 124/509] [PATCH] Remove "RV370 5B60 [Radeon X300 (PCIE)]" from
 DRI list

I get a machine check exception, triple fault, or NMI watchdog lockup
when DRI gets enabled on this card.

(And Mauro Tassinari <mtassinari@cmanet.it> reports hung kernels too in
http://lkml.org/lkml/2006/1/26/97)

[ Adrian Bunk also states that this is the only RV350 entry for an RV370
  in our lists, which implies that it's just buggy ]

Cc: Adrian Bunk <bunk@stusta.de>
Cc: Dave Jones <davej@redhat.com>
Cc: Mauro Tassinari <mtassinari@cmanet.it>
Cc: Dave Airlie <airlied@linux.ie>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/drm/drm_pciids.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/char/drm/drm_pciids.h b/drivers/char/drm/drm_pciids.h
index 8fd6357a48da..2c17e88a8847 100644
--- a/drivers/char/drm/drm_pciids.h
+++ b/drivers/char/drm/drm_pciids.h
@@ -85,7 +85,6 @@
 	{0x1002, 0x5969, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV100}, \
 	{0x1002, 0x596A, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV280}, \
 	{0x1002, 0x596B, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV280}, \
-	{0x1002, 0x5b60, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV350}, \
 	{0x1002, 0x5c61, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV280|CHIP_IS_MOBILITY}, \
 	{0x1002, 0x5c62, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV280}, \
 	{0x1002, 0x5c63, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RV280|CHIP_IS_MOBILITY}, \

From 7d2babc4874825027848db04d11784070da4456d Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Mon, 13 Feb 2006 12:19:44 -0800
Subject: [PATCH 125/509] IB/mthca: bump driver version and release date

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mthca/mthca_dev.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h
index 2a165fd06e57..e481037288d6 100644
--- a/drivers/infiniband/hw/mthca/mthca_dev.h
+++ b/drivers/infiniband/hw/mthca/mthca_dev.h
@@ -53,8 +53,8 @@
 
 #define DRV_NAME	"ib_mthca"
 #define PFX		DRV_NAME ": "
-#define DRV_VERSION	"0.06"
-#define DRV_RELDATE	"June 23, 2005"
+#define DRV_VERSION	"0.07"
+#define DRV_RELDATE	"February 13, 2006"
 
 enum {
 	MTHCA_FLAG_DDR_HIDDEN = 1 << 1,

From 7a11c4d0635d9f6995736390b8c3346fe6f63d57 Mon Sep 17 00:00:00 2001
From: Joe Perches <joe@perches.com>
Date: Mon, 13 Feb 2006 15:34:11 -0800
Subject: [PATCH 126/509] [IRDA]: Ratelimit messages.

From: Joe Perches <joe@perches.com>

Based upon a patch by Dave Jones.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/irda/irda.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/net/irda/irda.h b/include/net/irda/irda.h
index 05a840837fe7..1880e46ecc9b 100644
--- a/include/net/irda/irda.h
+++ b/include/net/irda/irda.h
@@ -82,9 +82,9 @@ do { if(!(expr)) { \
 #define IRDA_ASSERT_LABEL(label)
 #endif /* CONFIG_IRDA_DEBUG */
 
-#define IRDA_WARNING(args...) printk(KERN_WARNING args)
-#define IRDA_MESSAGE(args...) printk(KERN_INFO args)
-#define IRDA_ERROR(args...)   printk(KERN_ERR args)
+#define IRDA_WARNING(args...) do { if (net_ratelimit()) printk(KERN_WARNING args); } while (0)
+#define IRDA_MESSAGE(args...) do { if (net_ratelimit()) printk(KERN_INFO args); } while (0)
+#define IRDA_ERROR(args...)   do { if (net_ratelimit()) printk(KERN_ERR args); } while (0)
 
 /*
  *  Magic numbers used by Linux-IrDA. Random numbers which must be unique to 

From bf3883c12fece9189ab4f7bb6e2690451db1366e Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Mon, 13 Feb 2006 15:34:58 -0800
Subject: [PATCH 127/509] [ATM]: Ratelimit atmsvc failure messages

This seems to be trivial to trigger.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/atm/signaling.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/atm/signaling.c b/net/atm/signaling.c
index e7211a7f382c..93ad59a28ef5 100644
--- a/net/atm/signaling.c
+++ b/net/atm/signaling.c
@@ -56,7 +56,8 @@ static void sigd_put_skb(struct sk_buff *skb)
 	remove_wait_queue(&sigd_sleep,&wait);
 #else
 	if (!sigd) {
-		printk(KERN_WARNING "atmsvc: no signaling demon\n");
+		if (net_ratelimit())
+			printk(KERN_WARNING "atmsvc: no signaling demon\n");
 		kfree_skb(skb);
 		return;
 	}

From 77decfc716d460b3f7037bb19bd4eb12cd0dc996 Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Mon, 13 Feb 2006 15:36:21 -0800
Subject: [PATCH 128/509] [IPV4] ICMP: Invert default for invalid icmp msgs
 sysctl

isic can trigger these msgs to be spewed at a very high rate.
There's already a sysctl to turn them off. Given these messages
aren't useful for most people, this patch disables them by
default.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/icmp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 4d1c40972a4b..e7bbff4340bb 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -192,7 +192,7 @@ int sysctl_icmp_echo_ignore_all;
 int sysctl_icmp_echo_ignore_broadcasts = 1;
 
 /* Control parameter - ignore bogus broadcast responses? */
-int sysctl_icmp_ignore_bogus_error_responses;
+int sysctl_icmp_ignore_bogus_error_responses = 1;
 
 /*
  * 	Configurable global rate limit.

From 99e382afd297d91ab150ae46c28c4585f925818c Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Mon, 13 Feb 2006 15:38:42 -0800
Subject: [PATCH 129/509] [P8023]: Fix tainting of kernel.

Missing license tag.
I've assumed this is GPL.  (It could also use a MODULE_AUTHOR)

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/802/p8023.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/802/p8023.c b/net/802/p8023.c
index d23e906456eb..53cf05709283 100644
--- a/net/802/p8023.c
+++ b/net/802/p8023.c
@@ -59,3 +59,5 @@ void destroy_8023_client(struct datalink_proto *dl)
 
 EXPORT_SYMBOL(destroy_8023_client);
 EXPORT_SYMBOL(make_8023_client);
+
+MODULE_LICENSE("GPL");

From a6c1cd572642478528165ac44db4d2daae125a21 Mon Sep 17 00:00:00 2001
From: Harald Welte <laforge@netfilter.org>
Date: Mon, 13 Feb 2006 15:42:48 -0800
Subject: [PATCH 130/509] [NETFILTER] Fix Kconfig menu level for x_tables

The new x_tables related Kconfig options appear at the wrong menu level
without this patch.

Signed-off-by: Harald Welte <laforge@netfilter.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 99c0a0fa4a97..0e550127fa7e 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -102,8 +102,6 @@ config NF_CT_NETLINK
 	help
 	  This option enables support for a netlink-based userspace interface
 
-endmenu
-
 config NETFILTER_XTABLES
 	tristate "Netfilter Xtables support (required for ip_tables)"
 	help
@@ -361,3 +359,5 @@ config NETFILTER_XT_MATCH_TCPMSS
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+endmenu
+

From 178a3259f2508e786fb1bd6538365a167cee35c1 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Mon, 13 Feb 2006 15:43:58 -0800
Subject: [PATCH 131/509] [BRIDGE]: Better fix for netfilter missing symbol
 has_bridge_parent

Horms patch was the best of the three fixes. Dave, already applied
Harald's version, so this patch converts that to the better one.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_netfilter.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index c06cb0983530..6bb0c7eb1ef0 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -805,8 +805,8 @@ static unsigned int br_nf_post_routing(unsigned int hook, struct sk_buff **pskb,
 print_error:
 	if (skb->dev != NULL) {
 		printk("[%s]", skb->dev->name);
-		if (bridge_parent(skb->dev))
-			printk("[%s]", bridge_parent(skb->dev)->name);
+		if (realoutdev)
+			printk("[%s]", realoutdev->name);
 	}
 	printk(" head:%p, raw:%p, data:%p\n", skb->head, skb->mac.raw,
 					      skb->data);

From e200bd8065e4db6297cd8db071a9188cf9aa6b56 Mon Sep 17 00:00:00 2001
From: Jamal Hadi Salim <hadi@cyberus.ca>
Date: Mon, 13 Feb 2006 15:51:24 -0800
Subject: [PATCH 132/509] [NETLINK] genetlink: Fix bugs spotted by Andrew
 Morton.

- panic() doesn't return.

- Don't forget to unlock on genl_register_family() error path

- genl_rcv_msg() is called via pointer so there's no point in declaring it
  `inline'.

Notes:

genl_ctrl_event() ignores the genlmsg_multicast() return value.

lots of things ignore the genl_ctrl_event() return value.

Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netlink/genetlink.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/net/netlink/genetlink.c b/net/netlink/genetlink.c
index 4ae1538c54a9..43e72419c868 100644
--- a/net/netlink/genetlink.c
+++ b/net/netlink/genetlink.c
@@ -238,7 +238,7 @@ int genl_register_family(struct genl_family *family)
 					sizeof(struct nlattr *), GFP_KERNEL);
 		if (family->attrbuf == NULL) {
 			err = -ENOMEM;
-			goto errout;
+			goto errout_locked;
 		}
 	} else
 		family->attrbuf = NULL;
@@ -288,7 +288,7 @@ int genl_unregister_family(struct genl_family *family)
 	return -ENOENT;
 }
 
-static inline int genl_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
+static int genl_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 			       int *errp)
 {
 	struct genl_ops *ops;
@@ -375,7 +375,7 @@ static void genl_rcv(struct sock *sk, int len)
 	do {
 		if (genl_trylock())
 			return;
-		netlink_run_queue(sk, &qlen, &genl_rcv_msg);
+		netlink_run_queue(sk, &qlen, genl_rcv_msg);
 		genl_unlock();
 	} while (qlen && genl_sock && genl_sock->sk_receive_queue.qlen);
 }
@@ -549,10 +549,8 @@ static int __init genl_init(void)
 	netlink_set_nonroot(NETLINK_GENERIC, NL_NONROOT_RECV);
 	genl_sock = netlink_kernel_create(NETLINK_GENERIC, GENL_MAX_ID,
 					  genl_rcv, THIS_MODULE);
-	if (genl_sock == NULL) {
+	if (genl_sock == NULL)
 		panic("GENL: Cannot initialize generic netlink\n");
-		return -ENOMEM;
-	}
 
 	return 0;
 
@@ -560,7 +558,6 @@ errout_register:
 	genl_unregister_family(&genl_ctrl);
 errout:
 	panic("GENL: Cannot register controller: %d\n", err);
-	return err;
 }
 
 subsys_initcall(genl_init);

From 72fb630dad170084026bda6728b8d8f21ed48ff1 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Mon, 13 Feb 2006 15:53:41 -0800
Subject: [PATCH 133/509] [APPLETALK]: warning fix

drivers/net/appletalk/cops.c: In function `cops_load':
drivers/net/appletalk/cops.c:539: warning: assignment discards qualifiers from pointer target type
drivers/net/appletalk/cops.c:547: warning: assignment discards qualifiers from pointer target type

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/appletalk/cops.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/appletalk/cops.h b/drivers/net/appletalk/cops.h
index c68ba9c2ef46..fd2750b269c8 100644
--- a/drivers/net/appletalk/cops.h
+++ b/drivers/net/appletalk/cops.h
@@ -51,7 +51,7 @@
 struct ltfirmware
 {
         unsigned int length;
-        unsigned char * data;
+        const unsigned char *data;
 };
 
 #define DAYNA 1

From 6d3e85ecf22a5e3610df47b9c3fb2fc32cfd35bf Mon Sep 17 00:00:00 2001
From: Nicolas DICHTEL <nicolas.dichtel@6wind.com>
Date: Mon, 13 Feb 2006 15:56:13 -0800
Subject: [PATCH 134/509] [IPV6] Don't store dst_entry for RAW socket

Signed-off-by: Nicolas DICHTEL <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/raw.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 738376cf0c51..ae20a0ec9bd8 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -803,10 +803,7 @@ back_from_confirm:
 			err = rawv6_push_pending_frames(sk, &fl, rp);
 	}
 done:
-	ip6_dst_store(sk, dst,
-		      ipv6_addr_equal(&fl.fl6_dst, &np->daddr) ?
-		      &np->daddr : NULL);
-
+	dst_release(dst);
 	release_sock(sk);
 out:	
 	fl6_sock_release(flowlabel);

From 00de651d14baabc5c1d2f32c49d9a984d8891c8e Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 13 Feb 2006 16:01:27 -0800
Subject: [PATCH 135/509] [IPSEC]: Fix strange IPsec freeze.

Problem discovered and initial patch by Olaf Kirch:

	there's a problem with IPsec that has been bugging some of our users
	for the last couple of kernel revs. Every now and then, IPsec will
	freeze the machine completely. This is with openswan user land,
	and with kernels up to and including 2.6.16-rc2.

	I managed to debug this a little, and what happens is that we end
	up looping in xfrm_lookup, and never get out. With a bit of debug
	printks added, I can this happening:

		ip_route_output_flow calls xfrm_lookup

		xfrm_find_bundle returns NULL (apparently we're in the
			middle of negotiating a new SA or something)

		We therefore call xfrm_tmpl_resolve. This returns EAGAIN
			We go to sleep, waiting for a policy update.
			Then we loop back to the top

		Apparently, the dst_orig that was passed into xfrm_lookup
			has been dropped from the routing table (obsolete=2)
			This leads to the endless loop, because we now create
			a new bundle, check the new bundle and find it's stale
			(stale_bundle -> xfrm_bundle_ok -> dst_check() return 0)

	People have been testing with the patch below, which seems to fix the
	problem partially. They still see connection hangs however (things
	only clear up when they start a new ping or new ssh). So the patch
	is obvsiouly not sufficient, and something else seems to go wrong.

	I'm grateful for any hints you may have...

I suggest that we simply bail out always.  If the dst decides to die
on us later on, the packet will be dropped anyway.  So there is no
great urgency to retry here.  Once we have the proper resolution
queueing, we can then do the retry again.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Olaf Kirch <okir@suse.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/xfrm/xfrm_policy.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index dbf4620768d6..98ec53bd3ac7 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -889,7 +889,9 @@ restart:
 			xfrm_pol_put(policy);
 			if (dst)
 				dst_free(dst);
-			goto restart;
+
+			err = -EHOSTUNREACH;
+			goto error;
 		}
 		dst->next = policy->bundles;
 		policy->bundles = dst;

From b4d9eda028e8becbb5057b554e63eea12e496a88 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Mon, 13 Feb 2006 16:06:10 -0800
Subject: [PATCH 136/509] [NET]: Revert skb_copy_datagram_iovec() recursion
 elimination.

Revert the following changeset:

bc8dfcb93970ad7139c976356bfc99d7e251deaf

Recursive SKB frag lists are really possible and disallowing
them breaks things.

Noticed by: Jesse Brandeburg <jesse.brandeburg@intel.com>

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/datagram.c | 81 +++++++++++++++++++++++++++++----------------
 1 file changed, 53 insertions(+), 28 deletions(-)

diff --git a/net/core/datagram.c b/net/core/datagram.c
index f8d322e1ea92..b8ce6bf81188 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -247,49 +247,74 @@ EXPORT_SYMBOL(skb_kill_datagram);
 int skb_copy_datagram_iovec(const struct sk_buff *skb, int offset,
 			    struct iovec *to, int len)
 {
-	int i, err, fraglen, end = 0;
-	struct sk_buff *next = skb_shinfo(skb)->frag_list;
+	int start = skb_headlen(skb);
+	int i, copy = start - offset;
 
-	if (!len)
-		return 0;
+	/* Copy header. */
+	if (copy > 0) {
+		if (copy > len)
+			copy = len;
+		if (memcpy_toiovec(to, skb->data + offset, copy))
+			goto fault;
+		if ((len -= copy) == 0)
+			return 0;
+		offset += copy;
+	}
 
-next_skb:
-	fraglen = skb_headlen(skb);
-	i = -1;
+	/* Copy paged appendix. Hmm... why does this look so complicated? */
+	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
+		int end;
 
-	while (1) {
-		int start = end;
+		BUG_TRAP(start <= offset + len);
 
-		if ((end += fraglen) > offset) {
-			int copy = end - offset, o = offset - start;
+		end = start + skb_shinfo(skb)->frags[i].size;
+		if ((copy = end - offset) > 0) {
+			int err;
+			u8  *vaddr;
+			skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+			struct page *page = frag->page;
 
 			if (copy > len)
 				copy = len;
-			if (i == -1)
-				err = memcpy_toiovec(to, skb->data + o, copy);
-			else {
-				skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
-				struct page *page = frag->page;
-				void *p = kmap(page) + frag->page_offset + o;
-				err = memcpy_toiovec(to, p, copy);
-				kunmap(page);
-			}
+			vaddr = kmap(page);
+			err = memcpy_toiovec(to, vaddr + frag->page_offset +
+					     offset - start, copy);
+			kunmap(page);
 			if (err)
 				goto fault;
 			if (!(len -= copy))
 				return 0;
 			offset += copy;
 		}
-		if (++i >= skb_shinfo(skb)->nr_frags)
-			break;
-		fraglen = skb_shinfo(skb)->frags[i].size;
+		start = end;
 	}
-	if (next) {
-		skb = next;
-		BUG_ON(skb_shinfo(skb)->frag_list);
-		next = skb->next;
-		goto next_skb;
+
+	if (skb_shinfo(skb)->frag_list) {
+		struct sk_buff *list = skb_shinfo(skb)->frag_list;
+
+		for (; list; list = list->next) {
+			int end;
+
+			BUG_TRAP(start <= offset + len);
+
+			end = start + list->len;
+			if ((copy = end - offset) > 0) {
+				if (copy > len)
+					copy = len;
+				if (skb_copy_datagram_iovec(list,
+							    offset - start,
+							    to, copy))
+					goto fault;
+				if ((len -= copy) == 0)
+					return 0;
+				offset += copy;
+			}
+			start = end;
+		}
 	}
+	if (!len)
+		return 0;
+
 fault:
 	return -EFAULT;
 }

From 108dff7d2b721759b5e025ab18024967c1294792 Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@parisc-linux.org>
Date: Mon, 13 Feb 2006 22:44:22 -0500
Subject: [PATCH 137/509] [PATCH] sys_newfstatat -> sys_fstatat64

parisc defines ARCH_WANT_STAT64, so we want to use fstatat64. It does not
appear that it needs to be ENTRY_COMP, because struct stat64 is the same
on both 32-bit and 64-bit (unlike on other platforms which did define a
compat_sys_fstatat64.)

Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/parisc/kernel/syscall_table.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S
index 66224f710e85..71011eadb872 100644
--- a/arch/parisc/kernel/syscall_table.S
+++ b/arch/parisc/kernel/syscall_table.S
@@ -385,7 +385,7 @@
 	ENTRY_SAME(mknodat)
 	ENTRY_SAME(fchownat)
 	ENTRY_COMP(futimesat)
-	ENTRY_COMP(newfstatat)		/* 280 */
+	ENTRY_SAME(fstatat64)		/* 280 */
 	ENTRY_SAME(unlinkat)
 	ENTRY_SAME(renameat)
 	ENTRY_SAME(linkat)

From a38408cd8d6bc0e5d16e609d4b1fdf9ba2e099ce Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Thu, 9 Feb 2006 16:35:31 -0500
Subject: [PATCH 138/509] [PATCH] USB: fix up the usb early handoff logic for
 EHCI

Disable some dubious "early" USB handoff code that allegedly works around bugs
on some systems (we don't know which ones) but rudely breaks some others.

Also make the kernel warnings reporting BIOS handoff problems be more useful,
reporting the register whose value displays the trouble.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/pci-quirks.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/usb/host/pci-quirks.c b/drivers/usb/host/pci-quirks.c
index e9e5bc178cef..118288d94423 100644
--- a/drivers/usb/host/pci-quirks.c
+++ b/drivers/usb/host/pci-quirks.c
@@ -191,8 +191,9 @@ static void __devinit quirk_usb_handoff_ohci(struct pci_dev *pdev)
 		}
 		if (wait_time <= 0)
 			printk(KERN_WARNING "%s %s: BIOS handoff "
-					"failed (BIOS bug ?)\n",
-					pdev->dev.bus_id, "OHCI");
+					"failed (BIOS bug ?) %08x\n",
+					pdev->dev.bus_id, "OHCI",
+					readl(base + OHCI_CONTROL));
 
 		/* reset controller, preserving RWC */
 		writel(control & OHCI_CTRL_RWC, base + OHCI_CONTROL);
@@ -243,6 +244,12 @@ static void __devinit quirk_usb_disable_ehci(struct pci_dev *pdev)
 				pr_debug("%s %s: BIOS handoff\n",
 						pdev->dev.bus_id, "EHCI");
 
+#if 0
+/* aleksey_gorelov@phoenix.com reports that some systems need SMI forced on,
+ * but that seems dubious in general (the BIOS left it off intentionally)
+ * and is known to prevent some systems from booting.  so we won't do this
+ * unless maybe we can determine when we're on a system that needs SMI forced.
+ */
 				/* BIOS workaround (?): be sure the
 				 * pre-Linux code receives the SMI
 				 */
@@ -252,6 +259,7 @@ static void __devinit quirk_usb_disable_ehci(struct pci_dev *pdev)
 				pci_write_config_dword(pdev,
 						offset + EHCI_USBLEGCTLSTS,
 						val | EHCI_USBLEGCTLSTS_SOOE);
+#endif
 			}
 
 			/* always say Linux will own the hardware
@@ -274,8 +282,8 @@ static void __devinit quirk_usb_disable_ehci(struct pci_dev *pdev)
 				 * it down, and hope nothing goes too wrong
 				 */
 				printk(KERN_WARNING "%s %s: BIOS handoff "
-						"failed (BIOS bug ?)\n",
-					pdev->dev.bus_id, "EHCI");
+						"failed (BIOS bug ?) %08x\n",
+					pdev->dev.bus_id, "EHCI", cap);
 				pci_write_config_byte(pdev, offset + 2, 0);
 			}
 

From ba3e66e94b9fb8c2a370a90729e068314845549d Mon Sep 17 00:00:00 2001
From: Michael Hund <mhund@ld-didactic.de>
Date: Thu, 2 Feb 2006 09:36:43 +0100
Subject: [PATCH 139/509] [PATCH] USB: add new device ids to ldusb

Signed-off-by: Michael Hund <mhund@ld-didactic.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/input/hid-core.c | 50 +++++++++++++++++--------------
 drivers/usb/misc/ldusb.c     | 57 ++++++++++++++++++++----------------
 2 files changed, 60 insertions(+), 47 deletions(-)

diff --git a/drivers/usb/input/hid-core.c b/drivers/usb/input/hid-core.c
index 6f7a684c3e07..772478086bd3 100644
--- a/drivers/usb/input/hid-core.c
+++ b/drivers/usb/input/hid-core.c
@@ -1435,17 +1435,20 @@ void hid_init_reports(struct hid_device *hid)
 #define USB_DEVICE_ID_VERNIER_CYCLOPS	0x0004
 
 #define USB_VENDOR_ID_LD		0x0f11
-#define USB_DEVICE_ID_CASSY		0x1000
-#define USB_DEVICE_ID_POCKETCASSY	0x1010
-#define USB_DEVICE_ID_MOBILECASSY	0x1020
-#define USB_DEVICE_ID_JWM		0x1080
-#define USB_DEVICE_ID_DMMP		0x1081
-#define USB_DEVICE_ID_UMIP		0x1090
-#define USB_DEVICE_ID_VIDEOCOM		0x1200
-#define USB_DEVICE_ID_COM3LAB		0x2000
-#define USB_DEVICE_ID_TELEPORT		0x2010
-#define USB_DEVICE_ID_NETWORKANALYSER	0x2020
-#define USB_DEVICE_ID_POWERCONTROL	0x2030
+#define USB_DEVICE_ID_LD_CASSY		0x1000
+#define USB_DEVICE_ID_LD_POCKETCASSY	0x1010
+#define USB_DEVICE_ID_LD_MOBILECASSY	0x1020
+#define USB_DEVICE_ID_LD_JWM		0x1080
+#define USB_DEVICE_ID_LD_DMMP		0x1081
+#define USB_DEVICE_ID_LD_UMIP		0x1090
+#define USB_DEVICE_ID_LD_XRAY1		0x1100
+#define USB_DEVICE_ID_LD_XRAY2		0x1101
+#define USB_DEVICE_ID_LD_VIDEOCOM	0x1200
+#define USB_DEVICE_ID_LD_COM3LAB	0x2000
+#define USB_DEVICE_ID_LD_TELEPORT	0x2010
+#define USB_DEVICE_ID_LD_NETWORKANALYSER 0x2020
+#define USB_DEVICE_ID_LD_POWERCONTROL	0x2030
+#define USB_DEVICE_ID_LD_MACHINETEST	0x2040
 
 #define USB_VENDOR_ID_APPLE		0x05ac
 #define USB_DEVICE_ID_APPLE_POWERMOUSE	0x0304
@@ -1491,17 +1494,20 @@ static const struct hid_blacklist {
 	{ USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_POWERMATE, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_GRIFFIN, USB_DEVICE_ID_SOUNDKNOB, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_KBGEAR, USB_DEVICE_ID_KBGEAR_JAMSTUDIO, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_CASSY, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_POCKETCASSY, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_MOBILECASSY, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_JWM, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_DMMP, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_UMIP, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_VIDEOCOM, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_COM3LAB, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_TELEPORT, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_NETWORKANALYSER, HID_QUIRK_IGNORE },
-	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_POWERCONTROL, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_CASSY, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_POCKETCASSY, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_MOBILECASSY, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_JWM, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_DMMP, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_UMIP, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_XRAY1, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_XRAY2, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_VIDEOCOM, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_COM3LAB, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_TELEPORT, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_NETWORKANALYSER, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_POWERCONTROL, HID_QUIRK_IGNORE },
+	{ USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_MACHINETEST, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_MCC, USB_DEVICE_ID_MCC_PMD1024LS, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_MCC, USB_DEVICE_ID_MCC_PMD1208LS, HID_QUIRK_IGNORE },
 	{ USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS, HID_QUIRK_IGNORE },
diff --git a/drivers/usb/misc/ldusb.c b/drivers/usb/misc/ldusb.c
index 331d4ae949ed..e2d1198623eb 100644
--- a/drivers/usb/misc/ldusb.c
+++ b/drivers/usb/misc/ldusb.c
@@ -24,6 +24,7 @@
  * V0.1  (mh) Initial version
  * V0.11 (mh) Added raw support for HID 1.0 devices (no interrupt out endpoint)
  * V0.12 (mh) Added kmalloc check for string buffer
+ * V0.13 (mh) Added support for LD X-Ray and Machine Test System
  */
 
 #include <linux/config.h>
@@ -40,17 +41,20 @@
 
 /* Define these values to match your devices */
 #define USB_VENDOR_ID_LD		0x0f11	/* USB Vendor ID of LD Didactic GmbH */
-#define USB_DEVICE_ID_CASSY		0x1000	/* USB Product ID for all CASSY-S modules */
-#define USB_DEVICE_ID_POCKETCASSY	0x1010	/* USB Product ID for Pocket-CASSY */
-#define USB_DEVICE_ID_MOBILECASSY	0x1020	/* USB Product ID for Mobile-CASSY */
-#define USB_DEVICE_ID_JWM		0x1080	/* USB Product ID for Joule and Wattmeter */
-#define USB_DEVICE_ID_DMMP		0x1081	/* USB Product ID for Digital Multimeter P (reserved) */
-#define USB_DEVICE_ID_UMIP		0x1090	/* USB Product ID for UMI P */
-#define USB_DEVICE_ID_VIDEOCOM		0x1200	/* USB Product ID for VideoCom */
-#define USB_DEVICE_ID_COM3LAB		0x2000	/* USB Product ID for COM3LAB */
-#define USB_DEVICE_ID_TELEPORT		0x2010	/* USB Product ID for Terminal Adapter */
-#define USB_DEVICE_ID_NETWORKANALYSER	0x2020	/* USB Product ID for Network Analyser */
-#define USB_DEVICE_ID_POWERCONTROL	0x2030	/* USB Product ID for Controlling device for Power Electronics */
+#define USB_DEVICE_ID_LD_CASSY		0x1000	/* USB Product ID of CASSY-S */
+#define USB_DEVICE_ID_LD_POCKETCASSY	0x1010	/* USB Product ID of Pocket-CASSY */
+#define USB_DEVICE_ID_LD_MOBILECASSY	0x1020	/* USB Product ID of Mobile-CASSY */
+#define USB_DEVICE_ID_LD_JWM		0x1080	/* USB Product ID of Joule and Wattmeter */
+#define USB_DEVICE_ID_LD_DMMP		0x1081	/* USB Product ID of Digital Multimeter P (reserved) */
+#define USB_DEVICE_ID_LD_UMIP		0x1090	/* USB Product ID of UMI P */
+#define USB_DEVICE_ID_LD_XRAY1		0x1100	/* USB Product ID of X-Ray Apparatus */
+#define USB_DEVICE_ID_LD_XRAY2		0x1101	/* USB Product ID of X-Ray Apparatus */
+#define USB_DEVICE_ID_LD_VIDEOCOM	0x1200	/* USB Product ID of VideoCom */
+#define USB_DEVICE_ID_LD_COM3LAB	0x2000	/* USB Product ID of COM3LAB */
+#define USB_DEVICE_ID_LD_TELEPORT	0x2010	/* USB Product ID of Terminal Adapter */
+#define USB_DEVICE_ID_LD_NETWORKANALYSER 0x2020	/* USB Product ID of Network Analyser */
+#define USB_DEVICE_ID_LD_POWERCONTROL	0x2030	/* USB Product ID of Converter Control Unit */
+#define USB_DEVICE_ID_LD_MACHINETEST	0x2040	/* USB Product ID of Machine Test System */
 
 #define USB_VENDOR_ID_VERNIER		0x08f7
 #define USB_DEVICE_ID_VERNIER_LABPRO	0x0001
@@ -67,17 +71,20 @@
 
 /* table of devices that work with this driver */
 static struct usb_device_id ld_usb_table [] = {
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_CASSY) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_POCKETCASSY) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_MOBILECASSY) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_JWM) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_DMMP) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_UMIP) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_VIDEOCOM) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_COM3LAB) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_TELEPORT) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_NETWORKANALYSER) },
-	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_POWERCONTROL) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_CASSY) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_POCKETCASSY) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_MOBILECASSY) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_JWM) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_DMMP) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_UMIP) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_XRAY1) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_XRAY2) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_VIDEOCOM) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_COM3LAB) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_TELEPORT) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_NETWORKANALYSER) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_POWERCONTROL) },
+	{ USB_DEVICE(USB_VENDOR_ID_LD, USB_DEVICE_ID_LD_MACHINETEST) },
 	{ USB_DEVICE(USB_VENDOR_ID_VERNIER, USB_DEVICE_ID_VERNIER_LABPRO) },
 	{ USB_DEVICE(USB_VENDOR_ID_VERNIER, USB_DEVICE_ID_VERNIER_GOTEMP) },
 	{ USB_DEVICE(USB_VENDOR_ID_VERNIER, USB_DEVICE_ID_VERNIER_SKIP) },
@@ -85,7 +92,7 @@ static struct usb_device_id ld_usb_table [] = {
 	{ }					/* Terminating entry */
 };
 MODULE_DEVICE_TABLE(usb, ld_usb_table);
-MODULE_VERSION("V0.12");
+MODULE_VERSION("V0.13");
 MODULE_AUTHOR("Michael Hund <mhund@ld-didactic.de>");
 MODULE_DESCRIPTION("LD USB Driver");
 MODULE_LICENSE("GPL");
@@ -632,8 +639,8 @@ static int ld_usb_probe(struct usb_interface *intf, const struct usb_device_id *
 
 	/* workaround for early firmware versions on fast computers */
 	if ((le16_to_cpu(udev->descriptor.idVendor) == USB_VENDOR_ID_LD) &&
-	    ((le16_to_cpu(udev->descriptor.idProduct) == USB_DEVICE_ID_CASSY) ||
-	     (le16_to_cpu(udev->descriptor.idProduct) == USB_DEVICE_ID_COM3LAB)) &&
+	    ((le16_to_cpu(udev->descriptor.idProduct) == USB_DEVICE_ID_LD_CASSY) ||
+	     (le16_to_cpu(udev->descriptor.idProduct) == USB_DEVICE_ID_LD_COM3LAB)) &&
 	    (le16_to_cpu(udev->descriptor.bcdDevice) <= 0x103)) {
 		buffer = kmalloc(256, GFP_KERNEL);
 		if (buffer == NULL) {

From 343a65cadb3a6a102f08513d9c64eb7e317478f0 Mon Sep 17 00:00:00 2001
From: Michael Hund <mhund@ld-didactic.de>
Date: Thu, 2 Feb 2006 09:37:02 +0100
Subject: [PATCH 140/509] [PATCH] USB: change ldusb's experimental state

Signed-off-by: Michael Hund <mhund@ld-didactic.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/misc/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/usb/misc/Kconfig b/drivers/usb/misc/Kconfig
index 6649531fa824..8ba6a701e9c1 100644
--- a/drivers/usb/misc/Kconfig
+++ b/drivers/usb/misc/Kconfig
@@ -141,7 +141,7 @@ source "drivers/usb/misc/sisusbvga/Kconfig"
 
 config USB_LD
 	tristate "USB LD driver"
-	depends on USB && EXPERIMENTAL
+	depends on USB
 	help
 	  This driver is for generic USB devices that use interrupt transfers,
 	  like LD Didactic's USB devices.

From c6c27721a42b991965bb792d5c196b8331d008d5 Mon Sep 17 00:00:00 2001
From: Christian Lindner <christian.lindner@gmx.net>
Date: Wed, 1 Feb 2006 14:10:52 +0100
Subject: [PATCH 141/509] [PATCH] USB: PL2303: Leadtek 9531 GPS-Mouse

The patch adds the USB ID (0413:2101) for the Leadtek GPS-Mouse 9531 to
the driver pl2303.

Signed-off-by: Christian Lindner <christian.lindner@gmx.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/serial/pl2303.c | 5 +++--
 drivers/usb/serial/pl2303.h | 4 ++++
 2 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/serial/pl2303.c b/drivers/usb/serial/pl2303.c
index e8e575e037c1..37c81c08faad 100644
--- a/drivers/usb/serial/pl2303.c
+++ b/drivers/usb/serial/pl2303.c
@@ -73,9 +73,10 @@ static struct usb_device_id id_table [] = {
 	{ USB_DEVICE(SIEMENS_VENDOR_ID, SIEMENS_PRODUCT_ID_X65) },
 	{ USB_DEVICE(SIEMENS_VENDOR_ID, SIEMENS_PRODUCT_ID_X75) },
 	{ USB_DEVICE(SYNTECH_VENDOR_ID, SYNTECH_PRODUCT_ID) },
-	{ USB_DEVICE(NOKIA_CA42_VENDOR_ID, NOKIA_CA42_PRODUCT_ID ) },
-	{ USB_DEVICE(CA_42_CA42_VENDOR_ID, CA_42_CA42_PRODUCT_ID ) },
+	{ USB_DEVICE(NOKIA_CA42_VENDOR_ID, NOKIA_CA42_PRODUCT_ID) },
+	{ USB_DEVICE(CA_42_CA42_VENDOR_ID, CA_42_CA42_PRODUCT_ID) },
 	{ USB_DEVICE(SAGEM_VENDOR_ID, SAGEM_PRODUCT_ID) },
+	{ USB_DEVICE(LEADTEK_VENDOR_ID, LEADTEK_9531_PRODUCT_ID) },
 	{ }					/* Terminating entry */
 };
 
diff --git a/drivers/usb/serial/pl2303.h b/drivers/usb/serial/pl2303.h
index 1807087a76e3..9bc4755162ad 100644
--- a/drivers/usb/serial/pl2303.h
+++ b/drivers/usb/serial/pl2303.h
@@ -71,3 +71,7 @@
 
 #define SAGEM_VENDOR_ID		0x079b
 #define SAGEM_PRODUCT_ID	0x0027
+
+/* Leadtek GPS 9531 (ID 0413:2101) */
+#define LEADTEK_VENDOR_ID	0x0413
+#define LEADTEK_9531_PRODUCT_ID	0x2101

From a0c53f1dca10acc93462339cbd0bf24b10d60a13 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Mon, 6 Feb 2006 12:15:15 -0800
Subject: [PATCH 142/509] [PATCH] USB: sl811_cs needs platform_device
 conversion too

The switchover to "platform_driver" from "device_driver" missed
one rather essential usage, which broke the sl811_cs driver ...
this resolves the omission.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/host/sl811_cs.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/usb/host/sl811_cs.c b/drivers/usb/host/sl811_cs.c
index 466384d7c79f..134d2000128a 100644
--- a/drivers/usb/host/sl811_cs.c
+++ b/drivers/usb/host/sl811_cs.c
@@ -101,7 +101,7 @@ static struct resource resources[] = {
 	},
 };
 
-extern struct device_driver sl811h_driver;
+extern struct platform_driver sl811h_driver;
 
 static struct platform_device platform_dev = {
 	.id			= -1,
@@ -132,7 +132,7 @@ static int sl811_hc_init(struct device *parent, ioaddr_t base_addr, int irq)
 	 * initialized already because of the link order dependency created
 	 * by referencing "sl811h_driver".
 	 */
-	platform_dev.name = sl811h_driver.name;
+	platform_dev.name = sl811h_driver.driver.name;
 	return platform_device_register(&platform_dev);
 }
 

From a2149bce2535047371a1309f032f0320b05db791 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Thu, 2 Feb 2006 09:52:45 -0500
Subject: [PATCH 143/509] [PATCH] usb-storage: new unusual_devs entry

This patch (as631) for unusual_devs.h fixes bugzilla entry 5913.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Phil Dibowitz <phil@ipom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/unusual_devs.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index ee958f986eb8..c5d6d84b9b2b 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -1162,6 +1162,13 @@ UNUSUAL_DEV(  0x55aa, 0xa103, 0x0000, 0x9999,
 		US_FL_SINGLE_LUN),
 #endif
 
+/* Reported by Andrew Simmons <andrew.simmons@gmail.com> */
+UNUSUAL_DEV(  0xed06, 0x4500, 0x0001, 0x0001,
+		"DataStor",
+		"USB4500 FW1.04",
+		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		US_FL_FIX_CAPACITY),
+
 /* Control/Bulk transport for all SubClass values */
 USUAL_DEV(US_SC_RBC, US_PR_CB, USB_US_TYPE_STOR),
 USUAL_DEV(US_SC_8020, US_PR_CB, USB_US_TYPE_STOR),

From ba3e93ad3c6e511f68c1b3e46954feff6cdd5a62 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 30 Jan 2006 10:19:43 -0500
Subject: [PATCH 144/509] [PATCH] usb-storage: unusual_devs entry

Here is a new entry for unusual_devs.h (as630).

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/unusual_devs.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index c5d6d84b9b2b..6e326159c2dd 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -946,6 +946,12 @@ UNUSUAL_DEV(  0x084d, 0x0011, 0x0110, 0x0110,
 		US_SC_DEVICE, US_PR_DEVICE, NULL,
 		US_FL_BULK32),
 
+/* Submitted by Jan De Luyck <lkml@kcore.org> */
+UNUSUAL_DEV(  0x08bd, 0x1100, 0x0000, 0x0000,
+		"CITIZEN",
+		"X1DE-USB",
+		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		US_FL_SINGLE_LUN),
 
 /* Entry needed for flags. Moreover, all devices with this ID use
  * bulk-only transport, but _some_ falsely report Control/Bulk instead.

From 1d614a4b0d2613d83f7bf0978e213ba29aebc44f Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 13 Feb 2006 10:15:22 -0500
Subject: [PATCH 145/509] [PATCH] USB: unusual_devs.h entry: TrekStor i.Beat

A new unusual_devs.h entry (as651).

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/unusual_devs.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 6e326159c2dd..1d8a739ccc96 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -106,6 +106,13 @@ UNUSUAL_DEV(  0x0411, 0x001c, 0x0113, 0x0113,
 		US_SC_DEVICE, US_PR_DEVICE, NULL,
 		US_FL_FIX_INQUIRY ),
 
+/* Reported by Christian Leber <christian@leber.de> */
+UNUSUAL_DEV(  0x0419, 0xaaf5, 0x0100, 0x0100,
+		"TrekStor",
+		"i.Beat 115 2.0",
+		US_SC_DEVICE, US_PR_DEVICE, NULL,
+		US_FL_IGNORE_RESIDUE | US_FL_NOT_LOCKABLE ),
+
 /* Reported by Stefan Werner <dustbln@gmx.de> */
 UNUSUAL_DEV(  0x0419, 0xaaf6, 0x0100, 0x0100,
 		"TrekStor",

From 982db2a127b29c9e1f5896d2ba691e84446cc858 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Mon, 13 Feb 2006 10:16:04 -0500
Subject: [PATCH 146/509] [PATCH] USB: unusual_devs.h entry: iAUDIO M5

Another unusual_devs.h entry (as652).

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/unusual_devs.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 1d8a739ccc96..7e1cb03c2ed7 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -1098,6 +1098,13 @@ UNUSUAL_DEV(  0x0dda, 0x0301, 0x0012, 0x0012,
 		US_SC_DEVICE, US_PR_DEVICE, NULL,
 		US_FL_IGNORE_RESIDUE ),
 
+/* Reported by Jim McCloskey <mcclosk@ucsc.edu> */
+UNUSUAL_DEV( 0x0e21, 0x0520, 0x0100, 0x0100,
+		"Cowon Systems",
+		"iAUDIO M5",
+		US_SC_DEVICE, US_PR_BULK, NULL,
+		0 ),
+
 /* Submitted by Antoine Mairesse <antoine.mairesse@free.fr> */
 UNUSUAL_DEV( 0x0ed1, 0x6660, 0x0100, 0x0300,
 		"USB",

From 16f05be7be0bf121491d83bd97337fe179b3b323 Mon Sep 17 00:00:00 2001
From: Phil Dibowitz <phil@ipom.com>
Date: Mon, 13 Feb 2006 15:59:42 -0800
Subject: [PATCH 147/509] [PATCH] USB: unusual-devs bugfix

The following patch looks good to me. It adds an unusual_devs entry as
well as fixing an ordering bug. Please apply.

From: Bohdan Linda <bohdan.linda@gmail.com>
Signed-off-by: Phil Dibowitz <phil@ipom.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/usb/storage/unusual_devs.h | 25 ++++++++++++++++++-------
 1 file changed, 18 insertions(+), 7 deletions(-)

diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h
index 7e1cb03c2ed7..e71c5ca1a07b 100644
--- a/drivers/usb/storage/unusual_devs.h
+++ b/drivers/usb/storage/unusual_devs.h
@@ -134,6 +134,14 @@ UNUSUAL_DEV(  0x0436, 0x0005, 0x0100, 0x0100,
  		US_SC_SCSI, US_PR_DPCM_USB, NULL, 0 ),
 #endif
 
+/* Patch submitted by Daniel Drake <dsd@gentoo.org>
+ * Device reports nonsense bInterfaceProtocol 6 when connected over USB2 */
+UNUSUAL_DEV(  0x0451, 0x5416, 0x0100, 0x0100,
+		"Neuros Audio",
+		"USB 2.0 HD 2.5",
+		US_SC_DEVICE, US_PR_BULK, NULL,
+		US_FL_NEED_OVERRIDE ),
+
 /*
  * Pete Zaitcev <zaitcev@yahoo.com>, from Patrick C. F. Ernzer, bz#162559.
  * The key does not actually break, but it returns zero sense which
@@ -144,13 +152,16 @@ UNUSUAL_DEV(  0x0457, 0x0150, 0x0100, 0x0100,
 		"USB Mass Storage Device",
 		US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_NOT_LOCKABLE ),
 
-/* Patch submitted by Daniel Drake <dsd@gentoo.org>
- * Device reports nonsense bInterfaceProtocol 6 when connected over USB2 */
-UNUSUAL_DEV(  0x0451, 0x5416, 0x0100, 0x0100,
-		"Neuros Audio",
-		"USB 2.0 HD 2.5",
-		US_SC_DEVICE, US_PR_BULK, NULL,
-		US_FL_NEED_OVERRIDE ),
+/*
+* Bohdan Linda <bohdan.linda@gmail.com>
+* 1GB USB sticks MyFlash High Speed. I have restricted
+* the revision to my model only
+*/
+UNUSUAL_DEV(  0x0457, 0x0151, 0x0100, 0x0100,
+                "USB 2.0",
+                "Flash Disk",
+                US_SC_DEVICE, US_PR_DEVICE, NULL,
+                US_FL_NOT_LOCKABLE ),
 
 UNUSUAL_DEV(  0x045a, 0x5210, 0x0101, 0x0101,
 		"Rio",

From e3efa5a7392e07471b5d0ef7e7cd7ab862f70284 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 5 Feb 2006 23:11:16 +0100
Subject: [PATCH 148/509] [PATCH] vt8231: Fix sysfs temperature interface

The VT8231 low temperature limits are actually hysteresis temperatures
to the high limits.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Roger Lucas <roger@planbit.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/hwmon/vt8231.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/hwmon/vt8231.c b/drivers/hwmon/vt8231.c
index 3eb08f004c0f..271e9cb9532c 100644
--- a/drivers/hwmon/vt8231.c
+++ b/drivers/hwmon/vt8231.c
@@ -437,12 +437,12 @@ static SENSOR_DEVICE_ATTR(temp##offset##_input, S_IRUGO,		\
 		show_temp, NULL, offset - 1);				\
 static SENSOR_DEVICE_ATTR(temp##offset##_max, S_IRUGO | S_IWUSR,	\
 		show_temp_max, set_temp_max, offset - 1);		\
-static SENSOR_DEVICE_ATTR(temp##offset##_min, S_IRUGO | S_IWUSR,	\
+static SENSOR_DEVICE_ATTR(temp##offset##_max_hyst, S_IRUGO | S_IWUSR,	\
 		show_temp_min, set_temp_min, offset - 1)
 
 static DEVICE_ATTR(temp1_input, S_IRUGO, show_temp0, NULL);
 static DEVICE_ATTR(temp1_max, S_IRUGO | S_IWUSR, show_temp0_max, set_temp0_max);
-static DEVICE_ATTR(temp1_min, S_IRUGO | S_IWUSR, show_temp0_min, set_temp0_min);
+static DEVICE_ATTR(temp1_max_hyst, S_IRUGO | S_IWUSR, show_temp0_min, set_temp0_min);
 
 define_temperature_sysfs(2);
 define_temperature_sysfs(3);
@@ -451,7 +451,7 @@ define_temperature_sysfs(5);
 define_temperature_sysfs(6);
 
 #define CFG_INFO_TEMP(id)	{ &sensor_dev_attr_temp##id##_input.dev_attr, \
-				&sensor_dev_attr_temp##id##_min.dev_attr, \
+				&sensor_dev_attr_temp##id##_max_hyst.dev_attr, \
 				&sensor_dev_attr_temp##id##_max.dev_attr }
 #define CFG_INFO_VOLT(id)	{ &sensor_dev_attr_in##id##_input.dev_attr, \
 				&sensor_dev_attr_in##id##_min.dev_attr, \
@@ -464,7 +464,7 @@ struct str_device_attr_table {
 };
 
 static struct str_device_attr_table cfg_info_temp[] = {
-	{ &dev_attr_temp1_input, &dev_attr_temp1_min, &dev_attr_temp1_max },
+	{ &dev_attr_temp1_input, &dev_attr_temp1_max_hyst, &dev_attr_temp1_max },
 	CFG_INFO_TEMP(2),
 	CFG_INFO_TEMP(3),
 	CFG_INFO_TEMP(4),

From c7f5d7edd8b3fa4204389efc4c9081cc90a811d2 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 5 Feb 2006 23:13:48 +0100
Subject: [PATCH 149/509] [PATCH] w83781d: Use real-time status registers

Use the real-time status registers of the Winbond W83782D, W83783S and
W83627HF chips, instead of the interrupt status registers. Interrupts
cannot be trusted at least for voltage inputs, as they are two-times
triggers (as opposed to comparator mode, which we want.) The w83627hf
driver was fixed in a similar way some times ago.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/hwmon/w83781d.c | 43 ++++++++++++++++++++++++-----------------
 1 file changed, 25 insertions(+), 18 deletions(-)

diff --git a/drivers/hwmon/w83781d.c b/drivers/hwmon/w83781d.c
index 557114872f3c..64c1f8af5bb2 100644
--- a/drivers/hwmon/w83781d.c
+++ b/drivers/hwmon/w83781d.c
@@ -95,11 +95,16 @@ MODULE_PARM_DESC(init, "Set to zero to bypass chip initialization");
 						     (0x39)))
 
 #define W83781D_REG_CONFIG		0x40
+
+/* Interrupt status (W83781D, AS99127F) */
 #define W83781D_REG_ALARM1		0x41
 #define W83781D_REG_ALARM2		0x42
-#define W83781D_REG_ALARM3		0x450	/* not on W83781D */
 
-#define W83781D_REG_IRQ			0x4C
+/* Real-time status (W83782D, W83783S, W83627HF) */
+#define W83782D_REG_ALARM1		0x459
+#define W83782D_REG_ALARM2		0x45A
+#define W83782D_REG_ALARM3		0x45B
+
 #define W83781D_REG_BEEP_CONFIG		0x4D
 #define W83781D_REG_BEEP_INTS1		0x56
 #define W83781D_REG_BEEP_INTS2		0x57
@@ -1513,15 +1518,6 @@ w83781d_init_client(struct i2c_client *client)
 					W83781D_REG_TEMP3_CONFIG, tmp & 0xfe);
 			}
 		}
-
-		if (type != w83781d) {
-			/* enable comparator mode for temp2 and temp3 so
-			   alarm indication will work correctly */
-			i = w83781d_read_value(client, W83781D_REG_IRQ);
-			if (!(i & 0x40))
-				w83781d_write_value(client, W83781D_REG_IRQ,
-						    i | 0x40);
-		}
 	}
 
 	/* Start monitoring */
@@ -1612,14 +1608,25 @@ static struct w83781d_data *w83781d_update_device(struct device *dev)
 			data->fan_div[1] |= (i >> 4) & 0x04;
 			data->fan_div[2] |= (i >> 5) & 0x04;
 		}
-		data->alarms =
-		    w83781d_read_value(client,
-				       W83781D_REG_ALARM1) +
-		    (w83781d_read_value(client, W83781D_REG_ALARM2) << 8);
 		if ((data->type == w83782d) || (data->type == w83627hf)) {
-			data->alarms |=
-			    w83781d_read_value(client,
-					       W83781D_REG_ALARM3) << 16;
+			data->alarms = w83781d_read_value(client,
+						W83782D_REG_ALARM1)
+				     | (w83781d_read_value(client,
+						W83782D_REG_ALARM2) << 8)
+				     | (w83781d_read_value(client,
+						W83782D_REG_ALARM3) << 16);
+		} else if (data->type == w83783s) {
+			data->alarms = w83781d_read_value(client,
+						W83782D_REG_ALARM1)
+				     | (w83781d_read_value(client,
+						W83782D_REG_ALARM2) << 8);
+		} else {
+			/* No real-time status registers, fall back to
+			   interrupt status registers */
+			data->alarms = w83781d_read_value(client,
+						W83781D_REG_ALARM1)
+				     | (w83781d_read_value(client,
+						W83781D_REG_ALARM2) << 8);
 		}
 		i = w83781d_read_value(client, W83781D_REG_BEEP_INTS2);
 		data->beep_enable = i >> 7;

From a40f0b0f2466483fad94e62177272d6a5711e50a Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Sun, 5 Feb 2006 23:17:34 +0100
Subject: [PATCH 150/509] [PATCH] w83627hf: Document the reset module parameter

Document the reset module parameter which was recently added to the
w83627hf driver.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/hwmon/w83627hf | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/Documentation/hwmon/w83627hf b/Documentation/hwmon/w83627hf
index 5d23776e9907..bbeaba680443 100644
--- a/Documentation/hwmon/w83627hf
+++ b/Documentation/hwmon/w83627hf
@@ -36,6 +36,10 @@ Module Parameters
   (default is 1)
   Use 'init=0' to bypass initializing the chip.
   Try this if your computer crashes when you load the module.
+* reset: int
+  (default is 0)
+  The driver used to reset the chip on load, but does no more. Use
+  'reset=1' to restore the old behavior. Report if you need to do this.
 
 Description
 -----------

From be79c383254cd3eb50953d8c0a7cacdbf6db31c0 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Tue, 7 Feb 2006 17:53:32 +0100
Subject: [PATCH 151/509] [PATCH] it87: Fix oops on removal

Fix an oops on it87 module removal when no supported hardware was
found.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/hwmon/it87.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/hwmon/it87.c b/drivers/hwmon/it87.c
index e87d52c59940..d7a9401600bb 100644
--- a/drivers/hwmon/it87.c
+++ b/drivers/hwmon/it87.c
@@ -1186,7 +1186,8 @@ static int __init sm_it87_init(void)
 
 static void __exit sm_it87_exit(void)
 {
-	i2c_isa_del_driver(&it87_isa_driver);
+	if (isa_address)
+		i2c_isa_del_driver(&it87_isa_driver);
 	i2c_del_driver(&it87_driver);
 }
 

From 41d9c98fe76298cebc5907bcebfb2db28017a277 Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Wed, 8 Feb 2006 20:38:29 +0100
Subject: [PATCH 152/509] [PATCH] i2c: Drop outdated probe/remove code in
 i2c-isa

Probe and remove methods are now defined at bus level. No more need to
redefine them at driver level in i2c-isa.

This lets us get rid of these annoying messages:
Driver 'it87-isa' needs updating - please use bus_type methods

Thanks to Nicolas Mailhot for reporting the problem and testing the fix.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 drivers/i2c/busses/i2c-isa.c | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/i2c/busses/i2c-isa.c b/drivers/i2c/busses/i2c-isa.c
index 9f2ffef4d812..4344ae6b1fcb 100644
--- a/drivers/i2c/busses/i2c-isa.c
+++ b/drivers/i2c/busses/i2c-isa.c
@@ -72,16 +72,6 @@ static ssize_t show_adapter_name(struct device *dev,
 }
 static DEVICE_ATTR(name, S_IRUGO, show_adapter_name, NULL);
 
-static int i2c_isa_device_probe(struct device *dev)
-{
-	return -ENODEV;
-}
-
-static int i2c_isa_device_remove(struct device *dev)
-{
-	return 0;
-}
-
 
 /* We implement an interface which resembles i2c_{add,del}_driver,
    but for i2c-isa drivers. We don't have to remember and handle lists
@@ -93,8 +83,6 @@ int i2c_isa_add_driver(struct i2c_driver *driver)
 
 	/* Add the driver to the list of i2c drivers in the driver core */
 	driver->driver.bus = &i2c_bus_type;
-	driver->driver.probe = i2c_isa_device_probe;
-	driver->driver.remove = i2c_isa_device_remove;
 	res = driver_register(&driver->driver);
 	if (res)
 		return res;

From 303794400992b907b7cac0d91788603636c7e0fe Mon Sep 17 00:00:00 2001
From: Gerald Britton <gbritton@alum.mit.edu>
Date: Tue, 14 Feb 2006 10:19:04 -0500
Subject: [PATCH 153/509] [PATCH] x86: fix oprofile kernel callgraph regression

Fix x86 oprofile regression introduced by:
  commit c34d1b4d165c67b966bca4aba026443d7ff161eb
  [PATCH] mm: kill check_user_page_readable

That commit reorganized tests for the userspace stack walking moving all
those tests into dump_backtrace(), however, dump_backtrace() was used for
both userspace and kernel stalk walking.  The result is typically no
recorded callgraph information for kernel samples.

Revive the original function as dump_kernel_backtrace() and rename the
other to dump_user_backtrace() to avoid future confusion.

Signed-off-by: Gerald Britton <gbritton@alum.mit.edu>
Apology-from: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/oprofile/backtrace.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/arch/i386/oprofile/backtrace.c b/arch/i386/oprofile/backtrace.c
index acc18138fb22..c049ce414f01 100644
--- a/arch/i386/oprofile/backtrace.c
+++ b/arch/i386/oprofile/backtrace.c
@@ -20,7 +20,20 @@ struct frame_head {
 } __attribute__((packed));
 
 static struct frame_head *
-dump_backtrace(struct frame_head * head)
+dump_kernel_backtrace(struct frame_head * head)
+{
+	oprofile_add_trace(head->ret);
+
+	/* frame pointers should strictly progress back up the stack
+	 * (towards higher addresses) */
+	if (head >= head->ebp)
+		return NULL;
+
+	return head->ebp;
+}
+
+static struct frame_head *
+dump_user_backtrace(struct frame_head * head)
 {
 	struct frame_head bufhead[2];
 
@@ -105,10 +118,10 @@ x86_backtrace(struct pt_regs * const regs, unsigned int depth)
 
 	if (!user_mode_vm(regs)) {
 		while (depth-- && valid_kernel_stack(head, regs))
-			head = dump_backtrace(head);
+			head = dump_kernel_backtrace(head);
 		return;
 	}
 
 	while (depth-- && head)
-		head = dump_backtrace(head);
+		head = dump_user_backtrace(head);
 }

From faead26d7a06605add627f29aee73ba654ce11f9 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Tue, 14 Feb 2006 10:42:07 -0600
Subject: [PATCH 154/509] [PATCH] add scsi_execute_in_process_context() API

We have several points in the SCSI stack (primarily for our device
functions) where we need to guarantee process context, but (given the
place where the last reference was released) we cannot guarantee this.

This API gets around the issue by executing the function directly if
the caller has process context, but scheduling a workqueue to execute
in process context if the caller doesn't have it.  Unfortunately, it
requires memory allocation in interrupt context, but it's better than
what we have previously.  The true solution will require a bit of
re-engineering, so isn't appropriate for 2.6.16.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_lib.c | 59 +++++++++++++++++++++++++++++++++++++++++
 include/scsi/scsi.h     |  2 ++
 2 files changed, 61 insertions(+)

diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index 4a602853a98e..4362dcde74af 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -16,6 +16,7 @@
 #include <linux/init.h>
 #include <linux/pci.h>
 #include <linux/delay.h>
+#include <linux/hardirq.h>
 
 #include <scsi/scsi.h>
 #include <scsi/scsi_dbg.h>
@@ -2248,3 +2249,61 @@ scsi_target_unblock(struct device *dev)
 		device_for_each_child(dev, NULL, target_unblock);
 }
 EXPORT_SYMBOL_GPL(scsi_target_unblock);
+
+
+struct work_queue_work {
+	struct work_struct	work;
+	void			(*fn)(void *);
+	void			*data;
+};
+
+static void execute_in_process_context_work(void *data)
+{
+	void (*fn)(void *data);
+	struct work_queue_work *wqw = data;
+
+	fn = wqw->fn;
+	data = wqw->data;
+
+	kfree(wqw);
+
+	fn(data);
+}
+
+/**
+ * scsi_execute_in_process_context - reliably execute the routine with user context
+ * @fn:		the function to execute
+ * @data:	data to pass to the function
+ *
+ * Executes the function immediately if process context is available,
+ * otherwise schedules the function for delayed execution.
+ *
+ * Returns:	0 - function was executed
+ *		1 - function was scheduled for execution
+ *		<0 - error
+ */
+int scsi_execute_in_process_context(void (*fn)(void *data), void *data)
+{
+	struct work_queue_work *wqw;
+
+	if (!in_interrupt()) {
+		fn(data);
+		return 0;
+	}
+
+	wqw = kmalloc(sizeof(struct work_queue_work), GFP_ATOMIC);
+
+	if (unlikely(!wqw)) {
+		printk(KERN_ERR "Failed to allocate memory\n");
+		WARN_ON(1);
+		return -ENOMEM;
+	}
+
+	INIT_WORK(&wqw->work, execute_in_process_context_work, wqw);
+	wqw->fn = fn;
+	wqw->data = data;
+	schedule_work(&wqw->work);
+
+	return 1;
+}
+EXPORT_SYMBOL_GPL(scsi_execute_in_process_context);
diff --git a/include/scsi/scsi.h b/include/scsi/scsi.h
index c60b8ff2f5e4..9c331258bc27 100644
--- a/include/scsi/scsi.h
+++ b/include/scsi/scsi.h
@@ -433,4 +433,6 @@ struct scsi_lun {
 /* Used to obtain the PCI location of a device */
 #define SCSI_IOCTL_GET_PCI		0x5387
 
+int scsi_execute_in_process_context(void (*fn)(void *data), void *data);
+
 #endif /* _SCSI_SCSI_H */

From 65110b2168950a19cc78b5027ed18cb811fbdae8 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@steeleye.com>
Date: Tue, 14 Feb 2006 10:48:46 -0600
Subject: [PATCH 155/509] [SCSI] fix wrong context bugs in SCSI

There's a bug in releasing scsi_device where the release function
actually frees the block queue.  However, the block queue release
calls flush_work(), which requires process context (the scsi_device
structure may release from irq context).  Update the release function
to invoke via the execute_in_process_context() API.

Also clean up the scsi_target structure releasing via this API.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
---
 drivers/scsi/scsi_scan.c  | 26 ++++----------------------
 drivers/scsi/scsi_sysfs.c |  9 ++++++++-
 2 files changed, 12 insertions(+), 23 deletions(-)

diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c
index 752fb5da3de4..5acb83ca5ae5 100644
--- a/drivers/scsi/scsi_scan.c
+++ b/drivers/scsi/scsi_scan.c
@@ -387,19 +387,12 @@ static struct scsi_target *scsi_alloc_target(struct device *parent,
 	return found_target;
 }
 
-struct work_queue_wrapper {
-	struct work_struct	work;
-	struct scsi_target	*starget;
-};
-
-static void scsi_target_reap_work(void *data) {
-	struct work_queue_wrapper *wqw = (struct work_queue_wrapper *)data;
-	struct scsi_target *starget = wqw->starget;
+static void scsi_target_reap_usercontext(void *data)
+{
+	struct scsi_target *starget = data;
 	struct Scsi_Host *shost = dev_to_shost(starget->dev.parent);
 	unsigned long flags;
 
-	kfree(wqw);
-
 	spin_lock_irqsave(shost->host_lock, flags);
 
 	if (--starget->reap_ref == 0 && list_empty(&starget->devices)) {
@@ -428,18 +421,7 @@ static void scsi_target_reap_work(void *data) {
  */
 void scsi_target_reap(struct scsi_target *starget)
 {
-	struct work_queue_wrapper *wqw = 
-		kzalloc(sizeof(struct work_queue_wrapper), GFP_ATOMIC);
-
-	if (!wqw) {
-		starget_printk(KERN_ERR, starget,
-			       "Failed to allocate memory in scsi_reap_target()\n");
-		return;
-	}
-
-	INIT_WORK(&wqw->work, scsi_target_reap_work, wqw);
-	wqw->starget = starget;
-	schedule_work(&wqw->work);
+	scsi_execute_in_process_context(scsi_target_reap_usercontext, starget);
 }
 
 /**
diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c
index a77b32deaf8f..902a5def8e62 100644
--- a/drivers/scsi/scsi_sysfs.c
+++ b/drivers/scsi/scsi_sysfs.c
@@ -217,8 +217,9 @@ static void scsi_device_cls_release(struct class_device *class_dev)
 	put_device(&sdev->sdev_gendev);
 }
 
-static void scsi_device_dev_release(struct device *dev)
+static void scsi_device_dev_release_usercontext(void *data)
 {
+	struct device *dev = data;
 	struct scsi_device *sdev;
 	struct device *parent;
 	struct scsi_target *starget;
@@ -237,6 +238,7 @@ static void scsi_device_dev_release(struct device *dev)
 
 	if (sdev->request_queue) {
 		sdev->request_queue->queuedata = NULL;
+		/* user context needed to free queue */
 		scsi_free_queue(sdev->request_queue);
 		/* temporary expedient, try to catch use of queue lock
 		 * after free of sdev */
@@ -252,6 +254,11 @@ static void scsi_device_dev_release(struct device *dev)
 		put_device(parent);
 }
 
+static void scsi_device_dev_release(struct device *dev)
+{
+	scsi_execute_in_process_context(scsi_device_dev_release_usercontext,	dev);
+}
+
 static struct class sdev_class = {
 	.name		= "scsi_device",
 	.release	= scsi_device_cls_release,

From 5552c28f6937d2a2b873d06c6d09b96204722dd0 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Tue, 14 Feb 2006 17:50:54 +0000
Subject: [PATCH 156/509] [PATCH] Fix locking error in esp

Noted by Al Viro.

Also remove unused tmp_buf

Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/esp.c | 22 ++--------------------
 1 file changed, 2 insertions(+), 20 deletions(-)

diff --git a/drivers/char/esp.c b/drivers/char/esp.c
index 57539d8f9f7c..09dc4b01232c 100644
--- a/drivers/char/esp.c
+++ b/drivers/char/esp.c
@@ -150,17 +150,6 @@ static void rs_wait_until_sent(struct tty_struct *, int);
 /* Standard COM flags (except for COM4, because of the 8514 problem) */
 #define STD_COM_FLAGS (ASYNC_BOOT_AUTOCONF | ASYNC_SKIP_TEST)
 
-/*
- * tmp_buf is used as a temporary buffer by serial_write.  We need to
- * lock it in case the memcpy_fromfs blocks while swapping in a page,
- * and some other program tries to do a serial write at the same time.
- * Since the lock will only come under contention when the system is
- * swapping and available memory is low, it makes sense to share one
- * buffer across all the serial ports, since it significantly saves
- * memory if large numbers of serial ports are open.
- */
-static unsigned char *tmp_buf;
-
 static inline int serial_paranoia_check(struct esp_struct *info,
 					char *name, const char *routine)
 {
@@ -1267,7 +1256,7 @@ static int rs_write(struct tty_struct * tty,
 	if (serial_paranoia_check(info, tty->name, "rs_write"))
 		return 0;
 
-	if (!tty || !info->xmit_buf || !tmp_buf)
+	if (!tty || !info->xmit_buf)
 		return 0;
 	    
 	while (1) {
@@ -2291,11 +2280,7 @@ static int esp_open(struct tty_struct *tty, struct file * filp)
 	tty->driver_data = info;
 	info->tty = tty;
 
-	if (!tmp_buf) {
-		tmp_buf = (unsigned char *) get_zeroed_page(GFP_KERNEL);
-		if (!tmp_buf)
-			return -ENOMEM;
-	}
+	spin_unlock_irqrestore(&info->lock, flags);
 	
 	/*
 	 * Start up serial port
@@ -2602,9 +2587,6 @@ static void __exit espserial_exit(void)
 		free_pages((unsigned long)dma_buffer,
 			get_order(DMA_BUFFER_SZ));
 
-	if (tmp_buf)
-		free_page((unsigned long)tmp_buf);
-
 	while (free_pio_buf) {
 		pio_buf = free_pio_buf->next;
 		kfree(free_pio_buf);

From f32ec77b421ee15bf5a42082b60679e997c07993 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 28 Nov 2005 13:10:54 +0000
Subject: [PATCH 157/509] [MIPS] RM200: Give RM200 it's own timex.h.

So we can get rid of config.h and the #ifdef crapola in the generic
timex.h.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/mach-generic/timex.h | 11 +----------
 include/asm-mips/mach-rm200/timex.h   | 13 +++++++++++++
 2 files changed, 14 insertions(+), 10 deletions(-)
 create mode 100644 include/asm-mips/mach-rm200/timex.h

diff --git a/include/asm-mips/mach-generic/timex.h b/include/asm-mips/mach-generic/timex.h
index c6a2e5f0574a..48b4cfaa0d50 100644
--- a/include/asm-mips/mach-generic/timex.h
+++ b/include/asm-mips/mach-generic/timex.h
@@ -3,20 +3,11 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2003 by Ralf Baechle
+ * Copyright (C) 2003, 2005 by Ralf Baechle
  */
 #ifndef __ASM_MACH_GENERIC_TIMEX_H
 #define __ASM_MACH_GENERIC_TIMEX_H
 
-#include <linux/config.h>
-
-/*
- * Last remaining user of the i8254 PIC, will be converted, too ...
- */
-#ifdef CONFIG_SNI_RM200_PCI
-#define CLOCK_TICK_RATE		1193182
-#else
 #define CLOCK_TICK_RATE		500000
-#endif
 
 #endif /* __ASM_MACH_GENERIC_TIMEX_H */
diff --git a/include/asm-mips/mach-rm200/timex.h b/include/asm-mips/mach-rm200/timex.h
new file mode 100644
index 000000000000..11ff6cb0f214
--- /dev/null
+++ b/include/asm-mips/mach-rm200/timex.h
@@ -0,0 +1,13 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 2003, 2005 by Ralf Baechle
+ */
+#ifndef __ASM_MACH_RM200_TIMEX_H
+#define __ASM_MACH_RM200_TIMEX_H
+
+#define CLOCK_TICK_RATE		1193182
+
+#endif /* __ASM_MACH_RM200_TIMEX_H */

From a3c9dc38313d05e4254c100e86af205cf33e8cd3 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 14 Feb 2006 19:13:23 +0000
Subject: [PATCH 158/509] [MIPS] Update docs to reflect the latest status of
 the Alchemy IDE driver.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 Documentation/mips/AU1xxx_IDE.README | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/Documentation/mips/AU1xxx_IDE.README b/Documentation/mips/AU1xxx_IDE.README
index a7e4c4ea3560..afb31c141d9d 100644
--- a/Documentation/mips/AU1xxx_IDE.README
+++ b/Documentation/mips/AU1xxx_IDE.README
@@ -95,11 +95,13 @@ CONFIG_BLK_DEV_IDEDMA_PCI=y
 CONFIG_IDEDMA_PCI_AUTO=y
 CONFIG_BLK_DEV_IDE_AU1XXX=y
 CONFIG_BLK_DEV_IDE_AU1XXX_MDMA2_DBDMA=y
-CONFIG_BLK_DEV_IDE_AU1XXX_BURSTABLE_ON=y
 CONFIG_BLK_DEV_IDE_AU1XXX_SEQTS_PER_RQ=128
 CONFIG_BLK_DEV_IDEDMA=y
 CONFIG_IDEDMA_AUTO=y
 
+Also define 'IDE_AU1XXX_BURSTMODE' in 'drivers/ide/mips/au1xxx-ide.c' to enable
+the burst support on DBDMA controller.
+
 If the used system need the USB support enable the following kernel configs for
 high IDE to USB throughput.
 
@@ -115,6 +117,8 @@ CONFIG_BLK_DEV_IDE_AU1XXX_SEQTS_PER_RQ=128
 CONFIG_BLK_DEV_IDEDMA=y
 CONFIG_IDEDMA_AUTO=y
 
+Also undefine 'IDE_AU1XXX_BURSTMODE' in 'drivers/ide/mips/au1xxx-ide.c' to
+disable the burst support on DBDMA controller.
 
 ADD NEW HARD DISC TO WHITE OR BLACK LIST
 ----------------------------------------

From 359bbd42a5a205234d5943571fc7bf946967ee59 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 9 Feb 2006 12:13:28 +0000
Subject: [PATCH 159/509] [MIPS] Fold non-__mips64 case into CONFIG_32BIT case.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/unistd.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h
index e7ff9b187783..769305d20108 100644
--- a/include/asm-mips/unistd.h
+++ b/include/asm-mips/unistd.h
@@ -1184,10 +1184,8 @@ type name (atype a,btype b,ctype c,dtype d,etype e,ftype f) \
 #define __ARCH_WANT_SYS_SIGPENDING
 #define __ARCH_WANT_SYS_SIGPROCMASK
 #define __ARCH_WANT_SYS_RT_SIGACTION
-# ifndef __mips64
-#  define __ARCH_WANT_STAT64
-# endif
 # ifdef CONFIG_32BIT
+#  define __ARCH_WANT_STAT64
 #  define __ARCH_WANT_SYS_TIME
 # endif
 # ifdef CONFIG_MIPS32_O32

From 1bdfd0d9632f0254a4fc01d17c9933ae6107dec4 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 9 Feb 2006 12:26:35 +0000
Subject: [PATCH 160/509] [MIPS] Remove commented out function
 prom_build_cpu_map.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/smp_mt.c | 29 -----------------------------
 1 file changed, 29 deletions(-)

diff --git a/arch/mips/kernel/smp_mt.c b/arch/mips/kernel/smp_mt.c
index 794a1c3de2a4..186f5deced40 100644
--- a/arch/mips/kernel/smp_mt.c
+++ b/arch/mips/kernel/smp_mt.c
@@ -102,35 +102,6 @@ void __init sanitize_tlb_entries(void)
 	clear_c0_mvpcontrol(MVPCONTROL_VPC);
 }
 
-#if 0
-/*
- * Use c0_MVPConf0 to find out how many CPUs are available, setting up
- * phys_cpu_present_map and the logical/physical mappings.
- */
-void __init prom_build_cpu_map(void)
-{
-	int i, num, ncpus;
-
-	cpus_clear(phys_cpu_present_map);
-
-	/* assume we boot on cpu 0.... */
-	cpu_set(0, phys_cpu_present_map);
-	__cpu_number_map[0] = 0;
-	__cpu_logical_map[0] = 0;
-
-	if (cpu_has_mipsmt) {
-		ncpus = ((read_c0_mvpconf0() & (MVPCONF0_PVPE)) >> MVPCONF0_PVPE_SHIFT) + 1;
-		for (i=1, num=0; i< NR_CPUS && i<ncpus; i++) {
-			cpu_set(i, phys_cpu_present_map);
-			__cpu_number_map[i] = ++num;
-			__cpu_logical_map[num] = i;
-		}
-
-		printk(KERN_INFO "%i available secondary CPU(s)\n", num);
-	}
-}
-#endif
-
 static void ipi_resched_dispatch (struct pt_regs *regs)
 {
 	do_IRQ(MIPS_CPU_IPI_RESCHED_IRQ, regs);

From 63077519899721120b61d663a68adced068a459d Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Wed, 8 Feb 2006 01:48:03 +0900
Subject: [PATCH 161/509] [MIPS] Rewrite get_wchan and its helper functions
 using kallsyms_lookup.

Implement get_wchan() and frame_info_init() using kallsyms_lookup().
This fixes problem with static sched/lock functions and mfinfo[]
maintenance issue.  If CONFIG_KALLSYMS was disabled, get_wchan() just
returns thread_saved_pc() value.

Also unwind stackframe based on "addiu sp,-imm" analysis instead of
frame pointer.  This fixes problem with functions compiled without
-fomit-frame-pointer.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/process.c | 158 ++++++++++++++++++-------------------
 1 file changed, 77 insertions(+), 81 deletions(-)

diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 5232fc752935..092679c2dca9 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -25,6 +25,7 @@
 #include <linux/a.out.h>
 #include <linux/init.h>
 #include <linux/completion.h>
+#include <linux/kallsyms.h>
 
 #include <asm/abi.h>
 #include <asm/bootinfo.h>
@@ -272,46 +273,19 @@ long kernel_thread(int (*fn)(void *), void *arg, unsigned long flags)
 
 static struct mips_frame_info {
 	void *func;
-	int omit_fp;	/* compiled without fno-omit-frame-pointer */
-	int frame_offset;
+	unsigned long func_size;
+	int frame_size;
 	int pc_offset;
-} schedule_frame, mfinfo[] = {
-	{ schedule, 0 },	/* must be first */
-	/* arch/mips/kernel/semaphore.c */
-	{ __down, 1 },
-	{ __down_interruptible, 1 },
-	/* kernel/sched.c */
-#ifdef CONFIG_PREEMPT
-	{ preempt_schedule, 0 },
-#endif
-	{ wait_for_completion, 0 },
-	{ interruptible_sleep_on, 0 },
-	{ interruptible_sleep_on_timeout, 0 },
-	{ sleep_on, 0 },
-	{ sleep_on_timeout, 0 },
-	{ yield, 0 },
-	{ io_schedule, 0 },
-	{ io_schedule_timeout, 0 },
-#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT)
-	{ __preempt_spin_lock, 0 },
-	{ __preempt_write_lock, 0 },
-#endif
-	/* kernel/timer.c */
-	{ schedule_timeout, 1 },
-/*	{ nanosleep_restart, 1 }, */
-	/* lib/rwsem-spinlock.c */
-	{ __down_read, 1 },
-	{ __down_write, 1 },
-};
+} *schedule_frame, mfinfo[64];
+static int mfinfo_num;
 
-static int mips_frame_info_initialized;
 static int __init get_frame_info(struct mips_frame_info *info)
 {
 	int i;
 	void *func = info->func;
 	union mips_instruction *ip = (union mips_instruction *)func;
 	info->pc_offset = -1;
-	info->frame_offset = info->omit_fp ? 0 : -1;
+	info->frame_size = 0;
 	for (i = 0; i < 128; i++, ip++) {
 		/* if jal, jalr, jr, stop. */
 		if (ip->j_format.opcode == jal_op ||
@@ -320,6 +294,23 @@ static int __init get_frame_info(struct mips_frame_info *info)
 		      ip->r_format.func == jr_op)))
 			break;
 
+		if (info->func_size && i >= info->func_size / 4)
+			break;
+		if (
+#ifdef CONFIG_32BIT
+		    ip->i_format.opcode == addiu_op &&
+#endif
+#ifdef CONFIG_64BIT
+		    ip->i_format.opcode == daddiu_op &&
+#endif
+		    ip->i_format.rs == 29 &&
+		    ip->i_format.rt == 29) {
+			/* addiu/daddiu sp,sp,-imm */
+			if (info->frame_size)
+				continue;
+			info->frame_size = - ip->i_format.simmediate;
+		}
+
 		if (
 #ifdef CONFIG_32BIT
 		    ip->i_format.opcode == sw_op &&
@@ -327,31 +318,20 @@ static int __init get_frame_info(struct mips_frame_info *info)
 #ifdef CONFIG_64BIT
 		    ip->i_format.opcode == sd_op &&
 #endif
-		    ip->i_format.rs == 29)
-		{
+		    ip->i_format.rs == 29 &&
+		    ip->i_format.rt == 31) {
 			/* sw / sd $ra, offset($sp) */
-			if (ip->i_format.rt == 31) {
-				if (info->pc_offset != -1)
-					continue;
-				info->pc_offset =
-					ip->i_format.simmediate / sizeof(long);
-			}
-			/* sw / sd $s8, offset($sp) */
-			if (ip->i_format.rt == 30) {
-//#if 0	/* gcc 3.4 does aggressive optimization... */
-				if (info->frame_offset != -1)
-					continue;
-//#endif
-				info->frame_offset =
-					ip->i_format.simmediate / sizeof(long);
-			}
+			if (info->pc_offset != -1)
+				continue;
+			info->pc_offset =
+				ip->i_format.simmediate / sizeof(long);
 		}
 	}
-	if (info->pc_offset == -1 || info->frame_offset == -1) {
-		printk("Can't analyze prologue code at %p\n", func);
+	if (info->pc_offset == -1 || info->frame_size == 0) {
+		if (func == schedule)
+			printk("Can't analyze prologue code at %p\n", func);
 		info->pc_offset = -1;
-		info->frame_offset = -1;
-		return -1;
+		info->frame_size = 0;
 	}
 
 	return 0;
@@ -359,25 +339,36 @@ static int __init get_frame_info(struct mips_frame_info *info)
 
 static int __init frame_info_init(void)
 {
-	int i, found;
-	for (i = 0; i < ARRAY_SIZE(mfinfo); i++)
-		if (get_frame_info(&mfinfo[i]))
-			return -1;
-	schedule_frame = mfinfo[0];
-	/* bubble sort */
-	do {
-		struct mips_frame_info tmp;
-		found = 0;
-		for (i = 1; i < ARRAY_SIZE(mfinfo); i++) {
-			if (mfinfo[i-1].func > mfinfo[i].func) {
-				tmp = mfinfo[i];
-				mfinfo[i] = mfinfo[i-1];
-				mfinfo[i-1] = tmp;
-				found = 1;
-			}
-		}
-	} while (found);
-	mips_frame_info_initialized = 1;
+	int i;
+#ifdef CONFIG_KALLSYMS
+	char *modname;
+	char namebuf[KSYM_NAME_LEN + 1];
+	unsigned long start, size, ofs;
+	extern char __sched_text_start[], __sched_text_end[];
+	extern char __lock_text_start[], __lock_text_end[];
+
+	start = (unsigned long)__sched_text_start;
+	for (i = 0; i < ARRAY_SIZE(mfinfo); i++) {
+		if (start == (unsigned long)schedule)
+			schedule_frame = &mfinfo[i];
+		if (!kallsyms_lookup(start, &size, &ofs, &modname, namebuf))
+			break;
+		mfinfo[i].func = (void *)(start + ofs);
+		mfinfo[i].func_size = size;
+		start += size - ofs;
+		if (start >= (unsigned long)__lock_text_end)
+			break;
+		if (start == (unsigned long)__sched_text_end)
+			start = (unsigned long)__lock_text_start;
+	}
+#else
+	mfinfo[0].func = schedule;
+	schedule_frame = &mfinfo[0];
+#endif
+	for (i = 0; i < ARRAY_SIZE(mfinfo) && mfinfo[i].func; i++)
+		get_frame_info(&mfinfo[i]);
+
+	mfinfo_num = i;
 	return 0;
 }
 
@@ -394,47 +385,52 @@ unsigned long thread_saved_pc(struct task_struct *tsk)
 	if (t->reg31 == (unsigned long) ret_from_fork)
 		return t->reg31;
 
-	if (schedule_frame.pc_offset < 0)
+	if (!schedule_frame || schedule_frame->pc_offset < 0)
 		return 0;
-	return ((unsigned long *)t->reg29)[schedule_frame.pc_offset];
+	return ((unsigned long *)t->reg29)[schedule_frame->pc_offset];
 }
 
 /* get_wchan - a maintenance nightmare^W^Wpain in the ass ...  */
 unsigned long get_wchan(struct task_struct *p)
 {
 	unsigned long stack_page;
-	unsigned long frame, pc;
+	unsigned long pc;
+#ifdef CONFIG_KALLSYMS
+	unsigned long frame;
+#endif
 
 	if (!p || p == current || p->state == TASK_RUNNING)
 		return 0;
 
 	stack_page = (unsigned long)task_stack_page(p);
-	if (!stack_page || !mips_frame_info_initialized)
+	if (!stack_page || !mfinfo_num)
 		return 0;
 
 	pc = thread_saved_pc(p);
+#ifdef CONFIG_KALLSYMS
 	if (!in_sched_functions(pc))
 		return pc;
 
-	frame = ((unsigned long *)p->thread.reg30)[schedule_frame.frame_offset];
+	frame = p->thread.reg29 + schedule_frame->frame_size;
 	do {
 		int i;
 
 		if (frame < stack_page || frame > stack_page + THREAD_SIZE - 32)
 			return 0;
 
-		for (i = ARRAY_SIZE(mfinfo) - 1; i >= 0; i--) {
+		for (i = mfinfo_num - 1; i >= 0; i--) {
 			if (pc >= (unsigned long) mfinfo[i].func)
 				break;
 		}
 		if (i < 0)
 			break;
 
-		if (mfinfo[i].omit_fp)
-			break;
 		pc = ((unsigned long *)frame)[mfinfo[i].pc_offset];
-		frame = ((unsigned long *)frame)[mfinfo[i].frame_offset];
+		if (!mfinfo[i].frame_size)
+			break;
+		frame += mfinfo[i].frame_size;
 	} while (in_sched_functions(pc));
+#endif
 
 	return pc;
 }

From 41700e73995d6c814932cb55e12525bd34be1ca5 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Fri, 10 Feb 2006 00:39:06 +0900
Subject: [PATCH 162/509] [MIPS] Add protected_blast_icache_range,
 blast_icache_range, etc.

Add blast_xxx_range(), protected_blast_xxx_range() etc. for common
use.  They are built by __BUILD_BLAST_CACHE_RANGE().
Use protected_cache_op() macro for various protected_ routines.
Output code should be logically same.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mm/c-r4k.c        | 104 +++++-------------------------------
 arch/mips/mm/c-tx39.c       |  70 ++++--------------------
 include/asm-mips/r4kcache.h |  74 +++++++++++++------------
 3 files changed, 64 insertions(+), 184 deletions(-)

diff --git a/arch/mips/mm/c-r4k.c b/arch/mips/mm/c-r4k.c
index e51c38cef88e..1b71d91e8268 100644
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -471,61 +471,29 @@ struct flush_icache_range_args {
 static inline void local_r4k_flush_icache_range(void *args)
 {
 	struct flush_icache_range_args *fir_args = args;
-	unsigned long dc_lsize = cpu_dcache_line_size();
-	unsigned long ic_lsize = cpu_icache_line_size();
-	unsigned long sc_lsize = cpu_scache_line_size();
 	unsigned long start = fir_args->start;
 	unsigned long end = fir_args->end;
-	unsigned long addr, aend;
 
 	if (!cpu_has_ic_fills_f_dc) {
 		if (end - start > dcache_size) {
 			r4k_blast_dcache();
 		} else {
 			R4600_HIT_CACHEOP_WAR_IMPL;
-			addr = start & ~(dc_lsize - 1);
-			aend = (end - 1) & ~(dc_lsize - 1);
-
-			while (1) {
-				/* Hit_Writeback_Inv_D */
-				protected_writeback_dcache_line(addr);
-				if (addr == aend)
-					break;
-				addr += dc_lsize;
-			}
+			protected_blast_dcache_range(start, end);
 		}
 
 		if (!cpu_icache_snoops_remote_store) {
-			if (end - start > scache_size) {
+			if (end - start > scache_size)
 				r4k_blast_scache();
-			} else {
-				addr = start & ~(sc_lsize - 1);
-				aend = (end - 1) & ~(sc_lsize - 1);
-
-				while (1) {
-					/* Hit_Writeback_Inv_SD */
-					protected_writeback_scache_line(addr);
-					if (addr == aend)
-						break;
-					addr += sc_lsize;
-				}
-			}
+			else
+				protected_blast_scache_range(start, end);
 		}
 	}
 
 	if (end - start > icache_size)
 		r4k_blast_icache();
-	else {
-		addr = start & ~(ic_lsize - 1);
-		aend = (end - 1) & ~(ic_lsize - 1);
-		while (1) {
-			/* Hit_Invalidate_I */
-			protected_flush_icache_line(addr);
-			if (addr == aend)
-				break;
-			addr += ic_lsize;
-		}
-	}
+	else
+		protected_blast_icache_range(start, end);
 }
 
 static void r4k_flush_icache_range(unsigned long start, unsigned long end)
@@ -619,27 +587,14 @@ static void r4k_flush_icache_page(struct vm_area_struct *vma,
 
 static void r4k_dma_cache_wback_inv(unsigned long addr, unsigned long size)
 {
-	unsigned long end, a;
-
 	/* Catch bad driver code */
 	BUG_ON(size == 0);
 
 	if (cpu_has_subset_pcaches) {
-		unsigned long sc_lsize = cpu_scache_line_size();
-
-		if (size >= scache_size) {
+		if (size >= scache_size)
 			r4k_blast_scache();
-			return;
-		}
-
-		a = addr & ~(sc_lsize - 1);
-		end = (addr + size - 1) & ~(sc_lsize - 1);
-		while (1) {
-			flush_scache_line(a);	/* Hit_Writeback_Inv_SD */
-			if (a == end)
-				break;
-			a += sc_lsize;
-		}
+		else
+			blast_scache_range(addr, addr + size);
 		return;
 	}
 
@@ -651,17 +606,8 @@ static void r4k_dma_cache_wback_inv(unsigned long addr, unsigned long size)
 	if (size >= dcache_size) {
 		r4k_blast_dcache();
 	} else {
-		unsigned long dc_lsize = cpu_dcache_line_size();
-
 		R4600_HIT_CACHEOP_WAR_IMPL;
-		a = addr & ~(dc_lsize - 1);
-		end = (addr + size - 1) & ~(dc_lsize - 1);
-		while (1) {
-			flush_dcache_line(a);	/* Hit_Writeback_Inv_D */
-			if (a == end)
-				break;
-			a += dc_lsize;
-		}
+		blast_dcache_range(addr, addr + size);
 	}
 
 	bc_wback_inv(addr, size);
@@ -669,44 +615,22 @@ static void r4k_dma_cache_wback_inv(unsigned long addr, unsigned long size)
 
 static void r4k_dma_cache_inv(unsigned long addr, unsigned long size)
 {
-	unsigned long end, a;
-
 	/* Catch bad driver code */
 	BUG_ON(size == 0);
 
 	if (cpu_has_subset_pcaches) {
-		unsigned long sc_lsize = cpu_scache_line_size();
-
-		if (size >= scache_size) {
+		if (size >= scache_size)
 			r4k_blast_scache();
-			return;
-		}
-
-		a = addr & ~(sc_lsize - 1);
-		end = (addr + size - 1) & ~(sc_lsize - 1);
-		while (1) {
-			flush_scache_line(a);	/* Hit_Writeback_Inv_SD */
-			if (a == end)
-				break;
-			a += sc_lsize;
-		}
+		else
+			blast_scache_range(addr, addr + size);
 		return;
 	}
 
 	if (size >= dcache_size) {
 		r4k_blast_dcache();
 	} else {
-		unsigned long dc_lsize = cpu_dcache_line_size();
-
 		R4600_HIT_CACHEOP_WAR_IMPL;
-		a = addr & ~(dc_lsize - 1);
-		end = (addr + size - 1) & ~(dc_lsize - 1);
-		while (1) {
-			flush_dcache_line(a);	/* Hit_Writeback_Inv_D */
-			if (a == end)
-				break;
-			a += dc_lsize;
-		}
+		blast_dcache_range(addr, addr + size);
 	}
 
 	bc_inv(addr, size);
diff --git a/arch/mips/mm/c-tx39.c b/arch/mips/mm/c-tx39.c
index 0a97a9434eba..7c572bea4a98 100644
--- a/arch/mips/mm/c-tx39.c
+++ b/arch/mips/mm/c-tx39.c
@@ -44,8 +44,6 @@ __asm__ __volatile__( \
 /* TX39H-style cache flush routines. */
 static void tx39h_flush_icache_all(void)
 {
-	unsigned long start = KSEG0;
-	unsigned long end = (start + icache_size);
 	unsigned long flags, config;
 
 	/* disable icache (set ICE#) */
@@ -53,33 +51,18 @@ static void tx39h_flush_icache_all(void)
 	config = read_c0_conf();
 	write_c0_conf(config & ~TX39_CONF_ICE);
 	TX39_STOP_STREAMING();
-
-	/* invalidate icache */
-	while (start < end) {
-		cache16_unroll32(start, Index_Invalidate_I);
-		start += 0x200;
-	}
-
+	blast_icache16();
 	write_c0_conf(config);
 	local_irq_restore(flags);
 }
 
 static void tx39h_dma_cache_wback_inv(unsigned long addr, unsigned long size)
 {
-	unsigned long end, a;
-	unsigned long dc_lsize = current_cpu_data.dcache.linesz;
-
 	/* Catch bad driver code */
 	BUG_ON(size == 0);
 
 	iob();
-	a = addr & ~(dc_lsize - 1);
-	end = (addr + size - 1) & ~(dc_lsize - 1);
-	while (1) {
-		invalidate_dcache_line(a); /* Hit_Invalidate_D */
-		if (a == end) break;
-		a += dc_lsize;
-	}
+	blast_inv_dcache_range(addr, addr + size);
 }
 
 
@@ -241,42 +224,21 @@ static void tx39_flush_data_cache_page(unsigned long addr)
 
 static void tx39_flush_icache_range(unsigned long start, unsigned long end)
 {
-	unsigned long dc_lsize = current_cpu_data.dcache.linesz;
-	unsigned long addr, aend;
-
 	if (end - start > dcache_size)
 		tx39_blast_dcache();
-	else {
-		addr = start & ~(dc_lsize - 1);
-		aend = (end - 1) & ~(dc_lsize - 1);
-
-		while (1) {
-			/* Hit_Writeback_Inv_D */
-			protected_writeback_dcache_line(addr);
-			if (addr == aend)
-				break;
-			addr += dc_lsize;
-		}
-	}
+	else
+		protected_blast_dcache_range(start, end);
 
 	if (end - start > icache_size)
 		tx39_blast_icache();
 	else {
 		unsigned long flags, config;
-		addr = start & ~(dc_lsize - 1);
-		aend = (end - 1) & ~(dc_lsize - 1);
 		/* disable icache (set ICE#) */
 		local_irq_save(flags);
 		config = read_c0_conf();
 		write_c0_conf(config & ~TX39_CONF_ICE);
 		TX39_STOP_STREAMING();
-		while (1) {
-			/* Hit_Invalidate_I */
-			protected_flush_icache_line(addr);
-			if (addr == aend)
-				break;
-			addr += dc_lsize;
-		}
+		protected_blast_icache_range(start, end);
 		write_c0_conf(config);
 		local_irq_restore(flags);
 	}
@@ -311,7 +273,7 @@ static void tx39_flush_icache_page(struct vm_area_struct *vma, struct page *page
 
 static void tx39_dma_cache_wback_inv(unsigned long addr, unsigned long size)
 {
-	unsigned long end, a;
+	unsigned long end;
 
 	if (((size | addr) & (PAGE_SIZE - 1)) == 0) {
 		end = addr + size;
@@ -322,20 +284,13 @@ static void tx39_dma_cache_wback_inv(unsigned long addr, unsigned long size)
 	} else if (size > dcache_size) {
 		tx39_blast_dcache();
 	} else {
-		unsigned long dc_lsize = current_cpu_data.dcache.linesz;
-		a = addr & ~(dc_lsize - 1);
-		end = (addr + size - 1) & ~(dc_lsize - 1);
-		while (1) {
-			flush_dcache_line(a); /* Hit_Writeback_Inv_D */
-			if (a == end) break;
-			a += dc_lsize;
-		}
+		blast_dcache_range(addr, addr + size);
 	}
 }
 
 static void tx39_dma_cache_inv(unsigned long addr, unsigned long size)
 {
-	unsigned long end, a;
+	unsigned long end;
 
 	if (((size | addr) & (PAGE_SIZE - 1)) == 0) {
 		end = addr + size;
@@ -346,14 +301,7 @@ static void tx39_dma_cache_inv(unsigned long addr, unsigned long size)
 	} else if (size > dcache_size) {
 		tx39_blast_dcache();
 	} else {
-		unsigned long dc_lsize = current_cpu_data.dcache.linesz;
-		a = addr & ~(dc_lsize - 1);
-		end = (addr + size - 1) & ~(dc_lsize - 1);
-		while (1) {
-			invalidate_dcache_line(a); /* Hit_Invalidate_D */
-			if (a == end) break;
-			a += dc_lsize;
-		}
+		blast_inv_dcache_range(addr, addr + size);
 	}
 }
 
diff --git a/include/asm-mips/r4kcache.h b/include/asm-mips/r4kcache.h
index cc53196efa40..9632c27dad15 100644
--- a/include/asm-mips/r4kcache.h
+++ b/include/asm-mips/r4kcache.h
@@ -14,6 +14,7 @@
 
 #include <asm/asm.h>
 #include <asm/cacheops.h>
+#include <asm/cpu-features.h>
 
 /*
  * This macro return a properly sign-extended address suitable as base address
@@ -78,22 +79,25 @@ static inline void flush_scache_line(unsigned long addr)
 	cache_op(Hit_Writeback_Inv_SD, addr);
 }
 
+#define protected_cache_op(op,addr)				\
+	__asm__ __volatile__(					\
+	"	.set	push			\n"		\
+	"	.set	noreorder		\n"		\
+	"	.set	mips3			\n"		\
+	"1:	cache	%0, (%1)		\n"		\
+	"2:	.set	pop			\n"		\
+	"	.section __ex_table,\"a\"	\n"		\
+	"	"STR(PTR)" 1b, 2b		\n"		\
+	"	.previous"					\
+	:							\
+	: "i" (op), "r" (addr))
+
 /*
  * The next two are for badland addresses like signal trampolines.
  */
 static inline void protected_flush_icache_line(unsigned long addr)
 {
-	__asm__ __volatile__(
-		"	.set	push			\n"
-		"	.set	noreorder		\n"
-		"	.set	mips3			\n"
-		"1:	cache	%0, (%1)		\n"
-		"2:	.set	pop			\n"
-		"	.section __ex_table,\"a\"	\n"
-		"	"STR(PTR)" 1b, 2b		\n"
-		"	.previous"
-		:
-		: "i" (Hit_Invalidate_I), "r" (addr));
+	protected_cache_op(Hit_Invalidate_I, addr);
 }
 
 /*
@@ -104,32 +108,12 @@ static inline void protected_flush_icache_line(unsigned long addr)
  */
 static inline void protected_writeback_dcache_line(unsigned long addr)
 {
-	__asm__ __volatile__(
-		"	.set	push			\n"
-		"	.set	noreorder		\n"
-		"	.set	mips3			\n"
-		"1:	cache	%0, (%1)		\n"
-		"2:	.set	pop			\n"
-		"	.section __ex_table,\"a\"	\n"
-		"	"STR(PTR)" 1b, 2b		\n"
-		"	.previous"
-		:
-		: "i" (Hit_Writeback_Inv_D), "r" (addr));
+	protected_cache_op(Hit_Writeback_Inv_D, addr);
 }
 
 static inline void protected_writeback_scache_line(unsigned long addr)
 {
-	__asm__ __volatile__(
-		"	.set	push			\n"
-		"	.set	noreorder		\n"
-		"	.set	mips3			\n"
-		"1:	cache	%0, (%1)		\n"
-		"2:	.set	pop			\n"
-		"	.section __ex_table,\"a\"	\n"
-		"	"STR(PTR)" 1b, 2b		\n"
-		"	.previous"
-		:
-		: "i" (Hit_Writeback_Inv_SD), "r" (addr));
+	protected_cache_op(Hit_Writeback_Inv_SD, addr);
 }
 
 /*
@@ -295,4 +279,28 @@ __BUILD_BLAST_CACHE(i, icache, Index_Invalidate_I, Hit_Invalidate_I, 64)
 __BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 64)
 __BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 128)
 
+/* build blast_xxx_range, protected_blast_xxx_range */
+#define __BUILD_BLAST_CACHE_RANGE(pfx, desc, hitop, prot) \
+static inline void prot##blast_##pfx##cache##_range(unsigned long start, \
+						    unsigned long end)	\
+{									\
+	unsigned long lsize = cpu_##desc##_line_size();			\
+	unsigned long addr = start & ~(lsize - 1);			\
+	unsigned long aend = (end - 1) & ~(lsize - 1);			\
+	while (1) {							\
+		prot##cache_op(hitop, addr);				\
+		if (addr == aend)					\
+			break;						\
+		addr += lsize;						\
+	}								\
+}
+
+__BUILD_BLAST_CACHE_RANGE(d, dcache, Hit_Writeback_Inv_D, protected_)
+__BUILD_BLAST_CACHE_RANGE(s, scache, Hit_Writeback_Inv_SD, protected_)
+__BUILD_BLAST_CACHE_RANGE(i, icache, Hit_Invalidate_I, protected_)
+__BUILD_BLAST_CACHE_RANGE(d, dcache, Hit_Writeback_Inv_D, )
+__BUILD_BLAST_CACHE_RANGE(s, scache, Hit_Writeback_Inv_SD, )
+/* blast_inv_dcache_range */
+__BUILD_BLAST_CACHE_RANGE(inv_d, dcache, Hit_Invalidate_D, )
+
 #endif /* _ASM_R4KCACHE_H */

From 3218357c94af92478ef39163163a81e654385320 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 10 Feb 2006 01:31:24 +0000
Subject: [PATCH 163/509] [MIPS] More uaccess.h fixes with gcc >= 4.0.1.

From Richard Sandiford <richard@codesourcery.com>:

This patch caused a miscompilation of the restore_gp_regs() block
in restore_sigcontext().  This was in a 32-bit kernel compiled with
GCC CVS head.

restore_gp_regs() copies 64-bit user fields into 32-bit variables,
and in this combination, the new __get_user_asm_ll32() clobbers too
many registers.  It says:

/*
 * Get a long long 64 using 32 bit registers.
 */
{									\
	__asm__ __volatile__(						\
	"1:	lw	%1, (%3)				\n"	\
	"2:	lw	%D1, 4(%3)				\n"	\
	"	move	%0, $0					\n"	\
	"3:	.section	.fixup,\"ax\"			\n"	\
	"4:	li	%0, %4					\n"	\
	"	move	%1, $0					\n"	\
	"	move	%D1, $0					\n"	\
	"	j	3b					\n"	\
	"	.previous					\n"	\
	"	.section	__ex_table,\"a\"		\n"	\
	"	" __UA_ADDR "	1b, 4b				\n"	\
	"	" __UA_ADDR "	2b, 4b				\n"	\
	"	.previous					\n"	\
	: "=r" (__gu_err), "=&r" (val)					\
	: "0" (0), "r" (addr), "i" (-EFAULT));				\
}

and this requires val (%1) to be a 64-bit value.  In the case I saw,
gcc was using $3 for the 32-bit val, and wasn't expecting $4 to be
clobbered.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/uaccess.h | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/include/asm-mips/uaccess.h b/include/asm-mips/uaccess.h
index 91d813a37823..7a553e9d44d3 100644
--- a/include/asm-mips/uaccess.h
+++ b/include/asm-mips/uaccess.h
@@ -266,6 +266,8 @@ do {									\
  */
 #define __get_user_asm_ll32(val, addr)					\
 {									\
+        unsigned long long __gu_tmp;					\
+									\
 	__asm__ __volatile__(						\
 	"1:	lw	%1, (%3)				\n"	\
 	"2:	lw	%D1, 4(%3)				\n"	\
@@ -280,8 +282,9 @@ do {									\
 	"	" __UA_ADDR "	1b, 4b				\n"	\
 	"	" __UA_ADDR "	2b, 4b				\n"	\
 	"	.previous					\n"	\
-	: "=r" (__gu_err), "=&r" (val)					\
+	: "=r" (__gu_err), "=&r" (__gu_tmp)				\
 	: "0" (0), "r" (addr), "i" (-EFAULT));				\
+	(val) = __gu_tmp;						\
 }
 
 /*

From fbb6b3a4ac0ccf12a97c98881d9d873d6dc26fe5 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 10 Feb 2006 14:13:08 +0000
Subject: [PATCH 164/509] [MIPS] Get rid of kludgery needed to keep stdargs of
 old compilers working.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/Makefile             |  1 -
 include/asm-mips/gcc/sgidefs.h | 17 -----------------
 2 files changed, 18 deletions(-)
 delete mode 100644 include/asm-mips/gcc/sgidefs.h

diff --git a/arch/mips/Makefile b/arch/mips/Makefile
index 6a57407df1bc..38c0f3360d51 100644
--- a/arch/mips/Makefile
+++ b/arch/mips/Makefile
@@ -94,7 +94,6 @@ endif
 # machines may also.  Since BFD is incredibly buggy with respect to
 # crossformat linking we rely on the elf2ecoff tool for format conversion.
 #
-cflags-y			+= -I $(TOPDIR)/include/asm/gcc
 cflags-y			+= -G 0 -mno-abicalls -fno-pic -pipe
 LDFLAGS_vmlinux			+= -G 0 -static -n -nostdlib
 MODFLAGS			+= -mlong-calls
diff --git a/include/asm-mips/gcc/sgidefs.h b/include/asm-mips/gcc/sgidefs.h
deleted file mode 100644
index 05994371a2af..000000000000
--- a/include/asm-mips/gcc/sgidefs.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * include/sgidefs.h
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1996 by Ralf Baechle
- *
- * This file is here to satisfy GCC's expectations.
- */
-#ifndef __SGIDEFS_H
-#define __SGIDEFS_H
-
-#include <asm/sgidefs.h>
-
-#endif /* __SGIDEFS_H */

From fdc9bb16d3e16f674e52aa69306df5732d7f951b Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 10 Feb 2006 14:25:16 +0000
Subject: [PATCH 165/509] [MIPS] MT: Fix c0 back-to-back hazard.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/smp_mt.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/mips/kernel/smp_mt.c b/arch/mips/kernel/smp_mt.c
index 186f5deced40..184fd464686f 100644
--- a/arch/mips/kernel/smp_mt.c
+++ b/arch/mips/kernel/smp_mt.c
@@ -68,6 +68,8 @@ void __init sanitize_tlb_entries(void)
 
 	set_c0_mvpcontrol(MVPCONTROL_VPC);
 
+	back_to_back_c0_hazard();
+
 	/* Disable TLB sharing */
 	clear_c0_mvpcontrol(MVPCONTROL_STLB);
 

From 74a96d943ac96ce607a841a74ff0f84eeb0c3343 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Fri, 10 Feb 2006 16:03:47 +0000
Subject: [PATCH 166/509] [MIPS] MT: Propagate config7 into VPE.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/smp_mt.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/mips/kernel/smp_mt.c b/arch/mips/kernel/smp_mt.c
index 184fd464686f..c930364830d0 100644
--- a/arch/mips/kernel/smp_mt.c
+++ b/arch/mips/kernel/smp_mt.c
@@ -195,6 +195,9 @@ void prom_prepare_cpus(unsigned int max_cpus)
 
 				/* set config to be the same as vpe0, particularly kseg0 coherency alg */
 				write_vpe_c0_config( read_c0_config());
+
+				/* Propagate Config7 */
+				write_vpe_c0_config7(read_c0_config7());
 			}
 
 		}

From 387a154d0db113690ce85185a003e39e7c87009e Mon Sep 17 00:00:00 2001
From: Thomas Koeller <thomas.koeller@baslerweb.com>
Date: Fri, 10 Feb 2006 17:36:27 +0100
Subject: [PATCH 167/509] [MIPS] RM9000: Fix buggy I-cache workaround.

Signed-off-by: Thomas Koeller  <thomas.koeller@baslerweb.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/signal-common.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/kernel/signal-common.h b/arch/mips/kernel/signal-common.h
index 0fbc492d24b4..36bfc2588aa3 100644
--- a/arch/mips/kernel/signal-common.h
+++ b/arch/mips/kernel/signal-common.h
@@ -176,7 +176,7 @@ get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size)
 	if ((ka->sa.sa_flags & SA_ONSTACK) && (sas_ss_flags (sp) == 0))
 		sp = current->sas_ss_sp + current->sas_ss_size;
 
-	return (void __user *)((sp - frame_size) & (ICACHE_REFILLS_WORKAROUND_WAR ? 32 : ALMASK));
+	return (void __user *)((sp - frame_size) & (ICACHE_REFILLS_WORKAROUND_WAR ? ~(cpu_icache_line_size()-1) : ALMASK));
 }
 
 static inline int install_sigtramp(unsigned int __user *tramp,

From 9cf8ff96447f995d5ea18ec9f25dc8dae26501a2 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@linux-mips.org>
Date: Mon, 13 Feb 2006 09:15:49 +0000
Subject: [PATCH 168/509] [MIPS] Fix CPU type bitmasks for MIPS III, IV and V.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/cpu.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/asm-mips/cpu.h b/include/asm-mips/cpu.h
index 934e063e79f1..818b9a97e214 100644
--- a/include/asm-mips/cpu.h
+++ b/include/asm-mips/cpu.h
@@ -204,9 +204,9 @@
  */
 #define MIPS_CPU_ISA_I		0x00000001
 #define MIPS_CPU_ISA_II		0x00000002
-#define MIPS_CPU_ISA_III	0x00000003
-#define MIPS_CPU_ISA_IV		0x00000004
-#define MIPS_CPU_ISA_V		0x00000005
+#define MIPS_CPU_ISA_III	0x00000004
+#define MIPS_CPU_ISA_IV		0x00000008
+#define MIPS_CPU_ISA_V		0x00000010
 #define MIPS_CPU_ISA_M32R1	0x00000020
 #define MIPS_CPU_ISA_M32R2	0x00000040
 #define MIPS_CPU_ISA_M64R1	0x00000080

From 4cbf8767902c578481ff3df366c77d24fe68fd26 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Tue, 14 Feb 2006 22:40:45 +0900
Subject: [PATCH 169/509] [MIPS] Fix typo in _sys32_rt_sigreturn and
 _sysn32_rt_sigreturn.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/signal32.c   | 2 +-
 arch/mips/kernel/signal_n32.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/mips/kernel/signal32.c b/arch/mips/kernel/signal32.c
index da3271e1fdac..8a8b8dd90417 100644
--- a/arch/mips/kernel/signal32.c
+++ b/arch/mips/kernel/signal32.c
@@ -537,7 +537,7 @@ _sys32_rt_sigreturn(nabi_no_regargs struct pt_regs regs)
 	/* The ucontext contains a stack32_t, so we must convert!  */
 	if (__get_user(sp, &frame->rs_uc.uc_stack.ss_sp))
 		goto badframe;
-	st.ss_size = (long) sp;
+	st.ss_sp = (void *)(long) sp;
 	if (__get_user(st.ss_size, &frame->rs_uc.uc_stack.ss_size))
 		goto badframe;
 	if (__get_user(st.ss_flags, &frame->rs_uc.uc_stack.ss_flags))
diff --git a/arch/mips/kernel/signal_n32.c b/arch/mips/kernel/signal_n32.c
index 384fc4a639a4..5a3776096f07 100644
--- a/arch/mips/kernel/signal_n32.c
+++ b/arch/mips/kernel/signal_n32.c
@@ -108,7 +108,7 @@ _sysn32_rt_sigreturn(nabi_no_regargs struct pt_regs regs)
 	/* The ucontext contains a stack32_t, so we must convert!  */
 	if (__get_user(sp, &frame->rs_uc.uc_stack.ss_sp))
 		goto badframe;
-	st.ss_size = (long) sp;
+	st.ss_sp = (void *)(long) sp;
 	if (__get_user(st.ss_size, &frame->rs_uc.uc_stack.ss_size))
 		goto badframe;
 	if (__get_user(st.ss_flags, &frame->rs_uc.uc_stack.ss_flags))

From 69aa234b918c0d9bc4a20cd6d4453aaa3418f457 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Tue, 14 Feb 2006 15:01:11 -0800
Subject: [PATCH 170/509] [IA64] Dont set NR_CPUS for cpu_possible_map when CPU
 hotplug is enabled.

Do not set cpu_possible_map for NR_CPUS when ACPI_CONFIG_HOTPLUG_CPU is set.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/smpboot.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/arch/ia64/kernel/smpboot.c b/arch/ia64/kernel/smpboot.c
index 8f44e7d2df66..b681ef34a86e 100644
--- a/arch/ia64/kernel/smpboot.c
+++ b/arch/ia64/kernel/smpboot.c
@@ -129,7 +129,7 @@ DEFINE_PER_CPU(int, cpu_state);
 /* Bitmasks of currently online, and possible CPUs */
 cpumask_t cpu_online_map;
 EXPORT_SYMBOL(cpu_online_map);
-cpumask_t cpu_possible_map;
+cpumask_t cpu_possible_map = CPU_MASK_NONE;
 EXPORT_SYMBOL(cpu_possible_map);
 
 cpumask_t cpu_core_map[NR_CPUS] __cacheline_aligned;
@@ -506,9 +506,6 @@ smp_build_cpu_map (void)
 
 	for (cpu = 0; cpu < NR_CPUS; cpu++) {
 		ia64_cpu_to_sapicid[cpu] = -1;
-#ifdef CONFIG_HOTPLUG_CPU
-		cpu_set(cpu, cpu_possible_map);
-#endif
 	}
 
 	ia64_cpu_to_sapicid[0] = boot_cpu_id;

From a6b14fa6fdc01ab3519c2729624f808677539b59 Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Tue, 14 Feb 2006 15:01:12 -0800
Subject: [PATCH 171/509] [IA64] Count disabled cpus as potential hot-pluggable
 CPUs

Have a facility to account for potentially hot-pluggable CPUs. ACPI doesnt
give a determinstic method to find hot-pluggable CPUs. Hence we use 2 methods
to assist.

- BIOS can mark potentially hot-pluggable CPUs as disabled in the MADT tables.
- User can specify the number of hot-pluggable CPUs via parameter
  additional_cpus=X

The option is enabled only if ACPI_CONFIG_HOTPLUG_CPU=y which enables the
physical hotplug option. Without which user can still use logical onlining
and offlining of CPUs by enabling CONFIG_HOTPLUG_CPU=y

Adds more bits to cpu_possible_map for potentially hot-pluggable cpus.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/acpi.c  | 56 ++++++++++++++++++++++++++++++++++++++++
 arch/ia64/kernel/setup.c |  4 +++
 include/asm-ia64/acpi.h  |  2 ++
 3 files changed, 62 insertions(+)

diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index d2702c419cf8..34795ede72e0 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -761,6 +761,62 @@ int acpi_map_cpu2node(acpi_handle handle, int cpu, long physid)
 	return (0);
 }
 
+int additional_cpus __initdata = -1;
+
+static __init int setup_additional_cpus(char *s)
+{
+	if (s)
+		additional_cpus = simple_strtol(s, NULL, 0);
+
+	return 0;
+}
+
+early_param("additional_cpus", setup_additional_cpus);
+
+/*
+ * cpu_possible_map should be static, it cannot change as cpu's
+ * are onlined, or offlined. The reason is per-cpu data-structures
+ * are allocated by some modules at init time, and dont expect to
+ * do this dynamically on cpu arrival/departure.
+ * cpu_present_map on the other hand can change dynamically.
+ * In case when cpu_hotplug is not compiled, then we resort to current
+ * behaviour, which is cpu_possible == cpu_present.
+ * - Ashok Raj
+ *
+ * Three ways to find out the number of additional hotplug CPUs:
+ * - If the BIOS specified disabled CPUs in ACPI/mptables use that.
+ * - The user can overwrite it with additional_cpus=NUM
+ * - Otherwise don't reserve additional CPUs.
+ */
+__init void prefill_possible_map(void)
+{
+	int i;
+	int possible, disabled_cpus;
+
+	disabled_cpus = total_cpus - available_cpus;
+ 	if (additional_cpus == -1) {
+ 		if (disabled_cpus > 0) {
+ 			possible = total_cpus;
+			additional_cpus = disabled_cpus;
+		}
+ 		else {
+			possible = available_cpus;
+			additional_cpus = 0;
+		}
+ 	} else {
+		possible = available_cpus + additional_cpus;
+	}
+	if (possible > NR_CPUS)
+		possible = NR_CPUS;
+
+	printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n",
+		possible,
+	        max_t(int, additional_cpus, 0));
+
+	for (i = 0; i < possible; i++)
+		cpu_set(i, cpu_possible_map);
+}
+
 int acpi_map_lsapic(acpi_handle handle, int *pcpu)
 {
 	struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
diff --git a/arch/ia64/kernel/setup.c b/arch/ia64/kernel/setup.c
index 35f7835294a3..3258e09278d0 100644
--- a/arch/ia64/kernel/setup.c
+++ b/arch/ia64/kernel/setup.c
@@ -430,6 +430,7 @@ setup_arch (char **cmdline_p)
 	if (early_console_setup(*cmdline_p) == 0)
 		mark_bsp_online();
 
+	parse_early_param();
 #ifdef CONFIG_ACPI
 	/* Initialize the ACPI boot-time table parser */
 	acpi_table_init();
@@ -688,6 +689,9 @@ void
 setup_per_cpu_areas (void)
 {
 	/* start_kernel() requires this... */
+#ifdef CONFIG_ACPI_HOTPLUG_CPU
+	prefill_possible_map();
+#endif
 }
 
 /*
diff --git a/include/asm-ia64/acpi.h b/include/asm-ia64/acpi.h
index 3a544ffc5008..f7a517654308 100644
--- a/include/asm-ia64/acpi.h
+++ b/include/asm-ia64/acpi.h
@@ -106,6 +106,8 @@ extern unsigned int can_cpei_retarget(void);
 extern unsigned int is_cpu_cpei_target(unsigned int cpu);
 extern void set_cpei_target_cpu(unsigned int cpu);
 extern unsigned int get_cpei_target_cpu(void);
+extern void prefill_possible_map(void);
+extern int additional_cpus;
 
 #ifdef CONFIG_ACPI_NUMA
 /* Proximity bitmap length; _PXM is at most 255 (8 bit)*/

From 61a34937982cace51853b5dc88f86380e31998c0 Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Tue, 14 Feb 2006 13:52:54 -0800
Subject: [PATCH 172/509] [PATCH] pktcdvd: Don't spam the kernel log when
 nothing is wrong

Change some messages that don't indicate an error so that they are only
printed when debugging is enabled.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 4e7dbcc425ff..73999762582e 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -1548,7 +1548,7 @@ static int pkt_good_disc(struct pktcdvd_device *pd, disc_information *di)
 		case 0x12: /* DVD-RAM */
 			return 0;
 		default:
-			printk("pktcdvd: Wrong disc profile (%x)\n", pd->mmc3_profile);
+			VPRINTK("pktcdvd: Wrong disc profile (%x)\n", pd->mmc3_profile);
 			return 1;
 	}
 
@@ -1894,7 +1894,7 @@ static int pkt_open_write(struct pktcdvd_device *pd)
 	unsigned int write_speed, media_write_speed, read_speed;
 
 	if ((ret = pkt_probe_settings(pd))) {
-		DPRINTK("pktcdvd: %s failed probe\n", pd->name);
+		VPRINTK("pktcdvd: %s failed probe\n", pd->name);
 		return -EIO;
 	}
 
@@ -2440,7 +2440,7 @@ static int pkt_ioctl(struct inode *inode, struct file *file, unsigned int cmd, u
 		return blkdev_ioctl(pd->bdev->bd_inode, file, cmd, arg);
 
 	default:
-		printk("pktcdvd: Unknown ioctl for %s (%x)\n", pd->name, cmd);
+		VPRINTK("pktcdvd: Unknown ioctl for %s (%x)\n", pd->name, cmd);
 		return -ENOTTY;
 	}
 

From 01fd9fda2ce462b44bafdac2fe6aacacf23531f2 Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Tue, 14 Feb 2006 13:52:55 -0800
Subject: [PATCH 173/509] [PATCH] pktcdvd: Allow non-writable media to be
 mounted

If opening for write fails, the open method should return -EROFS.  This makes
"mount" try again with a read-only mount, instead of just giving up.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 73999762582e..edf6bf29002e 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -1895,7 +1895,7 @@ static int pkt_open_write(struct pktcdvd_device *pd)
 
 	if ((ret = pkt_probe_settings(pd))) {
 		VPRINTK("pktcdvd: %s failed probe\n", pd->name);
-		return -EIO;
+		return -EROFS;
 	}
 
 	if ((ret = pkt_set_write_settings(pd))) {
@@ -2053,10 +2053,9 @@ static int pkt_open(struct inode *inode, struct file *file)
 			goto out_dec;
 		}
 	} else {
-		if (pkt_open_dev(pd, file->f_mode & FMODE_WRITE)) {
-			ret = -EIO;
+		ret = pkt_open_dev(pd, file->f_mode & FMODE_WRITE);
+		if (ret)
 			goto out_dec;
-		}
 		/*
 		 * needed here as well, since ext2 (among others) may change
 		 * the blocksize at mount time

From 948423e5ccc33bc257384ad4b339214c577bc926 Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Tue, 14 Feb 2006 13:52:56 -0800
Subject: [PATCH 174/509] [PATCH] pktcdvd: Don't unlock the door if the disc is
 in use

Unlocking the door when the disc is in use is obviously not good, because then
it's possible to eject the disc at the wrong time and cause severe disc data
corruption.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index edf6bf29002e..89d8fe014efa 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -2435,7 +2435,8 @@ static int pkt_ioctl(struct inode *inode, struct file *file, unsigned int cmd, u
 		 * The door gets locked when the device is opened, so we
 		 * have to unlock it or else the eject command fails.
 		 */
-		pkt_lock_door(pd, 0);
+		if (pd->refcnt == 1)
+			pkt_lock_door(pd, 0);
 		return blkdev_ioctl(pd->bdev->bd_inode, file, cmd, arg);
 
 	default:

From 7277232374680595cdbc774fd246b206f56db015 Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Tue, 14 Feb 2006 13:52:56 -0800
Subject: [PATCH 175/509] [PATCH] pktcdvd: Reduce stack usage

Reduce stack usage in the pkt_start_write() function.  Even though it's not
currently a real problem, the pages and offsets arrays can be eliminated,
which saves approximately 1000 bytes of stack space.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 43 ++++++++++++++++-------------------------
 1 file changed, 17 insertions(+), 26 deletions(-)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 89d8fe014efa..93e44d0292ab 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -645,7 +645,7 @@ static void pkt_copy_bio_data(struct bio *src_bio, int seg, int offs, struct pag
  * b) The data can be used as cache to avoid read requests if we receive a
  *    new write request for the same zone.
  */
-static void pkt_make_local_copy(struct packet_data *pkt, struct page **pages, int *offsets)
+static void pkt_make_local_copy(struct packet_data *pkt, struct bio_vec *bvec)
 {
 	int f, p, offs;
 
@@ -653,15 +653,15 @@ static void pkt_make_local_copy(struct packet_data *pkt, struct page **pages, in
 	p = 0;
 	offs = 0;
 	for (f = 0; f < pkt->frames; f++) {
-		if (pages[f] != pkt->pages[p]) {
-			void *vfrom = kmap_atomic(pages[f], KM_USER0) + offsets[f];
+		if (bvec[f].bv_page != pkt->pages[p]) {
+			void *vfrom = kmap_atomic(bvec[f].bv_page, KM_USER0) + bvec[f].bv_offset;
 			void *vto = page_address(pkt->pages[p]) + offs;
 			memcpy(vto, vfrom, CD_FRAMESIZE);
 			kunmap_atomic(vfrom, KM_USER0);
-			pages[f] = pkt->pages[p];
-			offsets[f] = offs;
+			bvec[f].bv_page = pkt->pages[p];
+			bvec[f].bv_offset = offs;
 		} else {
-			BUG_ON(offsets[f] != offs);
+			BUG_ON(bvec[f].bv_offset != offs);
 		}
 		offs += CD_FRAMESIZE;
 		if (offs >= PAGE_SIZE) {
@@ -991,18 +991,17 @@ try_next_bio:
 static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt)
 {
 	struct bio *bio;
-	struct page *pages[PACKET_MAX_SIZE];
-	int offsets[PACKET_MAX_SIZE];
 	int f;
 	int frames_write;
+	struct bio_vec *bvec = pkt->w_bio->bi_io_vec;
 
 	for (f = 0; f < pkt->frames; f++) {
-		pages[f] = pkt->pages[(f * CD_FRAMESIZE) / PAGE_SIZE];
-		offsets[f] = (f * CD_FRAMESIZE) % PAGE_SIZE;
+		bvec[f].bv_page = pkt->pages[(f * CD_FRAMESIZE) / PAGE_SIZE];
+		bvec[f].bv_offset = (f * CD_FRAMESIZE) % PAGE_SIZE;
 	}
 
 	/*
-	 * Fill-in pages[] and offsets[] with data from orig_bios.
+	 * Fill-in bvec with data from orig_bios.
 	 */
 	frames_write = 0;
 	spin_lock(&pkt->lock);
@@ -1024,11 +1023,11 @@ static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt)
 			}
 
 			if (src_bvl->bv_len - src_offs >= CD_FRAMESIZE) {
-				pages[f] = src_bvl->bv_page;
-				offsets[f] = src_bvl->bv_offset + src_offs;
+				bvec[f].bv_page = src_bvl->bv_page;
+				bvec[f].bv_offset = src_bvl->bv_offset + src_offs;
 			} else {
 				pkt_copy_bio_data(bio, segment, src_offs,
-						  pages[f], offsets[f]);
+						  bvec[f].bv_page, bvec[f].bv_offset);
 			}
 			src_offs += CD_FRAMESIZE;
 			frames_write++;
@@ -1042,7 +1041,7 @@ static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt)
 	BUG_ON(frames_write != pkt->write_size);
 
 	if (test_bit(PACKET_MERGE_SEGS, &pd->flags) || (pkt->write_size < pkt->frames)) {
-		pkt_make_local_copy(pkt, pages, offsets);
+		pkt_make_local_copy(pkt, bvec);
 		pkt->cache_valid = 1;
 	} else {
 		pkt->cache_valid = 0;
@@ -1055,17 +1054,9 @@ static void pkt_start_write(struct pktcdvd_device *pd, struct packet_data *pkt)
 	pkt->w_bio->bi_bdev = pd->bdev;
 	pkt->w_bio->bi_end_io = pkt_end_io_packet_write;
 	pkt->w_bio->bi_private = pkt;
-	for (f = 0; f < pkt->frames; f++) {
-		if ((f + 1 < pkt->frames) && (pages[f + 1] == pages[f]) &&
-		    (offsets[f + 1] = offsets[f] + CD_FRAMESIZE)) {
-			if (!bio_add_page(pkt->w_bio, pages[f], CD_FRAMESIZE * 2, offsets[f]))
-				BUG();
-			f++;
-		} else {
-			if (!bio_add_page(pkt->w_bio, pages[f], CD_FRAMESIZE, offsets[f]))
-				BUG();
-		}
-	}
+	for (f = 0; f < pkt->frames; f++)
+		if (!bio_add_page(pkt->w_bio, bvec[f].bv_page, CD_FRAMESIZE, bvec[f].bv_offset))
+			BUG();
 	VPRINTK("pktcdvd: vcnt=%d\n", pkt->w_bio->bi_vcnt);
 
 	atomic_set(&pkt->io_wait, 1);

From 41d78ba55037468e6c86c53e3076d1a74841de39 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 14 Feb 2006 13:52:58 -0800
Subject: [PATCH 176/509] [PATCH] compound page: use page[1].lru

If a compound page has its own put_page_testzero destructor (the only current
example is free_huge_page), that is noted in page[1].mapping of the compound
page.  But that's rather a poor place to keep it: functions which call
set_page_dirty_lock after get_user_pages (e.g.  Infiniband's
__ib_umem_release) ought to be checking first, otherwise set_page_dirty is
liable to crash on what's not the address of a struct address_space.

And now I'm about to make that worse: it turns out that every compound page
needs a destructor, so we can no longer rely on hugetlb pages going their own
special way, to avoid further problems of page->mapping reuse.  For example,
not many people know that: on 50% of i386 -Os builds, the first tail page of a
compound page purports to be PageAnon (when its destructor has an odd
address), which surprises page_add_file_rmap.

Keep the compound page destructor in page[1].lru.next instead.  And to free up
the common pairing of mapping and index, also move compound page order from
index to lru.prev.  Slab reuses page->lru too: but if we ever need slab to use
compound pages, it can easily stack its use above this.

(akpm: decoded version of the above: the tail pages of a compound page now
have ->mapping==NULL, so there's no need for the set_page_dirty[_lock]()
caller to check that they're not compund pages before doing the dirty).

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/hugetlb.c    |  4 ++--
 mm/page_alloc.c | 15 ++++++---------
 mm/swap.c       |  2 +-
 3 files changed, 9 insertions(+), 12 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 67f29516662a..508707704d2c 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -85,7 +85,7 @@ void free_huge_page(struct page *page)
 	BUG_ON(page_count(page));
 
 	INIT_LIST_HEAD(&page->lru);
-	page[1].mapping = NULL;
+	page[1].lru.next = NULL;			/* reset dtor */
 
 	spin_lock(&hugetlb_lock);
 	enqueue_huge_page(page);
@@ -105,7 +105,7 @@ struct page *alloc_huge_page(struct vm_area_struct *vma, unsigned long addr)
 	}
 	spin_unlock(&hugetlb_lock);
 	set_page_count(page, 1);
-	page[1].mapping = (void *)free_huge_page;
+	page[1].lru.next = (void *)free_huge_page;	/* set dtor */
 	for (i = 0; i < (HPAGE_SIZE/PAGE_SIZE); ++i)
 		clear_user_highpage(&page[i], addr);
 	return page;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index dde04ff4be31..eec89ab39bb6 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -169,20 +169,17 @@ static void bad_page(struct page *page)
  * All pages have PG_compound set.  All pages have their ->private pointing at
  * the head page (even the head page has this).
  *
- * The first tail page's ->mapping, if non-zero, holds the address of the
- * compound page's put_page() function.
- *
- * The order of the allocation is stored in the first tail page's ->index
- * This is only for debug at present.  This usage means that zero-order pages
- * may not be compound.
+ * The first tail page's ->lru.next holds the address of the compound page's
+ * put_page() function.  Its ->lru.prev holds the order of allocation.
+ * This usage means that zero-order pages may not be compound.
  */
 static void prep_compound_page(struct page *page, unsigned long order)
 {
 	int i;
 	int nr_pages = 1 << order;
 
-	page[1].mapping = NULL;
-	page[1].index = order;
+	page[1].lru.next = NULL;			/* set dtor */
+	page[1].lru.prev = (void *)order;
 	for (i = 0; i < nr_pages; i++) {
 		struct page *p = page + i;
 
@@ -196,7 +193,7 @@ static void destroy_compound_page(struct page *page, unsigned long order)
 	int i;
 	int nr_pages = 1 << order;
 
-	if (unlikely(page[1].index != order))
+	if (unlikely((unsigned long)page[1].lru.prev != order))
 		bad_page(page);
 
 	for (i = 0; i < nr_pages; i++) {
diff --git a/mm/swap.c b/mm/swap.c
index 76247424dea1..cce3dda59c59 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -40,7 +40,7 @@ static void put_compound_page(struct page *page)
 	if (put_page_testzero(page)) {
 		void (*dtor)(struct page *page);
 
-		dtor = (void (*)(struct page *))page[1].mapping;
+		dtor = (void (*)(struct page *))page[1].lru.next;
 		(*dtor)(page);
 	}
 }

From d98c7a09843621f1b145ca5ae8ed03ff04085edb Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 14 Feb 2006 13:52:59 -0800
Subject: [PATCH 177/509] [PATCH] compound page: default destructor

Somehow I imagined that calling a NULL destructor would free a compound page
rather than oopsing.  No, we must supply a default destructor, __free_pages_ok
using the order noted by prep_compound_page.  hugetlb can still replace this
as before with its own free_huge_page pointer.

The case that needs this is not common: rarely does put_compound_page's
put_page_testzero bring the count down to 0.  But if get_user_pages is applied
to some part of a compound page, without immediate release (e.g.  AIO or
Infiniband), then it's possible for its put_page to come after the containing
vma has been unmapped and the driver done its free_pages.

That's just the kind of case compound pages are supposed to be guarding
against (but Nick points out, nor did PageReserved handle this right).

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index eec89ab39bb6..62c122528587 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -56,6 +56,7 @@ long nr_swap_pages;
 int percpu_pagelist_fraction;
 
 static void fastcall free_hot_cold_page(struct page *page, int cold);
+static void __free_pages_ok(struct page *page, unsigned int order);
 
 /*
  * results with 256, 32 in the lowmem_reserve sysctl:
@@ -173,12 +174,18 @@ static void bad_page(struct page *page)
  * put_page() function.  Its ->lru.prev holds the order of allocation.
  * This usage means that zero-order pages may not be compound.
  */
+
+static void free_compound_page(struct page *page)
+{
+	__free_pages_ok(page, (unsigned long)page[1].lru.prev);
+}
+
 static void prep_compound_page(struct page *page, unsigned long order)
 {
 	int i;
 	int nr_pages = 1 << order;
 
-	page[1].lru.next = NULL;			/* set dtor */
+	page[1].lru.next = (void *)free_compound_page;	/* set dtor */
 	page[1].lru.prev = (void *)order;
 	for (i = 0; i < nr_pages; i++) {
 		struct page *p = page + i;

From 16bf134840da3920ded1290973c56ec214636f12 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 14 Feb 2006 13:52:59 -0800
Subject: [PATCH 178/509] [PATCH] compound page: no access_process_vm check

The PageCompound check before access_process_vm's set_page_dirty_lock is no
longer necessary, so remove it.  But leave the PageCompound checks in
bio_set_pages_dirty, dio_bio_complete and nfs_free_user_pages: at least some
of those were introduced as a little optimization on hugetlb pages.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/ptrace.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index 5f33cdb6fff5..d2cf144d0af5 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -242,8 +242,7 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in
 		if (write) {
 			copy_to_user_page(vma, page, addr,
 					  maddr + offset, buf, bytes);
-			if (!PageCompound(page))
-				set_page_dirty_lock(page);
+			set_page_dirty_lock(page);
 		} else {
 			copy_from_user_page(vma, page, addr,
 					    buf, maddr + offset, bytes);

From da965822abd18a17d7cffe1d511f48951c82dfb6 Mon Sep 17 00:00:00 2001
From: Paul Fulghum <paulkf@microgate.com>
Date: Tue, 14 Feb 2006 13:53:00 -0800
Subject: [PATCH 179/509] [PATCH] tty reference count fix

Fix hole where tty structure can be released when reference count is non
zero.  Existing code can sleep without tty_sem protection between deciding
to release the tty structure (setting local variables tty_closing and
otty_closing) and setting TTY_CLOSING to prevent further opens.  An open
can occur during this interval causing release_dev() to free the tty
structure while it is still referenced.

This should fix bugzilla.kernel.org [Bug 6041] New: Unable to handle kernel
paging request

In Bug 6041, tty_open() oopes on accessing the tty structure it has
successfully claimed.  Bug was on SMP machine with the same tty being
opened and closed by multiple processes, and DEBUG_PAGEALLOC enabled.

Signed-off-by: Paul Fulghum <paulkf@microgate.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Jesper Juhl <jesper.juhl@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/tty_io.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index a23816d3e9a1..e9bba94fc898 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -1841,7 +1841,6 @@ static void release_dev(struct file * filp)
 		tty_closing = tty->count <= 1;
 		o_tty_closing = o_tty &&
 			(o_tty->count <= (pty_master ? 1 : 0));
-		up(&tty_sem);
 		do_sleep = 0;
 
 		if (tty_closing) {
@@ -1869,6 +1868,7 @@ static void release_dev(struct file * filp)
 
 		printk(KERN_WARNING "release_dev: %s: read/write wait queue "
 				    "active!\n", tty_name(tty, buf));
+		up(&tty_sem);
 		schedule();
 	}	
 
@@ -1877,8 +1877,6 @@ static void release_dev(struct file * filp)
 	 * both sides, and we've completed the last operation that could 
 	 * block, so it's safe to proceed with closing.
 	 */
-	 
-	down(&tty_sem);
 	if (pty_master) {
 		if (--o_tty->count < 0) {
 			printk(KERN_WARNING "release_dev: bad pty slave count "
@@ -1892,7 +1890,6 @@ static void release_dev(struct file * filp)
 		       tty->count, tty_name(tty, buf));
 		tty->count = 0;
 	}
-	up(&tty_sem);
 	
 	/*
 	 * We've decremented tty->count, so we need to remove this file
@@ -1937,6 +1934,8 @@ static void release_dev(struct file * filp)
 		read_unlock(&tasklist_lock);
 	}
 
+	up(&tty_sem);
+
 	/* check whether both sides are closing ... */
 	if (!tty_closing || (o_tty && !o_tty_closing))
 		return;

From be5efffb762fa4a7b9a7a45ebf34b13e3bf5a2d8 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Tue, 14 Feb 2006 13:53:02 -0800
Subject: [PATCH 180/509] [PATCH] HPET: handle multiple ACPI EXTENDED_IRQ
 resources

When the _CRS for a single HPET contains multiple EXTENDED_IRQ resources,
we overwrote hdp->hd_nirqs every time we found one.

So the driver worked when all the IRQs were described in a single
EXTENDED_IRQ resource, but failed when multiple resources were used.
(Strictly speaking, I think the latter is actually more correct, but both
styles have been used.)

Someday we should remove all the ACPI stuff from hpet.c and use PNP driver
registration instead.  But currently PNP_MAX_IRQ is 2, and HPETs often have
more IRQs.  Hint, hint, Adam :-)

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Bob Picco <robert.picco@hp.com>
Cc: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Cc: Len Brown <len.brown@intel.com>
Cc: Adam Belay <ambx1@neo.rr.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/hpet.c | 20 ++++++++------------
 1 file changed, 8 insertions(+), 12 deletions(-)

diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index 66a2fee06eb9..ef140ebde117 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -956,22 +956,18 @@ static acpi_status hpet_resources(struct acpi_resource *res, void *data)
 		}
 	} else if (res->type == ACPI_RESOURCE_TYPE_EXTENDED_IRQ) {
 		struct acpi_resource_extended_irq *irqp;
-		int i;
+		int i, irq;
 
 		irqp = &res->data.extended_irq;
 
-		if (irqp->interrupt_count > 0) {
-			hdp->hd_nirqs = irqp->interrupt_count;
+		for (i = 0; i < irqp->interrupt_count; i++) {
+			irq = acpi_register_gsi(irqp->interrupts[i],
+				      irqp->triggering, irqp->polarity);
+			if (irq < 0)
+				return AE_ERROR;
 
-			for (i = 0; i < hdp->hd_nirqs; i++) {
-				int rc =
-				    acpi_register_gsi(irqp->interrupts[i],
-						      irqp->triggering,
-						      irqp->polarity);
-				if (rc < 0)
-					return AE_ERROR;
-				hdp->hd_irq[i] = rc;
-			}
+			hdp->hd_irq[hdp->hd_nirqs] = irq;
+			hdp->hd_nirqs++;
 		}
 	}
 

From 7c8903f6373f9abecf060bad53ca36bc4ac037f2 Mon Sep 17 00:00:00 2001
From: Mark Fasheh <mark.fasheh@oracle.com>
Date: Tue, 14 Feb 2006 13:53:03 -0800
Subject: [PATCH 181/509] [PATCH] jbd: revert checkpoint list changes

This patch reverts commit f93ea411b73594f7d144855fd34278bcf34a9afc:
  [PATCH] jbd: split checkpoint lists

This broke journal_flush() for OCFS2, which is its method of being sure
that metadata is sent to disk for another node.

And two related commits 8d3c7fce2d20ecc3264c8d8c91ae3beacdeaed1b and
43c3e6f5abdf6acac9b90c86bf03f995bf7d3d92 with the subjects:
  [PATCH] jbd: log_do_checkpoint fix
  [PATCH] jbd: remove_transaction fix

These seem to be incremental bugfixes on the original patch and as such are
no longer needed.

Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
Cc: Jan Kara <jack@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/jbd/checkpoint.c | 420 +++++++++++++++++++-------------------------
 fs/jbd/commit.c     |   3 +-
 include/linux/jbd.h |   8 +-
 3 files changed, 180 insertions(+), 251 deletions(-)

diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index e6265a0b56b8..543ed543d1e5 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -24,28 +24,7 @@
 #include <linux/slab.h>
 
 /*
- * Unlink a buffer from a transaction checkpoint list.
- *
- * Called with j_list_lock held.
- */
-
-static void __buffer_unlink_first(struct journal_head *jh)
-{
-	transaction_t *transaction;
-
-	transaction = jh->b_cp_transaction;
-
-	jh->b_cpnext->b_cpprev = jh->b_cpprev;
-	jh->b_cpprev->b_cpnext = jh->b_cpnext;
-	if (transaction->t_checkpoint_list == jh) {
-		transaction->t_checkpoint_list = jh->b_cpnext;
-		if (transaction->t_checkpoint_list == jh)
-			transaction->t_checkpoint_list = NULL;
-	}
-}
-
-/*
- * Unlink a buffer from a transaction checkpoint(io) list.
+ * Unlink a buffer from a transaction.
  *
  * Called with j_list_lock held.
  */
@@ -55,44 +34,19 @@ static inline void __buffer_unlink(struct journal_head *jh)
 	transaction_t *transaction;
 
 	transaction = jh->b_cp_transaction;
+	jh->b_cp_transaction = NULL;
 
-	__buffer_unlink_first(jh);
-	if (transaction->t_checkpoint_io_list == jh) {
-		transaction->t_checkpoint_io_list = jh->b_cpnext;
-		if (transaction->t_checkpoint_io_list == jh)
-			transaction->t_checkpoint_io_list = NULL;
-	}
-}
-
-/*
- * Move a buffer from the checkpoint list to the checkpoint io list
- *
- * Called with j_list_lock held
- */
-
-static inline void __buffer_relink_io(struct journal_head *jh)
-{
-	transaction_t *transaction;
-
-	transaction = jh->b_cp_transaction;
-	__buffer_unlink_first(jh);
-
-	if (!transaction->t_checkpoint_io_list) {
-		jh->b_cpnext = jh->b_cpprev = jh;
-	} else {
-		jh->b_cpnext = transaction->t_checkpoint_io_list;
-		jh->b_cpprev = transaction->t_checkpoint_io_list->b_cpprev;
-		jh->b_cpprev->b_cpnext = jh;
-		jh->b_cpnext->b_cpprev = jh;
-	}
-	transaction->t_checkpoint_io_list = jh;
+	jh->b_cpnext->b_cpprev = jh->b_cpprev;
+	jh->b_cpprev->b_cpnext = jh->b_cpnext;
+	if (transaction->t_checkpoint_list == jh)
+		transaction->t_checkpoint_list = jh->b_cpnext;
+	if (transaction->t_checkpoint_list == jh)
+		transaction->t_checkpoint_list = NULL;
 }
 
 /*
  * Try to release a checkpointed buffer from its transaction.
- * Returns 1 if we released it and 2 if we also released the
- * whole transaction.
- *
+ * Returns 1 if we released it.
  * Requires j_list_lock
  * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
  */
@@ -103,11 +57,12 @@ static int __try_to_free_cp_buf(struct journal_head *jh)
 
 	if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh)) {
 		JBUFFER_TRACE(jh, "remove from checkpoint list");
-		ret = __journal_remove_checkpoint(jh) + 1;
+		__journal_remove_checkpoint(jh);
 		jbd_unlock_bh_state(bh);
 		journal_remove_journal_head(bh);
 		BUFFER_TRACE(bh, "release");
 		__brelse(bh);
+		ret = 1;
 	} else {
 		jbd_unlock_bh_state(bh);
 	}
@@ -162,53 +117,83 @@ static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh)
 }
 
 /*
- * Clean up transaction's list of buffers submitted for io.
- * We wait for any pending IO to complete and remove any clean
- * buffers. Note that we take the buffers in the opposite ordering
- * from the one in which they were submitted for IO.
+ * Clean up a transaction's checkpoint list.
+ *
+ * We wait for any pending IO to complete and make sure any clean
+ * buffers are removed from the transaction.
+ *
+ * Return 1 if we performed any actions which might have destroyed the
+ * checkpoint.  (journal_remove_checkpoint() deletes the transaction when
+ * the last checkpoint buffer is cleansed)
  *
  * Called with j_list_lock held.
  */
-
-static void __wait_cp_io(journal_t *journal, transaction_t *transaction)
+static int __cleanup_transaction(journal_t *journal, transaction_t *transaction)
 {
-	struct journal_head *jh;
+	struct journal_head *jh, *next_jh, *last_jh;
 	struct buffer_head *bh;
-	tid_t this_tid;
-	int released = 0;
+	int ret = 0;
 
-	this_tid = transaction->t_tid;
-restart:
-	/* Didn't somebody clean up the transaction in the meanwhile */
-	if (journal->j_checkpoint_transactions != transaction ||
-		transaction->t_tid != this_tid)
-		return;
-	while (!released && transaction->t_checkpoint_io_list) {
-		jh = transaction->t_checkpoint_io_list;
+	assert_spin_locked(&journal->j_list_lock);
+	jh = transaction->t_checkpoint_list;
+	if (!jh)
+		return 0;
+
+	last_jh = jh->b_cpprev;
+	next_jh = jh;
+	do {
+		jh = next_jh;
 		bh = jh2bh(jh);
-		if (!jbd_trylock_bh_state(bh)) {
-			jbd_sync_bh(journal, bh);
-			spin_lock(&journal->j_list_lock);
-			goto restart;
-		}
 		if (buffer_locked(bh)) {
 			atomic_inc(&bh->b_count);
 			spin_unlock(&journal->j_list_lock);
-			jbd_unlock_bh_state(bh);
 			wait_on_buffer(bh);
 			/* the journal_head may have gone by now */
 			BUFFER_TRACE(bh, "brelse");
 			__brelse(bh);
-			spin_lock(&journal->j_list_lock);
-			goto restart;
+			goto out_return_1;
 		}
+
 		/*
-		 * Now in whatever state the buffer currently is, we know that
-		 * it has been written out and so we can drop it from the list
+		 * This is foul
 		 */
-		released = __journal_remove_checkpoint(jh);
-		jbd_unlock_bh_state(bh);
-	}
+		if (!jbd_trylock_bh_state(bh)) {
+			jbd_sync_bh(journal, bh);
+			goto out_return_1;
+		}
+
+		if (jh->b_transaction != NULL) {
+			transaction_t *t = jh->b_transaction;
+			tid_t tid = t->t_tid;
+
+			spin_unlock(&journal->j_list_lock);
+			jbd_unlock_bh_state(bh);
+			log_start_commit(journal, tid);
+			log_wait_commit(journal, tid);
+			goto out_return_1;
+		}
+
+		/*
+		 * AKPM: I think the buffer_jbddirty test is redundant - it
+		 * shouldn't have NULL b_transaction?
+		 */
+		next_jh = jh->b_cpnext;
+		if (!buffer_dirty(bh) && !buffer_jbddirty(bh)) {
+			BUFFER_TRACE(bh, "remove from checkpoint");
+			__journal_remove_checkpoint(jh);
+			jbd_unlock_bh_state(bh);
+			journal_remove_journal_head(bh);
+			__brelse(bh);
+			ret = 1;
+		} else {
+			jbd_unlock_bh_state(bh);
+		}
+	} while (jh != last_jh);
+
+	return ret;
+out_return_1:
+	spin_lock(&journal->j_list_lock);
+	return 1;
 }
 
 #define NR_BATCH	64
@@ -218,7 +203,9 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
 {
 	int i;
 
+	spin_unlock(&journal->j_list_lock);
 	ll_rw_block(SWRITE, *batch_count, bhs);
+	spin_lock(&journal->j_list_lock);
 	for (i = 0; i < *batch_count; i++) {
 		struct buffer_head *bh = bhs[i];
 		clear_buffer_jwrite(bh);
@@ -234,46 +221,19 @@ __flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count)
  * Return 1 if something happened which requires us to abort the current
  * scan of the checkpoint list.  
  *
- * Called with j_list_lock held and drops it if 1 is returned
+ * Called with j_list_lock held.
  * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it
  */
-static int __process_buffer(journal_t *journal, struct journal_head *jh,
-			struct buffer_head **bhs, int *batch_count)
+static int __flush_buffer(journal_t *journal, struct journal_head *jh,
+			struct buffer_head **bhs, int *batch_count,
+			int *drop_count)
 {
 	struct buffer_head *bh = jh2bh(jh);
 	int ret = 0;
 
-	if (buffer_locked(bh)) {
-		get_bh(bh);
-		spin_unlock(&journal->j_list_lock);
-		jbd_unlock_bh_state(bh);
-		wait_on_buffer(bh);
-		/* the journal_head may have gone by now */
-		BUFFER_TRACE(bh, "brelse");
-		put_bh(bh);
-		ret = 1;
-	}
-	else if (jh->b_transaction != NULL) {
-		transaction_t *t = jh->b_transaction;
-		tid_t tid = t->t_tid;
+	if (buffer_dirty(bh) && !buffer_locked(bh) && jh->b_jlist == BJ_None) {
+		J_ASSERT_JH(jh, jh->b_transaction == NULL);
 
-		spin_unlock(&journal->j_list_lock);
-		jbd_unlock_bh_state(bh);
-		log_start_commit(journal, tid);
-		log_wait_commit(journal, tid);
-		ret = 1;
-	}
-	else if (!buffer_dirty(bh)) {
-		J_ASSERT_JH(jh, !buffer_jbddirty(bh));
-		BUFFER_TRACE(bh, "remove from checkpoint");
-		__journal_remove_checkpoint(jh);
-		spin_unlock(&journal->j_list_lock);
-		jbd_unlock_bh_state(bh);
-		journal_remove_journal_head(bh);
-		put_bh(bh);
-		ret = 1;
-	}
-	else {
 		/*
 		 * Important: we are about to write the buffer, and
 		 * possibly block, while still holding the journal lock.
@@ -286,30 +246,45 @@ static int __process_buffer(journal_t *journal, struct journal_head *jh,
 		J_ASSERT_BH(bh, !buffer_jwrite(bh));
 		set_buffer_jwrite(bh);
 		bhs[*batch_count] = bh;
-		__buffer_relink_io(jh);
 		jbd_unlock_bh_state(bh);
 		(*batch_count)++;
 		if (*batch_count == NR_BATCH) {
-			spin_unlock(&journal->j_list_lock);
 			__flush_batch(journal, bhs, batch_count);
 			ret = 1;
 		}
+	} else {
+		int last_buffer = 0;
+		if (jh->b_cpnext == jh) {
+			/* We may be about to drop the transaction.  Tell the
+			 * caller that the lists have changed.
+			 */
+			last_buffer = 1;
+		}
+		if (__try_to_free_cp_buf(jh)) {
+			(*drop_count)++;
+			ret = last_buffer;
+		}
 	}
 	return ret;
 }
 
 /*
- * Perform an actual checkpoint. We take the first transaction on the
- * list of transactions to be checkpointed and send all its buffers
- * to disk. We submit larger chunks of data at once.
+ * Perform an actual checkpoint.  We don't write out only enough to
+ * satisfy the current blocked requests: rather we submit a reasonably
+ * sized chunk of the outstanding data to disk at once for
+ * efficiency.  __log_wait_for_space() will retry if we didn't free enough.
  * 
+ * However, we _do_ take into account the amount requested so that once
+ * the IO has been queued, we can return as soon as enough of it has
+ * completed to disk.
+ *
  * The journal should be locked before calling this function.
  */
 int log_do_checkpoint(journal_t *journal)
 {
-	transaction_t *transaction;
-	tid_t this_tid;
 	int result;
+	int batch_count = 0;
+	struct buffer_head *bhs[NR_BATCH];
 
 	jbd_debug(1, "Start checkpoint\n");
 
@@ -324,70 +299,79 @@ int log_do_checkpoint(journal_t *journal)
 		return result;
 
 	/*
-	 * OK, we need to start writing disk blocks.  Take one transaction
-	 * and write it.
+	 * OK, we need to start writing disk blocks.  Try to free up a
+	 * quarter of the log in a single checkpoint if we can.
+	 */
+	/*
+	 * AKPM: check this code.  I had a feeling a while back that it
+	 * degenerates into a busy loop at unmount time.
 	 */
 	spin_lock(&journal->j_list_lock);
-	if (!journal->j_checkpoint_transactions)
-		goto out;
-	transaction = journal->j_checkpoint_transactions;
-	this_tid = transaction->t_tid;
-restart:
-	/*
-	 * If someone cleaned up this transaction while we slept, we're
-	 * done (maybe it's a new transaction, but it fell at the same
-	 * address).
-	 */
- 	if (journal->j_checkpoint_transactions == transaction &&
-			transaction->t_tid == this_tid) {
-		int batch_count = 0;
-		struct buffer_head *bhs[NR_BATCH];
-		struct journal_head *jh;
-		int retry = 0;
+	while (journal->j_checkpoint_transactions) {
+		transaction_t *transaction;
+		struct journal_head *jh, *last_jh, *next_jh;
+		int drop_count = 0;
+		int cleanup_ret, retry = 0;
+		tid_t this_tid;
 
-		while (!retry && transaction->t_checkpoint_list) {
+		transaction = journal->j_checkpoint_transactions;
+		this_tid = transaction->t_tid;
+		jh = transaction->t_checkpoint_list;
+		last_jh = jh->b_cpprev;
+		next_jh = jh;
+		do {
 			struct buffer_head *bh;
 
-			jh = transaction->t_checkpoint_list;
+			jh = next_jh;
+			next_jh = jh->b_cpnext;
 			bh = jh2bh(jh);
 			if (!jbd_trylock_bh_state(bh)) {
 				jbd_sync_bh(journal, bh);
+				spin_lock(&journal->j_list_lock);
 				retry = 1;
 				break;
 			}
-			retry = __process_buffer(journal, jh, bhs,
-						&batch_count);
-			if (!retry &&
-			    lock_need_resched(&journal->j_list_lock)) {
-				spin_unlock(&journal->j_list_lock);
+			retry = __flush_buffer(journal, jh, bhs, &batch_count, &drop_count);
+			if (cond_resched_lock(&journal->j_list_lock)) {
 				retry = 1;
 				break;
 			}
-		}
+		} while (jh != last_jh && !retry);
 
 		if (batch_count) {
-			if (!retry) {
-				spin_unlock(&journal->j_list_lock);
-				retry = 1;
-			}
 			__flush_batch(journal, bhs, &batch_count);
+			retry = 1;
 		}
 
-		if (retry) {
-			spin_lock(&journal->j_list_lock);
-			goto restart;
-		}
 		/*
-		 * Now we have cleaned up the first transaction's checkpoint
-		 * list.  Let's clean up the second one.
+		 * If someone cleaned up this transaction while we slept, we're
+		 * done
 		 */
-		__wait_cp_io(journal, transaction);
+		if (journal->j_checkpoint_transactions != transaction)
+			break;
+		if (retry)
+			continue;
+		/*
+		 * Maybe it's a new transaction, but it fell at the same
+		 * address
+		 */
+		if (transaction->t_tid != this_tid)
+			continue;
+		/*
+		 * We have walked the whole transaction list without
+		 * finding anything to write to disk.  We had better be
+		 * able to make some progress or we are in trouble.
+		 */
+		cleanup_ret = __cleanup_transaction(journal, transaction);
+		J_ASSERT(drop_count != 0 || cleanup_ret != 0);
+		if (journal->j_checkpoint_transactions != transaction)
+			break;
 	}
-out:
 	spin_unlock(&journal->j_list_lock);
 	result = cleanup_journal_tail(journal);
 	if (result < 0)
 		return result;
+
 	return 0;
 }
 
@@ -471,53 +455,6 @@ int cleanup_journal_tail(journal_t *journal)
 
 /* Checkpoint list management */
 
-/*
- * journal_clean_one_cp_list
- *
- * Find all the written-back checkpoint buffers in the given list and release them.
- *
- * Called with the journal locked.
- * Called with j_list_lock held.
- * Returns number of bufers reaped (for debug)
- */
-
-static int journal_clean_one_cp_list(struct journal_head *jh, int *released)
-{
-	struct journal_head *last_jh;
-	struct journal_head *next_jh = jh;
-	int ret, freed = 0;
-
-	*released = 0;
-	if (!jh)
-		return 0;
-
- 	last_jh = jh->b_cpprev;
-	do {
-		jh = next_jh;
-		next_jh = jh->b_cpnext;
-		/* Use trylock because of the ranking */
-		if (jbd_trylock_bh_state(jh2bh(jh))) {
-			ret = __try_to_free_cp_buf(jh);
-			if (ret) {
-				freed++;
-				if (ret == 2) {
-					*released = 1;
-					return freed;
-				}
-			}
-		}
-		/*
-		 * This function only frees up some memory if possible so we
-		 * dont have an obligation to finish processing. Bail out if
-		 * preemption requested:
-		 */
-		if (need_resched())
-			return freed;
-	} while (jh != last_jh);
-
-	return freed;
-}
-
 /*
  * journal_clean_checkpoint_list
  *
@@ -525,38 +462,46 @@ static int journal_clean_one_cp_list(struct journal_head *jh, int *released)
  *
  * Called with the journal locked.
  * Called with j_list_lock held.
- * Returns number of buffers reaped (for debug)
+ * Returns number of bufers reaped (for debug)
  */
 
 int __journal_clean_checkpoint_list(journal_t *journal)
 {
 	transaction_t *transaction, *last_transaction, *next_transaction;
-	int ret = 0, released;
+	int ret = 0;
 
 	transaction = journal->j_checkpoint_transactions;
-	if (!transaction)
+	if (transaction == 0)
 		goto out;
 
 	last_transaction = transaction->t_cpprev;
 	next_transaction = transaction;
 	do {
+		struct journal_head *jh;
+
 		transaction = next_transaction;
 		next_transaction = transaction->t_cpnext;
-		ret += journal_clean_one_cp_list(transaction->
-				t_checkpoint_list, &released);
-		if (need_resched())
-			goto out;
-		if (released)
-			continue;
-		/*
-		 * It is essential that we are as careful as in the case of
-		 * t_checkpoint_list with removing the buffer from the list as
-		 * we can possibly see not yet submitted buffers on io_list
-		 */
-		ret += journal_clean_one_cp_list(transaction->
-				t_checkpoint_io_list, &released);
-		if (need_resched())
-			goto out;
+		jh = transaction->t_checkpoint_list;
+		if (jh) {
+			struct journal_head *last_jh = jh->b_cpprev;
+			struct journal_head *next_jh = jh;
+
+			do {
+				jh = next_jh;
+				next_jh = jh->b_cpnext;
+				/* Use trylock because of the ranknig */
+				if (jbd_trylock_bh_state(jh2bh(jh)))
+					ret += __try_to_free_cp_buf(jh);
+				/*
+				 * This function only frees up some memory
+				 * if possible so we dont have an obligation
+				 * to finish processing. Bail out if preemption
+				 * requested:
+				 */
+				if (need_resched())
+					goto out;
+			} while (jh != last_jh);
+		}
 	} while (transaction != last_transaction);
 out:
 	return ret;
@@ -571,22 +516,18 @@ out:
  * buffer updates committed in that transaction have safely been stored
  * elsewhere on disk.  To achieve this, all of the buffers in a
  * transaction need to be maintained on the transaction's checkpoint
- * lists until they have been rewritten, at which point this function is
+ * list until they have been rewritten, at which point this function is
  * called to remove the buffer from the existing transaction's
- * checkpoint lists.
- *
- * The function returns 1 if it frees the transaction, 0 otherwise.
+ * checkpoint list.
  *
  * This function is called with the journal locked.
  * This function is called with j_list_lock held.
- * This function is called with jbd_lock_bh_state(jh2bh(jh))
  */
 
-int __journal_remove_checkpoint(struct journal_head *jh)
+void __journal_remove_checkpoint(struct journal_head *jh)
 {
 	transaction_t *transaction;
 	journal_t *journal;
-	int ret = 0;
 
 	JBUFFER_TRACE(jh, "entry");
 
@@ -597,10 +538,8 @@ int __journal_remove_checkpoint(struct journal_head *jh)
 	journal = transaction->t_journal;
 
 	__buffer_unlink(jh);
-	jh->b_cp_transaction = NULL;
 
-	if (transaction->t_checkpoint_list != NULL ||
-	    transaction->t_checkpoint_io_list != NULL)
+	if (transaction->t_checkpoint_list != NULL)
 		goto out;
 	JBUFFER_TRACE(jh, "transaction has no more buffers");
 
@@ -626,10 +565,8 @@ int __journal_remove_checkpoint(struct journal_head *jh)
 	/* Just in case anybody was waiting for more transactions to be
            checkpointed... */
 	wake_up(&journal->j_wait_logspace);
-	ret = 1;
 out:
 	JBUFFER_TRACE(jh, "exit");
-	return ret;
 }
 
 /*
@@ -691,7 +628,6 @@ void __journal_drop_transaction(journal_t *journal, transaction_t *transaction)
 	J_ASSERT(transaction->t_shadow_list == NULL);
 	J_ASSERT(transaction->t_log_list == NULL);
 	J_ASSERT(transaction->t_checkpoint_list == NULL);
-	J_ASSERT(transaction->t_checkpoint_io_list == NULL);
 	J_ASSERT(transaction->t_updates == 0);
 	J_ASSERT(journal->j_committing_transaction != transaction);
 	J_ASSERT(journal->j_running_transaction != transaction);
diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c
index 29e62d98bae6..002ad2bbc769 100644
--- a/fs/jbd/commit.c
+++ b/fs/jbd/commit.c
@@ -829,8 +829,7 @@ restart_loop:
 	journal->j_committing_transaction = NULL;
 	spin_unlock(&journal->j_state_lock);
 
-	if (commit_transaction->t_checkpoint_list == NULL &&
-	    commit_transaction->t_checkpoint_io_list == NULL) {
+	if (commit_transaction->t_checkpoint_list == NULL) {
 		__journal_drop_transaction(journal, commit_transaction);
 	} else {
 		if (journal->j_checkpoint_transactions == NULL) {
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index 0fe4aa891ddc..41ee79962bb2 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -497,12 +497,6 @@ struct transaction_s
 	 */
 	struct journal_head	*t_checkpoint_list;
 
-	/*
-	 * Doubly-linked circular list of all buffers submitted for IO while
-	 * checkpointing. [j_list_lock]
-	 */
-	struct journal_head	*t_checkpoint_io_list;
-
 	/*
 	 * Doubly-linked circular list of temporary buffers currently undergoing
 	 * IO in the log [j_list_lock]
@@ -852,7 +846,7 @@ extern void journal_commit_transaction(journal_t *);
 
 /* Checkpoint list management */
 int __journal_clean_checkpoint_list(journal_t *journal);
-int __journal_remove_checkpoint(struct journal_head *);
+void __journal_remove_checkpoint(struct journal_head *);
 void __journal_insert_checkpoint(struct journal_head *, transaction_t *);
 
 /* Buffer IO */

From 5ac5f9d1ce8492163dbde5d357dc5d03becf7e36 Mon Sep 17 00:00:00 2001
From: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Tue, 14 Feb 2006 13:53:04 -0800
Subject: [PATCH 182/509] [PATCH] NLM: Fix the NLM_GRANTED callback checks

If 2 threads attached to the same process are blocking on different locks on
different files (maybe even on different servers) but have the same lock
arguments (i.e.  same offset+length - actually quite common, since most
processes try to lock the entire file) then the first GRANTED call that wakes
one up will also wake the other.

Currently when the NLM_GRANTED callback comes in, lockd walks the list of
blocked locks in search of a match to the lock that the NLM server has
granted.  Although it checks the lock pid, start and end, it fails to check
the filehandle and the server address.

By checking the filehandle and server IP address, we ensure that this only
happens if the locks truly are referencing the same file.

Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/lockd/clntlock.c         | 27 +++++++++++++++++----------
 fs/lockd/svc4proc.c         |  2 +-
 fs/lockd/svcproc.c          |  2 +-
 include/linux/lockd/lockd.h |  6 +++---
 4 files changed, 22 insertions(+), 15 deletions(-)

diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c
index 3eaf6e701087..da6354baa0b8 100644
--- a/fs/lockd/clntlock.c
+++ b/fs/lockd/clntlock.c
@@ -111,9 +111,10 @@ long nlmclnt_block(struct nlm_rqst *req, long timeout)
 /*
  * The server lockd has called us back to tell us the lock was granted
  */
-u32
-nlmclnt_grant(struct nlm_lock *lock)
+u32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock)
 {
+	const struct file_lock *fl = &lock->fl;
+	const struct nfs_fh *fh = &lock->fh;
 	struct nlm_wait	*block;
 	u32 res = nlm_lck_denied;
 
@@ -122,14 +123,20 @@ nlmclnt_grant(struct nlm_lock *lock)
 	 * Warning: must not use cookie to match it!
 	 */
 	list_for_each_entry(block, &nlm_blocked, b_list) {
-		if (nlm_compare_locks(block->b_lock, &lock->fl)) {
-			/* Alright, we found a lock. Set the return status
-			 * and wake up the caller
-			 */
-			block->b_status = NLM_LCK_GRANTED;
-			wake_up(&block->b_wait);
-			res = nlm_granted;
-		}
+		struct file_lock *fl_blocked = block->b_lock;
+
+		if (!nlm_compare_locks(fl_blocked, fl))
+			continue;
+		if (!nlm_cmp_addr(&block->b_host->h_addr, addr))
+			continue;
+		if (nfs_compare_fh(NFS_FH(fl_blocked->fl_file->f_dentry->d_inode) ,fh) != 0)
+			continue;
+		/* Alright, we found a lock. Set the return status
+		 * and wake up the caller
+		 */
+		block->b_status = NLM_LCK_GRANTED;
+		wake_up(&block->b_wait);
+		res = nlm_granted;
 	}
 	return res;
 }
diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c
index 4063095d849e..b10f913aa06a 100644
--- a/fs/lockd/svc4proc.c
+++ b/fs/lockd/svc4proc.c
@@ -228,7 +228,7 @@ nlm4svc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp,
 	resp->cookie = argp->cookie;
 
 	dprintk("lockd: GRANTED       called\n");
-	resp->status = nlmclnt_grant(&argp->lock);
+	resp->status = nlmclnt_grant(&rqstp->rq_addr, &argp->lock);
 	dprintk("lockd: GRANTED       status %d\n", ntohl(resp->status));
 	return rpc_success;
 }
diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c
index 3bc437e0cf5b..35681d9cf1fc 100644
--- a/fs/lockd/svcproc.c
+++ b/fs/lockd/svcproc.c
@@ -256,7 +256,7 @@ nlmsvc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp,
 	resp->cookie = argp->cookie;
 
 	dprintk("lockd: GRANTED       called\n");
-	resp->status = nlmclnt_grant(&argp->lock);
+	resp->status = nlmclnt_grant(&rqstp->rq_addr, &argp->lock);
 	dprintk("lockd: GRANTED       status %d\n", ntohl(resp->status));
 	return rpc_success;
 }
diff --git a/include/linux/lockd/lockd.h b/include/linux/lockd/lockd.h
index 920766cea79c..ef21ed296039 100644
--- a/include/linux/lockd/lockd.h
+++ b/include/linux/lockd/lockd.h
@@ -149,7 +149,7 @@ struct nlm_rqst * nlmclnt_alloc_call(void);
 int		  nlmclnt_prepare_block(struct nlm_rqst *req, struct nlm_host *host, struct file_lock *fl);
 void		  nlmclnt_finish_block(struct nlm_rqst *req);
 long		  nlmclnt_block(struct nlm_rqst *req, long timeout);
-u32		  nlmclnt_grant(struct nlm_lock *);
+u32		  nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *);
 void		  nlmclnt_recovery(struct nlm_host *, u32);
 int		  nlmclnt_reclaim(struct nlm_host *, struct file_lock *);
 int		  nlmclnt_setgrantargs(struct nlm_rqst *, struct nlm_lock *);
@@ -204,7 +204,7 @@ nlmsvc_file_inode(struct nlm_file *file)
  * Compare two host addresses (needs modifying for ipv6)
  */
 static __inline__ int
-nlm_cmp_addr(struct sockaddr_in *sin1, struct sockaddr_in *sin2)
+nlm_cmp_addr(const struct sockaddr_in *sin1, const struct sockaddr_in *sin2)
 {
 	return sin1->sin_addr.s_addr == sin2->sin_addr.s_addr;
 }
@@ -214,7 +214,7 @@ nlm_cmp_addr(struct sockaddr_in *sin1, struct sockaddr_in *sin2)
  * When the second lock is of type F_UNLCK, this acts like a wildcard.
  */
 static __inline__ int
-nlm_compare_locks(struct file_lock *fl1, struct file_lock *fl2)
+nlm_compare_locks(const struct file_lock *fl1, const struct file_lock *fl2)
 {
 	return	fl1->fl_pid   == fl2->fl_pid
 	     && fl1->fl_start == fl2->fl_start

From 8b09fb34513225d87d511c7e8f29c0fd3cf860e0 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Tue, 14 Feb 2006 13:53:05 -0800
Subject: [PATCH 183/509] [PATCH] fix x86 topology export in sysfs for
 subarchitectures

The correct way to export hyperthreading based functions is to predicate
them on CONFIG_X86_HT.  Without this, the topology exporting patch breaks
the build on all non-PC x86 subarchitectures.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/topology.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-i386/topology.h b/include/asm-i386/topology.h
index af503a122b23..aa958c6ee83e 100644
--- a/include/asm-i386/topology.h
+++ b/include/asm-i386/topology.h
@@ -27,7 +27,7 @@
 #ifndef _ASM_I386_TOPOLOGY_H
 #define _ASM_I386_TOPOLOGY_H
 
-#ifdef CONFIG_SMP
+#ifdef CONFIG_X86_HT
 #define topology_physical_package_id(cpu)				\
 	(phys_proc_id[cpu] == BAD_APICID ? -1 : phys_proc_id[cpu])
 #define topology_core_id(cpu)						\

From 61b9a26ae6d308ade964db122e0e89299586422c Mon Sep 17 00:00:00 2001
From: Karsten Keil <kkeil@suse.de>
Date: Tue, 14 Feb 2006 13:53:06 -0800
Subject: [PATCH 184/509] [PATCH] Fix NULL pointer dereference in
 isdn_tty_at_cout

The changes in the tty related code introduced wrong parenthesis in a if
condition in the isdn_tty_at_cout function.  This caused access to index -1
in the dev->drv[] array.  This patch change it back to the correct
condition from the previous versions.

Signed-off-by: Karsten Keil <kkeil@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/isdn/i4l/isdn_tty.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/isdn/i4l/isdn_tty.c b/drivers/isdn/i4l/isdn_tty.c
index f190a99604f0..393633681f49 100644
--- a/drivers/isdn/i4l/isdn_tty.c
+++ b/drivers/isdn/i4l/isdn_tty.c
@@ -2359,8 +2359,8 @@ isdn_tty_at_cout(char *msg, modem_info * info)
 
 	/* use queue instead of direct, if online and */
 	/* data is in queue or buffer is full */
-	if ((info->online && tty_buffer_request_room(tty, l) < l) ||
-	    (!skb_queue_empty(&dev->drv[info->isdn_driver]->rpqueue[info->isdn_channel]))) {
+	if (info->online && ((tty_buffer_request_room(tty, l) < l) ||
+	    !skb_queue_empty(&dev->drv[info->isdn_driver]->rpqueue[info->isdn_channel]))) {
 		skb = alloc_skb(l, GFP_ATOMIC);
 		if (!skb) {
 			spin_unlock_irqrestore(&info->readlock, flags);

From 8861da31e3b3e3df7b05e7b157230de3d486e53b Mon Sep 17 00:00:00 2001
From: Jim Keniston <jkenisto@us.ibm.com>
Date: Tue, 14 Feb 2006 13:53:06 -0800
Subject: [PATCH 185/509] [PATCH] kprobes: Update Documentation/kprobes.txt

Update Documentation/kprobes.txt to reflect Kprobes enhancements and other
recent developments.

Acked-by: Ananth Mavinakayanahalli <mananth@in.ibm.com>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kprobes.txt | 77 ++++++++++++++++++++-------------------
 1 file changed, 40 insertions(+), 37 deletions(-)

diff --git a/Documentation/kprobes.txt b/Documentation/kprobes.txt
index 0ea5a0c6e827..2c3b1eae4280 100644
--- a/Documentation/kprobes.txt
+++ b/Documentation/kprobes.txt
@@ -136,17 +136,20 @@ Kprobes, jprobes, and return probes are implemented on the following
 architectures:
 
 - i386
-- x86_64 (AMD-64, E64MT)
+- x86_64 (AMD-64, EM64T)
 - ppc64
-- ia64 (Support for probes on certain instruction types is still in progress.)
+- ia64 (Does not support probes on instruction slot1.)
 - sparc64 (Return probes not yet implemented.)
 
 3. Configuring Kprobes
 
 When configuring the kernel using make menuconfig/xconfig/oldconfig,
-ensure that CONFIG_KPROBES is set to "y".  Under "Kernel hacking",
-look for "Kprobes".  You may have to enable "Kernel debugging"
-(CONFIG_DEBUG_KERNEL) before you can enable Kprobes.
+ensure that CONFIG_KPROBES is set to "y".  Under "Instrumentation
+Support", look for "Kprobes".
+
+So that you can load and unload Kprobes-based instrumentation modules,
+make sure "Loadable module support" (CONFIG_MODULES) and "Module
+unloading" (CONFIG_MODULE_UNLOAD) are set to "y".
 
 You may also want to ensure that CONFIG_KALLSYMS and perhaps even
 CONFIG_KALLSYMS_ALL are set to "y", since kallsyms_lookup_name()
@@ -262,18 +265,18 @@ at any time after the probe has been registered.
 
 5. Kprobes Features and Limitations
 
-As of Linux v2.6.12, Kprobes allows multiple probes at the same
-address.  Currently, however, there cannot be multiple jprobes on
-the same function at the same time.
+Kprobes allows multiple probes at the same address.  Currently,
+however, there cannot be multiple jprobes on the same function at
+the same time.
 
 In general, you can install a probe anywhere in the kernel.
 In particular, you can probe interrupt handlers.  Known exceptions
 are discussed in this section.
 
-For obvious reasons, it's a bad idea to install a probe in
-the code that implements Kprobes (mostly kernel/kprobes.c and
-arch/*/kernel/kprobes.c).  A patch in the v2.6.13 timeframe instructs
-Kprobes to reject such requests.
+The register_*probe functions will return -EINVAL if you attempt
+to install a probe in the code that implements Kprobes (mostly
+kernel/kprobes.c and arch/*/kernel/kprobes.c, but also functions such
+as do_page_fault and notifier_call_chain).
 
 If you install a probe in an inline-able function, Kprobes makes
 no attempt to chase down all inline instances of the function and
@@ -290,18 +293,14 @@ from the accidental ones.  Don't drink and probe.
 
 Kprobes makes no attempt to prevent probe handlers from stepping on
 each other -- e.g., probing printk() and then calling printk() from a
-probe handler.  As of Linux v2.6.12, if a probe handler hits a probe,
-that second probe's handlers won't be run in that instance.
+probe handler.  If a probe handler hits a probe, that second probe's
+handlers won't be run in that instance, and the kprobe.nmissed member
+of the second probe will be incremented.
 
-In Linux v2.6.12 and previous versions, Kprobes' data structures are
-protected by a single lock that is held during probe registration and
-unregistration and while handlers are run.  Thus, no two handlers
-can run simultaneously.  To improve scalability on SMP systems,
-this restriction will probably be removed soon, in which case
-multiple handlers (or multiple instances of the same handler) may
-run concurrently on different CPUs.  Code your handlers accordingly.
+As of Linux v2.6.15-rc1, multiple handlers (or multiple instances of
+the same handler) may run concurrently on different CPUs.
 
-Kprobes does not use semaphores or allocate memory except during
+Kprobes does not use mutexes or allocate memory except during
 registration and unregistration.
 
 Probe handlers are run with preemption disabled.  Depending on the
@@ -316,11 +315,18 @@ address instead of the real return address for kretprobed functions.
 (As far as we can tell, __builtin_return_address() is used only
 for instrumentation and error reporting.)
 
-If the number of times a function is called does not match the
-number of times it returns, registering a return probe on that
-function may produce undesirable results.  We have the do_exit()
-and do_execve() cases covered.  do_fork() is not an issue.  We're
-unaware of other specific cases where this could be a problem.
+If the number of times a function is called does not match the number
+of times it returns, registering a return probe on that function may
+produce undesirable results.  We have the do_exit() case covered.
+do_execve() and do_fork() are not an issue.  We're unaware of other
+specific cases where this could be a problem.
+
+If, upon entry to or exit from a function, the CPU is running on
+a stack other than that of the current task, registering a return
+probe on that function may produce undesirable results.  For this
+reason, Kprobes doesn't support return probes (or kprobes or jprobes)
+on the x86_64 version of __switch_to(); the registration functions
+return -EINVAL.
 
 6. Probe Overhead
 
@@ -347,14 +353,12 @@ k = 0.77 usec; j = 1.31; r = 1.26; kr = 1.45; jr = 1.99
 
 7. TODO
 
-a. SystemTap (http://sourceware.org/systemtap): Work in progress
-to provide a simplified programming interface for probe-based
-instrumentation.
-b. Improved SMP scalability: Currently, work is in progress to handle
-multiple kprobes in parallel.
-c. Kernel return probes for sparc64.
-d. Support for other architectures.
-e. User-space probes.
+a. SystemTap (http://sourceware.org/systemtap): Provides a simplified
+programming interface for probe-based instrumentation.  Try it out.
+b. Kernel return probes for sparc64.
+c. Support for other architectures.
+d. User-space probes.
+e. Watchpoint probes (which fire on data references).
 
 8. Kprobes Example
 
@@ -411,8 +415,7 @@ int init_module(void)
 		printk("Couldn't find %s to plant kprobe\n", "do_fork");
 		return -1;
 	}
-	ret = register_kprobe(&kp);
-	if (ret < 0) {
+	if ((ret = register_kprobe(&kp) < 0)) {
 		printk("register_kprobe failed, returned %d\n", ret);
 		return -1;
 	}

From f822566165dd46ff5de9bf895cfa6c51f53bb0c4 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@mellanox.co.il>
Date: Tue, 14 Feb 2006 13:53:08 -0800
Subject: [PATCH 186/509] [PATCH] madvise MADV_DONTFORK/MADV_DOFORK

Currently, copy-on-write may change the physical address of a page even if the
user requested that the page is pinned in memory (either by mlock or by
get_user_pages).  This happens if the process forks meanwhile, and the parent
writes to that page.  As a result, the page is orphaned: in case of
get_user_pages, the application will never see any data hardware DMA's into
this page after the COW.  In case of mlock'd memory, the parent is not getting
the realtime/security benefits of mlock.

In particular, this affects the Infiniband modules which do DMA from and into
user pages all the time.

This patch adds madvise options to control whether memory range is inherited
across fork.  Useful e.g.  for when hardware is doing DMA from/into these
pages.  Could also be useful to an application wanting to speed up its forks
by cutting large areas out of consideration.

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Acked-by: Hugh Dickins <hugh@veritas.com>
Cc: Michael Kerrisk <mtk-manpages@gmx.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-alpha/mman.h   |  2 ++
 include/asm-arm/mman.h     |  2 ++
 include/asm-arm26/mman.h   |  2 ++
 include/asm-cris/mman.h    |  2 ++
 include/asm-frv/mman.h     |  2 ++
 include/asm-h8300/mman.h   |  2 ++
 include/asm-i386/mman.h    |  2 ++
 include/asm-ia64/mman.h    |  2 ++
 include/asm-m32r/mman.h    |  2 ++
 include/asm-m68k/mman.h    |  2 ++
 include/asm-mips/mman.h    |  2 ++
 include/asm-parisc/mman.h  |  2 ++
 include/asm-powerpc/mman.h |  2 ++
 include/asm-s390/mman.h    |  2 ++
 include/asm-sh/mman.h      |  2 ++
 include/asm-sparc/mman.h   |  2 ++
 include/asm-sparc64/mman.h |  2 ++
 include/asm-v850/mman.h    |  2 ++
 include/asm-x86_64/mman.h  |  2 ++
 include/asm-xtensa/mman.h  |  2 ++
 mm/madvise.c               | 21 +++++++++++++++++----
 21 files changed, 57 insertions(+), 4 deletions(-)

diff --git a/include/asm-alpha/mman.h b/include/asm-alpha/mman.h
index f6439532a262..a21515c16a43 100644
--- a/include/asm-alpha/mman.h
+++ b/include/asm-alpha/mman.h
@@ -43,6 +43,8 @@
 #define	MADV_SPACEAVAIL	5		/* ensure resources are available */
 #define MADV_DONTNEED	6		/* don't need these pages */
 #define MADV_REMOVE	7		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-arm/mman.h b/include/asm-arm/mman.h
index f0bebca2ac21..693ed859e632 100644
--- a/include/asm-arm/mman.h
+++ b/include/asm-arm/mman.h
@@ -36,6 +36,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-arm26/mman.h b/include/asm-arm26/mman.h
index 0ed7780541fa..2096c50df888 100644
--- a/include/asm-arm26/mman.h
+++ b/include/asm-arm26/mman.h
@@ -36,6 +36,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-cris/mman.h b/include/asm-cris/mman.h
index 5a382b8bf3f7..deddfb239ff5 100644
--- a/include/asm-cris/mman.h
+++ b/include/asm-cris/mman.h
@@ -38,6 +38,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-frv/mman.h b/include/asm-frv/mman.h
index 8af4a41c255e..d3bca306da82 100644
--- a/include/asm-frv/mman.h
+++ b/include/asm-frv/mman.h
@@ -36,6 +36,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-h8300/mman.h b/include/asm-h8300/mman.h
index 744a8fb485c2..ac0346f7d11d 100644
--- a/include/asm-h8300/mman.h
+++ b/include/asm-h8300/mman.h
@@ -36,6 +36,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-i386/mman.h b/include/asm-i386/mman.h
index ba4941e6f643..ab2339a1d807 100644
--- a/include/asm-i386/mman.h
+++ b/include/asm-i386/mman.h
@@ -36,6 +36,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-ia64/mman.h b/include/asm-ia64/mman.h
index 828beb24a20e..357ebb780cc0 100644
--- a/include/asm-ia64/mman.h
+++ b/include/asm-ia64/mman.h
@@ -44,6 +44,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-m32r/mman.h b/include/asm-m32r/mman.h
index 12e29747bc84..6b02fe3fcff2 100644
--- a/include/asm-m32r/mman.h
+++ b/include/asm-m32r/mman.h
@@ -38,6 +38,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-m68k/mman.h b/include/asm-m68k/mman.h
index ea262ab88b3b..efd12bc4ccb7 100644
--- a/include/asm-m68k/mman.h
+++ b/include/asm-m68k/mman.h
@@ -36,6 +36,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-mips/mman.h b/include/asm-mips/mman.h
index dd17c8bd62a1..6d01e26830fa 100644
--- a/include/asm-mips/mman.h
+++ b/include/asm-mips/mman.h
@@ -66,6 +66,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON       MAP_ANONYMOUS
diff --git a/include/asm-parisc/mman.h b/include/asm-parisc/mman.h
index 736b0abcac05..a381cf5c8f55 100644
--- a/include/asm-parisc/mman.h
+++ b/include/asm-parisc/mman.h
@@ -49,6 +49,8 @@
 #define MADV_4M_PAGES   22              /* Use 4 Megabyte pages */
 #define MADV_16M_PAGES  24              /* Use 16 Megabyte pages */
 #define MADV_64M_PAGES  26              /* Use 64 Megabyte pages */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-powerpc/mman.h b/include/asm-powerpc/mman.h
index a2e34c21b44f..fcff25d13f13 100644
--- a/include/asm-powerpc/mman.h
+++ b/include/asm-powerpc/mman.h
@@ -45,6 +45,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-s390/mman.h b/include/asm-s390/mman.h
index c8d5409b5d56..d41ca1477010 100644
--- a/include/asm-s390/mman.h
+++ b/include/asm-s390/mman.h
@@ -44,6 +44,8 @@
 #define MADV_WILLNEED  0x3              /* pre-fault pages */
 #define MADV_DONTNEED  0x4              /* discard these pages */
 #define MADV_REMOVE    0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-sh/mman.h b/include/asm-sh/mman.h
index 693bd55a3710..0e08d0573abc 100644
--- a/include/asm-sh/mman.h
+++ b/include/asm-sh/mman.h
@@ -36,6 +36,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-sparc/mman.h b/include/asm-sparc/mman.h
index 98435ad8619e..4a298b2be859 100644
--- a/include/asm-sparc/mman.h
+++ b/include/asm-sparc/mman.h
@@ -55,6 +55,8 @@
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_FREE	0x5		/* (Solaris) contents can be freed */
 #define MADV_REMOVE	0x6		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-sparc64/mman.h b/include/asm-sparc64/mman.h
index cb4b6156194d..d705ec92da8b 100644
--- a/include/asm-sparc64/mman.h
+++ b/include/asm-sparc64/mman.h
@@ -55,6 +55,8 @@
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_FREE	0x5		/* (Solaris) contents can be freed */
 #define MADV_REMOVE	0x6		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-v850/mman.h b/include/asm-v850/mman.h
index edc79965193a..7b851c310e41 100644
--- a/include/asm-v850/mman.h
+++ b/include/asm-v850/mman.h
@@ -33,6 +33,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-x86_64/mman.h b/include/asm-x86_64/mman.h
index d0e97b74f735..b699a38c1c3c 100644
--- a/include/asm-x86_64/mman.h
+++ b/include/asm-x86_64/mman.h
@@ -37,6 +37,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-xtensa/mman.h b/include/asm-xtensa/mman.h
index 082a7504925e..e2d7afb679c8 100644
--- a/include/asm-xtensa/mman.h
+++ b/include/asm-xtensa/mman.h
@@ -73,6 +73,8 @@
 #define MADV_WILLNEED	0x3		/* pre-fault pages */
 #define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_REMOVE	0x5		/* remove these pages & resources */
+#define MADV_DONTFORK	0x30		/* dont inherit across fork */
+#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON       MAP_ANONYMOUS
diff --git a/mm/madvise.c b/mm/madvise.c
index ae0ae3ea299a..af3d573b0141 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -22,16 +22,23 @@ static long madvise_behavior(struct vm_area_struct * vma,
 	struct mm_struct * mm = vma->vm_mm;
 	int error = 0;
 	pgoff_t pgoff;
-	int new_flags = vma->vm_flags & ~VM_READHINTMASK;
+	int new_flags = vma->vm_flags;
 
 	switch (behavior) {
+	case MADV_NORMAL:
+		new_flags = new_flags & ~VM_RAND_READ & ~VM_SEQ_READ;
+		break;
 	case MADV_SEQUENTIAL:
-		new_flags |= VM_SEQ_READ;
+		new_flags = (new_flags & ~VM_RAND_READ) | VM_SEQ_READ;
 		break;
 	case MADV_RANDOM:
-		new_flags |= VM_RAND_READ;
+		new_flags = (new_flags & ~VM_SEQ_READ) | VM_RAND_READ;
 		break;
-	default:
+	case MADV_DONTFORK:
+		new_flags |= VM_DONTCOPY;
+		break;
+	case MADV_DOFORK:
+		new_flags &= ~VM_DONTCOPY;
 		break;
 	}
 
@@ -177,6 +184,12 @@ madvise_vma(struct vm_area_struct *vma, struct vm_area_struct **prev,
 	long error;
 
 	switch (behavior) {
+	case MADV_DOFORK:
+		if (vma->vm_flags & VM_IO) {
+			error = -EINVAL;
+			break;
+		}
+	case MADV_DONTFORK:
 	case MADV_NORMAL:
 	case MADV_SEQUENTIAL:
 	case MADV_RANDOM:

From d6077cb80cde4506720f9165eba99ee07438513f Mon Sep 17 00:00:00 2001
From: "Chen, Kenneth W" <kenneth.w.chen@intel.com>
Date: Tue, 14 Feb 2006 13:53:10 -0800
Subject: [PATCH 187/509] [PATCH] sched: revert "filter affine wakeups"

Revert commit d7102e95b7b9c00277562c29aad421d2d521c5f6:

    [PATCH] sched: filter affine wakeups

Apparently caused more than 10% performance regression for aim7 benchmark.
The setup in use is 16-cpu HP rx8620, 64Gb of memory and 12 MSA1000s with 144
disks.  Each disk is 72Gb with a single ext3 filesystem (courtesy of HP, who
supplied benchmark results).

The problem is, for aim7, the wake-up pattern is random, but it still needs
load balancing action in the wake-up path to achieve best performance.  With
the above commit, lack of load balancing hurts that workload.

However, for workloads like database transaction processing, the requirement
is exactly opposite.  In the wake up path, best performance is achieved with
absolutely zero load balancing.  We simply wake up the process on the CPU that
it was previously run.  Worst performance is obtained when we do load
balancing at wake up.

There isn't an easy way to auto detect the workload characteristics.  Ingo's
earlier patch that detects idle CPU and decide whether to load balance or not
doesn't perform with aim7 either since all CPUs are busy (it causes even
bigger perf.  regression).

Revert commit d7102e95b7b9c00277562c29aad421d2d521c5f6, which causes more
than 10% performance regression with aim7.

Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/sched.h |  5 +----
 kernel/sched.c        | 10 +---------
 2 files changed, 2 insertions(+), 13 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 9c1da0269a18..b6f51e3a38ec 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -697,11 +697,8 @@ struct task_struct {
 
 	int lock_depth;		/* BKL lock depth */
 
-#if defined(CONFIG_SMP)
-	int last_waker_cpu;	/* CPU that last woke this task up */
-#if defined(__ARCH_WANT_UNLOCKED_CTXSW)
+#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW)
 	int oncpu;
-#endif
 #endif
 	int prio, static_prio;
 	struct list_head run_list;
diff --git a/kernel/sched.c b/kernel/sched.c
index 87d93be336a1..66d957227de9 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1204,9 +1204,6 @@ static int try_to_wake_up(task_t *p, unsigned int state, int sync)
 		}
 	}
 
-	if (p->last_waker_cpu != this_cpu)
-		goto out_set_cpu;
-
 	if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed)))
 		goto out_set_cpu;
 
@@ -1277,8 +1274,6 @@ out_set_cpu:
 		cpu = task_cpu(p);
 	}
 
-	p->last_waker_cpu = this_cpu;
-
 out_activate:
 #endif /* CONFIG_SMP */
 	if (old_state == TASK_UNINTERRUPTIBLE) {
@@ -1360,12 +1355,9 @@ void fastcall sched_fork(task_t *p, int clone_flags)
 #ifdef CONFIG_SCHEDSTATS
 	memset(&p->sched_info, 0, sizeof(p->sched_info));
 #endif
-#if defined(CONFIG_SMP)
-	p->last_waker_cpu = cpu;
-#if defined(__ARCH_WANT_UNLOCKED_CTXSW)
+#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW)
 	p->oncpu = 0;
 #endif
-#endif
 #ifdef CONFIG_PREEMPT
 	/* Want to start with kernel preemption disabled. */
 	task_thread_info(p)->preempt_count = 1;

From 5a1342f77304da8dc698e0ecf09925438764d80f Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Tue, 14 Feb 2006 13:53:10 -0800
Subject: [PATCH 188/509] [PATCH] fix a typo in the CPU_H8300H dependencies

Jean-Luc Leger <reiga@dspnet.fr.eu.org> found this obvious typo.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/h8300/Kconfig.cpu | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/h8300/Kconfig.cpu b/arch/h8300/Kconfig.cpu
index a380167a13cf..582797db9603 100644
--- a/arch/h8300/Kconfig.cpu
+++ b/arch/h8300/Kconfig.cpu
@@ -169,7 +169,7 @@ endif
 
 config CPU_H8300H
 	bool
-	depends on (H8002 || H83007 || H83048 || H83068)
+	depends on (H83002 || H83007 || H83048 || H83068)
 	default y
 
 config CPU_H8S

From e35a6619e7be59aa38249346327c89207663bb37 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Tue, 14 Feb 2006 13:53:14 -0800
Subject: [PATCH 189/509] [PATCH] s390: fix __delay implementation

Fix __delay implementation.  Called with an argument "1" or "0" it would
loop nearly forever (since (1/2)-1 = 0xffffffff).

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/lib/delay.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c
index e96c35bddac7..71f0a2fb3078 100644
--- a/arch/s390/lib/delay.c
+++ b/arch/s390/lib/delay.c
@@ -30,7 +30,7 @@ void __delay(unsigned long loops)
          */
         __asm__ __volatile__(
                 "0: brct %0,0b"
-                : /* no outputs */ : "r" (loops/2) );
+                : /* no outputs */ : "r" ((loops/2) + 1));
 }
 
 /*

From 06027bdd278a32a84b273e41db68a5db8ffd2bb6 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Tue, 14 Feb 2006 13:53:15 -0800
Subject: [PATCH 190/509] [PATCH] hrtimer: round up relative start time on
 low-res arches

CONFIG_TIME_LOW_RES is a temporary way for architectures to signal that
they simply return xtime in do_gettimeoffset().  In this corner-case we
want to round up by resolution when starting a relative timer, to avoid
short timeouts.  This will go away with the GTOD framework.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/Kconfig       |  4 ++++
 arch/h8300/Kconfig     |  4 ++++
 arch/m68k/Kconfig      |  4 ++++
 arch/m68knommu/Kconfig |  4 ++++
 arch/parisc/Kconfig    |  5 +++++
 arch/v850/Kconfig      |  4 ++++
 kernel/hrtimer.c       | 13 ++++++++++++-
 7 files changed, 37 insertions(+), 1 deletion(-)

diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig
index 60a617aff8ba..e08383712370 100644
--- a/arch/frv/Kconfig
+++ b/arch/frv/Kconfig
@@ -25,6 +25,10 @@ config GENERIC_HARDIRQS
 	bool
 	default n
 
+config TIME_LOW_RES
+	bool
+	default y
+
 mainmenu "Fujitsu FR-V Kernel Configuration"
 
 source "init/Kconfig"
diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig
index 80940d712acf..98308b018a35 100644
--- a/arch/h8300/Kconfig
+++ b/arch/h8300/Kconfig
@@ -33,6 +33,10 @@ config GENERIC_CALIBRATE_DELAY
 	bool
 	default y
 
+config TIME_LOW_RES
+	bool
+	default y
+
 config ISA
 	bool
 	default y
diff --git a/arch/m68k/Kconfig b/arch/m68k/Kconfig
index 96b919828053..8849439e88dd 100644
--- a/arch/m68k/Kconfig
+++ b/arch/m68k/Kconfig
@@ -21,6 +21,10 @@ config GENERIC_CALIBRATE_DELAY
 	bool
 	default y
 
+config TIME_LOW_RES
+	bool
+	default y
+
 config ARCH_MAY_HAVE_PC_FDC
 	bool
 	depends on Q40 || (BROKEN && SUN3X)
diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig
index e2a6e8648960..e50858dbc237 100644
--- a/arch/m68knommu/Kconfig
+++ b/arch/m68knommu/Kconfig
@@ -29,6 +29,10 @@ config GENERIC_CALIBRATE_DELAY
 	bool
 	default y
 
+config TIME_LOW_RES
+	bool
+	default y
+
 source "init/Kconfig"
 
 menu "Processor type and features"
diff --git a/arch/parisc/Kconfig b/arch/parisc/Kconfig
index 7c914a4c67c3..eca33cfa8a4c 100644
--- a/arch/parisc/Kconfig
+++ b/arch/parisc/Kconfig
@@ -29,6 +29,11 @@ config GENERIC_CALIBRATE_DELAY
 	bool
 	default y
 
+config TIME_LOW_RES
+	bool
+	depends on SMP
+	default y
+
 config GENERIC_ISA_DMA
 	bool
 
diff --git a/arch/v850/Kconfig b/arch/v850/Kconfig
index 04494638b963..e7fc3e500342 100644
--- a/arch/v850/Kconfig
+++ b/arch/v850/Kconfig
@@ -28,6 +28,10 @@ config GENERIC_IRQ_PROBE
 	bool
 	default y
 
+config TIME_LOW_RES
+	bool
+	default y
+
 # Turn off some random 386 crap that can affect device config
 config ISA
 	bool
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 2b6e1757aedd..5ae51f1bc7c8 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -418,8 +418,19 @@ hrtimer_start(struct hrtimer *timer, ktime_t tim, const enum hrtimer_mode mode)
 	/* Switch the timer base, if necessary: */
 	new_base = switch_hrtimer_base(timer, base);
 
-	if (mode == HRTIMER_REL)
+	if (mode == HRTIMER_REL) {
 		tim = ktime_add(tim, new_base->get_time());
+		/*
+		 * CONFIG_TIME_LOW_RES is a temporary way for architectures
+		 * to signal that they simply return xtime in
+		 * do_gettimeoffset(). In this case we want to round up by
+		 * resolution when starting a relative timer, to avoid short
+		 * timeouts. This will go away with the GTOD framework.
+		 */
+#ifdef CONFIG_TIME_LOW_RES
+		tim = ktime_add(tim, base->resolution);
+#endif
+	}
 	timer->expires = tim;
 
 	enqueue_hrtimer(timer, new_base);

From 68f624fc8b9fa50de9cc0ebd612ef7b7b9fa32d0 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 14 Feb 2006 13:53:18 -0800
Subject: [PATCH 191/509] [PATCH] FRV: Miscellaneous fixes

Make various alterations and fixes to the FRV arch:

 (1) Resyncs the FRV system call collection with the i386 arch.

 (2) Discards __iounmap() as it's not used.

 (3) Fixes the use of the SWAP/SWAPI instruction to get the arguments the right
     way around in atomic.h, and also to get the asm constraints correct.

 (4) Moves copy_to/from_user_page() to asm/cacheflush.h to be consistent with
     other archs.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/frv/kernel/entry.S      | 26 +++++++++++++++++++++-----
 arch/frv/mm/kmap.c           |  9 ---------
 include/asm-frv/atomic.h     |  6 +++---
 include/asm-frv/cacheflush.h | 12 ++++++++++++
 include/asm-frv/io.h         |  1 -
 include/asm-frv/uaccess.h    |  3 ---
 include/asm-frv/unistd.h     | 28 ++++++++++++++++++++++------
 7 files changed, 58 insertions(+), 27 deletions(-)

diff --git a/arch/frv/kernel/entry.S b/arch/frv/kernel/entry.S
index 5f6548388b74..c69d499d28cf 100644
--- a/arch/frv/kernel/entry.S
+++ b/arch/frv/kernel/entry.S
@@ -1418,11 +1418,27 @@ sys_call_table:
 	.long sys_add_key
 	.long sys_request_key
 	.long sys_keyctl
-	.long sys_ni_syscall // sys_vperfctr_open
-	.long sys_ni_syscall // sys_vperfctr_control	/* 290 */
-	.long sys_ni_syscall // sys_vperfctr_unlink
-	.long sys_ni_syscall // sys_vperfctr_iresume
-	.long sys_ni_syscall // sys_vperfctr_read
+	.long sys_ioprio_set
+	.long sys_ioprio_get		/* 290 */
+	.long sys_inotify_init
+	.long sys_inotify_add_watch
+	.long sys_inotify_rm_watch
+	.long sys_migrate_pages
+	.long sys_openat		/* 295 */
+	.long sys_mkdirat
+	.long sys_mknodat
+	.long sys_fchownat
+	.long sys_futimesat
+	.long sys_newfstatat		/* 300 */
+	.long sys_unlinkat
+	.long sys_renameat
+	.long sys_linkat
+	.long sys_symlinkat
+	.long sys_readlinkat		/* 305 */
+	.long sys_fchmodat
+	.long sys_faccessat
+	.long sys_pselect6
+	.long sys_ppoll
 
 
 syscall_table_size = (. - sys_call_table)
diff --git a/arch/frv/mm/kmap.c b/arch/frv/mm/kmap.c
index 539f45e6d15e..c54f18e65ea6 100644
--- a/arch/frv/mm/kmap.c
+++ b/arch/frv/mm/kmap.c
@@ -43,15 +43,6 @@ void iounmap(void *addr)
 {
 }
 
-/*
- * __iounmap unmaps nearly everything, so be careful
- * it doesn't free currently pointer/page tables anymore but it
- * wans't used anyway and might be added later.
- */
-void __iounmap(void *addr, unsigned long size)
-{
-}
-
 /*
  * Set new cache mode for some kernel address space.
  * The caller must push data for that range itself, if such data may already
diff --git a/include/asm-frv/atomic.h b/include/asm-frv/atomic.h
index a59f684b4f33..5d9f84bfdcad 100644
--- a/include/asm-frv/atomic.h
+++ b/include/asm-frv/atomic.h
@@ -220,9 +220,9 @@ extern unsigned long atomic_test_and_XOR_mask(unsigned long mask, volatile unsig
 	switch (sizeof(__xg_orig)) {						\
 	case 4:									\
 		asm volatile(							\
-			"swap%I0 %2,%M0"					\
-			: "+m"(*__xg_ptr), "=&r"(__xg_orig)			\
-			: "r"(x)						\
+			"swap%I0 %M0,%1"					\
+			: "+m"(*__xg_ptr), "=r"(__xg_orig)			\
+			: "1"(x)						\
 			: "memory"						\
 			);							\
 		break;								\
diff --git a/include/asm-frv/cacheflush.h b/include/asm-frv/cacheflush.h
index 3007deccb490..eaa5826bc1c8 100644
--- a/include/asm-frv/cacheflush.h
+++ b/include/asm-frv/cacheflush.h
@@ -87,5 +87,17 @@ static inline void flush_icache_page(struct vm_area_struct *vma, struct page *pa
 	flush_icache_user_range(vma, page, page_to_phys(page), PAGE_SIZE);
 }
 
+/*
+ * permit ptrace to access another process's address space through the icache
+ * and the dcache
+ */
+#define copy_to_user_page(vma, page, vaddr, dst, src, len)	\
+do {								\
+	memcpy((dst), (src), (len));				\
+	flush_icache_user_range((vma), (page), (vaddr), (len));	\
+} while(0)
+
+#define copy_from_user_page(vma, page, vaddr, dst, src, len)	\
+	memcpy((dst), (src), (len))
 
 #endif /* _ASM_CACHEFLUSH_H */
diff --git a/include/asm-frv/io.h b/include/asm-frv/io.h
index 075369b1a34b..01247cb2bc39 100644
--- a/include/asm-frv/io.h
+++ b/include/asm-frv/io.h
@@ -251,7 +251,6 @@ static inline void writel(uint32_t datum, volatile void __iomem *addr)
 #define IOMAP_WRITETHROUGH		3
 
 extern void __iomem *__ioremap(unsigned long physaddr, unsigned long size, int cacheflag);
-extern void __iounmap(void __iomem *addr, unsigned long size);
 
 static inline void __iomem *ioremap(unsigned long physaddr, unsigned long size)
 {
diff --git a/include/asm-frv/uaccess.h b/include/asm-frv/uaccess.h
index b6bcbe01f6ee..a1d140438863 100644
--- a/include/asm-frv/uaccess.h
+++ b/include/asm-frv/uaccess.h
@@ -306,7 +306,4 @@ extern long strnlen_user(const char *src, long count);
 
 extern unsigned long search_exception_table(unsigned long addr);
 
-#define copy_to_user_page(vma, page, vaddr, dst, src, len)	memcpy(dst, src, len)
-#define copy_from_user_page(vma, page, vaddr, dst, src, len)	memcpy(dst, src, len)
-
 #endif /* _ASM_UACCESS_H */
diff --git a/include/asm-frv/unistd.h b/include/asm-frv/unistd.h
index 4d994d2e99e3..322531caa484 100644
--- a/include/asm-frv/unistd.h
+++ b/include/asm-frv/unistd.h
@@ -295,13 +295,29 @@
 #define __NR_add_key		286
 #define __NR_request_key	287
 #define __NR_keyctl		288
-#define __NR_vperfctr_open	289
-#define __NR_vperfctr_control	(__NR_perfctr_info+1)
-#define __NR_vperfctr_unlink	(__NR_perfctr_info+2)
-#define __NR_vperfctr_iresume	(__NR_perfctr_info+3)
-#define __NR_vperfctr_read	(__NR_perfctr_info+4)
+#define __NR_ioprio_set		289
+#define __NR_ioprio_get		290
+#define __NR_inotify_init	291
+#define __NR_inotify_add_watch	292
+#define __NR_inotify_rm_watch	293
+#define __NR_migrate_pages	294
+#define __NR_openat		295
+#define __NR_mkdirat		296
+#define __NR_mknodat		297
+#define __NR_fchownat		298
+#define __NR_futimesat		299
+#define __NR_newfstatat		300
+#define __NR_unlinkat		301
+#define __NR_renameat		302
+#define __NR_linkat		303
+#define __NR_symlinkat		304
+#define __NR_readlinkat		305
+#define __NR_fchmodat		306
+#define __NR_faccessat		307
+#define __NR_pselect6		308
+#define __NR_ppoll		309
 
-#define NR_syscalls 294
+#define NR_syscalls 310
 
 /*
  * process the return value of a syscall, consigning it to one of two possible fates

From 28baebae73c3ea8b75c7cae225a7db817ab825a9 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 14 Feb 2006 13:53:20 -0800
Subject: [PATCH 192/509] [PATCH] FRV: Use virtual interrupt disablement

Make the FRV arch use virtual interrupt disablement because accesses to the
processor status register (PSR) are relatively slow and because we will
soon have the need to deal with multiple interrupt controls at the same
time (separate h/w and inter-core interrupts).

The way this is done is to dedicate one of the four integer condition code
registers (ICC2) to maintaining a virtual interrupt disablement state
whilst inside the kernel.  This uses the ICC2.Z flag (Zero) to indicate
whether the interrupts are virtually disabled and the ICC2.C flag (Carry)
to indicate whether the interrupts are physically disabled.

ICC2.Z is set to indicate interrupts are virtually disabled.  ICC2.C is set
to indicate interrupts are physically enabled.  Under normal running
conditions Z==0 and C==1.

Disabling interrupts with local_irq_disable() doesn't then actually
physically disable interrupts - it merely sets ICC2.Z to 1.  Should an
interrupt then happen, the exception prologue will note ICC2.Z is set and
branch out of line using one instruction (an unlikely BEQ).  Here it will
physically disable interrupts and clear ICC2.C.

When it comes time to enable interrupts (local_irq_enable()), this simply
clears the ICC2.Z flag and invokes a trap #2 if both Z and C flags are
clear (the HI integer condition).  This can be done with the TIHI
conditional trap instruction.

The trap then physically reenables interrupts and sets ICC2.C again.  Upon
returning the interrupt will be taken as interrupts will then be enabled.
Note that whilst processing the trap, the whole exceptions system is
disabled, and so an interrupt can't happen till it returns.

If no pending interrupt had happened, ICC2.C would still be set, the HI
condition would not be fulfilled, and no trap will happen.

Saving interrupts (local_irq_save) is simply a matter of pulling the ICC2.Z
flag out of the CCR register, shifting it down and masking it off.  This
gives a result of 0 if interrupts were enabled and 1 if they weren't.

Restoring interrupts (local_irq_restore) is then a matter of taking the
saved value mentioned previously and XOR'ing it against 1.  If it was one,
the result will be zero, and if it was zero the result will be non-zero.
This result is then used to affect the ICC2.Z flag directly (it is a
condition code flag after all).  An XOR instruction does not affect the
Carry flag, and so that bit of state is unchanged.  The two flags can then
be sampled to see if they're both zero using the trap (TIHI) as for the
unconditional reenablement (local_irq_enable).

This patch also:

 (1) Modifies the debugging stub (break.S) to handle single-stepping crossing
     into the trap #2 handler and into virtually disabled interrupts.

 (2) Removes superseded fixup pointers from the second instructions in the trap
     tables (there's no a separate fixup table for this).

 (3) Declares the trap #3 vector for use in .org directives in the trap table.

 (4) Moves irq_enter() and irq_exit() in do_IRQ() to avoid problems with
     virtual interrupt handling, and removes the duplicate code that has now
     been folded into irq_exit() (softirq and preemption handling).

 (5) Tells the compiler in the arch Makefile that ICC2 is now reserved.

 (6) Documents the in-kernel ABI, including the virtual interrupts.

 (7) Renames the old irq management functions to different names.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/fujitsu/frv/kernel-ABI.txt | 234 +++++++++++++++++++++++
 arch/frv/Makefile                        |   2 +-
 arch/frv/kernel/break.S                  |  77 +++++++-
 arch/frv/kernel/entry-table.S            |  39 +++-
 arch/frv/kernel/entry.S                  |  65 ++++++-
 arch/frv/kernel/head.S                   |   3 +
 arch/frv/kernel/irq.c                    |  41 +---
 include/asm-frv/spr-regs.h               |   1 +
 include/asm-frv/system.h                 |  88 ++++++++-
 9 files changed, 489 insertions(+), 61 deletions(-)
 create mode 100644 Documentation/fujitsu/frv/kernel-ABI.txt

diff --git a/Documentation/fujitsu/frv/kernel-ABI.txt b/Documentation/fujitsu/frv/kernel-ABI.txt
new file mode 100644
index 000000000000..0ed9b0a779bc
--- /dev/null
+++ b/Documentation/fujitsu/frv/kernel-ABI.txt
@@ -0,0 +1,234 @@
+				 =================================
+				 INTERNAL KERNEL ABI FOR FR-V ARCH
+				 =================================
+
+The internal FRV kernel ABI is not quite the same as the userspace ABI. A number of the registers
+are used for special purposed, and the ABI is not consistent between modules vs core, and MMU vs
+no-MMU.
+
+This partly stems from the fact that FRV CPUs do not have a separate supervisor stack pointer, and
+most of them do not have any scratch registers, thus requiring at least one general purpose
+register to be clobbered in such an event. Also, within the kernel core, it is possible to simply
+jump or call directly between functions using a relative offset. This cannot be extended to modules
+for the displacement is likely to be too far. Thus in modules the address of a function to call
+must be calculated in a register and then used, requiring two extra instructions.
+
+This document has the following sections:
+
+ (*) System call register ABI
+ (*) CPU operating modes
+ (*) Internal kernel-mode register ABI
+ (*) Internal debug-mode register ABI
+ (*) Virtual interrupt handling
+
+
+========================
+SYSTEM CALL REGISTER ABI
+========================
+
+When a system call is made, the following registers are effective:
+
+	REGISTERS	CALL			RETURN
+	===============	=======================	=======================
+	GR7		System call number	Preserved
+	GR8		Syscall arg #1		Return value
+	GR9-GR13	Syscall arg #2-6	Preserved
+
+
+===================
+CPU OPERATING MODES
+===================
+
+The FR-V CPU has three basic operating modes. In order of increasing capability:
+
+  (1) User mode.
+
+      Basic userspace running mode.
+
+  (2) Kernel mode.
+
+      Normal kernel mode. There are many additional control registers available that may be
+      accessed in this mode, in addition to all the stuff available to user mode. This has two
+      submodes:
+
+      (a) Exceptions enabled (PSR.T == 1).
+
+      	  Exceptions will invoke the appropriate normal kernel mode handler. On entry to the
+      	  handler, the PSR.T bit will be cleared.
+
+      (b) Exceptions disabled (PSR.T == 0).
+
+      	  No exceptions or interrupts may happen. Any mandatory exceptions will cause the CPU to
+      	  halt unless the CPU is told to jump into debug mode instead.
+
+  (3) Debug mode.
+
+      No exceptions may happen in this mode. Memory protection and management exceptions will be
+      flagged for later consideration, but the exception handler won't be invoked. Debugging traps
+      such as hardware breakpoints and watchpoints will be ignored. This mode is entered only by
+      debugging events obtained from the other two modes.
+
+      All kernel mode registers may be accessed, plus a few extra debugging specific registers.
+
+
+=================================
+INTERNAL KERNEL-MODE REGISTER ABI
+=================================
+
+There are a number of permanent register assignments that are set up by entry.S in the exception
+prologue. Note that there is a complete set of exception prologues for each of user->kernel
+transition and kernel->kernel transition. There are also user->debug and kernel->debug mode
+transition prologues.
+
+
+	REGISTER	FLAVOUR	USE
+	===============	=======	====================================================
+	GR1			Supervisor stack pointer
+	GR15			Current thread info pointer
+	GR16			GP-Rel base register for small data
+	GR28			Current exception frame pointer (__frame)
+	GR29			Current task pointer (current)
+	GR30			Destroyed by kernel mode entry
+	GR31		NOMMU	Destroyed by debug mode entry
+	GR31		MMU	Destroyed by TLB miss kernel mode entry
+	CCR.ICC2		Virtual interrupt disablement tracking
+	CCCR.CC3		Cleared by exception prologue (atomic op emulation)
+	SCR0		MMU	See mmu-layout.txt.
+	SCR1		MMU	See mmu-layout.txt.
+	SCR2		MMU	Save for EAR0 (destroyed by icache insns in debug mode)
+	SCR3		MMU	Save for GR31 during debug exceptions
+	DAMR/IAMR	NOMMU	Fixed memory protection layout.
+	DAMR/IAMR	MMU	See mmu-layout.txt.
+
+
+Certain registers are also used or modified across function calls:
+
+	REGISTER	CALL				RETURN
+	===============	===============================	===============================
+	GR0		Fixed Zero			-
+	GR2		Function call frame pointer
+	GR3		Special				Preserved
+	GR3-GR7		-				Clobbered
+	GR8		Function call arg #1		Return value (or clobbered)
+	GR9		Function call arg #2		Return value MSW (or clobbered)
+	GR10-GR13	Function call arg #3-#6		Clobbered
+	GR14		-				Clobbered
+	GR15-GR16	Special				Preserved
+	GR17-GR27	-				Preserved
+	GR28-GR31	Special				Only accessed explicitly
+	LR		Return address after CALL	Clobbered
+	CCR/CCCR	-				Mostly Clobbered
+
+
+================================
+INTERNAL DEBUG-MODE REGISTER ABI
+================================
+
+This is the same as the kernel-mode register ABI for functions calls. The difference is that in
+debug-mode there's a different stack and a different exception frame. Almost all the global
+registers from kernel-mode (including the stack pointer) may be changed.
+
+	REGISTER	FLAVOUR	USE
+	===============	=======	====================================================
+	GR1			Debug stack pointer
+	GR16			GP-Rel base register for small data
+	GR31			Current debug exception frame pointer (__debug_frame)
+	SCR3		MMU	Saved value of GR31
+
+
+Note that debug mode is able to interfere with the kernel's emulated atomic ops, so it must be
+exceedingly careful not to do any that would interact with the main kernel in this regard. Hence
+the debug mode code (gdbstub) is almost completely self-contained. The only external code used is
+the sprintf family of functions.
+
+Futhermore, break.S is so complicated because single-step mode does not switch off on entry to an
+exception. That means unless manually disabled, single-stepping will blithely go on stepping into
+things like interrupts. See gdbstub.txt for more information.
+
+
+==========================
+VIRTUAL INTERRUPT HANDLING
+==========================
+
+Because accesses to the PSR is so slow, and to disable interrupts we have to access it twice (once
+to read and once to write), we don't actually disable interrupts at all if we don't have to. What
+we do instead is use the ICC2 condition code flags to note virtual disablement, such that if we
+then do take an interrupt, we note the flag, really disable interrupts, set another flag and resume
+execution at the point the interrupt happened. Setting condition flags as a side effect of an
+arithmetic or logical instruction is really fast. This use of the ICC2 only occurs within the
+kernel - it does not affect userspace.
+
+The flags we use are:
+
+ (*) CCR.ICC2.Z [Zero flag]
+
+     Set to virtually disable interrupts, clear when interrupts are virtually enabled. Can be
+     modified by logical instructions without affecting the Carry flag.
+
+ (*) CCR.ICC2.C [Carry flag]
+
+     Clear to indicate hardware interrupts are really disabled, set otherwise.
+
+
+What happens is this:
+
+ (1) Normal kernel-mode operation.
+
+	ICC2.Z is 0, ICC2.C is 1.
+
+ (2) An interrupt occurs. The exception prologue examines ICC2.Z and determines that nothing needs
+     doing. This is done simply with an unlikely BEQ instruction.
+
+ (3) The interrupts are disabled (local_irq_disable)
+
+	ICC2.Z is set to 1.
+
+ (4) If interrupts were then re-enabled (local_irq_enable):
+
+	ICC2.Z would be set to 0.
+
+     A TIHI #2 instruction (trap #2 if condition HI - Z==0 && C==0) would be used to trap if
+     interrupts were now virtually enabled, but physically disabled - which they're not, so the
+     trap isn't taken. The kernel would then be back to state (1).
+
+ (5) An interrupt occurs. The exception prologue examines ICC2.Z and determines that the interrupt
+     shouldn't actually have happened. It jumps aside, and there disabled interrupts by setting
+     PSR.PIL to 14 and then it clears ICC2.C.
+
+ (6) If interrupts were then saved and disabled again (local_irq_save):
+
+	ICC2.Z would be shifted into the save variable and masked off (giving a 1).
+
+	ICC2.Z would then be set to 1 (thus unchanged), and ICC2.C would be unaffected (ie: 0).
+
+ (7) If interrupts were then restored from state (6) (local_irq_restore):
+
+	ICC2.Z would be set to indicate the result of XOR'ing the saved value (ie: 1) with 1, which
+	gives a result of 0 - thus leaving ICC2.Z set.
+
+	ICC2.C would remain unaffected (ie: 0).
+
+     A TIHI #2 instruction would be used to again assay the current state, but this would do
+     nothing as Z==1.
+
+ (8) If interrupts were then enabled (local_irq_enable):
+
+	ICC2.Z would be cleared. ICC2.C would be left unaffected. Both flags would now be 0.
+
+     A TIHI #2 instruction again issued to assay the current state would then trap as both Z==0
+     [interrupts virtually enabled] and C==0 [interrupts really disabled] would then be true.
+
+ (9) The trap #2 handler would simply enable hardware interrupts (set PSR.PIL to 0), set ICC2.C to
+     1 and return.
+
+(10) Immediately upon returning, the pending interrupt would be taken.
+
+(11) The interrupt handler would take the path of actually processing the interrupt (ICC2.Z is
+     clear, BEQ fails as per step (2)).
+
+(12) The interrupt handler would then set ICC2.C to 1 since hardware interrupts are definitely
+     enabled - or else the kernel wouldn't be here.
+
+(13) On return from the interrupt handler, things would be back to state (1).
+
+This trap (#2) is only available in kernel mode. In user mode it will result in SIGILL.
diff --git a/arch/frv/Makefile b/arch/frv/Makefile
index 90c0fb8d9dc3..d163747d17c0 100644
--- a/arch/frv/Makefile
+++ b/arch/frv/Makefile
@@ -81,7 +81,7 @@ endif
 # - reserve CC3 for use with atomic ops
 # - all the extra registers are dealt with only at context switch time
 CFLAGS		+= -mno-fdpic -mgpr-32 -msoft-float -mno-media
-CFLAGS		+= -ffixed-fcc3 -ffixed-cc3 -ffixed-gr15
+CFLAGS		+= -ffixed-fcc3 -ffixed-cc3 -ffixed-gr15 -ffixed-icc2
 AFLAGS		+= -mno-fdpic
 ASFLAGS		+= -mno-fdpic
 
diff --git a/arch/frv/kernel/break.S b/arch/frv/kernel/break.S
index 33233dc23e29..687c48d62dde 100644
--- a/arch/frv/kernel/break.S
+++ b/arch/frv/kernel/break.S
@@ -200,12 +200,20 @@ __break_step:
 	movsg		bpcsr,gr2
 	sethi.p		%hi(__entry_kernel_external_interrupt),gr3
 	setlo		%lo(__entry_kernel_external_interrupt),gr3
-	subcc		gr2,gr3,gr0,icc0
+	subcc.p		gr2,gr3,gr0,icc0
+	sethi		%hi(__entry_uspace_external_interrupt),gr3
+	setlo.p		%lo(__entry_uspace_external_interrupt),gr3
 	beq		icc0,#2,__break_step_kernel_external_interrupt
-	sethi.p		%hi(__entry_uspace_external_interrupt),gr3
-	setlo		%lo(__entry_uspace_external_interrupt),gr3
-	subcc		gr2,gr3,gr0,icc0
+	subcc.p		gr2,gr3,gr0,icc0
+	sethi		%hi(__entry_kernel_external_interrupt_virtually_disabled),gr3
+	setlo.p		%lo(__entry_kernel_external_interrupt_virtually_disabled),gr3
 	beq		icc0,#2,__break_step_uspace_external_interrupt
+	subcc.p		gr2,gr3,gr0,icc0
+	sethi		%hi(__entry_kernel_external_interrupt_virtual_reenable),gr3
+	setlo.p		%lo(__entry_kernel_external_interrupt_virtual_reenable),gr3
+	beq		icc0,#2,__break_step_kernel_external_interrupt_virtually_disabled
+	subcc		gr2,gr3,gr0,icc0
+	beq		icc0,#2,__break_step_kernel_external_interrupt_virtual_reenable
 
 	LEDS		0x2007,gr2
 
@@ -254,6 +262,9 @@ __break_step_kernel_softprog_interrupt:
 # step through an external interrupt from kernel mode
 	.globl		__break_step_kernel_external_interrupt
 __break_step_kernel_external_interrupt:
+	# deal with virtual interrupt disablement
+	beq		icc2,#0,__break_step_kernel_external_interrupt_virtually_disabled
+
 	sethi.p		%hi(__entry_kernel_external_interrupt_reentry),gr3
 	setlo		%lo(__entry_kernel_external_interrupt_reentry),gr3
 
@@ -294,6 +305,64 @@ __break_return_as_kernel_prologue:
 #endif
 	rett		#1
 
+# we single-stepped into an interrupt handler whilst interrupts were merely virtually disabled
+# need to really disable interrupts, set flag, fix up and return
+__break_step_kernel_external_interrupt_virtually_disabled:
+	movsg		psr,gr2
+	andi		gr2,#~PSR_PIL,gr2
+	ori		gr2,#PSR_PIL_14,gr2	/* debugging interrupts only */
+	movgs		gr2,psr
+
+	ldi		@(gr31,#REG_CCR),gr3
+	movgs		gr3,ccr
+	subcc.p		gr0,gr0,gr0,icc2	/* leave Z set, clear C */
+
+	# exceptions must've been enabled and we must've been in supervisor mode
+	setlos		BPSR_BET|BPSR_BS,gr3
+	movgs		gr3,bpsr
+
+	# return to where the interrupt happened
+	movsg		pcsr,gr2
+	movgs		gr2,bpcsr
+
+	lddi.p		@(gr31,#REG_GR(2)),gr2
+
+	xor		gr31,gr31,gr31
+	movgs		gr0,brr
+#ifdef CONFIG_MMU
+	movsg		scr3,gr31
+#endif
+	rett		#1
+
+# we stepped through into the virtual interrupt reenablement trap
+#
+# we also want to single step anyway, but after fixing up so that we get an event on the
+# instruction after the broken-into exception returns
+	.globl		__break_step_kernel_external_interrupt_virtual_reenable
+__break_step_kernel_external_interrupt_virtual_reenable:
+	movsg		psr,gr2
+	andi		gr2,#~PSR_PIL,gr2
+	movgs		gr2,psr
+
+	ldi		@(gr31,#REG_CCR),gr3
+	movgs		gr3,ccr
+	subicc		gr0,#1,gr0,icc2		/* clear Z, set C */
+
+	# save the adjusted ICC2
+	movsg		ccr,gr3
+	sti		gr3,@(gr31,#REG_CCR)
+
+	# exceptions must've been enabled and we must've been in supervisor mode
+	setlos		BPSR_BET|BPSR_BS,gr3
+	movgs		gr3,bpsr
+
+	# return to where the trap happened
+	movsg		pcsr,gr2
+	movgs		gr2,bpcsr
+
+	# and then process the single step
+	bra		__break_continue
+
 # step through an internal exception from uspace mode
 	.globl		__break_step_uspace_softprog_interrupt
 __break_step_uspace_softprog_interrupt:
diff --git a/arch/frv/kernel/entry-table.S b/arch/frv/kernel/entry-table.S
index 9b9243e2103c..81568acea9cd 100644
--- a/arch/frv/kernel/entry-table.S
+++ b/arch/frv/kernel/entry-table.S
@@ -116,6 +116,8 @@ __break_kerneltrap_fixup_table:
 	.long		__break_step_uspace_external_interrupt
 	.section .trap.kernel
 	.org		\tbr_tt
+	# deal with virtual interrupt disablement
+	beq		icc2,#0,__entry_kernel_external_interrupt_virtually_disabled
 	bra		__entry_kernel_external_interrupt
 	.section .trap.fixup.kernel
 	.org		\tbr_tt >> 2
@@ -259,25 +261,52 @@ __trap_fixup_kernel_data_tlb_miss:
 	.org		TBR_TT_TRAP0
 	.rept		127
 	bra		__entry_uspace_softprog_interrupt
-	bra		__break_step_uspace_softprog_interrupt
-	.long		0,0
+	.long		0,0,0
 	.endr
 	.org		TBR_TT_BREAK
 	bra		__entry_break
 	.long		0,0,0
 
+	.section	.trap.fixup.user
+	.org		TBR_TT_TRAP0 >> 2
+	.rept		127
+	.long		__break_step_uspace_softprog_interrupt
+	.endr
+	.org		TBR_TT_BREAK >> 2
+	.long		0
+
 	# miscellaneous kernel mode entry points
 	.section	.trap.kernel
 	.org		TBR_TT_TRAP0
-	.rept		127
 	bra		__entry_kernel_softprog_interrupt
-	bra		__break_step_kernel_softprog_interrupt
-	.long		0,0
+	.org		TBR_TT_TRAP1
+	bra		__entry_kernel_softprog_interrupt
+
+	# trap #2 in kernel - reenable interrupts
+	.org		TBR_TT_TRAP2
+	bra		__entry_kernel_external_interrupt_virtual_reenable
+
+	# miscellaneous kernel traps
+	.org		TBR_TT_TRAP3
+	.rept		124
+	bra		__entry_kernel_softprog_interrupt
+	.long		0,0,0
 	.endr
 	.org		TBR_TT_BREAK
 	bra		__entry_break
 	.long		0,0,0
 
+	.section	.trap.fixup.kernel
+	.org		TBR_TT_TRAP0 >> 2
+	.long		__break_step_kernel_softprog_interrupt
+	.long		__break_step_kernel_softprog_interrupt
+	.long		__break_step_kernel_external_interrupt_virtual_reenable
+	.rept		124
+	.long		__break_step_kernel_softprog_interrupt
+	.endr
+	.org		TBR_TT_BREAK >> 2
+	.long		0
+
 	# miscellaneous debug mode entry points
 	.section	.trap.break
 	.org		TBR_TT_BREAK
diff --git a/arch/frv/kernel/entry.S b/arch/frv/kernel/entry.S
index c69d499d28cf..1d21c8d34d8a 100644
--- a/arch/frv/kernel/entry.S
+++ b/arch/frv/kernel/entry.S
@@ -141,7 +141,10 @@ __entry_uspace_external_interrupt_reentry:
 
 	movsg		gner0,gr4
 	movsg		gner1,gr5
-	stdi		gr4,@(gr28,#REG_GNER0)
+	stdi.p		gr4,@(gr28,#REG_GNER0)
+
+	# interrupts start off fully disabled in the interrupt handler
+	subcc		gr0,gr0,gr0,icc2		/* set Z and clear C */
 
 	# set up kernel global registers
 	sethi.p		%hi(__kernel_current_task),gr5
@@ -193,9 +196,8 @@ __entry_uspace_external_interrupt_reentry:
         .type		__entry_kernel_external_interrupt,@function
 __entry_kernel_external_interrupt:
 	LEDS		0x6210
-
-	sub		sp,gr15,gr31
-	LEDS32
+//	sub		sp,gr15,gr31
+//	LEDS32
 
 	# set up the stack pointer
 	or.p		sp,gr0,gr30
@@ -231,7 +233,10 @@ __entry_kernel_external_interrupt_reentry:
 	stdi		gr24,@(gr28,#REG_GR(24))
 	stdi		gr26,@(gr28,#REG_GR(26))
 	sti		gr29,@(gr28,#REG_GR(29))
-	stdi		gr30,@(gr28,#REG_GR(30))
+	stdi.p		gr30,@(gr28,#REG_GR(30))
+
+	# note virtual interrupts will be fully enabled upon return
+	subicc		gr0,#1,gr0,icc2			/* clear Z, set C */
 
 	movsg		tbr ,gr20
 	movsg		psr ,gr22
@@ -267,7 +272,10 @@ __entry_kernel_external_interrupt_reentry:
 
 	movsg		gner0,gr4
 	movsg		gner1,gr5
-	stdi		gr4,@(gr28,#REG_GNER0)
+	stdi.p		gr4,@(gr28,#REG_GNER0)
+
+	# interrupts start off fully disabled in the interrupt handler
+	subcc		gr0,gr0,gr0,icc2			/* set Z and clear C */
 
 	# set the return address
 	sethi.p		%hi(__entry_return_from_kernel_interrupt),gr4
@@ -291,6 +299,45 @@ __entry_kernel_external_interrupt_reentry:
 
 	.size		__entry_kernel_external_interrupt,.-__entry_kernel_external_interrupt
 
+###############################################################################
+#
+# deal with interrupts that were actually virtually disabled
+# - we need to really disable them, flag the fact and return immediately
+# - if you change this, you must alter break.S also
+#
+###############################################################################
+	.balign		L1_CACHE_BYTES
+	.globl		__entry_kernel_external_interrupt_virtually_disabled
+	.type		__entry_kernel_external_interrupt_virtually_disabled,@function
+__entry_kernel_external_interrupt_virtually_disabled:
+	movsg		psr,gr30
+	andi		gr30,#~PSR_PIL,gr30
+	ori		gr30,#PSR_PIL_14,gr30		; debugging interrupts only
+	movgs		gr30,psr
+	subcc		gr0,gr0,gr0,icc2		; leave Z set, clear C
+	rett		#0
+
+	.size		__entry_kernel_external_interrupt_virtually_disabled,.-__entry_kernel_external_interrupt_virtually_disabled
+
+###############################################################################
+#
+# deal with re-enablement of interrupts that were pending when virtually re-enabled
+# - set ICC2.C, re-enable the real interrupts and return
+# - we can clear ICC2.Z because we shouldn't be here if it's not 0 [due to TIHI]
+# - if you change this, you must alter break.S also
+#
+###############################################################################
+	.balign		L1_CACHE_BYTES
+	.globl		__entry_kernel_external_interrupt_virtual_reenable
+	.type		__entry_kernel_external_interrupt_virtual_reenable,@function
+__entry_kernel_external_interrupt_virtual_reenable:
+	movsg		psr,gr30
+	andi		gr30,#~PSR_PIL,gr30		; re-enable interrupts
+	movgs		gr30,psr
+	subicc		gr0,#1,gr0,icc2			; clear Z, set C
+	rett		#0
+
+	.size		__entry_kernel_external_interrupt_virtual_reenable,.-__entry_kernel_external_interrupt_virtual_reenable
 
 ###############################################################################
 #
@@ -335,6 +382,7 @@ __entry_uspace_softprog_interrupt_reentry:
 
 	sethi.p		%hi(__entry_return_from_user_exception),gr23
 	setlo		%lo(__entry_return_from_user_exception),gr23
+
 	bra		__entry_common
 
 	.size		__entry_uspace_softprog_interrupt,.-__entry_uspace_softprog_interrupt
@@ -495,7 +543,10 @@ __entry_common:
 
 	movsg		gner0,gr4
 	movsg		gner1,gr5
-	stdi		gr4,@(gr28,#REG_GNER0)
+	stdi.p		gr4,@(gr28,#REG_GNER0)
+
+	# set up virtual interrupt disablement
+	subicc		gr0,#1,gr0,icc2			/* clear Z flag, set C flag */
 
 	# set up kernel global registers
 	sethi.p		%hi(__kernel_current_task),gr5
diff --git a/arch/frv/kernel/head.S b/arch/frv/kernel/head.S
index c73b4fe9f6ca..29a5265489b7 100644
--- a/arch/frv/kernel/head.S
+++ b/arch/frv/kernel/head.S
@@ -513,6 +513,9 @@ __head_mmu_enabled:
 	movgs		gr0,ccr
 	movgs		gr0,cccr
 
+	# initialise the virtual interrupt handling
+	subcc		gr0,gr0,gr0,icc2		/* set Z, clear C */
+
 #ifdef CONFIG_MMU
 	movgs		gr3,scr2
 	movgs		gr3,scr3
diff --git a/arch/frv/kernel/irq.c b/arch/frv/kernel/irq.c
index 59580c59c62c..27ab4c30aac6 100644
--- a/arch/frv/kernel/irq.c
+++ b/arch/frv/kernel/irq.c
@@ -287,18 +287,11 @@ asmlinkage void do_IRQ(void)
 	struct irq_source *source;
 	int level, cpu;
 
+	irq_enter();
+
 	level = (__frame->tbr >> 4) & 0xf;
 	cpu = smp_processor_id();
 
-#if 0
-	{
-		static u32 irqcount;
-		*(volatile u32 *) 0xe1200004 = ~((irqcount++ << 8) | level);
-		*(volatile u16 *) 0xffc00100 = (u16) ~0x9999;
-		mb();
-	}
-#endif
-
 	if ((unsigned long) __frame - (unsigned long) (current + 1) < 512)
 		BUG();
 
@@ -308,40 +301,12 @@ asmlinkage void do_IRQ(void)
 
 	kstat_this_cpu.irqs[level]++;
 
-	irq_enter();
-
 	for (source = frv_irq_levels[level].sources; source; source = source->next)
 		source->doirq(source);
 
-	irq_exit();
-
 	__clr_MASK(level);
 
-	/* only process softirqs if we didn't interrupt another interrupt handler */
-	if ((__frame->psr & PSR_PIL) == PSR_PIL_0)
-		if (local_softirq_pending())
-			do_softirq();
-
-#ifdef CONFIG_PREEMPT
-	local_irq_disable();
-	while (--current->preempt_count == 0) {
-		if (!(__frame->psr & PSR_S) ||
-		    current->need_resched == 0 ||
-		    in_interrupt())
-			break;
-		current->preempt_count++;
-		local_irq_enable();
-		preempt_schedule();
-		local_irq_disable();
-	}
-#endif
-
-#if 0
-	{
-		*(volatile u16 *) 0xffc00100 = (u16) ~0x6666;
-		mb();
-	}
-#endif
+	irq_exit();
 
 } /* end do_IRQ() */
 
diff --git a/include/asm-frv/spr-regs.h b/include/asm-frv/spr-regs.h
index ef472f058d9c..c2a541ef828d 100644
--- a/include/asm-frv/spr-regs.h
+++ b/include/asm-frv/spr-regs.h
@@ -98,6 +98,7 @@
 #define TBR_TT_TRAP0		(0x80 << 4)
 #define TBR_TT_TRAP1		(0x81 << 4)
 #define TBR_TT_TRAP2		(0x82 << 4)
+#define TBR_TT_TRAP3		(0x83 << 4)
 #define TBR_TT_TRAP126		(0xfe << 4)
 #define TBR_TT_BREAK		(0xff << 4)
 
diff --git a/include/asm-frv/system.h b/include/asm-frv/system.h
index d2aea70a5f64..f72ff0c4dc0b 100644
--- a/include/asm-frv/system.h
+++ b/include/asm-frv/system.h
@@ -40,8 +40,84 @@ do {									\
 
 /*
  * interrupt flag manipulation
+ * - use virtual interrupt management since touching the PSR is slow
+ *   - ICC2.Z: T if interrupts virtually disabled
+ *   - ICC2.C: F if interrupts really disabled
+ * - if Z==1 upon interrupt:
+ *   - C is set to 0
+ *   - interrupts are really disabled
+ *   - entry.S returns immediately
+ * - uses TIHI (TRAP if Z==0 && C==0) #2 to really reenable interrupts
+ *   - if taken, the trap:
+ *     - sets ICC2.C
+ *     - enables interrupts
  */
-#define local_irq_disable()				\
+#define local_irq_disable()					\
+do {								\
+	/* set Z flag, but don't change the C flag */		\
+	asm volatile("	andcc	gr0,gr0,gr0,icc2	\n"	\
+		     :						\
+		     :						\
+		     : "memory", "icc2"				\
+		     );						\
+} while(0)
+
+#define local_irq_enable()					\
+do {								\
+	/* clear Z flag and then test the C flag */		\
+	asm volatile("  oricc	gr0,#1,gr0,icc2		\n"	\
+		     "	tihi	icc2,gr0,#2		\n"	\
+		     :						\
+		     :						\
+		     : "memory", "icc2"				\
+		     );						\
+} while(0)
+
+#define local_save_flags(flags)					\
+do {								\
+	typecheck(unsigned long, flags);			\
+	asm volatile("movsg ccr,%0"				\
+		     : "=r"(flags)				\
+		     :						\
+		     : "memory");				\
+								\
+	/* shift ICC2.Z to bit 0 */				\
+	flags >>= 26;						\
+								\
+	/* make flags 1 if interrupts disabled, 0 otherwise */	\
+	flags &= 1UL;						\
+} while(0)
+
+#define irqs_disabled() \
+	({unsigned long flags; local_save_flags(flags); flags; })
+
+#define	local_irq_save(flags)			\
+do {						\
+	typecheck(unsigned long, flags);	\
+	local_save_flags(flags);		\
+	local_irq_disable();			\
+} while(0)
+
+#define	local_irq_restore(flags)					\
+do {									\
+	typecheck(unsigned long, flags);				\
+									\
+	/* load the Z flag by turning 1 if disabled into 0 if disabled	\
+	 * and thus setting the Z flag but not the C flag */		\
+	asm volatile("  xoricc	%0,#1,gr0,icc2		\n"		\
+		     /* then test Z=0 and C=0 */			\
+		     "	tihi	icc2,gr0,#2		\n"		\
+		     :							\
+		     : "r"(flags)					\
+		     : "memory", "icc2"					\
+		     );							\
+									\
+} while(0)
+
+/*
+ * real interrupt flag manipulation
+ */
+#define __local_irq_disable()				\
 do {							\
 	unsigned long psr;				\
 	asm volatile("	movsg	psr,%0		\n"	\
@@ -53,7 +129,7 @@ do {							\
 		     : "memory");			\
 } while(0)
 
-#define local_irq_enable()				\
+#define __local_irq_enable()				\
 do {							\
 	unsigned long psr;				\
 	asm volatile("	movsg	psr,%0		\n"	\
@@ -64,7 +140,7 @@ do {							\
 		     : "memory");			\
 } while(0)
 
-#define local_save_flags(flags)			\
+#define __local_save_flags(flags)		\
 do {						\
 	typecheck(unsigned long, flags);	\
 	asm("movsg psr,%0"			\
@@ -73,7 +149,7 @@ do {						\
 	    : "memory");			\
 } while(0)
 
-#define	local_irq_save(flags)				\
+#define	__local_irq_save(flags)				\
 do {							\
 	unsigned long npsr;				\
 	typecheck(unsigned long, flags);		\
@@ -86,7 +162,7 @@ do {							\
 		     : "memory");			\
 } while(0)
 
-#define	local_irq_restore(flags)			\
+#define	__local_irq_restore(flags)			\
 do {							\
 	typecheck(unsigned long, flags);		\
 	asm volatile("	movgs	%0,psr		\n"	\
@@ -95,7 +171,7 @@ do {							\
 		     : "memory");			\
 } while(0)
 
-#define irqs_disabled() \
+#define __irqs_disabled() \
 	((__get_PSR() & PSR_PIL) >= PSR_PIL_14)
 
 /*

From 581141cb4b10ebd865dcb7b80f5e712e2def5408 Mon Sep 17 00:00:00 2001
From: "Albert D. Cahalan" <acahalan@cs.uml.edu>
Date: Tue, 14 Feb 2006 13:53:20 -0800
Subject: [PATCH 193/509] [PATCH] x86: document sysenter path

This path isn't obvious.  It looks as if the kernel will be taking three
args from the user stack, but it only takes one from there.

Signed-off-by: Albert Cahalan <acahalan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/vsyscall-sysenter.S | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/arch/i386/kernel/vsyscall-sysenter.S b/arch/i386/kernel/vsyscall-sysenter.S
index 4daefb2ec1b2..76b728159403 100644
--- a/arch/i386/kernel/vsyscall-sysenter.S
+++ b/arch/i386/kernel/vsyscall-sysenter.S
@@ -7,6 +7,21 @@
  *    for details.
  */
 
+/*
+ * The caller puts arg2 in %ecx, which gets pushed. The kernel will use
+ * %ecx itself for arg2. The pushing is because the sysexit instruction
+ * (found in entry.S) requires that we clobber %ecx with the desired %esp.
+ * User code might expect that %ecx is unclobbered though, as it would be
+ * for returning via the iret instruction, so we must push and pop.
+ *
+ * The caller puts arg3 in %edx, which the sysexit instruction requires
+ * for %eip. Thus, exactly as for arg2, we must push and pop.
+ *
+ * Arg6 is different. The caller puts arg6 in %ebp. Since the sysenter
+ * instruction clobbers %esp, the user's %esp won't even survive entry
+ * into the kernel. We store %esp in %ebp. Code in entry.S must fetch
+ * arg6 from the stack.
+ */
 	.text
 	.globl __kernel_vsyscall
 	.type __kernel_vsyscall,@function

From e2fbf1ace5cfefdd192f29fd4a027422f567c62d Mon Sep 17 00:00:00 2001
From: Thomas Meyer <thomas@m3y3r.de>
Date: Tue, 14 Feb 2006 13:53:21 -0800
Subject: [PATCH 194/509] [PATCH] x86: gitignore some autogenerated files for
 i386

Add some more gitignore files for i386 architecture.  This files are
created during the build process of a i386 kernel.

Signed-off-by: Thomas Meyer <thomas@m3y3r.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/boot/.gitignore       | 3 +++
 arch/i386/boot/tools/.gitignore | 1 +
 arch/i386/kernel/.gitignore     | 1 +
 3 files changed, 5 insertions(+)
 create mode 100644 arch/i386/boot/.gitignore
 create mode 100644 arch/i386/boot/tools/.gitignore
 create mode 100644 arch/i386/kernel/.gitignore

diff --git a/arch/i386/boot/.gitignore b/arch/i386/boot/.gitignore
new file mode 100644
index 000000000000..495f20c085de
--- /dev/null
+++ b/arch/i386/boot/.gitignore
@@ -0,0 +1,3 @@
+bootsect
+bzImage
+setup
diff --git a/arch/i386/boot/tools/.gitignore b/arch/i386/boot/tools/.gitignore
new file mode 100644
index 000000000000..378eac25d311
--- /dev/null
+++ b/arch/i386/boot/tools/.gitignore
@@ -0,0 +1 @@
+build
diff --git a/arch/i386/kernel/.gitignore b/arch/i386/kernel/.gitignore
new file mode 100644
index 000000000000..40836ad9079c
--- /dev/null
+++ b/arch/i386/kernel/.gitignore
@@ -0,0 +1 @@
+vsyscall.lds

From 10ee39fe3ff618d274e1cd0f6abbc2917b736bfd Mon Sep 17 00:00:00 2001
From: Christian Trefzer <ctrefzer@gmx.de>
Date: Tue, 14 Feb 2006 13:53:26 -0800
Subject: [PATCH 195/509] [PATCH] neofb: avoid resetting display config on
 unblank

Fix issues with the NeoMagic framebuffer driver.

It nicely complements my previous fix already in linus' tree.  The only
thing missing now is that the external CRT will not be activated at neofb
init when external-only is selected, either by register read or
module/kernel parameter.

Testing was done on a Dell Latitude CPi-A/NM2200 chip.

Previous behaviour:
- before booting linux, set the preferred display config X via FN+F8

- boot linux, neofb stores the register values in a private
  variable

- change the display config to Y via keystroke

- leave the machine in peace until display is blanked

- touching any key will result in display config X being restored

- booting up, the BIOS will acknowledge config Y, though...

Current behaviour:
At the time of unblanking, config Y is honoured because we now read back
register contents instead of just overwriting them with outdated values.

Signed-off by: Christian Trefzer <ctrefzer@gmx.de>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/neofb.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/drivers/video/neofb.c b/drivers/video/neofb.c
index 747602aa5615..b85e2b180a44 100644
--- a/drivers/video/neofb.c
+++ b/drivers/video/neofb.c
@@ -1334,6 +1334,12 @@ static int neofb_blank(int blank_mode, struct fb_info *info)
 	struct neofb_par *par = info->par;
 	int seqflags, lcdflags, dpmsflags, reg;
 
+	/*
+	 * Reload the value stored in the register, might have been changed via
+	 * FN keystroke
+	 */
+	par->PanelDispCntlReg1 = vga_rgfx(NULL, 0x20) & 0x03;
+
 	switch (blank_mode) {
 	case FB_BLANK_POWERDOWN:	/* powerdown - both sync lines down */
 		seqflags = VGA_SR01_SCREEN_OFF; /* Disable sequencer */
@@ -1366,7 +1372,7 @@ static int neofb_blank(int blank_mode, struct fb_info *info)
 	case FB_BLANK_NORMAL:		/* just blank screen (backlight stays on) */
 		seqflags = VGA_SR01_SCREEN_OFF;	/* Disable sequencer */
 		lcdflags = par->PanelDispCntlReg1 & 0x02; /* LCD normal */
-		dpmsflags = 0;			/* no hsync/vsync suppression */
+		dpmsflags = 0x00;	/* no hsync/vsync suppression */
 		break;
 	case FB_BLANK_UNBLANK:		/* unblank */
 		seqflags = 0;			/* Enable sequencer */

From 93544cc6486bea12e127ed58ca33477bb6ceafe6 Mon Sep 17 00:00:00 2001
From: Steve French <smfrench@austin.rr.com>
Date: Tue, 14 Feb 2006 22:30:52 -0600
Subject: [PATCH 196/509] [PATCH] CIFS: fix cifs_user_read oops when null SMB
 response on forcedirectio mount

This patch fixes an oops reported by Adrian Bunk in cifs_user_read when a null
read response is returned on a forcedirectio mount.

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/cifs/file.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index d17c97d07c80..675bd2568297 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -1442,13 +1442,15 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data,
 					 &bytes_read, &smb_read_data,
 					 &buf_type);
 			pSMBr = (struct smb_com_read_rsp *)smb_read_data;
-			if (copy_to_user(current_offset, 
-					 smb_read_data + 4 /* RFC1001 hdr */
-					 + le16_to_cpu(pSMBr->DataOffset), 
-					 bytes_read)) {
-				rc = -EFAULT;
-			}
 			if (smb_read_data) {
+				if (copy_to_user(current_offset,
+						smb_read_data +
+						4 /* RFC1001 length field */ +
+						le16_to_cpu(pSMBr->DataOffset),
+						bytes_read)) {
+					rc = -EFAULT;
+				}
+
 				if(buf_type == CIFS_SMALL_BUFFER)
 					cifs_small_buf_release(smb_read_data);
 				else if(buf_type == CIFS_LARGE_BUFFER)

From a09d31ff762a3671f2ae41b3bca50a100c5e4da6 Mon Sep 17 00:00:00 2001
From: Alessandro Zummo <alessandro.zummo@towertech.it>
Date: Wed, 15 Feb 2006 00:48:40 -0500
Subject: [PATCH 197/509] Input: ixp4xx-beeper - fix compile error

Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/misc/ixp4xx-beeper.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/input/misc/ixp4xx-beeper.c b/drivers/input/misc/ixp4xx-beeper.c
index d448bb5e4869..3a6ae85cd69c 100644
--- a/drivers/input/misc/ixp4xx-beeper.c
+++ b/drivers/input/misc/ixp4xx-beeper.c
@@ -19,6 +19,7 @@
 #include <linux/input.h>
 #include <linux/delay.h>
 #include <linux/platform_device.h>
+#include <linux/interrupt.h>
 #include <asm/hardware.h>
 
 MODULE_AUTHOR("Alessandro Zummo <a.zummo@towertech.it>");

From 50f6dde0ad05ee4ee8450feb731b15b716115c4d Mon Sep 17 00:00:00 2001
From: Meelis Roos <mroos@linux.ee>
Date: Wed, 15 Feb 2006 00:48:58 -0500
Subject: [PATCH 198/509] Input: logips2pp - add new signature (99)

Add Logitech mouse type 99 (Premium Optical Wheel Mouse, model M-BT58,
plain 3 buttons + wheel) to cure the following message: logips2pp: Detected
unknown logitech mouse model 99

Signed-off-by: Meelis Roos <mroos@linux.ee>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/mouse/logips2pp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/input/mouse/logips2pp.c b/drivers/input/mouse/logips2pp.c
index c88520d3d13c..40333d61093c 100644
--- a/drivers/input/mouse/logips2pp.c
+++ b/drivers/input/mouse/logips2pp.c
@@ -232,6 +232,7 @@ static struct ps2pp_info *get_model_info(unsigned char model)
 		{ 88,	PS2PP_KIND_WHEEL,	PS2PP_WHEEL },
 		{ 96,	0,			0 },
 		{ 97,	PS2PP_KIND_TP3,		PS2PP_WHEEL | PS2PP_HWHEEL },
+		{ 99,	PS2PP_KIND_WHEEL,	PS2PP_WHEEL },
 		{ 100,	PS2PP_KIND_MX,					/* MX510 */
 				PS2PP_WHEEL | PS2PP_SIDE_BTN | PS2PP_TASK_BTN |
 				PS2PP_EXTRA_BTN | PS2PP_NAV_BTN },

From b8044c74bcd64bd1a9d2e8cec58fdcd40f16f5a4 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Wed, 15 Feb 2006 00:49:09 -0500
Subject: [PATCH 199/509] Input: trackpoint - enable devices connected to
 external port

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/mouse/trackpoint.c | 20 ++++++++++++++------
 drivers/input/mouse/trackpoint.h |  4 ++--
 2 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/input/mouse/trackpoint.c b/drivers/input/mouse/trackpoint.c
index b4898d8a68e2..6d9ec9ab1b90 100644
--- a/drivers/input/mouse/trackpoint.c
+++ b/drivers/input/mouse/trackpoint.c
@@ -68,15 +68,19 @@ struct trackpoint_attr_data {
 	size_t field_offset;
 	unsigned char command;
 	unsigned char mask;
+	unsigned char inverted;
 };
 
 static ssize_t trackpoint_show_int_attr(struct psmouse *psmouse, void *data, char *buf)
 {
 	struct trackpoint_data *tp = psmouse->private;
 	struct trackpoint_attr_data *attr = data;
-	unsigned char *field = (unsigned char *)((char *)tp + attr->field_offset);
+	unsigned char value = *(unsigned char *)((char *)tp + attr->field_offset);
 
-	return sprintf(buf, "%u\n", *field);
+	if (attr->inverted)
+		value = !value;
+
+	return sprintf(buf, "%u\n", value);
 }
 
 static ssize_t trackpoint_set_int_attr(struct psmouse *psmouse, void *data,
@@ -120,6 +124,9 @@ static ssize_t trackpoint_set_bit_attr(struct psmouse *psmouse, void *data,
 	if (*rest || value > 1)
 		return -EINVAL;
 
+	if (attr->inverted)
+		value = !value;
+
 	if (*field != value) {
 		*field = value;
 		trackpoint_toggle_bit(&psmouse->ps2dev, attr->command, attr->mask);
@@ -129,11 +136,12 @@ static ssize_t trackpoint_set_bit_attr(struct psmouse *psmouse, void *data,
 }
 
 
-#define TRACKPOINT_BIT_ATTR(_name, _command, _mask)				\
+#define TRACKPOINT_BIT_ATTR(_name, _command, _mask, _inv)				\
 	static struct trackpoint_attr_data trackpoint_attr_##_name = {		\
 		.field_offset	= offsetof(struct trackpoint_data, _name),	\
 		.command	= _command,					\
 		.mask		= _mask,					\
+		.inverted	= _inv,						\
 	};									\
 	PSMOUSE_DEFINE_ATTR(_name, S_IWUSR | S_IRUGO,				\
 			    &trackpoint_attr_##_name,				\
@@ -150,9 +158,9 @@ TRACKPOINT_INT_ATTR(upthresh, TP_UP_THRESH);
 TRACKPOINT_INT_ATTR(ztime, TP_Z_TIME);
 TRACKPOINT_INT_ATTR(jenks, TP_JENKS_CURV);
 
-TRACKPOINT_BIT_ATTR(press_to_select, TP_TOGGLE_PTSON, TP_MASK_PTSON);
-TRACKPOINT_BIT_ATTR(skipback, TP_TOGGLE_SKIPBACK, TP_MASK_SKIPBACK);
-TRACKPOINT_BIT_ATTR(ext_dev, TP_TOGGLE_EXT_DEV, TP_MASK_EXT_DEV);
+TRACKPOINT_BIT_ATTR(press_to_select, TP_TOGGLE_PTSON, TP_MASK_PTSON, 0);
+TRACKPOINT_BIT_ATTR(skipback, TP_TOGGLE_SKIPBACK, TP_MASK_SKIPBACK, 0);
+TRACKPOINT_BIT_ATTR(ext_dev, TP_TOGGLE_EXT_DEV, TP_MASK_EXT_DEV, 1);
 
 static struct attribute *trackpoint_attrs[] = {
 	&psmouse_attr_sensitivity.dattr.attr,
diff --git a/drivers/input/mouse/trackpoint.h b/drivers/input/mouse/trackpoint.h
index 9857d8b6ad66..050298b1a09d 100644
--- a/drivers/input/mouse/trackpoint.h
+++ b/drivers/input/mouse/trackpoint.h
@@ -78,7 +78,7 @@
 
 #define TP_TOGGLE_MB		0x23	/* Disable/Enable Middle Button */
 #define TP_MASK_MB			0x01
-#define TP_TOGGLE_EXT_DEV	0x23	/* Toggle external device */
+#define TP_TOGGLE_EXT_DEV	0x23	/* Disable external device */
 #define TP_MASK_EXT_DEV			0x02
 #define TP_TOGGLE_DRIFT		0x23	/* Drift Correction */
 #define TP_MASK_DRIFT			0x80
@@ -125,7 +125,7 @@
 #define TP_DEF_MB		0x00
 #define TP_DEF_PTSON		0x00
 #define TP_DEF_SKIPBACK		0x00
-#define TP_DEF_EXT_DEV		0x01
+#define TP_DEF_EXT_DEV		0x00	/* 0 means enabled */
 
 #define MAKE_PS2_CMD(params, results, cmd) ((params<<12) | (results<<8) | (cmd))
 

From a90f7e98b7df3309ebc0e389076990456db20989 Mon Sep 17 00:00:00 2001
From: Dmitry Torokhov <dtor_core@ameritech.net>
Date: Wed, 15 Feb 2006 00:49:22 -0500
Subject: [PATCH 200/509] Input: ads7846 - convert to to dynamic input_dev
 allocation

Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/touchscreen/ads7846.c | 79 +++++++++++++++++------------
 1 file changed, 47 insertions(+), 32 deletions(-)

diff --git a/drivers/input/touchscreen/ads7846.c b/drivers/input/touchscreen/ads7846.c
index b45a45ca7cc9..72cf0a26d676 100644
--- a/drivers/input/touchscreen/ads7846.c
+++ b/drivers/input/touchscreen/ads7846.c
@@ -60,7 +60,7 @@ struct ts_event {
 };
 
 struct ads7846 {
-	struct input_dev	input;
+	struct input_dev	*input;
 	char			phys[32];
 
 	struct spi_device	*spi;
@@ -152,7 +152,7 @@ static int ads7846_read12_ser(struct device *dev, unsigned command)
 	struct ser_req		*req = kzalloc(sizeof *req, SLAB_KERNEL);
 	int			status;
 	int			sample;
-	int 			i;
+	int			i;
 
 	if (!req)
 		return -ENOMEM;
@@ -236,11 +236,12 @@ SHOW(vbatt)
 
 static void ads7846_rx(void *ads)
 {
-	struct ads7846	*ts = ads;
-	unsigned	Rt;
-	unsigned	sync = 0;
-	u16		x, y, z1, z2;
-	unsigned long	flags;
+	struct ads7846		*ts = ads;
+	struct input_dev	*input_dev = ts->input;
+	unsigned		Rt;
+	unsigned		sync = 0;
+	u16			x, y, z1, z2;
+	unsigned long		flags;
 
 	/* adjust:  12 bit samples (left aligned), built from
 	 * two 8 bit values writen msb-first.
@@ -276,21 +277,21 @@ static void ads7846_rx(void *ads)
 	 * won't notice that, even if nPENIRQ never fires ...
 	 */
 	if (!ts->pendown && Rt != 0) {
-		input_report_key(&ts->input, BTN_TOUCH, 1);
+		input_report_key(input_dev, BTN_TOUCH, 1);
 		sync = 1;
 	} else if (ts->pendown && Rt == 0) {
-		input_report_key(&ts->input, BTN_TOUCH, 0);
+		input_report_key(input_dev, BTN_TOUCH, 0);
 		sync = 1;
 	}
 
 	if (Rt) {
-		input_report_abs(&ts->input, ABS_X, x);
-		input_report_abs(&ts->input, ABS_Y, y);
-		input_report_abs(&ts->input, ABS_PRESSURE, Rt);
+		input_report_abs(input_dev, ABS_X, x);
+		input_report_abs(input_dev, ABS_Y, y);
+		input_report_abs(input_dev, ABS_PRESSURE, Rt);
 		sync = 1;
 	}
 	if (sync)
-		input_sync(&ts->input);
+		input_sync(input_dev);
 
 #ifdef	VERBOSE
 	if (Rt || ts->pendown)
@@ -396,9 +397,11 @@ static int ads7846_resume(struct spi_device *spi)
 static int __devinit ads7846_probe(struct spi_device *spi)
 {
 	struct ads7846			*ts;
+	struct input_dev		*input_dev;
 	struct ads7846_platform_data	*pdata = spi->dev.platform_data;
 	struct spi_transfer		*x;
 	int				i;
+	int				err;
 
 	if (!spi->irq) {
 		dev_dbg(&spi->dev, "no IRQ?\n");
@@ -423,13 +426,18 @@ static int __devinit ads7846_probe(struct spi_device *spi)
 	 * to discard the four garbage LSBs.
 	 */
 
-	if (!(ts = kzalloc(sizeof(struct ads7846), GFP_KERNEL)))
-		return -ENOMEM;
+	ts = kzalloc(sizeof(struct ads7846), GFP_KERNEL);
+	input_dev = input_allocate_device();
+	if (!ts || !input_dev) {
+		err = -ENOMEM;
+		goto err_free_mem;
+	}
 
 	dev_set_drvdata(&spi->dev, ts);
+	spi->dev.power.power_state = PMSG_ON;
 
 	ts->spi = spi;
-	spi->dev.power.power_state = PMSG_ON;
+	ts->input = input_dev;
 
 	init_timer(&ts->timer);
 	ts->timer.data = (unsigned long) ts;
@@ -439,28 +447,25 @@ static int __devinit ads7846_probe(struct spi_device *spi)
 	ts->vref_delay_usecs = pdata->vref_delay_usecs ? : 100;
 	ts->x_plate_ohms = pdata->x_plate_ohms ? : 400;
 
-	init_input_dev(&ts->input);
+	snprintf(ts->phys, sizeof(ts->phys), "%s/input0", spi->dev.bus_id);
 
-	ts->input.dev = &spi->dev;
-	ts->input.name = "ADS784x Touchscreen";
-	snprintf(ts->phys, sizeof ts->phys, "%s/input0", spi->dev.bus_id);
-	ts->input.phys = ts->phys;
+	input_dev->name = "ADS784x Touchscreen";
+	input_dev->phys = ts->phys;
+	input_dev->cdev.dev = &spi->dev;
 
-	ts->input.evbit[0] = BIT(EV_KEY) | BIT(EV_ABS);
-	ts->input.keybit[LONG(BTN_TOUCH)] = BIT(BTN_TOUCH);
-	input_set_abs_params(&ts->input, ABS_X,
+	input_dev->evbit[0] = BIT(EV_KEY) | BIT(EV_ABS);
+	input_dev->keybit[LONG(BTN_TOUCH)] = BIT(BTN_TOUCH);
+	input_set_abs_params(input_dev, ABS_X,
 			pdata->x_min ? : 0,
 			pdata->x_max ? : MAX_12BIT,
 			0, 0);
-	input_set_abs_params(&ts->input, ABS_Y,
+	input_set_abs_params(input_dev, ABS_Y,
 			pdata->y_min ? : 0,
 			pdata->y_max ? : MAX_12BIT,
 			0, 0);
-	input_set_abs_params(&ts->input, ABS_PRESSURE,
+	input_set_abs_params(input_dev, ABS_PRESSURE,
 			pdata->pressure_min, pdata->pressure_max, 0, 0);
 
-	input_register_device(&ts->input);
-
 	/* set up the transfers to read touchscreen state; this assumes we
 	 * use formula #2 for pressure, not #3.
 	 */
@@ -510,9 +515,8 @@ static int __devinit ads7846_probe(struct spi_device *spi)
 			SA_SAMPLE_RANDOM | SA_TRIGGER_FALLING,
 			spi->dev.bus_id, ts)) {
 		dev_dbg(&spi->dev, "irq %d busy?\n", spi->irq);
-		input_unregister_device(&ts->input);
-		kfree(ts);
-		return -EBUSY;
+		err = -EBUSY;
+		goto err_free_mem;
 	}
 
 	dev_info(&spi->dev, "touchscreen, irq %d\n", spi->irq);
@@ -534,7 +538,18 @@ static int __devinit ads7846_probe(struct spi_device *spi)
 		device_create_file(&spi->dev, &dev_attr_vbatt);
 	device_create_file(&spi->dev, &dev_attr_vaux);
 
+	err = input_register_device(input_dev);
+	if (err)
+		goto err_free_irq;
+
 	return 0;
+
+ err_free_irq:
+	free_irq(spi->irq, ts);
+ err_free_mem:
+	input_free_device(input_dev);
+	kfree(ts);
+	return err;
 }
 
 static int __devexit ads7846_remove(struct spi_device *spi)
@@ -554,7 +569,7 @@ static int __devexit ads7846_remove(struct spi_device *spi)
 		device_remove_file(&spi->dev, &dev_attr_vbatt);
 	device_remove_file(&spi->dev, &dev_attr_vaux);
 
-	input_unregister_device(&ts->input);
+	input_unregister_device(ts->input);
 	kfree(ts);
 
 	dev_dbg(&spi->dev, "unregistered touchscreen\n");

From d93f70b2d758e79ee4ac9d6d982e3f532453911f Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Wed, 15 Feb 2006 00:49:35 -0500
Subject: [PATCH 201/509] Input: ads7846 - assorted updates

This updates the ads7846 touchscreen driver:
  - to allow faster clocking (this driver doesn't push sample rates);
  - bugfixes the conversion of spi_transfer to lists;
  - some dma-unsafe command buffers are fixed.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/touchscreen/ads7846.c | 68 +++++++++++++++++++----------
 1 file changed, 44 insertions(+), 24 deletions(-)

diff --git a/drivers/input/touchscreen/ads7846.c b/drivers/input/touchscreen/ads7846.c
index 72cf0a26d676..8c12a974b411 100644
--- a/drivers/input/touchscreen/ads7846.c
+++ b/drivers/input/touchscreen/ads7846.c
@@ -48,10 +48,13 @@
 
 #define	TS_POLL_PERIOD	msecs_to_jiffies(10)
 
+/* this driver doesn't aim at the peak continuous sample rate */
+#define	SAMPLE_BITS	(8 /*cmd*/ + 16 /*sample*/ + 2 /* before, after */)
+
 struct ts_event {
 	/* For portability, we can't read 12 bit values using SPI (which
 	 * would make the controller deliver them as native byteorder u16
-	 * with msbs zeroed).  Instead, we read them as two 8-byte values,
+	 * with msbs zeroed).  Instead, we read them as two 8-bit values,
 	 * which need byteswapping then range adjustment.
 	 */
 	__be16 x;
@@ -68,6 +71,7 @@ struct ads7846 {
 	u16			vref_delay_usecs;
 	u16			x_plate_ohms;
 
+	u8			read_x, read_y, read_z1, read_z2;
 	struct ts_event		tc;
 
 	struct spi_transfer	xfer[8];
@@ -117,10 +121,10 @@ struct ads7846 {
 #define	READ_12BIT_DFR(x) (ADS_START | ADS_A2A1A0_d_ ## x \
 	| ADS_12_BIT | ADS_DFR)
 
-static const u8	read_y  = READ_12BIT_DFR(y)  | ADS_PD10_ADC_ON;
-static const u8	read_z1 = READ_12BIT_DFR(z1) | ADS_PD10_ADC_ON;
-static const u8	read_z2 = READ_12BIT_DFR(z2) | ADS_PD10_ADC_ON;
-static const u8	read_x  = READ_12BIT_DFR(x)  | ADS_PD10_PDOWN;	/* LAST */
+#define	READ_Y	(READ_12BIT_DFR(y)  | ADS_PD10_ADC_ON)
+#define	READ_Z1	(READ_12BIT_DFR(z1) | ADS_PD10_ADC_ON)
+#define	READ_Z2	(READ_12BIT_DFR(z2) | ADS_PD10_ADC_ON)
+#define	READ_X	(READ_12BIT_DFR(x)  | ADS_PD10_PDOWN)	/* LAST */
 
 /* single-ended samples need to first power up reference voltage;
  * we leave both ADC and VREF powered
@@ -128,8 +132,8 @@ static const u8	read_x  = READ_12BIT_DFR(x)  | ADS_PD10_PDOWN;	/* LAST */
 #define	READ_12BIT_SER(x) (ADS_START | ADS_A2A1A0_ ## x \
 	| ADS_12_BIT | ADS_SER)
 
-static const u8	ref_on = READ_12BIT_DFR(x) | ADS_PD10_ALL_ON;
-static const u8	ref_off = READ_12BIT_DFR(y) | ADS_PD10_PDOWN;
+#define	REF_ON	(READ_12BIT_DFR(x) | ADS_PD10_ALL_ON)
+#define	REF_OFF	(READ_12BIT_DFR(y) | ADS_PD10_PDOWN)
 
 /*--------------------------------------------------------------------------*/
 
@@ -138,7 +142,9 @@ static const u8	ref_off = READ_12BIT_DFR(y) | ADS_PD10_PDOWN;
  */
 
 struct ser_req {
+	u8			ref_on;
 	u8			command;
+	u8			ref_off;
 	u16			scratch;
 	__be16			sample;
 	struct spi_message	msg;
@@ -160,7 +166,8 @@ static int ads7846_read12_ser(struct device *dev, unsigned command)
 	INIT_LIST_HEAD(&req->msg.transfers);
 
 	/* activate reference, so it has time to settle; */
-	req->xfer[0].tx_buf = &ref_on;
+	req->ref_on = REF_ON;
+	req->xfer[0].tx_buf = &req->ref_on;
 	req->xfer[0].len = 1;
 	req->xfer[1].rx_buf = &req->scratch;
 	req->xfer[1].len = 2;
@@ -182,7 +189,8 @@ static int ads7846_read12_ser(struct device *dev, unsigned command)
 	/* REVISIT:  take a few more samples, and compare ... */
 
 	/* turn off reference */
-	req->xfer[4].tx_buf = &ref_off;
+	req->ref_off = REF_OFF;
+	req->xfer[4].tx_buf = &req->ref_off;
 	req->xfer[4].len = 1;
 	req->xfer[5].rx_buf = &req->scratch;
 	req->xfer[5].len = 2;
@@ -400,7 +408,6 @@ static int __devinit ads7846_probe(struct spi_device *spi)
 	struct input_dev		*input_dev;
 	struct ads7846_platform_data	*pdata = spi->dev.platform_data;
 	struct spi_transfer		*x;
-	int				i;
 	int				err;
 
 	if (!spi->irq) {
@@ -414,9 +421,9 @@ static int __devinit ads7846_probe(struct spi_device *spi)
 	}
 
 	/* don't exceed max specified sample rate */
-	if (spi->max_speed_hz > (125000 * 16)) {
+	if (spi->max_speed_hz > (125000 * SAMPLE_BITS)) {
 		dev_dbg(&spi->dev, "f(sample) %d KHz?\n",
-				(spi->max_speed_hz/16)/1000);
+				(spi->max_speed_hz/SAMPLE_BITS)/1000);
 		return -EINVAL;
 	}
 
@@ -469,45 +476,58 @@ static int __devinit ads7846_probe(struct spi_device *spi)
 	/* set up the transfers to read touchscreen state; this assumes we
 	 * use formula #2 for pressure, not #3.
 	 */
+	INIT_LIST_HEAD(&ts->msg.transfers);
 	x = ts->xfer;
 
 	/* y- still on; turn on only y+ (and ADC) */
-	x->tx_buf = &read_y;
+	ts->read_y = READ_Y;
+	x->tx_buf = &ts->read_y;
 	x->len = 1;
+	spi_message_add_tail(x, &ts->msg);
+
 	x++;
 	x->rx_buf = &ts->tc.y;
 	x->len = 2;
-	x++;
+	spi_message_add_tail(x, &ts->msg);
 
 	/* turn y+ off, x- on; we'll use formula #2 */
 	if (ts->model == 7846) {
-		x->tx_buf = &read_z1;
+		x++;
+		ts->read_z1 = READ_Z1;
+		x->tx_buf = &ts->read_z1;
 		x->len = 1;
+		spi_message_add_tail(x, &ts->msg);
+
 		x++;
 		x->rx_buf = &ts->tc.z1;
 		x->len = 2;
-		x++;
+		spi_message_add_tail(x, &ts->msg);
 
-		x->tx_buf = &read_z2;
+		x++;
+		ts->read_z2 = READ_Z2;
+		x->tx_buf = &ts->read_z2;
 		x->len = 1;
+		spi_message_add_tail(x, &ts->msg);
+
 		x++;
 		x->rx_buf = &ts->tc.z2;
 		x->len = 2;
-		x++;
+		spi_message_add_tail(x, &ts->msg);
 	}
 
 	/* turn y- off, x+ on, then leave in lowpower */
-	x->tx_buf = &read_x;
+	x++;
+	ts->read_x = READ_X;
+	x->tx_buf = &ts->read_x;
 	x->len = 1;
+	spi_message_add_tail(x, &ts->msg);
+
 	x++;
 	x->rx_buf = &ts->tc.x;
 	x->len = 2;
-	x++;
+	CS_CHANGE(*x);
+	spi_message_add_tail(x, &ts->msg);
 
-	CS_CHANGE(x[-1]);
-
-	for (i = 0; i < x - ts->xfer; i++)
-		spi_message_add_tail(&ts->xfer[i], &ts->msg);
 	ts->msg.complete = ads7846_rx;
 	ts->msg.context = ts;
 

From 02860ab6cd2c71dbe42fa70a65a97823c213635b Mon Sep 17 00:00:00 2001
From: Arthur Othieno <apgo@patchbomb.org>
Date: Wed, 15 Feb 2006 00:49:48 -0500
Subject: [PATCH 202/509] Input: kill remnants of 98kbd{,-io} and 98spkr

98kbd{,-io} and 98spkr all went out with PC98 subarch.  Remove stale Makefile
entries that remained.

Signed-off-by: Arthur Othieno <apgo@patchbomb.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dmitry Torokhov <dtor@mail.ru>
---
 drivers/input/keyboard/Makefile | 1 -
 drivers/input/misc/Makefile     | 1 -
 drivers/input/serio/Makefile    | 1 -
 3 files changed, 3 deletions(-)

diff --git a/drivers/input/keyboard/Makefile b/drivers/input/keyboard/Makefile
index 6e0afbb22383..2708167ba175 100644
--- a/drivers/input/keyboard/Makefile
+++ b/drivers/input/keyboard/Makefile
@@ -11,7 +11,6 @@ obj-$(CONFIG_KEYBOARD_XTKBD)		+= xtkbd.o
 obj-$(CONFIG_KEYBOARD_AMIGA)		+= amikbd.o
 obj-$(CONFIG_KEYBOARD_LOCOMO)		+= locomokbd.o
 obj-$(CONFIG_KEYBOARD_NEWTON)		+= newtonkbd.o
-obj-$(CONFIG_KEYBOARD_98KBD)		+= 98kbd.o
 obj-$(CONFIG_KEYBOARD_CORGI)		+= corgikbd.o
 obj-$(CONFIG_KEYBOARD_SPITZ)		+= spitzkbd.o
 obj-$(CONFIG_KEYBOARD_HIL)		+= hil_kbd.o
diff --git a/drivers/input/misc/Makefile b/drivers/input/misc/Makefile
index 184c4129470d..415c49178985 100644
--- a/drivers/input/misc/Makefile
+++ b/drivers/input/misc/Makefile
@@ -7,7 +7,6 @@
 obj-$(CONFIG_INPUT_SPARCSPKR)		+= sparcspkr.o
 obj-$(CONFIG_INPUT_PCSPKR)		+= pcspkr.o
 obj-$(CONFIG_INPUT_M68K_BEEP)		+= m68kspkr.o
-obj-$(CONFIG_INPUT_98SPKR)		+= 98spkr.o
 obj-$(CONFIG_INPUT_UINPUT)		+= uinput.o
 obj-$(CONFIG_INPUT_WISTRON_BTNS)	+= wistron_btns.o
 obj-$(CONFIG_HP_SDC_RTC)		+= hp_sdc_rtc.o
diff --git a/drivers/input/serio/Makefile b/drivers/input/serio/Makefile
index 678a8599f9ff..4155197867a3 100644
--- a/drivers/input/serio/Makefile
+++ b/drivers/input/serio/Makefile
@@ -13,7 +13,6 @@ obj-$(CONFIG_SERIO_RPCKBD)	+= rpckbd.o
 obj-$(CONFIG_SERIO_SA1111)	+= sa1111ps2.o
 obj-$(CONFIG_SERIO_AMBAKMI)	+= ambakmi.o
 obj-$(CONFIG_SERIO_Q40KBD)	+= q40kbd.o
-obj-$(CONFIG_SERIO_98KBD)	+= 98kbd-io.o
 obj-$(CONFIG_SERIO_GSCPS2)	+= gscps2.o
 obj-$(CONFIG_HP_SDC)		+= hp_sdc.o
 obj-$(CONFIG_HIL_MLC)		+= hp_sdc_mlc.o hil_mlc.o

From ee68cea2c26b7a8222f9020f54d22c6067011e8b Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 15 Feb 2006 01:34:23 -0800
Subject: [PATCH 203/509] [NETFILTER]: Fix xfrm lookup after SNAT

To find out if a packet needs to be handled by IPsec after SNAT, packets
are currently rerouted in POST_ROUTING and a new xfrm lookup is done. This
breaks SNAT of non-unicast packets to non-local addresses because the
packet is routed as incoming packet and no neighbour entry is bound to the
dst_entry. In general, it seems to be a bad idea to replace the dst_entry
after the packet was already sent to the output routine because its state
might not match what's expected.

This patch changes the xfrm lookup in POST_ROUTING to re-use the original
dst_entry without routing the packet again. This means no policy routing
can be used for transport mode transforms (which keep the original route)
when packets are SNATed to match the policy, but it looks like the best
we can do for now.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_ipv4.h         |  2 +-
 net/ipv4/netfilter.c                   | 41 ++++++++++++++++++++++++++
 net/ipv4/netfilter/ip_nat_standalone.c |  6 ++--
 3 files changed, 45 insertions(+), 4 deletions(-)

diff --git a/include/linux/netfilter_ipv4.h b/include/linux/netfilter_ipv4.h
index fdc4a9527343..43c09d790b83 100644
--- a/include/linux/netfilter_ipv4.h
+++ b/include/linux/netfilter_ipv4.h
@@ -79,7 +79,7 @@ enum nf_ip_hook_priorities {
 
 #ifdef __KERNEL__
 extern int ip_route_me_harder(struct sk_buff **pskb);
-
+extern int ip_xfrm_me_harder(struct sk_buff **pskb);
 #endif /*__KERNEL__*/
 
 #endif /*__LINUX_IP_NETFILTER_H*/
diff --git a/net/ipv4/netfilter.c b/net/ipv4/netfilter.c
index 52a3d7c57907..ed42cdc57cd9 100644
--- a/net/ipv4/netfilter.c
+++ b/net/ipv4/netfilter.c
@@ -78,6 +78,47 @@ int ip_route_me_harder(struct sk_buff **pskb)
 }
 EXPORT_SYMBOL(ip_route_me_harder);
 
+#ifdef CONFIG_XFRM
+int ip_xfrm_me_harder(struct sk_buff **pskb)
+{
+	struct flowi fl;
+	unsigned int hh_len;
+	struct dst_entry *dst;
+
+	if (IPCB(*pskb)->flags & IPSKB_XFRM_TRANSFORMED)
+		return 0;
+	if (xfrm_decode_session(*pskb, &fl, AF_INET) < 0)
+		return -1;
+
+	dst = (*pskb)->dst;
+	if (dst->xfrm)
+		dst = ((struct xfrm_dst *)dst)->route;
+	dst_hold(dst);
+
+	if (xfrm_lookup(&dst, &fl, (*pskb)->sk, 0) < 0)
+		return -1;
+
+	dst_release((*pskb)->dst);
+	(*pskb)->dst = dst;
+
+	/* Change in oif may mean change in hh_len. */
+	hh_len = (*pskb)->dst->dev->hard_header_len;
+	if (skb_headroom(*pskb) < hh_len) {
+		struct sk_buff *nskb;
+
+		nskb = skb_realloc_headroom(*pskb, hh_len);
+		if (!nskb)
+			return -1;
+		if ((*pskb)->sk)
+			skb_set_owner_w(nskb, (*pskb)->sk);
+		kfree_skb(*pskb);
+		*pskb = nskb;
+	}
+	return 0;
+}
+EXPORT_SYMBOL(ip_xfrm_me_harder);
+#endif
+
 void (*ip_nat_decode_session)(struct sk_buff *, struct flowi *);
 EXPORT_SYMBOL(ip_nat_decode_session);
 
diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index 92c54999a19d..7c3f7d380240 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -235,19 +235,19 @@ ip_nat_out(unsigned int hooknum,
 		return NF_ACCEPT;
 
 	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
+#ifdef CONFIG_XFRM
 	if (ret != NF_DROP && ret != NF_STOLEN
 	    && (ct = ip_conntrack_get(*pskb, &ctinfo)) != NULL) {
 		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
 
 		if (ct->tuplehash[dir].tuple.src.ip !=
 		    ct->tuplehash[!dir].tuple.dst.ip
-#ifdef CONFIG_XFRM
 		    || ct->tuplehash[dir].tuple.src.u.all !=
 		       ct->tuplehash[!dir].tuple.dst.u.all
-#endif
 		    )
-			return ip_route_me_harder(pskb) == 0 ? ret : NF_DROP;
+			return ip_xfrm_me_harder(pskb) == 0 ? ret : NF_DROP;
 	}
+#endif
 	return ret;
 }
 

From 78872ccb68335b14f0d1ac7338ecfcbf1cba1df4 Mon Sep 17 00:00:00 2001
From: Adrian Drzewiecki <z@drze.net>
Date: Wed, 15 Feb 2006 01:47:48 -0800
Subject: [PATCH 204/509] [BRIDGE]: Fix deadlock in br_stp_disable_bridge

Looks like somebody forgot to use the _bh spin_lock variant. We ran into a
deadlock where br->hello_timer expired while br_stp_disable_br() walked
br->port_list.

Signed-off-by: Adrian Drzewiecki <z@drze.net>
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_stp_if.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/bridge/br_stp_if.c b/net/bridge/br_stp_if.c
index cc047f7fb6ef..35cf3a074087 100644
--- a/net/bridge/br_stp_if.c
+++ b/net/bridge/br_stp_if.c
@@ -67,7 +67,7 @@ void br_stp_disable_bridge(struct net_bridge *br)
 {
 	struct net_bridge_port *p;
 
-	spin_lock(&br->lock);
+	spin_lock_bh(&br->lock);
 	list_for_each_entry(p, &br->port_list, list) {
 		if (p->state != BR_STATE_DISABLED)
 			br_stp_disable_port(p);
@@ -76,7 +76,7 @@ void br_stp_disable_bridge(struct net_bridge *br)
 
 	br->topology_change = 0;
 	br->topology_change_detected = 0;
-	spin_unlock(&br->lock);
+	spin_unlock_bh(&br->lock);
 
 	del_timer_sync(&br->hello_timer);
 	del_timer_sync(&br->topology_change_timer);

From a5f1e4edb3cdd90733893b8aec38fac5553db60a Mon Sep 17 00:00:00 2001
From: Arthur Othieno <apgo@patchbomb.org>
Date: Wed, 15 Feb 2006 09:52:46 +0000
Subject: [PATCH 205/509] [SERIAL] Documentation/jsm.txt is a no show.

In kernel bugzilla #5176 (http://bugzilla.kernel.org/show_bug.cgi?id=5176)
Harry R\374ter <h.rueter@gmx.de> points out Documentation/jsm.txt is missing.

No one at Digi seems to care, so just remove the stale reference.

Signed-off-by: Arthur Othieno <apgo@patchbomb.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/serial/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 0f4361c8466b..b3c561abe3f6 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -902,8 +902,8 @@ config SERIAL_JSM
 	  something like this to connect more than two modems to your Linux
 	  box, for instance in order to become a dial-in server. This driver
 	  supports PCI boards only.
-	  If you have a card like this, say Y here and read the file
-	  <file:Documentation/jsm.txt>.
+
+	  If you have a card like this, say Y here, otherwise say N.
 
 	  To compile this driver as a module, choose M here: the
 	  module will be called jsm.

From dc7bf130b8552a218e2f3ea0b58268e469f335da Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Wed, 15 Feb 2006 09:59:47 +0000
Subject: [PATCH 206/509] [SERIAL] Fix typo in comment

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/serial/8250.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index b1fc97d5f643..244e8ff11977 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -2198,7 +2198,7 @@ serial8250_console_write(struct console *co, const char *s, unsigned int count)
 	touch_nmi_watchdog();
 
 	/*
-	 *	First save the UER then disable the interrupts
+	 *	First save the IER then disable the interrupts
 	 */
 	ier = serial_in(up, UART_IER);
 

From ba09cf2bcf9b74d852dcb5ea957ac6af2bc0e057 Mon Sep 17 00:00:00 2001
From: Herbert Poetzl <herbert@13thfloor.at>
Date: Wed, 15 Feb 2006 10:13:02 +0000
Subject: [PATCH 207/509] [ARM] remove duplicate #includes

Signed-off-by: Herbert P?tzl <herbert@13thfloor.at>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-iop3xx/iop321-setup.c | 1 -
 arch/arm/mach-iop3xx/iop331-setup.c | 1 -
 arch/arm/plat-omap/pm.c             | 1 -
 drivers/video/s3c2410fb.c           | 1 -
 4 files changed, 4 deletions(-)

diff --git a/arch/arm/mach-iop3xx/iop321-setup.c b/arch/arm/mach-iop3xx/iop321-setup.c
index e4f4c52d93d4..0ebbcb20c6ae 100644
--- a/arch/arm/mach-iop3xx/iop321-setup.c
+++ b/arch/arm/mach-iop3xx/iop321-setup.c
@@ -13,7 +13,6 @@
 #include <linux/mm.h>
 #include <linux/init.h>
 #include <linux/config.h>
-#include <linux/init.h>
 #include <linux/major.h>
 #include <linux/fs.h>
 #include <linux/platform_device.h>
diff --git a/arch/arm/mach-iop3xx/iop331-setup.c b/arch/arm/mach-iop3xx/iop331-setup.c
index 63585485123e..2d6abe5be14d 100644
--- a/arch/arm/mach-iop3xx/iop331-setup.c
+++ b/arch/arm/mach-iop3xx/iop331-setup.c
@@ -12,7 +12,6 @@
 #include <linux/mm.h>
 #include <linux/init.h>
 #include <linux/config.h>
-#include <linux/init.h>
 #include <linux/major.h>
 #include <linux/fs.h>
 #include <linux/platform_device.h>
diff --git a/arch/arm/plat-omap/pm.c b/arch/arm/plat-omap/pm.c
index 1a24e2c10714..093efd786f21 100644
--- a/arch/arm/plat-omap/pm.c
+++ b/arch/arm/plat-omap/pm.c
@@ -38,7 +38,6 @@
 #include <linux/pm.h>
 #include <linux/sched.h>
 #include <linux/proc_fs.h>
-#include <linux/pm.h>
 #include <linux/interrupt.h>
 
 #include <asm/io.h>
diff --git a/drivers/video/s3c2410fb.c b/drivers/video/s3c2410fb.c
index d574dd3c9c8a..9451932fbaf2 100644
--- a/drivers/video/s3c2410fb.c
+++ b/drivers/video/s3c2410fb.c
@@ -82,7 +82,6 @@
 #include <linux/fb.h>
 #include <linux/init.h>
 #include <linux/dma-mapping.h>
-#include <linux/string.h>
 #include <linux/interrupt.h>
 #include <linux/workqueue.h>
 #include <linux/wait.h>

From 3f17da699431ec48540beabc55c54d4b5e66c8e7 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Wed, 15 Feb 2006 22:13:24 +0300
Subject: [PATCH 208/509] [PATCH] fix kill_proc_info() vs CLONE_THREAD race

There is a window after copy_process() unlocks ->sighand.siglock
and before it adds the new thread to the thread list.

In that window __group_complete_signal(SIGKILL) will not see the
new thread yet, so this thread will start running while the whole
thread group was supposed to exit.

I beleive we have another good reason to place attach_pid(PID/TGID)
under ->sighand.siglock. We can do the same for

	release_task()->__unhash_process()

	de_thread()->switch_exec_pids()

After that we don't need tasklist_lock to iterate over the thread
list, and we can simplify things, see for example do_sigaction()
or sys_times().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/fork.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/kernel/fork.c b/kernel/fork.c
index 8e88b374cee9..3683ce10f4a9 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1123,8 +1123,8 @@ static task_t *copy_process(unsigned long clone_flags,
 		p->real_parent = current;
 	p->parent = p->real_parent;
 
+	spin_lock(&current->sighand->siglock);
 	if (clone_flags & CLONE_THREAD) {
-		spin_lock(&current->sighand->siglock);
 		/*
 		 * Important: if an exit-all has been started then
 		 * do not create this new thread - the whole thread
@@ -1162,8 +1162,6 @@ static task_t *copy_process(unsigned long clone_flags,
 			 */
 			p->it_prof_expires = jiffies_to_cputime(1);
 		}
-
-		spin_unlock(&current->sighand->siglock);
 	}
 
 	/*
@@ -1189,6 +1187,7 @@ static task_t *copy_process(unsigned long clone_flags,
 
 	nr_threads++;
 	total_forks++;
+	spin_unlock(&current->sighand->siglock);
 	write_unlock_irq(&tasklist_lock);
 	proc_fork_connector(p);
 	return p;

From dadac81b1b86196fcc48fb87620403c4a7174f06 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Wed, 15 Feb 2006 22:13:26 +0300
Subject: [PATCH 209/509] [PATCH] fix kill_proc_info() vs fork() theoretical
 race

copy_process:

	attach_pid(p, PIDTYPE_PID, p->pid);
	attach_pid(p, PIDTYPE_TGID, p->tgid);

What if kill_proc_info(p->pid) happens in between?

copy_process() holds current->sighand.siglock, so we are safe
in CLONE_THREAD case, because current->sighand == p->sighand.

Otherwise, p->sighand is unlocked, the new process is already
visible to the find_task_by_pid(), but have a copy of parent's
'struct pid' in ->pids[PIDTYPE_TGID].

This means that __group_complete_signal() may hang while doing

	do ... while (next_thread() != p)

We can solve this problem if we reverse these 2 attach_pid()s:

	attach_pid() does wmb()

	group_send_sig_info() calls spin_lock(), which
	provides a read barrier. // Yes ?

I don't think we can hit this race in practice, but still.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/fork.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/fork.c b/kernel/fork.c
index 3683ce10f4a9..fbea12d7a943 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1173,8 +1173,6 @@ static task_t *copy_process(unsigned long clone_flags,
 	if (unlikely(p->ptrace & PT_PTRACED))
 		__ptrace_link(p, current->parent);
 
-	attach_pid(p, PIDTYPE_PID, p->pid);
-	attach_pid(p, PIDTYPE_TGID, p->tgid);
 	if (thread_group_leader(p)) {
 		p->signal->tty = current->signal->tty;
 		p->signal->pgrp = process_group(current);
@@ -1184,6 +1182,8 @@ static task_t *copy_process(unsigned long clone_flags,
 		if (p->pid)
 			__get_cpu_var(process_counts)++;
 	}
+	attach_pid(p, PIDTYPE_TGID, p->tgid);
+	attach_pid(p, PIDTYPE_PID, p->pid);
 
 	nr_threads++;
 	total_forks++;

From 5ecfbae093f0c37311e89b29bfc0c9d586eace87 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Wed, 15 Feb 2006 22:50:10 +0300
Subject: [PATCH 210/509] [PATCH] fix zap_thread's ptrace related problems

1. The tracee can go from ptrace_stop() to do_signal_stop()
   after __ptrace_unlink(p).

2. It is unsafe to __ptrace_unlink(p) while p->parent may wait
   for tasklist_lock in ptrace_detach().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/exec.c              |  2 +-
 include/linux/ptrace.h |  1 +
 kernel/ptrace.c        | 29 +++++++++++++++++------------
 3 files changed, 19 insertions(+), 13 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 055378d2513e..0e1c95074d42 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1403,7 +1403,7 @@ static void zap_threads (struct mm_struct *mm)
 		do_each_thread(g,p) {
 			if (mm == p->mm && p != tsk &&
 			    p->ptrace && p->parent->mm == mm) {
-				__ptrace_unlink(p);
+				__ptrace_detach(p, 0);
 			}
 		} while_each_thread(g,p);
 		write_unlock_irq(&tasklist_lock);
diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index 9d5cd106b344..0d36750fc0f1 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -84,6 +84,7 @@ extern int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __us
 extern int ptrace_writedata(struct task_struct *tsk, char __user *src, unsigned long dst, int len);
 extern int ptrace_attach(struct task_struct *tsk);
 extern int ptrace_detach(struct task_struct *, unsigned int);
+extern void __ptrace_detach(struct task_struct *, unsigned int);
 extern void ptrace_disable(struct task_struct *);
 extern int ptrace_check_attach(struct task_struct *task, int kill);
 extern int ptrace_request(struct task_struct *child, long request, long addr, long data);
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index d2cf144d0af5..d95a72c9279d 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -72,8 +72,8 @@ void ptrace_untrace(task_t *child)
  */
 void __ptrace_unlink(task_t *child)
 {
-	if (!child->ptrace)
-		BUG();
+	BUG_ON(!child->ptrace);
+
 	child->ptrace = 0;
 	if (!list_empty(&child->ptrace_list)) {
 		list_del_init(&child->ptrace_list);
@@ -184,22 +184,27 @@ bad:
 	return retval;
 }
 
-int ptrace_detach(struct task_struct *child, unsigned int data)
+void __ptrace_detach(struct task_struct *child, unsigned int data)
 {
-	if (!valid_signal(data))
-		return	-EIO;
-
-	/* Architecture-specific hardware disable .. */
-	ptrace_disable(child);
-
-	/* .. re-parent .. */
 	child->exit_code = data;
-
-	write_lock_irq(&tasklist_lock);
+	/* .. re-parent .. */
 	__ptrace_unlink(child);
 	/* .. and wake it up. */
 	if (child->exit_state != EXIT_ZOMBIE)
 		wake_up_process(child);
+}
+
+int ptrace_detach(struct task_struct *child, unsigned int data)
+{
+	if (!valid_signal(data))
+		return -EIO;
+
+	/* Architecture-specific hardware disable .. */
+	ptrace_disable(child);
+
+	write_lock_irq(&tasklist_lock);
+	if (child->ptrace)
+		__ptrace_detach(child, data);
 	write_unlock_irq(&tasklist_lock);
 
 	return 0;

From 50d8e59038703c4da5acaed9afaa37ae416d3153 Mon Sep 17 00:00:00 2001
From: Andreas Schwab <schwab@suse.de>
Date: Sun, 12 Feb 2006 17:01:35 +0100
Subject: [PATCH 211/509] [IA64] Remove duplicate EXPORT_SYMBOLs

Remove symbol exports from ia64_ksyms.c that are already exported in
lib/string.c.

Signed-off-by: Andreas Schwab <schwab@suse.de>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/ia64_ksyms.c | 15 ---------------
 1 file changed, 15 deletions(-)

diff --git a/arch/ia64/kernel/ia64_ksyms.c b/arch/ia64/kernel/ia64_ksyms.c
index e72de580ebbf..bbcfd08378a6 100644
--- a/arch/ia64/kernel/ia64_ksyms.c
+++ b/arch/ia64/kernel/ia64_ksyms.c
@@ -10,23 +10,8 @@
 
 #include <linux/string.h>
 EXPORT_SYMBOL(memset);
-EXPORT_SYMBOL(memchr);
-EXPORT_SYMBOL(memcmp);
 EXPORT_SYMBOL(memcpy);
-EXPORT_SYMBOL(memmove);
-EXPORT_SYMBOL(memscan);
-EXPORT_SYMBOL(strcat);
-EXPORT_SYMBOL(strchr);
-EXPORT_SYMBOL(strcmp);
-EXPORT_SYMBOL(strcpy);
 EXPORT_SYMBOL(strlen);
-EXPORT_SYMBOL(strncat);
-EXPORT_SYMBOL(strncmp);
-EXPORT_SYMBOL(strncpy);
-EXPORT_SYMBOL(strnlen);
-EXPORT_SYMBOL(strrchr);
-EXPORT_SYMBOL(strstr);
-EXPORT_SYMBOL(strpbrk);
 
 #include <asm/checksum.h>
 EXPORT_SYMBOL(ip_fast_csum);		/* hand-coded assembly */

From 8ed9b2c7a804335004e4bd3b4c6989c5b6bc243f Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Mon, 13 Feb 2006 05:29:57 -0500
Subject: [PATCH 212/509] [IA64-SGI] sn2 minor fixes and cleanups

General SN2 code cleanup:
 - Do not initialize global variables to zero
 - Use kzalloc instead of kmalloc+memset
 - Check kmalloc return values
 - Do not obfuscate spin lock calls
 - Remove some unused code
 - Various formatting cleanups

Signed-off-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/io_init.c           | 96 ++++++++++---------------
 arch/ia64/sn/kernel/sn2/prominfo_proc.c | 25 +++----
 arch/ia64/sn/kernel/sn2/sn2_smp.c       | 35 +++++----
 arch/ia64/sn/kernel/sn2/sn_proc_fs.c    | 22 +++---
 arch/ia64/sn/kernel/tiocx.c             |  4 +-
 arch/ia64/sn/pci/pci_dma.c              | 16 ++---
 arch/ia64/sn/pci/pcibr/pcibr_ate.c      | 29 +++-----
 arch/ia64/sn/pci/pcibr/pcibr_dma.c      | 14 ++--
 arch/ia64/sn/pci/pcibr/pcibr_provider.c |  9 ++-
 include/asm-ia64/sn/bte.h               |  6 +-
 include/asm-ia64/sn/pcibr_provider.h    | 14 +---
 include/asm-ia64/sn/sn_feature_sets.h   |  3 -
 12 files changed, 111 insertions(+), 162 deletions(-)

diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c
index 3437c2390429..dfb3f2902379 100644
--- a/arch/ia64/sn/kernel/io_init.c
+++ b/arch/ia64/sn/kernel/io_init.c
@@ -23,6 +23,10 @@
 #include "xtalk/hubdev.h"
 #include "xtalk/xwidgetdev.h"
 
+
+extern void sn_init_cpei_timer(void);
+extern void register_sn_procfs(void);
+
 static struct list_head sn_sysdata_list;
 
 /* sysdata list struct */
@@ -40,12 +44,12 @@ struct brick {
 	struct slab_info slab_info[MAX_SLABS + 1];
 };
 
-int sn_ioif_inited = 0;		/* SN I/O infrastructure initialized? */
+int sn_ioif_inited;		/* SN I/O infrastructure initialized? */
 
 struct sn_pcibus_provider *sn_pci_provider[PCIIO_ASIC_MAX_TYPES];	/* indexed by asic type */
 
-static int max_segment_number = 0; /* Default highest segment number */
-static int max_pcibus_number = 255; /* Default highest pci bus number */
+static int max_segment_number;		 /* Default highest segment number */
+static int max_pcibus_number = 255;	/* Default highest pci bus number */
 
 /*
  * Hooks and struct for unsupported pci providers
@@ -84,7 +88,6 @@ static inline u64
 sal_get_device_dmaflush_list(u64 nasid, u64 widget_num, u64 device_num,
 			     u64 address)
 {
-
 	struct ia64_sal_retval ret_stuff;
 	ret_stuff.status = 0;
 	ret_stuff.v0 = 0;
@@ -94,7 +97,6 @@ sal_get_device_dmaflush_list(u64 nasid, u64 widget_num, u64 device_num,
 			(u64) nasid, (u64) widget_num,
 			(u64) device_num, (u64) address, 0, 0, 0);
 	return ret_stuff.status;
-
 }
 
 /*
@@ -102,7 +104,6 @@ sal_get_device_dmaflush_list(u64 nasid, u64 widget_num, u64 device_num,
  */
 static inline u64 sal_get_hubdev_info(u64 handle, u64 address)
 {
-
 	struct ia64_sal_retval ret_stuff;
 	ret_stuff.status = 0;
 	ret_stuff.v0 = 0;
@@ -118,7 +119,6 @@ static inline u64 sal_get_hubdev_info(u64 handle, u64 address)
  */
 static inline u64 sal_get_pcibus_info(u64 segment, u64 busnum, u64 address)
 {
-
 	struct ia64_sal_retval ret_stuff;
 	ret_stuff.status = 0;
 	ret_stuff.v0 = 0;
@@ -215,7 +215,7 @@ static void __init sn_fixup_ionodes(void)
 	struct hubdev_info *hubdev;
 	u64 status;
 	u64 nasid;
-	int i, widget, device;
+	int i, widget, device, size;
 
 	/*
 	 * Get SGI Specific HUB chipset information.
@@ -251,48 +251,37 @@ static void __init sn_fixup_ionodes(void)
 		if (!hubdev->hdi_flush_nasid_list.widget_p)
 			continue;
 
+		size = (HUB_WIDGET_ID_MAX + 1) *
+			sizeof(struct sn_flush_device_kernel *);
 		hubdev->hdi_flush_nasid_list.widget_p =
-		    kmalloc((HUB_WIDGET_ID_MAX + 1) *
-			    sizeof(struct sn_flush_device_kernel *),
-			    GFP_KERNEL);
-		memset(hubdev->hdi_flush_nasid_list.widget_p, 0x0,
-		       (HUB_WIDGET_ID_MAX + 1) *
-		       sizeof(struct sn_flush_device_kernel *));
+			kzalloc(size, GFP_KERNEL);
+		if (!hubdev->hdi_flush_nasid_list.widget_p)
+			BUG();
 
 		for (widget = 0; widget <= HUB_WIDGET_ID_MAX; widget++) {
-			sn_flush_device_kernel = kmalloc(DEV_PER_WIDGET *
-						         sizeof(struct
-						        sn_flush_device_kernel),
-						        GFP_KERNEL);
+			size = DEV_PER_WIDGET *
+				sizeof(struct sn_flush_device_kernel);
+			sn_flush_device_kernel = kzalloc(size, GFP_KERNEL);
 			if (!sn_flush_device_kernel)
 				BUG();
-			memset(sn_flush_device_kernel, 0x0,
-			       DEV_PER_WIDGET *
-			       sizeof(struct sn_flush_device_kernel));
 
 			dev_entry = sn_flush_device_kernel;
 			for (device = 0; device < DEV_PER_WIDGET;
 			     device++,dev_entry++) {
-				dev_entry->common = kmalloc(sizeof(struct
-					      	        sn_flush_device_common),
-					                    GFP_KERNEL);
+				size = sizeof(struct sn_flush_device_common);
+				dev_entry->common = kzalloc(size, GFP_KERNEL);
 				if (!dev_entry->common)
 					BUG();
-				memset(dev_entry->common, 0x0, sizeof(struct
-					     	       sn_flush_device_common));
 
 				if (sn_prom_feature_available(
 						       PRF_DEVICE_FLUSH_LIST))
 					status = sal_get_device_dmaflush_list(
-									  nasid,
-									 widget,
-								       	 device,
-						      (u64)(dev_entry->common));
+						     nasid, widget, device,
+						     (u64)(dev_entry->common));
 				else
 					status = sn_device_fixup_war(nasid,
-								     widget,
-							    	     device,
-							     dev_entry->common);
+						     widget, device,
+						     dev_entry->common);
 				if (status != SALRET_OK)
 					panic("SAL call failed: %s\n",
 					      ia64_sal_strerror(status));
@@ -383,13 +372,12 @@ void sn_pci_fixup_slot(struct pci_dev *dev)
 
 	pci_dev_get(dev); /* for the sysdata pointer */
 	pcidev_info = kzalloc(sizeof(struct pcidev_info), GFP_KERNEL);
-	if (pcidev_info <= 0)
+	if (!pcidev_info)
 		BUG();		/* Cannot afford to run out of memory */
 
-	sn_irq_info = kmalloc(sizeof(struct sn_irq_info), GFP_KERNEL);
-	if (sn_irq_info <= 0)
+	sn_irq_info = kzalloc(sizeof(struct sn_irq_info), GFP_KERNEL);
+	if (!sn_irq_info)
 		BUG();		/* Cannot afford to run out of memory */
-	memset(sn_irq_info, 0, sizeof(struct sn_irq_info));
 
 	/* Call to retrieve pci device information needed by kernel. */
 	status = sal_get_pcidev_info((u64) segment, (u64) dev->bus->number, 
@@ -482,13 +470,13 @@ void sn_pci_fixup_slot(struct pci_dev *dev)
  */
 void sn_pci_controller_fixup(int segment, int busnum, struct pci_bus *bus)
 {
-	int status = 0;
+	int status;
 	int nasid, cnode;
 	struct pci_controller *controller;
 	struct sn_pci_controller *sn_controller;
 	struct pcibus_bussoft *prom_bussoft_ptr;
 	struct hubdev_info *hubdev_info;
-	void *provider_soft = NULL;
+	void *provider_soft;
 	struct sn_pcibus_provider *provider;
 
  	status = sal_get_pcibus_info((u64) segment, (u64) busnum,
@@ -535,6 +523,8 @@ void sn_pci_controller_fixup(int segment, int busnum, struct pci_bus *bus)
 	bus->sysdata = controller;
 	if (provider->bus_fixup)
 		provider_soft = (*provider->bus_fixup) (prom_bussoft_ptr, controller);
+	else
+		provider_soft = NULL;
 
 	if (provider_soft == NULL) {
 		/* fixup failed or not applicable */
@@ -638,13 +628,8 @@ void sn_bus_free_sysdata(void)
 
 static int __init sn_pci_init(void)
 {
-	int i = 0;
-	int j = 0;
+	int i, j;
 	struct pci_dev *pci_dev = NULL;
-	extern void sn_init_cpei_timer(void);
-#ifdef CONFIG_PROC_FS
-	extern void register_sn_procfs(void);
-#endif
 
 	if (!ia64_platform_is("sn2") || IS_RUNNING_ON_FAKE_PROM())
 		return 0;
@@ -700,32 +685,29 @@ static int __init sn_pci_init(void)
  */
 void hubdev_init_node(nodepda_t * npda, cnodeid_t node)
 {
-
 	struct hubdev_info *hubdev_info;
+	int size;
+	pg_data_t *pg;
+
+	size = sizeof(struct hubdev_info);
 
 	if (node >= num_online_nodes())	/* Headless/memless IO nodes */
-		hubdev_info =
-		    (struct hubdev_info *)alloc_bootmem_node(NODE_DATA(0),
-							     sizeof(struct
-								    hubdev_info));
+		pg = NODE_DATA(0);
 	else
-		hubdev_info =
-		    (struct hubdev_info *)alloc_bootmem_node(NODE_DATA(node),
-							     sizeof(struct
-								    hubdev_info));
-	npda->pdinfo = (void *)hubdev_info;
+		pg = NODE_DATA(node);
 
+	hubdev_info = (struct hubdev_info *)alloc_bootmem_node(pg, size);
+
+	npda->pdinfo = (void *)hubdev_info;
 }
 
 geoid_t
 cnodeid_get_geoid(cnodeid_t cnode)
 {
-
 	struct hubdev_info *hubdev;
 
 	hubdev = (struct hubdev_info *)(NODEPDA(cnode)->pdinfo);
 	return hubdev->hdi_geoid;
-
 }
 
 subsys_initcall(sn_pci_init);
diff --git a/arch/ia64/sn/kernel/sn2/prominfo_proc.c b/arch/ia64/sn/kernel/sn2/prominfo_proc.c
index 81c63b2f8ae9..6ae276d5d50c 100644
--- a/arch/ia64/sn/kernel/sn2/prominfo_proc.c
+++ b/arch/ia64/sn/kernel/sn2/prominfo_proc.c
@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 1999,2001-2004 Silicon Graphics, Inc.  All Rights Reserved.
+ * Copyright (C) 1999,2001-2004, 2006 Silicon Graphics, Inc.  All Rights Reserved.
  *
  * Module to export the system's Firmware Interface Tables, including
  * PROM revision numbers and banners, in /proc
@@ -190,7 +190,7 @@ static int
 read_version_entry(char *page, char **start, off_t off, int count, int *eof,
 		   void *data)
 {
-	int len = 0;
+	int len;
 
 	/* data holds the NASID of the node */
 	len = dump_version(page, (unsigned long)data);
@@ -202,7 +202,7 @@ static int
 read_fit_entry(char *page, char **start, off_t off, int count, int *eof,
 	       void *data)
 {
-	int len = 0;
+	int len;
 
 	/* data holds the NASID of the node */
 	len = dump_fit(page, (unsigned long)data);
@@ -229,13 +229,16 @@ int __init prominfo_init(void)
 	struct proc_dir_entry *p;
 	cnodeid_t cnodeid;
 	unsigned long nasid;
+	int size;
 	char name[NODE_NAME_LEN];
 
 	if (!ia64_platform_is("sn2"))
 		return 0;
 
-	proc_entries = kmalloc(num_online_nodes() * sizeof(struct proc_dir_entry *),
-			       GFP_KERNEL);
+	size = num_online_nodes() * sizeof(struct proc_dir_entry *);
+	proc_entries = kzalloc(size, GFP_KERNEL);
+	if (!proc_entries)
+		return -ENOMEM;
 
 	sgi_prominfo_entry = proc_mkdir("sgi_prominfo", NULL);
 
@@ -244,14 +247,12 @@ int __init prominfo_init(void)
 		sprintf(name, "node%d", cnodeid);
 		*entp = proc_mkdir(name, sgi_prominfo_entry);
 		nasid = cnodeid_to_nasid(cnodeid);
-		p = create_proc_read_entry(
-			"fit", 0, *entp, read_fit_entry,
-			(void *)nasid);
+		p = create_proc_read_entry("fit", 0, *entp, read_fit_entry,
+					   (void *)nasid);
 		if (p)
 			p->owner = THIS_MODULE;
-		p = create_proc_read_entry(
-			"version", 0, *entp, read_version_entry,
-			(void *)nasid);
+		p = create_proc_read_entry("version", 0, *entp,
+					   read_version_entry, (void *)nasid);
 		if (p)
 			p->owner = THIS_MODULE;
 		entp++;
@@ -263,7 +264,7 @@ int __init prominfo_init(void)
 void __exit prominfo_exit(void)
 {
 	struct proc_dir_entry **entp;
-	unsigned cnodeid;
+	unsigned int cnodeid;
 	char name[NODE_NAME_LEN];
 
 	entp = proc_entries;
diff --git a/arch/ia64/sn/kernel/sn2/sn2_smp.c b/arch/ia64/sn/kernel/sn2/sn2_smp.c
index f153a4c35c70..24eefb2fc55f 100644
--- a/arch/ia64/sn/kernel/sn2/sn2_smp.c
+++ b/arch/ia64/sn/kernel/sn2/sn2_smp.c
@@ -46,8 +46,14 @@ DECLARE_PER_CPU(struct ptc_stats, ptcstats);
 
 static  __cacheline_aligned DEFINE_SPINLOCK(sn2_global_ptc_lock);
 
-void sn2_ptc_deadlock_recovery(short *, short, short, int, volatile unsigned long *, unsigned long,
-	volatile unsigned long *, unsigned long);
+extern unsigned long
+sn2_ptc_deadlock_recovery_core(volatile unsigned long *, unsigned long,
+			       volatile unsigned long *, unsigned long,
+			       volatile unsigned long *, unsigned long);
+void
+sn2_ptc_deadlock_recovery(short *, short, short, int,
+			  volatile unsigned long *, unsigned long,
+			  volatile unsigned long *, unsigned long);
 
 /*
  * Note: some is the following is captured here to make degugging easier
@@ -59,16 +65,6 @@ void sn2_ptc_deadlock_recovery(short *, short, short, int, volatile unsigned lon
 #define reset_max_active_on_deadlock()	1
 #define PTC_LOCK(sh1)			((sh1) ? &sn2_global_ptc_lock : &sn_nodepda->ptc_lock)
 
-static inline void ptc_lock(int sh1, unsigned long *flagp)
-{
-	spin_lock_irqsave(PTC_LOCK(sh1), *flagp);
-}
-
-static inline void ptc_unlock(int sh1, unsigned long flags)
-{
-	spin_unlock_irqrestore(PTC_LOCK(sh1), flags);
-}
-
 struct ptc_stats {
 	unsigned long ptc_l;
 	unsigned long change_rid;
@@ -82,6 +78,8 @@ struct ptc_stats {
 	unsigned long shub_ptc_flushes_not_my_mm;
 };
 
+#define sn2_ptctest	0
+
 static inline unsigned long wait_piowc(void)
 {
 	volatile unsigned long *piows;
@@ -200,7 +198,7 @@ sn2_global_tlb_purge(struct mm_struct *mm, unsigned long start,
 	max_active = max_active_pio(shub1);
 
 	itc = ia64_get_itc();
-	ptc_lock(shub1, &flags);
+	spin_lock_irqsave(PTC_LOCK(shub1), flags);
 	itc2 = ia64_get_itc();
 
 	__get_cpu_var(ptcstats).lock_itc_clocks += itc2 - itc;
@@ -258,7 +256,7 @@ sn2_global_tlb_purge(struct mm_struct *mm, unsigned long start,
 		ia64_srlz_d();
 	}
 
-	ptc_unlock(shub1, flags);
+	spin_unlock_irqrestore(PTC_LOCK(shub1), flags);
 
 	preempt_enable();
 }
@@ -270,11 +268,12 @@ sn2_global_tlb_purge(struct mm_struct *mm, unsigned long start,
  * TLB flush transaction.  The recovery sequence is somewhat tricky & is
  * coded in assembly language.
  */
-void sn2_ptc_deadlock_recovery(short *nasids, short ib, short ie, int mynasid, volatile unsigned long *ptc0, unsigned long data0,
-	volatile unsigned long *ptc1, unsigned long data1)
+
+void
+sn2_ptc_deadlock_recovery(short *nasids, short ib, short ie, int mynasid,
+			  volatile unsigned long *ptc0, unsigned long data0,
+			  volatile unsigned long *ptc1, unsigned long data1)
 {
-	extern unsigned long sn2_ptc_deadlock_recovery_core(volatile unsigned long *, unsigned long,
-	        volatile unsigned long *, unsigned long, volatile unsigned long *, unsigned long);
 	short nasid, i;
 	unsigned long *piows, zeroval, n;
 
diff --git a/arch/ia64/sn/kernel/sn2/sn_proc_fs.c b/arch/ia64/sn/kernel/sn2/sn_proc_fs.c
index a06719d752a0..c686d9c12f7b 100644
--- a/arch/ia64/sn/kernel/sn2/sn_proc_fs.c
+++ b/arch/ia64/sn/kernel/sn2/sn_proc_fs.c
@@ -6,11 +6,11 @@
  * Copyright (C) 2000-2005 Silicon Graphics, Inc. All rights reserved.
  */
 #include <linux/config.h>
-#include <asm/uaccess.h>
 
 #ifdef CONFIG_PROC_FS
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
+#include <asm/uaccess.h>
 #include <asm/sn/sn_sal.h>
 
 static int partition_id_show(struct seq_file *s, void *p)
@@ -90,10 +90,10 @@ static int coherence_id_open(struct inode *inode, struct file *file)
 	return single_open(file, coherence_id_show, NULL);
 }
 
-static struct proc_dir_entry *sn_procfs_create_entry(
-	const char *name, struct proc_dir_entry *parent,
-	int (*openfunc)(struct inode *, struct file *),
-	int (*releasefunc)(struct inode *, struct file *))
+static struct proc_dir_entry
+*sn_procfs_create_entry(const char *name, struct proc_dir_entry *parent,
+			int (*openfunc)(struct inode *, struct file *),
+			int (*releasefunc)(struct inode *, struct file *))
 {
 	struct proc_dir_entry *e = create_proc_entry(name, 0444, parent);
 
@@ -126,24 +126,24 @@ void register_sn_procfs(void)
 		return;
 
 	sn_procfs_create_entry("partition_id", sgi_proc_dir,
-		partition_id_open, single_release);
+			       partition_id_open, single_release);
 
 	sn_procfs_create_entry("system_serial_number", sgi_proc_dir,
-		system_serial_number_open, single_release);
+			       system_serial_number_open, single_release);
 
 	sn_procfs_create_entry("licenseID", sgi_proc_dir, 
-		licenseID_open, single_release);
+			       licenseID_open, single_release);
 
 	e = sn_procfs_create_entry("sn_force_interrupt", sgi_proc_dir, 
-		sn_force_interrupt_open, single_release);
+				   sn_force_interrupt_open, single_release);
 	if (e) 
 		e->proc_fops->write = sn_force_interrupt_write_proc;
 
 	sn_procfs_create_entry("coherence_id", sgi_proc_dir, 
-		coherence_id_open, single_release);
+			       coherence_id_open, single_release);
 	
 	sn_procfs_create_entry("sn_topology", sgi_proc_dir,
-		sn_topology_open, sn_topology_release);
+			       sn_topology_open, sn_topology_release);
 }
 
 #endif /* CONFIG_PROC_FS */
diff --git a/arch/ia64/sn/kernel/tiocx.c b/arch/ia64/sn/kernel/tiocx.c
index d263d3e8fbb9..8a56f8b5ffa2 100644
--- a/arch/ia64/sn/kernel/tiocx.c
+++ b/arch/ia64/sn/kernel/tiocx.c
@@ -284,12 +284,10 @@ struct sn_irq_info *tiocx_irq_alloc(nasid_t nasid, int widget, int irq,
 	if ((nasid & 1) == 0)
 		return NULL;
 
-	sn_irq_info = kmalloc(sn_irq_size, GFP_KERNEL);
+	sn_irq_info = kzalloc(sn_irq_size, GFP_KERNEL);
 	if (sn_irq_info == NULL)
 		return NULL;
 
-	memset(sn_irq_info, 0x0, sn_irq_size);
-
 	status = tiocx_intr_alloc(nasid, widget, __pa(sn_irq_info), irq,
 				  req_nasid, slice);
 	if (status) {
diff --git a/arch/ia64/sn/pci/pci_dma.c b/arch/ia64/sn/pci/pci_dma.c
index 5a36292388eb..b4b84c269210 100644
--- a/arch/ia64/sn/pci/pci_dma.c
+++ b/arch/ia64/sn/pci/pci_dma.c
@@ -335,10 +335,10 @@ int sn_pci_legacy_read(struct pci_bus *bus, u16 port, u32 *val, u8 size)
 	 */
 
 	SAL_CALL(isrv, SN_SAL_IOIF_PCI_SAFE,
-		pci_domain_nr(bus), bus->number,
-		0, /* io */
-		0, /* read */
-		port, size, __pa(val));
+		 pci_domain_nr(bus), bus->number,
+		 0, /* io */
+		 0, /* read */
+		 port, size, __pa(val));
 
 	if (isrv.status == 0)
 		return size;
@@ -381,10 +381,10 @@ int sn_pci_legacy_write(struct pci_bus *bus, u16 port, u32 val, u8 size)
 	 */
 
 	SAL_CALL(isrv, SN_SAL_IOIF_PCI_SAFE,
-		pci_domain_nr(bus), bus->number,
-		0, /* io */
-		1, /* write */
-		port, size, __pa(&val));
+		 pci_domain_nr(bus), bus->number,
+		 0, /* io */
+		 1, /* write */
+		 port, size, __pa(&val));
 
 	if (isrv.status == 0)
 		return size;
diff --git a/arch/ia64/sn/pci/pcibr/pcibr_ate.c b/arch/ia64/sn/pci/pcibr/pcibr_ate.c
index aa3fa5152a32..1f0253bfe0a0 100644
--- a/arch/ia64/sn/pci/pcibr/pcibr_ate.c
+++ b/arch/ia64/sn/pci/pcibr/pcibr_ate.c
@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 2001-2004 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (C) 2001-2006 Silicon Graphics, Inc. All rights reserved.
  */
 
 #include <linux/types.h>
@@ -12,7 +12,7 @@
 #include <asm/sn/pcibus_provider_defs.h>
 #include <asm/sn/pcidev.h>
 
-int pcibr_invalidate_ate = 0;	/* by default don't invalidate ATE on free */
+int pcibr_invalidate_ate;	/* by default don't invalidate ATE on free */
 
 /*
  * mark_ate: Mark the ate as either free or inuse.
@@ -20,14 +20,12 @@ int pcibr_invalidate_ate = 0;	/* by default don't invalidate ATE on free */
 static void mark_ate(struct ate_resource *ate_resource, int start, int number,
 		     u64 value)
 {
-
 	u64 *ate = ate_resource->ate;
 	int index;
 	int length = 0;
 
 	for (index = start; length < number; index++, length++)
 		ate[index] = value;
-
 }
 
 /*
@@ -37,7 +35,6 @@ static void mark_ate(struct ate_resource *ate_resource, int start, int number,
 static int find_free_ate(struct ate_resource *ate_resource, int start,
 			 int count)
 {
-
 	u64 *ate = ate_resource->ate;
 	int index;
 	int start_free;
@@ -70,12 +67,10 @@ static int find_free_ate(struct ate_resource *ate_resource, int start,
 static inline void free_ate_resource(struct ate_resource *ate_resource,
 				     int start)
 {
-
 	mark_ate(ate_resource, start, ate_resource->ate[start], 0);
 	if ((ate_resource->lowest_free_index > start) ||
 	    (ate_resource->lowest_free_index < 0))
 		ate_resource->lowest_free_index = start;
-
 }
 
 /*
@@ -84,7 +79,6 @@ static inline void free_ate_resource(struct ate_resource *ate_resource,
 static inline int alloc_ate_resource(struct ate_resource *ate_resource,
 				     int ate_needed)
 {
-
 	int start_index;
 
 	/*
@@ -118,19 +112,12 @@ static inline int alloc_ate_resource(struct ate_resource *ate_resource,
  */
 int pcibr_ate_alloc(struct pcibus_info *pcibus_info, int count)
 {
-	int status = 0;
-	u64 flag;
+	int status;
+	unsigned long flags;
 
-	flag = pcibr_lock(pcibus_info);
+	spin_lock_irqsave(&pcibus_info->pbi_lock, flags);
 	status = alloc_ate_resource(&pcibus_info->pbi_int_ate_resource, count);
-
-	if (status < 0) {
-		/* Failed to allocate */
-		pcibr_unlock(pcibus_info, flag);
-		return -1;
-	}
-
-	pcibr_unlock(pcibus_info, flag);
+	spin_unlock_irqrestore(&pcibus_info->pbi_lock, flags);
 
 	return status;
 }
@@ -182,7 +169,7 @@ void pcibr_ate_free(struct pcibus_info *pcibus_info, int index)
 		ate_write(pcibus_info, index, count, (ate & ~PCI32_ATE_V));
 	}
 
-	flags = pcibr_lock(pcibus_info);
+	spin_lock_irqsave(&pcibus_info->pbi_lock, flags);
 	free_ate_resource(&pcibus_info->pbi_int_ate_resource, index);
-	pcibr_unlock(pcibus_info, flags);
+	spin_unlock_irqrestore(&pcibus_info->pbi_lock, flags);
 }
diff --git a/arch/ia64/sn/pci/pcibr/pcibr_dma.c b/arch/ia64/sn/pci/pcibr/pcibr_dma.c
index 54ce5b7ceed2..9f86bb6519aa 100644
--- a/arch/ia64/sn/pci/pcibr/pcibr_dma.c
+++ b/arch/ia64/sn/pci/pcibr/pcibr_dma.c
@@ -137,14 +137,12 @@ pcibr_dmatrans_direct64(struct pcidev_info * info, u64 paddr,
 		pci_addr |= PCI64_ATTR_VIRTUAL;
 
 	return pci_addr;
-
 }
 
 static dma_addr_t
 pcibr_dmatrans_direct32(struct pcidev_info * info,
 			u64 paddr, size_t req_size, u64 flags)
 {
-
 	struct pcidev_info *pcidev_info = info->pdi_host_pcidev_info;
 	struct pcibus_info *pcibus_info = (struct pcibus_info *)pcidev_info->
 	    pdi_pcibus_info;
@@ -171,7 +169,6 @@ pcibr_dmatrans_direct32(struct pcidev_info * info,
 	}
 
 	return PCI32_DIRECT_BASE | offset;
-
 }
 
 /*
@@ -218,9 +215,8 @@ void sn_dma_flush(u64 addr)
 	u64 flags;
 	u64 itte;
 	struct hubdev_info *hubinfo;
-	volatile struct sn_flush_device_kernel *p;
-	volatile struct sn_flush_device_common *common;
-
+	struct sn_flush_device_kernel *p;
+	struct sn_flush_device_common *common;
 	struct sn_flush_nasid_entry *flush_nasid_list;
 
 	if (!sn_ioif_inited)
@@ -310,8 +306,7 @@ void sn_dma_flush(u64 addr)
 					     (common->sfdl_slot - 1));
 		}
 	} else {
-		spin_lock_irqsave((spinlock_t *)&p->sfdl_flush_lock,
-				  flags);
+		spin_lock_irqsave(&p->sfdl_flush_lock, flags);
 		*common->sfdl_flush_addr = 0;
 
 		/* force an interrupt. */
@@ -322,8 +317,7 @@ void sn_dma_flush(u64 addr)
 			cpu_relax();
 
 		/* okay, everything is synched up. */
-		spin_unlock_irqrestore((spinlock_t *)&p->sfdl_flush_lock,
-				       flags);
+		spin_unlock_irqrestore(&p->sfdl_flush_lock, flags);
 	}
 	return;
 }
diff --git a/arch/ia64/sn/pci/pcibr/pcibr_provider.c b/arch/ia64/sn/pci/pcibr/pcibr_provider.c
index 2fac27049bf6..98f716bd92f0 100644
--- a/arch/ia64/sn/pci/pcibr/pcibr_provider.c
+++ b/arch/ia64/sn/pci/pcibr/pcibr_provider.c
@@ -163,9 +163,12 @@ pcibr_bus_fixup(struct pcibus_bussoft *prom_bussoft, struct pci_controller *cont
 	/* Setup the PMU ATE map */
 	soft->pbi_int_ate_resource.lowest_free_index = 0;
 	soft->pbi_int_ate_resource.ate =
-	    kmalloc(soft->pbi_int_ate_size * sizeof(u64), GFP_KERNEL);
-	memset(soft->pbi_int_ate_resource.ate, 0,
- 	       (soft->pbi_int_ate_size * sizeof(u64)));
+	    kzalloc(soft->pbi_int_ate_size * sizeof(u64), GFP_KERNEL);
+
+	if (!soft->pbi_int_ate_resource.ate) {
+		kfree(soft);
+		return NULL;
+	}
 
 	if (prom_bussoft->bs_asic_type == PCIIO_ASIC_TYPE_TIOCP) {
 		/* TIO PCI Bridge: find nearest node with CPUs */
diff --git a/include/asm-ia64/sn/bte.h b/include/asm-ia64/sn/bte.h
index 01e5b4103235..5335d87ca5f8 100644
--- a/include/asm-ia64/sn/bte.h
+++ b/include/asm-ia64/sn/bte.h
@@ -46,7 +46,7 @@
 #define BTES_PER_NODE (is_shub2() ? 4 : 2)
 #define MAX_BTES_PER_NODE 4
 
-#define BTE2OFF_CTRL	(0)
+#define BTE2OFF_CTRL	0
 #define BTE2OFF_SRC	(SH2_BT_ENG_SRC_ADDR_0 - SH2_BT_ENG_CSR_0)
 #define BTE2OFF_DEST	(SH2_BT_ENG_DEST_ADDR_0 - SH2_BT_ENG_CSR_0)
 #define BTE2OFF_NOTIFY	(SH2_BT_ENG_NOTIF_ADDR_0 - SH2_BT_ENG_CSR_0)
@@ -75,11 +75,11 @@
 		: base + (BTEOFF_NOTIFY/8))
 
 /* Define hardware modes */
-#define BTE_NOTIFY (IBCT_NOTIFY)
+#define BTE_NOTIFY IBCT_NOTIFY
 #define BTE_NORMAL BTE_NOTIFY
 #define BTE_ZERO_FILL (BTE_NOTIFY | IBCT_ZFIL_MODE)
 /* Use a reserved bit to let the caller specify a wait for any BTE */
-#define BTE_WACQUIRE (0x4000)
+#define BTE_WACQUIRE 0x4000
 /* Use the BTE on the node with the destination memory */
 #define BTE_USE_DEST (BTE_WACQUIRE << 1)
 /* Use any available BTE interface on any node for the transfer */
diff --git a/include/asm-ia64/sn/pcibr_provider.h b/include/asm-ia64/sn/pcibr_provider.h
index 9334078b089a..a601d3af39b6 100644
--- a/include/asm-ia64/sn/pcibr_provider.h
+++ b/include/asm-ia64/sn/pcibr_provider.h
@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 1992-1997,2000-2004 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (C) 1992-1997,2000-2006 Silicon Graphics, Inc. All rights reserved.
  */
 #ifndef _ASM_IA64_SN_PCI_PCIBR_PROVIDER_H
 #define _ASM_IA64_SN_PCI_PCIBR_PROVIDER_H
@@ -115,18 +115,6 @@ struct pcibus_info {
 	spinlock_t              pbi_lock;
 };
 
-/*
- * pcibus_info structure locking macros
- */
-inline static unsigned long
-pcibr_lock(struct pcibus_info *pcibus_info)
-{
-	unsigned long flag;
-	spin_lock_irqsave(&pcibus_info->pbi_lock, flag);
-	return(flag);
-}
-#define pcibr_unlock(pcibus_info, flag)  spin_unlock_irqrestore(&pcibus_info->pbi_lock, flag)
-
 extern int  pcibr_init_provider(void);
 extern void *pcibr_bus_fixup(struct pcibus_bussoft *, struct pci_controller *);
 extern dma_addr_t pcibr_dma_map(struct pci_dev *, unsigned long, size_t);
diff --git a/include/asm-ia64/sn/sn_feature_sets.h b/include/asm-ia64/sn/sn_feature_sets.h
index 9ca642cad338..ff33e3bd3f8e 100644
--- a/include/asm-ia64/sn/sn_feature_sets.h
+++ b/include/asm-ia64/sn/sn_feature_sets.h
@@ -12,9 +12,6 @@
  */
 
 
-#include <asm/types.h>
-#include <asm/bitops.h>
-
 /* --------------------- PROM Features -----------------------------*/
 extern int sn_prom_feature_available(int id);
 

From d3454344b3507042e5d561d0cfed19e99cf2fc88 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Mon, 13 Feb 2006 05:32:09 -0500
Subject: [PATCH 213/509] [IA64] remove obsolete corporate address

Remove obsolete SGI address

Signed-off-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/sn2/timer_interrupt.c | 7 +------
 drivers/ide/pci/sgiioc4.c                 | 5 -----
 include/asm-ia64/machvec_sn2.h            | 7 +------
 3 files changed, 2 insertions(+), 17 deletions(-)

diff --git a/arch/ia64/sn/kernel/sn2/timer_interrupt.c b/arch/ia64/sn/kernel/sn2/timer_interrupt.c
index adf5db2e2afe..fa7f69945917 100644
--- a/arch/ia64/sn/kernel/sn2/timer_interrupt.c
+++ b/arch/ia64/sn/kernel/sn2/timer_interrupt.c
@@ -1,7 +1,7 @@
 /*
  *
  *
- * Copyright (c) 2005 Silicon Graphics, Inc.  All Rights Reserved.
+ * Copyright (c) 2005, 2006 Silicon Graphics, Inc.  All Rights Reserved.
  * 
  * This program is free software; you can redistribute it and/or modify it 
  * under the terms of version 2 of the GNU General Public License 
@@ -22,11 +22,6 @@
  * License along with this program; if not, write the Free Software 
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
  * 
- * Contact information:  Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, 
- * Mountain View, CA  94043, or:
- * 
- * http://www.sgi.com 
- * 
  * For further information regarding this notice, see: 
  * 
  * http://oss.sgi.com/projects/GenInfo/NoticeExplan
diff --git a/drivers/ide/pci/sgiioc4.c b/drivers/ide/pci/sgiioc4.c
index 2b286e865163..43b96e298363 100644
--- a/drivers/ide/pci/sgiioc4.c
+++ b/drivers/ide/pci/sgiioc4.c
@@ -13,11 +13,6 @@
  * License along with this program; if not, write the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
  *
- * Contact information:  Silicon Graphics, Inc., 1600 Amphitheatre Pkwy,
- * Mountain View, CA  94043, or:
- *
- * http://www.sgi.com
- *
  * For further information regarding this notice, see:
  *
  * http://oss.sgi.com/projects/GenInfo/NoticeExplan
diff --git a/include/asm-ia64/machvec_sn2.h b/include/asm-ia64/machvec_sn2.h
index e1b6cd63f49e..03d00faf03b5 100644
--- a/include/asm-ia64/machvec_sn2.h
+++ b/include/asm-ia64/machvec_sn2.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2002-2003 Silicon Graphics, Inc.  All Rights Reserved.
+ * Copyright (c) 2002-2003, 2006 Silicon Graphics, Inc.  All Rights Reserved.
  * 
  * This program is free software; you can redistribute it and/or modify it 
  * under the terms of version 2 of the GNU General Public License 
@@ -20,11 +20,6 @@
  * License along with this program; if not, write the Free Software 
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307, USA.
  * 
- * Contact information:  Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, 
- * Mountain View, CA  94043, or:
- * 
- * http://www.sgi.com 
- * 
  * For further information regarding this notice, see: 
  * 
  * http://oss.sgi.com/projects/GenInfo/NoticeExplan

From 26d10915de3030a55253dba3b2b145402cdf6429 Mon Sep 17 00:00:00 2001
From: Jes Sorensen <jes@sgi.com>
Date: Mon, 13 Feb 2006 05:35:01 -0500
Subject: [PATCH 214/509] [IA64-SGI] remove compile time warning

This one falls into the "present for Andrew Morton" category to address
his wishlist for a compiler warning free build ;-)

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/setup.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c
index 48645ac120fc..1672ecad77e2 100644
--- a/arch/ia64/sn/kernel/setup.c
+++ b/arch/ia64/sn/kernel/setup.c
@@ -317,6 +317,7 @@ struct pcdp_vga_device {
 #define PCDP_PCI_TRANS_IOPORT	0x02
 #define PCDP_PCI_TRANS_MMIO	0x01
 
+#if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
 static void
 sn_scan_pcdp(void)
 {
@@ -358,6 +359,7 @@ sn_scan_pcdp(void)
 		break; /* once we find the primary, we're done */
 	}
 }
+#endif
 
 static unsigned long sn2_rtc_initial;
 

From 9c65cb9be62ac4993a5b392304b82e4f04f010fd Mon Sep 17 00:00:00 2001
From: Mark Maule <maule@sgi.com>
Date: Tue, 14 Feb 2006 10:23:37 -0600
Subject: [PATCH 215/509] [IA64-SGI] export sn_pcidev_info_get

Export sn_pcidev_info_get.

Signed-off-by Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/io_init.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/ia64/sn/kernel/io_init.c b/arch/ia64/sn/kernel/io_init.c
index dfb3f2902379..3edef0d32f86 100644
--- a/arch/ia64/sn/kernel/io_init.c
+++ b/arch/ia64/sn/kernel/io_init.c
@@ -716,3 +716,4 @@ EXPORT_SYMBOL(sn_pci_unfixup_slot);
 EXPORT_SYMBOL(sn_pci_controller_fixup);
 EXPORT_SYMBOL(sn_bus_store_sysdata);
 EXPORT_SYMBOL(sn_bus_free_sysdata);
+EXPORT_SYMBOL(sn_pcidev_info_get);

From c2a4969ba14e852bf4ee92c7db3b0cf82405a0c9 Mon Sep 17 00:00:00 2001
From: Dean Roe <roe@sgi.com>
Date: Tue, 14 Feb 2006 15:01:23 -0600
Subject: [PATCH 216/509] [IA64-SGI] fix the size of __sn_cnodeid_to_nasid

The __sn_cnodeid_to_nasid array was incorrectly sized at MAX_NUMNODES.
On a large system, this array could overflow.  The following patch
corrects this by defining it to MAX_COMPACT_NODES.

Signed-off-by: Dean Roe <roe@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/setup.c | 2 +-
 include/asm-ia64/sn/arch.h  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/ia64/sn/kernel/setup.c b/arch/ia64/sn/kernel/setup.c
index 1672ecad77e2..5b84836c2171 100644
--- a/arch/ia64/sn/kernel/setup.c
+++ b/arch/ia64/sn/kernel/setup.c
@@ -75,7 +75,7 @@ EXPORT_SYMBOL(sn_rtc_cycles_per_second);
 DEFINE_PER_CPU(struct sn_hub_info_s, __sn_hub_info);
 EXPORT_PER_CPU_SYMBOL(__sn_hub_info);
 
-DEFINE_PER_CPU(short, __sn_cnodeid_to_nasid[MAX_NUMNODES]);
+DEFINE_PER_CPU(short, __sn_cnodeid_to_nasid[MAX_COMPACT_NODES]);
 EXPORT_PER_CPU_SYMBOL(__sn_cnodeid_to_nasid);
 
 DEFINE_PER_CPU(struct nodepda_s *, __sn_nodepda);
diff --git a/include/asm-ia64/sn/arch.h b/include/asm-ia64/sn/arch.h
index 1a3831c04af6..91c31be87b13 100644
--- a/include/asm-ia64/sn/arch.h
+++ b/include/asm-ia64/sn/arch.h
@@ -70,7 +70,7 @@ DECLARE_PER_CPU(struct sn_hub_info_s, __sn_hub_info);
  * Compact node ID to nasid mappings kept in the per-cpu data areas of each
  * cpu.
  */
-DECLARE_PER_CPU(short, __sn_cnodeid_to_nasid[MAX_NUMNODES]);
+DECLARE_PER_CPU(short, __sn_cnodeid_to_nasid[MAX_COMPACT_NODES]);
 #define sn_cnodeid_to_nasid	(&__get_cpu_var(__sn_cnodeid_to_nasid[0]))
 
 

From 4c2cd96696ae0896ce4bcf725b9f0eaffafeb640 Mon Sep 17 00:00:00 2001
From: Dean Nelson <dcn@sgi.com>
Date: Wed, 15 Feb 2006 08:02:21 -0600
Subject: [PATCH 217/509] [IA64-SGI] enforce proper ordering of callouts by XPC

Fix XPC so that it does not deliver any messages until the connected
callout has returned, as well as, prevent the disconnected callout to
occur before the disconnecting callout has returned.

Signed-off-by: Dean Nelson <dcn@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/sn/kernel/xpc_channel.c |  8 +++++---
 arch/ia64/sn/kernel/xpc_main.c    | 20 +++++++++++++-------
 include/asm-ia64/sn/xpc.h         | 27 ++++++++++++++++-----------
 3 files changed, 34 insertions(+), 21 deletions(-)

diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c
index 36e5437a0fb6..cdf6856ce089 100644
--- a/arch/ia64/sn/kernel/xpc_channel.c
+++ b/arch/ia64/sn/kernel/xpc_channel.c
@@ -738,7 +738,9 @@ xpc_process_disconnect(struct xpc_channel *ch, unsigned long *irq_flags)
 
 	/* make sure all activity has settled down first */
 
-	if (atomic_read(&ch->references) > 0) {
+	if (atomic_read(&ch->references) > 0 ||
+			((ch->flags & XPC_C_CONNECTEDCALLOUT_MADE) &&
+			!(ch->flags & XPC_C_DISCONNECTINGCALLOUT_MADE))) {
 		return;
 	}
 	DBUG_ON(atomic_read(&ch->kthreads_assigned) != 0);
@@ -775,7 +777,7 @@ xpc_process_disconnect(struct xpc_channel *ch, unsigned long *irq_flags)
 
 	/* both sides are disconnected now */
 
-	if (ch->flags & XPC_C_CONNECTCALLOUT) {
+	if (ch->flags & XPC_C_DISCONNECTINGCALLOUT_MADE) {
 		spin_unlock_irqrestore(&ch->lock, *irq_flags);
 		xpc_disconnect_callout(ch, xpcDisconnected);
 		spin_lock_irqsave(&ch->lock, *irq_flags);
@@ -1300,7 +1302,7 @@ xpc_process_msg_IPI(struct xpc_partition *part, int ch_number)
 				"delivered=%d, partid=%d, channel=%d\n",
 				nmsgs_sent, ch->partid, ch->number);
 
-			if (ch->flags & XPC_C_CONNECTCALLOUT) {
+			if (ch->flags & XPC_C_CONNECTEDCALLOUT_MADE) {
 				xpc_activate_kthreads(ch, nmsgs_sent);
 			}
 		}
diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c
index 9cd460dfe27e..8cbf16432570 100644
--- a/arch/ia64/sn/kernel/xpc_main.c
+++ b/arch/ia64/sn/kernel/xpc_main.c
@@ -750,12 +750,16 @@ xpc_daemonize_kthread(void *args)
 		/* let registerer know that connection has been established */
 
 		spin_lock_irqsave(&ch->lock, irq_flags);
-		if (!(ch->flags & XPC_C_CONNECTCALLOUT)) {
-			ch->flags |= XPC_C_CONNECTCALLOUT;
+		if (!(ch->flags & XPC_C_CONNECTEDCALLOUT)) {
+			ch->flags |= XPC_C_CONNECTEDCALLOUT;
 			spin_unlock_irqrestore(&ch->lock, irq_flags);
 
 			xpc_connected_callout(ch);
 
+			spin_lock_irqsave(&ch->lock, irq_flags);
+			ch->flags |= XPC_C_CONNECTEDCALLOUT_MADE;
+			spin_unlock_irqrestore(&ch->lock, irq_flags);
+
 			/*
 			 * It is possible that while the callout was being
 			 * made that the remote partition sent some messages.
@@ -777,15 +781,17 @@ xpc_daemonize_kthread(void *args)
 
 	if (atomic_dec_return(&ch->kthreads_assigned) == 0) {
 		spin_lock_irqsave(&ch->lock, irq_flags);
-		if ((ch->flags & XPC_C_CONNECTCALLOUT) &&
-				!(ch->flags & XPC_C_DISCONNECTCALLOUT)) {
-			ch->flags |= XPC_C_DISCONNECTCALLOUT;
+		if ((ch->flags & XPC_C_CONNECTEDCALLOUT_MADE) &&
+				!(ch->flags & XPC_C_DISCONNECTINGCALLOUT)) {
+			ch->flags |= XPC_C_DISCONNECTINGCALLOUT;
 			spin_unlock_irqrestore(&ch->lock, irq_flags);
 
 			xpc_disconnect_callout(ch, xpcDisconnecting);
-		} else {
-			spin_unlock_irqrestore(&ch->lock, irq_flags);
+
+			spin_lock_irqsave(&ch->lock, irq_flags);
+			ch->flags |= XPC_C_DISCONNECTINGCALLOUT_MADE;
 		}
+		spin_unlock_irqrestore(&ch->lock, irq_flags);
 		if (atomic_dec_return(&part->nchannels_engaged) == 0) {
 			xpc_mark_partition_disengaged(part);
 			xpc_IPI_send_disengage(part);
diff --git a/include/asm-ia64/sn/xpc.h b/include/asm-ia64/sn/xpc.h
index 0c36928ffd8b..df7f5f4f3cde 100644
--- a/include/asm-ia64/sn/xpc.h
+++ b/include/asm-ia64/sn/xpc.h
@@ -508,19 +508,24 @@ struct xpc_channel {
 #define	XPC_C_OPENREQUEST	0x00000010 /* local open channel request */
 
 #define	XPC_C_SETUP		0x00000020 /* channel's msgqueues are alloc'd */
-#define	XPC_C_CONNECTCALLOUT	0x00000040 /* channel connected callout made */
-#define	XPC_C_CONNECTED		0x00000080 /* local channel is connected */
-#define	XPC_C_CONNECTING	0x00000100 /* channel is being connected */
+#define	XPC_C_CONNECTEDCALLOUT	0x00000040 /* connected callout initiated */
+#define	XPC_C_CONNECTEDCALLOUT_MADE \
+				0x00000080 /* connected callout completed */
+#define	XPC_C_CONNECTED		0x00000100 /* local channel is connected */
+#define	XPC_C_CONNECTING	0x00000200 /* channel is being connected */
 
-#define	XPC_C_RCLOSEREPLY	0x00000200 /* remote close channel reply */
-#define	XPC_C_CLOSEREPLY	0x00000400 /* local close channel reply */
-#define	XPC_C_RCLOSEREQUEST	0x00000800 /* remote close channel request */
-#define	XPC_C_CLOSEREQUEST	0x00001000 /* local close channel request */
+#define	XPC_C_RCLOSEREPLY	0x00000400 /* remote close channel reply */
+#define	XPC_C_CLOSEREPLY	0x00000800 /* local close channel reply */
+#define	XPC_C_RCLOSEREQUEST	0x00001000 /* remote close channel request */
+#define	XPC_C_CLOSEREQUEST	0x00002000 /* local close channel request */
 
-#define	XPC_C_DISCONNECTED	0x00002000 /* channel is disconnected */
-#define	XPC_C_DISCONNECTING	0x00004000 /* channel is being disconnected */
-#define	XPC_C_DISCONNECTCALLOUT	0x00008000 /* chan disconnected callout made */
-#define	XPC_C_WDISCONNECT	0x00010000 /* waiting for channel disconnect */
+#define	XPC_C_DISCONNECTED	0x00004000 /* channel is disconnected */
+#define	XPC_C_DISCONNECTING	0x00008000 /* channel is being disconnected */
+#define	XPC_C_DISCONNECTINGCALLOUT \
+				0x00010000 /* disconnecting callout initiated */
+#define	XPC_C_DISCONNECTINGCALLOUT_MADE \
+				0x00020000 /* disconnecting callout completed */
+#define	XPC_C_WDISCONNECT	0x00040000 /* waiting for channel disconnect */
 
 
 

From defbb2c929cbe89dc92239b303cd33d3c85e9a83 Mon Sep 17 00:00:00 2001
From: "hawkes@sgi.com" <hawkes@sgi.com>
Date: Tue, 14 Feb 2006 10:40:17 -0800
Subject: [PATCH 218/509] [IA64] ia64: simplify and fix udelay()

The original ia64 udelay() was simple, but flawed for platforms without
synchronized ITCs:  a preemption and migration to another CPU during the
while-loop likely resulted in too-early termination or very, very
lengthy looping.

The first fix (now in 2.6.15) broke the delay loop into smaller,
non-preemptible chunks, reenabling preemption between the chunks.  This
fix is flawed in that the total udelay is computed to be the sum of just
the non-premptible while-loop pieces, i.e., not counting the time spent
in the interim preemptible periods.  If an interrupt or a migration
occurs during one of these interim periods, then that time is invisible
and only serves to lengthen the effective udelay().

This new fix backs out the current flawed fix and returns to a simple
udelay(), fully preemptible and interruptible.  It implements two simple
alternative udelay() routines:  one a default generic version that uses
ia64_get_itc(), and the other an sn-specific version that uses that
platform's RTC.

Signed-off-by: John Hawkes <hawkes@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/time.c         | 39 ++++++++++++++-------------------
 arch/ia64/sn/kernel/sn2/timer.c | 19 ++++++++++++++++
 include/asm-ia64/timex.h        |  2 ++
 3 files changed, 38 insertions(+), 22 deletions(-)

diff --git a/arch/ia64/kernel/time.c b/arch/ia64/kernel/time.c
index a094ec49ccfa..307d01e15b2e 100644
--- a/arch/ia64/kernel/time.c
+++ b/arch/ia64/kernel/time.c
@@ -250,32 +250,27 @@ time_init (void)
 	set_normalized_timespec(&wall_to_monotonic, -xtime.tv_sec, -xtime.tv_nsec);
 }
 
-#define SMALLUSECS 100
+/*
+ * Generic udelay assumes that if preemption is allowed and the thread
+ * migrates to another CPU, that the ITC values are synchronized across
+ * all CPUs.
+ */
+static void
+ia64_itc_udelay (unsigned long usecs)
+{
+	unsigned long start = ia64_get_itc();
+	unsigned long end = start + usecs*local_cpu_data->cyc_per_usec;
+
+	while (time_before(ia64_get_itc(), end))
+		cpu_relax();
+}
+
+void (*ia64_udelay)(unsigned long usecs) = &ia64_itc_udelay;
 
 void
 udelay (unsigned long usecs)
 {
-	unsigned long start;
-	unsigned long cycles;
-	unsigned long smallusecs;
-
-	/*
-	 * Execute the non-preemptible delay loop (because the ITC might
-	 * not be synchronized between CPUS) in relatively short time
-	 * chunks, allowing preemption between the chunks.
-	 */
-	while (usecs > 0) {
-		smallusecs = (usecs > SMALLUSECS) ? SMALLUSECS : usecs;
-		preempt_disable();
-		cycles = smallusecs*local_cpu_data->cyc_per_usec;
-		start = ia64_get_itc();
-
-		while (ia64_get_itc() - start < cycles)
-			cpu_relax();
-
-		preempt_enable();
-		usecs -= smallusecs;
-	}
+	(*ia64_udelay)(usecs);
 }
 EXPORT_SYMBOL(udelay);
 
diff --git a/arch/ia64/sn/kernel/sn2/timer.c b/arch/ia64/sn/kernel/sn2/timer.c
index deb9baf4d473..56a88b6df4b4 100644
--- a/arch/ia64/sn/kernel/sn2/timer.c
+++ b/arch/ia64/sn/kernel/sn2/timer.c
@@ -14,6 +14,7 @@
 
 #include <asm/hw_irq.h>
 #include <asm/system.h>
+#include <asm/timex.h>
 
 #include <asm/sn/leds.h>
 #include <asm/sn/shub_mmr.h>
@@ -28,9 +29,27 @@ static struct time_interpolator sn2_interpolator = {
 	.source = TIME_SOURCE_MMIO64
 };
 
+/*
+ * sn udelay uses the RTC instead of the ITC because the ITC is not
+ * synchronized across all CPUs, and the thread may migrate to another CPU
+ * if preemption is enabled.
+ */
+static void
+ia64_sn_udelay (unsigned long usecs)
+{
+	unsigned long start = rtc_time();
+	unsigned long end = start +
+			usecs * sn_rtc_cycles_per_second / 1000000;
+
+	while (time_before((unsigned long)rtc_time(), end))
+		cpu_relax();
+}
+
 void __init sn_timer_init(void)
 {
 	sn2_interpolator.frequency = sn_rtc_cycles_per_second;
 	sn2_interpolator.addr = RTC_COUNTER_ADDR;
 	register_time_interpolator(&sn2_interpolator);
+
+	ia64_udelay = &ia64_sn_udelay;
 }
diff --git a/include/asm-ia64/timex.h b/include/asm-ia64/timex.h
index 414aae060440..05a6baf8a472 100644
--- a/include/asm-ia64/timex.h
+++ b/include/asm-ia64/timex.h
@@ -15,6 +15,8 @@
 
 typedef unsigned long cycles_t;
 
+extern void (*ia64_udelay)(unsigned long usecs);
+
 /*
  * For performance reasons, we don't want to define CLOCK_TICK_TRATE as
  * local_cpu_data->itc_rate.  Fortunately, we don't have to, either: according to George

From 48d5cad87c3a4998d0bda16ccfb5c60dfe4de5fb Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 15 Feb 2006 15:10:22 -0800
Subject: [PATCH 219/509] [XFRM]: Fix SNAT-related crash in xfrm4_output_finish

When a packet matching an IPsec policy is SNATed so it doesn't match any
policy anymore it looses its xfrm bundle, which makes xfrm4_output_finish
crash because of a NULL pointer dereference.

This patch directs these packets to the original output path instead. Since
the packets have already passed the POST_ROUTING hook, but need to start at
the beginning of the original output path which includes another
POST_ROUTING invocation, a flag is added to the IPCB to indicate that the
packet was rerouted and doesn't need to pass the POST_ROUTING hook again.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h | 19 +++++++++++++++----
 include/net/ip.h          |  1 +
 include/net/xfrm.h        |  1 -
 net/ipv4/ip_gre.c         |  3 ++-
 net/ipv4/ip_output.c      | 16 ++++++++++------
 net/ipv4/ipip.c           |  3 ++-
 net/ipv4/xfrm4_output.c   | 13 ++++++++++---
 7 files changed, 40 insertions(+), 16 deletions(-)

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 4cf6088625c1..3ca3d9ee78a9 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -184,8 +184,11 @@ static inline int nf_hook_thresh(int pf, unsigned int hook,
 				 struct sk_buff **pskb,
 				 struct net_device *indev,
 				 struct net_device *outdev,
-				 int (*okfn)(struct sk_buff *), int thresh)
+				 int (*okfn)(struct sk_buff *), int thresh,
+				 int cond)
 {
+	if (!cond)
+		return 1;
 #ifndef CONFIG_NETFILTER_DEBUG
 	if (list_empty(&nf_hooks[pf][hook]))
 		return 1;
@@ -197,7 +200,7 @@ static inline int nf_hook(int pf, unsigned int hook, struct sk_buff **pskb,
 			  struct net_device *indev, struct net_device *outdev,
 			  int (*okfn)(struct sk_buff *))
 {
-	return nf_hook_thresh(pf, hook, pskb, indev, outdev, okfn, INT_MIN);
+	return nf_hook_thresh(pf, hook, pskb, indev, outdev, okfn, INT_MIN, 1);
 }
                    
 /* Activate hook; either okfn or kfree_skb called, unless a hook
@@ -224,7 +227,13 @@ static inline int nf_hook(int pf, unsigned int hook, struct sk_buff **pskb,
 
 #define NF_HOOK_THRESH(pf, hook, skb, indev, outdev, okfn, thresh)	       \
 ({int __ret;								       \
-if ((__ret=nf_hook_thresh(pf, hook, &(skb), indev, outdev, okfn, thresh)) == 1)\
+if ((__ret=nf_hook_thresh(pf, hook, &(skb), indev, outdev, okfn, thresh, 1)) == 1)\
+	__ret = (okfn)(skb);						       \
+__ret;})
+
+#define NF_HOOK_COND(pf, hook, skb, indev, outdev, okfn, cond)		       \
+({int __ret;								       \
+if ((__ret=nf_hook_thresh(pf, hook, &(skb), indev, outdev, okfn, INT_MIN, cond)) == 1)\
 	__ret = (okfn)(skb);						       \
 __ret;})
 
@@ -295,11 +304,13 @@ extern struct proc_dir_entry *proc_net_netfilter;
 
 #else /* !CONFIG_NETFILTER */
 #define NF_HOOK(pf, hook, skb, indev, outdev, okfn) (okfn)(skb)
+#define NF_HOOK_COND(pf, hook, skb, indev, outdev, okfn, cond) (okfn)(skb)
 static inline int nf_hook_thresh(int pf, unsigned int hook,
 				 struct sk_buff **pskb,
 				 struct net_device *indev,
 				 struct net_device *outdev,
-				 int (*okfn)(struct sk_buff *), int thresh)
+				 int (*okfn)(struct sk_buff *), int thresh,
+				 int cond)
 {
 	return okfn(*pskb);
 }
diff --git a/include/net/ip.h b/include/net/ip.h
index 8de0697b364c..fab3d5b3ab1c 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -41,6 +41,7 @@ struct inet_skb_parm
 #define IPSKB_XFRM_TUNNEL_SIZE	2
 #define IPSKB_XFRM_TRANSFORMED	4
 #define IPSKB_FRAG_COMPLETE	8
+#define IPSKB_REROUTED		16
 };
 
 struct ipcm_cookie
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index d09ca0e7d139..d6111a2f0a23 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -866,7 +866,6 @@ extern int xfrm_state_mtu(struct xfrm_state *x, int mtu);
 extern int xfrm_init_state(struct xfrm_state *x);
 extern int xfrm4_rcv(struct sk_buff *skb);
 extern int xfrm4_output(struct sk_buff *skb);
-extern int xfrm4_output_finish(struct sk_buff *skb);
 extern int xfrm4_tunnel_register(struct xfrm_tunnel *handler);
 extern int xfrm4_tunnel_deregister(struct xfrm_tunnel *handler);
 extern int xfrm6_rcv_spi(struct sk_buff **pskb, u32 spi);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index abe23923e4e7..9981dcd68f11 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -830,7 +830,8 @@ static int ipgre_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	skb->h.raw = skb->nh.raw;
 	skb->nh.raw = skb_push(skb, gre_hlen);
 	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
-	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE|IPSKB_XFRM_TRANSFORMED);
+	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
+			      IPSKB_REROUTED);
 	dst_release(skb->dst);
 	skb->dst = &rt->u.dst;
 
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 3324fbfe528a..57d290d89ec2 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -207,8 +207,10 @@ static inline int ip_finish_output(struct sk_buff *skb)
 {
 #if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM)
 	/* Policy lookup after SNAT yielded a new policy */
-	if (skb->dst->xfrm != NULL)
-		return xfrm4_output_finish(skb);
+	if (skb->dst->xfrm != NULL) {
+		IPCB(skb)->flags |= IPSKB_REROUTED;
+		return dst_output(skb);
+	}
 #endif
 	if (skb->len > dst_mtu(skb->dst) &&
 	    !(skb_shinfo(skb)->ufo_size || skb_shinfo(skb)->tso_size))
@@ -271,8 +273,9 @@ int ip_mc_output(struct sk_buff *skb)
 				newskb->dev, ip_dev_loopback_xmit);
 	}
 
-	return NF_HOOK(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dev,
-		       ip_finish_output);
+	return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dev,
+			    ip_finish_output,
+			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
 
 int ip_output(struct sk_buff *skb)
@@ -284,8 +287,9 @@ int ip_output(struct sk_buff *skb)
 	skb->dev = dev;
 	skb->protocol = htons(ETH_P_IP);
 
-	return NF_HOOK(PF_INET, NF_IP_POST_ROUTING, skb, NULL, dev,
-		       ip_finish_output);
+	return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, dev,
+		            ip_finish_output,
+			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
 
 int ip_queue_xmit(struct sk_buff *skb, int ipfragok)
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index e5cbe72c6b80..03d13742a4b8 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -622,7 +622,8 @@ static int ipip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
 	skb->h.raw = skb->nh.raw;
 	skb->nh.raw = skb_push(skb, sizeof(struct iphdr));
 	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
-	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE|IPSKB_XFRM_TRANSFORMED);
+	IPCB(skb)->flags &= ~(IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED |
+			      IPSKB_REROUTED);
 	dst_release(skb->dst);
 	skb->dst = &rt->u.dst;
 
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index d4df0ddd424b..32ad229b4fed 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -152,10 +152,16 @@ error_nolock:
 	goto out_exit;
 }
 
-int xfrm4_output_finish(struct sk_buff *skb)
+static int xfrm4_output_finish(struct sk_buff *skb)
 {
 	int err;
 
+#ifdef CONFIG_NETFILTER
+	if (!skb->dst->xfrm) {
+		IPCB(skb)->flags |= IPSKB_REROUTED;
+		return dst_output(skb);
+	}
+#endif
 	while (likely((err = xfrm4_output_one(skb)) == 0)) {
 		nf_reset(skb);
 
@@ -178,6 +184,7 @@ int xfrm4_output_finish(struct sk_buff *skb)
 
 int xfrm4_output(struct sk_buff *skb)
 {
-	return NF_HOOK(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dst->dev,
-		       xfrm4_output_finish);
+	return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dst->dev,
+			    xfrm4_output_finish,
+			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }

From b05de01ae1c76b7d61da21bbcc26345bf7a9052f Mon Sep 17 00:00:00 2001
From: Horms <horms@verge.net.au>
Date: Wed, 15 Feb 2006 17:23:09 +0900
Subject: [PATCH 220/509] [IA64] support panic_on_oops sysctl

Trivial port of this feature from i386
As it stands, panic_on_oops but does nothing on ia64

Signed-Off-By: Horms <horms@verge.net.au>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/traps.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/arch/ia64/kernel/traps.c b/arch/ia64/kernel/traps.c
index 55391901b013..dabd6c32641e 100644
--- a/arch/ia64/kernel/traps.c
+++ b/arch/ia64/kernel/traps.c
@@ -16,6 +16,7 @@
 #include <linux/module.h>       /* for EXPORT_SYMBOL */
 #include <linux/hardirq.h>
 #include <linux/kprobes.h>
+#include <linux/delay.h>		/* for ssleep() */
 
 #include <asm/fpswa.h>
 #include <asm/ia32.h>
@@ -116,6 +117,13 @@ die (const char *str, struct pt_regs *regs, long err)
 	bust_spinlocks(0);
 	die.lock_owner = -1;
 	spin_unlock_irq(&die.lock);
+
+	if (panic_on_oops) {
+		printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n");
+		ssleep(5);
+		panic("Fatal exception");
+	}
+
   	do_exit(SIGSEGV);
 }
 

From 9c92d3486434e7310cb288587953e2dae4a79701 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 15 Feb 2006 15:18:19 -0800
Subject: [PATCH 221/509] [NETFILTER]: Don't invoke okfn in CONFIG_NETFILTER=n
 variant of nf_hook()

nf_hook() is supposed to call the netfilter hook and return control of the
packet back to the caller in case it may pass, the okfn is only used for
queueing.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/netfilter.h b/include/linux/netfilter.h
index 3ca3d9ee78a9..468896939843 100644
--- a/include/linux/netfilter.h
+++ b/include/linux/netfilter.h
@@ -318,7 +318,7 @@ static inline int nf_hook(int pf, unsigned int hook, struct sk_buff **pskb,
 			  struct net_device *indev, struct net_device *outdev,
 			  int (*okfn)(struct sk_buff *))
 {
-	return okfn(*pskb);
+	return 1;
 }
 static inline void nf_ct_attach(struct sk_buff *new, struct sk_buff *skb) {}
 struct flowi;

From deac0ccdb4da16b68539d75edecf26162de05150 Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Wed, 15 Feb 2006 15:21:31 -0800
Subject: [PATCH 222/509] [NETFILTER]: x_tables: fix dependencies of conntrack
 related modules

NF_CONNTRACK_MARK is bool and depends on NF_CONNTRACK which is
tristate.  If a variable depends on NF_CONNTRACK_MARK and doesn't take
care about NF_CONNTRACK, it can be y even if NF_CONNTRACK isn't y.
NF_CT_ACCT have same issue, too.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/Kconfig | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 0e550127fa7e..a8e5544da93e 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -126,7 +126,7 @@ config NETFILTER_XT_TARGET_CONNMARK
 	tristate  '"CONNMARK" target support'
 	depends on NETFILTER_XTABLES
 	depends on IP_NF_MANGLE || IP6_NF_MANGLE
-	depends on (IP_NF_CONNTRACK && IP_NF_CONNTRACK_MARK) || (NF_CONNTRACK_MARK && NF_CONNTRACK_IPV4)
+	depends on (IP_NF_CONNTRACK && IP_NF_CONNTRACK_MARK) || (NF_CONNTRACK_MARK && NF_CONNTRACK)
 	help
 	  This option adds a `CONNMARK' target, which allows one to manipulate
 	  the connection mark value.  Similar to the MARK target, but
@@ -187,7 +187,7 @@ config NETFILTER_XT_MATCH_COMMENT
 config NETFILTER_XT_MATCH_CONNBYTES
 	tristate  '"connbytes" per-connection counter match support'
 	depends on NETFILTER_XTABLES
-	depends on (IP_NF_CONNTRACK && IP_NF_CT_ACCT) || NF_CT_ACCT
+	depends on (IP_NF_CONNTRACK && IP_NF_CT_ACCT) || (NF_CT_ACCT && NF_CONNTRACK)
 	help
 	  This option adds a `connbytes' match, which allows you to match the
 	  number of bytes and/or packets for each direction within a connection.
@@ -198,7 +198,7 @@ config NETFILTER_XT_MATCH_CONNBYTES
 config NETFILTER_XT_MATCH_CONNMARK
 	tristate  '"connmark" connection mark match support'
 	depends on NETFILTER_XTABLES
-	depends on (IP_NF_CONNTRACK && IP_NF_CONNTRACK_MARK) || NF_CONNTRACK_MARK
+	depends on (IP_NF_CONNTRACK && IP_NF_CONNTRACK_MARK) || (NF_CONNTRACK_MARK && NF_CONNTRACK)
 	help
 	  This option adds a `connmark' match, which allows you to match the
 	  connection mark value previously set for the session by `CONNMARK'. 

From 7d3cdc6b554137a7a0534ce38b155a63a3117f27 Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Wed, 15 Feb 2006 15:22:21 -0800
Subject: [PATCH 223/509] [NETFILTER]: nf_conntrack: move registration of
 __nf_ct_attach

Move registration of __nf_ct_attach to nf_conntrack_core to make it usable
for IPv6 connection tracking as well.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c | 5 -----
 net/netfilter/nf_conntrack_core.c              | 5 +++++
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
index 167619f638c6..6c8624a54933 100644
--- a/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
+++ b/net/ipv4/netfilter/nf_conntrack_l3proto_ipv4.c
@@ -529,15 +529,10 @@ static int init_or_cleanup(int init)
 		goto cleanup_localinops;
 	}
 #endif
-
-	/* For use by REJECT target */
-	ip_ct_attach = __nf_conntrack_attach;
-
 	return ret;
 
  cleanup:
 	synchronize_net();
-	ip_ct_attach = NULL;
 #ifdef CONFIG_SYSCTL
  	unregister_sysctl_table(nf_ct_ipv4_sysctl_header);
  cleanup_localinops:
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 0ce337a1d974..d622ddf08bb0 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1556,6 +1556,8 @@ void nf_conntrack_cleanup(void)
 {
 	int i;
 
+	ip_ct_attach = NULL;
+
 	/* This makes sure all current packets have passed through
 	   netfilter framework.  Roll on, two-stage module
 	   delete... */
@@ -1715,6 +1717,9 @@ int __init nf_conntrack_init(void)
 		nf_ct_l3protos[i] = &nf_conntrack_generic_l3proto;
         write_unlock_bh(&nf_conntrack_lock);
 
+	/* For use by REJECT target */
+	ip_ct_attach = __nf_conntrack_attach;
+
 	/* Set up fake conntrack:
 	    - to never be deleted, not in any hashes */
 	atomic_set(&nf_conntrack_untracked.ct_general.use, 1);

From 08857fa745ab6ce46601960d2774490e1cef2cff Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Wed, 15 Feb 2006 15:23:28 -0800
Subject: [PATCH 224/509] [NETFILTER]: nf_conntrack: attach conntrack to TCP
 RST generated by ip6t_REJECT

TCP RSTs generated by the REJECT target should be associated with the
conntrack of the original TCP packet. Since the conntrack entry is
usually not is the hash tables, it must be manually attached.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/netfilter/ip6t_REJECT.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c
index c745717b4ce2..0e6d1d4bbd5c 100644
--- a/net/ipv6/netfilter/ip6t_REJECT.c
+++ b/net/ipv6/netfilter/ip6t_REJECT.c
@@ -160,6 +160,8 @@ static void send_reset(struct sk_buff *oldskb)
 				      csum_partial((char *)tcph,
 						   sizeof(struct tcphdr), 0));
 
+	nf_ct_attach(nskb, oldskb);
+
 	NF_HOOK(PF_INET6, NF_IP6_LOCAL_OUT, nskb, NULL, nskb->dst->dev,
 		dst_output);
 }

From 763ecff1879b3877f57f20fc9e79599aef59359f Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Wed, 15 Feb 2006 15:24:15 -0800
Subject: [PATCH 225/509] [NETFILTER]: nf_conntrack: attach conntrack to
 locally generated ICMPv6 error

Locally generated ICMPv6 errors should be associated with the conntrack
of the original packet. Since the conntrack entry may not be in the hash
tables (for the first packet), it must be manually attached.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/icmp.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index fcf883183cef..21eb725e885f 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -42,6 +42,7 @@
 #include <linux/net.h>
 #include <linux/skbuff.h>
 #include <linux/init.h>
+#include <linux/netfilter.h>
 
 #ifdef CONFIG_SYSCTL
 #include <linux/sysctl.h>
@@ -255,6 +256,7 @@ out:
 struct icmpv6_msg {
 	struct sk_buff	*skb;
 	int		offset;
+	uint8_t		type;
 };
 
 static int icmpv6_getfrag(void *from, char *to, int offset, int len, int odd, struct sk_buff *skb)
@@ -266,6 +268,8 @@ static int icmpv6_getfrag(void *from, char *to, int offset, int len, int odd, st
 	csum = skb_copy_and_csum_bits(org_skb, msg->offset + offset,
 				      to, len, csum);
 	skb->csum = csum_block_add(skb->csum, csum, odd);
+	if (!(msg->type & ICMPV6_INFOMSG_MASK))
+		nf_ct_attach(skb, org_skb);
 	return 0;
 }
 
@@ -403,6 +407,7 @@ void icmpv6_send(struct sk_buff *skb, int type, int code, __u32 info,
 
 	msg.skb = skb;
 	msg.offset = skb->nh.raw - skb->data;
+	msg.type = type;
 
 	len = skb->len - msg.offset;
 	len = min_t(unsigned int, len, IPV6_MIN_MTU - sizeof(struct ipv6hdr) -sizeof(struct icmp6hdr));
@@ -500,6 +505,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb)
 
 	msg.skb = skb;
 	msg.offset = 0;
+	msg.type = ICMPV6_ECHO_REPLY;
 
 	err = ip6_append_data(sk, icmpv6_getfrag, &msg, skb->len + sizeof(struct icmp6hdr),
 				sizeof(struct icmp6hdr), hlimit, tclass, NULL, &fl,

From 7c6de05884b9fcc7ef621e2ab198ba93d85f46aa Mon Sep 17 00:00:00 2001
From: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Date: Wed, 15 Feb 2006 15:25:18 -0800
Subject: [PATCH 226/509] [NETFILTER]: nf_conntrack: Fix TCP/UDP HW checksum
 handling for IPv6 packet

If skb->ip_summed is CHECKSUM_HW here, skb->csum includes checksum
of actual IPv6 header and extension headers. Then such excess
checksum must be subtruct when nf_conntrack calculates TCP/UDP checksum
with pseudo IPv6 header. Spotted by Ben Skeggs.

Signed-off-by: Yasuyuki Kozakai <yasuyuki.kozakai@toshiba.co.jp>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/nf_conntrack_proto_tcp.c | 4 +++-
 net/netfilter/nf_conntrack_proto_udp.c | 4 +++-
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index df99138c3b3b..6492ed66fb3c 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -864,7 +864,9 @@ static int csum6(const struct sk_buff *skb, unsigned int dataoff)
 {
 	return csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr,
 			       skb->len - dataoff, IPPROTO_TCP,
-			       skb->ip_summed == CHECKSUM_HW ? skb->csum
+			       skb->ip_summed == CHECKSUM_HW
+			       ? csum_sub(skb->csum,
+					  skb_checksum(skb, 0, dataoff, 0))
 			       : skb_checksum(skb, dataoff, skb->len - dataoff,
 					      0));
 }
diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c
index 4264dd079a16..831d206344e0 100644
--- a/net/netfilter/nf_conntrack_proto_udp.c
+++ b/net/netfilter/nf_conntrack_proto_udp.c
@@ -161,7 +161,9 @@ static int csum6(const struct sk_buff *skb, unsigned int dataoff)
 {
 	return csum_ipv6_magic(&skb->nh.ipv6h->saddr, &skb->nh.ipv6h->daddr,
 			       skb->len - dataoff, IPPROTO_UDP,
-			       skb->ip_summed == CHECKSUM_HW ? skb->csum
+			       skb->ip_summed == CHECKSUM_HW
+			       ? csum_sub(skb->csum,
+					  skb_checksum(skb, 0, dataoff, 0))
 			       : skb_checksum(skb, dataoff, skb->len - dataoff,
 					      0));
 }

From 9f672004ab1a8094bec1785b39ac683ab9eebebc Mon Sep 17 00:00:00 2001
From: Christian Trefzer <ctrefzer@gmx.de>
Date: Wed, 15 Feb 2006 15:17:34 -0800
Subject: [PATCH 227/509] [PATCH] neofb: avoid resetting display config on
 unblank (v2)

There were two mistakes in the register-read-on-(un)blank approach.

- First, without proper register (un)locking the value read back will always
  be zero, and this is what I missed entirely until just now.  Due to this,
  the logic could not be verified at all and I tried some bogus checks which
  are completely stupid.

- Second, the LCD status bit will always be set to zero when the backlight
  has been turned off.  Reading the value back during unblank will disable the
  LCD unconditionally, regardless of the state it is supposed to be in, since
  we set it to zero beforehand.

So this is what we do now:

- create a new variable in struct neofb_par, and use that to determine
  whether to read back registers (initialized to true)

- before actually blanking the screen, read back the register to sense any
  possible change made through Fn key combo

- use proper neoUnlock() / neoLock() to actually read something

- every call to neofb_blank() determines if we read back next time: blanking
  disables readback, unblanking (FB_BLANK_UNBLANK) enables it

This should give us a nice and clean state machine.  Has been thoroughly
tested on a Dell Latitude CPiA / NM220 Chip docked to a C/Dock2 with attached
CRT in all possible combinations of LCD/CRT on/off.  I changed the config via
Fn key, let the console blank, unblanked by keypress - works flawlessly.

Signed-off-by: Christian Trefzer <ctrefzer@gmx.de>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/neofb.c    | 15 ++++++++++++---
 include/video/neomagic.h |  1 +
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/video/neofb.c b/drivers/video/neofb.c
index b85e2b180a44..a2e201dc40f7 100644
--- a/drivers/video/neofb.c
+++ b/drivers/video/neofb.c
@@ -843,6 +843,9 @@ static int neofb_set_par(struct fb_info *info)
 
 	par->SysIfaceCntl2 = 0xc0;	/* VESA Bios sets this to 0x80! */
 
+	/* Initialize: by default, we want display config register to be read */
+	par->PanelDispCntlRegRead = 1;
+
 	/* Enable any user specified display devices. */
 	par->PanelDispCntlReg1 = 0x00;
 	if (par->internal_display)
@@ -1334,11 +1337,17 @@ static int neofb_blank(int blank_mode, struct fb_info *info)
 	struct neofb_par *par = info->par;
 	int seqflags, lcdflags, dpmsflags, reg;
 
+
 	/*
-	 * Reload the value stored in the register, might have been changed via
-	 * FN keystroke
+	 * Reload the value stored in the register, if sensible. It might have
+	 * been changed via FN keystroke.
 	 */
-	par->PanelDispCntlReg1 = vga_rgfx(NULL, 0x20) & 0x03;
+	if (par->PanelDispCntlRegRead) {
+		neoUnlock();
+		par->PanelDispCntlReg1 = vga_rgfx(NULL, 0x20) & 0x03;
+		neoLock(&par->state);
+	}
+	par->PanelDispCntlRegRead = !blank_mode;
 
 	switch (blank_mode) {
 	case FB_BLANK_POWERDOWN:	/* powerdown - both sync lines down */
diff --git a/include/video/neomagic.h b/include/video/neomagic.h
index 1d69049bd4c1..78b1f15a538f 100644
--- a/include/video/neomagic.h
+++ b/include/video/neomagic.h
@@ -159,6 +159,7 @@ struct neofb_par {
 	unsigned char PanelDispCntlReg1;
 	unsigned char PanelDispCntlReg2;
 	unsigned char PanelDispCntlReg3;
+	unsigned char PanelDispCntlRegRead;
 	unsigned char PanelVertCenterReg1;
 	unsigned char PanelVertCenterReg2;
 	unsigned char PanelVertCenterReg3;

From 36cbbe5eb9857730768aa5f54ad94d69e0b2133d Mon Sep 17 00:00:00 2001
From: Benjamin LaHaise <bcrl@kvack.org>
Date: Wed, 15 Feb 2006 15:17:35 -0800
Subject: [PATCH 228/509] [PATCH] kbuild: revert "fix make -jN with multiple
 targets with O=..."

Commit 296e0855b0f9a4ec9be17106ac541745a55b2ce1:

    "kbuild: fix make -jN with multiple targets with O=..."

causes a ~95% increase in build time for the kernel.  Before: 4m21s
after: 8m1.403s.  Can we revert this until another approach is found?

Cc: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Makefile | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/Makefile b/Makefile
index 74d67b2c35d9..48d569d46ae1 100644
--- a/Makefile
+++ b/Makefile
@@ -106,13 +106,12 @@ KBUILD_OUTPUT := $(shell cd $(KBUILD_OUTPUT) && /bin/pwd)
 $(if $(KBUILD_OUTPUT),, \
      $(error output directory "$(saved-output)" does not exist))
 
-.PHONY: $(MAKECMDGOALS) cdbuilddir
-$(MAKECMDGOALS) _all: cdbuilddir
+.PHONY: $(MAKECMDGOALS)
 
-cdbuilddir:
+$(filter-out _all,$(MAKECMDGOALS)) _all:
 	$(if $(KBUILD_VERBOSE:1=),@)$(MAKE) -C $(KBUILD_OUTPUT) \
 	KBUILD_SRC=$(CURDIR) \
-	KBUILD_EXTMOD="$(KBUILD_EXTMOD)" -f $(CURDIR)/Makefile $(MAKECMDGOALS)
+	KBUILD_EXTMOD="$(KBUILD_EXTMOD)" -f $(CURDIR)/Makefile $@
 
 # Leave processing to above invocation of make
 skip-makefile := 1

From 651c29a17f7ea0204dacbc2a5042d57b1c9e2e37 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 15 Feb 2006 15:17:37 -0800
Subject: [PATCH 229/509] [PATCH] ide: touch softlockup detector during pio

We're getting some softlockup false positives during heavy PIO operations.  So
poke the lockup detector.

Cc: Bartlomiej Zolnierkiewicz <B.Zolnierkiewicz@elka.pw.edu.pl>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/ide/ide-taskfile.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index 9834dce4e20f..0606bd2f6020 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -34,6 +34,7 @@
 #include <linux/kernel.h>
 #include <linux/timer.h>
 #include <linux/mm.h>
+#include <linux/sched.h>
 #include <linux/interrupt.h>
 #include <linux/major.h>
 #include <linux/errno.h>
@@ -314,6 +315,8 @@ static void ide_pio_datablock(ide_drive_t *drive, struct request *rq,
 	if (rq->bio)	/* fs request */
 		rq->errors = 0;
 
+	touch_softlockup_watchdog();
+
 	switch (drive->hwif->data_phase) {
 	case TASKFILE_MULTI_IN:
 	case TASKFILE_MULTI_OUT:

From 06fed33849c13af637c4d09e9ba27828fac9edd5 Mon Sep 17 00:00:00 2001
From: Paul Jackson <pj@sgi.com>
Date: Wed, 15 Feb 2006 15:17:38 -0800
Subject: [PATCH 230/509] [PATCH] cpuset: oops in exit on null cpuset fix

Fix a latent bug in cpuset_exit() handling.  If a task tried to allocate
memory after calling cpuset_exit(), it oops'd in
cpuset_update_task_memory_state() on a NULL cpuset pointer.

So set the exiting tasks cpuset to the root cpuset instead of to NULL.

A distro kernel hit this with an added kernel package that had just such a
hook (allocating memory) in the exit code path.

Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/cpuset.c | 35 ++++++++++++++++++++++++++++++++++-
 1 file changed, 34 insertions(+), 1 deletion(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index ba42b0a76961..12815d3f1a05 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -1977,6 +1977,39 @@ void cpuset_fork(struct task_struct *child)
  * We don't need to task_lock() this reference to tsk->cpuset,
  * because tsk is already marked PF_EXITING, so attach_task() won't
  * mess with it, or task is a failed fork, never visible to attach_task.
+ *
+ * Hack:
+ *
+ *    Set the exiting tasks cpuset to the root cpuset (top_cpuset).
+ *
+ *    Don't leave a task unable to allocate memory, as that is an
+ *    accident waiting to happen should someone add a callout in
+ *    do_exit() after the cpuset_exit() call that might allocate.
+ *    If a task tries to allocate memory with an invalid cpuset,
+ *    it will oops in cpuset_update_task_memory_state().
+ *
+ *    We call cpuset_exit() while the task is still competent to
+ *    handle notify_on_release(), then leave the task attached to
+ *    the root cpuset (top_cpuset) for the remainder of its exit.
+ *
+ *    To do this properly, we would increment the reference count on
+ *    top_cpuset, and near the very end of the kernel/exit.c do_exit()
+ *    code we would add a second cpuset function call, to drop that
+ *    reference.  This would just create an unnecessary hot spot on
+ *    the top_cpuset reference count, to no avail.
+ *
+ *    Normally, holding a reference to a cpuset without bumping its
+ *    count is unsafe.   The cpuset could go away, or someone could
+ *    attach us to a different cpuset, decrementing the count on
+ *    the first cpuset that we never incremented.  But in this case,
+ *    top_cpuset isn't going away, and either task has PF_EXITING set,
+ *    which wards off any attach_task() attempts, or task is a failed
+ *    fork, never visible to attach_task.
+ *
+ *    Another way to do this would be to set the cpuset pointer
+ *    to NULL here, and check in cpuset_update_task_memory_state()
+ *    for a NULL pointer.  This hack avoids that NULL check, for no
+ *    cost (other than this way too long comment ;).
  **/
 
 void cpuset_exit(struct task_struct *tsk)
@@ -1984,7 +2017,7 @@ void cpuset_exit(struct task_struct *tsk)
 	struct cpuset *cs;
 
 	cs = tsk->cpuset;
-	tsk->cpuset = NULL;
+	tsk->cpuset = &top_cpuset;	/* Hack - see comment above */
 
 	if (notify_on_release(cs)) {
 		char *pathbuf = NULL;

From 5f6164f3092832e0d9b12eed52e09a76bf39c64a Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <mst@mellanox.co.il>
Date: Wed, 15 Feb 2006 15:17:39 -0800
Subject: [PATCH 231/509] [PATCH] add asm-generic/mman.h

Make new MADV_REMOVE, MADV_DONTFORK, MADV_DOFORK consistent across all
arches.  The idea is to make it possible to use them portably even before
distros include them in libc headers.

Move common flags to asm-generic/mman.h

Signed-off-by: Michael S. Tsirkin <mst@mellanox.co.il>
Cc: Roland Dreier <rolandd@cisco.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-alpha/mman.h   |  8 +++++---
 include/asm-arm/mman.h     | 31 +---------------------------
 include/asm-arm26/mman.h   | 31 +---------------------------
 include/asm-cris/mman.h    | 31 +---------------------------
 include/asm-frv/mman.h     | 31 +---------------------------
 include/asm-generic/mman.h | 42 ++++++++++++++++++++++++++++++++++++++
 include/asm-h8300/mman.h   | 31 +---------------------------
 include/asm-i386/mman.h    | 31 +---------------------------
 include/asm-ia64/mman.h    | 31 +---------------------------
 include/asm-m32r/mman.h    | 33 ++----------------------------
 include/asm-m68k/mman.h    | 31 +---------------------------
 include/asm-mips/mman.h    | 22 +++++++++++---------
 include/asm-parisc/mman.h  |  8 +++++---
 include/asm-powerpc/mman.h | 32 ++---------------------------
 include/asm-s390/mman.h    | 31 +---------------------------
 include/asm-sh/mman.h      | 31 +---------------------------
 include/asm-sparc/mman.h   | 31 ++--------------------------
 include/asm-sparc64/mman.h | 31 ++--------------------------
 include/asm-v850/mman.h    | 30 +--------------------------
 include/asm-x86_64/mman.h  | 30 +--------------------------
 include/asm-xtensa/mman.h  | 22 +++++++++++---------
 21 files changed, 96 insertions(+), 503 deletions(-)
 create mode 100644 include/asm-generic/mman.h

diff --git a/include/asm-alpha/mman.h b/include/asm-alpha/mman.h
index a21515c16a43..5f24c755f577 100644
--- a/include/asm-alpha/mman.h
+++ b/include/asm-alpha/mman.h
@@ -42,9 +42,11 @@
 #define MADV_WILLNEED	3		/* will need these pages */
 #define	MADV_SPACEAVAIL	5		/* ensure resources are available */
 #define MADV_DONTNEED	6		/* don't need these pages */
-#define MADV_REMOVE	7		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
+
+/* common/generic parameters */
+#define MADV_REMOVE	9		/* remove these pages & resources */
+#define MADV_DONTFORK	10		/* don't inherit across fork */
+#define MADV_DOFORK	11		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-arm/mman.h b/include/asm-arm/mman.h
index 693ed859e632..54570d2e95b7 100644
--- a/include/asm-arm/mman.h
+++ b/include/asm-arm/mman.h
@@ -1,19 +1,7 @@
 #ifndef __ARM_MMAN_H__
 #define __ARM_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -23,24 +11,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) page tables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __ARM_MMAN_H__ */
diff --git a/include/asm-arm26/mman.h b/include/asm-arm26/mman.h
index 2096c50df888..4000a6c1b76b 100644
--- a/include/asm-arm26/mman.h
+++ b/include/asm-arm26/mman.h
@@ -1,19 +1,7 @@
 #ifndef __ARM_MMAN_H__
 #define __ARM_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -23,24 +11,7 @@
 #define MAP_POPULATE    0x8000          /* populate (prefault) page tables */
 #define MAP_NONBLOCK    0x10000         /* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __ARM_MMAN_H__ */
diff --git a/include/asm-cris/mman.h b/include/asm-cris/mman.h
index deddfb239ff5..1c35e1b66b46 100644
--- a/include/asm-cris/mman.h
+++ b/include/asm-cris/mman.h
@@ -3,19 +3,7 @@
 
 /* verbatim copy of asm-i386/ version */
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -25,24 +13,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __CRIS_MMAN_H__ */
diff --git a/include/asm-frv/mman.h b/include/asm-frv/mman.h
index d3bca306da82..b4371e928683 100644
--- a/include/asm-frv/mman.h
+++ b/include/asm-frv/mman.h
@@ -1,19 +1,7 @@
 #ifndef __ASM_MMAN_H__
 #define __ASM_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -23,25 +11,8 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __ASM_MMAN_H__ */
 
diff --git a/include/asm-generic/mman.h b/include/asm-generic/mman.h
new file mode 100644
index 000000000000..3b41d2bb70da
--- /dev/null
+++ b/include/asm-generic/mman.h
@@ -0,0 +1,42 @@
+#ifndef _ASM_GENERIC_MMAN_H
+#define _ASM_GENERIC_MMAN_H
+
+/*
+ Author: Michael S. Tsirkin <mst@mellanox.co.il>, Mellanox Technologies Ltd.
+ Based on: asm-xxx/mman.h
+*/
+
+#define PROT_READ	0x1		/* page can be read */
+#define PROT_WRITE	0x2		/* page can be written */
+#define PROT_EXEC	0x4		/* page can be executed */
+#define PROT_SEM	0x8		/* page may be used for atomic ops */
+#define PROT_NONE	0x0		/* page can not be accessed */
+#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
+#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
+
+#define MAP_SHARED	0x01		/* Share changes */
+#define MAP_PRIVATE	0x02		/* Changes are private */
+#define MAP_TYPE	0x0f		/* Mask for type of mapping */
+#define MAP_FIXED	0x10		/* Interpret addr exactly */
+#define MAP_ANONYMOUS	0x20		/* don't use a file */
+
+#define MS_ASYNC	1		/* sync memory asynchronously */
+#define MS_INVALIDATE	2		/* invalidate the caches */
+#define MS_SYNC		4		/* synchronous memory sync */
+
+#define MADV_NORMAL	0		/* no further special treatment */
+#define MADV_RANDOM	1		/* expect random page references */
+#define MADV_SEQUENTIAL	2		/* expect sequential page references */
+#define MADV_WILLNEED	3		/* will need these pages */
+#define MADV_DONTNEED	4		/* don't need these pages */
+
+/* common parameters: try to keep these consistent across architectures */
+#define MADV_REMOVE	9		/* remove these pages & resources */
+#define MADV_DONTFORK	10		/* don't inherit across fork */
+#define MADV_DOFORK	11		/* do inherit across fork */
+
+/* compatibility flags */
+#define MAP_ANON	MAP_ANONYMOUS
+#define MAP_FILE	0
+
+#endif
diff --git a/include/asm-h8300/mman.h b/include/asm-h8300/mman.h
index ac0346f7d11d..b9f104f22a36 100644
--- a/include/asm-h8300/mman.h
+++ b/include/asm-h8300/mman.h
@@ -1,19 +1,7 @@
 #ifndef __H8300_MMAN_H__
 #define __H8300_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -23,24 +11,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __H8300_MMAN_H__ */
diff --git a/include/asm-i386/mman.h b/include/asm-i386/mman.h
index ab2339a1d807..8fd9d7ab7faf 100644
--- a/include/asm-i386/mman.h
+++ b/include/asm-i386/mman.h
@@ -1,19 +1,7 @@
 #ifndef __I386_MMAN_H__
 #define __I386_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -23,24 +11,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __I386_MMAN_H__ */
diff --git a/include/asm-ia64/mman.h b/include/asm-ia64/mman.h
index 357ebb780cc0..6ba179f12718 100644
--- a/include/asm-ia64/mman.h
+++ b/include/asm-ia64/mman.h
@@ -8,19 +8,7 @@
  *	David Mosberger-Tang <davidm@hpl.hp.com>, Hewlett-Packard Co
  */
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x00100		/* stack-like segment */
 #define MAP_GROWSUP	0x00200		/* register stack-like segment */
@@ -31,24 +19,7 @@
 #define MAP_POPULATE	0x08000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* _ASM_IA64_MMAN_H */
diff --git a/include/asm-m32r/mman.h b/include/asm-m32r/mman.h
index 6b02fe3fcff2..695a860c024f 100644
--- a/include/asm-m32r/mman.h
+++ b/include/asm-m32r/mman.h
@@ -1,22 +1,10 @@
 #ifndef __M32R_MMAN_H__
 #define __M32R_MMAN_H__
 
+#include <asm-generic/mman.h>
+
 /* orig : i386 2.6.0-test6 */
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
-
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
 #define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
@@ -25,24 +13,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __M32R_MMAN_H__ */
diff --git a/include/asm-m68k/mman.h b/include/asm-m68k/mman.h
index efd12bc4ccb7..1626d37f4898 100644
--- a/include/asm-m68k/mman.h
+++ b/include/asm-m68k/mman.h
@@ -1,19 +1,7 @@
 #ifndef __M68K_MMAN_H__
 #define __M68K_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -23,24 +11,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __M68K_MMAN_H__ */
diff --git a/include/asm-mips/mman.h b/include/asm-mips/mman.h
index 6d01e26830fa..046cf686bee7 100644
--- a/include/asm-mips/mman.h
+++ b/include/asm-mips/mman.h
@@ -60,17 +60,19 @@
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
+#define MADV_NORMAL	0		/* no further special treatment */
+#define MADV_RANDOM	1		/* expect random page references */
+#define MADV_SEQUENTIAL	2		/* expect sequential page references */
+#define MADV_WILLNEED	3		/* will need these pages */
+#define MADV_DONTNEED	4		/* don't need these pages */
+
+/* common parameters: try to keep these consistent across architectures */
+#define MADV_REMOVE	9		/* remove these pages & resources */
+#define MADV_DONTFORK	10		/* don't inherit across fork */
+#define MADV_DOFORK	11		/* do inherit across fork */
 
 /* compatibility flags */
-#define MAP_ANON       MAP_ANONYMOUS
-#define MAP_FILE       0
+#define MAP_ANON	MAP_ANONYMOUS
+#define MAP_FILE	0
 
 #endif /* _ASM_MMAN_H */
diff --git a/include/asm-parisc/mman.h b/include/asm-parisc/mman.h
index a381cf5c8f55..0ef15ee0f17e 100644
--- a/include/asm-parisc/mman.h
+++ b/include/asm-parisc/mman.h
@@ -38,7 +38,11 @@
 #define MADV_SPACEAVAIL 5               /* insure that resources are reserved */
 #define MADV_VPS_PURGE  6               /* Purge pages from VM page cache */
 #define MADV_VPS_INHERIT 7              /* Inherit parents page size */
-#define MADV_REMOVE     8		/* remove these pages & resources */
+
+/* common/generic parameters */
+#define MADV_REMOVE	9		/* remove these pages & resources */
+#define MADV_DONTFORK	10		/* don't inherit across fork */
+#define MADV_DOFORK	11		/* do inherit across fork */
 
 /* The range 12-64 is reserved for page size specification. */
 #define MADV_4K_PAGES   12              /* Use 4K pages  */
@@ -49,8 +53,6 @@
 #define MADV_4M_PAGES   22              /* Use 4 Megabyte pages */
 #define MADV_16M_PAGES  24              /* Use 16 Megabyte pages */
 #define MADV_64M_PAGES  26              /* Use 64 Megabyte pages */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
 
 /* compatibility flags */
 #define MAP_ANON	MAP_ANONYMOUS
diff --git a/include/asm-powerpc/mman.h b/include/asm-powerpc/mman.h
index fcff25d13f13..24cf664a8295 100644
--- a/include/asm-powerpc/mman.h
+++ b/include/asm-powerpc/mman.h
@@ -1,6 +1,8 @@
 #ifndef _ASM_POWERPC_MMAN_H
 #define _ASM_POWERPC_MMAN_H
 
+#include <asm-generic/mman.h>
+
 /*
  * This program is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License
@@ -8,19 +10,6 @@
  * 2 of the License, or (at your option) any later version.
  */
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
 #define MAP_RENAME      MAP_ANONYMOUS   /* In SunOS terminology */
 #define MAP_NORESERVE   0x40            /* don't reserve swap pages */
 #define MAP_LOCKED	0x80
@@ -29,27 +18,10 @@
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
 #define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT     0x2000          /* lock all currently mapped pages */
 #define MCL_FUTURE      0x4000          /* lock all additions to address space */
 
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif	/* _ASM_POWERPC_MMAN_H */
diff --git a/include/asm-s390/mman.h b/include/asm-s390/mman.h
index d41ca1477010..7839767d837e 100644
--- a/include/asm-s390/mman.h
+++ b/include/asm-s390/mman.h
@@ -9,19 +9,7 @@
 #ifndef __S390_MMAN_H__
 #define __S390_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -31,24 +19,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL    0x0              /* default page-in behavior */
-#define MADV_RANDOM    0x1              /* page-in minimum required */
-#define MADV_SEQUENTIAL        0x2             /* read-ahead aggressively */
-#define MADV_WILLNEED  0x3              /* pre-fault pages */
-#define MADV_DONTNEED  0x4              /* discard these pages */
-#define MADV_REMOVE    0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __S390_MMAN_H__ */
diff --git a/include/asm-sh/mman.h b/include/asm-sh/mman.h
index 0e08d0573abc..156eb0225cf6 100644
--- a/include/asm-sh/mman.h
+++ b/include/asm-sh/mman.h
@@ -1,19 +1,7 @@
 #ifndef __ASM_SH_MMAN_H
 #define __ASM_SH_MMAN_H
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -23,24 +11,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) page tables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __ASM_SH_MMAN_H */
diff --git a/include/asm-sparc/mman.h b/include/asm-sparc/mman.h
index 4a298b2be859..88d1886abf3b 100644
--- a/include/asm-sparc/mman.h
+++ b/include/asm-sparc/mman.h
@@ -2,21 +2,10 @@
 #ifndef __SPARC_MMAN_H__
 #define __SPARC_MMAN_H__
 
+#include <asm-generic/mman.h>
+
 /* SunOS'ified... */
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
 #define MAP_RENAME      MAP_ANONYMOUS   /* In SunOS terminology */
 #define MAP_NORESERVE   0x40            /* don't reserve swap pages */
 #define MAP_INHERIT     0x80            /* SunOS doesn't do this, but... */
@@ -27,10 +16,6 @@
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
 #define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT     0x2000          /* lock all currently mapped pages */
 #define MCL_FUTURE      0x4000          /* lock all additions to address space */
 
@@ -48,18 +33,6 @@
 #define MC_LOCKAS       5  /* Lock an entire address space of the calling process */
 #define MC_UNLOCKAS     6  /* Unlock entire address space of calling process */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_FREE	0x5		/* (Solaris) contents can be freed */
-#define MADV_REMOVE	0x6		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
 
 #endif /* __SPARC_MMAN_H__ */
diff --git a/include/asm-sparc64/mman.h b/include/asm-sparc64/mman.h
index d705ec92da8b..6fd878e61435 100644
--- a/include/asm-sparc64/mman.h
+++ b/include/asm-sparc64/mman.h
@@ -2,21 +2,10 @@
 #ifndef __SPARC64_MMAN_H__
 #define __SPARC64_MMAN_H__
 
+#include <asm-generic/mman.h>
+
 /* SunOS'ified... */
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_SEM	0x8		/* page may be used for atomic ops */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
 #define MAP_RENAME      MAP_ANONYMOUS   /* In SunOS terminology */
 #define MAP_NORESERVE   0x40            /* don't reserve swap pages */
 #define MAP_INHERIT     0x80            /* SunOS doesn't do this, but... */
@@ -27,10 +16,6 @@
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
 #define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT     0x2000          /* lock all currently mapped pages */
 #define MCL_FUTURE      0x4000          /* lock all additions to address space */
 
@@ -48,18 +33,6 @@
 #define MC_LOCKAS       5  /* Lock an entire address space of the calling process */
 #define MC_UNLOCKAS     6  /* Unlock entire address space of calling process */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
 #define MADV_FREE	0x5		/* (Solaris) contents can be freed */
-#define MADV_REMOVE	0x6		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
 
 #endif /* __SPARC64_MMAN_H__ */
diff --git a/include/asm-v850/mman.h b/include/asm-v850/mman.h
index 7b851c310e41..edbf6edbfb37 100644
--- a/include/asm-v850/mman.h
+++ b/include/asm-v850/mman.h
@@ -1,18 +1,7 @@
 #ifndef __V850_MMAN_H__
 #define __V850_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
-
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
+#include <asm-generic/mman.h>
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
 #define MAP_DENYWRITE	0x0800		/* ETXTBSY */
@@ -20,24 +9,7 @@
 #define MAP_LOCKED	0x2000		/* pages are locked */
 #define MAP_NORESERVE	0x4000		/* don't check for reservations */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif /* __V850_MMAN_H__ */
diff --git a/include/asm-x86_64/mman.h b/include/asm-x86_64/mman.h
index b699a38c1c3c..dd5cb0534d37 100644
--- a/include/asm-x86_64/mman.h
+++ b/include/asm-x86_64/mman.h
@@ -1,19 +1,8 @@
 #ifndef __X8664_MMAN_H__
 #define __X8664_MMAN_H__
 
-#define PROT_READ	0x1		/* page can be read */
-#define PROT_WRITE	0x2		/* page can be written */
-#define PROT_EXEC	0x4		/* page can be executed */
-#define PROT_NONE	0x0		/* page can not be accessed */
-#define PROT_SEM	0x8
-#define PROT_GROWSDOWN	0x01000000	/* mprotect flag: extend change to start of growsdown vma */
-#define PROT_GROWSUP	0x02000000	/* mprotect flag: extend change to end of growsup vma */
+#include <asm-generic/mman.h>
 
-#define MAP_SHARED	0x01		/* Share changes */
-#define MAP_PRIVATE	0x02		/* Changes are private */
-#define MAP_TYPE	0x0f		/* Mask for type of mapping */
-#define MAP_FIXED	0x10		/* Interpret addr exactly */
-#define MAP_ANONYMOUS	0x20		/* don't use a file */
 #define MAP_32BIT	0x40		/* only give out 32bit addresses */
 
 #define MAP_GROWSDOWN	0x0100		/* stack-like segment */
@@ -24,24 +13,7 @@
 #define MAP_POPULATE	0x8000		/* populate (prefault) pagetables */
 #define MAP_NONBLOCK	0x10000		/* do not block on IO */
 
-#define MS_ASYNC	1		/* sync memory asynchronously */
-#define MS_INVALIDATE	2		/* invalidate the caches */
-#define MS_SYNC		4		/* synchronous memory sync */
-
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
-
-/* compatibility flags */
-#define MAP_ANON	MAP_ANONYMOUS
-#define MAP_FILE	0
-
 #endif
diff --git a/include/asm-xtensa/mman.h b/include/asm-xtensa/mman.h
index e2d7afb679c8..ba394cbb4807 100644
--- a/include/asm-xtensa/mman.h
+++ b/include/asm-xtensa/mman.h
@@ -67,17 +67,19 @@
 #define MCL_CURRENT	1		/* lock all current mappings */
 #define MCL_FUTURE	2		/* lock all future mappings */
 
-#define MADV_NORMAL	0x0		/* default page-in behavior */
-#define MADV_RANDOM	0x1		/* page-in minimum required */
-#define MADV_SEQUENTIAL	0x2		/* read-ahead aggressively */
-#define MADV_WILLNEED	0x3		/* pre-fault pages */
-#define MADV_DONTNEED	0x4		/* discard these pages */
-#define MADV_REMOVE	0x5		/* remove these pages & resources */
-#define MADV_DONTFORK	0x30		/* dont inherit across fork */
-#define MADV_DOFORK	0x31		/* do inherit across fork */
+#define MADV_NORMAL	0		/* no further special treatment */
+#define MADV_RANDOM	1		/* expect random page references */
+#define MADV_SEQUENTIAL	2		/* expect sequential page references */
+#define MADV_WILLNEED	3		/* will need these pages */
+#define MADV_DONTNEED	4		/* don't need these pages */
+
+/* common parameters: try to keep these consistent across architectures */
+#define MADV_REMOVE	9		/* remove these pages & resources */
+#define MADV_DONTFORK	10		/* don't inherit across fork */
+#define MADV_DOFORK	11		/* do inherit across fork */
 
 /* compatibility flags */
-#define MAP_ANON       MAP_ANONYMOUS
-#define MAP_FILE       0
+#define MAP_ANON	MAP_ANONYMOUS
+#define MAP_FILE	0
 
 #endif /* _XTENSA_MMAN_H */

From b2ee9dbfad14ba8e34a589d552ddc67300a26bec Mon Sep 17 00:00:00 2001
From: Roman Zippel <zippel@linux-m68k.org>
Date: Wed, 15 Feb 2006 15:17:40 -0800
Subject: [PATCH 232/509] [PATCH] hrtimer: fix multiple macro argument
 expansion

For two macros the arguments were expanded twice, change them to inline
functions to avoid it.

Signed-off-by: Roman Zippel <zippel@linux-m68k.org>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/ktime.h | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/include/linux/ktime.h b/include/linux/ktime.h
index 6aca67a569a2..f3dec45ef874 100644
--- a/include/linux/ktime.h
+++ b/include/linux/ktime.h
@@ -96,10 +96,16 @@ static inline ktime_t ktime_set(const long secs, const unsigned long nsecs)
 		({ (ktime_t){ .tv64 = (kt).tv64 + (nsval) }; })
 
 /* convert a timespec to ktime_t format: */
-#define timespec_to_ktime(ts)		ktime_set((ts).tv_sec, (ts).tv_nsec)
+static inline ktime_t timespec_to_ktime(struct timespec ts)
+{
+	return ktime_set(ts.tv_sec, ts.tv_nsec);
+}
 
 /* convert a timeval to ktime_t format: */
-#define timeval_to_ktime(tv)		ktime_set((tv).tv_sec, (tv).tv_usec * 1000)
+static inline ktime_t timeval_to_ktime(struct timeval tv)
+{
+	return ktime_set(tv.tv_sec, tv.tv_usec * NSEC_PER_USEC);
+}
 
 /* Map the ktime_t to timespec conversion to ns_to_timespec function */
 #define ktime_to_timespec(kt)		ns_to_timespec((kt).tv64)

From d1db4ec86c7b1bf5b44d2ed3bf84a4bb53c33b1c Mon Sep 17 00:00:00 2001
From: Daniel Yeisley <dan.yeisley@unisys.com>
Date: Wed, 15 Feb 2006 15:17:41 -0800
Subject: [PATCH 233/509] [PATCH] x86_64: early initialization of cpu_to_node

The early initialization of cpu_to_node code as it is now only updates the
cpu_to_node array, and does not update cpu_pda()->nodemember.  This will
cause numa_node_id() to return 0 on systems where CPU 0 is not on Node 0.
This leads to a kernel panic in slab.c.

I've tested the patch below on a 16 processor x86_64 ES7000-600 server, and
no longer see the panic I saw with the original 2.6.16-rc3.

Signed-off-by: Dan Yeisley <dan.yeisley@unisys.com>
Acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/mm/numa.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/mm/numa.c b/arch/x86_64/mm/numa.c
index 6ef9f9a76235..22e51beee8d3 100644
--- a/arch/x86_64/mm/numa.c
+++ b/arch/x86_64/mm/numa.c
@@ -351,7 +351,7 @@ void __init init_cpu_to_node(void)
 			continue;
 		if (apicid_to_node[apicid] == NUMA_NO_NODE)
 			continue;
- 		cpu_to_node[i] = apicid_to_node[apicid];
+		numa_set_node(i,apicid_to_node[apicid]);
 	}
 }
 

From c8adb494a6df6b2be8e50a8dafd5bab231df3505 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Wed, 15 Feb 2006 15:17:42 -0800
Subject: [PATCH 234/509] [PATCH] swsusp: nuke noisy message

I get about 88 squillion of these when suspending an old ad450nx server.

Cc: Pavel Roskin <proski@gnu.org>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/snapshot.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 41f66365f0d8..8d5a5986d621 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -91,10 +91,8 @@ static int save_highmem_zone(struct zone *zone)
 		 * corrected eventually when the cases giving rise to this
 		 * are better understood.
 		 */
-		if (PageReserved(page)) {
-			printk("highmem reserved page?!\n");
+		if (PageReserved(page))
 			continue;
-		}
 		BUG_ON(PageNosave(page));
 		if (PageNosaveFree(page))
 			continue;

From 61be6d660093edde709ed638c7e1c458bd88c941 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Wed, 15 Feb 2006 15:17:43 -0800
Subject: [PATCH 235/509] [PATCH] mmconfig: add kernel parameter documentation

Mention the "pci=nommconf" option in kernel-parameters.txt.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 84370363da80..ac75b57edf2e 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1133,6 +1133,8 @@ running once the system is up.
 				Mechanism 1.
 		conf2		[IA-32] Force use of PCI Configuration
 				Mechanism 2.
+		nommconf	[IA-32,X86_64] Disable use of MMCONFIG for PCI
+				Configuration
 		nosort		[IA-32] Don't sort PCI devices according to
 				order given by the PCI BIOS. This sorting is
 				done to get a device order compatible with

From 7bbb79403163e047c6e333ff169db34e3c969e65 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Thu, 16 Feb 2006 11:08:09 +0000
Subject: [PATCH 236/509] [ARM] Fix SMP initialisation oops

A change to the SMP initialisation caused the following oops:

 CPU1: Booted secondary processor
 CPU1: D VIPT write-back cache
 CPU1: I cache: 32768 bytes, associativity 4, 32 byte lines, 256 sets
 CPU1: D cache: 32768 bytes, associativity 4, 32 byte lines, 256 sets
 <7>Calibrating delay loop... 83.14 BogoMIPS (lpj=415744)
 <1>Unable to handle kernel NULL pointer dereference at virtual address 0000001c
 ...
 PC is at enqueue_task+0x1c/0x64
 LR is at activate_task+0xcc/0xe4

SMP initialisation now requires cpu_possible_map to be initialised in
setup_arch().  Move this from smp_prepare_cpus() to smp_init_cpus()
and call it from our setup_arch() if CONFIG_SMP is enabled.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/setup.c            |  5 +++++
 arch/arm/kernel/smp.c              |  1 -
 arch/arm/mach-integrator/platsmp.c | 21 +++++++++++++++------
 arch/arm/mach-realview/platsmp.c   | 21 +++++++++++++++------
 include/asm-arm/smp.h              |  5 +++++
 5 files changed, 40 insertions(+), 13 deletions(-)

diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index c45d10d07bde..68273b4dc882 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -23,6 +23,7 @@
 #include <linux/root_dev.h>
 #include <linux/cpu.h>
 #include <linux/interrupt.h>
+#include <linux/smp.h>
 
 #include <asm/cpu.h>
 #include <asm/elf.h>
@@ -771,6 +772,10 @@ void __init setup_arch(char **cmdline_p)
 	paging_init(&meminfo, mdesc);
 	request_standard_resources(&meminfo, mdesc);
 
+#ifdef CONFIG_SMP
+	smp_init_cpus();
+#endif
+
 	cpu_init();
 
 	/*
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 7338948bd7d3..02aa300c4633 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -338,7 +338,6 @@ void __init smp_prepare_boot_cpu(void)
 
 	per_cpu(cpu_data, cpu).idle = current;
 
-	cpu_set(cpu, cpu_possible_map);
 	cpu_set(cpu, cpu_present_map);
 	cpu_set(cpu, cpu_online_map);
 }
diff --git a/arch/arm/mach-integrator/platsmp.c b/arch/arm/mach-integrator/platsmp.c
index ea10bd8c972c..1bc8534ef0c6 100644
--- a/arch/arm/mach-integrator/platsmp.c
+++ b/arch/arm/mach-integrator/platsmp.c
@@ -140,6 +140,18 @@ static void __init poke_milo(void)
 	mb();
 }
 
+/*
+ * Initialise the CPU possible map early - this describes the CPUs
+ * which may be present or become present in the system.
+ */
+void __init smp_init_cpus(void)
+{
+	unsigned int i, ncores = get_core_count();
+
+	for (i = 0; i < ncores; i++)
+		cpu_set(i, cpu_possible_map);
+}
+
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
 	unsigned int ncores = get_core_count();
@@ -176,14 +188,11 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 		max_cpus = ncores;
 
 	/*
-	 * Initialise the possible/present maps.
-	 * cpu_possible_map describes the set of CPUs which may be present
-	 * cpu_present_map describes the set of CPUs populated
+	 * Initialise the present map, which describes the set of CPUs
+	 * actually populated at the present time.
 	 */
-	for (i = 0; i < max_cpus; i++) {
-		cpu_set(i, cpu_possible_map);
+	for (i = 0; i < max_cpus; i++)
 		cpu_set(i, cpu_present_map);
-	}
 
 	/*
 	 * Do we need any more CPUs? If so, then let them know where
diff --git a/arch/arm/mach-realview/platsmp.c b/arch/arm/mach-realview/platsmp.c
index a8fbd76d8be5..b8484e15dacb 100644
--- a/arch/arm/mach-realview/platsmp.c
+++ b/arch/arm/mach-realview/platsmp.c
@@ -143,6 +143,18 @@ static void __init poke_milo(void)
 	mb();
 }
 
+/*
+ * Initialise the CPU possible map early - this describes the CPUs
+ * which may be present or become present in the system.
+ */
+void __init smp_init_cpus(void)
+{
+	unsigned int i, ncores = get_core_count();
+
+	for (i = 0; i < ncores; i++)
+		cpu_set(i, cpu_possible_map);
+}
+
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
 	unsigned int ncores = get_core_count();
@@ -179,14 +191,11 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	local_timer_setup(cpu);
 
 	/*
-	 * Initialise the possible/present maps.
-	 * cpu_possible_map describes the set of CPUs which may be present
-	 * cpu_present_map describes the set of CPUs populated
+	 * Initialise the present map, which describes the set of CPUs
+	 * actually populated at the present time.
 	 */
-	for (i = 0; i < max_cpus; i++) {
-		cpu_set(i, cpu_possible_map);
+	for (i = 0; i < max_cpus; i++)
 		cpu_set(i, cpu_present_map);
-	}
 
 	/*
 	 * Do we need any more CPUs? If so, then let them know where
diff --git a/include/asm-arm/smp.h b/include/asm-arm/smp.h
index 5a72e50ca9fc..fe45f7f61223 100644
--- a/include/asm-arm/smp.h
+++ b/include/asm-arm/smp.h
@@ -41,6 +41,11 @@ extern void show_ipi_list(struct seq_file *p);
  */
 asmlinkage void do_IPI(struct pt_regs *regs);
 
+/*
+ * Setup the SMP cpu_possible_map
+ */
+extern void smp_init_cpus(void);
+
 /*
  * Move global data into per-processor storage.
  */

From 90f9dd8f72773152b69042debd6b9ed6d224703a Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.de>
Date: Thu, 16 Feb 2006 14:43:01 +1100
Subject: [PATCH 237/509] [PATCH] Fix over-zealous tag clearing in
 radix_tree_delete

If a tag is set for a node being deleted from a radix_tree, then that
tag gets cleared from the parent of the node, even if it is set for some
siblings of the node begin deleted.

This patch changes the logic to include a test for any_tag_set similar
to the logic a little futher down.  Care is taken to ensure that
'nr_cleared_tags' remains equals to the number of entries in the 'tags'
array which are set to '0' (which means that this tag is not set in the
tree below pathp->node, and should be cleared at pathp->node and
possibly above.

[ Nick says: "Linus FYI, I was able to modify the radix tree test
  harness to catch the bug and can no longer trigger it after the fix.
  Resulting code passes all other harness tests as well of course." ]

Signed-off-by: Neil Brown <neilb@suse.de>
Acked-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 lib/radix-tree.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/lib/radix-tree.c b/lib/radix-tree.c
index c0bd4a914803..1e5b17dc7e3d 100644
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -752,12 +752,14 @@ void *radix_tree_delete(struct radix_tree_root *root, unsigned long index)
 	 */
 	nr_cleared_tags = 0;
 	for (tag = 0; tag < RADIX_TREE_TAGS; tag++) {
+		tags[tag] = 1;
 		if (tag_get(pathp->node, tag, pathp->offset)) {
 			tag_clear(pathp->node, tag, pathp->offset);
-			tags[tag] = 0;
-			nr_cleared_tags++;
-		} else
-			tags[tag] = 1;
+			if (!any_tag_set(pathp->node, tag)) {
+				tags[tag] = 0;
+				nr_cleared_tags++;
+			}
+		}
 	}
 
 	for (pathp--; nr_cleared_tags && pathp->node; pathp--) {

From 0425a14213f373595bd23cacdc675f2b973a28d4 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Thu, 16 Feb 2006 16:48:31 +0000
Subject: [PATCH 238/509] [MMC] mmci: allow small data transfers

If a data transfer is small (less than a FIFO size) we would
hang waiting for the data to be read due to the PIO interrupt
not occuring.  We allowed for this in our PIO interrupt handler,
but not when setting up a data transfer.

Apply the "fix" when setting up a data transfer as well.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/mmc/mmci.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/mmc/mmci.c b/drivers/mmc/mmci.c
index 37ee7f8dc82f..9fef29d978b5 100644
--- a/drivers/mmc/mmci.c
+++ b/drivers/mmc/mmci.c
@@ -97,6 +97,13 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
 	if (data->flags & MMC_DATA_READ) {
 		datactrl |= MCI_DPSM_DIRECTION;
 		irqmask = MCI_RXFIFOHALFFULLMASK;
+
+		/*
+		 * If we have less than a FIFOSIZE of bytes to transfer,
+		 * trigger a PIO interrupt as soon as any data is available.
+		 */
+		if (host->size < MCI_FIFOSIZE)
+			irqmask |= MCI_RXDATAAVLBLMASK;
 	} else {
 		/*
 		 * We don't actually need to include "FIFO empty" here

From 6f6d75825dc49b082906b84537b4df28293c2977 Mon Sep 17 00:00:00 2001
From: Jack Steiner <steiner@sgi.com>
Date: Wed, 15 Feb 2006 19:46:50 -0600
Subject: [PATCH 239/509] [IA64] Missing check for TIF_WORK if trace/audit
 enabled

It appears that if auditing is enabled, the kernel fails to
check for pending signals before returning to user mode.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Acked-by: Ken Chen <kenneth.w.chen@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 27b222c277e4..930fdfca6ddb 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -569,7 +569,9 @@ GLOBAL_ENTRY(ia64_trace_syscall)
 .mem.offset 0,0; st8.spill [r2]=r8		// store return value in slot for r8
 .mem.offset 8,0; st8.spill [r3]=r10		// clear error indication in slot for r10
 	br.call.sptk.many rp=syscall_trace_leave // give parent a chance to catch return value
-.ret3:	br.cond.sptk .work_pending_syscall_end
+.ret3:
+(pUStk)	cmp.eq.unc p6,p0=r0,r0			// p6 <- pUStk
+	br.cond.sptk .work_pending_syscall_end
 
 strace_error:
 	ld8 r3=[r2]				// load pt_regs.r8

From 898efface1a5076cbae5af87b935212b1869971b Mon Sep 17 00:00:00 2001
From: Kurt Hackel <kurt.hackel@oracle.com>
Date: Wed, 18 Jan 2006 17:01:25 -0800
Subject: [PATCH 240/509] [PATCH] ocfs2: recheck recovery state after getting
 lock

* after successfully taking the $RECOVERY lock in EX mode, recheck to make
  sure that recovery has not already begun or completed on another node

Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
---
 fs/ocfs2/dlm/dlmrecovery.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index 186e9a76aa58..f9ce864966ec 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -2032,6 +2032,30 @@ again:
 			     dlm->reco.new_master);
 			status = -EEXIST;
 		} else {
+			status = 0;
+
+			/* see if recovery was already finished elsewhere */
+			spin_lock(&dlm->spinlock);
+			if (dlm->reco.dead_node == O2NM_INVALID_NODE_NUM) {
+				status = -EINVAL;	
+				mlog(0, "%s: got reco EX lock, but "
+				     "node got recovered already\n", dlm->name);
+				if (dlm->reco.new_master != O2NM_INVALID_NODE_NUM) {
+					mlog(ML_ERROR, "%s: new master is %u "
+					     "but no dead node!\n", 
+					     dlm->name, dlm->reco.new_master);
+					BUG();
+				}
+			}
+			spin_unlock(&dlm->spinlock);
+		}
+
+		/* if this node has actually become the recovery master,
+		 * set the master and send the messages to begin recovery */
+		if (!status) {
+			mlog(0, "%s: dead=%u, this=%u, sending "
+			     "begin_reco now\n", dlm->name, 
+			     dlm->reco.dead_node, dlm->node_num);
 			status = dlm_send_begin_reco_message(dlm,
 				      dlm->reco.dead_node);
 			/* this always succeeds */

From e2b5e4506f5c5187b91d7a79fbad28fe3ebd2fc5 Mon Sep 17 00:00:00 2001
From: Kurt Hackel <kurt.hackel@oracle.com>
Date: Wed, 18 Jan 2006 17:02:56 -0800
Subject: [PATCH 241/509] [PATCH] ocfs2: fix release of ast never reserved

* fix a bug in dlm_convert_lock_handler where dlm_lockres_release_ast was
  being called even if no ast was ever reserved

Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
---
 fs/ocfs2/dlm/dlmconvert.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/ocfs2/dlm/dlmconvert.c b/fs/ocfs2/dlm/dlmconvert.c
index 6001b22a997d..f5c2f1979ad3 100644
--- a/fs/ocfs2/dlm/dlmconvert.c
+++ b/fs/ocfs2/dlm/dlmconvert.c
@@ -421,7 +421,7 @@ int dlm_convert_lock_handler(struct o2net_msg *msg, u32 len, void *data)
 	struct dlm_lockstatus *lksb;
 	enum dlm_status status = DLM_NORMAL;
 	u32 flags;
-	int call_ast = 0, kick_thread = 0;
+	int call_ast = 0, kick_thread = 0, ast_reserved = 0;
 
 	if (!dlm_grab(dlm)) {
 		dlm_error(DLM_REJECTED);
@@ -490,6 +490,7 @@ int dlm_convert_lock_handler(struct o2net_msg *msg, u32 len, void *data)
 	status = __dlm_lockres_state_to_status(res);
 	if (status == DLM_NORMAL) {
 		__dlm_lockres_reserve_ast(res);
+		ast_reserved = 1;
 		res->state |= DLM_LOCK_RES_IN_PROGRESS;
 		status = __dlmconvert_master(dlm, res, lock, flags,
 					     cnv->requested_type,
@@ -512,10 +513,10 @@ leave:
 	else
 		dlm_lock_put(lock);
 
-	/* either queue the ast or release it */
+	/* either queue the ast or release it, if reserved */
 	if (call_ast)
 		dlm_queue_ast(dlm, lock);
-	else
+	else if (ast_reserved)
 		dlm_lockres_release_ast(dlm, res);
 
 	if (kick_thread)

From 44465a7daf7c4e34199b2b0ebb3c5101619dcb9d Mon Sep 17 00:00:00 2001
From: Kurt Hackel <kurt.hackel@oracle.com>
Date: Wed, 18 Jan 2006 17:05:38 -0800
Subject: [PATCH 242/509] [PATCH] ocfs2: add dlm_wait_for_node_death

* add dlm_wait_for_node_death function to be used after receiving a network
  error.  this will wait for the given timeout to allow the heartbeat
  callbacks to update the domain map.  without this, some paths may spin
  and consume enough cpu that the heartbeat gets starved and never updates.

Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
---
 fs/ocfs2/dlm/dlmcommon.h   |  4 ++++
 fs/ocfs2/dlm/dlmconvert.c  |  5 +++++
 fs/ocfs2/dlm/dlmlock.c     | 14 +++++++++++++-
 fs/ocfs2/dlm/dlmrecovery.c | 18 ++++++++++++++++++
 4 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h
index 42eb53b5293b..23ceaa7127b4 100644
--- a/fs/ocfs2/dlm/dlmcommon.h
+++ b/fs/ocfs2/dlm/dlmcommon.h
@@ -208,6 +208,9 @@ static inline void __dlm_set_joining_node(struct dlm_ctxt *dlm,
 #define DLM_LOCK_RES_IN_PROGRESS          0x00000010
 #define DLM_LOCK_RES_MIGRATING            0x00000020
 
+/* max milliseconds to wait to sync up a network failure with a node death */
+#define DLM_NODE_DEATH_WAIT_MAX (5 * 1000)
+
 #define DLM_PURGE_INTERVAL_MS   (8 * 1000)
 
 struct dlm_lock_resource
@@ -658,6 +661,7 @@ int dlm_launch_recovery_thread(struct dlm_ctxt *dlm);
 void dlm_complete_recovery_thread(struct dlm_ctxt *dlm);
 void dlm_wait_for_recovery(struct dlm_ctxt *dlm);
 int dlm_is_node_dead(struct dlm_ctxt *dlm, u8 node);
+int dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout);
 
 void dlm_put(struct dlm_ctxt *dlm);
 struct dlm_ctxt *dlm_grab(struct dlm_ctxt *dlm);
diff --git a/fs/ocfs2/dlm/dlmconvert.c b/fs/ocfs2/dlm/dlmconvert.c
index f5c2f1979ad3..f66e2d818ccd 100644
--- a/fs/ocfs2/dlm/dlmconvert.c
+++ b/fs/ocfs2/dlm/dlmconvert.c
@@ -392,6 +392,11 @@ static enum dlm_status dlm_send_remote_convert_request(struct dlm_ctxt *dlm,
 	} else {
 		mlog_errno(tmpret);
 		if (dlm_is_host_down(tmpret)) {
+			/* instead of logging the same network error over
+			 * and over, sleep here and wait for the heartbeat
+			 * to notice the node is dead.  times out after 5s. */
+			dlm_wait_for_node_death(dlm, res->owner, 
+						DLM_NODE_DEATH_WAIT_MAX);
 			ret = DLM_RECOVERING;
 			mlog(0, "node %u died so returning DLM_RECOVERING "
 			     "from convert message!\n", res->owner);
diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c
index d1a0038557a3..e709412e6e32 100644
--- a/fs/ocfs2/dlm/dlmlock.c
+++ b/fs/ocfs2/dlm/dlmlock.c
@@ -646,7 +646,19 @@ retry_lock:
 			mlog(0, "retrying lock with migration/"
 			     "recovery/in progress\n");
 			msleep(100);
-			dlm_wait_for_recovery(dlm);
+			/* no waiting for dlm_reco_thread */
+			if (recovery) {
+				if (status == DLM_RECOVERING) {
+					mlog(0, "%s: got RECOVERING "
+					     "for $REOCVERY lock, master "
+					     "was %u\n", dlm->name, 
+					     res->owner);
+					dlm_wait_for_node_death(dlm, res->owner, 
+							DLM_NODE_DEATH_WAIT_MAX);
+				}
+			} else {
+				dlm_wait_for_recovery(dlm);
+			}
 			goto retry_lock;
 		}
 
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index f9ce864966ec..ed76bda1a534 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -278,6 +278,24 @@ int dlm_is_node_dead(struct dlm_ctxt *dlm, u8 node)
 	return dead;
 }
 
+int dlm_wait_for_node_death(struct dlm_ctxt *dlm, u8 node, int timeout)
+{
+	if (timeout) {
+		mlog(ML_NOTICE, "%s: waiting %dms for notification of "
+		     "death of node %u\n", dlm->name, timeout, node);
+		wait_event_timeout(dlm->dlm_reco_thread_wq,
+			   dlm_is_node_dead(dlm, node),
+			   msecs_to_jiffies(timeout));
+	} else {
+		mlog(ML_NOTICE, "%s: waiting indefinitely for notification "
+		     "of death of node %u\n", dlm->name, node);
+		wait_event(dlm->dlm_reco_thread_wq,
+			   dlm_is_node_dead(dlm, node));
+	}
+	/* for now, return 0 */
+	return 0;
+}
+
 /* callers of the top-level api calls (dlmlock/dlmunlock) should
  * block on the dlm->reco.event when recovery is in progress.
  * the dlm recovery thread will set this state when it begins

From 558c70c59b75a5a53ba496fe3bccea80a9e3e6fb Mon Sep 17 00:00:00 2001
From: Kurt Hackel <kurt.hackel@oracle.com>
Date: Wed, 18 Jan 2006 17:07:47 -0800
Subject: [PATCH 243/509] [PATCH] ocfs2: manually grant remote recovery lock

* fix a hang in recovery that occurred in dlmlock_remote.  the $RECOVERY
  lock was never moved to the granted queue even after getting DLM_NORMAL
  back from the master node.

Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
---
 fs/ocfs2/dlm/dlmlock.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/fs/ocfs2/dlm/dlmlock.c b/fs/ocfs2/dlm/dlmlock.c
index e709412e6e32..671d4ff222cc 100644
--- a/fs/ocfs2/dlm/dlmlock.c
+++ b/fs/ocfs2/dlm/dlmlock.c
@@ -220,6 +220,17 @@ static enum dlm_status dlmlock_remote(struct dlm_ctxt *dlm,
 			dlm_error(status);
 		dlm_revert_pending_lock(res, lock);
 		dlm_lock_put(lock);
+	} else if (dlm_is_recovery_lock(res->lockname.name, 
+					res->lockname.len)) {
+		/* special case for the $RECOVERY lock.
+		 * there will never be an AST delivered to put
+		 * this lock on the proper secondary queue
+		 * (granted), so do it manually. */
+		mlog(0, "%s: $RECOVERY lock for this node (%u) is "
+		     "mastered by %u; got lock, manually granting (no ast)\n",
+		     dlm->name, dlm->node_num, res->owner);
+		list_del_init(&lock->list);
+		list_add_tail(&lock->list, &res->granted);
 	}
 	spin_unlock(&res->spinlock);
 

From 745ae8ba29e729ec922393fa4d9448c385673599 Mon Sep 17 00:00:00 2001
From: Mark Fasheh <mark.fasheh@oracle.com>
Date: Thu, 9 Feb 2006 13:23:39 -0800
Subject: [PATCH 244/509] [PATCH] ocfs2: only checkpoint journal when asked to

Disable automatic checkpointing of the journal - this is a relic from older
ocfs2 days. Worth quite a bit of performance on longer running single node
tests.

Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
---
 fs/ocfs2/journal.c | 7 +++----
 fs/ocfs2/journal.h | 2 --
 2 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c
index fa0bcac5ceae..d329c9df90ae 100644
--- a/fs/ocfs2/journal.c
+++ b/fs/ocfs2/journal.c
@@ -1584,10 +1584,9 @@ static int ocfs2_commit_thread(void *arg)
 	while (!(kthread_should_stop() &&
 		 atomic_read(&journal->j_num_trans) == 0)) {
 
-		wait_event_interruptible_timeout(osb->checkpoint_event,
-						 atomic_read(&journal->j_num_trans)
-						 || kthread_should_stop(),
-						 OCFS2_CHECKPOINT_INTERVAL);
+		wait_event_interruptible(osb->checkpoint_event,
+					 atomic_read(&journal->j_num_trans)
+					 || kthread_should_stop());
 
 		status = ocfs2_commit_cache(osb);
 		if (status < 0)
diff --git a/fs/ocfs2/journal.h b/fs/ocfs2/journal.h
index 7d0a816184fa..2f3a6acdac45 100644
--- a/fs/ocfs2/journal.h
+++ b/fs/ocfs2/journal.h
@@ -29,8 +29,6 @@
 #include <linux/fs.h>
 #include <linux/jbd.h>
 
-#define OCFS2_CHECKPOINT_INTERVAL        (8 * HZ)
-
 enum ocfs2_journal_state {
 	OCFS2_JOURNAL_FREE = 0,
 	OCFS2_JOURNAL_LOADED,

From f671c09bce88ea253d576c842f8f39d9a2a29028 Mon Sep 17 00:00:00 2001
From: Kurt Hackel <Kurt.Hackel@oracle.com>
Date: Tue, 14 Feb 2006 11:45:21 -0800
Subject: [PATCH 245/509] [PATCH] ocfs2: detach from heartbeat events before
 freeing mle

Signed-off-by: Kurt Hackel <Kurt.Hackel@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
---
 fs/ocfs2/dlm/dlmmaster.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index a3194fe173d9..2e2e95e69499 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -2482,7 +2482,9 @@ top:
 				atomic_set(&mle->woken, 1);
 				spin_unlock(&mle->spinlock);
 				wake_up(&mle->wq);
-				/* final put will take care of list removal */
+				/* do not need events any longer, so detach 
+				 * from heartbeat */
+				__dlm_mle_detach_hb_events(dlm, mle);
 				__dlm_put_mle(mle);
 			}
 			continue;
@@ -2537,6 +2539,9 @@ top:
 			spin_unlock(&res->spinlock);
 			dlm_lockres_put(res);
 
+			/* about to get rid of mle, detach from heartbeat */
+			__dlm_mle_detach_hb_events(dlm, mle);
+
 			/* dump the mle */
 			spin_lock(&dlm->master_lock);
 			__dlm_put_mle(mle);

From 10487fbd74f2f9c9b5fb2d77efd50eab48db191a Mon Sep 17 00:00:00 2001
From: Francois Romieu <romieu@electric-eye.fr.zoreil.com>
Date: Thu, 16 Feb 2006 22:17:00 +0100
Subject: [PATCH 246/509] sis190: early setting of the pci driver private data

Below this point, the error path will proceed through
sis190_release_board(). It will happily oops if
pci_set_drvdata() has not been issued.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
---
 drivers/net/sis190.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/sis190.c b/drivers/net/sis190.c
index b420182eec4b..ed4bc91638d2 100644
--- a/drivers/net/sis190.c
+++ b/drivers/net/sis190.c
@@ -1791,6 +1791,8 @@ static int __devinit sis190_init_one(struct pci_dev *pdev,
 		goto out;
 	}
 
+	pci_set_drvdata(pdev, dev);
+
 	tp = netdev_priv(dev);
 	ioaddr = tp->mmio_addr;
 
@@ -1827,8 +1829,6 @@ static int __devinit sis190_init_one(struct pci_dev *pdev,
 	if (rc < 0)
 		goto err_remove_mii;
 
-	pci_set_drvdata(pdev, dev);
-
 	net_probe(tp, KERN_INFO "%s: %s at %p (IRQ: %d), "
 	       "%2.2x:%2.2x:%2.2x:%2.2x:%2.2x:%2.2x\n",
 	       pci_name(pdev), sis_chip_info[ent->driver_data].name,

From 8f8b1138fc9f65e3591aac83a4ee394fef34ac1d Mon Sep 17 00:00:00 2001
From: Ashok Raj <ashok.raj@intel.com>
Date: Thu, 16 Feb 2006 14:01:48 -0800
Subject: [PATCH 247/509] [IA64] Count disabled cpus as potential hot-pluggable
 CPUs

Minor updates to earlier patch.
- Added to documentation to add ia64 as well.
- Minor clarification on how to use disabled cpus
- used plain max instead of max_t per Andew Morton.

Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 Documentation/cpu-hotplug.txt | 14 ++++++++++++--
 arch/ia64/kernel/acpi.c       | 19 ++++++++-----------
 2 files changed, 20 insertions(+), 13 deletions(-)

diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt
index 08c5d04f3086..e05278087ffa 100644
--- a/Documentation/cpu-hotplug.txt
+++ b/Documentation/cpu-hotplug.txt
@@ -44,10 +44,20 @@ maxcpus=n    Restrict boot time cpus to n. Say if you have 4 cpus, using
              maxcpus=2 will only boot 2. You can choose to bring the
              other cpus later online, read FAQ's for more info.
 
-additional_cpus=n	[x86_64 only] use this to limit hotpluggable cpus.
-                        This option sets
+additional_cpus*=n	Use this to limit hotpluggable cpus. This option sets
 			cpu_possible_map = cpu_present_map + additional_cpus
 
+(*) Option valid only for following architectures
+- x86_64, ia64
+
+ia64 and x86_64 use the number of disabled local apics in ACPI tables MADT
+to determine the number of potentially hot-pluggable cpus. The implementation
+should only rely on this to count the #of cpus, but *MUST* not rely on the
+apicid values in those tables for disabled apics. In the event BIOS doesnt
+mark such hot-pluggable cpus as disabled entries, one could use this
+parameter "additional_cpus=x" to represent those cpus in the cpu_possible_map.
+
+
 CPU maps and such
 -----------------
 [More on cpumaps and primitive to manipulate, please check
diff --git a/arch/ia64/kernel/acpi.c b/arch/ia64/kernel/acpi.c
index 34795ede72e0..ecd44bdc8394 100644
--- a/arch/ia64/kernel/acpi.c
+++ b/arch/ia64/kernel/acpi.c
@@ -794,24 +794,21 @@ __init void prefill_possible_map(void)
 	int possible, disabled_cpus;
 
 	disabled_cpus = total_cpus - available_cpus;
+
  	if (additional_cpus == -1) {
- 		if (disabled_cpus > 0) {
- 			possible = total_cpus;
+ 		if (disabled_cpus > 0)
 			additional_cpus = disabled_cpus;
-		}
- 		else {
-			possible = available_cpus;
+ 		else
 			additional_cpus = 0;
-		}
- 	} else {
-		possible = available_cpus + additional_cpus;
-	}
+ 	}
+
+	possible = available_cpus + additional_cpus;
+
 	if (possible > NR_CPUS)
 		possible = NR_CPUS;
 
 	printk(KERN_INFO "SMP: Allowing %d CPUs, %d hotplug CPUs\n",
-		possible,
-	        max_t(int, additional_cpus, 0));
+		possible, max((possible - available_cpus), 0));
 
 	for (i = 0; i < possible; i++)
 		cpu_set(i, cpu_possible_map);

From 3dfaf7a68e275a1a6bee4861fdd61f911e6eb7a2 Mon Sep 17 00:00:00 2001
From: Martin Michlmayr <tbm@cyrius.com>
Date: Thu, 16 Feb 2006 22:36:12 +0000
Subject: [PATCH 248/509] [ARM] 3337/1: Fix NSLU2 flash support according to
 window size configuration patch
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Patch from Martin Michlmayr

ARM patch 3226/1 (IXP4xx runtime expansion bus window size configuration)
forgot to update mach-ixp4xx/nslu2-setup.c which leads to the following
compilation error.  Update NSLU2 flash support following patch 3226/1.

  CC      arch/arm/mach-ixp4xx/nslu2-setup.o
arch/arm/mach-ixp4xx/nslu2-setup.c:30: error: ‘NSLU2_FLASH_BASE’ undeclared here (not in a function)
arch/arm/mach-ixp4xx/nslu2-setup.c:31: error: ‘NSLU2_FLASH_SIZE’ undeclared here (not in a function)
make[1]: *** [arch/arm/mach-ixp4xx/nslu2-setup.o] Error 1
make: *** [arch/arm/mach-ixp4xx] Error 2

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>

---

 nslu2-setup.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp4xx/nslu2-setup.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-ixp4xx/nslu2-setup.c b/arch/arm/mach-ixp4xx/nslu2-setup.c
index da9340a53434..f260a9d34f70 100644
--- a/arch/arm/mach-ixp4xx/nslu2-setup.c
+++ b/arch/arm/mach-ixp4xx/nslu2-setup.c
@@ -27,8 +27,6 @@ static struct flash_platform_data nslu2_flash_data = {
 };
 
 static struct resource nslu2_flash_resource = {
-	.start			= NSLU2_FLASH_BASE,
-	.end			= NSLU2_FLASH_BASE + NSLU2_FLASH_SIZE,
 	.flags			= IORESOURCE_MEM,
 };
 
@@ -116,6 +114,10 @@ static void __init nslu2_init(void)
 {
 	ixp4xx_sys_init();
 
+	nslu2_flash_resource.start = IXP4XX_EXP_BUS_BASE(0);
+	nslu2_flash_resource.end =
+		IXP4XX_EXP_BUS_BASE(0) + ixp4xx_exp_bus_size - 1;
+
 	pm_power_off = nslu2_power_off;
 
 	platform_add_devices(nslu2_devices, ARRAY_SIZE(nslu2_devices));

From 6c0fa49b18b09ba9e69c0999f89bc38fad95d8a6 Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Thu, 16 Feb 2006 22:36:13 +0000
Subject: [PATCH 249/509] [ARM] 3338/1: old ABI compat: sys_socketcall

Patch from Nicolas Pitre

Commit 99595d0237926b5aba1fe4c844a011a1ba1ee1f8 forgot to intercept
sys_socketcall as well.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/calls.S           |  2 +-
 arch/arm/kernel/sys_oabi-compat.c | 30 ++++++++++++++++++++++++++++++
 2 files changed, 31 insertions(+), 1 deletion(-)

diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S
index 8c3035d5ffc9..3173924a9b60 100644
--- a/arch/arm/kernel/calls.S
+++ b/arch/arm/kernel/calls.S
@@ -111,7 +111,7 @@
 		CALL(sys_statfs)
 /* 100 */	CALL(sys_fstatfs)
 		CALL(sys_ni_syscall)
-		CALL(OBSOLETE(sys_socketcall))
+		CALL(OBSOLETE(ABI(sys_socketcall, sys_oabi_socketcall)))
 		CALL(sys_syslog)
 		CALL(sys_setitimer)
 /* 105 */	CALL(sys_getitimer)
diff --git a/arch/arm/kernel/sys_oabi-compat.c b/arch/arm/kernel/sys_oabi-compat.c
index 9d4b76409c64..8e2f9bc3368b 100644
--- a/arch/arm/kernel/sys_oabi-compat.c
+++ b/arch/arm/kernel/sys_oabi-compat.c
@@ -64,6 +64,7 @@
  * sys_connect:
  * sys_sendmsg:
  * sys_sendto:
+ * sys_socketcall:
  *
  *   struct sockaddr_un loses its padding with EABI.  Since the size of the
  *   structure is used as a validation test in unix_mkname(), we need to
@@ -78,6 +79,7 @@
 #include <linux/eventpoll.h>
 #include <linux/sem.h>
 #include <linux/socket.h>
+#include <linux/net.h>
 #include <asm/ipc.h>
 #include <asm/uaccess.h>
 
@@ -408,3 +410,31 @@ asmlinkage long sys_oabi_sendmsg(int fd, struct msghdr __user *msg, unsigned fla
 	return sys_sendmsg(fd, msg, flags);
 }
 
+asmlinkage long sys_oabi_socketcall(int call, unsigned long __user *args)
+{
+	unsigned long r = -EFAULT, a[6];
+
+	switch (call) {
+	case SYS_BIND:
+		if (copy_from_user(a, args, 3 * sizeof(long)) == 0)
+			r = sys_oabi_bind(a[0], (struct sockaddr __user *)a[1], a[2]);
+		break;
+	case SYS_CONNECT:
+		if (copy_from_user(a, args, 3 * sizeof(long)) == 0)
+			r = sys_oabi_connect(a[0], (struct sockaddr __user *)a[1], a[2]);
+		break;
+	case SYS_SENDTO:
+		if (copy_from_user(a, args, 6 * sizeof(long)) == 0)
+			r = sys_oabi_sendto(a[0], (void __user *)a[1], a[2], a[3],
+					    (struct sockaddr __user *)a[4], a[5]);
+		break;
+	case SYS_SENDMSG:
+		if (copy_from_user(a, args, 3 * sizeof(long)) == 0)
+			r = sys_oabi_sendmsg(a[0], (struct msghdr __user *)a[1], a[2]);
+		break;
+	default:
+		r = sys_socketcall(call, args);
+	}
+
+	return r;
+}

From d9db950cfa3d674ee834d980c329efdf8e4a0568 Mon Sep 17 00:00:00 2001
From: Nicolas Pitre <nico@cam.org>
Date: Thu, 16 Feb 2006 22:36:15 +0000
Subject: [PATCH 250/509] [ARM] 3339/1: ARM EABI: make unmuxed syscalls visible

Patch from Nicolas Pitre

With EABI the multiplex sys_ipc and sys_socketcall syscalls are
unavailable and their support code even removed from the compiled
kernel, and the new unmuxed syscalls must be used instead.

Make those syscall numbers visible.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/unistd.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/asm-arm/unistd.h b/include/asm-arm/unistd.h
index 77430d6178ae..8f331bbd39a8 100644
--- a/include/asm-arm/unistd.h
+++ b/include/asm-arm/unistd.h
@@ -309,7 +309,7 @@
 #define __NR_mq_getsetattr		(__NR_SYSCALL_BASE+279)
 #define __NR_waitid			(__NR_SYSCALL_BASE+280)
 
-#if 0 /* reserve these for un-muxing socketcall */
+#if defined(__ARM_EABI__)  /* reserve these for un-muxing socketcall */
 #define __NR_socket			(__NR_SYSCALL_BASE+281)
 #define __NR_bind			(__NR_SYSCALL_BASE+282)
 #define __NR_connect			(__NR_SYSCALL_BASE+283)
@@ -329,7 +329,7 @@
 #define __NR_recvmsg			(__NR_SYSCALL_BASE+297)
 #endif
 
-#if 0 /* reserve these for un-muxing ipc */
+#if defined(__ARM_EABI__)  /* reserve these for un-muxing ipc */
 #define __NR_semop			(__NR_SYSCALL_BASE+298)
 #define __NR_semget			(__NR_SYSCALL_BASE+299)
 #define __NR_semctl			(__NR_SYSCALL_BASE+300)
@@ -347,7 +347,7 @@
 #define __NR_request_key		(__NR_SYSCALL_BASE+310)
 #define __NR_keyctl			(__NR_SYSCALL_BASE+311)
 
-#if 0 /* reserved for un-muxing ipc */
+#if defined(__ARM_EABI__)  /* reserved for un-muxing ipc */
 #define __NR_semtimedop			(__NR_SYSCALL_BASE+312)
 #endif
 

From 0d467502b7fc2656f01d7f18ab290c8d41762018 Mon Sep 17 00:00:00 2001
From: Dan Williams <dcbw@redhat.com>
Date: Sun, 5 Feb 2006 17:52:21 -0500
Subject: [PATCH 251/509] [PATCH] wireless/atmel: fix setting TX key only in
 ENCODEEXT

The previous patch that added ENCODEEXT and AUTH support to the atmel
driver contained a slight error which would cause just setting the TX
key index to also set the encryption key again.  This patch allows any
combination of setting the TX key index and setting an encryption key.

Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/atmel.c | 63 +++++++++++++++++++-----------------
 1 file changed, 33 insertions(+), 30 deletions(-)

diff --git a/drivers/net/wireless/atmel.c b/drivers/net/wireless/atmel.c
index 98a76f10a0f7..c1c27e10b8df 100644
--- a/drivers/net/wireless/atmel.c
+++ b/drivers/net/wireless/atmel.c
@@ -1872,7 +1872,7 @@ static int atmel_set_encodeext(struct net_device *dev,
 	struct atmel_private *priv = netdev_priv(dev);
 	struct iw_point *encoding = &wrqu->encoding;
 	struct iw_encode_ext *ext = (struct iw_encode_ext *)extra;
-	int idx, key_len;
+	int idx, key_len, alg = ext->alg, set_key = 1;
 
 	/* Determine and validate the key index */
 	idx = encoding->flags & IW_ENCODE_INDEX;
@@ -1883,39 +1883,42 @@ static int atmel_set_encodeext(struct net_device *dev,
 	} else
 		idx = priv->default_key;
 
-	if ((encoding->flags & IW_ENCODE_DISABLED) ||
-	    ext->alg == IW_ENCODE_ALG_NONE) {
-		priv->wep_is_on = 0;
-		priv->encryption_level = 0;
-		priv->pairwise_cipher_suite = CIPHER_SUITE_NONE;
+	if (encoding->flags & IW_ENCODE_DISABLED)
+	    alg = IW_ENCODE_ALG_NONE;
+
+	if (ext->ext_flags & IW_ENCODE_EXT_SET_TX_KEY) {
+		priv->default_key = idx;
+		set_key = ext->key_len > 0 ? 1 : 0;
 	}
 
-	if (ext->ext_flags & IW_ENCODE_EXT_SET_TX_KEY)
-		priv->default_key = idx;
-
-	/* Set the requested key */
-	switch (ext->alg) {
-	case IW_ENCODE_ALG_NONE:
-		break;
-	case IW_ENCODE_ALG_WEP:
-		if (ext->key_len > 5) {
-			priv->wep_key_len[idx] = 13;
-			priv->pairwise_cipher_suite = CIPHER_SUITE_WEP_128;
-			priv->encryption_level = 2;
-		} else if (ext->key_len > 0) {
-			priv->wep_key_len[idx] = 5;
-			priv->pairwise_cipher_suite = CIPHER_SUITE_WEP_64;
-			priv->encryption_level = 1;
-		} else {
+	if (set_key) {
+		/* Set the requested key first */
+		switch (alg) {
+		case IW_ENCODE_ALG_NONE:
+			priv->wep_is_on = 0;
+			priv->encryption_level = 0;
+			priv->pairwise_cipher_suite = CIPHER_SUITE_NONE;
+			break;
+		case IW_ENCODE_ALG_WEP:
+			if (ext->key_len > 5) {
+				priv->wep_key_len[idx] = 13;
+				priv->pairwise_cipher_suite = CIPHER_SUITE_WEP_128;
+				priv->encryption_level = 2;
+			} else if (ext->key_len > 0) {
+				priv->wep_key_len[idx] = 5;
+				priv->pairwise_cipher_suite = CIPHER_SUITE_WEP_64;
+				priv->encryption_level = 1;
+			} else {
+				return -EINVAL;
+			}
+			priv->wep_is_on = 1;
+			memset(priv->wep_keys[idx], 0, 13);
+			key_len = min ((int)ext->key_len, priv->wep_key_len[idx]);
+			memcpy(priv->wep_keys[idx], ext->key, key_len);
+			break;
+		default:
 			return -EINVAL;
 		}
-		priv->wep_is_on = 1;
-		memset(priv->wep_keys[idx], 0, 13);
-		key_len = min ((int)ext->key_len, priv->wep_key_len[idx]);
-		memcpy(priv->wep_keys[idx], ext->key, key_len);
-		break;
-	default:
-		return -EINVAL;
 	}
 
 	return -EINPROGRESS;

From 7345137930907ba747781636c60112f7c2789aa8 Mon Sep 17 00:00:00 2001
From: Dan Williams <dcbw@redhat.com>
Date: Sun, 5 Feb 2006 17:55:16 -0500
Subject: [PATCH 252/509] [PATCH] wireless/atmel: fix Open System
 authentication process bugs

This patch fixes a number of bugs in the authentication process:

1) When falling back to Shared Key authentication mode from Open System,
a missing 'return' would cause the auth request to be sent, but would
drop the card into Management Error state.  When falling back, the
driver should also indicate that it is switching to Shared Key mode by
setting exclude_unencrypted.

2) Initial authentication modes were apparently wrong in some cases,
causing the driver to attempt Shared Key authentication mode when in
fact the access point didn't support that mode or even had WEP disabled.
The driver should set the correct initial authentication mode based on
wep_is_on and exclude_unencrypted.

3) Authentication response packets from the access point in Open System
mode were getting ignored because the driver was expecting the sequence
number of a Shared Key mode response.  The patch separates the OS and SK
mode handling to provide the correct behavior.

Signed-off-by: Dan Williams <dcbw@redhat.com>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
---
 drivers/net/wireless/atmel.c | 37 ++++++++++++++++++++++++++----------
 1 file changed, 27 insertions(+), 10 deletions(-)

diff --git a/drivers/net/wireless/atmel.c b/drivers/net/wireless/atmel.c
index c1c27e10b8df..dfc24016ba81 100644
--- a/drivers/net/wireless/atmel.c
+++ b/drivers/net/wireless/atmel.c
@@ -3064,17 +3064,26 @@ static void authenticate(struct atmel_private *priv, u16 frame_len)
 	}
 
 	if (status == C80211_MGMT_SC_Success && priv->wep_is_on) {
+		int should_associate = 0;
 		/* WEP */
 		if (trans_seq_no != priv->ExpectedAuthentTransactionSeqNum)
 			return;
 
-		if (trans_seq_no == 0x0002 &&
-		    auth->el_id == C80211_MGMT_ElementID_ChallengeText) {
-			send_authentication_request(priv, system, auth->chall_text, auth->chall_text_len);
-			return;
+		if (system == C80211_MGMT_AAN_OPENSYSTEM) {
+			if (trans_seq_no == 0x0002) {
+				should_associate = 1;
+			}
+		} else if (system == C80211_MGMT_AAN_SHAREDKEY) {
+			if (trans_seq_no == 0x0002 &&
+			    auth->el_id == C80211_MGMT_ElementID_ChallengeText) {
+				send_authentication_request(priv, system, auth->chall_text, auth->chall_text_len);
+				return;
+			} else if (trans_seq_no == 0x0004) {
+				should_associate = 1;
+			}
 		}
 
-		if (trans_seq_no == 0x0004) {
+		if (should_associate) {
 			if(priv->station_was_associated) {
 				atmel_enter_state(priv, STATION_STATE_REASSOCIATING);
 				send_association_request(priv, 1);
@@ -3087,11 +3096,13 @@ static void authenticate(struct atmel_private *priv, u16 frame_len)
 		}
 	}
 
-	if (status == C80211_MGMT_SC_AuthAlgNotSupported) {
+	if (status == WLAN_STATUS_NOT_SUPPORTED_AUTH_ALG) {
 		/* Do opensystem first, then try sharedkey */
-		if (system ==  C80211_MGMT_AAN_OPENSYSTEM) {
+		if (system == WLAN_AUTH_OPEN) {
 			priv->CurrentAuthentTransactionSeqNum = 0x001;
-			send_authentication_request(priv, C80211_MGMT_AAN_SHAREDKEY, NULL, 0);
+			priv->exclude_unencrypted = 1;
+			send_authentication_request(priv, WLAN_AUTH_SHARED_KEY, NULL, 0);
+			return;
 		} else if (priv->connect_to_any_BSS) {
 			int bss_index;
 
@@ -3442,10 +3453,13 @@ static void atmel_management_timer(u_long a)
 			priv->AuthenticationRequestRetryCnt = 0;
 			restart_search(priv);
 		} else {
+			int auth = C80211_MGMT_AAN_OPENSYSTEM;
 			priv->AuthenticationRequestRetryCnt++;
 			priv->CurrentAuthentTransactionSeqNum = 0x0001;
 			mod_timer(&priv->management_timer, jiffies + MGMT_JIFFIES);
-			send_authentication_request(priv, C80211_MGMT_AAN_OPENSYSTEM, NULL, 0);
+			if (priv->wep_is_on && priv->exclude_unencrypted)
+				auth = C80211_MGMT_AAN_SHAREDKEY;
+			send_authentication_request(priv, auth, NULL, 0);
 	  }
 	  break;
 
@@ -3544,12 +3558,15 @@ static void atmel_command_irq(struct atmel_private *priv)
 				priv->station_was_associated = priv->station_is_associated;
 				atmel_enter_state(priv, STATION_STATE_READY);
 			} else {
+				int auth = C80211_MGMT_AAN_OPENSYSTEM;
 				priv->AuthenticationRequestRetryCnt = 0;
 				atmel_enter_state(priv, STATION_STATE_AUTHENTICATING);
 
 				mod_timer(&priv->management_timer, jiffies + MGMT_JIFFIES);
 				priv->CurrentAuthentTransactionSeqNum = 0x0001;
-				send_authentication_request(priv, C80211_MGMT_AAN_SHAREDKEY, NULL, 0);
+				if (priv->wep_is_on && priv->exclude_unencrypted)
+					auth = C80211_MGMT_AAN_SHAREDKEY;
+				send_authentication_request(priv, auth, NULL, 0);
 			}
 			return;
 		}

From e4444d1a3039354c388135073786efeb64d8ef0c Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Thu, 16 Feb 2006 23:41:52 +0100
Subject: [PATCH 253/509] [PATCH] x86_64: Update defconfig

... and enable 1394 by default.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/defconfig | 42 +++++++++++++++++++++++++++++++++---------
 1 file changed, 33 insertions(+), 9 deletions(-)

diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig
index 56832929a543..b337136f28b6 100644
--- a/arch/x86_64/defconfig
+++ b/arch/x86_64/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.16-rc1-git2
-# Thu Jan 19 10:05:21 2006
+# Linux kernel version: 2.6.16-rc3
+# Mon Feb 13 22:31:24 2006
 #
 CONFIG_X86_64=y
 CONFIG_64BIT=y
@@ -21,7 +21,6 @@ CONFIG_DMI=y
 # Code maturity level options
 #
 CONFIG_EXPERIMENTAL=y
-CONFIG_CLEAN_COMPILE=y
 CONFIG_LOCK_KERNEL=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
 
@@ -267,6 +266,7 @@ CONFIG_NET=y
 #
 # Networking options
 #
+# CONFIG_NETDEBUG is not set
 CONFIG_PACKET=y
 # CONFIG_PACKET_MMAP is not set
 CONFIG_UNIX=y
@@ -446,7 +446,6 @@ CONFIG_BLK_DEV_PIIX=y
 # CONFIG_BLK_DEV_NS87415 is not set
 # CONFIG_BLK_DEV_PDC202XX_OLD is not set
 CONFIG_BLK_DEV_PDC202XX_NEW=y
-# CONFIG_PDC202XX_FORCE is not set
 # CONFIG_BLK_DEV_SVWKS is not set
 # CONFIG_BLK_DEV_SIIMAGE is not set
 # CONFIG_BLK_DEV_SIS5513 is not set
@@ -573,7 +572,33 @@ CONFIG_FUSION_MAX_SGE=128
 #
 # IEEE 1394 (FireWire) support
 #
-# CONFIG_IEEE1394 is not set
+CONFIG_IEEE1394=y
+
+#
+# Subsystem Options
+#
+# CONFIG_IEEE1394_VERBOSEDEBUG is not set
+# CONFIG_IEEE1394_OUI_DB is not set
+# CONFIG_IEEE1394_EXTRA_CONFIG_ROMS is not set
+# CONFIG_IEEE1394_EXPORT_FULL_API is not set
+
+#
+# Device Drivers
+#
+
+#
+# Texas Instruments PCILynx requires I2C
+#
+CONFIG_IEEE1394_OHCI1394=y
+
+#
+# Protocol Drivers
+#
+# CONFIG_IEEE1394_VIDEO1394 is not set
+# CONFIG_IEEE1394_SBP2 is not set
+# CONFIG_IEEE1394_ETH1394 is not set
+# CONFIG_IEEE1394_DV1394 is not set
+CONFIG_IEEE1394_RAWIO=y
 
 #
 # I2O device support
@@ -772,6 +797,7 @@ CONFIG_SERIAL_8250_RUNTIME_UARTS=4
 #
 CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_JSM is not set
 CONFIG_UNIX98_PTYS=y
 CONFIG_LEGACY_PTYS=y
 CONFIG_LEGACY_PTY_COUNT=256
@@ -871,6 +897,7 @@ CONFIG_HPET_MMAP=y
 #
 CONFIG_HWMON=y
 # CONFIG_HWMON_VID is not set
+# CONFIG_SENSORS_F71805F is not set
 # CONFIG_SENSORS_HDAPS is not set
 # CONFIG_HWMON_DEBUG_CHIP is not set
 
@@ -1101,7 +1128,6 @@ CONFIG_USB_MON=y
 # EDAC - error detection and reporting (RAS)
 #
 # CONFIG_EDAC is not set
-# CONFIG_EDAC_POLL is not set
 
 #
 # Firmware Drivers
@@ -1291,14 +1317,12 @@ CONFIG_DETECT_SOFTLOCKUP=y
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
-# CONFIG_DEBUG_INFO is not set
+CONFIG_DEBUG_INFO=y
 CONFIG_DEBUG_FS=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_FRAME_POINTER is not set
 # CONFIG_FORCED_INLINING is not set
-# CONFIG_UNWIND_INFO is not set
 # CONFIG_RCU_TORTURE_TEST is not set
-CONFIG_INIT_DEBUG=y
 # CONFIG_DEBUG_RODATA is not set
 # CONFIG_IOMMU_DEBUG is not set
 

From 99019e919969be88e7e4042f3afa296bd55ad9ec Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Thu, 16 Feb 2006 23:41:55 +0100
Subject: [PATCH 254/509] [PATCH] x86_64: make touch_nmi_watchdog() not touch
 impossible cpus' private data

Along with that, also suppress the memory touching altogether when the
watchdog is not running, to eliminate needless crosstalk. Plus ad a call
to it to make things consistent (one could also consider removing the call
in enable_timer_nmi_watchdog()).

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/nmi.c | 19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)

diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 8be407a1f62d..5bf17e41cd2d 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -236,6 +236,7 @@ static void enable_lapic_nmi_watchdog(void)
 {
 	if (nmi_active < 0) {
 		nmi_watchdog = NMI_LOCAL_APIC;
+		touch_nmi_watchdog();
 		setup_apic_nmi_watchdog();
 	}
 }
@@ -456,15 +457,17 @@ static DEFINE_PER_CPU(int, nmi_touch);
 
 void touch_nmi_watchdog (void)
 {
-	int i;
+	if (nmi_watchdog > 0) {
+		unsigned cpu;
 
-	/*
- 	 * Tell other CPUs to reset their alert counters. We cannot
-	 * do it ourselves because the alert count increase is not
-	 * atomic.
-	 */
-	for (i = 0; i < NR_CPUS; i++)
-		per_cpu(nmi_touch, i) = 1;
+		/*
+ 		 * Tell other CPUs to reset their alert counters. We cannot
+		 * do it ourselves because the alert count increase is not
+		 * atomic.
+		 */
+		for_each_present_cpu (cpu)
+			per_cpu(nmi_touch, cpu) = 1;
+	}
 
  	touch_softlockup_watchdog();
 }

From a62eaf151d9cb478d127cfbc2e93c498869785b0 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Thu, 16 Feb 2006 23:41:58 +0100
Subject: [PATCH 255/509] [PATCH] x86_64: Add boot option to disable randomized
 mappings and cleanup

AMD SimNow!'s JIT doesn't like them at all in the guest. For distribution
installation it's easiest if it's a boot time option.

Also I moved the variable to a more appropiate place and make
it independent from sysctl

And marked __read_mostly which it is.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/kernel-parameters.txt |  3 +++
 arch/i386/kernel/cpu/transmeta.c    |  1 +
 include/linux/kernel.h              |  6 ------
 include/linux/mm.h                  |  2 ++
 kernel/sysctl.c                     |  2 --
 mm/memory.c                         | 10 ++++++++++
 6 files changed, 16 insertions(+), 8 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index ac75b57edf2e..b874771385cd 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1638,6 +1638,9 @@ running once the system is up.
 			Format:
 			<irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]]
 
+	norandmaps	Don't use address space randomization
+			Equivalent to echo 0 > /proc/sys/kernel/randomize_va_space
+
 
 ______________________________________________________________________
 Changelog:
diff --git a/arch/i386/kernel/cpu/transmeta.c b/arch/i386/kernel/cpu/transmeta.c
index bdbeb77f4e22..7214c9b577ab 100644
--- a/arch/i386/kernel/cpu/transmeta.c
+++ b/arch/i386/kernel/cpu/transmeta.c
@@ -1,4 +1,5 @@
 #include <linux/kernel.h>
+#include <linux/mm.h>
 #include <linux/init.h>
 #include <asm/processor.h>
 #include <asm/msr.h>
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index b49affa0ac5a..3b507bf05d09 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -326,12 +326,6 @@ struct sysinfo {
 /* Force a compilation error if condition is true */
 #define BUILD_BUG_ON(condition) ((void)sizeof(char[1 - 2*!!(condition)]))
 
-#ifdef CONFIG_SYSCTL
-extern int randomize_va_space;
-#else
-#define randomize_va_space 1
-#endif
-
 /* Trap pasters of __FUNCTION__ at compile-time */
 #define __FUNCTION__ (__func__)
 
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 75e9f0724997..26e1663a5cbe 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1051,5 +1051,7 @@ int shrink_slab(unsigned long scanned, gfp_t gfp_mask,
 void drop_pagecache(void);
 void drop_slab(void);
 
+extern int randomize_va_space;
+
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MM_H */
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 71dd6f62efec..7654d55c47f5 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -126,8 +126,6 @@ extern int sysctl_hz_timer;
 extern int acct_parm[];
 #endif
 
-int randomize_va_space = 1;
-
 static int parse_table(int __user *, int, void __user *, size_t __user *, void __user *, size_t,
 		       ctl_table *, void **);
 static int proc_doutsstring(ctl_table *table, int write, struct file *filp,
diff --git a/mm/memory.c b/mm/memory.c
index 2bee1f21aa8a..9abc6008544b 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -82,6 +82,16 @@ EXPORT_SYMBOL(num_physpages);
 EXPORT_SYMBOL(high_memory);
 EXPORT_SYMBOL(vmalloc_earlyreserve);
 
+int randomize_va_space __read_mostly = 1;
+
+static int __init disable_randmaps(char *s)
+{
+	randomize_va_space = 0;
+	return 0;
+}
+__setup("norandmaps", disable_randmaps);
+
+
 /*
  * If a p?d_bad entry is found while walking page tables, report
  * the error, before resetting entry to p?d_none.  Usually (but

From 2391c4b594eb28abd58102de8f4e5d7a4fa39f4c Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Thu, 16 Feb 2006 23:42:01 +0100
Subject: [PATCH 256/509] [PATCH] x86_64: Don't call do_exit with interrupts
 disabled after IRET exception

This caused a sigreturn with bad argument on a preemptible kernel
to complain with

Debug: sleeping function called from invalid context at /home/lsrc/quilt/linux/include/linux/rwsem.h:43
in_atomic():0, irqs_disabled():1

Call Trace: {__might_sleep+190} {profile_task_exit+21}
       {__do_exit+34} {do_wait+0}

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/entry.S | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S
index b150c87a08c6..7c10e9009d61 100644
--- a/arch/x86_64/kernel/entry.S
+++ b/arch/x86_64/kernel/entry.S
@@ -554,6 +554,7 @@ iret_label:
 	/* running with kernel gs */
 bad_iret:
 	movq $-9999,%rdi	/* better code? */
+	sti
 	jmp do_exit			
 	.previous	
 	

From ab68805955ee3dd84a6aa76cd70e61fde996968d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Thu, 16 Feb 2006 23:42:04 +0100
Subject: [PATCH 257/509] [PATCH] x86_64: Don't enable ATI apicmaintimer
 workaround when the machine has C2 or C3

Many laptops have problems with ticking the local APIC timer in C2/C3.
The code added earlier to use it by default on ATI didn't really work
for them. Don't enable it when the system supports C2/C3.

This doesn't fix the problem fully, but at least it's not worse than before.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/io_apic.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 4282d72b2a26..2585c1d92b26 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -30,6 +30,9 @@
 #include <linux/mc146818rtc.h>
 #include <linux/acpi.h>
 #include <linux/sysdev.h>
+#ifdef CONFIG_ACPI
+#include <acpi/acpi_bus.h>
+#endif
 
 #include <asm/io.h>
 #include <asm/smp.h>
@@ -260,6 +263,8 @@ __setup("apic", enable_ioapic_setup);
 
    And another hack to disable the IOMMU on VIA chipsets.
 
+   ... and others. Really should move this somewhere else.
+
    Kludge-O-Rama. */
 void __init check_ioapic(void) 
 { 
@@ -307,6 +312,17 @@ void __init check_ioapic(void)
 				case PCI_VENDOR_ID_ATI:
 					if (apic_runs_main_timer != 0)
 						break;
+#ifdef CONFIG_ACPI
+					/* Don't do this for laptops right
+					   right now because their timer
+					   doesn't necessarily tick in C2/3 */
+					if (acpi_fadt.revision >= 3 &&
+			(acpi_fadt.plvl2_lat + acpi_fadt.plvl3_lat) < 1100) {
+						printk(KERN_INFO
+"ATI board detected, but seems to be a laptop. Timer might be shakey, sorry\n");
+						break;
+					}
+#endif					
 					printk(KERN_INFO
 	     "ATI board detected. Using APIC/PM timer.\n");
 					apic_runs_main_timer = 1;

From 7fd67843b96f90f59c9a244a1bc25137978a3ff9 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Thu, 16 Feb 2006 23:42:07 +0100
Subject: [PATCH 258/509] [PATCH] x86_64: Disable tsc when apicpmtimer is
 active

Otherwise it has no effect anyways.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/apic.c  | 1 +
 arch/x86_64/kernel/time.c  | 3 +--
 include/asm-x86_64/proto.h | 1 +
 3 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index 7a0a3e8d5d72..e5b14c57eaa0 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -1152,6 +1152,7 @@ __setup("noapicmaintimer", setup_noapicmaintimer);
 static __init int setup_apicpmtimer(char *s)
 {
 	apic_calibrate_pmtmr = 1;
+	notsc_setup(NULL);
 	return setup_apicmaintimer(NULL);
 }
 __setup("apicpmtimer", setup_apicpmtimer);
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 3c58c30506a1..67841d11ed1f 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -1327,8 +1327,7 @@ static int __init nohpet_setup(char *s)
 
 __setup("nohpet", nohpet_setup);
 
-
-static int __init notsc_setup(char *s)
+int __init notsc_setup(char *s)
 {
 	notsc = 1;
 	return 0;
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index c99832e7bf3f..eca3f2d633db 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -133,6 +133,7 @@ extern int fix_aperture;
 extern int force_iommu;
 
 extern int reboot_force;
+extern int notsc_setup(char *);
 
 extern void smp_local_timer_interrupt(struct pt_regs * regs);
 

From 6574ffd74b03e35026f428a2c820e6ddf45d426c Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Thu, 16 Feb 2006 23:42:10 +0100
Subject: [PATCH 259/509] [PATCH] x86_64: Resolve the RIP of an early exception
 using kallsyms

But do it after everything else to risk less from recursive
crashes.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/head.S | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/x86_64/kernel/head.S b/arch/x86_64/kernel/head.S
index 692c737feddb..02fc7fa0ea28 100644
--- a/arch/x86_64/kernel/head.S
+++ b/arch/x86_64/kernel/head.S
@@ -213,6 +213,11 @@ ENTRY(early_idt_handler)
 	cmpl $2,early_recursion_flag(%rip)
 	jz  1f
 	call dump_stack
+#ifdef CONFIG_KALLSYMS	
+	leaq early_idt_ripmsg(%rip),%rdi
+	movq 8(%rsp),%rsi	# get rip again
+	call __print_symbol
+#endif
 1:	hlt
 	jmp 1b
 early_recursion_flag:
@@ -220,6 +225,8 @@ early_recursion_flag:
 
 early_idt_msg:
 	.asciz "PANIC: early exception rip %lx error %lx cr2 %lx\n"
+early_idt_ripmsg:
+	.asciz "RIP %s\n"
 
 .code32
 ENTRY(no_long_mode)

From fdb9df942437c6c5d1a6928d5fff824466c3af67 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Thu, 16 Feb 2006 23:42:13 +0100
Subject: [PATCH 260/509] [PATCH] x86_64: Relax SRAT covers all memory check a
 bit

Code was refusing good SRATs because about 12K got lost somewhere.
Allow less than 1MB of difference before rejecting it.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/mm/srat.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index cd25300726fc..809dc70675f7 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -228,7 +228,8 @@ static int nodes_cover_memory(void)
 	}
 
 	e820ram = end_pfn - e820_hole_size(0, end_pfn);
-	if (pxmram < e820ram) {
+	/* We seem to lose 3 pages somewhere. Allow a bit of slack. */
+	if ((long)(e820ram - pxmram) >= 1*1024*1024) {
 		printk(KERN_ERR
 	"SRAT: PXMs only cover %luMB of your %luMB e820 RAM. Not used.\n",
 			(pxmram << PAGE_SHIFT) >> 20,

From 2aed711a399cbc4a9bf239c13f05a8a8b460f215 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Thu, 16 Feb 2006 23:42:16 +0100
Subject: [PATCH 261/509] [PATCH] x86_64: Always pass full number of nodes to
 NUMA hash computation

Previously the numa hash code would be confused by holes in the node space
and stop early. This is the first part of the fix for the non boot issue
with empty nodes on Opterons.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/mm/k8topology.c | 2 +-
 arch/x86_64/mm/srat.c       | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/mm/k8topology.c b/arch/x86_64/mm/k8topology.c
index a5663e0bb01c..dd60e71fdba6 100644
--- a/arch/x86_64/mm/k8topology.c
+++ b/arch/x86_64/mm/k8topology.c
@@ -155,7 +155,7 @@ int __init k8_scan_nodes(unsigned long start, unsigned long end)
 	if (!found)
 		return -1; 
 
-	memnode_shift = compute_hash_shift(nodes, numnodes);
+	memnode_shift = compute_hash_shift(nodes, 8);
 	if (memnode_shift < 0) { 
 		printk(KERN_ERR "No NUMA node hash function found. Contact maintainer\n"); 
 		return -1; 
diff --git a/arch/x86_64/mm/srat.c b/arch/x86_64/mm/srat.c
index 809dc70675f7..482c25767369 100644
--- a/arch/x86_64/mm/srat.c
+++ b/arch/x86_64/mm/srat.c
@@ -271,7 +271,7 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end)
 		return -1;
 	}
 
-	memnode_shift = compute_hash_shift(nodes, nodes_weight(nodes_parsed));
+	memnode_shift = compute_hash_shift(nodes, MAX_NUMNODES);
 	if (memnode_shift < 0) {
 		printk(KERN_ERR
 		     "SRAT: No NUMA node hash function found. Contact maintainer\n");

From dd942ae331425812930cd01766178b7e28e65f2d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Fri, 17 Feb 2006 01:39:16 +0100
Subject: [PATCH 262/509] [PATCH] Handle all and empty zones when setting up
 custom zonelists for mbind

The memory allocator doesn't like empty zones (which have an
uninitialized freelist), so a x86-64 system with a node fully
in GFP_DMA32 only would crash on mbind.

Fix that up by putting all possible zones as fallback into the zonelist
and skipping the empty ones.

In fact the code always enough allocated space for all zones,
but only used it for the highest. This change just uses all the
memory that was allocated before.

This should work fine for now, but whoever implements node hot removal
needs to fix this somewhere else too (or make sure zone datastructures
by itself never go away, only their memory)

Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/mempolicy.c | 18 ++++++++++++++----
 1 file changed, 14 insertions(+), 4 deletions(-)

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 3bd7fb7e4b75..323fdcf128c4 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -132,19 +132,29 @@ static int mpol_check_policy(int mode, nodemask_t *nodes)
 	}
 	return nodes_subset(*nodes, node_online_map) ? 0 : -EINVAL;
 }
+
 /* Generate a custom zonelist for the BIND policy. */
 static struct zonelist *bind_zonelist(nodemask_t *nodes)
 {
 	struct zonelist *zl;
-	int num, max, nd;
+	int num, max, nd, k;
 
 	max = 1 + MAX_NR_ZONES * nodes_weight(*nodes);
-	zl = kmalloc(sizeof(void *) * max, GFP_KERNEL);
+	zl = kmalloc(sizeof(struct zone *) * max, GFP_KERNEL);
 	if (!zl)
 		return NULL;
 	num = 0;
-	for_each_node_mask(nd, *nodes)
-		zl->zones[num++] = &NODE_DATA(nd)->node_zones[policy_zone];
+	/* First put in the highest zones from all nodes, then all the next 
+	   lower zones etc. Avoid empty zones because the memory allocator
+	   doesn't like them. If you implement node hot removal you
+	   have to fix that. */
+	for (k = policy_zone; k >= 0; k--) { 
+		for_each_node_mask(nd, *nodes) { 
+			struct zone *z = &NODE_DATA(nd)->node_zones[k];
+			if (z->present_pages > 0) 
+				zl->zones[num++] = z;
+		}
+	}
 	zl->zones[num] = NULL;
 	return zl;
 }

From 726c14bf499e91e7ede4f1728830aba05c675061 Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@samba.org>
Date: Fri, 17 Feb 2006 10:30:23 +1100
Subject: [PATCH 263/509] [PATCH] Provide an interface for getting the current
 tick length

This provides an interface for arch code to find out how many
nanoseconds are going to be added on to xtime by the next call to
do_timer.  The value returned is a fixed-point number in 52.12 format
in nanoseconds.  The reason for this format is that it gives the
full precision that the timekeeping code is using internally.

The motivation for this is to fix a problem that has arisen on 32-bit
powerpc in that the value returned by do_gettimeofday drifts apart
from xtime if NTP is being used.  PowerPC is now using a lockless
do_gettimeofday based on reading the timebase register and performing
some simple arithmetic.  (This method of getting the time is also
exported to userspace via the VDSO.)  However, the factor and offset
it uses were calculated based on the nominal tick length and weren't
being adjusted when NTP varied the tick length.

Note that 64-bit powerpc has had the lockless do_gettimeofday for a
long time now.  It also had an extremely hairy routine that got called
from the 32-bit compat routine for adjtimex, which adjusted the
factor and offset according to what it thought the timekeeping code
was going to do.  Not only was this only called if a 32-bit task did
adjtimex (i.e. not if a 64-bit task did adjtimex), it was also
duplicating computations from kernel/timer.c and it wasn't clear that
it was (still) correct.

The simple solution is to ask the timekeeping code how long the
current jiffy will be on each timer interrupt, after calling
do_timer.  If this jiffy will be a different length from the last one,
we then need to compute new values for the factor and offset used in
the lockless do_gettimeofday.  In this way we can keep xtime and
do_gettimeofday in sync, even when NTP is varying the tick length.

Note that when adjtimex varies the tick length, it almost always
introduces the variation from the next tick on.  The only case I could
see where adjtimex would vary the length of the current tick is when
an old-style adjtime adjustment is being cancelled.  (It's not clear
to me why the adjustment has to be cancelled immediately rather than
from the next tick on.)  Thus I don't see any real need for a hook in
adjtimex; the rare case of an old-style adjustment being cancelled can
be fixed up at the next tick.

Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: john stultz <johnstul@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/timex.h |  3 +++
 kernel/timer.c        | 39 ++++++++++++++++++++++++++++++++++-----
 2 files changed, 37 insertions(+), 5 deletions(-)

diff --git a/include/linux/timex.h b/include/linux/timex.h
index 04a4a8cb4ed3..b7ca1204e42a 100644
--- a/include/linux/timex.h
+++ b/include/linux/timex.h
@@ -345,6 +345,9 @@ time_interpolator_reset(void)
 
 #endif /* !CONFIG_TIME_INTERPOLATION */
 
+/* Returns how long ticks are at present, in ns / 2^(SHIFT_SCALE-10). */
+extern u64 current_tick_length(void);
+
 #endif /* KERNEL */
 
 #endif /* LINUX_TIMEX_H */
diff --git a/kernel/timer.c b/kernel/timer.c
index b9dad3994676..fe3a9a9f8328 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -717,12 +717,16 @@ static void second_overflow(void)
 #endif
 }
 
-/* in the NTP reference this is called "hardclock()" */
-static void update_wall_time_one_tick(void)
+/*
+ * Returns how many microseconds we need to add to xtime this tick
+ * in doing an adjustment requested with adjtime.
+ */
+static long adjtime_adjustment(void)
 {
-	long time_adjust_step, delta_nsec;
+	long time_adjust_step;
 
-	if ((time_adjust_step = time_adjust) != 0 ) {
+	time_adjust_step = time_adjust;
+	if (time_adjust_step) {
 		/*
 		 * We are doing an adjtime thing.  Prepare time_adjust_step to
 		 * be within bounds.  Note that a positive time_adjust means we
@@ -733,10 +737,19 @@ static void update_wall_time_one_tick(void)
 		 */
 		time_adjust_step = min(time_adjust_step, (long)tickadj);
 		time_adjust_step = max(time_adjust_step, (long)-tickadj);
+	}
+	return time_adjust_step;
+}
 
+/* in the NTP reference this is called "hardclock()" */
+static void update_wall_time_one_tick(void)
+{
+	long time_adjust_step, delta_nsec;
+
+	time_adjust_step = adjtime_adjustment();
+	if (time_adjust_step)
 		/* Reduce by this step the amount of time left  */
 		time_adjust -= time_adjust_step;
-	}
 	delta_nsec = tick_nsec + time_adjust_step * 1000;
 	/*
 	 * Advance the phase, once it gets to one microsecond, then
@@ -758,6 +771,22 @@ static void update_wall_time_one_tick(void)
 	}
 }
 
+/*
+ * Return how long ticks are at the moment, that is, how much time
+ * update_wall_time_one_tick will add to xtime next time we call it
+ * (assuming no calls to do_adjtimex in the meantime).
+ * The return value is in fixed-point nanoseconds with SHIFT_SCALE-10
+ * bits to the right of the binary point.
+ * This function has no side-effects.
+ */
+u64 current_tick_length(void)
+{
+	long delta_nsec;
+
+	delta_nsec = tick_nsec + adjtime_adjustment() * 1000;
+	return ((u64) delta_nsec << (SHIFT_SCALE - 10)) + time_adj;
+}
+
 /*
  * Using a loop looks inefficient, but "ticks" is
  * usually just one (we shouldn't be losing ticks,

From d30864392823d5f38002fa32950689e651ee11da Mon Sep 17 00:00:00 2001
From: Joshua Kinard <kumba@gentoo.org>
Date: Fri, 17 Feb 2006 03:52:25 +0000
Subject: [PATCH 264/509] [PATCH] Fix SGI O2 compile error in
 drivers/video/gbefb.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A sysfs function call uses the wrong parameter, and thus breaks a build on
SGI O2.

  CC      drivers/video/gbefb.o
drivers/video/gbefb.c: In function ‘gbefb_remove’:
drivers/video/gbefb.c:1246: error: ‘dev’ undeclared (first use in this function)
drivers/video/gbefb.c:1246: error: (Each undeclared identifier is reported only once
drivers/video/gbefb.c:1246: error: for each function it appears in.)
make[2]: *** [drivers/video/gbefb.o] Error 1

Signed-off-by: Joshua Kinard <kumba@gentoo.org>
Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/gbefb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/gbefb.c b/drivers/video/gbefb.c
index 38d22729b129..c9a7cdf6d543 100644
--- a/drivers/video/gbefb.c
+++ b/drivers/video/gbefb.c
@@ -1243,7 +1243,7 @@ static int __devexit gbefb_remove(struct platform_device* p_dev)
 			  (void *)gbe_tiles.cpu, gbe_tiles.dma);
 	release_mem_region(GBE_BASE, sizeof(struct sgi_gbe));
 	iounmap(gbe);
-	gbefb_remove_sysfs(dev);
+	gbefb_remove_sysfs(&p_dev->dev);
 	framebuffer_release(info);
 
 	return 0;

From cfe91f9ce297e23e6fbdf61c02bdd8ab9af7c8a8 Mon Sep 17 00:00:00 2001
From: Chuck Ebbert <76306.1226@compuserve.com>
Date: Fri, 17 Feb 2006 03:16:55 -0500
Subject: [PATCH 265/509] [PATCH] i386: fix singlestepping though a syscall

Do not mask TIF_SINGLESTEP bit in _TIF_WORK_MASK. Masking this stopped
do_notify_resume() from being called when it should have been.

Signed-off-by: Chuck Ebbert <76306.1226@compuserve.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-i386/thread_info.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/asm-i386/thread_info.h b/include/asm-i386/thread_info.h
index e20e99551d71..1f7d48c9ba3f 100644
--- a/include/asm-i386/thread_info.h
+++ b/include/asm-i386/thread_info.h
@@ -158,8 +158,8 @@ register unsigned long current_stack_pointer asm("esp") __attribute_used__;
 
 /* work to do on interrupt/exception return */
 #define _TIF_WORK_MASK \
-  (0x0000FFFF & ~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP|\
-		  _TIF_SECCOMP|_TIF_SYSCALL_EMU))
+  (0x0000FFFF & ~(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | \
+		  _TIF_SECCOMP | _TIF_SYSCALL_EMU))
 /* work to do on any return to u-space */
 #define _TIF_ALLWORK_MASK	(0x0000FFFF & ~_TIF_SECCOMP)
 

From aca0b510cdbf81d52e15014a720be2b8dfd26aea Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Thu, 16 Feb 2006 17:44:54 -0800
Subject: [PATCH 266/509] [PATCH] Wavelan_cs bitfield fixes

	Some bitfields were incorrectly initialised in wavelan_cs,
causing some compiler warning. Also killed a error message that should
not be there...

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/net/wireless/wavelan_cs.c | 16 ++++------------
 1 file changed, 4 insertions(+), 12 deletions(-)

diff --git a/drivers/net/wireless/wavelan_cs.c b/drivers/net/wireless/wavelan_cs.c
index cf373625fc70..98122f3a4bc2 100644
--- a/drivers/net/wireless/wavelan_cs.c
+++ b/drivers/net/wireless/wavelan_cs.c
@@ -950,16 +950,8 @@ wv_82593_cmd(struct net_device *	dev,
 static inline int
 wv_diag(struct net_device *	dev)
 {
-  int		ret = FALSE;
-
-  if(wv_82593_cmd(dev, "wv_diag(): diagnose",
-		  OP0_DIAGNOSE, SR0_DIAGNOSE_PASSED))
-    ret = TRUE;
-
-#ifdef DEBUG_CONFIG_ERRORS
-  printk(KERN_INFO "wavelan_cs: i82593 Self Test failed!\n");
-#endif
-  return(ret);
+  return(wv_82593_cmd(dev, "wv_diag(): diagnose",
+		      OP0_DIAGNOSE, SR0_DIAGNOSE_PASSED));
 } /* wv_diag */
 
 /*------------------------------------------------------------------*/
@@ -3604,8 +3596,8 @@ wv_82593_config(struct net_device *	dev)
   cfblk.lin_prio = 0;   	/* conform to 802.3 backoff algoritm */
   cfblk.exp_prio = 5;	        /* conform to 802.3 backoff algoritm */
   cfblk.bof_met = 1;	        /* conform to 802.3 backoff algoritm */
-  cfblk.ifrm_spc = 0x20;	/* 32 bit times interframe spacing */
-  cfblk.slottim_low = 0x20;	/* 32 bit times slot time */
+  cfblk.ifrm_spc = 0x20 >> 4;	/* 32 bit times interframe spacing */
+  cfblk.slottim_low = 0x20 >> 5;	/* 32 bit times slot time */
   cfblk.slottim_hi = 0x0;
   cfblk.max_retr = 15;
   cfblk.prmisc = ((lp->promiscuous) ? TRUE: FALSE);	/* Promiscuous mode */

From 56c8f7e290a1b3249c6a5b464fada8a167f2e5ff Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Mon, 13 Feb 2006 15:49:55 -0800
Subject: [PATCH 267/509] [PATCH] sk98lin: no d-link support (kconfig)

The sk98lin driver was changed a while ago to remove support for the
D-Link 530T card because that hardware has no working VPD data.  The help
text for Kconfig was not updated.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/net/Kconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index 47c72a63dfe1..ae8037acf489 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2082,7 +2082,6 @@ config SK98LIN
 	    - Allied Telesyn AT-2971SX Gigabit Ethernet Adapter
 	    - Allied Telesyn AT-2971T Gigabit Ethernet Adapter
 	    - Belkin Gigabit Desktop Card 10/100/1000Base-T Adapter, Copper RJ-45
-	    - DGE-530T Gigabit Ethernet Adapter
 	    - EG1032 v2 Instant Gigabit Network Adapter
 	    - EG1064 v2 Instant Gigabit Network Adapter
 	    - Marvell 88E8001 Gigabit LOM Ethernet Adapter (Abit)

From 7a160c735a264de400cbbf4cf0fb3250007ae0c6 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Mon, 13 Feb 2006 15:48:08 -0800
Subject: [PATCH 268/509] [PATCH] skge: no longer experimental

Take the experimental dependency of skge driver, it is as stable as the
others.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/net/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig
index ae8037acf489..e45a8f959719 100644
--- a/drivers/net/Kconfig
+++ b/drivers/net/Kconfig
@@ -2020,8 +2020,8 @@ config SIS190
 	  will be called sis190.  This is recommended.
 
 config SKGE
-	tristate "New SysKonnect GigaEthernet support (EXPERIMENTAL)"
-	depends on PCI && EXPERIMENTAL
+	tristate "New SysKonnect GigaEthernet support"
+	depends on PCI
 	select CRC32
 	---help---
 	  This driver support the Marvell Yukon or SysKonnect SK-98xx/SK-95xx

From 564f9abb34c936e5d0c7682129042dad14dbbb95 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Mon, 13 Feb 2006 15:46:48 -0800
Subject: [PATCH 269/509] [PATCH] skge: speed setting

This is a clone of John Linville's fixed for speed setting on sky2 driver.
The skge driver has the same code (and bug). It would not allow manually forcing
100 and 10 mbit.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/net/skge.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/skge.c b/drivers/net/skge.c
index bf55a4cfb3d2..67fb19b8fde9 100644
--- a/drivers/net/skge.c
+++ b/drivers/net/skge.c
@@ -1697,6 +1697,7 @@ static void yukon_mac_init(struct skge_hw *hw, int port)
 	skge_write32(hw, SK_REG(port, GPHY_CTRL), reg | GPC_RST_SET);
 	skge_write32(hw, SK_REG(port, GPHY_CTRL), reg | GPC_RST_CLR);
 	skge_write32(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_ON | GMC_RST_CLR);
+
 	if (skge->autoneg == AUTONEG_DISABLE) {
 		reg = GM_GPCR_AU_ALL_DIS;
 		gma_write16(hw, port, GM_GP_CTRL,
@@ -1704,16 +1705,23 @@ static void yukon_mac_init(struct skge_hw *hw, int port)
 
 		switch (skge->speed) {
 		case SPEED_1000:
+			reg &= ~GM_GPCR_SPEED_100;
 			reg |= GM_GPCR_SPEED_1000;
-			/* fallthru */
+			break;
 		case SPEED_100:
+			reg &= ~GM_GPCR_SPEED_1000;
 			reg |= GM_GPCR_SPEED_100;
+			break;
+		case SPEED_10:
+			reg &= ~(GM_GPCR_SPEED_1000 | GM_GPCR_SPEED_100);
+			break;
 		}
 
 		if (skge->duplex == DUPLEX_FULL)
 			reg |= GM_GPCR_DUP_FULL;
 	} else
 		reg = GM_GPCR_SPEED_1000 | GM_GPCR_SPEED_100 | GM_GPCR_DUP_FULL;
+
 	switch (skge->flow_control) {
 	case FLOW_MODE_NONE:
 		skge_write32(hw, SK_REG(port, GMAC_CTRL), GMC_PAUSE_OFF);

From 6f4c56b2ae10b680be518cc99f5308fc59236db2 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Fri, 10 Feb 2006 15:58:59 -0800
Subject: [PATCH 270/509] [PATCH] sky2: speed setting fix

Users report problems w/ auto-negotiation disabled and the link set
to 100/Half or 10/Half.  Problems range from poor performance to no
link at all.

The current sky2 code does not set things properly on link up if
autonegotiation is disabled.  Plus it does not contemplate a 10Mbit
setting at all.  This patch corrects that.

Signed-off-by: John W. Linville <linville@tuxdriver.com>
Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/net/sky2.c | 31 ++++++++++++++++++++++++++++++-
 1 file changed, 30 insertions(+), 1 deletion(-)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index cae2edf23004..bfeba5b9cd7a 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -520,10 +520,16 @@ static void sky2_mac_init(struct sky2_hw *hw, unsigned port)
 
 		switch (sky2->speed) {
 		case SPEED_1000:
+			reg &= ~GM_GPCR_SPEED_100;
 			reg |= GM_GPCR_SPEED_1000;
-			/* fallthru */
+			break;
 		case SPEED_100:
+			reg &= ~GM_GPCR_SPEED_1000;
 			reg |= GM_GPCR_SPEED_100;
+			break;
+		case SPEED_10:
+			reg &= ~(GM_GPCR_SPEED_1000 | GM_GPCR_SPEED_100);
+			break;
 		}
 
 		if (sky2->duplex == DUPLEX_FULL)
@@ -1446,6 +1452,29 @@ static void sky2_link_up(struct sky2_port *sky2)
 	sky2_write8(hw, SK_REG(port, GMAC_IRQ_MSK), GMAC_DEF_MSK);
 
 	reg = gma_read16(hw, port, GM_GP_CTRL);
+	if (sky2->autoneg == AUTONEG_DISABLE) {
+		reg |= GM_GPCR_AU_ALL_DIS;
+
+		/* Is write/read necessary?  Copied from sky2_mac_init */
+		gma_write16(hw, port, GM_GP_CTRL, reg);
+		gma_read16(hw, port, GM_GP_CTRL);
+
+		switch (sky2->speed) {
+		case SPEED_1000:
+			reg &= ~GM_GPCR_SPEED_100;
+			reg |= GM_GPCR_SPEED_1000;
+			break;
+		case SPEED_100:
+			reg &= ~GM_GPCR_SPEED_1000;
+			reg |= GM_GPCR_SPEED_100;
+			break;
+		case SPEED_10:
+			reg &= ~(GM_GPCR_SPEED_1000 | GM_GPCR_SPEED_100);
+			break;
+		}
+	} else
+		reg &= ~GM_GPCR_AU_ALL_DIS;
+
 	if (sky2->duplex == DUPLEX_FULL || sky2->autoneg == AUTONEG_ENABLE)
 		reg |= GM_GPCR_DUP_FULL;
 

From ca5b0ec8ae9f11c85d1f27b19f182a054303f324 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Fri, 10 Feb 2006 02:00:43 -0800
Subject: [PATCH 271/509] [PATCH] smctr warning fix

drivers/net/tokenring/smctr.c: In function `smctr_load_firmware':
drivers/net/tokenring/smctr.c:2981: warning: assignment discards qualifiers from pointer target type

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/net/tokenring/smctr.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/tokenring/smctr.h b/drivers/net/tokenring/smctr.h
index b306c7e4c793..88dfa2e01d6e 100644
--- a/drivers/net/tokenring/smctr.h
+++ b/drivers/net/tokenring/smctr.h
@@ -1042,7 +1042,7 @@ typedef struct net_local {
         __u16            functional_address[2];
         __u16            bitwise_group_address[2];
 
-	__u8            *ptr_ucode;
+	const __u8       *ptr_ucode;
 
 	__u8		cleanup;
 

From 0d613a27cc753bfacd20e6eaa2183bb7fef4c76e Mon Sep 17 00:00:00 2001
From: Frank Pavlic <fpavlic@de.ibm.com>
Date: Tue, 7 Feb 2006 17:04:36 +0100
Subject: [PATCH 272/509] [PATCH] s390: lcs performance enhancements

[patch 1/2] s390: lcs performance enhancements

From: Klaus Wacker <kdwacker@de.ibm.com>
	- When flood pinging (with large packet size) an LCS device,
	  about 90 % of all packets are dropped by driver.
	- increased number of lcs IO buffers to 32.
	- use netif_stop_queue/netif_wake_queue in lcs_start_xmit routine
	- don't lock the whole xmit routine but just the piece of code where
	  tx_buffer is touched.

Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com>

diffstat:
 lcs.c |   31 +++++++++++++++++--------------
 lcs.h |    2 +-
 2 files changed, 18 insertions(+), 15 deletions(-)
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/s390/net/lcs.c | 31 +++++++++++++++++--------------
 drivers/s390/net/lcs.h |  2 +-
 2 files changed, 18 insertions(+), 15 deletions(-)

diff --git a/drivers/s390/net/lcs.c b/drivers/s390/net/lcs.c
index 6229ba4995ad..9cf88d7201d3 100644
--- a/drivers/s390/net/lcs.c
+++ b/drivers/s390/net/lcs.c
@@ -98,9 +98,9 @@ lcs_register_debug_facility(void)
 		return -ENOMEM;
 	}
 	debug_register_view(lcs_dbf_setup, &debug_hex_ascii_view);
-	debug_set_level(lcs_dbf_setup, 4);
+	debug_set_level(lcs_dbf_setup, 2);
 	debug_register_view(lcs_dbf_trace, &debug_hex_ascii_view);
-	debug_set_level(lcs_dbf_trace, 4);
+	debug_set_level(lcs_dbf_trace, 2);
 	return 0;
 }
 
@@ -1292,9 +1292,8 @@ lcs_set_multicast_list(struct net_device *dev)
         LCS_DBF_TEXT(4, trace, "setmulti");
         card = (struct lcs_card *) dev->priv;
 
-        if (!lcs_set_thread_start_bit(card, LCS_SET_MC_THREAD)) {
+        if (!lcs_set_thread_start_bit(card, LCS_SET_MC_THREAD)) 
 		schedule_work(&card->kernel_thread_starter);
-	}
 }
 
 #endif /* CONFIG_IP_MULTICAST */
@@ -1459,6 +1458,8 @@ lcs_txbuffer_cb(struct lcs_channel *channel, struct lcs_buffer *buffer)
 	lcs_release_buffer(channel, buffer);
 	card = (struct lcs_card *)
 		((char *) channel - offsetof(struct lcs_card, write));
+	if (netif_queue_stopped(card->dev))
+		netif_wake_queue(card->dev);
 	spin_lock(&card->lock);
 	card->tx_emitted--;
 	if (card->tx_emitted <= 0 && card->tx_buffer != NULL)
@@ -1478,6 +1479,7 @@ __lcs_start_xmit(struct lcs_card *card, struct sk_buff *skb,
 		 struct net_device *dev)
 {
 	struct lcs_header *header;
+	int rc = 0;
 
 	LCS_DBF_TEXT(5, trace, "hardxmit");
 	if (skb == NULL) {
@@ -1492,10 +1494,8 @@ __lcs_start_xmit(struct lcs_card *card, struct sk_buff *skb,
 		card->stats.tx_carrier_errors++;
 		return 0;
 	}
-	if (netif_queue_stopped(dev) ) {
-		card->stats.tx_dropped++;
-		return -EBUSY;
-	}
+	netif_stop_queue(card->dev);
+	spin_lock(&card->lock);
 	if (card->tx_buffer != NULL &&
 	    card->tx_buffer->count + sizeof(struct lcs_header) +
 	    skb->len + sizeof(u16) > LCS_IOBUFFERSIZE)
@@ -1506,7 +1506,8 @@ __lcs_start_xmit(struct lcs_card *card, struct sk_buff *skb,
 		card->tx_buffer = lcs_get_buffer(&card->write);
 		if (card->tx_buffer == NULL) {
 			card->stats.tx_dropped++;
-			return -EBUSY;
+			rc = -EBUSY;
+			goto out;
 		}
 		card->tx_buffer->callback = lcs_txbuffer_cb;
 		card->tx_buffer->count = 0;
@@ -1518,13 +1519,18 @@ __lcs_start_xmit(struct lcs_card *card, struct sk_buff *skb,
 	header->type = card->lan_type;
 	header->slot = card->portno;
 	memcpy(header + 1, skb->data, skb->len);
+	spin_unlock(&card->lock);
 	card->stats.tx_bytes += skb->len;
 	card->stats.tx_packets++;
 	dev_kfree_skb(skb);
-	if (card->tx_emitted <= 0)
+	netif_wake_queue(card->dev);
+	spin_lock(&card->lock);
+	if (card->tx_emitted <= 0 && card->tx_buffer != NULL)
 		/* If this is the first tx buffer emit it immediately. */
 		__lcs_emit_txbuffer(card);
-	return 0;
+out:
+	spin_unlock(&card->lock);
+	return rc;
 }
 
 static int
@@ -1535,9 +1541,7 @@ lcs_start_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	LCS_DBF_TEXT(5, trace, "pktxmit");
 	card = (struct lcs_card *) dev->priv;
-	spin_lock(&card->lock);
 	rc = __lcs_start_xmit(card, skb, dev);
-	spin_unlock(&card->lock);
 	return rc;
 }
 
@@ -2319,7 +2323,6 @@ __init lcs_init_module(void)
 		PRINT_ERR("Initialization failed\n");
 		return rc;
 	}
-
 	return 0;
 }
 
diff --git a/drivers/s390/net/lcs.h b/drivers/s390/net/lcs.h
index 08e60ad43916..2fad5e40c2e4 100644
--- a/drivers/s390/net/lcs.h
+++ b/drivers/s390/net/lcs.h
@@ -95,7 +95,7 @@ do {                                       \
  */
 #define LCS_ILLEGAL_OFFSET		0xffff
 #define LCS_IOBUFFERSIZE		0x5000
-#define LCS_NUM_BUFFS			8	/* needs to be power of 2 */
+#define LCS_NUM_BUFFS			32	/* needs to be power of 2 */
 #define LCS_MAC_LENGTH			6
 #define LCS_INVALID_PORT_NO		-1
 #define LCS_LANCMD_TIMEOUT_DEFAULT      5

From 66cc5d5aee1ea427b3aeacdabd006a4195c81eee Mon Sep 17 00:00:00 2001
From: Frank Pavlic <fpavlic@de.ibm.com>
Date: Tue, 7 Feb 2006 17:04:38 +0100
Subject: [PATCH 273/509] [PATCH] s390: some qeth driver fixes

[patch 2/2] s390: some qeth driver fixes

From: Frank Pavlic <fpavlic@de.ibm.com>
	- fixed kernel panic when using EDDP support in Layer 2 mode
	- NULL pointer exception in qeth_set_offline fixed.
	- setting EDDP in Layer 2 mode did not set NETIF_F_(SG/TSO)
	  flags when device became online.
	- use sscanf for parsing and converting IPv4 addresses
	  from string to __u8 values.
	- qeth_string_to_ipaddr6 fixed. in case of double colon
	  the converted IPv6 address out from the string was not correct
	  in previous implementation.

Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com>

diffstat:
 qeth.h      |  112 +++++++++++++++++++++++++-----------------------------------
 qeth_eddp.c |   11 ++++-
 qeth_main.c |   17 +++------
 3 files changed, 63 insertions(+), 77 deletions(-)
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/s390/net/qeth.h      | 108 +++++++++++++++--------------------
 drivers/s390/net/qeth_eddp.c |  11 +++-
 drivers/s390/net/qeth_main.c |  17 +++---
 3 files changed, 61 insertions(+), 75 deletions(-)

diff --git a/drivers/s390/net/qeth.h b/drivers/s390/net/qeth.h
index 9a064d4727ad..4df0fcd7b10b 100644
--- a/drivers/s390/net/qeth.h
+++ b/drivers/s390/net/qeth.h
@@ -1075,16 +1075,6 @@ qeth_get_qdio_q_format(struct qeth_card *card)
 	}
 }
 
-static inline int
-qeth_isdigit(char * buf)
-{
-	while (*buf) {
-		if (!isdigit(*buf++))
-			return 0;
-	}
-	return 1;
-}
-
 static inline int
 qeth_isxdigit(char * buf)
 {
@@ -1104,33 +1094,17 @@ qeth_ipaddr4_to_string(const __u8 *addr, char *buf)
 static inline int
 qeth_string_to_ipaddr4(const char *buf, __u8 *addr)
 {
-	const char *start, *end;
-	char abuf[4];
-	char *tmp;
-	int len;
-	int i;
+	int count = 0, rc = 0;
+	int in[4];
 
-	start = buf;
-	for (i = 0; i < 4; i++) {
-		if (i == 3) {
-			end = strchr(start,0xa);
-			if (end)
-				len = end - start;
-			else		
-				len = strlen(start);
-		}
-		else {
-			end = strchr(start, '.');
-			len = end - start;
-		}
-		if ((len <= 0) || (len > 3))
+	rc = sscanf(buf, "%d.%d.%d.%d%n", 
+		    &in[0], &in[1], &in[2], &in[3], &count);
+	if (rc != 4  || count) 
+		return -EINVAL;
+	for (count = 0; count < 4; count++) {
+		if (in[count] > 255)
 			return -EINVAL;
-		memset(abuf, 0, 4);
-		strncpy(abuf, start, len);
-		if (!qeth_isdigit(abuf))
-			return -EINVAL;
-		addr[i] = simple_strtoul(abuf, &tmp, 10);
-		start = end + 1;
+		addr[count] = in[count];
 	}
 	return 0;
 }
@@ -1149,36 +1123,44 @@ qeth_ipaddr6_to_string(const __u8 *addr, char *buf)
 static inline int
 qeth_string_to_ipaddr6(const char *buf, __u8 *addr)
 {
-	const char *start, *end;
-	u16 *tmp_addr;
-	char abuf[5];
-	char *tmp;
-	int len;
-	int i;
+	char *end, *start;
+	__u16 *in;
+        char num[5];
+        int num2, cnt, out, found, save_cnt;
+        unsigned short in_tmp[8] = {0, };
 
-	tmp_addr = (u16 *)addr;
-	start = buf;
-	for (i = 0; i < 8; i++) {
-		if (i == 7) {
-			end = strchr(start,0xa);
-			if (end)
-				len = end - start;
-			else
-				len = strlen(start);
+	cnt = out = found = save_cnt = num2 = 0;
+        end = start = (char *) buf;
+	in = (__u16 *) addr;	
+	memset(in, 0, 16);
+        while (end) {
+                end = strchr(end,':');
+                if (end == NULL) {
+                        end = (char *)buf + (strlen(buf));
+                        out = 1;
+                }
+                if ((end - start)) { 
+                        memset(num, 0, 5);
+                        memcpy(num, start, end - start);
+			if (!qeth_isxdigit(num))
+				return -EINVAL;
+                        sscanf(start, "%x", &num2);
+                        if (found)
+                                in_tmp[save_cnt++] = num2;
+                        else
+                                in[cnt++] = num2;
+                        if (out)
+                                break;
+                } else {
+			if (found)
+				return -EINVAL;
+                        found = 1;
 		}
-		else {
-			end = strchr(start, ':');
-			len = end - start;
-		}
-		if ((len <= 0) || (len > 4))
-			return -EINVAL;
-		memset(abuf, 0, 5);
-		strncpy(abuf, start, len);
-		if (!qeth_isxdigit(abuf))
-			return -EINVAL;
-		tmp_addr[i] = simple_strtoul(abuf, &tmp, 16);
-		start = end + 1;
-	}
+		start = ++end;
+        }
+        cnt = 7;
+	while (save_cnt)
+                in[cnt--] = in_tmp[--save_cnt];
 	return 0;
 }
 
diff --git a/drivers/s390/net/qeth_eddp.c b/drivers/s390/net/qeth_eddp.c
index b02313127780..82cb4af2f0e7 100644
--- a/drivers/s390/net/qeth_eddp.c
+++ b/drivers/s390/net/qeth_eddp.c
@@ -59,8 +59,7 @@ qeth_eddp_free_context(struct qeth_eddp_context *ctx)
 	for (i = 0; i < ctx->num_pages; ++i)
 		free_page((unsigned long)ctx->pages[i]);
 	kfree(ctx->pages);
-	if (ctx->elements != NULL)
-		kfree(ctx->elements);
+	kfree(ctx->elements);
 	kfree(ctx);
 }
 
@@ -413,6 +412,13 @@ __qeth_eddp_fill_context_tcp(struct qeth_eddp_context *ctx,
 	
 	QETH_DBF_TEXT(trace, 5, "eddpftcp");
 	eddp->skb_offset = sizeof(struct qeth_hdr) + eddp->nhl + eddp->thl;
+       if (eddp->qh.hdr.l2.id == QETH_HEADER_TYPE_LAYER2) {
+               eddp->skb_offset += sizeof(struct ethhdr);
+#ifdef CONFIG_QETH_VLAN
+               if (eddp->mac.h_proto == __constant_htons(ETH_P_8021Q))
+                       eddp->skb_offset += VLAN_HLEN;
+#endif /* CONFIG_QETH_VLAN */
+       }
 	tcph = eddp->skb->h.th;
 	while (eddp->skb_offset < eddp->skb->len) {
 		data_len = min((int)skb_shinfo(eddp->skb)->tso_size,
@@ -483,6 +489,7 @@ qeth_eddp_fill_context_tcp(struct qeth_eddp_context *ctx,
 		return -ENOMEM;
 	}
 	if (qhdr->hdr.l2.id == QETH_HEADER_TYPE_LAYER2) {
+		skb->mac.raw = (skb->data) + sizeof(struct qeth_hdr);
 		memcpy(&eddp->mac, eth_hdr(skb), ETH_HLEN);
 #ifdef CONFIG_QETH_VLAN
 		if (eddp->mac.h_proto == __constant_htons(ETH_P_8021Q)) {
diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c
index 410abeada6c4..dba7f7f02e79 100644
--- a/drivers/s390/net/qeth_main.c
+++ b/drivers/s390/net/qeth_main.c
@@ -516,7 +516,8 @@ __qeth_set_offline(struct ccwgroup_device *cgdev, int recovery_mode)
 	QETH_DBF_TEXT(setup, 3, "setoffl");
 	QETH_DBF_HEX(setup, 3, &card, sizeof(void *));
 	
-	netif_carrier_off(card->dev);
+	if (card->dev && netif_carrier_ok(card->dev))
+		netif_carrier_off(card->dev);
 	recover_flag = card->state;
 	if (qeth_stop_card(card, recovery_mode) == -ERESTARTSYS){
 		PRINT_WARN("Stopping card %s interrupted by user!\n",
@@ -1679,6 +1680,7 @@ qeth_cmd_timeout(unsigned long data)
 	spin_unlock_irqrestore(&reply->card->lock, flags);
 }
 
+
 static struct qeth_ipa_cmd *
 qeth_check_ipa_data(struct qeth_card *card, struct qeth_cmd_buffer *iob)
 {
@@ -1699,7 +1701,8 @@ qeth_check_ipa_data(struct qeth_card *card, struct qeth_cmd_buffer *iob)
 					   QETH_CARD_IFNAME(card),
 					   card->info.chpid);
 				card->lan_online = 0;
-				netif_carrier_off(card->dev);
+				if (card->dev && netif_carrier_ok(card->dev))
+					netif_carrier_off(card->dev);
 				return NULL;
 			case IPA_CMD_STARTLAN:
 				PRINT_INFO("Link reestablished on %s "
@@ -5562,7 +5565,7 @@ qeth_set_multicast_list(struct net_device *dev)
 	if (card->info.type == QETH_CARD_TYPE_OSN)
 		return ;
 	 
-	QETH_DBF_TEXT(trace,3,"setmulti");
+	QETH_DBF_TEXT(trace, 3, "setmulti");
 	qeth_delete_mc_addresses(card);
 	if (card->options.layer2) {
 		qeth_layer2_add_multicast(card);
@@ -5579,7 +5582,6 @@ out:
 		return;
 	if (qeth_set_thread_start_bit(card, QETH_SET_PROMISC_MODE_THREAD)==0)
 		schedule_work(&card->kernel_thread_starter);
-
 }
 
 static int
@@ -7452,6 +7454,7 @@ qeth_softsetup_card(struct qeth_card *card)
 		card->lan_online = 1;
 	if (card->info.type==QETH_CARD_TYPE_OSN)
 		goto out;
+	qeth_set_large_send(card, card->options.large_send);
 	if (card->options.layer2) {
 		card->dev->features |=
 			NETIF_F_HW_VLAN_FILTER |
@@ -7468,12 +7471,6 @@ qeth_softsetup_card(struct qeth_card *card)
 #endif
 		goto out;
 	}
-	if ((card->options.large_send == QETH_LARGE_SEND_EDDP) ||
-	    (card->options.large_send == QETH_LARGE_SEND_TSO))
-		card->dev->features |= NETIF_F_TSO | NETIF_F_SG;
-	else
-		card->dev->features &= ~(NETIF_F_TSO | NETIF_F_SG);
-
 	if ((rc = qeth_setadapter_parms(card)))
 		QETH_DBF_TEXT_(setup, 2, "2err%d", rc);
 	if ((rc = qeth_start_ipassists(card)))

From f5e2a7b22e7d7dfda8794906d0fddeaaa09bb944 Mon Sep 17 00:00:00 2001
From: Jay Vosburgh <fubar@us.ibm.com>
Date: Tue, 7 Feb 2006 21:17:22 -0800
Subject: [PATCH 274/509] [PATCH] bonding: fix a locking bug in bond_release

bond_release returns EINVAL without releasing the bond lock if the
slave device is not being bonded by the bond.  The following patch
ensures that the lock is released in this case.

Signed-off-by: Stephen J. Bevan <stephen@dino.dnsalias.com>
Acked-by: Jay Vosburgh <fubar@us.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/net/bonding/bond_main.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index e0f51afec778..bcf9f17daf0d 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1581,6 +1581,7 @@ int bond_release(struct net_device *bond_dev, struct net_device *slave_dev)
 		printk(KERN_INFO DRV_NAME
 		       ": %s: %s not enslaved\n",
 		       bond_dev->name, slave_dev->name);
+		write_unlock_bh(&bond->lock);
 		return -EINVAL;
 	}
 

From 4cf808eb443ead42777a0230b73aec0cee7fb298 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@osdl.org>
Date: Fri, 17 Feb 2006 20:38:21 +0100
Subject: [PATCH 275/509] [PATCH] Handle holes in node mask in node fallback
 list setup

Change the find_next_best_node algorithm to correctly skip
over holes in the node online mask. Previously it would not handle
missing nodes correctly and cause crashes at boot.

[Written by Linus, tested by AK]

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/page_alloc.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 62c122528587..208812b25597 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1541,29 +1541,29 @@ static int __initdata node_load[MAX_NUMNODES];
  */
 static int __init find_next_best_node(int node, nodemask_t *used_node_mask)
 {
-	int i, n, val;
+	int n, val;
 	int min_val = INT_MAX;
 	int best_node = -1;
 
-	for_each_online_node(i) {
-		cpumask_t tmp;
+	/* Use the local node if we haven't already */
+	if (!node_isset(node, *used_node_mask)) {
+		node_set(node, *used_node_mask);
+		return node;
+	}
 
-		/* Start from local node */
-		n = (node+i) % num_online_nodes();
+	for_each_online_node(n) {
+		cpumask_t tmp;
 
 		/* Don't want a node to appear more than once */
 		if (node_isset(n, *used_node_mask))
 			continue;
 
-		/* Use the local node if we haven't already */
-		if (!node_isset(node, *used_node_mask)) {
-			best_node = node;
-			break;
-		}
-
 		/* Use the distance array to find the distance */
 		val = node_distance(node, n);
 
+		/* Penalize nodes under us ("prefer the next node") */
+		val += (n < node);
+
 		/* Give preference to headless and unused nodes */
 		tmp = node_to_cpumask(n);
 		if (!cpus_empty(tmp))

From 2ae5b30ff08cee422c7f6388a759f743633c7542 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Wed, 14 Dec 2005 13:10:49 -0700
Subject: [PATCH 276/509] [PATCH] Necessary evil to get sata_vsc to initialize
 with Intel iq3124h hba

* libata does not care about error interrupts, so handle them locally
* the interrupts that are ignored only appear to happen at init time

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/scsi/sata_vsc.c | 30 +++++++++++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

diff --git a/drivers/scsi/sata_vsc.c b/drivers/scsi/sata_vsc.c
index 2e2c3b7acb0c..e484e8db6810 100644
--- a/drivers/scsi/sata_vsc.c
+++ b/drivers/scsi/sata_vsc.c
@@ -81,6 +81,19 @@
 /* Port stride */
 #define VSC_SATA_PORT_OFFSET		0x200
 
+/* Error interrupt status bit offsets */
+#define VSC_SATA_INT_ERROR_E_OFFSET	2
+#define VSC_SATA_INT_ERROR_P_OFFSET	4
+#define VSC_SATA_INT_ERROR_T_OFFSET	5
+#define VSC_SATA_INT_ERROR_M_OFFSET	1
+#define is_vsc_sata_int_err(port_idx, int_status) \
+	 (int_status & ((1 << (VSC_SATA_INT_ERROR_E_OFFSET + (8 * port_idx))) | \
+		        (1 << (VSC_SATA_INT_ERROR_P_OFFSET + (8 * port_idx))) | \
+		        (1 << (VSC_SATA_INT_ERROR_T_OFFSET + (8 * port_idx))) | \
+		        (1 << (VSC_SATA_INT_ERROR_M_OFFSET + (8 * port_idx)))   \
+		       )\
+ 	 )
+
 
 static u32 vsc_sata_scr_read (struct ata_port *ap, unsigned int sc_reg)
 {
@@ -201,13 +214,28 @@ static irqreturn_t vsc_sata_interrupt (int irq, void *dev_instance,
 			struct ata_port *ap;
 
 			ap = host_set->ports[i];
+
+			if (is_vsc_sata_int_err(i, int_status)) {
+				u32 err_status;
+				printk(KERN_DEBUG "%s: ignoring interrupt(s)\n", __FUNCTION__);
+				err_status = ap ? vsc_sata_scr_read(ap, SCR_ERROR) : 0;
+				vsc_sata_scr_write(ap, SCR_ERROR, err_status);
+				handled++;
+			}
+
 			if (ap && !(ap->flags &
 				    (ATA_FLAG_PORT_DISABLED|ATA_FLAG_NOINTR))) {
 				struct ata_queued_cmd *qc;
 
 				qc = ata_qc_from_tag(ap, ap->active_tag);
-				if (qc && (!(qc->tf.ctl & ATA_NIEN)))
+				if (qc && (!(qc->tf.ctl & ATA_NIEN))) {
 					handled += ata_host_intr(ap, qc);
+				} else {
+					printk(KERN_DEBUG "%s: ignoring interrupt(s)\n", __FUNCTION__);
+					ata_chk_status(ap);
+					handled++;
+				}
+
 			}
 		}
 	}

From 0565c26de7b2c069add553106f210b3128b67785 Mon Sep 17 00:00:00 2001
From: Albert Lee <albertcc@tw.ibm.com>
Date: Mon, 13 Feb 2006 18:55:25 +0800
Subject: [PATCH 277/509] [PATCH] libata: minor fix for 2.6.16-rc3

 - Fix the array index value in ata_rwcmd_protocol() for the added FUA commands.
 - Filter out ATAPI packet command error messages in ata_pio_error()

Signed-off-by: Albert Lee <albertcc@tw.ibm.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/scsi/libata-core.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index 46c4cdbaee86..7ddd5a69352a 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -614,7 +614,7 @@ int ata_rwcmd_protocol(struct ata_queued_cmd *qc)
 	} else if (lba48 && (qc->ap->flags & ATA_FLAG_PIO_LBA48)) {
 		/* Unable to use DMA due to host limitation */
 		tf->protocol = ATA_PROT_PIO;
-		index = dev->multi_count ? 0 : 4;
+		index = dev->multi_count ? 0 : 8;
 	} else {
 		tf->protocol = ATA_PROT_DMA;
 		index = 16;
@@ -3357,11 +3357,12 @@ static void ata_pio_error(struct ata_port *ap)
 {
 	struct ata_queued_cmd *qc;
 
-	printk(KERN_WARNING "ata%u: PIO error\n", ap->id);
-
 	qc = ata_qc_from_tag(ap, ap->active_tag);
 	assert(qc != NULL);
 
+	if (qc->tf.command != ATA_CMD_PACKET)
+		printk(KERN_WARNING "ata%u: PIO error\n", ap->id);
+
 	/* make sure qc->err_mask is available to 
 	 * know what's wrong and recover
 	 */

From c15d85c8f3f73b5f20aae7928e25b6996f16b328 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@suse.de>
Date: Wed, 15 Feb 2006 15:59:25 +0100
Subject: [PATCH 278/509] [PATCH] Add missing FUA write to sata_mv dma command
 list

Signed-off-by: Jens Axboe <axboe@suse.de>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/scsi/sata_mv.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/scsi/sata_mv.c b/drivers/scsi/sata_mv.c
index 6fddf17a3b70..2770005324b4 100644
--- a/drivers/scsi/sata_mv.c
+++ b/drivers/scsi/sata_mv.c
@@ -997,6 +997,7 @@ static void mv_qc_prep(struct ata_queued_cmd *qc)
 	case ATA_CMD_READ_EXT:
 	case ATA_CMD_WRITE:
 	case ATA_CMD_WRITE_EXT:
+	case ATA_CMD_WRITE_FUA_EXT:
 		mv_crqb_pack_cmd(cw++, tf->hob_nsect, ATA_REG_NSECT, 0);
 		break;
 #ifdef LIBATA_NCQ		/* FIXME: remove this line when NCQ added */

From b2f49033d80c952a0ffc2d5647bc1a0b8a09c1b3 Mon Sep 17 00:00:00 2001
From: Peter Staubach <staubach@redhat.com>
Date: Fri, 17 Feb 2006 13:52:36 -0800
Subject: [PATCH 279/509] [PATCH] fix deadlock in ext2

Fix a deadlock possible in the ext2 file system implementation.  This
deadlock occurs when a file is removed from an ext2 file system which was
mounted with the "sync" mount option.

The problem is that ext2_xattr_delete_inode() was invoking the routine,
sync_dirty_buffer(), using a buffer head which was previously locked via
lock_buffer().  The first thing that sync_dirty_buffer() does is to lock
the buffer head that it was passed.  It does this via lock_buffer().  Oops.

The solution is to unlock the buffer head in ext2_xattr_delete_inode()
before invoking sync_dirty_buffer().  This makes the code in
ext2_xattr_delete_inode() obey the same locking rules as all other callers
of sync_dirty_buffer() in the ext2 file system implementation.

Signed-off-by: Peter Staubach <staubach@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ext2/xattr.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index a2ca3107d475..86ae8e93adb9 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -792,18 +792,20 @@ ext2_xattr_delete_inode(struct inode *inode)
 		ext2_free_blocks(inode, EXT2_I(inode)->i_file_acl, 1);
 		get_bh(bh);
 		bforget(bh);
+		unlock_buffer(bh);
 	} else {
 		HDR(bh)->h_refcount = cpu_to_le32(
 			le32_to_cpu(HDR(bh)->h_refcount) - 1);
 		if (ce)
 			mb_cache_entry_release(ce);
+		ea_bdebug(bh, "refcount now=%d",
+			le32_to_cpu(HDR(bh)->h_refcount));
+		unlock_buffer(bh);
 		mark_buffer_dirty(bh);
 		if (IS_SYNC(inode))
 			sync_dirty_buffer(bh);
 		DQUOT_FREE_BLOCK(inode, 1);
 	}
-	ea_bdebug(bh, "refcount now=%d", le32_to_cpu(HDR(bh)->h_refcount) - 1);
-	unlock_buffer(bh);
 	EXT2_I(inode)->i_file_acl = 0;
 
 cleanup:

From 8c9e877949d953e80d0d400bc4d1d1195a2028a4 Mon Sep 17 00:00:00 2001
From: Marcel Selhorst <selhorst@crypto.rub.de>
Date: Fri, 17 Feb 2006 13:52:41 -0800
Subject: [PATCH 280/509] [PATCH] Infineon TPM: IO-port leakage fix, WTX-bugfix

Fix IO-port leakage from request_region in case of error during TPM
initialization, adds more pnp-verification and fixes a WTX-bug.

Signed-off-by: Marcel Selhorst <selhorst@crypto.rub.de>
Acked-by: Kylene Jo Hall <kjhall@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/tpm/tpm_infineon.c | 48 +++++++++++++++++++++++----------
 1 file changed, 34 insertions(+), 14 deletions(-)

diff --git a/drivers/char/tpm/tpm_infineon.c b/drivers/char/tpm/tpm_infineon.c
index ec7590951af5..24095f6ee6da 100644
--- a/drivers/char/tpm/tpm_infineon.c
+++ b/drivers/char/tpm/tpm_infineon.c
@@ -33,6 +33,7 @@
 static int TPM_INF_DATA;
 static int TPM_INF_ADDR;
 static int TPM_INF_BASE;
+static int TPM_INF_ADDR_LEN;
 static int TPM_INF_PORT_LEN;
 
 /* TPM header definitions */
@@ -195,6 +196,7 @@ static int tpm_inf_recv(struct tpm_chip *chip, u8 * buf, size_t count)
 	int i;
 	int ret;
 	u32 size = 0;
+	number_of_wtx = 0;
 
 recv_begin:
 	/* start receiving header */
@@ -378,24 +380,35 @@ static int __devinit tpm_inf_pnp_probe(struct pnp_dev *dev,
 	if (pnp_port_valid(dev, 0) && pnp_port_valid(dev, 1) &&
 	    !(pnp_port_flags(dev, 0) & IORESOURCE_DISABLED)) {
 		TPM_INF_ADDR = pnp_port_start(dev, 0);
+		TPM_INF_ADDR_LEN = pnp_port_len(dev, 0);
 		TPM_INF_DATA = (TPM_INF_ADDR + 1);
 		TPM_INF_BASE = pnp_port_start(dev, 1);
 		TPM_INF_PORT_LEN = pnp_port_len(dev, 1);
-		if (!TPM_INF_PORT_LEN)
-			return -EINVAL;
+		if ((TPM_INF_PORT_LEN < 4) || (TPM_INF_ADDR_LEN < 2)) {
+			rc = -EINVAL;
+			goto err_last;
+		}
 		dev_info(&dev->dev, "Found %s with ID %s\n",
 			 dev->name, dev_id->id);
-		if (!((TPM_INF_BASE >> 8) & 0xff))
-			return -EINVAL;
+		if (!((TPM_INF_BASE >> 8) & 0xff)) {
+			rc = -EINVAL;
+			goto err_last;
+		}
 		/* publish my base address and request region */
 		tpm_inf.base = TPM_INF_BASE;
 		if (request_region
 		    (tpm_inf.base, TPM_INF_PORT_LEN, "tpm_infineon0") == NULL) {
-			release_region(tpm_inf.base, TPM_INF_PORT_LEN);
-			return -EINVAL;
+			rc = -EINVAL;
+			goto err_last;
+		}
+		if (request_region(TPM_INF_ADDR, TPM_INF_ADDR_LEN,
+				"tpm_infineon0") == NULL) {
+			rc = -EINVAL;
+			goto err_last;
 		}
 	} else {
-		return -EINVAL;
+		rc = -EINVAL;
+		goto err_last;
 	}
 
 	/* query chip for its vendor, its version number a.s.o. */
@@ -443,8 +456,8 @@ static int __devinit tpm_inf_pnp_probe(struct pnp_dev *dev,
 			dev_err(&dev->dev,
 				"Could not set IO-ports to 0x%lx\n",
 				tpm_inf.base);
-			release_region(tpm_inf.base, TPM_INF_PORT_LEN);
-			return -EIO;
+			rc = -EIO;
+			goto err_release_region;
 		}
 
 		/* activate register */
@@ -471,14 +484,21 @@ static int __devinit tpm_inf_pnp_probe(struct pnp_dev *dev,
 
 		rc = tpm_register_hardware(&dev->dev, &tpm_inf);
 		if (rc < 0) {
-			release_region(tpm_inf.base, TPM_INF_PORT_LEN);
-			return -ENODEV;
+			rc = -ENODEV;
+			goto err_release_region;
 		}
 		return 0;
 	} else {
-		dev_info(&dev->dev, "No Infineon TPM found!\n");
-		return -ENODEV;
+		rc = -ENODEV;
+		goto err_release_region;
 	}
+
+err_release_region:
+	release_region(tpm_inf.base, TPM_INF_PORT_LEN);
+	release_region(TPM_INF_ADDR, TPM_INF_ADDR_LEN);
+
+err_last:
+	return rc;
 }
 
 static __devexit void tpm_inf_pnp_remove(struct pnp_dev *dev)
@@ -518,5 +538,5 @@ module_exit(cleanup_inf);
 
 MODULE_AUTHOR("Marcel Selhorst <selhorst@crypto.rub.de>");
 MODULE_DESCRIPTION("Driver for Infineon TPM SLD 9630 TT 1.1 / SLB 9635 TT 1.2");
-MODULE_VERSION("1.6");
+MODULE_VERSION("1.7");
 MODULE_LICENSE("GPL");

From 05efc67d100ff6c3364604b72729addf1a86fdab Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Fri, 17 Feb 2006 13:52:42 -0800
Subject: [PATCH 281/509] [PATCH] arch/sh/Kconfig: fix the ISA_DMA_API
 dependencies

Jean-Luc Leger <reiga@dspnet.fr.eu.org> found this obvious typo.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/sh/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 504d56f8ca7f..e73621d03a28 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -446,7 +446,7 @@ endmenu
 
 config ISA_DMA_API
 	bool
-	depends on MPC1211
+	depends on SH_MPC1211
 	default y
 
 menu "Kernel features"

From 4bbf39c29bc3409d6454faf0dfa1b3b0aa2ac2af Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 17 Feb 2006 13:52:44 -0800
Subject: [PATCH 282/509] [PATCH] Introduce CONFIG_DEFAULT_MIGRATION_COST

Heiko Carstens <heiko.carstens@de.ibm.com> wrote:

  The boot sequence on s390 sometimes takes ages and we spend a very long
  time (up to one or two minutes) in calibrate_migration_costs.  The time
  spent there differs from boot to boot.  Also the calculated costs differ
  a lot.  I've seen differences by up to a factor of 15 (yes, factor not
  percent).  Also I doubt that making these measurements make much sense on
  a completely virtualized architecture where you cannot tell how much cpu
  time you will get anyway.

So introduce the CONFIG_DEFAULT_MIGRATION_COST method for an architecture
to set the scheduler migration costs.  This turns off automatic detection
of migration costs.  Makes sense on virtual platforms, where migration
costs are hard to measure accurately.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/Kconfig |  4 ++++
 kernel/sched.c    | 13 ++++++++++++-
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index b66602ad7b33..b7ca5bf9acfc 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -80,6 +80,10 @@ config HOTPLUG_CPU
 	  can be controlled through /sys/devices/system/cpu/cpu#.
 	  Say N if you want to disable CPU hotplug.
 
+config DEFAULT_MIGRATION_COST
+	int
+	default "1000000"
+
 config MATHEMU
 	bool "IEEE FPU emulation"
 	depends on MARCH_G5
diff --git a/kernel/sched.c b/kernel/sched.c
index 66d957227de9..12d291bf3379 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5058,7 +5058,18 @@ static void init_sched_build_groups(struct sched_group groups[], cpumask_t span,
 #define MAX_DOMAIN_DISTANCE 32
 
 static unsigned long long migration_cost[MAX_DOMAIN_DISTANCE] =
-		{ [ 0 ... MAX_DOMAIN_DISTANCE-1 ] = -1LL };
+		{ [ 0 ... MAX_DOMAIN_DISTANCE-1 ] =
+/*
+ * Architectures may override the migration cost and thus avoid
+ * boot-time calibration. Unit is nanoseconds. Mostly useful for
+ * virtualized hardware:
+ */
+#ifdef CONFIG_DEFAULT_MIGRATION_COST
+			CONFIG_DEFAULT_MIGRATION_COST
+#else
+			-1LL
+#endif
+};
 
 /*
  * Allow override of migration cost - in units of microseconds.

From 6d751c43b29deb1d990fb9644c13ca941c9d1305 Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Fri, 17 Feb 2006 13:52:45 -0800
Subject: [PATCH 283/509] [PATCH] s390: ccw device disbanding

If __ccw_device_disband_start() fails to initiate disbanding, it should finish
with ccw_device_disband_done() (which leaves the device in offline state)
instead of ccw_device_verify_done() (which leaves the device in online state).

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/s390/cio/device_pgid.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/s390/cio/device_pgid.c b/drivers/s390/cio/device_pgid.c
index d2a5b04d7cba..85b1020a1fcc 100644
--- a/drivers/s390/cio/device_pgid.c
+++ b/drivers/s390/cio/device_pgid.c
@@ -405,7 +405,7 @@ __ccw_device_disband_start(struct ccw_device *cdev)
 		cdev->private->iretry = 5;
 		cdev->private->imask >>= 1;
 	}
-	ccw_device_verify_done(cdev, (sch->lpm != 0) ? 0 : -ENODEV);
+	ccw_device_disband_done(cdev, (sch->lpm != 0) ? 0 : -ENODEV);
 }
 
 /*

From 1fca251f36fac3fae7d9cf10de69c2c93f6c0000 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 17 Feb 2006 13:52:46 -0800
Subject: [PATCH 284/509] [PATCH] s390: fix preempt_count of idle thread with
 cpu hotplug

Set preempt_count of idle_thread to zero before switching off cpu.  Otherwise
the preempt_count will be wrong if the cpu is switched on again since the
thread will be reused.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/process.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 008c74526fd3..da6fbae8df91 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -128,8 +128,10 @@ void default_idle(void)
 	__ctl_set_bit(8, 15);
 
 #ifdef CONFIG_HOTPLUG_CPU
-	if (cpu_is_offline(cpu))
+	if (cpu_is_offline(cpu)) {
+		preempt_enable_no_resched();
 		cpu_die();
+	}
 #endif
 
 	local_mcck_disable();

From 255acee706b333b79f593dd366f16e1f107cccc3 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 17 Feb 2006 13:52:46 -0800
Subject: [PATCH 285/509] [PATCH] s390: additional_cpus parameter

Introduce additional_cpus command line option.  By default no additional cpu
can be attached to the system anymore.  Only the cpus present at IPL time can
be switched on/off.  If it is desired that additional cpus can be attached to
the system the maximum number of additional cpus needs to be specified with
this option.

This change is necessary in order to limit the waste of per_cpu data
structures.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/cpu-hotplug.txt | 10 +++---
 arch/s390/kernel/setup.c      |  2 ++
 arch/s390/kernel/smp.c        | 58 ++++++++++++++++++++++++-----------
 include/asm-s390/smp.h        |  2 ++
 4 files changed, 49 insertions(+), 23 deletions(-)

diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt
index e05278087ffa..4d3355da0e26 100644
--- a/Documentation/cpu-hotplug.txt
+++ b/Documentation/cpu-hotplug.txt
@@ -11,6 +11,8 @@
 			Joel Schopp <jschopp@austin.ibm.com>
 		ia64/x86_64:
 			Ashok Raj <ashok.raj@intel.com>
+		s390:
+			Heiko Carstens <heiko.carstens@de.ibm.com>
 
 Authors: Ashok Raj <ashok.raj@intel.com>
 Lots of feedback: Nathan Lynch <nathanl@austin.ibm.com>,
@@ -44,11 +46,9 @@ maxcpus=n    Restrict boot time cpus to n. Say if you have 4 cpus, using
              maxcpus=2 will only boot 2. You can choose to bring the
              other cpus later online, read FAQ's for more info.
 
-additional_cpus*=n	Use this to limit hotpluggable cpus. This option sets
-			cpu_possible_map = cpu_present_map + additional_cpus
-
-(*) Option valid only for following architectures
-- x86_64, ia64
+additional_cpus=n	[x86_64, s390 only] use this to limit hotpluggable cpus.
+                          This option sets
+  			cpu_possible_map = cpu_present_map + additional_cpus
 
 ia64 and x86_64 use the number of disabled local apics in ACPI tables MADT
 to determine the number of potentially hot-pluggable cpus. The implementation
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index de8784267473..24f62f16c0e5 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -600,6 +600,7 @@ setup_arch(char **cmdline_p)
 	init_mm.brk = (unsigned long) &_end;
 
 	parse_cmdline_early(cmdline_p);
+	parse_early_param();
 
 	setup_memory();
 	setup_resources();
@@ -607,6 +608,7 @@ setup_arch(char **cmdline_p)
 
         cpu_init();
         __cpu_logical_map[0] = S390_lowcore.cpu_data.cpu_addr;
+	smp_setup_cpu_possible_map();
 
 	/*
 	 * Create kernel page tables and switch to virtual addressing.
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 0d1ad5dbe2b1..53291e94ac7b 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -1,8 +1,7 @@
 /*
  *  arch/s390/kernel/smp.c
  *
- *  S390 version
- *    Copyright (C) 1999,2000 IBM Deutschland Entwicklung GmbH, IBM Corporation
+ *    Copyright (C) IBM Corp. 1999,2006
  *    Author(s): Denis Joseph Barrow (djbarrow@de.ibm.com,barrow_dj@yahoo.com),
  *               Martin Schwidefsky (schwidefsky@de.ibm.com)
  *               Heiko Carstens (heiko.carstens@de.ibm.com)
@@ -41,8 +40,6 @@
 #include <asm/cpcmd.h>
 #include <asm/tlbflush.h>
 
-/* prototypes */
-
 extern volatile int __cpu_logical_map[];
 
 /*
@@ -51,13 +48,11 @@ extern volatile int __cpu_logical_map[];
 
 struct _lowcore *lowcore_ptr[NR_CPUS];
 
-cpumask_t cpu_online_map;
-cpumask_t cpu_possible_map = CPU_MASK_ALL;
+cpumask_t cpu_online_map = CPU_MASK_NONE;
+cpumask_t cpu_possible_map = CPU_MASK_NONE;
 
 static struct task_struct *current_set[NR_CPUS];
 
-EXPORT_SYMBOL(cpu_online_map);
-
 /*
  * Reboot, halt and power_off routines for SMP.
  */
@@ -490,10 +485,10 @@ void smp_ctl_clear_bit(int cr, int bit) {
  * Lets check how many CPUs we have.
  */
 
-void
-__init smp_check_cpus(unsigned int max_cpus)
+static unsigned int
+__init smp_count_cpus(void)
 {
-	int cpu, num_cpus;
+	unsigned int cpu, num_cpus;
 	__u16 boot_cpu_addr;
 
 	/*
@@ -503,19 +498,20 @@ __init smp_check_cpus(unsigned int max_cpus)
 	boot_cpu_addr = S390_lowcore.cpu_data.cpu_addr;
 	current_thread_info()->cpu = 0;
 	num_cpus = 1;
-	for (cpu = 0; cpu <= 65535 && num_cpus < max_cpus; cpu++) {
+	for (cpu = 0; cpu <= 65535; cpu++) {
 		if ((__u16) cpu == boot_cpu_addr)
 			continue;
-		__cpu_logical_map[num_cpus] = (__u16) cpu;
-		if (signal_processor(num_cpus, sigp_sense) ==
+		__cpu_logical_map[1] = (__u16) cpu;
+		if (signal_processor(1, sigp_sense) ==
 		    sigp_not_operational)
 			continue;
-		cpu_set(num_cpus, cpu_present_map);
 		num_cpus++;
 	}
 
 	printk("Detected %d CPU's\n",(int) num_cpus);
 	printk("Boot cpu address %2X\n", boot_cpu_addr);
+
+	return num_cpus;
 }
 
 /*
@@ -676,6 +672,32 @@ __cpu_up(unsigned int cpu)
 	return 0;
 }
 
+static unsigned int __initdata additional_cpus;
+
+void __init smp_setup_cpu_possible_map(void)
+{
+	unsigned int pcpus, cpu;
+
+	pcpus = smp_count_cpus() + additional_cpus;
+
+	if (pcpus > NR_CPUS)
+		pcpus = NR_CPUS;
+
+	for (cpu = 0; cpu < pcpus; cpu++)
+		cpu_set(cpu, cpu_possible_map);
+
+	cpu_present_map = cpu_possible_map;
+}
+
+#ifdef CONFIG_HOTPLUG_CPU
+
+static int __init setup_additional_cpus(char *s)
+{
+	additional_cpus = simple_strtoul(s, NULL, 0);
+	return 0;
+}
+early_param("additional_cpus", setup_additional_cpus);
+
 int
 __cpu_disable(void)
 {
@@ -744,6 +766,8 @@ cpu_die(void)
 	for(;;);
 }
 
+#endif /* CONFIG_HOTPLUG_CPU */
+
 /*
  *	Cycle through the processors and setup structures.
  */
@@ -757,7 +781,6 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
         /* request the 0x1201 emergency signal external interrupt */
         if (register_external_interrupt(0x1201, do_ext_call_interrupt) != 0)
                 panic("Couldn't request external interrupt 0x1201");
-        smp_check_cpus(max_cpus);
         memset(lowcore_ptr,0,sizeof(lowcore_ptr));  
         /*
          *  Initialize prefix pages and stacks for all possible cpus
@@ -806,14 +829,12 @@ void __devinit smp_prepare_boot_cpu(void)
 	BUG_ON(smp_processor_id() != 0);
 
 	cpu_set(0, cpu_online_map);
-	cpu_set(0, cpu_present_map);
 	S390_lowcore.percpu_offset = __per_cpu_offset[0];
 	current_set[0] = current;
 }
 
 void smp_cpus_done(unsigned int max_cpus)
 {
-	cpu_present_map = cpu_possible_map;
 }
 
 /*
@@ -845,6 +866,7 @@ static int __init topology_init(void)
 
 subsys_initcall(topology_init);
 
+EXPORT_SYMBOL(cpu_online_map);
 EXPORT_SYMBOL(cpu_possible_map);
 EXPORT_SYMBOL(lowcore_ptr);
 EXPORT_SYMBOL(smp_ctl_set_bit);
diff --git a/include/asm-s390/smp.h b/include/asm-s390/smp.h
index 9c6e9c300eb9..444dae5912e6 100644
--- a/include/asm-s390/smp.h
+++ b/include/asm-s390/smp.h
@@ -31,6 +31,7 @@ typedef struct
 	__u16      cpu;
 } sigp_info;
 
+extern void smp_setup_cpu_possible_map(void);
 extern int smp_call_function_on(void (*func) (void *info), void *info,
 				int nonatomic, int wait, int cpu);
 #define NO_PROC_ID		0xFF		/* No processor magic marker */
@@ -104,6 +105,7 @@ smp_call_function_on(void (*func) (void *info), void *info,
 #define smp_cpu_not_running(cpu)	1
 #define smp_get_cpu(cpu) ({ 0; })
 #define smp_put_cpu(cpu) ({ 0; })
+#define smp_setup_cpu_possible_map()
 #endif
 
 #endif

From 37a3302618a51520e2056494715ea6b4776dd8ab Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 17 Feb 2006 13:52:47 -0800
Subject: [PATCH 286/509] [PATCH] s390: possible_cpus parameter

Introduce possible_cpus command line option.  Hard sets the number of bits set
in cpu_possible_map.  Unlike the additional_cpus parameter this one guarantees
that num_possible_cpus() will stay constant even if the system gets rebooted
and a different number of cpus are present at startup.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/cpu-hotplug.txt |  6 ++++++
 arch/s390/kernel/smp.c        | 14 +++++++++++---
 2 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt
index 4d3355da0e26..e71bc6cbbc5e 100644
--- a/Documentation/cpu-hotplug.txt
+++ b/Documentation/cpu-hotplug.txt
@@ -58,6 +58,12 @@ mark such hot-pluggable cpus as disabled entries, one could use this
 parameter "additional_cpus=x" to represent those cpus in the cpu_possible_map.
 
 
+possible_cpus=n		[s390 only] use this to set hotpluggable cpus.
+			This option sets possible_cpus bits in
+			cpu_possible_map. Thus keeping the numbers of bits set
+			constant even if the machine gets rebooted.
+			This option overrides additional_cpus.
+
 CPU maps and such
 -----------------
 [More on cpumaps and primitive to manipulate, please check
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 53291e94ac7b..d0a2745aec7f 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -673,15 +673,16 @@ __cpu_up(unsigned int cpu)
 }
 
 static unsigned int __initdata additional_cpus;
+static unsigned int __initdata possible_cpus;
 
 void __init smp_setup_cpu_possible_map(void)
 {
 	unsigned int pcpus, cpu;
 
-	pcpus = smp_count_cpus() + additional_cpus;
+	pcpus = min(smp_count_cpus() + additional_cpus, (unsigned int) NR_CPUS);
 
-	if (pcpus > NR_CPUS)
-		pcpus = NR_CPUS;
+	if (possible_cpus)
+		pcpus = min(possible_cpus, (unsigned int) NR_CPUS);
 
 	for (cpu = 0; cpu < pcpus; cpu++)
 		cpu_set(cpu, cpu_possible_map);
@@ -698,6 +699,13 @@ static int __init setup_additional_cpus(char *s)
 }
 early_param("additional_cpus", setup_additional_cpus);
 
+static int __init setup_possible_cpus(char *s)
+{
+	possible_cpus = simple_strtoul(s, NULL, 0);
+	return 0;
+}
+early_param("possible_cpus", setup_possible_cpus);
+
 int
 __cpu_disable(void)
 {

From 54330456b2e3398743586254f6d7695061ea0d49 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 17 Feb 2006 13:52:48 -0800
Subject: [PATCH 287/509] [PATCH] s390: smp initialization speed

The last changes that introduced the additional_cpus command line parameter
also introduced a regression regarding smp initialization speed.  In
smp_setup_cpu_possible_map() cpu_present_map is set to the same value as
cpu_possible_map.  Especially that means that bits in the present map will be
set for cpus that are not present.  This will cause a slow down in the initial
cpu_up() loop in smp_init() since trying to take cpus online that aren't
present takes a while.

Fix this by setting only bits for present cpus in cpu_present_map and set
cpu_present_map to cpu_possible_map in smp_cpus_done().

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/smp.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index d0a2745aec7f..7dbe00c76c6b 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -677,17 +677,21 @@ static unsigned int __initdata possible_cpus;
 
 void __init smp_setup_cpu_possible_map(void)
 {
-	unsigned int pcpus, cpu;
+	unsigned int phy_cpus, pos_cpus, cpu;
 
-	pcpus = min(smp_count_cpus() + additional_cpus, (unsigned int) NR_CPUS);
+	phy_cpus = smp_count_cpus();
+	pos_cpus = min(phy_cpus + additional_cpus, (unsigned int) NR_CPUS);
 
 	if (possible_cpus)
-		pcpus = min(possible_cpus, (unsigned int) NR_CPUS);
+		pos_cpus = min(possible_cpus, (unsigned int) NR_CPUS);
 
-	for (cpu = 0; cpu < pcpus; cpu++)
+	for (cpu = 0; cpu < pos_cpus; cpu++)
 		cpu_set(cpu, cpu_possible_map);
 
-	cpu_present_map = cpu_possible_map;
+	phy_cpus = min(phy_cpus, pos_cpus);
+
+	for (cpu = 0; cpu < phy_cpus; cpu++)
+		cpu_set(cpu, cpu_present_map);
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
@@ -843,6 +847,7 @@ void __devinit smp_prepare_boot_cpu(void)
 
 void smp_cpus_done(unsigned int max_cpus)
 {
+	cpu_present_map = cpu_possible_map;
 }
 
 /*

From 6cadb78b3bec0a439a99db8fb550dc568e924ae6 Mon Sep 17 00:00:00 2001
From: Cornelia Huck <cornelia.huck@de.ibm.com>
Date: Fri, 17 Feb 2006 13:52:49 -0800
Subject: [PATCH 288/509] [PATCH] s390: fix assignment instead of check in
 ccw_device_set_online()

Fix assignment instead of check in ccw_device_set_online().  Also remove
unneeded assignment in ccw_device_do_sense().

Signed-off-by: Cornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/s390/cio/device.c        | 2 +-
 drivers/s390/cio/device_status.c | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/s390/cio/device.c b/drivers/s390/cio/device.c
index 062fb100d94c..afc4e88551ad 100644
--- a/drivers/s390/cio/device.c
+++ b/drivers/s390/cio/device.c
@@ -359,7 +359,7 @@ ccw_device_set_online(struct ccw_device *cdev)
 	else 
 		pr_debug("ccw_device_offline returned %d, device %s\n",
 			 ret, cdev->dev.bus_id);
-	return (ret = 0) ? -ENODEV : ret;
+	return (ret == 0) ? -ENODEV : ret;
 }
 
 static ssize_t
diff --git a/drivers/s390/cio/device_status.c b/drivers/s390/cio/device_status.c
index dad4dd9887c9..6c762b43f921 100644
--- a/drivers/s390/cio/device_status.c
+++ b/drivers/s390/cio/device_status.c
@@ -317,7 +317,6 @@ ccw_device_do_sense(struct ccw_device *cdev, struct irb *irb)
 	/*
 	 * We have ending status but no sense information. Do a basic sense.
 	 */
-	sch = to_subchannel(cdev->dev.parent);
 	sch->sense_ccw.cmd_code = CCW_CMD_BASIC_SENSE;
 	sch->sense_ccw.cda = (__u32) __pa(cdev->private->irb.ecw);
 	sch->sense_ccw.count = SENSE_MAX_COUNT;

From ed3d021b823336a2e0c5090a91d083243f756e6c Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Fri, 17 Feb 2006 13:52:50 -0800
Subject: [PATCH 289/509] [PATCH] s390: sys32_fstatat -> sys32_fstatat64

Just rename the compat system call to keep the name consistent with all the
other *64 compat system calls.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/s390/kernel/compat_linux.c   | 4 ++--
 arch/s390/kernel/compat_wrapper.S | 6 +++---
 arch/s390/kernel/syscalls.S       | 2 +-
 3 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c
index 2d021626c1a6..cc058dc3bc8b 100644
--- a/arch/s390/kernel/compat_linux.c
+++ b/arch/s390/kernel/compat_linux.c
@@ -905,8 +905,8 @@ asmlinkage long sys32_fstat64(unsigned long fd, struct stat64_emu31 __user * sta
 	return ret;
 }
 
-asmlinkage long sys32_fstatat(unsigned int dfd, char __user *filename,
-			      struct stat64_emu31 __user* statbuf, int flag)
+asmlinkage long sys32_fstatat64(unsigned int dfd, char __user *filename,
+				struct stat64_emu31 __user* statbuf, int flag)
 {
 	struct kstat stat;
 	int error = -EINVAL;
diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index dd2d6c3e8df8..615964cca15f 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -1523,13 +1523,13 @@ compat_sys_futimesat_wrapper:
 	llgtr	%r4,%r4			# struct timeval *
 	jg	compat_sys_futimesat
 
-	.globl sys32_fstatat_wrapper
-sys32_fstatat_wrapper:
+	.globl sys32_fstatat64_wrapper
+sys32_fstatat64_wrapper:
 	llgfr	%r2,%r2			# unsigned int
 	llgtr	%r3,%r3			# char *
 	llgtr	%r4,%r4			# struct stat64 *
 	lgfr	%r5,%r5			# int
-	jg	sys32_fstatat
+	jg	sys32_fstatat64
 
 	.globl sys_unlinkat_wrapper
 sys_unlinkat_wrapper:
diff --git a/arch/s390/kernel/syscalls.S b/arch/s390/kernel/syscalls.S
index 84921fe8d266..7c88d85c3597 100644
--- a/arch/s390/kernel/syscalls.S
+++ b/arch/s390/kernel/syscalls.S
@@ -301,7 +301,7 @@ SYSCALL(sys_mkdirat,sys_mkdirat,sys_mkdirat_wrapper)
 SYSCALL(sys_mknodat,sys_mknodat,sys_mknodat_wrapper)	/* 290 */
 SYSCALL(sys_fchownat,sys_fchownat,sys_fchownat_wrapper)
 SYSCALL(sys_futimesat,sys_futimesat,compat_sys_futimesat_wrapper)
-SYSCALL(sys_fstatat64,sys_newfstatat,sys32_fstatat_wrapper)
+SYSCALL(sys_fstatat64,sys_newfstatat,sys32_fstatat64_wrapper)
 SYSCALL(sys_unlinkat,sys_unlinkat,sys_unlinkat_wrapper)
 SYSCALL(sys_renameat,sys_renameat,sys_renameat_wrapper)	/* 295 */
 SYSCALL(sys_linkat,sys_linkat,sys_linkat_wrapper)

From a8534adb74e23374889b84b3d97eb18da542a1b5 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Fri, 17 Feb 2006 13:52:51 -0800
Subject: [PATCH 290/509] [PATCH] swsusp: fix breakage with swap on LVM

Restore the compatibility with the older code and make it possible to
suspend if the kernel command line doesn't contain the "resume=" argument

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/power/swsusp.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 4e90905f0e87..2d9d08f72f76 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -153,13 +153,11 @@ static int swsusp_swap_check(void) /* This is called before saving image */
 {
 	int i;
 
-	if (!swsusp_resume_device)
-		return -ENODEV;
 	spin_lock(&swap_lock);
 	for (i = 0; i < MAX_SWAPFILES; i++) {
 		if (!(swap_info[i].flags & SWP_WRITEOK))
 			continue;
-		if (is_resume_device(swap_info + i)) {
+		if (!swsusp_resume_device || is_resume_device(swap_info + i)) {
 			spin_unlock(&swap_lock);
 			root_swap = i;
 			return 0;

From 77e7f250f88cd62844e24c42aff4d0e95969c746 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <miklos@szeredi.hu>
Date: Fri, 17 Feb 2006 13:52:52 -0800
Subject: [PATCH 291/509] [PATCH] fuse: fix bug in aborted fuse_release_end()

There's a rather theoretical case of the BUG triggering in
fuse_reset_request():

  - iget() fails because of OOM after a successful CREATE_OPEN request
  - during IO on the resulting RELEASE request the connection is aborted

Fix and add warning to fuse_reset_request().

Signed-off-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/fuse/dev.c  |  6 ++++++
 fs/fuse/file.c | 11 ++++++++---
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index f556a0d5c0d3..0c9a2ee54c91 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -66,6 +66,12 @@ static void restore_sigs(sigset_t *oldset)
 	sigprocmask(SIG_SETMASK, oldset, NULL);
 }
 
+/*
+ * Reset request, so that it can be reused
+ *
+ * The caller must be _very_ careful to make sure, that it is holding
+ * the only reference to req
+ */
 void fuse_reset_request(struct fuse_req *req)
 {
 	int preallocated = req->preallocated;
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 296351615b00..6f05379b0a0d 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -116,9 +116,14 @@ int fuse_open_common(struct inode *inode, struct file *file, int isdir)
 /* Special case for failed iget in CREATE */
 static void fuse_release_end(struct fuse_conn *fc, struct fuse_req *req)
 {
-	u64 nodeid = req->in.h.nodeid;
-	fuse_reset_request(req);
-	fuse_send_forget(fc, req, nodeid, 1);
+	/* If called from end_io_requests(), req has more than one
+	   reference and fuse_reset_request() cannot work */
+	if (fc->connected) {
+		u64 nodeid = req->in.h.nodeid;
+		fuse_reset_request(req);
+		fuse_send_forget(fc, req, nodeid, 1);
+	} else
+		fuse_put_request(fc, req);
 }
 
 void fuse_send_release(struct fuse_conn *fc, struct fuse_file *ff,

From 9127dd1aace4e89acb48fbcafd0ed27d3869847b Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Fri, 17 Feb 2006 13:52:54 -0800
Subject: [PATCH 292/509] [PATCH] allow windfarm_pm112 module to load

The windfarm_pm112 module relies on smu_sat_get_sdb_partition which is in
windfarm_smu_sat.c but is not exported to modules, so despite Kconfig
having the option to build the pm112 as modules, this can never be loaded.

This patch fixes that by exporting smu_sat_get_sdb_partition with
EXPORT_SYMBOL_GPL

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/macintosh/windfarm_smu_sat.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/macintosh/windfarm_smu_sat.c b/drivers/macintosh/windfarm_smu_sat.c
index 3a32c59494f2..24e51d5e97fc 100644
--- a/drivers/macintosh/windfarm_smu_sat.c
+++ b/drivers/macintosh/windfarm_smu_sat.c
@@ -151,6 +151,7 @@ struct smu_sdbp_header *smu_sat_get_sdb_partition(unsigned int sat_id, int id,
 	kfree(buf);
 	return NULL;
 }
+EXPORT_SYMBOL_GPL(smu_sat_get_sdb_partition);
 
 /* refresh the cache */
 static int wf_sat_read_cache(struct wf_sat *sat)

From 9ff4ced4676d3cd1f28b14d93a339f263ca304b0 Mon Sep 17 00:00:00 2001
From: Tim Hockin <thockin@google.com>
Date: Fri, 17 Feb 2006 13:52:54 -0800
Subject: [PATCH 293/509] [PATCH] Remove KERN_INFO from middle of printk line

Don't print KERN_INFO in the middle of a printk line.
	printk(KERN_INFO "OEM ID: %s ",str);
is just above this. This is already fixed up in i386 copy.

Signed-off-by: Martin J. Bligh <mbligh@google.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/mpparse.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/x86_64/kernel/mpparse.c b/arch/x86_64/kernel/mpparse.c
index dc49bfb6db0a..9013a90b5c2e 100644
--- a/arch/x86_64/kernel/mpparse.c
+++ b/arch/x86_64/kernel/mpparse.c
@@ -288,9 +288,9 @@ static int __init smp_read_mpc(struct mp_config_table *mpc)
 
 	memcpy(str,mpc->mpc_productid,12);
 	str[12]=0;
-	printk(KERN_INFO "Product ID: %s ",str);
+	printk("Product ID: %s ",str);
 
-	printk(KERN_INFO "APIC at: 0x%X\n",mpc->mpc_lapic);
+	printk("APIC at: 0x%X\n",mpc->mpc_lapic);
 
 	/* save the local APIC address, it might be non-default */
 	if (!acpi_lapic)

From 200a4552af34b9a32e1f68a881a9ed5c7ec699cc Mon Sep 17 00:00:00 2001
From: David Gibson <david@gibson.dropbear.id.au>
Date: Fri, 17 Feb 2006 13:52:56 -0800
Subject: [PATCH 294/509] [PATCH] powerpc: Fix accidentally-working typo in
 __pud_free_tlb

One of the parameters to the __pud_free_tlb() macro for powerpc is
incorrect (see patch) .  We get away with it by accident, because the one
place the macro is called, the second parameter is a variable named "pud".

Signed-off-by: David Gibson <dwg@au1.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-powerpc/pgalloc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-powerpc/pgalloc.h b/include/asm-powerpc/pgalloc.h
index 9f5b052784a5..a00ee002cd11 100644
--- a/include/asm-powerpc/pgalloc.h
+++ b/include/asm-powerpc/pgalloc.h
@@ -146,7 +146,7 @@ extern void pgtable_free_tlb(struct mmu_gather *tlb, pgtable_free_t pgf);
 	pgtable_free_tlb(tlb, pgtable_free_cache(pmd, \
 		PMD_CACHE_NUM, PMD_TABLE_SIZE-1))
 #ifndef CONFIG_PPC_64K_PAGES
-#define __pud_free_tlb(tlb, pmd)	\
+#define __pud_free_tlb(tlb, pud)	\
 	pgtable_free_tlb(tlb, pgtable_free_cache(pud, \
 		PUD_CACHE_NUM, PUD_TABLE_SIZE-1))
 #endif /* CONFIG_PPC_64K_PAGES */

From 74910e6c7dc7471b286a883c1a7af70483ffd2ba Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Fri, 17 Feb 2006 13:52:58 -0800
Subject: [PATCH 295/509] [PATCH] select: time comparison fixes

I got all of these backwards.  We want to return

	min(input timeout, new timeout)

to userspace to prevent increasing the time-remaining value.

Thanks to Ernst Herzberg <earny@net4u.de> for reporting and diagnosing.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/compat.c | 6 +++---
 fs/select.c | 6 +++---
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/fs/compat.c b/fs/compat.c
index a2ba78bdf7f7..5333c7d7427f 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -1757,7 +1757,7 @@ asmlinkage long compat_sys_select(int n, compat_ulong_t __user *inp,
 			goto sticky;
 		rtv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ));
 		rtv.tv_sec = timeout;
-		if (compat_timeval_compare(&rtv, &tv) < 0)
+		if (compat_timeval_compare(&rtv, &tv) >= 0)
 			rtv = tv;
 		if (copy_to_user(tvp, &rtv, sizeof(rtv))) {
 sticky:
@@ -1834,7 +1834,7 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp,
 			rts.tv_sec++;
 			rts.tv_nsec -= NSEC_PER_SEC;
 		}
-		if (compat_timespec_compare(&rts, &ts) < 0)
+		if (compat_timespec_compare(&rts, &ts) >= 0)
 			rts = ts;
 		copy_to_user(tsp, &rts, sizeof(rts));
 	}
@@ -1934,7 +1934,7 @@ asmlinkage long compat_sys_ppoll(struct pollfd __user *ufds,
 		rts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) *
 					1000;
 		rts.tv_sec = timeout;
-		if (compat_timespec_compare(&rts, &ts) < 0)
+		if (compat_timespec_compare(&rts, &ts) >= 0)
 			rts = ts;
 		if (copy_to_user(tsp, &rts, sizeof(rts))) {
 sticky:
diff --git a/fs/select.c b/fs/select.c
index 6ce68a9c8976..1815a57d2255 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -404,7 +404,7 @@ asmlinkage long sys_select(int n, fd_set __user *inp, fd_set __user *outp,
 			goto sticky;
 		rtv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ));
 		rtv.tv_sec = timeout;
-		if (timeval_compare(&rtv, &tv) < 0)
+		if (timeval_compare(&rtv, &tv) >= 0)
 			rtv = tv;
 		if (copy_to_user(tvp, &rtv, sizeof(rtv))) {
 sticky:
@@ -471,7 +471,7 @@ asmlinkage long sys_pselect7(int n, fd_set __user *inp, fd_set __user *outp,
 		rts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) *
 						1000;
 		rts.tv_sec = timeout;
-		if (timespec_compare(&rts, &ts) < 0)
+		if (timespec_compare(&rts, &ts) >= 0)
 			rts = ts;
 		if (copy_to_user(tsp, &rts, sizeof(rts))) {
 sticky:
@@ -775,7 +775,7 @@ asmlinkage long sys_ppoll(struct pollfd __user *ufds, unsigned int nfds,
 		rts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) *
 						1000;
 		rts.tv_sec = timeout;
-		if (timespec_compare(&rts, &ts) < 0)
+		if (timespec_compare(&rts, &ts) >= 0)
 			rts = ts;
 		if (copy_to_user(tsp, &rts, sizeof(rts))) {
 		sticky:

From 636f13c174dd7c84a437d3c3e8fa66f03f7fda63 Mon Sep 17 00:00:00 2001
From: Chris Wright <chrisw@sous-sol.org>
Date: Fri, 17 Feb 2006 13:59:36 -0800
Subject: [PATCH 296/509] [PATCH] sys_mbind sanity checking

Make sure maxnodes is safe size before calculating nlongs in
get_nodes().

Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/mempolicy.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 323fdcf128c4..bedfa4f09c80 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -808,6 +808,8 @@ static int get_nodes(nodemask_t *nodes, const unsigned long __user *nmask,
 	nodes_clear(*nodes);
 	if (maxnode == 0 || !nmask)
 		return 0;
+	if (maxnode > PAGE_SIZE)
+		return -EINVAL;
 
 	nlongs = BITS_TO_LONGS(maxnode);
 	if ((maxnode % BITS_PER_LONG) == 0)

From 35b73ceb9a7d10c81bd9e79e8485f7079ef2b40e Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Fri, 17 Feb 2006 13:59:50 -0800
Subject: [PATCH 297/509] [PATCH] ACPI: fix vendor resource length computation

acpi_rs_get_list_length() needs to account for all the vendor-defined data
bytes.  Failing to include these causes buffers to be sized too small,
which causes slab corruption when we later convert AML to resources and run
off the end of the buffer.

This causes slab corruption on machines that use ACPI vendor-defined
resources.  All HP ia64 machines do, and I'm told that some NEC machines
may as well.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/acpi/resources/rscalc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/acpi/resources/rscalc.c b/drivers/acpi/resources/rscalc.c
index 7d6481d9fbec..4038dbfa63a0 100644
--- a/drivers/acpi/resources/rscalc.c
+++ b/drivers/acpi/resources/rscalc.c
@@ -391,8 +391,7 @@ acpi_rs_get_list_length(u8 * aml_buffer,
 			 * Ensure a 32-bit boundary for the structure
 			 */
 			extra_struct_bytes =
-			    ACPI_ROUND_UP_to_32_bITS(resource_length) -
-			    resource_length;
+			    ACPI_ROUND_UP_to_32_bITS(resource_length);
 			break;
 
 		case ACPI_RESOURCE_NAME_END_TAG:
@@ -408,8 +407,7 @@ acpi_rs_get_list_length(u8 * aml_buffer,
 			 * Add vendor data and ensure a 32-bit boundary for the structure
 			 */
 			extra_struct_bytes =
-			    ACPI_ROUND_UP_to_32_bITS(resource_length) -
-			    resource_length;
+			    ACPI_ROUND_UP_to_32_bITS(resource_length);
 			break;
 
 		case ACPI_RESOURCE_NAME_ADDRESS32:

From bd71c2b17468a2531fb4c81ec1d73520845e97e1 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@g5.osdl.org>
Date: Fri, 17 Feb 2006 14:23:45 -0800
Subject: [PATCH 298/509] Linux v2.6.16-rc4

---
 Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Makefile b/Makefile
index 48d569d46ae1..77a448c8e776 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 16
-EXTRAVERSION =-rc3
+EXTRAVERSION =-rc4
 NAME=Sliding Snow Leopard
 
 # *DOCUMENTATION*

From 91e3738ebc6d858e784090382e02afeae5a93b08 Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@linux.ie>
Date: Sat, 18 Feb 2006 15:17:04 +1100
Subject: [PATCH 299/509] drm: fixup i915 interrupt on X server exit

Fixes: IRQ disabled (i915?) when switchig between gnome themes (gnome-theme-manager)

Signed-off-by: Dave Airlie <airlied@linux.ie>
---
 drivers/char/drm/i915_irq.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/char/drm/i915_irq.c b/drivers/char/drm/i915_irq.c
index a1381c61aa63..d3879ac9970f 100644
--- a/drivers/char/drm/i915_irq.c
+++ b/drivers/char/drm/i915_irq.c
@@ -202,10 +202,15 @@ void i915_driver_irq_postinstall(drm_device_t * dev)
 void i915_driver_irq_uninstall(drm_device_t * dev)
 {
 	drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
+	u16 temp;
+
 	if (!dev_priv)
 		return;
 
 	I915_WRITE16(I915REG_HWSTAM, 0xffff);
 	I915_WRITE16(I915REG_INT_MASK_R, 0xffff);
 	I915_WRITE16(I915REG_INT_ENABLE_R, 0x0);
+
+	temp = I915_READ16(I915REG_INT_IDENTITY_R);
+	I915_WRITE16(I915REG_INT_IDENTITY_R, temp);
 }

From 4e5e2e2560aa1d1d01f7af97af2f72706f61da27 Mon Sep 17 00:00:00 2001
From: Dave Airlie <airlied@linux.ie>
Date: Sat, 18 Feb 2006 15:51:35 +1100
Subject: [PATCH 300/509] drm: radeon add r300 TX_CNTL and verify bitblt
 packets

The Xgl on r300 doesn't work unless you add a verify bitblt function to the
DRM, and we need to pass TX_CNTL to flush texture caches.

Signed-off-by: Dave Airlie <airlied@linux.ie>
---
 drivers/char/drm/r300_cmdbuf.c | 50 ++++++++++++++++++++++++++++++++++
 drivers/char/drm/r300_reg.h    |  3 ++
 drivers/char/drm/radeon_drv.h  |  3 +-
 3 files changed, 55 insertions(+), 1 deletion(-)

diff --git a/drivers/char/drm/r300_cmdbuf.c b/drivers/char/drm/r300_cmdbuf.c
index 291dbf4c8186..6dd21754ba67 100644
--- a/drivers/char/drm/r300_cmdbuf.c
+++ b/drivers/char/drm/r300_cmdbuf.c
@@ -161,6 +161,7 @@ void r300_init_reg_flags(void)
 	ADD_RANGE(R300_VAP_PVS_CNTL_1, 3);
 	ADD_RANGE(R300_GB_ENABLE, 1);
 	ADD_RANGE(R300_GB_MSPOS0, 5);
+	ADD_RANGE(R300_TX_CNTL, 1);
 	ADD_RANGE(R300_TX_ENABLE, 1);
 	ADD_RANGE(0x4200, 4);
 	ADD_RANGE(0x4214, 1);
@@ -489,6 +490,52 @@ static __inline__ int r300_emit_3d_load_vbpntr(drm_radeon_private_t *dev_priv,
 
 	return 0;
 }
+static __inline__ int r300_emit_bitblt_multi(drm_radeon_private_t *dev_priv,
+					     drm_radeon_kcmd_buffer_t *cmdbuf)
+{
+	u32 *cmd = (u32 *) cmdbuf->buf;
+	int count, ret;
+	RING_LOCALS;
+
+	count=(cmd[0]>>16) & 0x3fff;
+
+	if (cmd[0] & 0x8000) {
+		u32 offset;
+
+		if (cmd[1] & (RADEON_GMC_SRC_PITCH_OFFSET_CNTL 
+			      | RADEON_GMC_DST_PITCH_OFFSET_CNTL)) {
+			offset = cmd[2] << 10;
+			ret = r300_check_offset(dev_priv, offset);
+			if (ret)
+			{
+				DRM_ERROR("Invalid bitblt first offset is %08X\n", offset);
+				return DRM_ERR(EINVAL);
+			}
+		}
+
+		if ((cmd[1] & RADEON_GMC_SRC_PITCH_OFFSET_CNTL) &&
+		    (cmd[1] & RADEON_GMC_DST_PITCH_OFFSET_CNTL)) {
+			offset = cmd[3] << 10;
+			ret = r300_check_offset(dev_priv, offset);
+			if (ret)
+			{
+				DRM_ERROR("Invalid bitblt second offset is %08X\n", offset);
+				return DRM_ERR(EINVAL);
+			}
+			
+		}
+	}
+
+	BEGIN_RING(count+2);
+	OUT_RING(cmd[0]);
+	OUT_RING_TABLE((int *)(cmdbuf->buf + 4), count + 1);
+	ADVANCE_RING();
+
+	cmdbuf->buf += (count+2)*4;
+	cmdbuf->bufsz -= (count+2)*4;
+
+	return 0;
+}
 
 static __inline__ int r300_emit_raw_packet3(drm_radeon_private_t *dev_priv,
 					    drm_radeon_kcmd_buffer_t *cmdbuf)
@@ -527,6 +574,9 @@ static __inline__ int r300_emit_raw_packet3(drm_radeon_private_t *dev_priv,
 	case RADEON_3D_LOAD_VBPNTR:	/* load vertex array pointers */
 		return r300_emit_3d_load_vbpntr(dev_priv, cmdbuf, header);
 
+	case RADEON_CNTL_BITBLT_MULTI:
+		return r300_emit_bitblt_multi(dev_priv, cmdbuf);
+
 	case RADEON_CP_3D_DRAW_IMMD_2:	/* triggers drawing using in-packet vertex data */
 	case RADEON_CP_3D_DRAW_VBUF_2:	/* triggers drawing of vertex buffers setup elsewhere */
 	case RADEON_CP_3D_DRAW_INDX_2:	/* triggers drawing using indices to vertex buffer */
diff --git a/drivers/char/drm/r300_reg.h b/drivers/char/drm/r300_reg.h
index a0ed20e25221..d1e19954406b 100644
--- a/drivers/char/drm/r300_reg.h
+++ b/drivers/char/drm/r300_reg.h
@@ -451,6 +451,9 @@ I am fairly certain that they are correct unless stated otherwise in comments.
 /* END */
 
 /* gap */
+/* Zero to flush caches. */
+#define R300_TX_CNTL                        0x4100
+
 /* The upper enable bits are guessed, based on fglrx reported limits. */
 #define R300_TX_ENABLE                      0x4104
 #       define R300_TX_ENABLE_0                  (1 << 0)
diff --git a/drivers/char/drm/radeon_drv.h b/drivers/char/drm/radeon_drv.h
index 498b19b1d641..1f7d2ab8c4fc 100644
--- a/drivers/char/drm/radeon_drv.h
+++ b/drivers/char/drm/radeon_drv.h
@@ -90,9 +90,10 @@
  * 1.19- Add support for gart table in FB memory and PCIE r300
  * 1.20- Add support for r300 texrect
  * 1.21- Add support for card type getparam
+ * 1.22- Add support for texture cache flushes (R300_TX_CNTL)
  */
 #define DRIVER_MAJOR		1
-#define DRIVER_MINOR		21
+#define DRIVER_MINOR		22
 #define DRIVER_PATCHLEVEL	0
 
 /*

From 73d72cffe53407e447df0cbb0bf15a2c931108b3 Mon Sep 17 00:00:00 2001
From: Dave Airlie <Ben Herrenschmidt benh@kernel.crashing.org>
Date: Sat, 18 Feb 2006 16:30:54 +1100
Subject: [PATCH 301/509] drm: fix brace placement

Signed-off-by: Dave Airlie <airlied@linux.ie>
---
 drivers/char/drm/r300_cmdbuf.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/char/drm/r300_cmdbuf.c b/drivers/char/drm/r300_cmdbuf.c
index 6dd21754ba67..c08fa5076f05 100644
--- a/drivers/char/drm/r300_cmdbuf.c
+++ b/drivers/char/drm/r300_cmdbuf.c
@@ -506,8 +506,7 @@ static __inline__ int r300_emit_bitblt_multi(drm_radeon_private_t *dev_priv,
 			      | RADEON_GMC_DST_PITCH_OFFSET_CNTL)) {
 			offset = cmd[2] << 10;
 			ret = r300_check_offset(dev_priv, offset);
-			if (ret)
-			{
+			if (ret) {
 				DRM_ERROR("Invalid bitblt first offset is %08X\n", offset);
 				return DRM_ERR(EINVAL);
 			}
@@ -517,8 +516,7 @@ static __inline__ int r300_emit_bitblt_multi(drm_radeon_private_t *dev_priv,
 		    (cmd[1] & RADEON_GMC_DST_PITCH_OFFSET_CNTL)) {
 			offset = cmd[3] << 10;
 			ret = r300_check_offset(dev_priv, offset);
-			if (ret)
-			{
+			if (ret) {
 				DRM_ERROR("Invalid bitblt second offset is %08X\n", offset);
 				return DRM_ERR(EINVAL);
 			}

From ef20c8c197df9b8d5bd4af0679123826da028861 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 18 Feb 2006 15:41:50 -0500
Subject: [PATCH 302/509] [PATCH] GFP_KERNEL allocations in atomic (auditsc)

audit_log_exit() is called from atomic contexts and gets explicit
gfp_mask argument; it should use it for all allocations rather
than doing some with gfp_mask and some with GFP_KERNEL.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 kernel/auditsc.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 685c25175d96..d7e7e637b92a 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -841,7 +841,7 @@ static void audit_log_exit(struct audit_context *context, gfp_t gfp_mask)
 
 	for (aux = context->aux; aux; aux = aux->next) {
 
-		ab = audit_log_start(context, GFP_KERNEL, aux->type);
+		ab = audit_log_start(context, gfp_mask, aux->type);
 		if (!ab)
 			continue; /* audit_panic has been called */
 
@@ -878,14 +878,14 @@ static void audit_log_exit(struct audit_context *context, gfp_t gfp_mask)
 	}
 
 	if (context->pwd && context->pwdmnt) {
-		ab = audit_log_start(context, GFP_KERNEL, AUDIT_CWD);
+		ab = audit_log_start(context, gfp_mask, AUDIT_CWD);
 		if (ab) {
 			audit_log_d_path(ab, "cwd=", context->pwd, context->pwdmnt);
 			audit_log_end(ab);
 		}
 	}
 	for (i = 0; i < context->name_count; i++) {
-		ab = audit_log_start(context, GFP_KERNEL, AUDIT_PATH);
+		ab = audit_log_start(context, gfp_mask, AUDIT_PATH);
 		if (!ab)
 			continue; /* audit_panic has been called */
 

From e30809fde59d591809f00caa1a4c960cca5916af Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 8 Feb 2006 14:09:00 -0500
Subject: [PATCH 303/509] [PATCH] don't mangle INQUIRY if cmddt or evpd bits
 are set

sbp2.c mangles INQUIRY response in a way that only applies to standard
inquiry data (i.e. when both cmddt and evpd bits are 0).  Leave other cases
alone; e.g. when asking for VPD the length of reply is in byte 3, not 4
and byte 4 is the first byte of device serial number.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 drivers/ieee1394/sbp2.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c
index 18d7eda38851..c2c776fbda01 100644
--- a/drivers/ieee1394/sbp2.c
+++ b/drivers/ieee1394/sbp2.c
@@ -2082,9 +2082,7 @@ static void sbp2_check_sbp2_response(struct scsi_id_instance_data *scsi_id,
 
 	SBP2_DEBUG("sbp2_check_sbp2_response");
 
-	switch (SCpnt->cmnd[0]) {
-
-	case INQUIRY:
+	if (SCpnt->cmnd[0] == INQUIRY && (SCpnt->cmnd[1] & 3) == 0) {
 		/*
 		 * Make sure data length is ok. Minimum length is 36 bytes
 		 */
@@ -2097,13 +2095,7 @@ static void sbp2_check_sbp2_response(struct scsi_id_instance_data *scsi_id,
 		 */
 		scsi_buf[2] |= 2;
 		scsi_buf[3] = (scsi_buf[3] & 0xf0) | 2;
-
-		break;
-
-	default:
-		break;
 	}
-	return;
 }
 
 /*

From 76b6159ba094544e003a237cedcf555d82fa3bfe Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 8 Feb 2006 14:37:40 -0500
Subject: [PATCH 304/509] [PATCH] fix handling of st_nlink on procfs root

1) it should use nr_processes(), not nr_threads; otherwise we are getting
very confused find(1) and friends, among other things.
2) better do that at stat() time than at every damn lookup in procfs root.

Patch had been sitting in FC4 kernels for many months now...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/proc/inode.c |  4 ----
 fs/proc/root.c  | 17 +++++++++--------
 2 files changed, 9 insertions(+), 12 deletions(-)

diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 6573f31f1fd9..075d3e945602 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -204,10 +204,6 @@ int proc_fill_super(struct super_block *s, void *data, int silent)
 	root_inode = proc_get_inode(s, PROC_ROOT_INO, &proc_root);
 	if (!root_inode)
 		goto out_no_root;
-	/*
-	 * Fixup the root inode's nlink value
-	 */
-	root_inode->i_nlink += nr_processes();
 	root_inode->i_uid = 0;
 	root_inode->i_gid = 0;
 	s->s_root = d_alloc_root(root_inode);
diff --git a/fs/proc/root.c b/fs/proc/root.c
index 68896283c8ae..c3fd3611112f 100644
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -80,16 +80,16 @@ void __init proc_root_init(void)
 	proc_bus = proc_mkdir("bus", NULL);
 }
 
+static int proc_root_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat
+)
+{
+	generic_fillattr(dentry->d_inode, stat);
+	stat->nlink = proc_root.nlink + nr_processes();
+	return 0;
+}
+
 static struct dentry *proc_root_lookup(struct inode * dir, struct dentry * dentry, struct nameidata *nd)
 {
-	/*
-	 * nr_threads is actually protected by the tasklist_lock;
-	 * however, it's conventional to do reads, especially for
-	 * reporting, without any locking whatsoever.
-	 */
-	if (dir->i_ino == PROC_ROOT_INO) /* check for safety... */
-		dir->i_nlink = proc_root.nlink + nr_threads;
-
 	if (!proc_lookup(dir, dentry, nd)) {
 		return NULL;
 	}
@@ -134,6 +134,7 @@ static struct file_operations proc_root_operations = {
  */
 static struct inode_operations proc_root_inode_operations = {
 	.lookup		= proc_root_lookup,
+	.getattr	= proc_root_getattr,
 };
 
 /*

From 00fc00df9e7b637cd13fe1f163da0a2957273947 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 18 Jan 2006 21:22:55 -0500
Subject: [PATCH 305/509] [PATCH] m68k: restore disable_irq_nosync()

Patch claiming to remove enable_irq_nosync() had left it alive but killed
disable_irq_nosync() instead...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/asm-m68k/irq.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-m68k/irq.h b/include/asm-m68k/irq.h
index 325c86f8512d..9ac047c400c4 100644
--- a/include/asm-m68k/irq.h
+++ b/include/asm-m68k/irq.h
@@ -79,7 +79,7 @@ static __inline__ int irq_canonicalize(int irq)
 
 extern void (*enable_irq)(unsigned int);
 extern void (*disable_irq)(unsigned int);
-#define enable_irq_nosync	enable_irq
+#define disable_irq_nosync	disable_irq
 
 struct pt_regs;
 

From cc6cdac0cf11955dc81d6c50b6738d05e1dca12b Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 18 Feb 2006 16:02:18 -0500
Subject: [PATCH 306/509] [PATCH] missing ntohs() in ip6_tunnel

->payload_len is net-endian

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 net/ipv6/ip6_tunnel.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 92ead3cf956b..faea8a120ee2 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -458,7 +458,7 @@ ip6ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 			mtu = IPV6_MIN_MTU;
 		t->dev->mtu = mtu;
 
-		if ((len = sizeof (*ipv6h) + ipv6h->payload_len) > mtu) {
+		if ((len = sizeof (*ipv6h) + ntohs(ipv6h->payload_len)) > mtu) {
 			rel_type = ICMPV6_PKT_TOOBIG;
 			rel_code = 0;
 			rel_info = mtu;

From cead14da59fc261534fa749886c12c16757711fd Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 18 Jan 2006 19:41:35 -0500
Subject: [PATCH 307/509] [PATCH] m68k: pm_power_off() breakage

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 arch/m68k/kernel/process.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/m68k/kernel/process.c b/arch/m68k/kernel/process.c
index 3f9cb55d0356..2d8ad0727b6b 100644
--- a/arch/m68k/kernel/process.c
+++ b/arch/m68k/kernel/process.c
@@ -129,6 +129,9 @@ void machine_power_off(void)
 	for (;;);
 }
 
+void (*pm_power_off)(void) = machine_power_off;
+EXPORT_SYMBOL(pm_power_off);
+
 void show_regs(struct pt_regs * regs)
 {
 	printk("\n");

From ad6b97fc929e5844bfd1d708ab1d74d131d7960d Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Fri, 3 Feb 2006 02:06:42 -0500
Subject: [PATCH 308/509] [PATCH] iomap_copy fallout (m68k)

added __raw_writel(), sanitized include order in iomap_copy.c

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/asm-m68k/raw_io.h | 1 +
 lib/iomap_copy.c          | 2 +-
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/asm-m68k/raw_io.h b/include/asm-m68k/raw_io.h
index 5439bcaa57c6..811ccd25d4a6 100644
--- a/include/asm-m68k/raw_io.h
+++ b/include/asm-m68k/raw_io.h
@@ -336,6 +336,7 @@ static inline void raw_outsw_swapw(volatile u16 __iomem *port, const u16 *buf,
 		: "d0", "a0", "a1", "d6");
 }
 
+#define __raw_writel raw_outl
 
 #endif /* __KERNEL__ */
 
diff --git a/lib/iomap_copy.c b/lib/iomap_copy.c
index a6b1e271d53c..351045f4f63c 100644
--- a/lib/iomap_copy.c
+++ b/lib/iomap_copy.c
@@ -15,8 +15,8 @@
  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA.
  */
 
-#include <linux/io.h>
 #include <linux/module.h>
+#include <linux/io.h>
 
 /**
  * __iowrite32_copy - copy data to MMIO space, in 32-bit units

From 092b8f3488a3e50a4ab5f2f3f7c8bbf56b3144e1 Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@samba.org>
Date: Mon, 20 Feb 2006 10:38:56 +1100
Subject: [PATCH 309/509] powerpc: Keep xtime and gettimeofday in sync

This fixes a regression which was introduced by moving ppc32 to use
the same sort of lockless gettimeofday as ppc64 has been using for
some time.  This involves getting the timebase and performing some
simple arithmetic to convert it to seconds and microseconds.  However,
the factor and offset used there weren't being updated when NTP
varied the tick length using adjtimex.  64-bit didn't notice the
problem because it had a hook in the 32-bit adjtimex compat routine
that attempted to work out what the generic timekeeping code would
do and alter the factor and offset to match.  However, that code
was very complex and it wasn't clear that it still matched what the
generic code would do.

Now we use the generic current_tick_length() routine that was recently
added to check that the current tick will be as long as we expect; if
not we recompute the factor and offset.  This keeps gettimeofday and
xtime in sync.  In addition we check that gettimeofday hasn't got ahead
of xtime on each timer interrupt; if it has, we resync.

Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/sys_ppc32.c |   4 -
 arch/powerpc/kernel/time.c      | 282 +++++++++++---------------------
 2 files changed, 99 insertions(+), 187 deletions(-)

diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c
index 475249dc2350..cd75ab2908fa 100644
--- a/arch/powerpc/kernel/sys_ppc32.c
+++ b/arch/powerpc/kernel/sys_ppc32.c
@@ -176,7 +176,6 @@ struct timex32 {
 };
 
 extern int do_adjtimex(struct timex *);
-extern void ppc_adjtimex(void);
 
 asmlinkage long compat_sys_adjtimex(struct timex32 __user *utp)
 {
@@ -209,9 +208,6 @@ asmlinkage long compat_sys_adjtimex(struct timex32 __user *utp)
 
 	ret = do_adjtimex(&txc);
 
-	/* adjust the conversion of TB to time of day to track adjtimex */
-	ppc_adjtimex();
-
 	if(put_user(txc.modes, &utp->modes) ||
 	   __put_user(txc.offset, &utp->offset) ||
 	   __put_user(txc.freq, &utp->freq) ||
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 1886045a2fd8..2a7ddc579379 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -50,6 +50,7 @@
 #include <linux/security.h>
 #include <linux/percpu.h>
 #include <linux/rtc.h>
+#include <linux/jiffies.h>
 
 #include <asm/io.h>
 #include <asm/processor.h>
@@ -99,7 +100,15 @@ EXPORT_SYMBOL(tb_ticks_per_usec);
 unsigned long tb_ticks_per_sec;
 u64 tb_to_xs;
 unsigned tb_to_us;
-unsigned long processor_freq;
+
+#define TICKLEN_SCALE	(SHIFT_SCALE - 10)
+u64 last_tick_len;	/* units are ns / 2^TICKLEN_SCALE */
+u64 ticklen_to_xs;	/* 0.64 fraction */
+
+/* If last_tick_len corresponds to about 1/HZ seconds, then
+   last_tick_len << TICKLEN_SHIFT will be about 2^63. */
+#define TICKLEN_SHIFT	(63 - 30 - TICKLEN_SCALE + SHIFT_HZ)
+
 DEFINE_SPINLOCK(rtc_lock);
 EXPORT_SYMBOL_GPL(rtc_lock);
 
@@ -113,10 +122,6 @@ extern unsigned long wall_jiffies;
 extern struct timezone sys_tz;
 static long timezone_offset;
 
-void ppc_adjtimex(void);
-
-static unsigned adjusting_time = 0;
-
 unsigned long ppc_proc_freq;
 unsigned long ppc_tb_freq;
 
@@ -178,8 +183,7 @@ static __inline__ void timer_check_rtc(void)
          */
         if (ppc_md.set_rtc_time && ntp_synced() &&
 	    xtime.tv_sec - last_rtc_update >= 659 &&
-	    abs((xtime.tv_nsec/1000) - (1000000-1000000/HZ)) < 500000/HZ &&
-	    jiffies - wall_jiffies == 1) {
+	    abs((xtime.tv_nsec/1000) - (1000000-1000000/HZ)) < 500000/HZ) {
 		struct rtc_time tm;
 		to_tm(xtime.tv_sec + 1 + timezone_offset, &tm);
 		tm.tm_year -= 1900;
@@ -226,15 +230,14 @@ void do_gettimeofday(struct timeval *tv)
 	if (__USE_RTC()) {
 		/* do this the old way */
 		unsigned long flags, seq;
-		unsigned int sec, nsec, usec, lost;
+		unsigned int sec, nsec, usec;
 
 		do {
 			seq = read_seqbegin_irqsave(&xtime_lock, flags);
 			sec = xtime.tv_sec;
 			nsec = xtime.tv_nsec + tb_ticks_since(tb_last_stamp);
-			lost = jiffies - wall_jiffies;
 		} while (read_seqretry_irqrestore(&xtime_lock, seq, flags));
-		usec = nsec / 1000 + lost * (1000000 / HZ);
+		usec = nsec / 1000;
 		while (usec >= 1000000) {
 			usec -= 1000000;
 			++sec;
@@ -248,23 +251,6 @@ void do_gettimeofday(struct timeval *tv)
 
 EXPORT_SYMBOL(do_gettimeofday);
 
-/* Synchronize xtime with do_gettimeofday */ 
-
-static inline void timer_sync_xtime(unsigned long cur_tb)
-{
-#ifdef CONFIG_PPC64
-	/* why do we do this? */
-	struct timeval my_tv;
-
-	__do_gettimeofday(&my_tv, cur_tb);
-
-	if (xtime.tv_sec <= my_tv.tv_sec) {
-		xtime.tv_sec = my_tv.tv_sec;
-		xtime.tv_nsec = my_tv.tv_usec * 1000;
-	}
-#endif
-}
-
 /*
  * There are two copies of tb_to_xs and stamp_xsec so that no
  * lock is needed to access and use these values in
@@ -323,15 +309,30 @@ static __inline__ void timer_recalc_offset(u64 cur_tb)
 {
 	unsigned long offset;
 	u64 new_stamp_xsec;
+	u64 tlen, t2x;
 
 	if (__USE_RTC())
 		return;
+	tlen = current_tick_length();
 	offset = cur_tb - do_gtod.varp->tb_orig_stamp;
-	if ((offset & 0x80000000u) == 0)
-		return;
-	new_stamp_xsec = do_gtod.varp->stamp_xsec
-		+ mulhdu(offset, do_gtod.varp->tb_to_xs);
-	update_gtod(cur_tb, new_stamp_xsec, do_gtod.varp->tb_to_xs);
+	if (tlen == last_tick_len && offset < 0x80000000u) {
+		/* check that we're still in sync; if not, resync */
+		struct timeval tv;
+		__do_gettimeofday(&tv, cur_tb);
+		if (tv.tv_sec <= xtime.tv_sec &&
+		    (tv.tv_sec < xtime.tv_sec ||
+		     tv.tv_usec * 1000 <= xtime.tv_nsec))
+			return;
+	}
+	if (tlen != last_tick_len) {
+		t2x = mulhdu(tlen << TICKLEN_SHIFT, ticklen_to_xs);
+		last_tick_len = tlen;
+	} else
+		t2x = do_gtod.varp->tb_to_xs;
+	new_stamp_xsec = (u64) xtime.tv_nsec * XSEC_PER_SEC;
+	do_div(new_stamp_xsec, 1000000000);
+	new_stamp_xsec += (u64) xtime.tv_sec * XSEC_PER_SEC;
+	update_gtod(cur_tb, new_stamp_xsec, t2x);
 }
 
 #ifdef CONFIG_SMP
@@ -462,13 +463,10 @@ void timer_interrupt(struct pt_regs * regs)
 		write_seqlock(&xtime_lock);
 		tb_last_jiffy += tb_ticks_per_jiffy;
 		tb_last_stamp = per_cpu(last_jiffy, cpu);
-		timer_recalc_offset(tb_last_jiffy);
 		do_timer(regs);
-		timer_sync_xtime(tb_last_jiffy);
+		timer_recalc_offset(tb_last_jiffy);
 		timer_check_rtc();
 		write_sequnlock(&xtime_lock);
-		if (adjusting_time && (time_adjust == 0))
-			ppc_adjtimex();
 	}
 	
 	next_dec = tb_ticks_per_jiffy - ticks;
@@ -492,16 +490,18 @@ void timer_interrupt(struct pt_regs * regs)
 
 void wakeup_decrementer(void)
 {
-	int i;
+	unsigned long ticks;
 
-	set_dec(tb_ticks_per_jiffy);
 	/*
-	 * We don't expect this to be called on a machine with a 601,
-	 * so using get_tbl is fine.
+	 * The timebase gets saved on sleep and restored on wakeup,
+	 * so all we need to do is to reset the decrementer.
 	 */
-	tb_last_stamp = tb_last_jiffy = get_tb();
-	for_each_cpu(i)
-		per_cpu(last_jiffy, i) = tb_last_stamp;
+	ticks = tb_ticks_since(__get_cpu_var(last_jiffy));
+	if (ticks < tb_ticks_per_jiffy)
+		ticks = tb_ticks_per_jiffy - ticks;
+	else
+		ticks = 1;
+	set_dec(ticks);
 }
 
 #ifdef CONFIG_SMP
@@ -541,8 +541,8 @@ int do_settimeofday(struct timespec *tv)
 	time_t wtm_sec, new_sec = tv->tv_sec;
 	long wtm_nsec, new_nsec = tv->tv_nsec;
 	unsigned long flags;
-	long int tb_delta;
-	u64 new_xsec, tb_delta_xs;
+	u64 new_xsec;
+	unsigned long tb_delta;
 
 	if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
 		return -EINVAL;
@@ -563,9 +563,19 @@ int do_settimeofday(struct timespec *tv)
 		first_settimeofday = 0;
 	}
 #endif
+
+	/*
+	 * Subtract off the number of nanoseconds since the
+	 * beginning of the last tick.
+	 * Note that since we don't increment jiffies_64 anywhere other
+	 * than in do_timer (since we don't have a lost tick problem),
+	 * wall_jiffies will always be the same as jiffies,
+	 * and therefore the (jiffies - wall_jiffies) computation
+	 * has been removed.
+	 */
 	tb_delta = tb_ticks_since(tb_last_stamp);
-	tb_delta += (jiffies - wall_jiffies) * tb_ticks_per_jiffy;
-	tb_delta_xs = mulhdu(tb_delta, do_gtod.varp->tb_to_xs);
+	tb_delta = mulhdu(tb_delta, do_gtod.varp->tb_to_xs); /* in xsec */
+	new_nsec -= SCALE_XSEC(tb_delta, 1000000000);
 
 	wtm_sec  = wall_to_monotonic.tv_sec + (xtime.tv_sec - new_sec);
 	wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - new_nsec);
@@ -580,12 +590,12 @@ int do_settimeofday(struct timespec *tv)
 
 	ntp_clear();
 
-	new_xsec = 0;
-	if (new_nsec != 0) {
-		new_xsec = (u64)new_nsec * XSEC_PER_SEC;
+	new_xsec = xtime.tv_nsec;
+	if (new_xsec != 0) {
+		new_xsec *= XSEC_PER_SEC;
 		do_div(new_xsec, NSEC_PER_SEC);
 	}
-	new_xsec += (u64)new_sec * XSEC_PER_SEC - tb_delta_xs;
+	new_xsec += (u64)xtime.tv_sec * XSEC_PER_SEC;
 	update_gtod(tb_last_jiffy, new_xsec, do_gtod.varp->tb_to_xs);
 
 	vdso_data->tz_minuteswest = sys_tz.tz_minuteswest;
@@ -671,7 +681,7 @@ void __init time_init(void)
 	unsigned long flags;
 	unsigned long tm = 0;
 	struct div_result res;
-	u64 scale;
+	u64 scale, x;
 	unsigned shift;
 
         if (ppc_md.time_init != NULL)
@@ -693,11 +703,36 @@ void __init time_init(void)
 	}
 
 	tb_ticks_per_jiffy = ppc_tb_freq / HZ;
-	tb_ticks_per_sec = tb_ticks_per_jiffy * HZ;
+	tb_ticks_per_sec = ppc_tb_freq;
 	tb_ticks_per_usec = ppc_tb_freq / 1000000;
 	tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000);
-	div128_by_32(1024*1024, 0, tb_ticks_per_sec, &res);
-	tb_to_xs = res.result_low;
+
+	/*
+	 * Calculate the length of each tick in ns.  It will not be
+	 * exactly 1e9/HZ unless ppc_tb_freq is divisible by HZ.
+	 * We compute 1e9 * tb_ticks_per_jiffy / ppc_tb_freq,
+	 * rounded up.
+	 */
+	x = (u64) NSEC_PER_SEC * tb_ticks_per_jiffy + ppc_tb_freq - 1;
+	do_div(x, ppc_tb_freq);
+	tick_nsec = x;
+	last_tick_len = x << TICKLEN_SCALE;
+
+	/*
+	 * Compute ticklen_to_xs, which is a factor which gets multiplied
+	 * by (last_tick_len << TICKLEN_SHIFT) to get a tb_to_xs value.
+	 * It is computed as:
+	 * ticklen_to_xs = 2^N / (tb_ticks_per_jiffy * 1e9)
+	 * where N = 64 + 20 - TICKLEN_SCALE - TICKLEN_SHIFT
+	 * so as to give the result as a 0.64 fixed-point fraction.
+	 */
+	div128_by_32(1ULL << (64 + 20 - TICKLEN_SCALE - TICKLEN_SHIFT), 0,
+		     tb_ticks_per_jiffy, &res);
+	div128_by_32(res.result_high, res.result_low, NSEC_PER_SEC, &res);
+	ticklen_to_xs = res.result_low;
+
+	/* Compute tb_to_xs from tick_nsec */
+	tb_to_xs = mulhdu(last_tick_len << TICKLEN_SHIFT, ticklen_to_xs);
 
 	/*
 	 * Compute scale factor for sched_clock.
@@ -724,6 +759,14 @@ void __init time_init(void)
 		tm = get_boot_time();
 
 	write_seqlock_irqsave(&xtime_lock, flags);
+
+	/* If platform provided a timezone (pmac), we correct the time */
+        if (timezone_offset) {
+		sys_tz.tz_minuteswest = -timezone_offset / 60;
+		sys_tz.tz_dsttime = 0;
+		tm -= timezone_offset;
+        }
+
 	xtime.tv_sec = tm;
 	xtime.tv_nsec = 0;
 	do_gtod.varp = &do_gtod.vars[0];
@@ -738,18 +781,11 @@ void __init time_init(void)
 	vdso_data->tb_orig_stamp = tb_last_jiffy;
 	vdso_data->tb_update_count = 0;
 	vdso_data->tb_ticks_per_sec = tb_ticks_per_sec;
-	vdso_data->stamp_xsec = xtime.tv_sec * XSEC_PER_SEC;
+	vdso_data->stamp_xsec = (u64) xtime.tv_sec * XSEC_PER_SEC;
 	vdso_data->tb_to_xs = tb_to_xs;
 
 	time_freq = 0;
 
-	/* If platform provided a timezone (pmac), we correct the time */
-        if (timezone_offset) {
-		sys_tz.tz_minuteswest = -timezone_offset / 60;
-		sys_tz.tz_dsttime = 0;
-		xtime.tv_sec -= timezone_offset;
-        }
-
 	last_rtc_update = xtime.tv_sec;
 	set_normalized_timespec(&wall_to_monotonic,
 	                        -xtime.tv_sec, -xtime.tv_nsec);
@@ -759,126 +795,6 @@ void __init time_init(void)
 	set_dec(tb_ticks_per_jiffy);
 }
 
-/* 
- * After adjtimex is called, adjust the conversion of tb ticks
- * to microseconds to keep do_gettimeofday synchronized 
- * with ntpd.
- *
- * Use the time_adjust, time_freq and time_offset computed by adjtimex to 
- * adjust the frequency.
- */
-
-/* #define DEBUG_PPC_ADJTIMEX 1 */
-
-void ppc_adjtimex(void)
-{
-#ifdef CONFIG_PPC64
-	unsigned long den, new_tb_ticks_per_sec, tb_ticks, old_xsec,
-		new_tb_to_xs, new_xsec, new_stamp_xsec;
-	unsigned long tb_ticks_per_sec_delta;
-	long delta_freq, ltemp;
-	struct div_result divres; 
-	unsigned long flags;
-	long singleshot_ppm = 0;
-
-	/*
-	 * Compute parts per million frequency adjustment to
-	 * accomplish the time adjustment implied by time_offset to be
-	 * applied over the elapsed time indicated by time_constant.
-	 * Use SHIFT_USEC to get it into the same units as
-	 * time_freq.
-	 */
-	if ( time_offset < 0 ) {
-		ltemp = -time_offset;
-		ltemp <<= SHIFT_USEC - SHIFT_UPDATE;
-		ltemp >>= SHIFT_KG + time_constant;
-		ltemp = -ltemp;
-	} else {
-		ltemp = time_offset;
-		ltemp <<= SHIFT_USEC - SHIFT_UPDATE;
-		ltemp >>= SHIFT_KG + time_constant;
-	}
-	
-	/* If there is a single shot time adjustment in progress */
-	if ( time_adjust ) {
-#ifdef DEBUG_PPC_ADJTIMEX
-		printk("ppc_adjtimex: ");
-		if ( adjusting_time == 0 )
-			printk("starting ");
-		printk("single shot time_adjust = %ld\n", time_adjust);
-#endif	
-	
-		adjusting_time = 1;
-		
-		/*
-		 * Compute parts per million frequency adjustment
-		 * to match time_adjust
-		 */
-		singleshot_ppm = tickadj * HZ;	
-		/*
-		 * The adjustment should be tickadj*HZ to match the code in
-		 * linux/kernel/timer.c, but experiments show that this is too
-		 * large. 3/4 of tickadj*HZ seems about right
-		 */
-		singleshot_ppm -= singleshot_ppm / 4;
-		/* Use SHIFT_USEC to get it into the same units as time_freq */
-		singleshot_ppm <<= SHIFT_USEC;
-		if ( time_adjust < 0 )
-			singleshot_ppm = -singleshot_ppm;
-	}
-	else {
-#ifdef DEBUG_PPC_ADJTIMEX
-		if ( adjusting_time )
-			printk("ppc_adjtimex: ending single shot time_adjust\n");
-#endif
-		adjusting_time = 0;
-	}
-	
-	/* Add up all of the frequency adjustments */
-	delta_freq = time_freq + ltemp + singleshot_ppm;
-	
-	/*
-	 * Compute a new value for tb_ticks_per_sec based on
-	 * the frequency adjustment
-	 */
-	den = 1000000 * (1 << (SHIFT_USEC - 8));
-	if ( delta_freq < 0 ) {
-		tb_ticks_per_sec_delta = ( tb_ticks_per_sec * ( (-delta_freq) >> (SHIFT_USEC - 8))) / den;
-		new_tb_ticks_per_sec = tb_ticks_per_sec + tb_ticks_per_sec_delta;
-	}
-	else {
-		tb_ticks_per_sec_delta = ( tb_ticks_per_sec * ( delta_freq >> (SHIFT_USEC - 8))) / den;
-		new_tb_ticks_per_sec = tb_ticks_per_sec - tb_ticks_per_sec_delta;
-	}
-	
-#ifdef DEBUG_PPC_ADJTIMEX
-	printk("ppc_adjtimex: ltemp = %ld, time_freq = %ld, singleshot_ppm = %ld\n", ltemp, time_freq, singleshot_ppm);
-	printk("ppc_adjtimex: tb_ticks_per_sec - base = %ld  new = %ld\n", tb_ticks_per_sec, new_tb_ticks_per_sec);
-#endif
-
-	/*
-	 * Compute a new value of tb_to_xs (used to convert tb to
-	 * microseconds) and a new value of stamp_xsec which is the
-	 * time (in 1/2^20 second units) corresponding to
-	 * tb_orig_stamp.  This new value of stamp_xsec compensates
-	 * for the change in frequency (implied by the new tb_to_xs)
-	 * which guarantees that the current time remains the same.
-	 */
-	write_seqlock_irqsave( &xtime_lock, flags );
-	tb_ticks = get_tb() - do_gtod.varp->tb_orig_stamp;
-	div128_by_32(1024*1024, 0, new_tb_ticks_per_sec, &divres);
-	new_tb_to_xs = divres.result_low;
-	new_xsec = mulhdu(tb_ticks, new_tb_to_xs);
-
-	old_xsec = mulhdu(tb_ticks, do_gtod.varp->tb_to_xs);
-	new_stamp_xsec = do_gtod.varp->stamp_xsec + old_xsec - new_xsec;
-
-	update_gtod(do_gtod.varp->tb_orig_stamp, new_stamp_xsec, new_tb_to_xs);
-
-	write_sequnlock_irqrestore( &xtime_lock, flags );
-#endif /* CONFIG_PPC64 */
-}
-
 
 #define FEBRUARY	2
 #define	STARTOFTIME	1970

From 0728a2f99ef6efd1984f9e0ed59834c1cc602e6f Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Sat, 11 Feb 2006 18:21:47 +0100
Subject: [PATCH 310/509] [PATCH] powerpc: remove duplicate exports

A few symbols are exported twice, remove them from ppc_ksyms.c
Remove users of sys_ctrler in arch/ppc/

WARNING: vmlinux: duplicate symbol '__delay' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol '__up' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol '__down' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol '__down_interruptible' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'sys_ctrler' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'strncat' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'strncmp' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'strchr' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'strrchr' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'strnlen' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'strpbrk' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'memscan' previous definition was in vmlinux
WARNING: vmlinux: duplicate symbol 'strstr' previous definition was in vmlinux

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/ppc_ksyms.c | 16 ----------------
 arch/ppc/kernel/ppc_ksyms.c     |  8 --------
 arch/ppc/xmon/start.c           | 15 +--------------
 include/asm-ppc/machdep.h       | 13 -------------
 4 files changed, 1 insertion(+), 51 deletions(-)

diff --git a/arch/powerpc/kernel/ppc_ksyms.c b/arch/powerpc/kernel/ppc_ksyms.c
index d9a459c144d8..8a731ea877b7 100644
--- a/arch/powerpc/kernel/ppc_ksyms.c
+++ b/arch/powerpc/kernel/ppc_ksyms.c
@@ -79,15 +79,8 @@ EXPORT_SYMBOL(sys_sigreturn);
 EXPORT_SYMBOL(strcpy);
 EXPORT_SYMBOL(strncpy);
 EXPORT_SYMBOL(strcat);
-EXPORT_SYMBOL(strncat);
-EXPORT_SYMBOL(strchr);
-EXPORT_SYMBOL(strrchr);
-EXPORT_SYMBOL(strpbrk);
-EXPORT_SYMBOL(strstr);
 EXPORT_SYMBOL(strlen);
-EXPORT_SYMBOL(strnlen);
 EXPORT_SYMBOL(strcmp);
-EXPORT_SYMBOL(strncmp);
 EXPORT_SYMBOL(strcasecmp);
 
 EXPORT_SYMBOL(csum_partial);
@@ -185,9 +178,6 @@ EXPORT_SYMBOL(adb_try_handler_change);
 EXPORT_SYMBOL(cuda_request);
 EXPORT_SYMBOL(cuda_poll);
 #endif /* CONFIG_ADB_CUDA */
-#ifdef CONFIG_PPC_PMAC
-EXPORT_SYMBOL(sys_ctrler);
-#endif
 #ifdef CONFIG_VT
 EXPORT_SYMBOL(kd_mksound);
 #endif
@@ -205,7 +195,6 @@ EXPORT_SYMBOL(__lshrdi3);
 EXPORT_SYMBOL(memcpy);
 EXPORT_SYMBOL(memset);
 EXPORT_SYMBOL(memmove);
-EXPORT_SYMBOL(memscan);
 EXPORT_SYMBOL(memcmp);
 EXPORT_SYMBOL(memchr);
 
@@ -214,7 +203,6 @@ EXPORT_SYMBOL(screen_info);
 #endif
 
 #ifdef CONFIG_PPC32
-EXPORT_SYMBOL(__delay);
 EXPORT_SYMBOL(timer_interrupt);
 EXPORT_SYMBOL(irq_desc);
 EXPORT_SYMBOL(tb_ticks_per_jiffy);
@@ -222,10 +210,6 @@ EXPORT_SYMBOL(console_drivers);
 EXPORT_SYMBOL(cacheable_memcpy);
 #endif
 
-EXPORT_SYMBOL(__up);
-EXPORT_SYMBOL(__down);
-EXPORT_SYMBOL(__down_interruptible);
-
 #ifdef  CONFIG_8xx
 EXPORT_SYMBOL(cpm_install_handler);
 EXPORT_SYMBOL(cpm_free_handler);
diff --git a/arch/ppc/kernel/ppc_ksyms.c b/arch/ppc/kernel/ppc_ksyms.c
index 15bd9b448a48..82adb4601348 100644
--- a/arch/ppc/kernel/ppc_ksyms.c
+++ b/arch/ppc/kernel/ppc_ksyms.c
@@ -93,15 +93,8 @@ EXPORT_SYMBOL(test_and_change_bit);
 EXPORT_SYMBOL(strcpy);
 EXPORT_SYMBOL(strncpy);
 EXPORT_SYMBOL(strcat);
-EXPORT_SYMBOL(strncat);
-EXPORT_SYMBOL(strchr);
-EXPORT_SYMBOL(strrchr);
-EXPORT_SYMBOL(strpbrk);
-EXPORT_SYMBOL(strstr);
 EXPORT_SYMBOL(strlen);
-EXPORT_SYMBOL(strnlen);
 EXPORT_SYMBOL(strcmp);
-EXPORT_SYMBOL(strncmp);
 EXPORT_SYMBOL(strcasecmp);
 EXPORT_SYMBOL(__div64_32);
 
@@ -253,7 +246,6 @@ EXPORT_SYMBOL(memcpy);
 EXPORT_SYMBOL(cacheable_memcpy);
 EXPORT_SYMBOL(memset);
 EXPORT_SYMBOL(memmove);
-EXPORT_SYMBOL(memscan);
 EXPORT_SYMBOL(memcmp);
 EXPORT_SYMBOL(memchr);
 
diff --git a/arch/ppc/xmon/start.c b/arch/ppc/xmon/start.c
index 4344cbe9b5c5..484f5bb1aa3e 100644
--- a/arch/ppc/xmon/start.c
+++ b/arch/ppc/xmon/start.c
@@ -146,19 +146,6 @@ xmon_map_scc(void)
 static int scc_initialized = 0;
 
 void xmon_init_scc(void);
-extern void cuda_poll(void);
-
-static inline void do_poll_adb(void)
-{
-#ifdef CONFIG_ADB_PMU
-	if (sys_ctrler == SYS_CTRLER_PMU)
-		pmu_poll_adb();
-#endif /* CONFIG_ADB_PMU */
-#ifdef CONFIG_ADB_CUDA
-	if (sys_ctrler == SYS_CTRLER_CUDA)
-		cuda_poll();
-#endif /* CONFIG_ADB_CUDA */
-}
 
 int
 xmon_write(void *handle, void *ptr, int nb)
@@ -189,7 +176,7 @@ xmon_write(void *handle, void *ptr, int nb)
 	ct = 0;
 	for (i = 0; i < nb; ++i) {
 		while ((*sccc & TXRDY) == 0)
-			do_poll_adb();
+			;
 		c = p[i];
 		if (c == '\n' && !ct) {
 			c = '\r';
diff --git a/include/asm-ppc/machdep.h b/include/asm-ppc/machdep.h
index 39200def8d11..a3e8a45e45a9 100644
--- a/include/asm-ppc/machdep.h
+++ b/include/asm-ppc/machdep.h
@@ -154,19 +154,6 @@ extern char cmd_line[COMMAND_LINE_SIZE];
 
 extern void setup_pci_ptrs(void);
 
-/*
- * Power macintoshes have either a CUDA or a PMU controlling
- * system reset, power, NVRAM, RTC.
- */
-typedef enum sys_ctrler_kind {
-	SYS_CTRLER_UNKNOWN = 0,
-	SYS_CTRLER_CUDA = 1,
-	SYS_CTRLER_PMU = 2,
-	SYS_CTRLER_SMU = 3,
-} sys_ctrler_t;
-
-extern sys_ctrler_t sys_ctrler;
-
 #ifdef CONFIG_SMP
 struct smp_ops_t {
 	void  (*message_pass)(int target, int msg);

From 2b9a32edba3af9ad4ccb23574bea0cc34455dc43 Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Wed, 15 Feb 2006 21:40:44 -0600
Subject: [PATCH 311/509] [PATCH] powerpc: Fix OOPS in lparcfg on G5

Fallback gracefully when reading /proc/ppc64/lparcfg when the /rtas
device node can't be found.

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/lparcfg.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 1ae96a8ed7e2..e789fef4eb8a 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -341,7 +341,7 @@ static int lparcfg_data(struct seq_file *m, void *v)
 	const char *system_id = "";
 	unsigned int *lp_index_ptr, lp_index = 0;
 	struct device_node *rtas_node;
-	int *lrdrp;
+	int *lrdrp = NULL;
 
 	rootdn = find_path_device("/");
 	if (rootdn) {
@@ -362,7 +362,9 @@ static int lparcfg_data(struct seq_file *m, void *v)
 	seq_printf(m, "partition_id=%d\n", (int)lp_index);
 
 	rtas_node = find_path_device("/rtas");
-	lrdrp = (int *)get_property(rtas_node, "ibm,lrdr-capacity", NULL);
+	if (rtas_node)
+		lrdrp = (int *)get_property(rtas_node, "ibm,lrdr-capacity",
+		                            NULL);
 
 	if (lrdrp == NULL) {
 		partition_potential_processors = vdso_data->processorCount;

From f018b36f3e1f21318066de8d01740d30e38b03d5 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 16 Feb 2006 14:13:50 +1100
Subject: [PATCH 312/509] [PATCH] powerpc: Don't start secondary CPUs in a UP
 && KEXEC kernel

Because smp_release_cpus() is built for SMP || KEXEC, it's not safe to
unconditionally call it from setup_system(). On a UP && KEXEC kernel we'll
start up the secondary CPUs which will then go beserk and we die.

Simple fix is to conditionally call smp_release_cpus() in setup_system(). With
that in place we don't need the dummy definition of smp_release_cpus() because
all call sites are #ifdef'ed either SMP or KEXEC.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/setup_64.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index a717dff695ef..f96c49b03ba0 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -311,8 +311,6 @@ void smp_release_cpus(void)
 
 	DBG(" <- smp_release_cpus()\n");
 }
-#else
-#define smp_release_cpus()
 #endif /* CONFIG_SMP || CONFIG_KEXEC */
 
 /*
@@ -473,10 +471,12 @@ void __init setup_system(void)
 	check_smt_enabled();
 	smp_setup_cpu_maps();
 
+#ifdef CONFIG_SMP
 	/* Release secondary cpus out of their spinloops at 0x60 now that
 	 * we can map physical -> logical CPU ids
 	 */
 	smp_release_cpus();
+#endif
 
 	printk("Starting Linux PPC64 %s\n", system_utsname.version);
 

From 8fca92705ef462f39e7db5a0f7100bcaae91bfd3 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 16 Feb 2006 14:13:51 +1100
Subject: [PATCH 313/509] [PATCH] powerpc: Make UP -> SMP kexec work again

For UP to SMP kexec to work we need to jump into pSeries_secondary_smp_init
event on a UP + KEXEC kernel. The secondary cpus will not find their hw_cpu_id
in the paca and so they'll jump into kexec_wait, ready for a kexec.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/head_64.S | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 415659629394..2b03a09fe5e9 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -157,8 +157,7 @@ _GLOBAL(__secondary_hold)
 	SET_REG_IMMEDIATE(r4, .hmt_init)
 	mtctr	r4
 	bctr
-#else
-#ifdef CONFIG_SMP
+#elif defined(CONFIG_SMP) || defined(CONFIG_KEXEC)
 	LOAD_REG_IMMEDIATE(r4, .pSeries_secondary_smp_init)
 	mtctr	r4
 	mr	r3,r24
@@ -166,7 +165,6 @@ _GLOBAL(__secondary_hold)
 #else
 	BUG_OPCODE
 #endif
-#endif
 
 /* This value is used to mark exception frames on the stack. */
 	.section ".toc","aw"

From 496b7a5159b8366b003bbc17f8c4e27f69b6779e Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Thu, 16 Feb 2006 14:13:53 +1100
Subject: [PATCH 314/509] [PATCH] powerpc: Fix bug in spinup of renumbered
 secondary threads

If the logical and physical cpu ids of a secondary thread don't match, we will
fail to spin the thread up on pSeries machines due to a bug in pseries/smp.c

We call the RTAS "start-cpu" method with the physical cpu id, the address of
pSeries_secondary_smp_init and the value to pass that function in r3. Currently
we pass "lcpu", the logical cpu id, but pSeries_secondary_smp_init expects
the physical cpu id in r3.

We should be passing pcpu instead.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/pseries/smp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/platforms/pseries/smp.c b/arch/powerpc/platforms/pseries/smp.c
index 8e6b1ed1396e..8d710af50756 100644
--- a/arch/powerpc/platforms/pseries/smp.c
+++ b/arch/powerpc/platforms/pseries/smp.c
@@ -292,7 +292,7 @@ static inline int __devinit smp_startup_cpu(unsigned int lcpu)
 	if (start_cpu == RTAS_UNKNOWN_SERVICE)
 		return 1;
 
-	status = rtas_call(start_cpu, 3, 1, NULL, pcpu, start_here, lcpu);
+	status = rtas_call(start_cpu, 3, 1, NULL, pcpu, start_here, pcpu);
 	if (status != 0) {
 		printk(KERN_ERR "start-cpu failed: %i\n", status);
 		return 0;

From 995110143880fd9cb255fa5df05f8950c56fb43a Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sun, 19 Feb 2006 22:11:50 -0800
Subject: [PATCH 315/509] [XFRM]: Fix policy double put

The policy is put once immediately and once at the error label, which results
in the following Oops:

kernel BUG at net/xfrm/xfrm_policy.c:250!
invalid opcode: 0000 [#2]
PREEMPT
[...]
CPU:    0
EIP:    0060:[<c028caf7>]    Not tainted VLI
EFLAGS: 00210246   (2.6.16-rc3 #39)
EIP is at __xfrm_policy_destroy+0xf/0x46
eax: d49f2000   ebx: d49f2000   ecx: f74bd880   edx: f74bd280
esi: d49f2000   edi: 00000001   ebp: cd506dcc   esp: cd506dc8
ds: 007b   es: 007b   ss: 0068
Process ssh (pid: 31970, threadinfo=cd506000 task=cfb04a70)
Stack: <0>cd506000 cd506e34 c028e92b ebde7280 cd506e58 cd506ec0 f74bd280 00000000
       00000214 0000000a 0000000a 00000000 00000002 f7ae6000 00000000 cd506e58
       cd506e14 c0299e36 f74bd280 e873fe00 c02943fd cd506ec0 ebde7280 f271f440
Call Trace:
 [<c0103a44>] show_stack_log_lvl+0xaa/0xb5
 [<c0103b75>] show_registers+0x126/0x18c
 [<c0103e68>] die+0x14e/0x1db
 [<c02b6809>] do_trap+0x7c/0x96
 [<c0104237>] do_invalid_op+0x89/0x93
 [<c01035af>] error_code+0x4f/0x54
 [<c028e92b>] xfrm_lookup+0x349/0x3c2
 [<c02b0b0d>] ip6_datagram_connect+0x317/0x452
 [<c0281749>] inet_dgram_connect+0x49/0x54
 [<c02404d2>] sys_connect+0x51/0x68
 [<c0240928>] sys_socketcall+0x6f/0x166
 [<c0102aa1>] syscall_call+0x7/0xb

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/xfrm/xfrm_policy.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 98ec53bd3ac7..5e6b05ac1260 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -885,8 +885,6 @@ restart:
 			 * We can't enlist stable bundles either.
 			 */
 			write_unlock_bh(&policy->lock);
-
-			xfrm_pol_put(policy);
 			if (dst)
 				dst_free(dst);
 

From bc6e14b6f0b06fe93d809d22e257ddd275feeda9 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sun, 19 Feb 2006 22:26:40 -0800
Subject: [PATCH 316/509] [NETFILTER]: Fix NAT PMTUD problems

ICMP errors are only SNATed when their source matches the source of the
connection they are related to, otherwise the source address is not
changed. This creates problems with ICMP frag. required messages
originating from a router behind the NAT, if private IPs are used the
packet has a good change of getting dropped on the path to its destination.

Always NAT ICMP errors similar to the original connection.

Based on report by Al Viro.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/netfilter/ip_nat_core.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/net/ipv4/netfilter/ip_nat_core.c b/net/ipv4/netfilter/ip_nat_core.c
index c1a61462507f..1741d555ad0d 100644
--- a/net/ipv4/netfilter/ip_nat_core.c
+++ b/net/ipv4/netfilter/ip_nat_core.c
@@ -434,6 +434,7 @@ int ip_nat_icmp_reply_translation(struct sk_buff **pskb,
 	} *inside;
 	struct ip_conntrack_tuple inner, target;
 	int hdrlen = (*pskb)->nh.iph->ihl * 4;
+	unsigned long statusbit;
 
 	if (!skb_make_writable(pskb, hdrlen + sizeof(*inside)))
 		return 0;
@@ -495,17 +496,16 @@ int ip_nat_icmp_reply_translation(struct sk_buff **pskb,
 
 	/* Change outer to look the reply to an incoming packet
 	 * (proto 0 means don't invert per-proto part). */
+	if (manip == IP_NAT_MANIP_SRC)
+		statusbit = IPS_SRC_NAT;
+	else
+		statusbit = IPS_DST_NAT;
 
-	/* Obviously, we need to NAT destination IP, but source IP
-	   should be NAT'ed only if it is from a NAT'd host.
+	/* Invert if this is reply dir. */
+	if (dir == IP_CT_DIR_REPLY)
+		statusbit ^= IPS_NAT_MASK;
 
-	   Explanation: some people use NAT for anonymizing.  Also,
-	   CERT recommends dropping all packets from private IP
-	   addresses (although ICMP errors from internal links with
-	   such addresses are not too uncommon, as Alan Cox points
-	   out) */
-	if (manip != IP_NAT_MANIP_SRC
-	    || ((*pskb)->nh.iph->saddr == ct->tuplehash[dir].tuple.src.ip)) {
+	if (ct->status & statusbit) {
 		invert_tuplepr(&target, &ct->tuplehash[!dir].tuple);
 		if (!manip_pkt(0, pskb, 0, &target, manip))
 			return 0;

From 669d32a293a348e692c365ddac2b23f3b907fcf1 Mon Sep 17 00:00:00 2001
From: Jean Tourrilhes <jt@hpl.hp.com>
Date: Sun, 19 Feb 2006 22:28:25 -0800
Subject: [PATCH 317/509] [IRDA]: irda-usb bug fixes

This patch fixes 2 bugs in the USB-IrDA code.

The first one is a buffer overrun in the RX path. We are now using
IRDA_SKB_MAX_MTU when initializing the Rx URB.

The second one is a potential stack recursion when unplugging the USB
dongle.  It seems that first we get the Rx URB with a generic error
code, and after a while the Rx URB comes again with a "disconnect"
error code.  Since we are resubmitting the Rx URB immediately after
receiving the first error one, we might enter an endless loop.

When getting an error Rx URB, the patch defers the Rx URB resubmitting
so that it gives us a chance to catch the disconnect one, in case the
dongle has juts been unplugged.

Tested against 2.6.16-rc2.

Patch from Jean Tourrilhes

Signed-off-by: Jean Tourrilhes <jt@hpl.hp.com>
Signed-off-by: Samuel Ortiz <samuel.ortiz@nokia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/irda/irda-usb.c | 90 +++++++++++++++++++++++++++++++------
 drivers/net/irda/irda-usb.h |  7 ++-
 2 files changed, 79 insertions(+), 18 deletions(-)

diff --git a/drivers/net/irda/irda-usb.c b/drivers/net/irda/irda-usb.c
index fa176ffb4ad5..8936058a3cce 100644
--- a/drivers/net/irda/irda-usb.c
+++ b/drivers/net/irda/irda-usb.c
@@ -108,6 +108,7 @@ static void irda_usb_close(struct irda_usb_cb *self);
 static void speed_bulk_callback(struct urb *urb, struct pt_regs *regs);
 static void write_bulk_callback(struct urb *urb, struct pt_regs *regs);
 static void irda_usb_receive(struct urb *urb, struct pt_regs *regs);
+static void irda_usb_rx_defer_expired(unsigned long data);
 static int irda_usb_net_open(struct net_device *dev);
 static int irda_usb_net_close(struct net_device *dev);
 static int irda_usb_net_ioctl(struct net_device *dev, struct ifreq *rq, int cmd);
@@ -677,6 +678,12 @@ static void irda_usb_net_timeout(struct net_device *netdev)
  * on the interrupt pipe and hang the Rx URB only when an interrupt is
  * received.
  * Jean II
+ *
+ * Note : don't read the above as what we are currently doing, but as
+ * something we could do with KC dongle. Also don't forget that the
+ * interrupt pipe is not part of the original standard, so this would
+ * need to be optional...
+ * Jean II
  */
 
 /*------------------------------------------------------------------*/
@@ -704,10 +711,8 @@ static void irda_usb_submit(struct irda_usb_cb *self, struct sk_buff *skb, struc
 	/* Reinitialize URB */
 	usb_fill_bulk_urb(urb, self->usbdev, 
 		      usb_rcvbulkpipe(self->usbdev, self->bulk_in_ep), 
-		      skb->data, skb->truesize,
+		      skb->data, IRDA_SKB_MAX_MTU,
                       irda_usb_receive, skb);
-	/* Note : unlink *must* be synchronous because of the code in 
-	 * irda_usb_net_close() -> free the skb - Jean II */
 	urb->status = 0;
 
 	/* Can be called from irda_usb_receive (irq handler) -> GFP_ATOMIC */
@@ -734,6 +739,7 @@ static void irda_usb_receive(struct urb *urb, struct pt_regs *regs)
 	struct irda_skb_cb *cb;
 	struct sk_buff *newskb;
 	struct sk_buff *dataskb;
+	struct urb *next_urb;
 	int		docopy;
 
 	IRDA_DEBUG(2, "%s(), len=%d\n", __FUNCTION__, urb->actual_length);
@@ -755,20 +761,37 @@ static void irda_usb_receive(struct urb *urb, struct pt_regs *regs)
 	if (urb->status != 0) {
 		switch (urb->status) {
 		case -EILSEQ:
-			self->stats.rx_errors++;
 			self->stats.rx_crc_errors++;	
-			break;
+			/* Also precursor to a hot-unplug on UHCI. */
+			/* Fallthrough... */
 		case -ECONNRESET:		/* -104 */
-			IRDA_DEBUG(0, "%s(), Connection Reset (-104), transfer_flags 0x%04X \n", __FUNCTION__, urb->transfer_flags);
+			/* Random error, if I remember correctly */
 			/* uhci_cleanup_unlink() is going to kill the Rx
 			 * URB just after we return. No problem, at this
 			 * point the URB will be idle ;-) - Jean II */
-			break;
+		case -ESHUTDOWN:		/* -108 */
+			/* That's usually a hot-unplug. Submit will fail... */
+		case -ETIMEDOUT:		/* -110 */
+			/* Usually precursor to a hot-unplug on OHCI. */
 		default:
-			IRDA_DEBUG(0, "%s(), RX status %d,transfer_flags 0x%04X \n", __FUNCTION__, urb->status, urb->transfer_flags);
+			self->stats.rx_errors++;
+			IRDA_DEBUG(0, "%s(), RX status %d, transfer_flags 0x%04X \n", __FUNCTION__, urb->status, urb->transfer_flags);
 			break;
 		}
-		goto done;
+		/* If we received an error, we don't want to resubmit the
+		 * Rx URB straight away but to give the USB layer a little
+		 * bit of breathing room.
+		 * We are in the USB thread context, therefore there is a
+		 * danger of recursion (new URB we submit fails, we come
+		 * back here).
+		 * With recent USB stack (2.6.15+), I'm seeing that on
+		 * hot unplug of the dongle...
+		 * Lowest effective timer is 10ms...
+		 * Jean II */
+		self->rx_defer_timer.function = &irda_usb_rx_defer_expired;
+		self->rx_defer_timer.data = (unsigned long) urb;
+		mod_timer(&self->rx_defer_timer, jiffies + (10 * HZ / 1000));
+		return;
 	}
 	
 	/* Check for empty frames */
@@ -845,13 +868,45 @@ done:
 	 * idle slot....
 	 * Jean II */
 	/* Note : with this scheme, we could submit the idle URB before
-	 * processing the Rx URB. Another time... Jean II */
+	 * processing the Rx URB. I don't think it would buy us anything as
+	 * we are running in the USB thread context. Jean II */
+	next_urb = self->idle_rx_urb;
 
-	/* Submit the idle URB to replace the URB we've just received */
-	irda_usb_submit(self, skb, self->idle_rx_urb);
 	/* Recycle Rx URB : Now, the idle URB is the present one */
 	urb->context = NULL;
 	self->idle_rx_urb = urb;
+
+	/* Submit the idle URB to replace the URB we've just received.
+	 * Do it last to avoid race conditions... Jean II */
+	irda_usb_submit(self, skb, next_urb);
+}
+
+/*------------------------------------------------------------------*/
+/*
+ * In case of errors, we want the USB layer to have time to recover.
+ * Now, it is time to resubmit ouur Rx URB...
+ */
+static void irda_usb_rx_defer_expired(unsigned long data)
+{
+	struct urb *urb = (struct urb *) data;
+	struct sk_buff *skb = (struct sk_buff *) urb->context;
+	struct irda_usb_cb *self; 
+	struct irda_skb_cb *cb;
+	struct urb *next_urb;
+
+	IRDA_DEBUG(2, "%s()\n", __FUNCTION__);
+
+	/* Find ourselves */
+	cb = (struct irda_skb_cb *) skb->cb;
+	IRDA_ASSERT(cb != NULL, return;);
+	self = (struct irda_usb_cb *) cb->context;
+	IRDA_ASSERT(self != NULL, return;);
+
+	/* Same stuff as when Rx is done, see above... */
+	next_urb = self->idle_rx_urb;
+	urb->context = NULL;
+	self->idle_rx_urb = urb;
+	irda_usb_submit(self, skb, next_urb);
 }
 
 /*------------------------------------------------------------------*/
@@ -990,6 +1045,9 @@ static int irda_usb_net_close(struct net_device *netdev)
 	/* Stop network Tx queue */
 	netif_stop_queue(netdev);
 
+	/* Kill defered Rx URB */
+	del_timer(&self->rx_defer_timer);
+
 	/* Deallocate all the Rx path buffers (URBs and skb) */
 	for (i = 0; i < IU_MAX_RX_URBS; i++) {
 		struct urb *urb = self->rx_urb[i];
@@ -1365,6 +1423,7 @@ static int irda_usb_probe(struct usb_interface *intf,
 	self = net->priv;
 	self->netdev = net;
 	spin_lock_init(&self->lock);
+	init_timer(&self->rx_defer_timer);
 
 	/* Create all of the needed urbs */
 	for (i = 0; i < IU_MAX_RX_URBS; i++) {
@@ -1498,6 +1557,9 @@ static void irda_usb_disconnect(struct usb_interface *intf)
 	 * This will stop/desactivate the Tx path. - Jean II */
 	self->present = 0;
 
+	/* Kill defered Rx URB */
+	del_timer(&self->rx_defer_timer);
+
 	/* We need to have irq enabled to unlink the URBs. That's OK,
 	 * at this point the Tx path is gone - Jean II */
 	spin_unlock_irqrestore(&self->lock, flags);
@@ -1507,11 +1569,11 @@ static void irda_usb_disconnect(struct usb_interface *intf)
 		/* Accept no more transmissions */
 		/*netif_device_detach(self->netdev);*/
 		netif_stop_queue(self->netdev);
-		/* Stop all the receive URBs */
+		/* Stop all the receive URBs. Must be synchronous. */
 		for (i = 0; i < IU_MAX_RX_URBS; i++)
 			usb_kill_urb(self->rx_urb[i]);
 		/* Cancel Tx and speed URB.
-		 * Toggle flags to make sure it's synchronous. */
+		 * Make sure it's synchronous to avoid races. */
 		usb_kill_urb(self->tx_urb);
 		usb_kill_urb(self->speed_urb);
 	}
diff --git a/drivers/net/irda/irda-usb.h b/drivers/net/irda/irda-usb.h
index bd8f66542322..4026af42dd47 100644
--- a/drivers/net/irda/irda-usb.h
+++ b/drivers/net/irda/irda-usb.h
@@ -136,8 +136,6 @@ struct irda_usb_cb {
 	__u16 bulk_out_mtu;		/* Max Tx packet size in bytes */
 	__u8  bulk_int_ep;		/* Interrupt Endpoint assignments */
 
-	wait_queue_head_t wait_q;	/* for timeouts */
-
 	struct urb *rx_urb[IU_MAX_RX_URBS];	/* URBs used to receive data frames */
 	struct urb *idle_rx_urb;	/* Pointer to idle URB in Rx path */
 	struct urb *tx_urb;		/* URB used to send data frames */
@@ -147,17 +145,18 @@ struct irda_usb_cb {
 	struct net_device_stats stats;
 	struct irlap_cb   *irlap;	/* The link layer we are binded to */
 	struct qos_info qos;
-	hashbin_t *tx_list;		/* Queued transmit skb's */
 	char *speed_buff;		/* Buffer for speed changes */
 
 	struct timeval stamp;
 	struct timeval now;
 
-	spinlock_t lock;		/* For serializing operations */
+	spinlock_t lock;		/* For serializing Tx operations */
 
 	__u16 xbofs;			/* Current xbofs setting */
 	__s16 new_xbofs;		/* xbofs we need to set */
 	__u32 speed;			/* Current speed */
 	__s32 new_speed;		/* speed we need to set */
+
+	struct timer_list rx_defer_timer;	/* Wait for Rx error to clear */
 };
 

From 8e249f088131cde5f77fd073bf0b0e8b3e9ea4ac Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sun, 19 Feb 2006 22:29:47 -0800
Subject: [PATCH 318/509] [NETFILTER]: Fix outgoing redirects to loopback

When redirecting an outgoing packet to loopback, it keeps the original
conntrack reference and information from the outgoing path, which
falsely triggers the check for DNAT on input and the dst_entry is
released to trigger rerouting. ip_route_input refuses to route the
packet because it has a local source address and it is dropped.

Look at the packet itself to dermine if it was NATed. Also fix a
missing inversion that causes unneccesary xfrm lookups.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/netfilter/ip_nat_standalone.c | 16 +++++-----------
 1 file changed, 5 insertions(+), 11 deletions(-)

diff --git a/net/ipv4/netfilter/ip_nat_standalone.c b/net/ipv4/netfilter/ip_nat_standalone.c
index 7c3f7d380240..ab1f88fa21ec 100644
--- a/net/ipv4/netfilter/ip_nat_standalone.c
+++ b/net/ipv4/netfilter/ip_nat_standalone.c
@@ -200,20 +200,14 @@ ip_nat_in(unsigned int hooknum,
           const struct net_device *out,
           int (*okfn)(struct sk_buff *))
 {
-	struct ip_conntrack *ct;
-	enum ip_conntrack_info ctinfo;
 	unsigned int ret;
+	u_int32_t daddr = (*pskb)->nh.iph->daddr;
 
 	ret = ip_nat_fn(hooknum, pskb, in, out, okfn);
 	if (ret != NF_DROP && ret != NF_STOLEN
-	    && (ct = ip_conntrack_get(*pskb, &ctinfo)) != NULL) {
-		enum ip_conntrack_dir dir = CTINFO2DIR(ctinfo);
-
-		if (ct->tuplehash[dir].tuple.dst.ip !=
-		    ct->tuplehash[!dir].tuple.src.ip) {
-			dst_release((*pskb)->dst);
-			(*pskb)->dst = NULL;
-		}
+	    && daddr != (*pskb)->nh.iph->daddr) {
+		dst_release((*pskb)->dst);
+		(*pskb)->dst = NULL;
 	}
 	return ret;
 }
@@ -276,7 +270,7 @@ ip_nat_local_fn(unsigned int hooknum,
 		    ct->tuplehash[!dir].tuple.src.ip
 #ifdef CONFIG_XFRM
 		    || ct->tuplehash[dir].tuple.dst.u.all !=
-		       ct->tuplehash[dir].tuple.src.u.all
+		       ct->tuplehash[!dir].tuple.src.u.all
 #endif
 		    )
 			return ip_route_me_harder(pskb) == 0 ? ret : NF_DROP;

From a8372f035aa2f6717123eb30679a08b619321dd4 Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Date: Sun, 19 Feb 2006 22:32:06 -0800
Subject: [PATCH 319/509] [NET]: NETFILTER: remove duplicated lines and fix
 order in skb_clone().

Some of netfilter-related members are initalized / copied twice in
skb_clone(). Remove one.

Pointed out by Olivier MATZ <olivier.matz@6wind.com>.

And this patch also fixes order of copying / clearing members.

Signed-off-by: YOSHIFUJI Hideaki <yoshfuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/skbuff.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 6766f118f070..2144952d1c6c 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -411,6 +411,9 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask)
 	C(pkt_type);
 	C(ip_summed);
 	C(priority);
+#if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE)
+	C(ipvs_property);
+#endif
 	C(protocol);
 	n->destructor = NULL;
 #ifdef CONFIG_NETFILTER
@@ -422,13 +425,6 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask)
 	C(nfct_reasm);
 	nf_conntrack_get_reasm(skb->nfct_reasm);
 #endif
-#if defined(CONFIG_IP_VS) || defined(CONFIG_IP_VS_MODULE)
-	C(ipvs_property);
-#endif
-#if defined(CONFIG_NF_CONNTRACK) || defined(CONFIG_NF_CONNTRACK_MODULE)
-	C(nfct_reasm);
-	nf_conntrack_get_reasm(skb->nfct_reasm);
-#endif
 #ifdef CONFIG_BRIDGE_NETFILTER
 	C(nf_bridge);
 	nf_bridge_get(skb->nf_bridge);

From 9ae61c6cb69f5251d160576c324948805f97e901 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 20 Feb 2006 23:48:37 +0900
Subject: [PATCH 320/509] [PATCH] libata: fix WARN_ON() condition in
 *_fill_sg()

For ATAPI commands, padding can reduce qc->n_elem by one and thus to
zero making assert(qc->n_elem > 0)'s in ata_fill_sg() and qs_fill_sg()
fail for legal commands.  This patch fixes the assert()'s to take
qc->pad_len into account.

Although the condition check seems a bit excessive, as this part of
code isn't still stable yet, I think it's worth to keep those.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/scsi/libata-core.c | 2 +-
 drivers/scsi/sata_qstor.c  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index 7ddd5a69352a..bbac87a13d57 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -2570,7 +2570,7 @@ static void ata_fill_sg(struct ata_queued_cmd *qc)
 	unsigned int idx;
 
 	assert(qc->__sg != NULL);
-	assert(qc->n_elem > 0);
+	assert(qc->n_elem > 0 || qc->pad_len > 0);
 
 	idx = 0;
 	ata_for_each_sg(sg, qc) {
diff --git a/drivers/scsi/sata_qstor.c b/drivers/scsi/sata_qstor.c
index de05e2883f9c..80480f0fb2b8 100644
--- a/drivers/scsi/sata_qstor.c
+++ b/drivers/scsi/sata_qstor.c
@@ -277,7 +277,7 @@ static unsigned int qs_fill_sg(struct ata_queued_cmd *qc)
 	u8 *prd = pp->pkt + QS_CPB_BYTES;
 
 	assert(qc->__sg != NULL);
-	assert(qc->n_elem > 0);
+	assert(qc->n_elem > 0 || qc->pad_len > 0);
 
 	nelem = 0;
 	ata_for_each_sg(sg, qc) {

From cc1887f3d8ae8ea61efa1a75af8ec0467b9dd546 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 20 Feb 2006 23:48:38 +0900
Subject: [PATCH 321/509] [PATCH] libata: fix qc->n_elem == 0 case handling in
 ata_qc_next_sg

This patch makes ata_for_each_sg() start with pad_sgent when
qc->n_elem is zero.  Previously, ata_for_each_sg() unconditionally
started with qc->__sg, handling the first sg to fill_sg() routines
even when the entry was invalid.  And while at it, unwind ?: in
ata_qc_next_sg() into if statement.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 include/linux/libata.h | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/include/linux/libata.h b/include/linux/libata.h
index 9e5db2949c58..c91be5e64ede 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -556,6 +556,16 @@ ata_sg_is_last(struct scatterlist *sg, struct ata_queued_cmd *qc)
 	return 0;
 }
 
+static inline struct scatterlist *
+ata_qc_first_sg(struct ata_queued_cmd *qc)
+{
+	if (qc->n_elem)
+		return qc->__sg;
+	if (qc->pad_len)
+		return &qc->pad_sgent;
+	return NULL;
+}
+
 static inline struct scatterlist *
 ata_qc_next_sg(struct scatterlist *sg, struct ata_queued_cmd *qc)
 {
@@ -563,11 +573,13 @@ ata_qc_next_sg(struct scatterlist *sg, struct ata_queued_cmd *qc)
 		return NULL;
 	if (++sg - qc->__sg < qc->n_elem)
 		return sg;
-	return qc->pad_len ? &qc->pad_sgent : NULL;
+	if (qc->pad_len)
+		return &qc->pad_sgent;
+	return NULL;
 }
 
 #define ata_for_each_sg(sg, qc) \
-	for (sg = qc->__sg; sg; sg = ata_qc_next_sg(sg, qc))
+	for (sg = ata_qc_first_sg(qc); sg; sg = ata_qc_next_sg(sg, qc))
 
 static inline unsigned int ata_tag_valid(unsigned int tag)
 {

From 2e242fa994428bd1a40b6a7e97430413246d0a16 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Mon, 20 Feb 2006 23:48:38 +0900
Subject: [PATCH 322/509] [PATCH] libata: make ata_sg_setup_one() trim zero
 length sg

This patch makes ata_sg_setup_one() trim sg entry (thus making
qc->n_elem zero) if padding results in zero length sg entry.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jgarzik@pobox.com>
---
 drivers/scsi/libata-core.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index bbac87a13d57..5f1d7580218d 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -2514,7 +2514,7 @@ static void ata_sg_clean(struct ata_queued_cmd *qc)
 	assert(sg != NULL);
 
 	if (qc->flags & ATA_QCFLAG_SINGLE)
-		assert(qc->n_elem == 1);
+		assert(qc->n_elem <= 1);
 
 	VPRINTK("unmapping %u sg elements\n", qc->n_elem);
 
@@ -2537,7 +2537,7 @@ static void ata_sg_clean(struct ata_queued_cmd *qc)
 			kunmap_atomic(addr, KM_IRQ0);
 		}
 	} else {
-		if (sg_dma_len(&sg[0]) > 0)
+		if (qc->n_elem)
 			dma_unmap_single(ap->host_set->dev,
 				sg_dma_address(&sg[0]), sg_dma_len(&sg[0]),
 				dir);
@@ -2715,6 +2715,7 @@ static int ata_sg_setup_one(struct ata_queued_cmd *qc)
 	int dir = qc->dma_dir;
 	struct scatterlist *sg = qc->__sg;
 	dma_addr_t dma_address;
+	int trim_sg = 0;
 
 	/* we must lengthen transfers to end on a 32-bit boundary */
 	qc->pad_len = sg->length & 3;
@@ -2734,13 +2735,15 @@ static int ata_sg_setup_one(struct ata_queued_cmd *qc)
 		sg_dma_len(psg) = ATA_DMA_PAD_SZ;
 		/* trim sg */
 		sg->length -= qc->pad_len;
+		if (sg->length == 0)
+			trim_sg = 1;
 
 		DPRINTK("padding done, sg->length=%u pad_len=%u\n",
 			sg->length, qc->pad_len);
 	}
 
-	if (!sg->length) {
-		sg_dma_address(sg) = 0;
+	if (trim_sg) {
+		qc->n_elem--;
 		goto skip_map;
 	}
 
@@ -2753,9 +2756,9 @@ static int ata_sg_setup_one(struct ata_queued_cmd *qc)
 	}
 
 	sg_dma_address(sg) = dma_address;
-skip_map:
 	sg_dma_len(sg) = sg->length;
 
+skip_map:
 	DPRINTK("mapped buffer of %d bytes for %s\n", sg_dma_len(sg),
 		qc->tf.flags & ATA_TFLAG_WRITE ? "write" : "read");
 

From b41c82eb5fb49912ce26c51ec221ba35e06c7d9b Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Mon, 20 Feb 2006 18:34:37 -0500
Subject: [PATCH 323/509] [AGPGART] Add some informational printk to nforce
 GART failure path.

Signed-off-by: Dave Jones <davej@redhat.com>
---
 drivers/char/agp/amd64-agp.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/char/agp/amd64-agp.c b/drivers/char/agp/amd64-agp.c
index 9964c508c111..1251b2515bbe 100644
--- a/drivers/char/agp/amd64-agp.c
+++ b/drivers/char/agp/amd64-agp.c
@@ -516,8 +516,10 @@ static int __devinit nforce3_agp_init(struct pci_dev *pdev)
 	pci_read_config_dword (hammers[0], AMD64_GARTAPERTUREBASE, &apbase);
 
 	/* if x86-64 aperture base is beyond 4G, exit here */
-	if ( (apbase & 0x7fff) >> (32 - 25) )
-		 return -ENODEV;
+	if ( (apbase & 0x7fff) >> (32 - 25) ) {
+		printk(KERN_INFO PFX "aperture base > 4G\n");
+		return -ENODEV;
+	}
 
 	apbase = (apbase & 0x7fff) << 25;
 

From 9827b781f20828e5ceb911b879f268f78fe90815 Mon Sep 17 00:00:00 2001
From: Kurt Garloff <garloff@suse.de>
Date: Mon, 20 Feb 2006 18:27:51 -0800
Subject: [PATCH 324/509] [PATCH] OOM kill: children accounting

In the badness() calculation, there's currently this piece of code:

        /*
         * Processes which fork a lot of child processes are likely
         * a good choice. We add the vmsize of the children if they
         * have an own mm. This prevents forking servers to flood the
         * machine with an endless amount of children
         */
        list_for_each(tsk, &p->children) {
                struct task_struct *chld;
                chld = list_entry(tsk, struct task_struct, sibling);
                if (chld->mm = p->mm && chld->mm)
                        points += chld->mm->total_vm;
        }

The intention is clear: If some server (apache) keeps spawning new children
and we run OOM, we want to kill the father rather than picking a child.

This -- to some degree -- also helps a bit with getting fork bombs under
control, though I'd consider this a desirable side-effect rather than a
feature.

There's one problem with this: No matter how many or few children there are,
if just one of them misbehaves, and all others (including the father) do
everything right, we still always kill the whole family.  This hits in real
life; whether it's javascript in konqueror resulting in kdeinit (and thus the
whole KDE session) being hit or just a classical server that spawns children.

Sidenote: The killer does kill all direct children as well, not only the
selected father, see oom_kill_process().

The idea in attached patch is that we do want to account the memory
consumption of the (direct) children to the father -- however not fully.
This maintains the property that fathers with too many children will still
very likely be picked, whereas a single misbehaving child has the chance to
be picked by the OOM killer.

In the patch I account only half (rounded up) of the children's vm_size to
the parent.  This means that if one child eats more mem than the rest of
the family, it will be picked, otherwise it's still the father and thus the
whole family that gets selected.

This is heuristics -- we could debate whether accounting for a fourth would
be better than for half of it.  Or -- if people would consider it worth the
trouble -- make it a sysctl.  For now I sticked to accounting for half,
which should IMHO be a significant improvement.

The patch does one more thing: As users tend to be irritated by the choice
of killed processes (mainly because the children are killed first, despite
some of them having a very low OOM score), I added some more output: The
selected (father) process will be reported first and it's oom_score printed
to syslog.

Description:

Only account for half of children's vm size in oom score calculation

This should still give the parent enough point in case of fork bombs.  If
any child however has more than 50% of the vm size of all children
together, it'll get a higher score and be elected.

This patch also makes the kernel display the oom_score.

Signed-off-by: Kurt Garloff <garloff@suse.de>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/oom_kill.c | 26 ++++++++++++++++----------
 1 file changed, 16 insertions(+), 10 deletions(-)

diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index b05ab8f2a562..949eba1d5ba3 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -58,15 +58,17 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
 
 	/*
 	 * Processes which fork a lot of child processes are likely
-	 * a good choice. We add the vmsize of the children if they
+	 * a good choice. We add half the vmsize of the children if they
 	 * have an own mm. This prevents forking servers to flood the
-	 * machine with an endless amount of children
+	 * machine with an endless amount of children. In case a single
+	 * child is eating the vast majority of memory, adding only half
+	 * to the parents will make the child our kill candidate of choice.
 	 */
 	list_for_each(tsk, &p->children) {
 		struct task_struct *chld;
 		chld = list_entry(tsk, struct task_struct, sibling);
 		if (chld->mm != p->mm && chld->mm)
-			points += chld->mm->total_vm;
+			points += chld->mm->total_vm/2 + 1;
 	}
 
 	/*
@@ -136,12 +138,12 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
  *
  * (not docbooked, we don't want this one cluttering up the manual)
  */
-static struct task_struct * select_bad_process(void)
+static struct task_struct *select_bad_process(unsigned long *ppoints)
 {
-	unsigned long maxpoints = 0;
 	struct task_struct *g, *p;
 	struct task_struct *chosen = NULL;
 	struct timespec uptime;
+	*ppoints = 0;
 
 	do_posix_clock_monotonic_gettime(&uptime);
 	do_each_thread(g, p) {
@@ -169,9 +171,9 @@ static struct task_struct * select_bad_process(void)
 			return p;
 
 		points = badness(p, uptime.tv_sec);
-		if (points > maxpoints || !chosen) {
+		if (points > *ppoints || !chosen) {
 			chosen = p;
-			maxpoints = points;
+			*ppoints = points;
 		}
 	} while_each_thread(g, p);
 	return chosen;
@@ -237,12 +239,15 @@ static struct mm_struct *oom_kill_task(task_t *p)
 	return mm;
 }
 
-static struct mm_struct *oom_kill_process(struct task_struct *p)
+static struct mm_struct *oom_kill_process(struct task_struct *p,
+					  unsigned long points)
 {
  	struct mm_struct *mm;
 	struct task_struct *c;
 	struct list_head *tsk;
 
+	printk(KERN_ERR "Out of Memory: Kill process %d (%s) score %li and "
+		"children.\n", p->pid, p->comm, points);
 	/* Try to kill a child first */
 	list_for_each(tsk, &p->children) {
 		c = list_entry(tsk, struct task_struct, sibling);
@@ -267,6 +272,7 @@ void out_of_memory(gfp_t gfp_mask, int order)
 {
 	struct mm_struct *mm = NULL;
 	task_t * p;
+	unsigned long points;
 
 	if (printk_ratelimit()) {
 		printk("oom-killer: gfp_mask=0x%x, order=%d\n",
@@ -278,7 +284,7 @@ void out_of_memory(gfp_t gfp_mask, int order)
 	cpuset_lock();
 	read_lock(&tasklist_lock);
 retry:
-	p = select_bad_process();
+	p = select_bad_process(&points);
 
 	if (PTR_ERR(p) == -1UL)
 		goto out;
@@ -290,7 +296,7 @@ retry:
 		panic("Out of memory and no killable processes...\n");
 	}
 
-	mm = oom_kill_process(p);
+	mm = oom_kill_process(p, points);
 	if (!mm)
 		goto retry;
 

From 9b0f8b040acd8dfd23860754c0d09ff4f44e2cbc Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Mon, 20 Feb 2006 18:27:52 -0800
Subject: [PATCH 325/509] [PATCH] Terminate process that fails on a constrained
 allocation

Some allocations are restricted to a limited set of nodes (due to memory
policies or cpuset constraints).  If the page allocator is not able to find
enough memory then that does not mean that overall system memory is low.

In particular going postal and more or less randomly shooting at processes
is not likely going to help the situation but may just lead to suicide (the
whole system coming down).

It is better to signal to the process that no memory exists given the
constraints that the process (or the configuration of the process) has
placed on the allocation behavior.  The process may be killed but then the
sysadmin or developer can investigate the situation.  The solution is
similar to what we do when running out of hugepages.

This patch adds a check before we kill processes.  At that point
performance considerations do not matter much so we just scan the zonelist
and reconstruct a list of nodes.  If the list of nodes does not contain all
online nodes then this is a constrained allocation and we should kill the
current process.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/sysrq.c |   2 +-
 include/linux/swap.h |   2 +-
 mm/oom_kill.c        | 103 ++++++++++++++++++++++++++++++++-----------
 mm/page_alloc.c      |   2 +-
 4 files changed, 81 insertions(+), 28 deletions(-)

diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c
index 5765f672e853..d58f82318853 100644
--- a/drivers/char/sysrq.c
+++ b/drivers/char/sysrq.c
@@ -243,7 +243,7 @@ static struct sysrq_key_op sysrq_term_op = {
 
 static void moom_callback(void *ignored)
 {
-	out_of_memory(GFP_KERNEL, 0);
+	out_of_memory(&NODE_DATA(0)->node_zonelists[ZONE_NORMAL], GFP_KERNEL, 0);
 }
 
 static DECLARE_WORK(moom_work, moom_callback, NULL);
diff --git a/include/linux/swap.h b/include/linux/swap.h
index f3e17d5963c3..d572b19afb7d 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -147,7 +147,7 @@ struct swap_list_t {
 #define vm_swap_full() (nr_swap_pages*2 < total_swap_pages)
 
 /* linux/mm/oom_kill.c */
-extern void out_of_memory(gfp_t gfp_mask, int order);
+extern void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order);
 
 /* linux/mm/memory.c */
 extern void swapin_readahead(swp_entry_t, unsigned long, struct vm_area_struct *);
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 949eba1d5ba3..8123fad5a485 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -132,6 +132,36 @@ unsigned long badness(struct task_struct *p, unsigned long uptime)
 	return points;
 }
 
+/*
+ * Types of limitations to the nodes from which allocations may occur
+ */
+#define CONSTRAINT_NONE 1
+#define CONSTRAINT_MEMORY_POLICY 2
+#define CONSTRAINT_CPUSET 3
+
+/*
+ * Determine the type of allocation constraint.
+ */
+static inline int constrained_alloc(struct zonelist *zonelist, gfp_t gfp_mask)
+{
+#ifdef CONFIG_NUMA
+	struct zone **z;
+	nodemask_t nodes = node_online_map;
+
+	for (z = zonelist->zones; *z; z++)
+		if (cpuset_zone_allowed(*z, gfp_mask))
+			node_clear((*z)->zone_pgdat->node_id,
+					nodes);
+		else
+			return CONSTRAINT_CPUSET;
+
+	if (!nodes_empty(nodes))
+		return CONSTRAINT_MEMORY_POLICY;
+#endif
+
+	return CONSTRAINT_NONE;
+}
+
 /*
  * Simple selection loop. We chose the process with the highest
  * number of 'points'. We expect the caller will lock the tasklist.
@@ -184,7 +214,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
  * CAP_SYS_RAW_IO set, send SIGTERM instead (but it's unlikely that
  * we select a process with CAP_SYS_RAW_IO set).
  */
-static void __oom_kill_task(task_t *p)
+static void __oom_kill_task(task_t *p, const char *message)
 {
 	if (p->pid == 1) {
 		WARN_ON(1);
@@ -200,8 +230,8 @@ static void __oom_kill_task(task_t *p)
 		return;
 	}
 	task_unlock(p);
-	printk(KERN_ERR "Out of Memory: Killed process %d (%s).\n",
-							p->pid, p->comm);
+	printk(KERN_ERR "%s: Killed process %d (%s).\n",
+				message, p->pid, p->comm);
 
 	/*
 	 * We give our sacrificial lamb high priority and access to
@@ -214,7 +244,7 @@ static void __oom_kill_task(task_t *p)
 	force_sig(SIGKILL, p);
 }
 
-static struct mm_struct *oom_kill_task(task_t *p)
+static struct mm_struct *oom_kill_task(task_t *p, const char *message)
 {
 	struct mm_struct *mm = get_task_mm(p);
 	task_t * g, * q;
@@ -226,21 +256,21 @@ static struct mm_struct *oom_kill_task(task_t *p)
 		return NULL;
 	}
 
-	__oom_kill_task(p);
+	__oom_kill_task(p, message);
 	/*
 	 * kill all processes that share the ->mm (i.e. all threads),
 	 * but are in a different thread group
 	 */
 	do_each_thread(g, q)
 		if (q->mm == mm && q->tgid != p->tgid)
-			__oom_kill_task(q);
+			__oom_kill_task(q, message);
 	while_each_thread(g, q);
 
 	return mm;
 }
 
 static struct mm_struct *oom_kill_process(struct task_struct *p,
-					  unsigned long points)
+				unsigned long points, const char *message)
 {
  	struct mm_struct *mm;
 	struct task_struct *c;
@@ -253,11 +283,11 @@ static struct mm_struct *oom_kill_process(struct task_struct *p,
 		c = list_entry(tsk, struct task_struct, sibling);
 		if (c->mm == p->mm)
 			continue;
-		mm = oom_kill_task(c);
+		mm = oom_kill_task(c, message);
 		if (mm)
 			return mm;
 	}
-	return oom_kill_task(p);
+	return oom_kill_task(p, message);
 }
 
 /**
@@ -268,10 +298,10 @@ static struct mm_struct *oom_kill_process(struct task_struct *p,
  * OR try to be smart about which process to kill. Note that we
  * don't have to be perfect here, we just have to be good.
  */
-void out_of_memory(gfp_t gfp_mask, int order)
+void out_of_memory(struct zonelist *zonelist, gfp_t gfp_mask, int order)
 {
 	struct mm_struct *mm = NULL;
-	task_t * p;
+	task_t *p;
 	unsigned long points;
 
 	if (printk_ratelimit()) {
@@ -283,25 +313,48 @@ void out_of_memory(gfp_t gfp_mask, int order)
 
 	cpuset_lock();
 	read_lock(&tasklist_lock);
+
+	/*
+	 * Check if there were limitations on the allocation (only relevant for
+	 * NUMA) that may require different handling.
+	 */
+	switch (constrained_alloc(zonelist, gfp_mask)) {
+	case CONSTRAINT_MEMORY_POLICY:
+		mm = oom_kill_process(current, points,
+				"No available memory (MPOL_BIND)");
+		break;
+
+	case CONSTRAINT_CPUSET:
+		mm = oom_kill_process(current, points,
+				"No available memory in cpuset");
+		break;
+
+	case CONSTRAINT_NONE:
 retry:
-	p = select_bad_process(&points);
+		/*
+		 * Rambo mode: Shoot down a process and hope it solves whatever
+		 * issues we may have.
+		 */
+		p = select_bad_process(&points);
 
-	if (PTR_ERR(p) == -1UL)
-		goto out;
+		if (PTR_ERR(p) == -1UL)
+			goto out;
 
-	/* Found nothing?!?! Either we hang forever, or we panic. */
-	if (!p) {
-		read_unlock(&tasklist_lock);
-		cpuset_unlock();
-		panic("Out of memory and no killable processes...\n");
+		/* Found nothing?!?! Either we hang forever, or we panic. */
+		if (!p) {
+			read_unlock(&tasklist_lock);
+			cpuset_unlock();
+			panic("Out of memory and no killable processes...\n");
+		}
+
+		mm = oom_kill_process(p, points, "Out of memory");
+		if (!mm)
+			goto retry;
+
+		break;
 	}
 
-	mm = oom_kill_process(p, points);
-	if (!mm)
-		goto retry;
-
- out:
-	read_unlock(&tasklist_lock);
+out:
 	cpuset_unlock();
 	if (mm)
 		mmput(mm);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 208812b25597..791690d7d3fa 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1015,7 +1015,7 @@ rebalance:
 		if (page)
 			goto got_pg;
 
-		out_of_memory(gfp_mask, order);
+		out_of_memory(zonelist, gfp_mask, order);
 		goto restart;
 	}
 

From 7d4c8e56109e0799ab9fb644c08a8daf4a026675 Mon Sep 17 00:00:00 2001
From: Daniel Yeisley <dan.yeisley@unisys.com>
Date: Mon, 20 Feb 2006 18:27:54 -0800
Subject: [PATCH 326/509] [PATCH] i386: need to pass virtual address to
 smp_read_mpc()

I'm seeing a kernel panic on an ES7000-600 when booting in virtual wire
mode.  The panic happens because smp_read_mpc() is passed a physical
address, and it should be virtual.  I tested the attached patch on the
ES7000-600 and on a 2 cpu Dell box, and saw no problems on either.

Signed-off-by: Dan Yeisley <dan.yeisley@unisys.com>
Acked-by: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/mpparse.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index 0102f3d50e57..e7609abf3796 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -710,7 +710,7 @@ void __init get_smp_config (void)
 		 * Read the physical hardware table.  Anything here will
 		 * override the defaults.
 		 */
-		if (!smp_read_mpc((void *)mpf->mpf_physptr)) {
+		if (!smp_read_mpc(phys_to_virt(mpf->mpf_physptr))) {
 			smp_found_config = 0;
 			printk(KERN_ERR "BIOS bug, MP table errors detected!...\n");
 			printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n");

From d86d43706a27bb87c2873de369f94a10f8758063 Mon Sep 17 00:00:00 2001
From: Alexey Korolev <alexey.korolev@intel.com>
Date: Mon, 20 Feb 2006 18:27:55 -0800
Subject: [PATCH 327/509] [PATCH] cfi_cmdset_0001: fix range for cache
 invalidation

I found an issue in cfi_cmdset0001.c.  It is related to cache region
invalidation in the buffered write procedure.

The code performs cache invalidation from "cmd_addr" to "cmd_adr + len" in
do_write_buffer() while we modify region from "adr" to "adr+len".

This issue affects writes + reads of data by small chunks.

Signed-off-by: Nicolas Pitre <nico@cam.org>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/mtd/chips/cfi_cmdset_0001.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index 69c04945591f..ded2c33f5b85 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -1019,8 +1019,8 @@ static void __xipram xip_udelay(struct map_info *map, struct flchip *chip,
 #define XIP_INVAL_CACHED_RANGE(map, from, size)  \
 	INVALIDATE_CACHED_RANGE(map, from, size)
 
-#define INVALIDATE_CACHE_UDELAY(map, chip, adr, len, usec)  \
-	UDELAY(map, chip, adr, usec)
+#define INVALIDATE_CACHE_UDELAY(map, chip, cmd_adr, adr, len, usec)  \
+	UDELAY(map, chip, cmd_adr, usec)
 
 /*
  * Extra notes:
@@ -1052,7 +1052,7 @@ do {  \
 	spin_lock(chip->mutex);  \
 } while (0)
 
-#define INVALIDATE_CACHE_UDELAY(map, chip, adr, len, usec)  \
+#define INVALIDATE_CACHE_UDELAY(map, chip, cmd_adr, adr, len, usec)  \
 do {  \
 	spin_unlock(chip->mutex);  \
 	INVALIDATE_CACHED_RANGE(map, adr, len);  \
@@ -1284,7 +1284,7 @@ static int __xipram do_write_oneword(struct map_info *map, struct flchip *chip,
 	map_write(map, datum, adr);
 	chip->state = mode;
 
-	INVALIDATE_CACHE_UDELAY(map, chip,
+	INVALIDATE_CACHE_UDELAY(map, chip, adr,
 				adr, map_bankwidth(map),
 				chip->word_write_time);
 
@@ -1572,8 +1572,8 @@ static int __xipram do_write_buffer(struct map_info *map, struct flchip *chip,
 	map_write(map, CMD(0xd0), cmd_adr);
 	chip->state = FL_WRITING;
 
-	INVALIDATE_CACHE_UDELAY(map, chip,
-				cmd_adr, len,
+	INVALIDATE_CACHE_UDELAY(map, chip, cmd_adr,
+				adr, len,
 				chip->buffer_write_time);
 
 	timeo = jiffies + (HZ/2);
@@ -1744,7 +1744,7 @@ static int __xipram do_erase_oneblock(struct map_info *map, struct flchip *chip,
 	chip->state = FL_ERASING;
 	chip->erase_suspended = 0;
 
-	INVALIDATE_CACHE_UDELAY(map, chip,
+	INVALIDATE_CACHE_UDELAY(map, chip, adr,
 				adr, len,
 				chip->erase_time*1000/2);
 

From d2799f083dcad0413ad1a396e9bc32d9afb70535 Mon Sep 17 00:00:00 2001
From: Stephen Street <stephen@streetfiresound.com>
Date: Mon, 20 Feb 2006 18:27:56 -0800
Subject: [PATCH 328/509] [PATCH] spi: Fix modular master driver remove and
 device suspend/remove

Fix two problems in the spi subsystem:

1) spi subsystem core dumps when modular spi master is unloaded.
2) spi subsystem core dumps when spi slave device is suspended/resumed and
   module slave driver is not loaded.

Signed-off-by: Stephen Street <stephen@streetfiresound.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/spi/spi.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 791c4dc550ae..94f5e8ed83a7 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -90,7 +90,7 @@ static int spi_suspend(struct device *dev, pm_message_t message)
 	int			value;
 	struct spi_driver	*drv = to_spi_driver(dev->driver);
 
-	if (!drv->suspend)
+	if (!drv || !drv->suspend)
 		return 0;
 
 	/* suspend will stop irqs and dma; no more i/o */
@@ -105,7 +105,7 @@ static int spi_resume(struct device *dev)
 	int			value;
 	struct spi_driver	*drv = to_spi_driver(dev->driver);
 
-	if (!drv->resume)
+	if (!drv || !drv->resume)
 		return 0;
 
 	/* resume may restart the i/o queue */
@@ -449,7 +449,6 @@ void spi_unregister_master(struct spi_master *master)
 {
 	(void) device_for_each_child(master->cdev.dev, NULL, __unregister);
 	class_device_unregister(&master->cdev);
-	master->cdev.dev = NULL;
 }
 EXPORT_SYMBOL_GPL(spi_unregister_master);
 

From 2e2b42636614f3c152672b5da67947ccbcbe0d32 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Mon, 20 Feb 2006 18:27:57 -0800
Subject: [PATCH 329/509] [PATCH] x86_64: Don't set CONFIG_DEBUG_INFO in
 defconfig

Undo setting of CONFIG_DEBUG_INFO in the previous defconfig update.  It
will make every build much slower and need more disk space and isn't a good
default.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/defconfig | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/x86_64/defconfig b/arch/x86_64/defconfig
index b337136f28b6..ce4de61ed85d 100644
--- a/arch/x86_64/defconfig
+++ b/arch/x86_64/defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.16-rc3
-# Mon Feb 13 22:31:24 2006
+# Linux kernel version: 2.6.16-rc3-git9
+# Sat Feb 18 00:27:03 2006
 #
 CONFIG_X86_64=y
 CONFIG_64BIT=y
@@ -1317,7 +1317,7 @@ CONFIG_DETECT_SOFTLOCKUP=y
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
-CONFIG_DEBUG_INFO=y
+# CONFIG_DEBUG_INFO is not set
 CONFIG_DEBUG_FS=y
 # CONFIG_DEBUG_VM is not set
 # CONFIG_FRAME_POINTER is not set

From 6303dbf570e410067380daec670fdb4137ac0d1d Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 20 Feb 2006 18:27:58 -0800
Subject: [PATCH 330/509] [PATCH] cpu hotplug documentation fix

Looks like there was a merge conflict when patches
8f8b1138fc9f65e3591aac83a4ee394fef34ac1d and
255acee706b333b79f593dd366f16e1f107cccc3 were applied which wasn't properly
resolved. Fix this and add some additional description.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/cpu-hotplug.txt | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/Documentation/cpu-hotplug.txt b/Documentation/cpu-hotplug.txt
index e71bc6cbbc5e..57a09f99ecb0 100644
--- a/Documentation/cpu-hotplug.txt
+++ b/Documentation/cpu-hotplug.txt
@@ -46,10 +46,12 @@ maxcpus=n    Restrict boot time cpus to n. Say if you have 4 cpus, using
              maxcpus=2 will only boot 2. You can choose to bring the
              other cpus later online, read FAQ's for more info.
 
-additional_cpus=n	[x86_64, s390 only] use this to limit hotpluggable cpus.
-                          This option sets
+additional_cpus*=n	Use this to limit hotpluggable cpus. This option sets
   			cpu_possible_map = cpu_present_map + additional_cpus
 
+(*) Option valid only for following architectures
+- x86_64, ia64, s390
+
 ia64 and x86_64 use the number of disabled local apics in ACPI tables MADT
 to determine the number of potentially hot-pluggable cpus. The implementation
 should only rely on this to count the #of cpus, but *MUST* not rely on the
@@ -57,6 +59,9 @@ apicid values in those tables for disabled apics. In the event BIOS doesnt
 mark such hot-pluggable cpus as disabled entries, one could use this
 parameter "additional_cpus=x" to represent those cpus in the cpu_possible_map.
 
+s390 uses the number of cpus it detects at IPL time to also the number of bits
+in cpu_possible_map. If it is desired to add additional cpus at a later time
+the number should be specified using this option or the possible_cpus option.
 
 possible_cpus=n		[s390 only] use this to set hotpluggable cpus.
 			This option sets possible_cpus bits in

From c255d844dd73616f23e4b4733edcc2e5fa4042b2 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@ucw.cz>
Date: Mon, 20 Feb 2006 18:27:58 -0800
Subject: [PATCH 331/509] [PATCH] suspend-to-ram: allow video options to be set
 at runtime

Currently, acpi video options can only be set on kernel command line.  That's
little inflexible; I'd like userland s2ram application that just works, and
modifying kernel command line according to whitelist is not fun.  It is better
to just allow s2ram application to set video options just before suspend
(according to the whitelist).

This implements sysctl to allow setting suspend video options without reboot.

(akpm: Documentation updates for this new sysctl are pending..)

Signed-off-by: Pavel Machek <pavel@suse.cz>
Cc: "Brown, Len" <len.brown@intel.com>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/sysctl/kernel.txt | 10 ++++++++++
 include/linux/acpi.h            |  3 ++-
 include/linux/sysctl.h          |  1 +
 kernel/sysctl.c                 | 16 ++++++++++++----
 4 files changed, 25 insertions(+), 5 deletions(-)

diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
index 9f11d36a8c10..b0c7ab93dcb9 100644
--- a/Documentation/sysctl/kernel.txt
+++ b/Documentation/sysctl/kernel.txt
@@ -16,6 +16,7 @@ before actually making adjustments.
 
 Currently, these files might (depending on your configuration)
 show up in /proc/sys/kernel:
+- acpi_video_flags
 - acct
 - core_pattern
 - core_uses_pid
@@ -57,6 +58,15 @@ show up in /proc/sys/kernel:
 
 ==============================================================
 
+acpi_video_flags:
+
+flags
+
+See Doc*/kernel/power/video.txt, it allows mode of video boot to be
+set during run time.
+
+==============================================================
+
 acct:
 
 highwater lowwater frequency
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 84d3d9f034ce..d3bc25e6d27d 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -427,7 +427,8 @@ extern int acpi_mp_config;
 extern struct acpi_table_mcfg_config *pci_mmcfg_config;
 extern int pci_mmcfg_config_num;
 
-extern int sbf_port ;
+extern int sbf_port;
+extern unsigned long acpi_video_flags;
 
 #else	/* !CONFIG_ACPI */
 
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 32a4139c4ad8..0e92bf7ec28e 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -146,6 +146,7 @@ enum
 	KERN_RANDOMIZE=68, /* int: randomize virtual address space */
 	KERN_SETUID_DUMPABLE=69, /* int: behaviour of dumps for setuid core */
 	KERN_SPIN_RETRY=70,	/* int: number of spinlock retries */
+	KERN_ACPI_VIDEO_FLAGS=71, /* int: flags for setting up video after ACPI sleep */
 };
 
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 7654d55c47f5..ebc41bf22f1e 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -44,14 +44,12 @@
 #include <linux/limits.h>
 #include <linux/dcache.h>
 #include <linux/syscalls.h>
+#include <linux/nfs_fs.h>
+#include <linux/acpi.h>
 
 #include <asm/uaccess.h>
 #include <asm/processor.h>
 
-#ifdef CONFIG_ROOT_NFS
-#include <linux/nfs_fs.h>
-#endif
-
 #if defined(CONFIG_SYSCTL)
 
 /* External variables not in a header file. */
@@ -655,6 +653,16 @@ static ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+#endif
+#ifdef CONFIG_ACPI_SLEEP
+	{
+		.ctl_name	= KERN_ACPI_VIDEO_FLAGS,
+		.procname	= "acpi_video_flags",
+		.data		= &acpi_video_flags,
+		.maxlen		= sizeof (unsigned long),
+		.mode		= 0644,
+		.proc_handler	= &proc_dointvec,
+	},
 #endif
 	{ .ctl_name = 0 }
 };

From a9c930bac163c5e616ca0ba9378e7dc746c93227 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Mon, 20 Feb 2006 18:27:59 -0800
Subject: [PATCH 332/509] [PATCH] Fix units in mbind check

maxnode is a bit index and can't be directly compared against a byte length
like PAGE_SIZE

Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/mempolicy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index bedfa4f09c80..6422fe478113 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -808,7 +808,7 @@ static int get_nodes(nodemask_t *nodes, const unsigned long __user *nmask,
 	nodes_clear(*nodes);
 	if (maxnode == 0 || !nmask)
 		return 0;
-	if (maxnode > PAGE_SIZE)
+	if (maxnode > PAGE_SIZE*BITS_PER_BYTE)
 		return -EINVAL;
 
 	nlongs = BITS_TO_LONGS(maxnode);

From aa657ca9245a06fa435e00332a13da1fce182abc Mon Sep 17 00:00:00 2001
From: Juergen Kreileder <jk@blackdown.de>
Date: Mon, 20 Feb 2006 18:28:00 -0800
Subject: [PATCH 333/509] [PATCH] Fix snd-usb-audio in 32-bit compat
 environment

I'm getting oopses with snd-usb-audio in 32-bit compat environments:
control_compat.c:get_ctl_type() doesn't initialize 'info', so
'itemlist[uinfo->value.enumerated.item]' in
usbmixer.c:mixer_ctl_selector_info() might access random memory (The 'if
((int)uinfo->value.enumerated.item >= cval->max)' doesn't fix all problems
because of the unsigned -> signed conversion.)

Signed-off-by: Juergen Kreileder <jk@blackdown.de>
Cc: Jaroslav Kysela <perex@suse.cz>
Acked-by: Takashi Iwai <tiwai@suse.de>
Cc: Greg KH <greg@kroah.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 sound/core/control_compat.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/sound/core/control_compat.c b/sound/core/control_compat.c
index 418c6d4e5daf..a529b62972b4 100644
--- a/sound/core/control_compat.c
+++ b/sound/core/control_compat.c
@@ -167,7 +167,7 @@ static int get_ctl_type(struct snd_card *card, struct snd_ctl_elem_id *id,
 			int *countp)
 {
 	struct snd_kcontrol *kctl;
-	struct snd_ctl_elem_info info;
+	struct snd_ctl_elem_info *info;
 	int err;
 
 	down_read(&card->controls_rwsem);
@@ -176,13 +176,19 @@ static int get_ctl_type(struct snd_card *card, struct snd_ctl_elem_id *id,
 		up_read(&card->controls_rwsem);
 		return -ENXIO;
 	}
-	info.id = *id;
-	err = kctl->info(kctl, &info);
+	info = kzalloc(sizeof(*info), GFP_KERNEL);
+	if (info == NULL) {
+		up_read(&card->controls_rwsem);
+		return -ENOMEM;
+	}
+	info->id = *id;
+	err = kctl->info(kctl, info);
 	up_read(&card->controls_rwsem);
 	if (err >= 0) {
-		err = info.type;
-		*countp = info.count;
+		err = info->type;
+		*countp = info->count;
 	}
+	kfree(info);
 	return err;
 }
 

From cef289633a9321cd99dd5f6cc935657dc487e9f0 Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Mon, 20 Feb 2006 18:28:01 -0800
Subject: [PATCH 334/509] [PATCH] pktcdvd: Correctly set rq->cmd_len in
 pkt_generic_packet()

It looks like the code in pkt_generic_packet() worked by luck in the past, but
after commit 186d330e682210100c671355580a8592e4a21692 leaving rq->cmd_len
uninitialized doesn't work any more.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 93e44d0292ab..f40310a42ba1 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -58,6 +58,7 @@
 #include <linux/suspend.h>
 #include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_ioctl.h>
+#include <scsi/scsi.h>
 
 #include <asm/uaccess.h>
 
@@ -380,6 +381,7 @@ static int pkt_generic_packet(struct pktcdvd_device *pd, struct packet_command *
 	memcpy(rq->cmd, cgc->cmd, CDROM_PACKET_SIZE);
 	if (sizeof(rq->cmd) > CDROM_PACKET_SIZE)
 		memset(rq->cmd + CDROM_PACKET_SIZE, 0, sizeof(rq->cmd) - CDROM_PACKET_SIZE);
+	rq->cmd_len = COMMAND_SIZE(rq->cmd[0]);
 
 	rq->ref_count++;
 	rq->flags |= REQ_NOMERGE;

From 7c613d593370292d1685f5794c743a2323be3a09 Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Mon, 20 Feb 2006 18:28:02 -0800
Subject: [PATCH 335/509] [PATCH] pktcdvd: Rename functions and make their
 return values sane

Boolean functions should return non-zero when they mean "true", otherwise the
calling code looks weird.  (As suggested by Linus.)

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 36 ++++++++++++++++++------------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index f40310a42ba1..fb2c31edca24 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -1497,9 +1497,9 @@ static int pkt_set_write_settings(struct pktcdvd_device *pd)
 }
 
 /*
- * 0 -- we can write to this track, 1 -- we can't
+ * 1 -- we can write to this track, 0 -- we can't
  */
-static int pkt_good_track(track_information *ti)
+static int pkt_writable_track(track_information *ti)
 {
 	/*
 	 * only good for CD-RW at the moment, not DVD-RW
@@ -1509,28 +1509,28 @@ static int pkt_good_track(track_information *ti)
 	 * FIXME: only for FP
 	 */
 	if (ti->fp == 0)
-		return 0;
+		return 1;
 
 	/*
 	 * "good" settings as per Mt Fuji.
 	 */
 	if (ti->rt == 0 && ti->blank == 0 && ti->packet == 1)
-		return 0;
+		return 1;
 
 	if (ti->rt == 0 && ti->blank == 1 && ti->packet == 1)
-		return 0;
+		return 1;
 
 	if (ti->rt == 1 && ti->blank == 0 && ti->packet == 1)
-		return 0;
+		return 1;
 
 	printk("pktcdvd: bad state %d-%d-%d\n", ti->rt, ti->blank, ti->packet);
-	return 1;
+	return 0;
 }
 
 /*
- * 0 -- we can write to this disc, 1 -- we can't
+ * 1 -- we can write to this disc, 0 -- we can't
  */
-static int pkt_good_disc(struct pktcdvd_device *pd, disc_information *di)
+static int pkt_writable_disc(struct pktcdvd_device *pd, disc_information *di)
 {
 	switch (pd->mmc3_profile) {
 		case 0x0a: /* CD-RW */
@@ -1539,10 +1539,10 @@ static int pkt_good_disc(struct pktcdvd_device *pd, disc_information *di)
 		case 0x1a: /* DVD+RW */
 		case 0x13: /* DVD-RW */
 		case 0x12: /* DVD-RAM */
-			return 0;
+			return 1;
 		default:
 			VPRINTK("pktcdvd: Wrong disc profile (%x)\n", pd->mmc3_profile);
-			return 1;
+			return 0;
 	}
 
 	/*
@@ -1551,25 +1551,25 @@ static int pkt_good_disc(struct pktcdvd_device *pd, disc_information *di)
 	 */
 	if (di->disc_type == 0xff) {
 		printk("pktcdvd: Unknown disc. No track?\n");
-		return 1;
+		return 0;
 	}
 
 	if (di->disc_type != 0x20 && di->disc_type != 0) {
 		printk("pktcdvd: Wrong disc type (%x)\n", di->disc_type);
-		return 1;
+		return 0;
 	}
 
 	if (di->erasable == 0) {
 		printk("pktcdvd: Disc not erasable\n");
-		return 1;
+		return 0;
 	}
 
 	if (di->border_status == PACKET_SESSION_RESERVED) {
 		printk("pktcdvd: Can't write to last track (reserved)\n");
-		return 1;
+		return 0;
 	}
 
-	return 0;
+	return 1;
 }
 
 static int pkt_probe_settings(struct pktcdvd_device *pd)
@@ -1594,7 +1594,7 @@ static int pkt_probe_settings(struct pktcdvd_device *pd)
 		return ret;
 	}
 
-	if (pkt_good_disc(pd, &di))
+	if (!pkt_writable_disc(pd, &di))
 		return -ENXIO;
 
 	switch (pd->mmc3_profile) {
@@ -1619,7 +1619,7 @@ static int pkt_probe_settings(struct pktcdvd_device *pd)
 		return ret;
 	}
 
-	if (pkt_good_track(&ti)) {
+	if (!pkt_writable_track(&ti)) {
 		printk("pktcdvd: can't write to this track\n");
 		return -ENXIO;
 	}

From 3b4828047d4ebe3703dedbfc739958c319ff0b24 Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Mon, 20 Feb 2006 18:28:03 -0800
Subject: [PATCH 336/509] [PATCH] pktcdvd: Remove useless printk statements

Writing the detected disc type in the kernel log is not useful during normal
use of the driver, so remove the printk statements.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 14 --------------
 1 file changed, 14 deletions(-)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index fb2c31edca24..fb08ae6de403 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -1597,20 +1597,6 @@ static int pkt_probe_settings(struct pktcdvd_device *pd)
 	if (!pkt_writable_disc(pd, &di))
 		return -ENXIO;
 
-	switch (pd->mmc3_profile) {
-		case 0x1a: /* DVD+RW */
-			printk("pktcdvd: inserted media is DVD+RW\n");
-			break;
-		case 0x13: /* DVD-RW */
-			printk("pktcdvd: inserted media is DVD-RW\n");
-			break;
-		case 0x12: /* DVD-RAM */
-			printk("pktcdvd: inserted media is DVD-RAM\n");
-			break;
-		default:
-			printk("pktcdvd: inserted media is CD-R%s\n", di.erasable ? "W" : "");
-			break;
-	}
 	pd->type = di.erasable ? PACKET_CDRW : PACKET_CDR;
 
 	track = 1; /* (di.last_track_msb << 8) | di.last_track_lsb; */

From ab863ec342cf148d02ed180b8ecf3a71a024e4be Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Mon, 20 Feb 2006 18:28:04 -0800
Subject: [PATCH 337/509] [PATCH] pktcdvd: Fix the logic in the
 pkt_writable_track function

Fix the pkt_writable_track() function to make it work correctly for all types
of CD/DVD discs.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 28 +++++++++++++++-------------
 1 file changed, 15 insertions(+), 13 deletions(-)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index fb08ae6de403..eb83197cc36c 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -1499,28 +1499,30 @@ static int pkt_set_write_settings(struct pktcdvd_device *pd)
 /*
  * 1 -- we can write to this track, 0 -- we can't
  */
-static int pkt_writable_track(track_information *ti)
+static int pkt_writable_track(struct pktcdvd_device *pd, track_information *ti)
 {
-	/*
-	 * only good for CD-RW at the moment, not DVD-RW
-	 */
+	switch (pd->mmc3_profile) {
+		case 0x1a: /* DVD+RW */
+		case 0x12: /* DVD-RAM */
+			/* The track is always writable on DVD+RW/DVD-RAM */
+			return 1;
+		default:
+			break;
+	}
 
-	/*
-	 * FIXME: only for FP
-	 */
-	if (ti->fp == 0)
-		return 1;
+	if (!ti->packet || !ti->fp)
+		return 0;
 
 	/*
 	 * "good" settings as per Mt Fuji.
 	 */
-	if (ti->rt == 0 && ti->blank == 0 && ti->packet == 1)
+	if (ti->rt == 0 && ti->blank == 0)
 		return 1;
 
-	if (ti->rt == 0 && ti->blank == 1 && ti->packet == 1)
+	if (ti->rt == 0 && ti->blank == 1)
 		return 1;
 
-	if (ti->rt == 1 && ti->blank == 0 && ti->packet == 1)
+	if (ti->rt == 1 && ti->blank == 0)
 		return 1;
 
 	printk("pktcdvd: bad state %d-%d-%d\n", ti->rt, ti->blank, ti->packet);
@@ -1605,7 +1607,7 @@ static int pkt_probe_settings(struct pktcdvd_device *pd)
 		return ret;
 	}
 
-	if (!pkt_writable_track(&ti)) {
+	if (!pkt_writable_track(pd, &ti)) {
 		printk("pktcdvd: can't write to this track\n");
 		return -ENXIO;
 	}

From 9db91546570ca1b3bc90b4c2d25d5bb74a44be24 Mon Sep 17 00:00:00 2001
From: Peter Osterlund <petero2@telia.com>
Date: Mon, 20 Feb 2006 18:28:04 -0800
Subject: [PATCH 338/509] [PATCH] pktcdvd: Only return -EROFS when appropriate

When attempting to open the device for writing, only return -EROFS if the disc
appears to be readable but not writable.

Signed-off-by: Peter Osterlund <petero2@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/block/pktcdvd.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index eb83197cc36c..bc9b2bcd7dba 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -1597,7 +1597,7 @@ static int pkt_probe_settings(struct pktcdvd_device *pd)
 	}
 
 	if (!pkt_writable_disc(pd, &di))
-		return -ENXIO;
+		return -EROFS;
 
 	pd->type = di.erasable ? PACKET_CDRW : PACKET_CDR;
 
@@ -1609,7 +1609,7 @@ static int pkt_probe_settings(struct pktcdvd_device *pd)
 
 	if (!pkt_writable_track(pd, &ti)) {
 		printk("pktcdvd: can't write to this track\n");
-		return -ENXIO;
+		return -EROFS;
 	}
 
 	/*
@@ -1623,7 +1623,7 @@ static int pkt_probe_settings(struct pktcdvd_device *pd)
 	}
 	if (pd->settings.size > PACKET_MAX_SECTORS) {
 		printk("pktcdvd: packet size is too big\n");
-		return -ENXIO;
+		return -EROFS;
 	}
 	pd->settings.fp = ti.fp;
 	pd->offset = (be32_to_cpu(ti.track_start) << 2) & (pd->settings.size - 1);
@@ -1665,7 +1665,7 @@ static int pkt_probe_settings(struct pktcdvd_device *pd)
 			break;
 		default:
 			printk("pktcdvd: unknown data mode\n");
-			return 1;
+			return -EROFS;
 	}
 	return 0;
 }
@@ -1876,7 +1876,7 @@ static int pkt_open_write(struct pktcdvd_device *pd)
 
 	if ((ret = pkt_probe_settings(pd))) {
 		VPRINTK("pktcdvd: %s failed probe\n", pd->name);
-		return -EROFS;
+		return ret;
 	}
 
 	if ((ret = pkt_set_write_settings(pd))) {

From e1c92117558261d5504c59712751f6c7925ff3ba Mon Sep 17 00:00:00 2001
From: Eric Van Hensbergen <ericvh@hera.kernel.org>
Date: Mon, 20 Feb 2006 18:28:05 -0800
Subject: [PATCH 339/509] [PATCH] v9fs: update documentation and fix debug flag

Minor updates to the documentation to bring them into sync with current
websites and available features.  The debug flag was switched back to hex
to match the documentation.

Signed-off-by: Eric Van Hensbergen <ericvh@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/filesystems/v9fs.txt | 16 ++++++++++------
 fs/9p/v9fs.c                       |  2 +-
 2 files changed, 11 insertions(+), 7 deletions(-)

diff --git a/Documentation/filesystems/v9fs.txt b/Documentation/filesystems/v9fs.txt
index 4e92feb6b507..24c7a9c41f0d 100644
--- a/Documentation/filesystems/v9fs.txt
+++ b/Documentation/filesystems/v9fs.txt
@@ -57,8 +57,6 @@ OPTIONS
 
   port=n	port to connect to on the remote server
 
-  timeout=n	request timeouts (in ms) (default 60000ms)
-
   noextend	force legacy mode (no 9P2000.u semantics)
 
   uid		attempt to mount as a particular uid
@@ -74,10 +72,16 @@ OPTIONS
 RESOURCES
 =========
 
-The Linux version of the 9P server, along with some client-side utilities
-can be found at http://v9fs.sf.net (along with a CVS repository of the
-development branch of this module).  There are user and developer mailing
-lists here, as well as a bug-tracker.
+The Linux version of the 9P server is now maintained under the npfs project
+on sourceforge (http://sourceforge.net/projects/npfs).
+
+There are user and developer mailing lists available through the v9fs project
+on sourceforge (http://sourceforge.net/projects/v9fs).
+
+News and other information is maintained on SWiK (http://swik.net/v9fs).
+
+Bug reports may be issued through the kernel.org bugzilla 
+(http://bugzilla.kernel.org)
 
 For more information on the Plan 9 Operating System check out
 http://plan9.bell-labs.com/plan9
diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c
index 5250c428fc1f..ef3386549140 100644
--- a/fs/9p/v9fs.c
+++ b/fs/9p/v9fs.c
@@ -66,7 +66,7 @@ static match_table_t tokens = {
 	{Opt_afid, "afid=%u"},
 	{Opt_rfdno, "rfdno=%u"},
 	{Opt_wfdno, "wfdno=%u"},
-	{Opt_debug, "debug=%u"},
+	{Opt_debug, "debug=%x"},
 	{Opt_name, "name=%s"},
 	{Opt_remotename, "aname=%s"},
 	{Opt_unix, "proto=unix"},

From c8b8b1f2e0eeb91cca22211950742b5f51564672 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Mon, 20 Feb 2006 18:28:06 -0800
Subject: [PATCH 340/509] [PATCH] powermac: Fix loss of ethernet PHY on sleep

Some recent PowerBook models tend to lose the ethernet PHY on
suspend/resume.  It -seems- that they use a combo ethernet-firewire PHY
chip and the firewire PHY seems to die the same way when that happens.  Not
trying to toggle the firewire cable power appears to fix it.  So this patch
disables changes to the firewire cable power control GPIO on those models.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Jeff Garzik <jgarzik@pobox.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/platforms/powermac/feature.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/platforms/powermac/feature.c b/arch/powerpc/platforms/powermac/feature.c
index 558dd0692092..34714d3ea69a 100644
--- a/arch/powerpc/platforms/powermac/feature.c
+++ b/arch/powerpc/platforms/powermac/feature.c
@@ -1646,10 +1646,10 @@ static void intrepid_shutdown(struct macio_chip *macio, int sleep_mode)
 		  KL0_SCC_CELL_ENABLE);
 
 	MACIO_BIC(KEYLARGO_FCR1,
-		  /*KL1_USB2_CELL_ENABLE |*/
 		KL1_I2S0_CELL_ENABLE | KL1_I2S0_CLK_ENABLE_BIT |
 		KL1_I2S0_ENABLE | KL1_I2S1_CELL_ENABLE |
-		KL1_I2S1_CLK_ENABLE_BIT | KL1_I2S1_ENABLE);
+		KL1_I2S1_CLK_ENABLE_BIT | KL1_I2S1_ENABLE |
+		KL1_EIDE0_ENABLE);
 	if (pmac_mb.board_flags & PMAC_MB_MOBILE)
 		MACIO_BIC(KEYLARGO_FCR1, KL1_UIDE_RESET_N);
 
@@ -2183,7 +2183,7 @@ static struct pmac_mb_def pmac_mb_defs[] = {
 	},
 	{	"PowerMac10,1",			"Mac mini",
 		PMAC_TYPE_UNKNOWN_INTREPID,	intrepid_features,
-		PMAC_MB_MAY_SLEEP | PMAC_MB_HAS_FW_POWER,
+		PMAC_MB_MAY_SLEEP,
 	},
 	{	"iMac,1",			"iMac (first generation)",
 		PMAC_TYPE_ORIG_IMAC,		paddington_features,
@@ -2295,11 +2295,11 @@ static struct pmac_mb_def pmac_mb_defs[] = {
 	},
 	{	"PowerBook5,8",			"PowerBook G4 15\"",
 		PMAC_TYPE_UNKNOWN_INTREPID,	intrepid_features,
-		PMAC_MB_MAY_SLEEP | PMAC_MB_HAS_FW_POWER | PMAC_MB_MOBILE,
+		PMAC_MB_MAY_SLEEP  | PMAC_MB_MOBILE,
 	},
 	{	"PowerBook5,9",			"PowerBook G4 17\"",
 		PMAC_TYPE_UNKNOWN_INTREPID,	intrepid_features,
-		PMAC_MB_MAY_SLEEP | PMAC_MB_HAS_FW_POWER | PMAC_MB_MOBILE,
+		PMAC_MB_MAY_SLEEP | PMAC_MB_MOBILE,
 	},
 	{	"PowerBook6,1",			"PowerBook G4 12\"",
 		PMAC_TYPE_UNKNOWN_INTREPID,	intrepid_features,

From 7a9166e3b037296366cea6f3c97f705d33e209e6 Mon Sep 17 00:00:00 2001
From: Luke Yang <luke.adi@gmail.com>
Date: Mon, 20 Feb 2006 18:28:07 -0800
Subject: [PATCH 341/509] [PATCH] Fix undefined symbols for nommu architecture

Signed-off-by: Luke Yang <luke.adi@gmail.com>
Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/mm.h | 4 ++++
 kernel/sysctl.c    | 2 ++
 mm/nommu.c         | 2 ++
 3 files changed, 8 insertions(+)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 26e1663a5cbe..498ff8778fb6 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1051,7 +1051,11 @@ int shrink_slab(unsigned long scanned, gfp_t gfp_mask,
 void drop_pagecache(void);
 void drop_slab(void);
 
+#ifndef CONFIG_MMU
+#define randomize_va_space 0
+#else
 extern int randomize_va_space;
+#endif
 
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MM_H */
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index ebc41bf22f1e..c05a2b7125e1 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -636,6 +636,7 @@ static ctl_table kern_table[] = {
 		.proc_handler	= &proc_dointvec,
 	},
 #endif
+#if defined(CONFIG_MMU)
 	{
 		.ctl_name	= KERN_RANDOMIZE,
 		.procname	= "randomize_va_space",
@@ -644,6 +645,7 @@ static ctl_table kern_table[] = {
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
 	},
+#endif
 #if defined(CONFIG_S390) && defined(CONFIG_SMP)
 	{
 		.ctl_name	= KERN_SPIN_RETRY,
diff --git a/mm/nommu.c b/mm/nommu.c
index c10262d68232..99d21020ec9d 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -57,6 +57,8 @@ EXPORT_SYMBOL(vmalloc);
 EXPORT_SYMBOL(vfree);
 EXPORT_SYMBOL(vmalloc_to_page);
 EXPORT_SYMBOL(vmalloc_32);
+EXPORT_SYMBOL(vmap);
+EXPORT_SYMBOL(vunmap);
 
 /*
  * Handle all mappings that got truncated by a "truncate()"

From 7fd105e758c8d746d57ab7e77f100e096bf153c8 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Mon, 20 Feb 2006 18:28:08 -0800
Subject: [PATCH 342/509] [PATCH] Fix compile for CONFIG_SYSVIPC=n or
 CONFIG_SYSCTL=n

The compat syscalls are added to sys_ni.c since they are not defined if the
above CONFIG options are off.  Also, nfs would not build with CONFIG_SYSCTL
off.

Noticed by Arthur Othieno.

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/linux/nfs_fs.h | 2 +-
 kernel/sys_ni.c        | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 547d649b274e..b4dc6e2e10c9 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -398,7 +398,7 @@ extern struct inode_operations nfs_symlink_inode_operations;
 extern int nfs_register_sysctl(void);
 extern void nfs_unregister_sysctl(void);
 #else
-#define nfs_register_sysctl() do { } while(0)
+#define nfs_register_sysctl() 0
 #define nfs_unregister_sysctl() do { } while(0)
 #endif
 
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 17313b99e53d..1067090db6b1 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -104,6 +104,8 @@ cond_syscall(sys_setreuid16);
 cond_syscall(sys_setuid16);
 cond_syscall(sys_vm86old);
 cond_syscall(sys_vm86);
+cond_syscall(compat_sys_ipc);
+cond_syscall(compat_sys_sysctl);
 
 /* arch-specific weak syscall entries */
 cond_syscall(sys_pciconfig_read);

From 1dd31b6c89611ee91c0ff309c8733c0af61579e8 Mon Sep 17 00:00:00 2001
From: Zhu Yi <yi.zhu@intel.com>
Date: Mon, 20 Feb 2006 18:28:09 -0800
Subject: [PATCH 343/509] [PATCH] ipw2200: Suppress warning message

The following message will be only printed if DEBUG_NOTIF is on.  "Unknown
notification: subtype=40,flags=0xa0,size=40"

Signed-off-by: Zhu Yi <yi.zhu@intel.com>
Cc: James Ketrenos <jketreno@linux.intel.com>
Cc: Jeff Garzik <jgarzik@pobox.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/net/wireless/ipw2200.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ipw2200.c b/drivers/net/wireless/ipw2200.c
index 14beab4bc91c..287676ad80df 100644
--- a/drivers/net/wireless/ipw2200.c
+++ b/drivers/net/wireless/ipw2200.c
@@ -4616,9 +4616,9 @@ static void ipw_rx_notification(struct ipw_priv *priv,
 		}
 
 	default:
-		IPW_ERROR("Unknown notification: "
-			  "subtype=%d,flags=0x%2x,size=%d\n",
-			  notif->subtype, notif->flags, notif->size);
+		IPW_DEBUG_NOTIF("Unknown notification: "
+				"subtype=%d,flags=0x%2x,size=%d\n",
+				notif->subtype, notif->flags, notif->size);
 	}
 }
 

From fcab6f351305029fc5e3c632209d45cae57e4835 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Mon, 20 Feb 2006 18:28:10 -0800
Subject: [PATCH 344/509] [PATCH] mm/mempolicy.c: fix 'if ();' typo

[akpm; it happens that the code was still correct, only inefficient ]

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Christoph Lameter <christoph@lameter.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/mempolicy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 6422fe478113..880831bd3003 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -587,7 +587,7 @@ redo:
 		}
 		list_add(&page->lru, &newlist);
 		nr_pages++;
-		if (nr_pages > MIGRATE_CHUNK_SIZE);
+		if (nr_pages > MIGRATE_CHUNK_SIZE)
 			break;
 	}
 	err = migrate_pages(pagelist, &newlist, &moved, &failed);

From ec33b5fe1a6f68c0a494aab476b9667945e029c4 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Mon, 20 Feb 2006 18:28:10 -0800
Subject: [PATCH 345/509] [PATCH] drivers/fc4/fc.c: memset correct length

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/fc4/fc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/fc4/fc.c b/drivers/fc4/fc.c
index 5c8943509cc1..66d03f242d3c 100644
--- a/drivers/fc4/fc.c
+++ b/drivers/fc4/fc.c
@@ -1053,7 +1053,7 @@ static int fc_do_els(fc_channel *fc, unsigned int alpa, void *data, int len)
 	int i;
 
 	fcmd = &_fcmd;
-	memset(fcmd, 0, sizeof(fcmd));
+	memset(fcmd, 0, sizeof(fcp_cmnd));
 	FCD(("PLOGI SID %d DID %d\n", fc->sid, alpa))
 	fch = &fcmd->fch;
 	FILL_FCHDR_RCTL_DID(fch, R_CTL_ELS_REQ, alpa);

From a1909e631caa3a02c25e493ac4004cd67984e0e0 Mon Sep 17 00:00:00 2001
From: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Date: Mon, 20 Feb 2006 18:28:11 -0800
Subject: [PATCH 346/509] [PATCH] radeonfb: resume support for Samsung P35
 laptops

Make resume from suspend-to-ram possible for Samsung P35 laptops.

The radeon mobility 9700 chip on Samsung P35 laptops locks up everything on
resume from suspend-to-ram if it is not reinitialized.

VGA compatible controller: ATI Technologies Inc RV350 [Mobility Radeon 9600 M10]
Class 0300: 1002:4e50
       Subsystem: 144d:c00c

Unfortunately, the DMI strings are mostly identical for all Samsung
laptops.  So we match the PCI ID and subsystem ID of the graphics card
which is unique for each Samsung laptop model.

Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006@gmx.net>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/aty/radeon_pm.c | 19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/video/aty/radeon_pm.c b/drivers/video/aty/radeon_pm.c
index 556895e99645..1f8d805c61e5 100644
--- a/drivers/video/aty/radeon_pm.c
+++ b/drivers/video/aty/radeon_pm.c
@@ -1321,8 +1321,6 @@ static void radeon_pm_full_reset_sdram(struct radeonfb_info *rinfo)
 	mdelay( 15);
 }
 
-#ifdef CONFIG_PPC_OF
-
 static void radeon_pm_reset_pad_ctlr_strength(struct radeonfb_info *rinfo)
 {
 	u32 tmp, tmp2;
@@ -1836,6 +1834,8 @@ static void radeon_reinitialize_M10(struct radeonfb_info *rinfo)
 	radeon_pm_m10_enable_lvds_spread_spectrum(rinfo);
 }
 
+#ifdef CONFIG_PPC_OF
+
 static void radeon_pm_m9p_reconfigure_mc(struct radeonfb_info *rinfo)
 {
 	OUTREG(MC_CNTL, rinfo->save_regs[46]);
@@ -2728,13 +2728,23 @@ void radeonfb_pm_init(struct radeonfb_info *rinfo, int dynclk)
 		printk("radeonfb: Dynamic Clock Power Management disabled\n");
 	}
 
+#if defined(CONFIG_PM)
 	/* Check if we can power manage on suspend/resume. We can do
 	 * D2 on M6, M7 and M9, and we can resume from D3 cold a few other
 	 * "Mac" cards, but that's all. We need more infos about what the
 	 * BIOS does tho. Right now, all this PM stuff is pmac-only for that
 	 * reason. --BenH
 	 */
-#if defined(CONFIG_PM) && defined(CONFIG_PPC_PMAC)
+	/* Special case for Samsung P35 laptops
+	 */
+	if ((rinfo->pdev->vendor == PCI_VENDOR_ID_ATI) &&
+	    (rinfo->pdev->device == PCI_CHIP_RV350_NP) &&
+	    (rinfo->pdev->subsystem_vendor == PCI_VENDOR_ID_SAMSUNG) &&
+	    (rinfo->pdev->subsystem_device == 0xc00c)) {
+		rinfo->reinit_func = radeon_reinitialize_M10;
+		rinfo->pm_mode |= radeon_pm_off;
+	}
+#if defined(CONFIG_PPC_PMAC)
 	if (_machine == _MACH_Pmac && rinfo->of_node) {
 		if (rinfo->is_mobility && rinfo->pm_reg &&
 		    rinfo->family <= CHIP_FAMILY_RV250)
@@ -2778,7 +2788,8 @@ void radeonfb_pm_init(struct radeonfb_info *rinfo, int dynclk)
 		OUTREG(TV_DAC_CNTL, INREG(TV_DAC_CNTL) | 0x07000000);
 #endif
 	}
-#endif /* defined(CONFIG_PM) && defined(CONFIG_PPC_PMAC) */
+#endif /* defined(CONFIG_PPC_PMAC) */
+#endif /* defined(CONFIG_PM) */
 }
 
 void radeonfb_pm_exit(struct radeonfb_info *rinfo)

From 15c73691780252a5571bfa7902b4dc227ec66c84 Mon Sep 17 00:00:00 2001
From: Frank Pavlic <fpavlic@de.ibm.com>
Date: Mon, 20 Feb 2006 18:28:12 -0800
Subject: [PATCH 347/509] [PATCH] s390: V=V qdio fixes

Using FCP devices with V=V support, the input queue stalled when CCQ 97 had
been returned in qdio_do_eqbs.  When this happen we have to reissue the eqbs
instruction.

Another bug was when V=V was enabled we checked if hardware has SIGA-sync
support.  If not we returned with 0 from tiqdio_is_inbound_q_done.  Thus qdio
lost initiative on FCP devices and input queue stalled.  Running devices in
V=V there is no SIGA-sync support but nevertheless we have to process
tiqdio_is_inbound_q_done either.

Signed-off-by: Frank Pavlic <fpavlic@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/s390/cio/qdio.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/s390/cio/qdio.c b/drivers/s390/cio/qdio.c
index 45ce032772f4..9ed37dc9a1b0 100644
--- a/drivers/s390/cio/qdio.c
+++ b/drivers/s390/cio/qdio.c
@@ -165,8 +165,13 @@ qdio_do_eqbs(struct qdio_q *q, unsigned char *state,
 	q_no = q->q_no;
 	if(!q->is_input_q)
 		q_no += irq->no_input_qs;
+again:
 	ccq = do_eqbs(irq->sch_token, state, q_no, start, cnt);
 	rc = qdio_check_ccq(q, ccq);
+	if (rc == 1) {
+		QDIO_DBF_TEXT5(1,trace,"eqAGAIN");
+		goto again;
+	}
 	if (rc < 0) {
                 QDIO_DBF_TEXT2(1,trace,"eqberr");
                 sprintf(dbf_text,"%2x,%2x,%d,%d",tmp_cnt, *cnt, ccq, q_no);
@@ -195,8 +200,13 @@ qdio_do_sqbs(struct qdio_q *q, unsigned char state,
 	q_no = q->q_no;
 	if(!q->is_input_q)
 		q_no += irq->no_input_qs;
+again:
 	ccq = do_sqbs(irq->sch_token, state, q_no, start, cnt);
 	rc = qdio_check_ccq(q, ccq);
+	if (rc == 1) {
+		QDIO_DBF_TEXT5(1,trace,"sqAGAIN");
+		goto again;
+	}
 	if (rc < 0) {
                 QDIO_DBF_TEXT3(1,trace,"sqberr");
                 sprintf(dbf_text,"%2x,%2x,%d,%d",tmp_cnt,*cnt,ccq,q_no);
@@ -1187,8 +1197,7 @@ tiqdio_is_inbound_q_done(struct qdio_q *q)
 
 	if (!no_used)
 		return 1;
-
-	if (!q->siga_sync)
+	if (!q->siga_sync && !irq->is_qebsm)
 		/* we'll check for more primed buffers in qeth_stop_polling */
 		return 0;
 	if (irq->is_qebsm) {

From aa88861fc3184a7d830954661dd281de4ae8d2ba Mon Sep 17 00:00:00 2001
From: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Date: Mon, 20 Feb 2006 18:28:13 -0800
Subject: [PATCH 348/509] [PATCH] s390: dasd reference counting

When using the dasd diag discipline, the base discipline module (eckd or fba)
can be unloaded, even though the dasd driver requires both discipline modules
(base and diag) to work correctly.

Implement reference counting for both base and diag discipline modules in
order to fix this.

Signed-off-by: Peter Oberparleiter <peter.oberparleiter@de.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/s390/block/dasd.c     | 21 ++++++++++++++++++++-
 drivers/s390/block/dasd_int.h |  1 +
 2 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index 08c88fcd8963..06bb992a4c6c 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -156,7 +156,12 @@ dasd_state_known_to_new(struct dasd_device * device)
 	/* disable extended error reporting for this device */
 	dasd_disable_eer(device);
 	/* Forget the discipline information. */
+	if (device->discipline)
+		module_put(device->discipline->owner);
 	device->discipline = NULL;
+	if (device->base_discipline)
+		module_put(device->base_discipline->owner);
+	device->base_discipline = NULL;
 	device->state = DASD_STATE_NEW;
 
 	dasd_free_queue(device);
@@ -1880,9 +1885,10 @@ dasd_generic_remove (struct ccw_device *cdev)
  */
 int
 dasd_generic_set_online (struct ccw_device *cdev,
-			 struct dasd_discipline *discipline)
+			 struct dasd_discipline *base_discipline)
 
 {
+	struct dasd_discipline *discipline;
 	struct dasd_device *device;
 	int rc;
 
@@ -1890,6 +1896,7 @@ dasd_generic_set_online (struct ccw_device *cdev,
 	if (IS_ERR(device))
 		return PTR_ERR(device);
 
+	discipline = base_discipline;
 	if (device->features & DASD_FEATURE_USEDIAG) {
 	  	if (!dasd_diag_discipline_pointer) {
 		        printk (KERN_WARNING
@@ -1901,6 +1908,16 @@ dasd_generic_set_online (struct ccw_device *cdev,
 		}
 		discipline = dasd_diag_discipline_pointer;
 	}
+	if (!try_module_get(base_discipline->owner)) {
+		dasd_delete_device(device);
+		return -EINVAL;
+	}
+	if (!try_module_get(discipline->owner)) {
+		module_put(base_discipline->owner);
+		dasd_delete_device(device);
+		return -EINVAL;
+	}
+	device->base_discipline = base_discipline;
 	device->discipline = discipline;
 
 	rc = discipline->check_device(device);
@@ -1909,6 +1926,8 @@ dasd_generic_set_online (struct ccw_device *cdev,
 			"dasd_generic couldn't online device %s "
 			"with discipline %s rc=%i\n",
 			cdev->dev.bus_id, discipline->name, rc);
+		module_put(discipline->owner);
+		module_put(base_discipline->owner);
 		dasd_delete_device(device);
 		return rc;
 	}
diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h
index d1b08fa13fd2..5efac1b97ece 100644
--- a/drivers/s390/block/dasd_int.h
+++ b/drivers/s390/block/dasd_int.h
@@ -321,6 +321,7 @@ struct dasd_device {
 
 	/* Device discipline stuff. */
 	struct dasd_discipline *discipline;
+	struct dasd_discipline *base_discipline;
 	char *private;
 
 	/* Device state and target state. */

From 49d9c81a699b57a5b6488f3a761669d05e116588 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 20 Feb 2006 18:28:14 -0800
Subject: [PATCH 349/509] [PATCH] s390: revert dasd eer module

Revert dasd eer module until we have a common understanding of how the
interface should be.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/s390/block/Kconfig         |   14 +-
 drivers/s390/block/Makefile        |    2 -
 drivers/s390/block/dasd.c          |   76 +-
 drivers/s390/block/dasd_3990_erp.c |    3 -
 drivers/s390/block/dasd_eckd.h     |    1 -
 drivers/s390/block/dasd_eer.c      | 1090 ----------------------------
 drivers/s390/block/dasd_int.h      |   37 -
 include/asm-s390/dasd.h            |   13 +-
 8 files changed, 5 insertions(+), 1231 deletions(-)
 delete mode 100644 drivers/s390/block/dasd_eer.c

diff --git a/drivers/s390/block/Kconfig b/drivers/s390/block/Kconfig
index 6912399d0937..6f50cc9323d9 100644
--- a/drivers/s390/block/Kconfig
+++ b/drivers/s390/block/Kconfig
@@ -55,21 +55,13 @@ config DASD_DIAG
 	  Disks under VM.  If you are not running under VM or unsure what it is,
 	  say "N".
 
-config DASD_EER
-	tristate "Extended error reporting (EER)"
-	depends on DASD
-	help
-	  This driver provides a character device interface to the
-          DASD extended error reporting. This is only needed if you want to
-          use applications written for the EER facility.
-
 config DASD_CMB
 	tristate "Compatibility interface for DASD channel measurement blocks"
 	depends on DASD
 	help
-	  This driver provides an additional interface to the channel
-          measurement facility, which is normally accessed though sysfs, with
-          a set of ioctl functions specific to the dasd driver.
+	  This driver provides an additional interface to the channel measurement
+	  facility, which is normally accessed though sysfs, with a set of
+	  ioctl functions specific to the dasd driver.
 	  This is only needed if you want to use applications written for
 	  linux-2.4 dasd channel measurement facility interface.
 
diff --git a/drivers/s390/block/Makefile b/drivers/s390/block/Makefile
index 0c0d871e8f51..58c6780134f7 100644
--- a/drivers/s390/block/Makefile
+++ b/drivers/s390/block/Makefile
@@ -5,7 +5,6 @@
 dasd_eckd_mod-objs := dasd_eckd.o dasd_3990_erp.o dasd_9343_erp.o
 dasd_fba_mod-objs  := dasd_fba.o dasd_3370_erp.o dasd_9336_erp.o
 dasd_diag_mod-objs := dasd_diag.o
-dasd_eer_mod-objs := dasd_eer.o
 dasd_mod-objs      := dasd.o dasd_ioctl.o dasd_proc.o dasd_devmap.o \
 			dasd_genhd.o dasd_erp.o
 
@@ -14,6 +13,5 @@ obj-$(CONFIG_DASD_DIAG) += dasd_diag_mod.o
 obj-$(CONFIG_DASD_ECKD) += dasd_eckd_mod.o
 obj-$(CONFIG_DASD_FBA)  += dasd_fba_mod.o
 obj-$(CONFIG_DASD_CMB)  += dasd_cmb.o
-obj-$(CONFIG_DASD_EER)  += dasd_eer.o
 obj-$(CONFIG_BLK_DEV_XPRAM) += xpram.o
 obj-$(CONFIG_DCSSBLK) += dcssblk.o
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index 06bb992a4c6c..af1d5b404cee 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -18,7 +18,6 @@
 #include <linux/slab.h>
 #include <linux/buffer_head.h>
 #include <linux/hdreg.h>
-#include <linux/notifier.h>
 
 #include <asm/ccwdev.h>
 #include <asm/ebcdic.h>
@@ -58,7 +57,6 @@ static void dasd_int_handler(struct ccw_device *, unsigned long, struct irb *);
 static void dasd_flush_ccw_queue(struct dasd_device *, int);
 static void dasd_tasklet(struct dasd_device *);
 static void do_kick_device(void *data);
-static void dasd_disable_eer(struct dasd_device *device);
 
 /*
  * SECTION: Operations on the device structure.
@@ -153,8 +151,6 @@ dasd_state_new_to_known(struct dasd_device *device)
 static inline void
 dasd_state_known_to_new(struct dasd_device * device)
 {
-	/* disable extended error reporting for this device */
-	dasd_disable_eer(device);
 	/* Forget the discipline information. */
 	if (device->discipline)
 		module_put(device->discipline->owner);
@@ -876,9 +872,6 @@ dasd_handle_state_change_pending(struct dasd_device *device)
 	struct dasd_ccw_req *cqr;
 	struct list_head *l, *n;
 
-	/* first of all call extended error reporting */
-	dasd_write_eer_trigger(DASD_EER_STATECHANGE, device, NULL);
-
 	device->stopped &= ~DASD_STOPPED_PENDING;
 
         /* restart all 'running' IO on queue */
@@ -1098,19 +1091,6 @@ restart:
 			}
 			goto restart;
 		}
-
-		/* first of all call extended error reporting */
-		if (device->eer && cqr->status == DASD_CQR_FAILED) {
-			dasd_write_eer_trigger(DASD_EER_FATALERROR,
-					       device, cqr);
-
-			/* restart request  */
-			cqr->status = DASD_CQR_QUEUED;
-			cqr->retries = 255;
-			device->stopped |= DASD_STOPPED_QUIESCE;
-			goto restart;
-		}
-
 		/* Process finished ERP request. */
 		if (cqr->refers) {
 			__dasd_process_erp(device, cqr);
@@ -1248,8 +1228,7 @@ __dasd_start_head(struct dasd_device * device)
 	cqr = list_entry(device->ccw_queue.next, struct dasd_ccw_req, list);
         /* check FAILFAST */
 	if (device->stopped & ~DASD_STOPPED_PENDING &&
-	    test_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags) &&
-	    (!device->eer)) {
+	    test_bit(DASD_CQR_FLAGS_FAILFAST, &cqr->flags)) {
 		cqr->status = DASD_CQR_FAILED;
 		dasd_schedule_bh(device);
 	}
@@ -2005,9 +1984,6 @@ dasd_generic_notify(struct ccw_device *cdev, int event)
 	switch (event) {
 	case CIO_GONE:
 	case CIO_NO_PATH:
-		/* first of all call extended error reporting */
-		dasd_write_eer_trigger(DASD_EER_NOPATH, device, NULL);
-
 		if (device->state < DASD_STATE_BASIC)
 			break;
 		/* Device is active. We want to keep it. */
@@ -2065,51 +2041,6 @@ dasd_generic_auto_online (struct ccw_driver *dasd_discipline_driver)
 	put_driver(drv);
 }
 
-/*
- * notifications for extended error reports
- */
-static struct notifier_block *dasd_eer_chain;
-
-int
-dasd_register_eer_notifier(struct notifier_block *nb)
-{
-	return notifier_chain_register(&dasd_eer_chain, nb);
-}
-
-int
-dasd_unregister_eer_notifier(struct notifier_block *nb)
-{
-	return notifier_chain_unregister(&dasd_eer_chain, nb);
-}
-
-/*
- * Notify the registered error reporting module of a problem
- */
-void
-dasd_write_eer_trigger(unsigned int id, struct dasd_device *device,
-		       struct dasd_ccw_req *cqr)
-{
-	if (device->eer) {
-		struct dasd_eer_trigger temp;
-		temp.id = id;
-		temp.device = device;
-		temp.cqr = cqr;
-		notifier_call_chain(&dasd_eer_chain, DASD_EER_TRIGGER,
-				    (void *)&temp);
-	}
-}
-
-/*
- * Tell the registered error reporting module to disable error reporting for
- * a given device and to cleanup any private data structures on that device.
- */
-static void
-dasd_disable_eer(struct dasd_device *device)
-{
-	notifier_call_chain(&dasd_eer_chain, DASD_EER_DISABLE, (void *)device);
-}
-
-
 static int __init
 dasd_init(void)
 {
@@ -2191,11 +2122,6 @@ EXPORT_SYMBOL_GPL(dasd_generic_set_online);
 EXPORT_SYMBOL_GPL(dasd_generic_set_offline);
 EXPORT_SYMBOL_GPL(dasd_generic_auto_online);
 
-EXPORT_SYMBOL(dasd_register_eer_notifier);
-EXPORT_SYMBOL(dasd_unregister_eer_notifier);
-EXPORT_SYMBOL(dasd_write_eer_trigger);
-
-
 /*
  * Overrides for Emacs so that we follow Linus's tabbing style.
  * Emacs will notice this stuff at the end of the file and automatically
diff --git a/drivers/s390/block/dasd_3990_erp.c b/drivers/s390/block/dasd_3990_erp.c
index c811380b9079..4ee0f934e325 100644
--- a/drivers/s390/block/dasd_3990_erp.c
+++ b/drivers/s390/block/dasd_3990_erp.c
@@ -1108,9 +1108,6 @@ dasd_3990_handle_env_data(struct dasd_ccw_req * erp, char *sense)
 		case 0x0B:
 			DEV_MESSAGE(KERN_WARNING, device, "%s",
 				    "FORMAT F - Volume is suspended duplex");
-			/* call extended error reporting (EER) */
-			dasd_write_eer_trigger(DASD_EER_PPRCSUSPEND, device,
-					       erp->refers);
 			break;
 		case 0x0C:
 			DEV_MESSAGE(KERN_WARNING, device, "%s",
diff --git a/drivers/s390/block/dasd_eckd.h b/drivers/s390/block/dasd_eckd.h
index e15dd7978050..bc3823d35223 100644
--- a/drivers/s390/block/dasd_eckd.h
+++ b/drivers/s390/block/dasd_eckd.h
@@ -29,7 +29,6 @@
 #define DASD_ECKD_CCW_PSF		 0x27
 #define DASD_ECKD_CCW_RSSD		 0x3e
 #define DASD_ECKD_CCW_LOCATE_RECORD	 0x47
-#define DASD_ECKD_CCW_SNSS               0x54
 #define DASD_ECKD_CCW_DEFINE_EXTENT	 0x63
 #define DASD_ECKD_CCW_WRITE_MT		 0x85
 #define DASD_ECKD_CCW_READ_MT		 0x86
diff --git a/drivers/s390/block/dasd_eer.c b/drivers/s390/block/dasd_eer.c
deleted file mode 100644
index f70cd7716b24..000000000000
--- a/drivers/s390/block/dasd_eer.c
+++ /dev/null
@@ -1,1090 +0,0 @@
-/*
- *	character device driver for extended error reporting
- *
- *
- *	Copyright (C) 2005 IBM Corporation
- *	extended error reporting for DASD ECKD devices
- *	Author(s): Stefan Weinhuber <wein@de.ibm.com>
- *
- */
-
-#include <linux/init.h>
-#include <linux/fs.h>
-#include <linux/kernel.h>
-#include <linux/miscdevice.h>
-#include <linux/module.h>
-#include <linux/moduleparam.h>
-#include <linux/device.h>
-#include <linux/workqueue.h>
-#include <linux/poll.h>
-#include <linux/notifier.h>
-
-#include <asm/uaccess.h>
-#include <asm/semaphore.h>
-#include <asm/atomic.h>
-#include <asm/ebcdic.h>
-
-#include "dasd_int.h"
-#include "dasd_eckd.h"
-
-
-MODULE_LICENSE("GPL");
-
-MODULE_AUTHOR("Stefan Weinhuber <wein@de.ibm.com>");
-MODULE_DESCRIPTION("DASD extended error reporting module");
-
-
-#ifdef PRINTK_HEADER
-#undef PRINTK_HEADER
-#endif				/* PRINTK_HEADER */
-#define PRINTK_HEADER "dasd(eer):"
-
-
-
-
-
-/*****************************************************************************/
-/*      the internal buffer                                                  */
-/*****************************************************************************/
-
-/*
- * The internal buffer is meant to store obaque blobs of data, so it doesn't
- * know of higher level concepts like triggers.
- * It consists of a number of pages that are used as a ringbuffer. Each data
- * blob is stored in a simple record that consists of an integer, which
- * contains the size of the following data, and the data bytes themselfes.
- *
- * To allow for multiple independent readers we create one internal buffer
- * each time the device is opened and destroy the buffer when the file is
- * closed again.
- *
- * One record can be written to a buffer by using the functions
- * - dasd_eer_start_record (one time per record to write the size to the buffer
- *                          and reserve the space for the data)
- * - dasd_eer_write_buffer (one or more times per record to write the data)
- * The data can be written in several steps but you will have to compute
- * the total size up front for the invocation of dasd_eer_start_record.
- * If the ringbuffer is full, dasd_eer_start_record will remove the required
- * number of old records.
- *
- * A record is typically read in two steps, first read the integer that
- * specifies the size of the following data, then read the data.
- * Both can be done by
- * - dasd_eer_read_buffer
- *
- * For all mentioned functions you need to get the bufferlock first and keep it
- * until a complete record is written or read.
- */
-
-
-/*
- * Alle information necessary to keep track of an internal buffer is kept in
- * a struct eerbuffer. The buffer specific to a file pointer is strored in
- * the private_data field of that file. To be able to write data to all
- * existing buffers, each buffer is also added to the bufferlist.
- * If the user doesn't want to read a complete record in one go, we have to
- * keep track of the rest of the record. residual stores the number of bytes
- * that are still to deliver. If the rest of the record is invalidated between
- * two reads then residual will be set to -1 so that the next read will fail.
- * All entries in the eerbuffer structure are protected with the bufferlock.
- * To avoid races between writing to a buffer on the one side and creating
- * and destroying buffers on the other side, the bufferlock must also be used
- * to protect the bufferlist.
- */
-
-struct eerbuffer {
-	struct list_head list;
-	char **buffer;
-	int buffersize;
-	int buffer_page_count;
-	int head;
-        int tail;
-	int residual;
-};
-
-LIST_HEAD(bufferlist);
-
-static spinlock_t bufferlock = SPIN_LOCK_UNLOCKED;
-
-DECLARE_WAIT_QUEUE_HEAD(dasd_eer_read_wait_queue);
-
-/*
- * How many free bytes are available on the buffer.
- * needs to be called with bufferlock held
- */
-static int
-dasd_eer_get_free_bytes(struct eerbuffer *eerb)
-{
-	if (eerb->head < eerb->tail) {
-		return eerb->tail - eerb->head - 1;
-	} else
-		return eerb->buffersize - eerb->head + eerb->tail -1;
-}
-
-/*
- * How many bytes of buffer space are used.
- * needs to be called with bufferlock held
- */
-static int
-dasd_eer_get_filled_bytes(struct eerbuffer *eerb)
-{
-
-	if (eerb->head >= eerb->tail) {
-		return eerb->head - eerb->tail;
-	} else
-		return eerb->buffersize - eerb->tail + eerb->head;
-}
-
-/*
- * The dasd_eer_write_buffer function just copies count bytes of data
- * to the buffer. Make sure to call dasd_eer_start_record first, to
- * make sure that enough free space is available.
- * needs to be called with bufferlock held
- */
-static void
-dasd_eer_write_buffer(struct eerbuffer *eerb, int count, char *data)
-{
-
-	unsigned long headindex,localhead;
-	unsigned long rest, len;
-	char *nextdata;
-
-	nextdata = data;
-	rest = count;
-	while (rest > 0) {
- 		headindex = eerb->head / PAGE_SIZE;
- 		localhead = eerb->head % PAGE_SIZE;
-		len = min(rest, (PAGE_SIZE - localhead));
-		memcpy(eerb->buffer[headindex]+localhead, nextdata, len);
-		nextdata += len;
-		rest -= len;
-		eerb->head += len;
-		if ( eerb->head == eerb->buffersize )
-			eerb->head = 0; /* wrap around */
-		if (eerb->head > eerb->buffersize) {
-			MESSAGE(KERN_ERR, "%s", "runaway buffer head.");
-			BUG();
-		}
-	}
-}
-
-/*
- * needs to be called with bufferlock held
- */
-static int
-dasd_eer_read_buffer(struct eerbuffer *eerb, int count, char *data)
-{
-
-	unsigned long tailindex,localtail;
-	unsigned long rest, len, finalcount;
-	char *nextdata;
-
-	finalcount = min(count, dasd_eer_get_filled_bytes(eerb));
-	nextdata = data;
-	rest = finalcount;
-	while (rest > 0) {
- 		tailindex = eerb->tail / PAGE_SIZE;
- 		localtail = eerb->tail % PAGE_SIZE;
-		len = min(rest, (PAGE_SIZE - localtail));
-		memcpy(nextdata, eerb->buffer[tailindex]+localtail, len);
-		nextdata += len;
-		rest -= len;
-		eerb->tail += len;
-		if ( eerb->tail == eerb->buffersize )
-			eerb->tail = 0; /* wrap around */
-		if (eerb->tail > eerb->buffersize) {
-			MESSAGE(KERN_ERR, "%s", "runaway buffer tail.");
-			BUG();
-		}
-	}
-	return finalcount;
-}
-
-/*
- * Whenever you want to write a blob of data to the internal buffer you
- * have to start by using this function first. It will write the number
- * of bytes that will be written to the buffer. If necessary it will remove
- * old records to make room for the new one.
- * needs to be called with bufferlock held
- */
-static int
-dasd_eer_start_record(struct eerbuffer *eerb, int count)
-{
-	int tailcount;
-	if (count + sizeof(count) > eerb->buffersize)
-		return -ENOMEM;
-	while (dasd_eer_get_free_bytes(eerb) < count + sizeof(count)) {
-		if (eerb->residual > 0) {
-			eerb->tail += eerb->residual;
-			if (eerb->tail >= eerb->buffersize)
-				eerb->tail -= eerb->buffersize;
-			eerb->residual = -1;
-		}
-		dasd_eer_read_buffer(eerb, sizeof(tailcount),
-				     (char*)(&tailcount));
-		eerb->tail += tailcount;
-		if (eerb->tail >= eerb->buffersize)
-			eerb->tail -= eerb->buffersize;
-	}
-	dasd_eer_write_buffer(eerb, sizeof(count), (char*)(&count));
-
-	return 0;
-};
-
-/*
- * release pages that are not used anymore
- */
-static void
-dasd_eer_free_buffer_pages(char **buf, int no_pages)
-{
-	int i;
-
-	for (i = 0; i < no_pages; ++i) {
-		free_page((unsigned long)buf[i]);
-	}
-}
-
-/*
- * allocate a new set of memory pages
- */
-static int
-dasd_eer_allocate_buffer_pages(char **buf, int no_pages)
-{
-	int i;
-
-	for (i = 0; i < no_pages; ++i) {
-		buf[i] = (char *) get_zeroed_page(GFP_KERNEL);
-		if (!buf[i]) {
-			dasd_eer_free_buffer_pages(buf, i);
-			return -ENOMEM;
-		}
-	}
-	return 0;
-}
-
-/*
- * empty the buffer by resetting head and tail
- * In case there is a half read data blob in the buffer, we set residual
- * to -1 to indicate that the remainder of the blob is lost.
- */
-static void
-dasd_eer_purge_buffer(struct eerbuffer *eerb)
-{
-	unsigned long flags;
-
-	spin_lock_irqsave(&bufferlock, flags);
-	if (eerb->residual > 0)
-		eerb->residual = -1;
-	eerb->tail=0;
-	eerb->head=0;
-	spin_unlock_irqrestore(&bufferlock, flags);
-}
-
-/*
- * set the size of the buffer, newsize is the new number of pages to be used
- * we don't try to copy any data back an forth, so any resize will also purge
- * the buffer
- */
-static int
-dasd_eer_resize_buffer(struct eerbuffer *eerb, int newsize)
-{
-	int i, oldcount, reuse;
-	char **new;
-	char **old;
-	unsigned long flags;
-
-	if (newsize < 1)
-		return -EINVAL;
-	if (eerb->buffer_page_count == newsize) {
-		/* documented behaviour is that any successfull invocation
-                 * will purge all records */
-		dasd_eer_purge_buffer(eerb);
-		return 0;
-	}
-	new = kmalloc(newsize*sizeof(char*), GFP_KERNEL);
-	if (!new)
-		return -ENOMEM;
-
-	reuse=min(eerb->buffer_page_count, newsize);
-	for (i = 0; i < reuse; ++i) {
-		new[i] = eerb->buffer[i];
-	}
-	if (eerb->buffer_page_count < newsize) {
-		if (dasd_eer_allocate_buffer_pages(
-			    &new[eerb->buffer_page_count],
-			    newsize - eerb->buffer_page_count)) {
-			kfree(new);
-			return -ENOMEM;
-		}
-	}
-
-	spin_lock_irqsave(&bufferlock, flags);
-	old = eerb->buffer;
-	eerb->buffer = new;
-	if (eerb->residual > 0)
-		eerb->residual = -1;
-	eerb->tail = 0;
-	eerb->head = 0;
-	oldcount = eerb->buffer_page_count;
-	eerb->buffer_page_count = newsize;
-	spin_unlock_irqrestore(&bufferlock, flags);
-
-	if (oldcount > newsize) {
-		for (i = newsize; i < oldcount; ++i) {
-			free_page((unsigned long)old[i]);
-		}
-	}
-	kfree(old);
-
-	return 0;
-}
-
-
-/*****************************************************************************/
-/*      The extended error reporting functionality                           */
-/*****************************************************************************/
-
-/*
- * When a DASD device driver wants to report an error, it calls the
- * function dasd_eer_write_trigger (via a notifier mechanism) and gives the
- * respective trigger ID as parameter.
- * Currently there are four kinds of triggers:
- *
- * DASD_EER_FATALERROR:  all kinds of unrecoverable I/O problems
- * DASD_EER_PPRCSUSPEND: PPRC was suspended
- * DASD_EER_NOPATH:      There is no path to the device left.
- * DASD_EER_STATECHANGE: The state of the device has changed.
- *
- * For the first three triggers all required information can be supplied by
- * the caller. For these triggers a record is written by the function
- * dasd_eer_write_standard_trigger.
- *
- * When dasd_eer_write_trigger is called to write a DASD_EER_STATECHANGE
- * trigger, we have to gather the necessary sense data first. We cannot queue
- * the necessary SNSS (sense subsystem status) request immediatly, since we
- * are likely to run in a deadlock situation. Instead, we schedule a
- * work_struct that calls the function dasd_eer_sense_subsystem_status to
- * create and start an SNSS  request asynchronously.
- *
- * To avoid memory allocations at runtime, the necessary memory is allocated
- * when the extended error reporting is enabled for a device (by
- * dasd_eer_probe). There is one private eer data structure for each eer
- * enabled DASD device. It contains memory for the work_struct, one SNSS cqr
- * and a flags field that is used to coordinate the use of the cqr. The call
- * to write a state change trigger can come in at any time, so we have one flag
- * CQR_IN_USE that protects the cqr itself. When this flag indicates that the
- * cqr is currently in use, dasd_eer_sense_subsystem_status cannot start a
- * second request but sets the SNSS_REQUESTED flag instead.
- *
- * When the request is finished, the callback function dasd_eer_SNSS_cb
- * is called. This function will invoke the function
- * dasd_eer_write_SNSS_trigger to finally write the trigger. It will also
- * check the SNSS_REQUESTED flag and if it is set it will call
- * dasd_eer_sense_subsystem_status again.
- *
- * To avoid race conditions during the handling of the lock, the flags must
- * be protected by the snsslock.
- */
-
-struct dasd_eer_private {
-	struct dasd_ccw_req *cqr;
-	unsigned long flags;
-	struct work_struct worker;
-};
-
-static void dasd_eer_destroy(struct dasd_device *device,
-			     struct dasd_eer_private *eer);
-static int
-dasd_eer_write_trigger(struct dasd_eer_trigger *trigger);
-static void dasd_eer_sense_subsystem_status(void *data);
-static int dasd_eer_notify(struct notifier_block *self,
-			   unsigned long action, void *data);
-
-struct workqueue_struct *dasd_eer_workqueue;
-
-#define SNSS_DATA_SIZE 44
-static spinlock_t snsslock = SPIN_LOCK_UNLOCKED;
-
-#define DASD_EER_BUSID_SIZE 10
-struct dasd_eer_header {
-	__u32 total_size;
-	__u32 trigger;
-	__u64 tv_sec;
-	__u64 tv_usec;
-	char busid[DASD_EER_BUSID_SIZE];
-} __attribute__ ((packed));
-
-static struct notifier_block dasd_eer_nb = {
-	.notifier_call = dasd_eer_notify,
-};
-
-/*
- * flags for use with dasd_eer_private
- */
-#define CQR_IN_USE     0
-#define SNSS_REQUESTED 1
-
-/*
- * This function checks if extended error reporting is available for a given
- * dasd_device. If yes, then it creates and returns a struct dasd_eer,
- * otherwise it returns an -EPERM error pointer.
- */
-struct dasd_eer_private *
-dasd_eer_probe(struct dasd_device *device)
-{
-	struct dasd_eer_private *private;
-
-	if (!(device && device->discipline
-	      && !strcmp(device->discipline->name, "ECKD"))) {
-		return ERR_PTR(-EPERM);
-	}
-	/* allocate the private data structure */
-	private = (struct dasd_eer_private *)kmalloc(
-		sizeof(struct dasd_eer_private), GFP_KERNEL);
-	if (!private) {
-		return ERR_PTR(-ENOMEM);
-	}
-	INIT_WORK(&private->worker, dasd_eer_sense_subsystem_status,
-		  (void *)device);
-	private->cqr = dasd_kmalloc_request("ECKD",
-					    1 /* SNSS */ ,
-					    SNSS_DATA_SIZE ,
-					    device);
-	if (!private->cqr) {
-		kfree(private);
-		return ERR_PTR(-ENOMEM);
-	}
-	private->flags = 0;
-	return private;
-};
-
-/*
- * If our private SNSS request is queued, remove it from the
- * dasd ccw queue so we can free the requests memory.
- */
-static void
-dasd_eer_dequeue_SNSS_request(struct dasd_device *device,
-			      struct dasd_eer_private *eer)
-{
-	struct list_head *lst, *nxt;
-	struct dasd_ccw_req *cqr, *erpcqr;
-	dasd_erp_fn_t erp_fn;
-
-	spin_lock_irq(get_ccwdev_lock(device->cdev));
-	list_for_each_safe(lst, nxt, &device->ccw_queue) {
-		cqr = list_entry(lst, struct dasd_ccw_req, list);
-		/* we are looking for two kinds or requests */
-		/* first kind: our SNSS request: */
-		if (cqr == eer->cqr) {
-			if (cqr->status == DASD_CQR_IN_IO)
-				device->discipline->term_IO(cqr);
-			list_del(&cqr->list);
-			break;
-		}
-		/* second kind: ERP requests for our SNSS request */
-		if (cqr->refers) {
-			/* If this erp request chain ends in our cqr, then */
-                        /* cal the erp_postaction to clean it up  */
-			erpcqr = cqr;
-			while (erpcqr->refers) {
-				erpcqr = erpcqr->refers;
-			}
-			if (erpcqr == eer->cqr) {
-				erp_fn = device->discipline->erp_postaction(
-					 cqr);
-				erp_fn(cqr);
-			}
-			continue;
-		}
-	}
-	spin_unlock_irq(get_ccwdev_lock(device->cdev));
-}
-
-/*
- * This function dismantles a struct dasd_eer that was created by
- * dasd_eer_probe. Since we want to free our private data structure,
- * we must make sure that the memory is not in use anymore.
- * We have to flush the work queue and remove a possible SNSS request
- * from the dasd queue.
- */
-static void
-dasd_eer_destroy(struct dasd_device *device, struct dasd_eer_private *eer)
-{
-	flush_workqueue(dasd_eer_workqueue);
-	dasd_eer_dequeue_SNSS_request(device, eer);
-	dasd_kfree_request(eer->cqr, device);
-	kfree(eer);
-};
-
-/*
- * enable the extended error reporting for a particular device
- */
-static int
-dasd_eer_enable_on_device(struct dasd_device *device)
-{
-	void *eer;
-	if (!device)
-		return -ENODEV;
-	if (device->eer)
-		return 0;
-	if (!try_module_get(THIS_MODULE)) {
-		return -EINVAL;
-	}
-	eer = (void *)dasd_eer_probe(device);
-	if (IS_ERR(eer)) {
-		module_put(THIS_MODULE);
-		return PTR_ERR(eer);
-	}
-	device->eer = eer;
-	return 0;
-}
-
-/*
- * enable the extended error reporting for a particular device
- */
-static int
-dasd_eer_disable_on_device(struct dasd_device *device)
-{
-	struct dasd_eer_private *eer = device->eer;
-
-	if (!device)
-		return -ENODEV;
-	if (!device->eer)
-		return 0;
-	device->eer = NULL;
-	dasd_eer_destroy(device,eer);
-	module_put(THIS_MODULE);
-
-	return 0;
-}
-
-/*
- * Set extended error reporting (eer)
- * Note: This will be registered as a DASD ioctl, to be called on DASD devices.
- */
-static int
-dasd_ioctl_set_eer(struct block_device *bdev, int no, long args)
-{
-	struct dasd_device *device;
-	int intval;
-
-	if (!capable(CAP_SYS_ADMIN))
-		return -EACCES;
-	if (bdev != bdev->bd_contains)
-		/* Error-reporting is not allowed for partitions */
-		return -EINVAL;
-	if (get_user(intval, (int __user *) args))
-		return -EFAULT;
-	device =  bdev->bd_disk->private_data;
-	if (device == NULL)
-		return -ENODEV;
-
-	intval = (intval != 0);
-	DEV_MESSAGE (KERN_DEBUG, device,
-		     "set eer on device to %d", intval);
-	if (intval)
-		return dasd_eer_enable_on_device(device);
-	else
-		return dasd_eer_disable_on_device(device);
-}
-
-/*
- * Get value of extended error reporting.
- * Note: This will be registered as a DASD ioctl, to be called on DASD devices.
- */
-static int
-dasd_ioctl_get_eer(struct block_device *bdev, int no, long args)
-{
-	struct dasd_device *device;
-
-	device =  bdev->bd_disk->private_data;
-	if (device == NULL)
-		return -ENODEV;
-	return put_user((device->eer != NULL), (int __user *) args);
-}
-
-/*
- * The following function can be used for those triggers that have
- * all necessary data available when the function is called.
- * If the parameter cqr is not NULL, the chain of requests will be searched
- * for valid sense data, and all valid sense data sets will be added to
- * the triggers data.
- */
-static int
-dasd_eer_write_standard_trigger(int trigger, struct dasd_device *device,
-				struct dasd_ccw_req *cqr)
-{
-	struct dasd_ccw_req *temp_cqr;
-	int data_size;
-	struct timeval tv;
-	struct dasd_eer_header header;
-	unsigned long flags;
-	struct eerbuffer *eerb;
-
-	/* go through cqr chain and count the valid sense data sets */
-	temp_cqr = cqr;
-	data_size = 0;
-	while (temp_cqr) {
-		if (temp_cqr->irb.esw.esw0.erw.cons)
-			data_size += 32;
-		temp_cqr = temp_cqr->refers;
-	}
-
-	header.total_size = sizeof(header) + data_size + 4; /* "EOR" */
-	header.trigger = trigger;
-	do_gettimeofday(&tv);
-	header.tv_sec = tv.tv_sec;
-	header.tv_usec = tv.tv_usec;
-	strncpy(header.busid, device->cdev->dev.bus_id, DASD_EER_BUSID_SIZE);
-
-	spin_lock_irqsave(&bufferlock, flags);
-	list_for_each_entry(eerb, &bufferlist, list) {
-		dasd_eer_start_record(eerb, header.total_size);
-		dasd_eer_write_buffer(eerb, sizeof(header), (char*)(&header));
-		temp_cqr = cqr;
-		while (temp_cqr) {
-			if (temp_cqr->irb.esw.esw0.erw.cons)
-				dasd_eer_write_buffer(eerb, 32, cqr->irb.ecw);
-			temp_cqr = temp_cqr->refers;
-		}
-		dasd_eer_write_buffer(eerb, 4,"EOR");
-	}
-	spin_unlock_irqrestore(&bufferlock, flags);
-
-	wake_up_interruptible(&dasd_eer_read_wait_queue);
-
-	return 0;
-}
-
-/*
- * This function writes a DASD_EER_STATECHANGE trigger.
- */
-static void
-dasd_eer_write_SNSS_trigger(struct dasd_device *device,
-			    struct dasd_ccw_req *cqr)
-{
-	int data_size;
-	int snss_rc;
-	struct timeval tv;
-	struct dasd_eer_header header;
-	unsigned long flags;
-	struct eerbuffer *eerb;
-
-	snss_rc = (cqr->status == DASD_CQR_FAILED) ? -EIO : 0;
-	if (snss_rc)
-		data_size = 0;
-	else
-		data_size = SNSS_DATA_SIZE;
-
-	header.total_size = sizeof(header) + data_size + 4; /* "EOR" */
-	header.trigger = DASD_EER_STATECHANGE;
-	do_gettimeofday(&tv);
-	header.tv_sec = tv.tv_sec;
-	header.tv_usec = tv.tv_usec;
-	strncpy(header.busid, device->cdev->dev.bus_id, DASD_EER_BUSID_SIZE);
-
-	spin_lock_irqsave(&bufferlock, flags);
-	list_for_each_entry(eerb, &bufferlist, list) {
-		dasd_eer_start_record(eerb, header.total_size);
-		dasd_eer_write_buffer(eerb, sizeof(header),(char*)(&header));
-		if (!snss_rc)
-			dasd_eer_write_buffer(eerb, SNSS_DATA_SIZE, cqr->data);
-		dasd_eer_write_buffer(eerb, 4,"EOR");
-	}
-	spin_unlock_irqrestore(&bufferlock, flags);
-
-	wake_up_interruptible(&dasd_eer_read_wait_queue);
-}
-
-/*
- * callback function for use with SNSS request
- */
-static void
-dasd_eer_SNSS_cb(struct dasd_ccw_req *cqr, void *data)
-{
-        struct dasd_device *device;
-	struct dasd_eer_private *private;
-	unsigned long irqflags;
-
-        device = (struct dasd_device *)data;
-	private = (struct dasd_eer_private *)device->eer;
-	dasd_eer_write_SNSS_trigger(device, cqr);
-	spin_lock_irqsave(&snsslock, irqflags);
-	if(!test_and_clear_bit(SNSS_REQUESTED, &private->flags)) {
-		clear_bit(CQR_IN_USE, &private->flags);
-		spin_unlock_irqrestore(&snsslock, irqflags);
-		return;
-	};
-	clear_bit(CQR_IN_USE, &private->flags);
-	spin_unlock_irqrestore(&snsslock, irqflags);
-	dasd_eer_sense_subsystem_status(device);
-	return;
-}
-
-/*
- * clean a used cqr before using it again
- */
-static void
-dasd_eer_clean_SNSS_request(struct dasd_ccw_req *cqr)
-{
-	struct ccw1 *cpaddr = cqr->cpaddr;
-	void *data = cqr->data;
-
-	memset(cqr, 0, sizeof(struct dasd_ccw_req));
-	memset(cpaddr, 0, sizeof(struct ccw1));
-	memset(data, 0, SNSS_DATA_SIZE);
-	cqr->cpaddr = cpaddr;
-	cqr->data = data;
-	strncpy((char *) &cqr->magic, "ECKD", 4);
-	ASCEBC((char *) &cqr->magic, 4);
-	set_bit(DASD_CQR_FLAGS_USE_ERP, &cqr->flags);
-}
-
-/*
- * build and start an SNSS request
- * This function is called from a work queue so we have to
- * pass the dasd_device pointer as a void pointer.
- */
-static void
-dasd_eer_sense_subsystem_status(void *data)
-{
-	struct dasd_device *device;
-	struct dasd_eer_private *private;
-	struct dasd_ccw_req *cqr;
-	struct ccw1 *ccw;
-	unsigned long irqflags;
-
-	device = (struct dasd_device *)data;
-	private = (struct dasd_eer_private *)device->eer;
-	if (!private) /* device not eer enabled any more */
-		return;
-	cqr = private->cqr;
-	spin_lock_irqsave(&snsslock, irqflags);
-	if(test_and_set_bit(CQR_IN_USE, &private->flags)) {
-		set_bit(SNSS_REQUESTED, &private->flags);
-		spin_unlock_irqrestore(&snsslock, irqflags);
-		return;
-	};
-	spin_unlock_irqrestore(&snsslock, irqflags);
-	dasd_eer_clean_SNSS_request(cqr);
-	cqr->device = device;
-	cqr->retries = 255;
-	cqr->expires = 10 * HZ;
-
-	ccw = cqr->cpaddr;
-	ccw->cmd_code = DASD_ECKD_CCW_SNSS;
-	ccw->count = SNSS_DATA_SIZE;
-	ccw->flags = 0;
-	ccw->cda = (__u32)(addr_t)cqr->data;
-
-	cqr->buildclk = get_clock();
-	cqr->status = DASD_CQR_FILLED;
-	cqr->callback = dasd_eer_SNSS_cb;
-	cqr->callback_data = (void *)device;
-        dasd_add_request_head(cqr);
-
-	return;
-}
-
-/*
- * This function is called for all triggers. It calls the appropriate
- * function that writes the actual trigger records.
- */
-static int
-dasd_eer_write_trigger(struct dasd_eer_trigger *trigger)
-{
-	int rc;
-	struct dasd_eer_private *private = trigger->device->eer;
-
-	switch (trigger->id) {
-	case DASD_EER_FATALERROR:
-	case DASD_EER_PPRCSUSPEND:
-		rc = dasd_eer_write_standard_trigger(
-			trigger->id, trigger->device, trigger->cqr);
-		break;
-	case DASD_EER_NOPATH:
-		rc = dasd_eer_write_standard_trigger(
-			trigger->id, trigger->device, NULL);
-		break;
-	case DASD_EER_STATECHANGE:
-                if (queue_work(dasd_eer_workqueue, &private->worker)) {
-                        rc=0;
-                } else {
-                        /* If the work_struct was already queued, it can't
-                         * be queued again. But this is OK since we don't
-                         * need to have it queued twice.
-                         */
-                        rc = -EBUSY;
-                }
-		break;
-	default: /* unknown trigger, so we write it without any sense data */
-		rc = dasd_eer_write_standard_trigger(
-			trigger->id, trigger->device, NULL);
-		break;
-	}
-	return rc;
-}
-
-/*
- * This function is registered with the dasd device driver and gets called
- * for all dasd eer notifications.
- */
-static int dasd_eer_notify(struct notifier_block *self,
-			    unsigned long action, void *data)
-{
-	switch (action) {
-	case DASD_EER_DISABLE:
-		dasd_eer_disable_on_device((struct dasd_device *)data);
-		break;
-	case DASD_EER_TRIGGER:
-		dasd_eer_write_trigger((struct dasd_eer_trigger *)data);
-		break;
-	}
-	return NOTIFY_OK;
-}
-
-
-/*****************************************************************************/
-/*      the device operations                                                */
-/*****************************************************************************/
-
-/*
- * On the one side we need a lock to access our internal buffer, on the
- * other side a copy_to_user can sleep. So we need to copy the data we have
- * to transfer in a readbuffer, which is protected by the readbuffer_mutex.
- */
-static char readbuffer[PAGE_SIZE];
-DECLARE_MUTEX(readbuffer_mutex);
-
-
-static int
-dasd_eer_open(struct inode *inp, struct file *filp)
-{
-	struct eerbuffer *eerb;
-	unsigned long flags;
-
-	eerb = kmalloc(sizeof(struct eerbuffer), GFP_KERNEL);
-	eerb->head = 0;
-	eerb->tail = 0;
-	eerb->residual = 0;
-	eerb->buffer_page_count = 1;
-	eerb->buffersize = eerb->buffer_page_count * PAGE_SIZE;
-        eerb->buffer = kmalloc(eerb->buffer_page_count*sizeof(char*),
-			       GFP_KERNEL);
-        if (!eerb->buffer)
-                return -ENOMEM;
-	if (dasd_eer_allocate_buffer_pages(eerb->buffer,
-					   eerb->buffer_page_count)) {
-		kfree(eerb->buffer);
-		return -ENOMEM;
-	}
-	filp->private_data = eerb;
-	spin_lock_irqsave(&bufferlock, flags);
-	list_add(&eerb->list, &bufferlist);
-	spin_unlock_irqrestore(&bufferlock, flags);
-
-	return nonseekable_open(inp,filp);
-}
-
-static int
-dasd_eer_close(struct inode *inp, struct file *filp)
-{
-	struct eerbuffer *eerb;
-	unsigned long flags;
-
-	eerb = (struct eerbuffer *)filp->private_data;
-	spin_lock_irqsave(&bufferlock, flags);
-	list_del(&eerb->list);
-	spin_unlock_irqrestore(&bufferlock, flags);
-	dasd_eer_free_buffer_pages(eerb->buffer, eerb->buffer_page_count);
-	kfree(eerb->buffer);
-	kfree(eerb);
-
-	return 0;
-}
-
-static long
-dasd_eer_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
-{
-	int intval;
-	struct eerbuffer *eerb;
-
-	eerb = (struct eerbuffer *)filp->private_data;
-	switch (cmd) {
-	case DASD_EER_PURGE:
-		dasd_eer_purge_buffer(eerb);
-		return 0;
-	case DASD_EER_SETBUFSIZE:
-		if (get_user(intval, (int __user *)arg))
-			return -EFAULT;
-		return dasd_eer_resize_buffer(eerb, intval);
-	default:
-		return -ENOIOCTLCMD;
-	}
-}
-
-static ssize_t
-dasd_eer_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos)
-{
-	int tc,rc;
-	int tailcount,effective_count;
-        unsigned long flags;
-	struct eerbuffer *eerb;
-
-	eerb = (struct eerbuffer *)filp->private_data;
-	if(down_interruptible(&readbuffer_mutex))
-		return -ERESTARTSYS;
-
-	spin_lock_irqsave(&bufferlock, flags);
-
-	if (eerb->residual < 0) { /* the remainder of this record */
-		                  /* has been deleted             */
-		eerb->residual = 0;
-		spin_unlock_irqrestore(&bufferlock, flags);
-		up(&readbuffer_mutex);
-		return -EIO;
-	} else if (eerb->residual > 0) {
-		/* OK we still have a second half of a record to deliver */
-		effective_count = min(eerb->residual, (int)count);
-		eerb->residual -= effective_count;
-	} else {
-		tc = 0;
-		while (!tc) {
-			tc = dasd_eer_read_buffer(eerb,
-				sizeof(tailcount), (char*)(&tailcount));
-			if (!tc) {
-				/* no data available */
-				spin_unlock_irqrestore(&bufferlock, flags);
-				up(&readbuffer_mutex);
-				if (filp->f_flags & O_NONBLOCK)
-					return -EAGAIN;
-				rc = wait_event_interruptible(
-					dasd_eer_read_wait_queue,
-					eerb->head != eerb->tail);
-				if (rc) {
-					return rc;
-				}
-				if(down_interruptible(&readbuffer_mutex))
-					return -ERESTARTSYS;
-				spin_lock_irqsave(&bufferlock, flags);
-			}
-		}
-		WARN_ON(tc != sizeof(tailcount));
-		effective_count = min(tailcount,(int)count);
-		eerb->residual = tailcount - effective_count;
-	}
-
-	tc = dasd_eer_read_buffer(eerb, effective_count, readbuffer);
-	WARN_ON(tc != effective_count);
-
-	spin_unlock_irqrestore(&bufferlock, flags);
-
-	if (copy_to_user(buf, readbuffer, effective_count)) {
-		up(&readbuffer_mutex);
-		return -EFAULT;
-	}
-
-	up(&readbuffer_mutex);
-	return effective_count;
-}
-
-static unsigned int
-dasd_eer_poll (struct file *filp, poll_table *ptable)
-{
-	unsigned int mask;
-	unsigned long flags;
-	struct eerbuffer *eerb;
-
-	eerb = (struct eerbuffer *)filp->private_data;
-	poll_wait(filp, &dasd_eer_read_wait_queue, ptable);
-	spin_lock_irqsave(&bufferlock, flags);
-	if (eerb->head != eerb->tail)
-		mask = POLLIN | POLLRDNORM ;
-	else
-		mask = 0;
-	spin_unlock_irqrestore(&bufferlock, flags);
-	return mask;
-}
-
-static struct file_operations dasd_eer_fops = {
-	.open		= &dasd_eer_open,
-	.release	= &dasd_eer_close,
-	.unlocked_ioctl = &dasd_eer_ioctl,
-	.compat_ioctl	= &dasd_eer_ioctl,
-	.read		= &dasd_eer_read,
-	.poll		= &dasd_eer_poll,
-	.owner		= THIS_MODULE,
-};
-
-static struct miscdevice dasd_eer_dev = {
-	.minor	    = MISC_DYNAMIC_MINOR,
-	.name	    = "dasd_eer",
-	.fops	    = &dasd_eer_fops,
-};
-
-
-/*****************************************************************************/
-/*	Init and exit							     */
-/*****************************************************************************/
-
-static int
-__init dasd_eer_init(void)
-{
-	int rc;
-
-	dasd_eer_workqueue = create_singlethread_workqueue("dasd_eer");
-	if (!dasd_eer_workqueue) {
-		MESSAGE(KERN_ERR , "%s", "dasd_eer_init could not "
-		       "create workqueue \n");
-		rc = -ENOMEM;
-		goto out;
-	}
-
-	rc = dasd_register_eer_notifier(&dasd_eer_nb);
-	if (rc) {
-		MESSAGE(KERN_ERR, "%s", "dasd_eer_init could not "
-		       "register error reporting");
-		goto queue;
-	}
-
-	dasd_ioctl_no_register(THIS_MODULE, BIODASDEERSET, dasd_ioctl_set_eer);
-	dasd_ioctl_no_register(THIS_MODULE, BIODASDEERGET, dasd_ioctl_get_eer);
-
-	/* we don't need our own character device,
-	 * so we just register as misc device */
-	rc = misc_register(&dasd_eer_dev);
-	if (rc) {
-		MESSAGE(KERN_ERR, "%s", "dasd_eer_init could not "
-		       "register misc device");
-		goto unregister;
-	}
-
-	return 0;
-
-unregister:
-	dasd_unregister_eer_notifier(&dasd_eer_nb);
-	dasd_ioctl_no_unregister(THIS_MODULE, BIODASDEERSET,
-				 dasd_ioctl_set_eer);
-	dasd_ioctl_no_unregister(THIS_MODULE, BIODASDEERGET,
-				 dasd_ioctl_get_eer);
-queue:
-	destroy_workqueue(dasd_eer_workqueue);
-out:
-	return rc;
-
-}
-module_init(dasd_eer_init);
-
-static void
-__exit dasd_eer_exit(void)
-{
-	dasd_unregister_eer_notifier(&dasd_eer_nb);
-	dasd_ioctl_no_unregister(THIS_MODULE, BIODASDEERSET,
-				 dasd_ioctl_set_eer);
-	dasd_ioctl_no_unregister(THIS_MODULE, BIODASDEERGET,
-				 dasd_ioctl_get_eer);
-	destroy_workqueue(dasd_eer_workqueue);
-
-	WARN_ON(misc_deregister(&dasd_eer_dev) != 0);
-}
-module_exit(dasd_eer_exit);
diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h
index 5efac1b97ece..0592354cc604 100644
--- a/drivers/s390/block/dasd_int.h
+++ b/drivers/s390/block/dasd_int.h
@@ -275,34 +275,6 @@ struct dasd_discipline {
 
 extern struct dasd_discipline *dasd_diag_discipline_pointer;
 
-
-/*
- * Notification numbers for extended error reporting notifications:
- * The DASD_EER_DISABLE notification is sent before a dasd_device (and it's
- * eer pointer) is freed. The error reporting module needs to do all necessary
- * cleanup steps.
- * The DASD_EER_TRIGGER notification sends the actual error reports (triggers).
- */
-#define DASD_EER_DISABLE 0
-#define DASD_EER_TRIGGER 1
-
-/* Trigger IDs for extended error reporting DASD_EER_TRIGGER notification */
-#define DASD_EER_FATALERROR  1
-#define DASD_EER_NOPATH      2
-#define DASD_EER_STATECHANGE 3
-#define DASD_EER_PPRCSUSPEND 4
-
-/*
- * The dasd_eer_trigger structure contains all data that we need to send
- * along with an DASD_EER_TRIGGER notification.
- */
-struct dasd_eer_trigger {
-	unsigned int id;
-	struct dasd_device *device;
-	struct dasd_ccw_req *cqr;
-};
-
-
 struct dasd_device {
 	/* Block device stuff. */
 	struct gendisk *gdp;
@@ -316,9 +288,6 @@ struct dasd_device {
 	unsigned long flags;		/* per device flags */
 	unsigned short features;        /* copy of devmap-features (read-only!) */
 
-	/* extended error reporting stuff (eer) */
-	void *eer;
-
 	/* Device discipline stuff. */
 	struct dasd_discipline *discipline;
 	struct dasd_discipline *base_discipline;
@@ -520,12 +489,6 @@ int dasd_generic_set_online(struct ccw_device *, struct dasd_discipline *);
 int dasd_generic_set_offline (struct ccw_device *cdev);
 int dasd_generic_notify(struct ccw_device *, int);
 void dasd_generic_auto_online (struct ccw_driver *);
-int dasd_register_eer_notifier(struct notifier_block *);
-int dasd_unregister_eer_notifier(struct notifier_block *);
-void dasd_write_eer_trigger(unsigned int , struct dasd_device *,
-			struct dasd_ccw_req *);
-
-
 
 /* externals in dasd_devmap.c */
 extern int dasd_max_devindex;
diff --git a/include/asm-s390/dasd.h b/include/asm-s390/dasd.h
index c744ff33b1df..1630c26e8f45 100644
--- a/include/asm-s390/dasd.h
+++ b/include/asm-s390/dasd.h
@@ -204,8 +204,7 @@ typedef struct attrib_data_t {
  *
  * Here ist how the ioctl-nr should be used:
  *    0 -   31   DASD driver itself
- *   32 -  229   still open
- *  230 -  239   DASD extended error reporting
+ *   32 -  239   still open
  *  240 -  255   reserved for EMC 
  *******************************************************************************/
 
@@ -237,22 +236,12 @@ typedef struct attrib_data_t {
 #define BIODASDPSRD    _IOR(DASD_IOCTL_LETTER,4,dasd_rssd_perf_stats_t)
 /* Get Attributes (cache operations) */
 #define BIODASDGATTR   _IOR(DASD_IOCTL_LETTER,5,attrib_data_t) 
-/* retrieve extended error-reporting value */
-#define BIODASDEERGET  _IOR(DASD_IOCTL_LETTER,6,int)
 
 
 /* #define BIODASDFORMAT  _IOW(IOCTL_LETTER,0,format_data_t) , deprecated */
 #define BIODASDFMT     _IOW(DASD_IOCTL_LETTER,1,format_data_t) 
 /* Set Attributes (cache operations) */
 #define BIODASDSATTR   _IOW(DASD_IOCTL_LETTER,2,attrib_data_t) 
-/* retrieve extended error-reporting value */
-#define BIODASDEERSET  _IOW(DASD_IOCTL_LETTER,3,int)
-
-
-/* remove all records from the eer buffer */
-#define DASD_EER_PURGE       _IO(DASD_IOCTL_LETTER,230)
-/* set the number of pages that are used for the internal eer buffer */
-#define DASD_EER_SETBUFSIZE  _IOW(DASD_IOCTL_LETTER,230,int)
 
 
 #endif				/* DASD_H */

From b04ec261bd64f927bf3fce5cf9eeb0225557939d Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Mon, 20 Feb 2006 18:28:15 -0800
Subject: [PATCH 350/509] [PATCH] m32r: __cmpxchg_u32 fix

This patch fixes a bug of include/asm-m32r/system.h:__cmpxchg_u32().

  static __inline__ unsigned long
  __cmpxchg_u32(volatile unsigned int *p, unsigned int old, unsigned int new);

In __cmpxchg_u32(), the "old" value must not be changed to the previous "*p"
value.  But the former code modifies the previous "*p" value.

A deadlock at _atomic_dec_and_lock sometimes happened due to this bug.

Signed-off-by: Hayato Fujiwara <fujiwara@linux-m32r.org>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-m32r/system.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-m32r/system.h b/include/asm-m32r/system.h
index 06c12a037cba..d6a2c613be68 100644
--- a/include/asm-m32r/system.h
+++ b/include/asm-m32r/system.h
@@ -239,7 +239,7 @@ __cmpxchg_u32(volatile unsigned int *p, unsigned int old, unsigned int new)
 		"	bra	2f;		\n"
                 "       .fillinsn		\n"
 		"1:"
-			M32R_UNLOCK" %2, @%1;	\n"
+			M32R_UNLOCK" %0, @%1;	\n"
                 "       .fillinsn		\n"
 		"2:"
 			: "=&r" (retval)

From cf535ea52e68e3ee6f4a90cc383faa1ee857f14d Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Mon, 20 Feb 2006 18:28:17 -0800
Subject: [PATCH 351/509] [PATCH] m32r: update sys_tas() routine

This patch updates and fixes sys_tas() routine for m32r.

In the previous implementation, a lockup rarely caused at sys_tas()
routine in SMP environment.

> > The problem is that touching *addr will generate an oops if that page isn't
> > paged in.  If we convert it to use get_user() then that's an improvement,
> > but we must not run get_user() under spinlock or local_irq_disable().

I rewrote sys_tas() routine by using "lock -> unlock" instructions, and
utilizing the m32r's interrupt handling characteristics; the m32r processor
can accept interrupts only at the 32-bit instruction boundary.  So, the
"unlock" instruction can be executed continuously after the "lock"
instruction execution without any interruptions.

In addition, to solve such a page_fault problem, I use a fixup code like
get_user().

And, as for the kernel lockup problem, we found that a calling
do_page_fault() routine with disabling interrupts might cause a lockup at
flush_tlb_others(), because we checked a completion of IPI handler's
operations in a spin-locked critical section.

Therefore, by using "lock -> unlock" code, we can implement the sys_tas()
rouitine without disabling interrupts explicitly, then no lockups would
happen at flush_tlb_others(), I hope.

Compile check and some working test in SMP environment have done.

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/m32r/kernel/sys_m32r.c | 61 +++++++++++++++++++++----------------
 1 file changed, 34 insertions(+), 27 deletions(-)

diff --git a/arch/m32r/kernel/sys_m32r.c b/arch/m32r/kernel/sys_m32r.c
index fe55b28d3725..670cb49210af 100644
--- a/arch/m32r/kernel/sys_m32r.c
+++ b/arch/m32r/kernel/sys_m32r.c
@@ -29,28 +29,7 @@
 
 /*
  * sys_tas() - test-and-set
- * linuxthreads testing version
  */
-#ifndef CONFIG_SMP
-asmlinkage int sys_tas(int *addr)
-{
-	int oldval;
-	unsigned long flags;
-
-	if (!access_ok(VERIFY_WRITE, addr, sizeof (int)))
-		return -EFAULT;
-	local_irq_save(flags);
-	oldval = *addr;
-	if (!oldval)
-		*addr = 1;
-	local_irq_restore(flags);
-	return oldval;
-}
-#else /* CONFIG_SMP */
-#include <linux/spinlock.h>
-
-static DEFINE_SPINLOCK(tas_lock);
-
 asmlinkage int sys_tas(int *addr)
 {
 	int oldval;
@@ -58,15 +37,43 @@ asmlinkage int sys_tas(int *addr)
 	if (!access_ok(VERIFY_WRITE, addr, sizeof (int)))
 		return -EFAULT;
 
-	_raw_spin_lock(&tas_lock);
-	oldval = *addr;
-	if (!oldval)
-		*addr = 1;
-	_raw_spin_unlock(&tas_lock);
+	/* atomic operation:
+	 *   oldval = *addr; *addr = 1;
+	 */
+	__asm__ __volatile__ (
+		DCACHE_CLEAR("%0", "r4", "%1")
+		"	.fillinsn\n"
+		"1:\n"
+		"	lock	%0, @%1	    ->	unlock	%2, @%1\n"
+		"2:\n"
+		/* NOTE:
+		 *   The m32r processor can accept interrupts only
+		 *   at the 32-bit instruction boundary.
+		 *   So, in the above code, the "unlock" instruction
+		 *   can be executed continuously after the "lock"
+		 *   instruction execution without any interruptions.
+		 */
+		".section .fixup,\"ax\"\n"
+		"	.balign 4\n"
+		"3:	ldi	%0, #%3\n"
+		"	seth	r14, #high(2b)\n"
+		"	or3	r14, r14, #low(2b)\n"
+		"	jmp	r14\n"
+		".previous\n"
+		".section __ex_table,\"a\"\n"
+		"	.balign 4\n"
+		"	.long 1b,3b\n"
+		".previous\n"
+		: "=&r" (oldval)
+		: "r" (addr), "r" (1), "i"(-EFAULT)
+		: "r14", "memory"
+#ifdef CONFIG_CHIP_M32700_TS1
+		  , "r4"
+#endif /* CONFIG_CHIP_M32700_TS1 */
+	);
 
 	return oldval;
 }
-#endif /* CONFIG_SMP */
 
 /*
  * sys_pipe() is the normal C calling standard for creating

From 35e622a67e6d2c5f7e3d3e2da92ebdb2f46db783 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Mon, 20 Feb 2006 12:41:55 +0000
Subject: [PATCH 352/509] [PATCH] H8/300: CONFIG_CONFIG_ doesn't fly.

All actual uses of the symbol refer to CONFIG_SH_STANDARD_BIOS so this
option could never be activated on H8/300.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/h8300/Kconfig.debug | 2 +-
 arch/h8300/defconfig     | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/h8300/Kconfig.debug b/arch/h8300/Kconfig.debug
index 55034d08abff..e0e9bcb015a9 100644
--- a/arch/h8300/Kconfig.debug
+++ b/arch/h8300/Kconfig.debug
@@ -34,7 +34,7 @@ config GDB_DEBUG
 	help
 	  gdb stub exception support
 
-config CONFIG_SH_STANDARD_BIOS
+config SH_STANDARD_BIOS
 	bool "Use gdb protocol serial console"
 	depends on (!H8300H_SIM && !H8S_SIM)
 	help
diff --git a/arch/h8300/defconfig b/arch/h8300/defconfig
index 9d9b491cfc2c..8f1ec3297150 100644
--- a/arch/h8300/defconfig
+++ b/arch/h8300/defconfig
@@ -328,7 +328,7 @@ CONFIG_FULLDEBUG=y
 CONFIG_NO_KERNEL_MSG=y
 # CONFIG_SYSCALL_PRINT is not set
 # CONFIG_GDB_DEBUG is not set
-# CONFIG_CONFIG_SH_STANDARD_BIOS is not set
+# CONFIG_SH_STANDARD_BIOS is not set
 # CONFIG_DEFAULT_CMDLINE is not set
 # CONFIG_BLKDEV_RESERVE is not set
 

From 5914811acf36c3ff091f860a6964808f668f27d0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Bj=F6rn=20Steinbrink?= <B.Steinbrink@gmx.de>
Date: Sat, 18 Feb 2006 18:12:43 +0100
Subject: [PATCH 353/509] [PATCH] kjournald keeps reference to namespace
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

In daemonize() a new thread gets cleaned up and 'merged' with init_task.
The current fs_struct is handled there, but not the current namespace.

This adds the namespace part.

[ Eric Biederman pointed out the namespace wrappers, and also notes that
  we can't ever count on using our parents namespace because we already
  have called exit_fs(), which is the only way to the namespace from a
  process. ]

Signed-off-by: Björn Steinbrink <B.Steinbrink@gmx.de>
Acked-by: Eric Biederman <ebiederm@xmission.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 kernel/exit.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/kernel/exit.c b/kernel/exit.c
index 93cee3671332..531aadca5530 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -360,6 +360,9 @@ void daemonize(const char *name, ...)
 	fs = init_task.fs;
 	current->fs = fs;
 	atomic_inc(&fs->count);
+	exit_namespace(current);
+	current->namespace = init_task.namespace;
+	get_namespace(current->namespace);
  	exit_files(current);
 	current->files = init_task.files;
 	atomic_inc(&current->files->count);

From 36ccf1c0e3917f1f73abc17c38ad704c59f8d1b6 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 14 Feb 2006 21:04:54 +0000
Subject: [PATCH 354/509] [MIPS] Make integer overflow exceptions in kernel
 mode fatal.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/traps.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index c9d2b5147ca3..005debbfbe84 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -3,7 +3,7 @@
  * License.  See the file "COPYING" in the main directory of this archive
  * for more details.
  *
- * Copyright (C) 1994 - 1999, 2000, 01 Ralf Baechle
+ * Copyright (C) 1994 - 1999, 2000, 01, 06 Ralf Baechle
  * Copyright (C) 1995, 1996 Paul M. Antoine
  * Copyright (C) 1998 Ulf Carlsson
  * Copyright (C) 1999 Silicon Graphics, Inc.
@@ -548,6 +548,8 @@ asmlinkage void do_ov(struct pt_regs *regs)
 {
 	siginfo_t info;
 
+	die_if_kernel("Integer overflow", regs);
+
 	info.si_code = FPE_INTOVF;
 	info.si_signo = SIGFPE;
 	info.si_errno = 0;

From 8ecbbcaf08c13c57d6602472478739d64650ee0e Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Tue, 14 Feb 2006 15:57:50 +0900
Subject: [PATCH 355/509] [MIPS] Fixes for uaccess.h with gcc >= 4.0.1

It seems current get_user() incorrectly sign-extend an unsigned int
value on 64bit kernel.  I think this is because '(__typeof__(val))'
cast in final assignment.  I suppose the cast should be
'(__typeof__(*(addr))'.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/uaccess.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/asm-mips/uaccess.h b/include/asm-mips/uaccess.h
index 7a553e9d44d3..b96f3e0f3933 100644
--- a/include/asm-mips/uaccess.h
+++ b/include/asm-mips/uaccess.h
@@ -233,7 +233,7 @@ do {									\
 #define __get_user_check(x,ptr,size)					\
 ({									\
 	long __gu_err = -EFAULT;					\
-	const void __user * __gu_ptr = (ptr);				\
+	const __typeof__(*(ptr)) __user * __gu_ptr = (ptr);		\
 									\
 	if (likely(access_ok(VERIFY_READ,  __gu_ptr, size)))		\
 		__get_user_common((x), size, __gu_ptr);			\
@@ -258,7 +258,7 @@ do {									\
 	: "=r" (__gu_err), "=r" (__gu_tmp)				\
 	: "0" (0), "o" (__m(addr)), "i" (-EFAULT));			\
 									\
-	(val) = (__typeof__(val)) __gu_tmp;				\
+	(val) = (__typeof__(*(addr))) __gu_tmp;				\
 }
 
 /*
@@ -284,7 +284,7 @@ do {									\
 	"	.previous					\n"	\
 	: "=r" (__gu_err), "=&r" (__gu_tmp)				\
 	: "0" (0), "r" (addr), "i" (-EFAULT));				\
-	(val) = __gu_tmp;						\
+	(val) = (__typeof__(*(addr))) __gu_tmp;				\
 }
 
 /*

From 68fa383f3e58b5adf1d8089c93c83ab8d0d00e8d Mon Sep 17 00:00:00 2001
From: Martin Michlmayr <tbm@cyrius.com>
Date: Sat, 18 Feb 2006 14:55:45 +0000
Subject: [PATCH 356/509] [MIPS] Add support for TIF_RESTORE_SIGMASK for
 signal32

Following the recent implementation of TIF_RESTORE_SIGMASK in
arch/mips/kernel/signal.c, 64-bit kernels with 32-bit user-land
compatibility oops when starting init.  signal32.c needs to be
converted to use TIF_RESTORE_SIGMASK too.

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/signal32.c | 68 ++++++++++++++++++++++---------------
 1 file changed, 41 insertions(+), 27 deletions(-)

diff --git a/arch/mips/kernel/signal32.c b/arch/mips/kernel/signal32.c
index 8a8b8dd90417..c07019575214 100644
--- a/arch/mips/kernel/signal32.c
+++ b/arch/mips/kernel/signal32.c
@@ -106,8 +106,6 @@ typedef struct compat_siginfo {
 
 #define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
 
-extern int do_signal32(sigset_t *oldset, struct pt_regs *regs);
-
 /* 32-bit compatibility types */
 
 #define _NSIG_BPW32	32
@@ -198,7 +196,7 @@ __attribute_used__ noinline static int
 _sys32_sigsuspend(nabi_no_regargs struct pt_regs regs)
 {
 	compat_sigset_t *uset;
-	sigset_t newset, saveset;
+	sigset_t newset;
 
 	uset = (compat_sigset_t *) regs.regs[4];
 	if (get_sigset(&newset, uset))
@@ -206,19 +204,15 @@ _sys32_sigsuspend(nabi_no_regargs struct pt_regs regs)
 	sigdelsetmask(&newset, ~_BLOCKABLE);
 
 	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
+	current->saved_sigmask = current->blocked;
 	current->blocked = newset;
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs.regs[2] = EINTR;
-	regs.regs[7] = 1;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal32(&saveset, &regs))
-			return -EINTR;
-	}
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
 }
 
 save_static_function(sys32_rt_sigsuspend);
@@ -226,7 +220,7 @@ __attribute_used__ noinline static int
 _sys32_rt_sigsuspend(nabi_no_regargs struct pt_regs regs)
 {
 	compat_sigset_t *uset;
-	sigset_t newset, saveset;
+	sigset_t newset;
         size_t sigsetsize;
 
 	/* XXX Don't preclude handling different sized sigset_t's.  */
@@ -240,19 +234,15 @@ _sys32_rt_sigsuspend(nabi_no_regargs struct pt_regs regs)
 	sigdelsetmask(&newset, ~_BLOCKABLE);
 
 	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
+	current->saved_sigmask = current->blocked;
 	current->blocked = newset;
         recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs.regs[2] = EINTR;
-	regs.regs[7] = 1;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal32(&saveset, &regs))
-			return -EINTR;
-	}
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
 }
 
 asmlinkage int sys32_sigaction(int sig, const struct sigaction32 *act,
@@ -783,7 +773,7 @@ static inline int handle_signal(unsigned long sig, siginfo_t *info,
 		regs->regs[2] = EINTR;
 		break;
 	case ERESTARTSYS:
-		if(!(ka->sa.sa_flags & SA_RESTART)) {
+		if (!(ka->sa.sa_flags & SA_RESTART)) {
 			regs->regs[2] = EINTR;
 			break;
 		}
@@ -810,9 +800,10 @@ static inline int handle_signal(unsigned long sig, siginfo_t *info,
 	return ret;
 }
 
-int do_signal32(sigset_t *oldset, struct pt_regs *regs)
+int do_signal32(struct pt_regs *regs)
 {
 	struct k_sigaction ka;
+	sigset_t *oldset;
 	siginfo_t info;
 	int signr;
 
@@ -827,12 +818,25 @@ int do_signal32(sigset_t *oldset, struct pt_regs *regs)
 	if (try_to_freeze())
 		goto no_signal;
 
-	if (!oldset)
+	if (test_thread_flag(TIF_RESTORE_SIGMASK))
+		oldset = &current->saved_sigmask;
+	else
 		oldset = &current->blocked;
 
 	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
-	if (signr > 0)
-		return handle_signal(signr, &info, &ka, oldset, regs);
+	if (signr > 0) {
+		/* Whee! Actually deliver the signal. */
+		if (handle_signal(signr, &info, &ka, oldset, regs) == 0) {
+			/*
+			* A signal was successfully delivered; the saved
+			* sigmask will have been stored in the signal frame,
+			* and will be restored by sigreturn, so we can simply
+			* clear the TIF_RESTORE_SIGMASK flag.
+			*/
+			if (test_thread_flag(TIF_RESTORE_SIGMASK))
+				clear_thread_flag(TIF_RESTORE_SIGMASK);
+		}
+	}
 
 no_signal:
 	/*
@@ -853,6 +857,16 @@ no_signal:
 			regs->cp0_epc -= 4;
 		}
 	}
+
+	/*
+	* If there's no signal to deliver, we just put the saved sigmask
+	* back
+	*/
+	if (test_thread_flag(TIF_RESTORE_SIGMASK)) {
+		clear_thread_flag(TIF_RESTORE_SIGMASK);
+		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
+	}
+
 	return 0;
 }
 

From dda73d0bb1d358e4337d2c4da9c61903873664cf Mon Sep 17 00:00:00 2001
From: Martin Michlmayr <tbm@cyrius.com>
Date: Sat, 18 Feb 2006 15:21:30 +0000
Subject: [PATCH 357/509] [MIPS] Make do_signal32 return void.

do_signal has been changed to return void since the "return value is
ignored everywhere".  Convert do_signal32 accordingly.

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/signal32.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/arch/mips/kernel/signal32.c b/arch/mips/kernel/signal32.c
index c07019575214..118a0a9d9a25 100644
--- a/arch/mips/kernel/signal32.c
+++ b/arch/mips/kernel/signal32.c
@@ -4,7 +4,7 @@
  * for more details.
  *
  * Copyright (C) 1991, 1992  Linus Torvalds
- * Copyright (C) 1994 - 2000  Ralf Baechle
+ * Copyright (C) 1994 - 2000, 2006  Ralf Baechle
  * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
  */
 #include <linux/cache.h>
@@ -800,7 +800,7 @@ static inline int handle_signal(unsigned long sig, siginfo_t *info,
 	return ret;
 }
 
-int do_signal32(struct pt_regs *regs)
+void do_signal32(struct pt_regs *regs)
 {
 	struct k_sigaction ka;
 	sigset_t *oldset;
@@ -813,7 +813,7 @@ int do_signal32(struct pt_regs *regs)
 	 * if so.
 	 */
 	if (!user_mode(regs))
-		return 1;
+		return;
 
 	if (try_to_freeze())
 		goto no_signal;
@@ -866,8 +866,6 @@ no_signal:
 		clear_thread_flag(TIF_RESTORE_SIGMASK);
 		sigprocmask(SIG_SETMASK, &current->saved_sigmask, NULL);
 	}
-
-	return 0;
 }
 
 asmlinkage int sys32_rt_sigaction(int sig, const struct sigaction32 *act,

From 304416da860b8cd548e61ee7038f852418008a85 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 18 Feb 2006 18:20:47 +0000
Subject: [PATCH 358/509] [MIPS] Reformat _sys32_rt_sigsuspend with tabs
 instead of space for consistency.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/signal32.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/kernel/signal32.c b/arch/mips/kernel/signal32.c
index 118a0a9d9a25..237cd8a2cd32 100644
--- a/arch/mips/kernel/signal32.c
+++ b/arch/mips/kernel/signal32.c
@@ -221,7 +221,7 @@ _sys32_rt_sigsuspend(nabi_no_regargs struct pt_regs regs)
 {
 	compat_sigset_t *uset;
 	sigset_t newset;
-        size_t sigsetsize;
+	size_t sigsetsize;
 
 	/* XXX Don't preclude handling different sized sigset_t's.  */
 	sigsetsize = regs.regs[5];

From 82ad93f4a002294820e9a5e6f84beef2222a54b7 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 18 Feb 2006 22:47:26 +0000
Subject: [PATCH 359/509] [MIPS] N32: Fix N32 rt_sigtimedwait and rt_sigsuspend
 breakage.

Originally found through an oops in the Gentoo N32 userland build; patch
based on original patch by Daniel Jacobwitz.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/linux32.c     | 19 -------------------
 arch/mips/kernel/scall64-n32.S |  4 ++--
 arch/mips/kernel/signal_n32.c  | 33 +++++++++++++++++++++++++++++++++
 3 files changed, 35 insertions(+), 21 deletions(-)

diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c
index 60353f5acc48..9996b6e84585 100644
--- a/arch/mips/kernel/linux32.c
+++ b/arch/mips/kernel/linux32.c
@@ -1450,25 +1450,6 @@ sys32_timer_create(u32 clock, struct sigevent32 __user *se32, timer_t __user *ti
 	return sys_timer_create(clock, p, timer_id);
 }
 
-asmlinkage long
-sysn32_rt_sigtimedwait(const sigset_t __user *uthese,
-		       siginfo_t __user *uinfo,
-		       const struct compat_timespec __user *uts32,
-		       size_t sigsetsize)
-{
-	struct timespec __user *uts = NULL;
-
-	if (uts32) {
-		struct timespec ts;
-		uts = compat_alloc_user_space(sizeof(struct timespec));
-		if (get_user(ts.tv_sec, &uts32->tv_sec) ||
-		    get_user(ts.tv_nsec, &uts32->tv_nsec) ||
-		    copy_to_user (uts, &ts, sizeof (ts)))
-			return -EFAULT;
-	}
-	return sys_rt_sigtimedwait(uthese, uinfo, uts, sigsetsize);
-}
-
 save_static_function(sys32_clone);
 __attribute_used__ noinline static int
 _sys32_clone(nabi_no_regargs struct pt_regs regs)
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index bc4980cefc8b..d87b5446fa13 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -245,9 +245,9 @@ EXPORT(sysn32_call_table)
 	PTR	sys_capget
 	PTR	sys_capset
 	PTR	sys32_rt_sigpending		/* 6125 */
-	PTR	sysn32_rt_sigtimedwait
+	PTR	compat_sys_rt_sigtimedwait
 	PTR	sys_rt_sigqueueinfo
-	PTR	sys32_rt_sigsuspend
+	PTR	sysn32_rt_sigsuspend
 	PTR	sys32_sigaltstack
 	PTR	compat_sys_utime		/* 6130 */
 	PTR	sys_mknod
diff --git a/arch/mips/kernel/signal_n32.c b/arch/mips/kernel/signal_n32.c
index 5a3776096f07..3e168c08a3a8 100644
--- a/arch/mips/kernel/signal_n32.c
+++ b/arch/mips/kernel/signal_n32.c
@@ -81,6 +81,39 @@ struct rt_sigframe_n32 {
 #endif
 };
 
+extern void sigset_from_compat (sigset_t *set, compat_sigset_t *compat);
+
+save_static_function(sysn32_rt_sigsuspend);
+__attribute_used__ noinline static int
+_sysn32_rt_sigsuspend(nabi_no_regargs struct pt_regs regs)
+{
+	compat_sigset_t __user *unewset, uset;
+	size_t sigsetsize;
+	sigset_t newset;
+
+	/* XXX Don't preclude handling different sized sigset_t's.  */
+	sigsetsize = regs.regs[5];
+	if (sigsetsize != sizeof(sigset_t))
+		return -EINVAL;
+
+	unewset = (compat_sigset_t __user *) regs.regs[4];
+	if (copy_from_user(&uset, unewset, sizeof(uset)))
+		return -EFAULT;
+	sigset_from_compat (&newset, &uset);
+	sigdelsetmask(&newset, ~_BLOCKABLE);
+
+	spin_lock_irq(&current->sighand->siglock);
+	current->saved_sigmask = current->blocked;
+	current->blocked = newset;
+        recalc_sigpending();
+	spin_unlock_irq(&current->sighand->siglock);
+
+	current->state = TASK_INTERRUPTIBLE;
+	schedule();
+	set_thread_flag(TIF_RESTORE_SIGMASK);
+	return -ERESTARTNOHAND;
+}
+
 save_static_function(sysn32_rt_sigreturn);
 __attribute_used__ noinline static void
 _sysn32_rt_sigreturn(nabi_no_regargs struct pt_regs regs)

From f3468e0c34c8de919062582575a01e2434c8e727 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sun, 19 Feb 2006 03:42:11 +0000
Subject: [PATCH 360/509] [MIPS] N32: Make sure pointer is good before passing
 it to sys_waitid().

After all we're calling sys_waitid() with fs set to KERNEL_DS ...

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/linux32.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c
index 9996b6e84585..5f68b220c26d 100644
--- a/arch/mips/kernel/linux32.c
+++ b/arch/mips/kernel/linux32.c
@@ -230,6 +230,9 @@ sysn32_waitid(int which, compat_pid_t pid,
 	long ret;
 	mm_segment_t old_fs = get_fs();
 
+	if (!access_ok(VERIFY_WRITE, uinfo, sizeof(*uinfo)))
+		return -EFAULT;
+
 	set_fs (KERNEL_DS);
 	ret = sys_waitid(which, pid, uinfo, options,
 			 uru ? (struct rusage __user *) &ru : NULL);

From 51939fbb79f66cc471f5bde9845be797ea61ad0b Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 10 Nov 2005 16:35:03 +0000
Subject: [PATCH 361/509] [MIPS] Sibyte: #if CONFIG_* doesn't fly.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/mm/cex-sb1.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/mm/cex-sb1.S b/arch/mips/mm/cex-sb1.S
index 0e71580774ff..e54a62f2807c 100644
--- a/arch/mips/mm/cex-sb1.S
+++ b/arch/mips/mm/cex-sb1.S
@@ -64,7 +64,7 @@ LEAF(except_vec2_sb1)
 	sd	k0,0x170($0)
 	sd	k1,0x178($0)
 
-#if CONFIG_SB1_CEX_ALWAYS_FATAL
+#ifdef CONFIG_SB1_CEX_ALWAYS_FATAL
 	j	handle_vec2_sb1
 	 nop
 #else

From 77607635c3f15e6bf6366e6d7db731a5cb209fb1 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 10 Nov 2005 16:32:14 +0000
Subject: [PATCH 362/509] [MIPS] Sibyte: Config option names shouldn't be
 prefixed with CONFIG_

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/sibyte/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/mips/sibyte/Kconfig b/arch/mips/sibyte/Kconfig
index de46f62ac462..816aee7fcd25 100644
--- a/arch/mips/sibyte/Kconfig
+++ b/arch/mips/sibyte/Kconfig
@@ -102,11 +102,11 @@ config SIMULATION
 	  Build a kernel suitable for running under the GDB simulator.
 	  Primarily adjusts the kernel's notion of time.
 
-config CONFIG_SB1_CEX_ALWAYS_FATAL
+config SB1_CEX_ALWAYS_FATAL
 	bool "All cache exceptions considered fatal (no recovery attempted)"
 	depends on SIBYTE_SB1xxx_SOC
 
-config CONFIG_SB1_CERR_STALL
+config SB1_CERR_STALL
 	bool "Stall (rather than panic) on fatal cache error"
 	depends on SIBYTE_SB1xxx_SOC
 

From 124273773596cbf8aa9c79304b01091d4693f368 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 14 Feb 2006 14:22:10 +0000
Subject: [PATCH 363/509] [MIPS] Follow Uli's latest *at syscall changes.

(This really is only the half of the patch which was forgotten in
326a625748535c4cdb1c632b1dcb07030989a393 ...)

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/unistd.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/include/asm-mips/unistd.h b/include/asm-mips/unistd.h
index 769305d20108..b5c78a4a0192 100644
--- a/include/asm-mips/unistd.h
+++ b/include/asm-mips/unistd.h
@@ -313,7 +313,7 @@
 #define __NR_mknodat			(__NR_Linux + 290)
 #define __NR_fchownat			(__NR_Linux + 291)
 #define __NR_futimesat			(__NR_Linux + 292)
-#define __NR_newfstatat			(__NR_Linux + 293)
+#define __NR_fstatat			(__NR_Linux + 293)
 #define __NR_unlinkat			(__NR_Linux + 294)
 #define __NR_renameat			(__NR_Linux + 295)
 #define __NR_linkat			(__NR_Linux + 296)
@@ -593,7 +593,7 @@
 #define __NR_mknodat			(__NR_Linux + 249)
 #define __NR_fchownat			(__NR_Linux + 250)
 #define __NR_futimesat			(__NR_Linux + 251)
-#define __NR_newfstatat			(__NR_Linux + 252)
+#define __NR_fstatat			(__NR_Linux + 252)
 #define __NR_unlinkat			(__NR_Linux + 253)
 #define __NR_renameat			(__NR_Linux + 254)
 #define __NR_linkat			(__NR_Linux + 255)
@@ -877,7 +877,7 @@
 #define __NR_mknodat			(__NR_Linux + 253)
 #define __NR_fchownat			(__NR_Linux + 254)
 #define __NR_futimesat			(__NR_Linux + 255)
-#define __NR_newfstatat			(__NR_Linux + 256)
+#define __NR_fstatat			(__NR_Linux + 256)
 #define __NR_unlinkat			(__NR_Linux + 257)
 #define __NR_renameat			(__NR_Linux + 258)
 #define __NR_linkat			(__NR_Linux + 259)

From 76e1daee7db153400aaed55646e05a312da353b3 Mon Sep 17 00:00:00 2001
From: Martin Michlmayr <tbm@cyrius.com>
Date: Mon, 20 Feb 2006 04:57:00 +0000
Subject: [PATCH 364/509] [MIPS] Fix compiler warnings in
 arch/mips/sibyte/bcm1480/irq.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix the following compiler warnings:

  CC      arch/mips/sibyte/bcm1480/irq.o
arch/mips/sibyte/bcm1480/irq.c: In function ‘bcm1480_set_affinity’:
arch/mips/sibyte/bcm1480/irq.c:168: warning: ISO C90 forbids mixed declarations and code
arch/mips/sibyte/bcm1480/irq.c: In function ‘ack_bcm1480_irq’:
arch/mips/sibyte/bcm1480/irq.c:230: warning: ISO C90 forbids mixed declarations and code

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/sibyte/bcm1480/irq.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/arch/mips/sibyte/bcm1480/irq.c b/arch/mips/sibyte/bcm1480/irq.c
index b2a1ba5d23df..9cf7d713b13c 100644
--- a/arch/mips/sibyte/bcm1480/irq.c
+++ b/arch/mips/sibyte/bcm1480/irq.c
@@ -139,7 +139,7 @@ void bcm1480_unmask_irq(int cpu, int irq)
 #ifdef CONFIG_SMP
 static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask)
 {
-	int i = 0, old_cpu, cpu, int_on;
+	int i = 0, old_cpu, cpu, int_on, k;
 	u64 cur_ints;
 	irq_desc_t *desc = irq_desc + irq;
 	unsigned long flags;
@@ -165,7 +165,6 @@ static void bcm1480_set_affinity(unsigned int irq, cpumask_t mask)
 		irq_dirty -= BCM1480_NR_IRQS_HALF;
 	}
 
-	int k;
 	for (k=0; k<2; k++) { /* Loop through high and low interrupt mask register */
 		cur_ints = ____raw_readq(IOADDR(A_BCM1480_IMR_MAPPER(old_cpu) + R_BCM1480_IMR_INTERRUPT_MASK_H + (k*BCM1480_IMR_HL_SPACING)));
 		int_on = !(cur_ints & (((u64) 1) << irq_dirty));
@@ -216,6 +215,7 @@ static void ack_bcm1480_irq(unsigned int irq)
 {
 	u64 pending;
 	unsigned int irq_dirty;
+	int k;
 
 	/*
 	 * If the interrupt was an HT interrupt, now is the time to
@@ -227,7 +227,6 @@ static void ack_bcm1480_irq(unsigned int irq)
 	if ((irq_dirty >= BCM1480_NR_IRQS_HALF) && (irq_dirty <= BCM1480_NR_IRQS)) {
 		irq_dirty -= BCM1480_NR_IRQS_HALF;
 	}
-	int k;
 	for (k=0; k<2; k++) { /* Loop through high and low LDT interrupts */
 		pending = __raw_readq(IOADDR(A_BCM1480_IMR_REGISTER(bcm1480_irq_owner[irq],
 						R_BCM1480_IMR_LDT_INTERRUPT_H + (k*BCM1480_IMR_HL_SPACING))));

From 1e35aabab82f8a6f7ab884b642e68be260f92f2c Mon Sep 17 00:00:00 2001
From: Rojhalat Ibrahim <imr@rtschenk.de>
Date: Mon, 20 Feb 2006 13:35:27 +0000
Subject: [PATCH 365/509] [MIPS] Add topology_init.

A recent patch introduced cpu topology in sysfs.  When you run a kernel
with SMP and sysfs enabled, you now get an Oops on boot. The following
patch fixes that by adding topology_init to arch/mips/kernel/smp.c. The
code is copied from arch/s390/kernel/smp.c.

Signed-off-by: Rojhalat Ibrahim <imr@rtschenk.de>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/smp.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 25472fcaf715..5e189862e523 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -29,6 +29,7 @@
 #include <linux/timex.h>
 #include <linux/sched.h>
 #include <linux/cpumask.h>
+#include <linux/cpu.h>
 
 #include <asm/atomic.h>
 #include <asm/cpu.h>
@@ -424,6 +425,25 @@ void flush_tlb_one(unsigned long vaddr)
 	local_flush_tlb_one(vaddr);
 }
 
+static DEFINE_PER_CPU(struct cpu, cpu_devices);
+
+static int __init topology_init(void)
+{
+	int cpu;
+	int ret;
+
+	for_each_cpu(cpu) {
+		ret = register_cpu(&per_cpu(cpu_devices, cpu), cpu, NULL);
+		if (ret)
+			printk(KERN_WARNING "topology_init: register_cpu %d "
+			       "failed (%d)\n", cpu, ret);
+	}
+
+	return 0;
+}
+
+subsys_initcall(topology_init);
+
 EXPORT_SYMBOL(flush_tlb_page);
 EXPORT_SYMBOL(flush_tlb_one);
 EXPORT_SYMBOL(cpu_data);

From 909fa64c8bf0ef20722cddaa402404fb1aa14a4a Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Tue, 21 Feb 2006 01:25:06 +0900
Subject: [PATCH 366/509] [MIPS] jiffies_to_compat_timeval fix

The last argument of div_long_long_rem() must be long.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/binfmt_elfn32.c | 5 +++--
 arch/mips/kernel/binfmt_elfo32.c | 5 +++--
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/arch/mips/kernel/binfmt_elfn32.c b/arch/mips/kernel/binfmt_elfn32.c
index d8e2674a1543..4a9f1ecefaf2 100644
--- a/arch/mips/kernel/binfmt_elfn32.c
+++ b/arch/mips/kernel/binfmt_elfn32.c
@@ -103,8 +103,9 @@ jiffies_to_compat_timeval(unsigned long jiffies, struct compat_timeval *value)
 	 * one divide.
 	 */
 	u64 nsec = (u64)jiffies * TICK_NSEC;
-	value->tv_sec = div_long_long_rem(nsec, NSEC_PER_SEC, &value->tv_usec);
-	value->tv_usec /= NSEC_PER_USEC;
+	long rem;
+	value->tv_sec = div_long_long_rem(nsec, NSEC_PER_SEC, &rem);
+	value->tv_usec = rem / NSEC_PER_USEC;
 }
 
 #define ELF_CORE_EFLAGS EF_MIPS_ABI2
diff --git a/arch/mips/kernel/binfmt_elfo32.c b/arch/mips/kernel/binfmt_elfo32.c
index cec5f327e360..e31813779895 100644
--- a/arch/mips/kernel/binfmt_elfo32.c
+++ b/arch/mips/kernel/binfmt_elfo32.c
@@ -105,8 +105,9 @@ jiffies_to_compat_timeval(unsigned long jiffies, struct compat_timeval *value)
 	 * one divide.
 	 */
 	u64 nsec = (u64)jiffies * TICK_NSEC;
-	value->tv_sec = div_long_long_rem(nsec, NSEC_PER_SEC, &value->tv_usec);
-	value->tv_usec /= NSEC_PER_USEC;
+	long rem;
+	value->tv_sec = div_long_long_rem(nsec, NSEC_PER_SEC, &rem);
+	value->tv_usec = rem / NSEC_PER_USEC;
 }
 
 #undef ELF_CORE_COPY_REGS

From 1e93e70d035a26c0e108f0a9e8f735dabcf948ac Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 21 Feb 2006 00:17:50 +0000
Subject: [PATCH 367/509] [MIPS] Yosemite: Fix build damage by
 dc8f6029cd51af1b148846a32e68d69013a5cc0f.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/pmc-sierra/yosemite/smp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/pmc-sierra/yosemite/smp.c b/arch/mips/pmc-sierra/yosemite/smp.c
index f17f575f58f0..7f8fda962190 100644
--- a/arch/mips/pmc-sierra/yosemite/smp.c
+++ b/arch/mips/pmc-sierra/yosemite/smp.c
@@ -94,7 +94,7 @@ void __init prom_prepare_cpus(unsigned int max_cpus)
 void prom_boot_secondary(int cpu, struct task_struct *idle)
 {
 	unsigned long gp = (unsigned long) task_thread_info(idle);
-	unsigned long sp = __KSTK_TOP(idle);
+	unsigned long sp = __KSTK_TOS(idle);
 
 	secondary_sp = sp;
 	secondary_gp = gp;

From 8db41685c73ad1893d8571861171e183a551e90d Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 21 Feb 2006 13:46:01 +0100
Subject: [PATCH 368/509] [MIPS] Disable CONFIG_ISCSI_TCP; it triggers a gcc
 3.4 endless loop.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/configs/ip27_defconfig | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)

diff --git a/arch/mips/configs/ip27_defconfig b/arch/mips/configs/ip27_defconfig
index e17d3adff021..58c22cd344d3 100644
--- a/arch/mips/configs/ip27_defconfig
+++ b/arch/mips/configs/ip27_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.15-rc2
-# Thu Nov 24 01:06:21 2005
+# Linux kernel version: 2.6.16-rc4
+# Tue Feb 21 13:44:31 2006
 #
 CONFIG_MIPS=y
 
@@ -144,7 +144,6 @@ CONFIG_PREEMPT_BKL=y
 # Code maturity level options
 #
 CONFIG_EXPERIMENTAL=y
-CONFIG_CLEAN_COMPILE=y
 CONFIG_LOCK_KERNEL=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
 
@@ -250,6 +249,7 @@ CONFIG_NET=y
 #
 # Networking options
 #
+# CONFIG_NETDEBUG is not set
 CONFIG_PACKET=y
 CONFIG_PACKET_MMAP=y
 CONFIG_UNIX=y
@@ -289,6 +289,7 @@ CONFIG_TCP_CONG_BIC=y
 # SCTP Configuration (EXPERIMENTAL)
 #
 # CONFIG_IP_SCTP is not set
+
 # CONFIG_ATM is not set
 # CONFIG_BRIDGE is not set
 # CONFIG_VLAN_8021Q is not set
@@ -448,7 +449,7 @@ CONFIG_SCSI_SAS_ATTRS=m
 #
 # SCSI low-level drivers
 #
-CONFIG_ISCSI_TCP=m
+# CONFIG_ISCSI_TCP is not set
 # CONFIG_BLK_DEV_3W_XXXX_RAID is not set
 # CONFIG_SCSI_3W_9XXX is not set
 # CONFIG_SCSI_ACARD is not set
@@ -773,6 +774,10 @@ CONFIG_USB_ARCH_HAS_OHCI=y
 # SN Devices
 #
 
+#
+# EDAC - error detection and reporting (RAS)
+#
+
 #
 # File systems
 #

From b00dc3ad74fdb676552d46ee573b88e927240d0c Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Tue, 21 Feb 2006 23:49:47 +0000
Subject: [PATCH 369/509] [PATCH] tmpfs: fix mount mpol nodelist parsing

I've been dissatisfied with the mpol_nodelist mount option which was
added to tmpfs earlier in -rc.  Replace it by mpol=policy:nodelist.

And it was broken: a nodelist is a comma-separated list of numbers and
ranges; the mount options are a comma-separated list of token=values.
Whoops, blindly strsep'ing on commas doesn't work so well: since we've
no numeric tokens, and unlikely to add them, use that to distinguish.

Move the mpol= parsing to shmem_parse_mpol under CONFIG_NUMA, reject
all its options as invalid if not NUMA.  /proc shows MPOL_PREFERRED
as "prefer", so use that name for the policy instead of "preferred".

Enforce that mpol=default has no nodelist; that mpol=prefer has one
node only; that mpol=bind has a nodelist; but let mpol=interleave use
node_online_map if no nodelist given.  Describe this in tmpfs.txt.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Robin Holt <holt@sgi.com>
Acked-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/filesystems/tmpfs.txt | 21 ++++----
 mm/shmem.c                          | 81 ++++++++++++++++++++++++-----
 2 files changed, 81 insertions(+), 21 deletions(-)

diff --git a/Documentation/filesystems/tmpfs.txt b/Documentation/filesystems/tmpfs.txt
index dbe4d87d2615..8a155418c705 100644
--- a/Documentation/filesystems/tmpfs.txt
+++ b/Documentation/filesystems/tmpfs.txt
@@ -79,15 +79,18 @@ that instance in a system with many cpus making intensive use of it.
 
 
 tmpfs has a mount option to set the NUMA memory allocation policy for
-all files in that instance:
-mpol=interleave		prefers to allocate memory from each node in turn
-mpol=default		prefers to allocate memory from the local node
-mpol=bind		prefers to allocate from mpol_nodelist
-mpol=preferred		prefers to allocate from first node in mpol_nodelist
+all files in that instance (if CONFIG_NUMA is enabled) - which can be
+adjusted on the fly via 'mount -o remount ...'
 
-The following mount option is used in conjunction with mpol=interleave,
-mpol=bind or mpol=preferred:
-mpol_nodelist:	nodelist suitable for parsing with nodelist_parse.
+mpol=default             prefers to allocate memory from the local node
+mpol=prefer:Node         prefers to allocate memory from the given Node
+mpol=bind:NodeList       allocates memory only from nodes in NodeList
+mpol=interleave          prefers to allocate from each node in turn
+mpol=interleave:NodeList allocates from each node of NodeList in turn
+
+NodeList format is a comma-separated list of decimal numbers and ranges,
+a range being two hyphen-separated decimal numbers, the smallest and
+largest node numbers in the range.  For example, mpol=bind:0-3,5,7,9-15
 
 
 To specify the initial root directory you can use the following mount
@@ -109,4 +112,4 @@ RAM/SWAP in 10240 inodes and it is only accessible by root.
 Author:
    Christoph Rohland <cr@sap.com>, 1.12.01
 Updated:
-   Hugh Dickins <hugh@veritas.com>, 13 March 2005
+   Hugh Dickins <hugh@veritas.com>, 19 February 2006
diff --git a/mm/shmem.c b/mm/shmem.c
index f7ac7b812f92..7c455fbaff7b 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -45,6 +45,7 @@
 #include <linux/swapops.h>
 #include <linux/mempolicy.h>
 #include <linux/namei.h>
+#include <linux/ctype.h>
 #include <asm/uaccess.h>
 #include <asm/div64.h>
 #include <asm/pgtable.h>
@@ -874,6 +875,51 @@ redirty:
 }
 
 #ifdef CONFIG_NUMA
+static int shmem_parse_mpol(char *value, int *policy, nodemask_t *policy_nodes)
+{
+	char *nodelist = strchr(value, ':');
+	int err = 1;
+
+	if (nodelist) {
+		/* NUL-terminate policy string */
+		*nodelist++ = '\0';
+		if (nodelist_parse(nodelist, *policy_nodes))
+			goto out;
+	}
+	if (!strcmp(value, "default")) {
+		*policy = MPOL_DEFAULT;
+		/* Don't allow a nodelist */
+		if (!nodelist)
+			err = 0;
+	} else if (!strcmp(value, "prefer")) {
+		*policy = MPOL_PREFERRED;
+		/* Insist on a nodelist of one node only */
+		if (nodelist) {
+			char *rest = nodelist;
+			while (isdigit(*rest))
+				rest++;
+			if (!*rest)
+				err = 0;
+		}
+	} else if (!strcmp(value, "bind")) {
+		*policy = MPOL_BIND;
+		/* Insist on a nodelist */
+		if (nodelist)
+			err = 0;
+	} else if (!strcmp(value, "interleave")) {
+		*policy = MPOL_INTERLEAVE;
+		/* Default to nodes online if no nodelist */
+		if (!nodelist)
+			*policy_nodes = node_online_map;
+		err = 0;
+	}
+out:
+	/* Restore string for error message */
+	if (nodelist)
+		*--nodelist = ':';
+	return err;
+}
+
 static struct page *shmem_swapin_async(struct shared_policy *p,
 				       swp_entry_t entry, unsigned long idx)
 {
@@ -926,6 +972,11 @@ shmem_alloc_page(gfp_t gfp, struct shmem_inode_info *info,
 	return page;
 }
 #else
+static inline int shmem_parse_mpol(char *value, int *policy, nodemask_t *policy_nodes)
+{
+	return 1;
+}
+
 static inline struct page *
 shmem_swapin(struct shmem_inode_info *info,swp_entry_t entry,unsigned long idx)
 {
@@ -1859,7 +1910,23 @@ static int shmem_parse_options(char *options, int *mode, uid_t *uid,
 {
 	char *this_char, *value, *rest;
 
-	while ((this_char = strsep(&options, ",")) != NULL) {
+	while (options != NULL) {
+		this_char = options;
+		for (;;) {
+			/*
+			 * NUL-terminate this option: unfortunately,
+			 * mount options form a comma-separated list,
+			 * but mpol's nodelist may also contain commas.
+			 */
+			options = strchr(options, ',');
+			if (options == NULL)
+				break;
+			options++;
+			if (!isdigit(*options)) {
+				options[-1] = '\0';
+				break;
+			}
+		}
 		if (!*this_char)
 			continue;
 		if ((value = strchr(this_char,'=')) != NULL) {
@@ -1910,18 +1977,8 @@ static int shmem_parse_options(char *options, int *mode, uid_t *uid,
 			if (*rest)
 				goto bad_val;
 		} else if (!strcmp(this_char,"mpol")) {
-			if (!strcmp(value,"default"))
-				*policy = MPOL_DEFAULT;
-			else if (!strcmp(value,"preferred"))
-				*policy = MPOL_PREFERRED;
-			else if (!strcmp(value,"bind"))
-				*policy = MPOL_BIND;
-			else if (!strcmp(value,"interleave"))
-				*policy = MPOL_INTERLEAVE;
-			else
+			if (shmem_parse_mpol(value,policy,policy_nodes))
 				goto bad_val;
-		} else if (!strcmp(this_char,"mpol_nodelist")) {
-			nodelist_parse(value, *policy_nodes);
 		} else {
 			printk(KERN_ERR "tmpfs: Bad mount option %s\n",
 			       this_char);

From 5bd546aa78b5d74f3162815e41940f862215d9e3 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Fri, 17 Feb 2006 20:23:29 +0000
Subject: [PATCH 370/509] [MMC] Fix mmc_cmd_type() mask

It's MMC_CMD_MASK not MMC_CMD_TYPE.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/linux/mmc/mmc.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/mmc/mmc.h b/include/linux/mmc/mmc.h
index f38872abc126..bdc556d88498 100644
--- a/include/linux/mmc/mmc.h
+++ b/include/linux/mmc/mmc.h
@@ -49,7 +49,7 @@ struct mmc_command {
 /*
  * These are the command types.
  */
-#define mmc_cmd_type(cmd)	((cmd)->flags & MMC_CMD_TYPE)
+#define mmc_cmd_type(cmd)	((cmd)->flags & MMC_CMD_MASK)
 
 	unsigned int		retries;	/* max number of retries */
 	unsigned int		error;		/* command error */

From 31867499b21b2374eb0cc6b3d1ea6b4ade4d1cc2 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Sun, 19 Feb 2006 19:53:56 +0000
Subject: [PATCH 371/509] [ARM] Add panic-on-oops support

Although you could ask the kernel for panic-on-oops, it remained
non-functional because the architecture specific code fragment had
not been implemented.  Add it, so it works as advertised.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/traps.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/arch/arm/kernel/traps.c b/arch/arm/kernel/traps.c
index 10235b01582e..03924bcc6129 100644
--- a/arch/arm/kernel/traps.c
+++ b/arch/arm/kernel/traps.c
@@ -19,6 +19,7 @@
 #include <linux/personality.h>
 #include <linux/ptrace.h>
 #include <linux/kallsyms.h>
+#include <linux/delay.h>
 #include <linux/init.h>
 
 #include <asm/atomic.h>
@@ -231,6 +232,13 @@ NORET_TYPE void die(const char *str, struct pt_regs *regs, int err)
 	__die(str, err, thread, regs);
 	bust_spinlocks(0);
 	spin_unlock_irq(&die_lock);
+
+	if (panic_on_oops) {
+		printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n");
+		ssleep(5);
+		panic("Fatal exception");
+	}
+
 	do_exit(SIGSEGV);
 }
 

From a43c7ff8bafe841502cab52b7795e2825c2e42c4 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Mon, 20 Feb 2006 10:18:38 +0000
Subject: [PATCH 372/509] [ARM] Update mach-types

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/tools/mach-types | 35 +++++++++++++++++++++++++++++++++--
 1 file changed, 33 insertions(+), 2 deletions(-)

diff --git a/arch/arm/tools/mach-types b/arch/arm/tools/mach-types
index d0f9bb5e9023..8ab5300dcb94 100644
--- a/arch/arm/tools/mach-types
+++ b/arch/arm/tools/mach-types
@@ -12,7 +12,7 @@
 #
 #   http://www.arm.linux.org.uk/developer/machines/?action=new
 #
-# Last update: Mon Jan 9 12:56:42 2006
+# Last update: Mon Feb 20 10:18:02 2006
 #
 # machine_is_xxx	CONFIG_xxxx		MACH_TYPE_xxx		number
 #
@@ -904,7 +904,7 @@ wg302v2			MACH_WG302V2		WG302V2			890
 eb42x			MACH_EB42X		EB42X			891
 iq331es			MACH_IQ331ES		IQ331ES			892
 cosydsp			MACH_COSYDSP		COSYDSP			893
-uplat7d			MACH_UPLAT7D		UPLAT7D			894
+uplat7d_proto		MACH_UPLAT7D		UPLAT7D			894
 ptdavinci		MACH_PTDAVINCI		PTDAVINCI		895
 mbus			MACH_MBUS		MBUS			896
 nadia2vb		MACH_NADIA2VB		NADIA2VB		897
@@ -938,3 +938,34 @@ auckland		MACH_AUCKLAND		AUCKLAND		924
 ak3220m			MACH_AK3320M		AK3320M			925
 duramax			MACH_DURAMAX		DURAMAX			926
 n35			MACH_N35		N35			927
+pronghorn		MACH_PRONGHORN		PRONGHORN		928
+fundy			MACH_FUNDY		FUNDY			929
+logicpd_pxa270		MACH_LOGICPD_PXA270	LOGICPD_PXA270		930
+cpu777			MACH_CPU777		CPU777			931
+simicon9201		MACH_SIMICON9201	SIMICON9201		932
+leap2_hpm		MACH_LEAP2_HPM		LEAP2_HPM		933
+cm922txa10		MACH_CM922TXA10		CM922TXA10		934
+sandgate		MACH_PXA		PXA			935
+sandgate2		MACH_SANDGATE2		SANDGATE2		936
+sandgate2g		MACH_SANDGATE2G		SANDGATE2G		937
+sandgate2p		MACH_SANDGATE2P		SANDGATE2P		938
+fred_jack		MACH_FRED_JACK		FRED_JACK		939
+ttg_color1		MACH_TTG_COLOR1		TTG_COLOR1		940
+nxeb500hmi		MACH_NXEB500HMI		NXEB500HMI		941
+netdcu8			MACH_NETDCU8		NETDCU8			942
+ml675050_cpu_boa	MACH_ML675050_CPU_BOA	ML675050_CPU_BOA	943
+ng_fvx538		MACH_NG_FVX538		NG_FVX538		944
+ng_fvs338		MACH_NG_FVS338		NG_FVS338		945
+pnx4103			MACH_PNX4103		PNX4103			946
+hesdb			MACH_HESDB		HESDB			947
+xsilo			MACH_XSILO		XSILO			948
+espresso		MACH_ESPRESSO		ESPRESSO		949
+emlc			MACH_EMLC		EMLC			950
+sisteron		MACH_SISTERON		SISTERON		951
+rx1950			MACH_RX1950		RX1950			952
+tsc_venus		MACH_TSC_VENUS		TSC_VENUS		953
+ds101j			MACH_DS101J		DS101J			954
+mxc300_30ads		MACH_MXC30030ADS	MXC30030ADS		955
+fujitsu_wimaxsoc	MACH_FUJITSU_WIMAXSOC	FUJITSU_WIMAXSOC	956
+dualpcmodem		MACH_DUALPCMODEM	DUALPCMODEM		957
+gesbc9312		MACH_GESBC9312		GESBC9312		958

From 102d60a2d8a6b54a20317a855dca3b598a2fd581 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 22 Feb 2006 23:43:40 +1100
Subject: [PATCH 373/509] [PATCH] padlock: Fix typo that broke 256-bit keys

A typo crept into the le32_to_cpu patch which broke 256-bit keys
in the padlock driver.  The following patch based on observations
by Michael Heyse fixes the problem.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/crypto/padlock-aes.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/crypto/padlock-aes.c b/drivers/crypto/padlock-aes.c
index 64819aa7cac4..0c08c58252be 100644
--- a/drivers/crypto/padlock-aes.c
+++ b/drivers/crypto/padlock-aes.c
@@ -348,10 +348,10 @@ aes_set_key(void *ctx_arg, const uint8_t *in_key, unsigned int key_len, uint32_t
 		break;
 
 	case 32:
-		E_KEY[4] = le32_to_cpu(in_key[4]);
-		E_KEY[5] = le32_to_cpu(in_key[5]);
-		E_KEY[6] = le32_to_cpu(in_key[6]);
-		t = E_KEY[7] = le32_to_cpu(in_key[7]);
+		E_KEY[4] = le32_to_cpu(key[4]);
+		E_KEY[5] = le32_to_cpu(key[5]);
+		E_KEY[6] = le32_to_cpu(key[6]);
+		t = E_KEY[7] = le32_to_cpu(key[7]);
 		for (i = 0; i < 7; ++i)
 			loop8 (i);
 		break;

From fa675765afed59bb89adba3369094ebd428b930b Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 22 Feb 2006 09:39:02 -0800
Subject: [PATCH 374/509] Revert mount/umount uevent removal

This change reverts the 033b96fd30db52a710d97b06f87d16fc59fee0f1 commit
from Kay Sievers that removed the mount/umount uevents from the kernel.
Some older versions of HAL still depend on these events to detect when a
new device has been mounted.  These events are not correctly emitted,
and are broken by design, and so, should not be relied upon by any
future program.  Instead, the /proc/mounts file should be polled to
properly detect this kind of event.

A feature-removal-schedule.txt entry has been added, noting when this
interface will be removed from the kernel.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
 Documentation/feature-removal-schedule.txt |  9 +++++++++
 fs/super.c                                 | 15 ++++++++++++++-
 include/linux/kobject.h                    |  6 ++++--
 lib/kobject_uevent.c                       |  4 ++++
 4 files changed, 31 insertions(+), 3 deletions(-)

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index b730d765b525..be5ae600f533 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -171,3 +171,12 @@ Why:	The ISA interface is faster and should be always available. The I2C
 	probing is also known to cause trouble in at least one case (see
 	bug #5889.)
 Who:	Jean Delvare <khali@linux-fr.org>
+
+---------------------------
+
+What:	mount/umount uevents
+When:	February 2007
+Why:	These events are not correct, and do not properly let userspace know
+	when a file system has been mounted or unmounted.  Userspace should
+	poll the /proc/mounts file instead to detect this properly.
+Who:	Greg Kroah-Hartman <gregkh@suse.de>
diff --git a/fs/super.c b/fs/super.c
index 30294218fa63..e20b5580afd5 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -666,6 +666,16 @@ static int test_bdev_super(struct super_block *s, void *data)
 	return (void *)s->s_bdev == data;
 }
 
+static void bdev_uevent(struct block_device *bdev, enum kobject_action action)
+{
+	if (bdev->bd_disk) {
+		if (bdev->bd_part)
+			kobject_uevent(&bdev->bd_part->kobj, action);
+		else
+			kobject_uevent(&bdev->bd_disk->kobj, action);
+	}
+}
+
 struct super_block *get_sb_bdev(struct file_system_type *fs_type,
 	int flags, const char *dev_name, void *data,
 	int (*fill_super)(struct super_block *, void *, int))
@@ -707,8 +717,10 @@ struct super_block *get_sb_bdev(struct file_system_type *fs_type,
 			up_write(&s->s_umount);
 			deactivate_super(s);
 			s = ERR_PTR(error);
-		} else
+		} else {
 			s->s_flags |= MS_ACTIVE;
+			bdev_uevent(bdev, KOBJ_MOUNT);
+		}
 	}
 
 	return s;
@@ -724,6 +736,7 @@ void kill_block_super(struct super_block *sb)
 {
 	struct block_device *bdev = sb->s_bdev;
 
+	bdev_uevent(bdev, KOBJ_UMOUNT);
 	generic_shutdown_super(sb);
 	sync_blockdev(bdev);
 	close_bdev_excl(bdev);
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 2a8d8da70961..c374b5fa8d3b 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -41,8 +41,10 @@ enum kobject_action {
 	KOBJ_ADD	= (__force kobject_action_t) 0x01,	/* exclusive to core */
 	KOBJ_REMOVE	= (__force kobject_action_t) 0x02,	/* exclusive to core */
 	KOBJ_CHANGE	= (__force kobject_action_t) 0x03,	/* device state change */
-	KOBJ_OFFLINE	= (__force kobject_action_t) 0x04,	/* device offline */
-	KOBJ_ONLINE	= (__force kobject_action_t) 0x05,	/* device online */
+	KOBJ_MOUNT	= (__force kobject_action_t) 0x04,	/* mount event for block devices (broken) */
+	KOBJ_UMOUNT	= (__force kobject_action_t) 0x05,	/* umount event for block devices (broken) */
+	KOBJ_OFFLINE	= (__force kobject_action_t) 0x06,	/* device offline */
+	KOBJ_ONLINE	= (__force kobject_action_t) 0x07,	/* device online */
 };
 
 struct kobject {
diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c
index 1b1985c136ec..086a0c6e888e 100644
--- a/lib/kobject_uevent.c
+++ b/lib/kobject_uevent.c
@@ -38,6 +38,10 @@ static char *action_to_string(enum kobject_action action)
 		return "remove";
 	case KOBJ_CHANGE:
 		return "change";
+	case KOBJ_MOUNT:
+		return "mount";
+	case KOBJ_UMOUNT:
+		return "umount";
 	case KOBJ_OFFLINE:
 		return "offline";
 	case KOBJ_ONLINE:

From c27a2164a32cfda80dc1647638a940103669af3d Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Wed, 22 Feb 2006 19:51:38 +0000
Subject: [PATCH 375/509] [ARM] 3340/1: Fix the PCI setup for direct master
 access to SDRAM

Patch from Catalin Marinas

The initial code did not configure the inbound memory windows for direct
master access to the SDRAM. This patch creates a 1:1 mapping between the
Versatile/PB PCI memory windows and its SDRAM. Note that an updated FPGA
image is needed for Versatile/PB since the original windows were 1MB and
not able to cover the whole SDRAM (now extended to 256MB). The patch also
fixes the PCI IRQ mapping for slot #2.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-versatile/pci.c | 93 ++++++++++++++++++-----------------
 1 file changed, 48 insertions(+), 45 deletions(-)

diff --git a/arch/arm/mach-versatile/pci.c b/arch/arm/mach-versatile/pci.c
index b80d57d51699..722fbabc9cfb 100644
--- a/arch/arm/mach-versatile/pci.c
+++ b/arch/arm/mach-versatile/pci.c
@@ -240,6 +240,14 @@ int __init pci_versatile_setup(int nr, struct pci_sys_data *sys)
         int i;
         int myslot = -1;
 	unsigned long val;
+	void __iomem *local_pci_cfg_base;
+
+	val = __raw_readl(SYS_PCICTL);
+	if (!(val & 1)) {
+		printk("Not plugged into PCI backplane!\n");
+		ret = -EIO;
+		goto out;
+	}
 
 	if (nr == 0) {
 		sys->mem_offset = 0;
@@ -253,48 +261,45 @@ int __init pci_versatile_setup(int nr, struct pci_sys_data *sys)
 		goto out;
 	}
 
-	__raw_writel(VERSATILE_PCI_MEM_BASE0 >> 28,PCI_IMAP0);
-	__raw_writel(VERSATILE_PCI_MEM_BASE1 >> 28,PCI_IMAP1);
-	__raw_writel(VERSATILE_PCI_MEM_BASE2 >> 28,PCI_IMAP2);
-
-	__raw_writel(1, SYS_PCICTL);
-
-	val = __raw_readl(SYS_PCICTL);
-	if (!(val & 1)) {
-		printk("Not plugged into PCI backplane!\n");
-		ret = -EIO;
-		goto out;
-	}
-
 	/*
 	 *  We need to discover the PCI core first to configure itself
 	 *  before the main PCI probing is performed
 	 */
-	for (i=0; i<32; i++) {
+	for (i=0; i<32; i++)
 		if ((__raw_readl(VERSATILE_PCI_VIRT_BASE+(i<<11)+DEVICE_ID_OFFSET) == VP_PCI_DEVICE_ID) &&
 		    (__raw_readl(VERSATILE_PCI_VIRT_BASE+(i<<11)+CLASS_ID_OFFSET) == VP_PCI_CLASS_ID)) {
 			myslot = i;
-
-			__raw_writel(myslot, PCI_SELFID);
-			val = __raw_readl(VERSATILE_PCI_CFG_VIRT_BASE+(myslot<<11)+CSR_OFFSET);
-			val |= (1<<2);
-			__raw_writel(val, VERSATILE_PCI_CFG_VIRT_BASE+(myslot<<11)+CSR_OFFSET);
 			break;
 		}
-	}
 
 	if (myslot == -1) {
 		printk("Cannot find PCI core!\n");
 		ret = -EIO;
-	} else {
-		printk("PCI core found (slot %d)\n",myslot);
-		/* Do not to map Versatile FPGA PCI device
-		   into memory space as we are short of
-		   mappable memory */
-		pci_slot_ignore |= (1 << myslot);
-		ret = 1;
+		goto out;
 	}
 
+	printk("PCI core found (slot %d)\n",myslot);
+
+	__raw_writel(myslot, PCI_SELFID);
+	local_pci_cfg_base = (void *) VERSATILE_PCI_CFG_VIRT_BASE + (myslot << 11);
+
+	val = __raw_readl(local_pci_cfg_base + CSR_OFFSET);
+	val |= PCI_COMMAND_MEMORY | PCI_COMMAND_MASTER | PCI_COMMAND_INVALIDATE;
+	__raw_writel(val, local_pci_cfg_base + CSR_OFFSET);
+
+	/*
+	 * Configure the PCI inbound memory windows to be 1:1 mapped to SDRAM
+	 */
+	__raw_writel(PHYS_OFFSET, local_pci_cfg_base + PCI_BASE_ADDRESS_0);
+	__raw_writel(PHYS_OFFSET, local_pci_cfg_base + PCI_BASE_ADDRESS_1);
+	__raw_writel(PHYS_OFFSET, local_pci_cfg_base + PCI_BASE_ADDRESS_2);
+
+	/*
+	 * Do not to map Versatile FPGA PCI device into memory space
+	 */
+	pci_slot_ignore |= (1 << myslot);
+	ret = 1;
+
  out:
 	return ret;
 }
@@ -305,18 +310,18 @@ struct pci_bus *pci_versatile_scan_bus(int nr, struct pci_sys_data *sys)
 	return pci_scan_bus(sys->busnr, &pci_versatile_ops, sys);
 }
 
-/*
- * V3_LB_BASE? - local bus address
- * V3_LB_MAP?  - pci bus address
- */
 void __init pci_versatile_preinit(void)
 {
-}
+	__raw_writel(VERSATILE_PCI_MEM_BASE0 >> 28, PCI_IMAP0);
+	__raw_writel(VERSATILE_PCI_MEM_BASE1 >> 28, PCI_IMAP1);
+	__raw_writel(VERSATILE_PCI_MEM_BASE2 >> 28, PCI_IMAP2);
 
-void __init pci_versatile_postinit(void)
-{
-}
+	__raw_writel(PHYS_OFFSET >> 28, PCI_SMAP0);
+	__raw_writel(PHYS_OFFSET >> 28, PCI_SMAP1);
+	__raw_writel(PHYS_OFFSET >> 28, PCI_SMAP2);
 
+	__raw_writel(1, SYS_PCICTL);
+}
 
 /*
  * map the specified device/slot/pin to an IRQ.   Different backplanes may need to modify this.
@@ -326,16 +331,15 @@ static int __init versatile_map_irq(struct pci_dev *dev, u8 slot, u8 pin)
 	int irq;
 	int devslot = PCI_SLOT(dev->devfn);
 
-	/* slot,  pin,  irq
-	    24	  1	27
-	    25    1	28	untested
-	    26	  1	29
-	    27    1	30	untested
-	*/
+	/* slot,  pin,	irq
+	 *  24     1     27
+	 *  25     1     28
+	 *  26     1     29
+	 *  27     1     30
+	 */
+	irq = 27 + ((slot + pin - 1) & 3);
 
-	irq = 27 + ((slot + pin + 2) % 3);	/* Fudged */
-
-	printk("map irq: slot %d, pin %d, devslot %d, irq: %d\n",slot,pin,devslot,irq);
+	printk("PCI map irq: slot %d, pin %d, devslot %d, irq: %d\n",slot,pin,devslot,irq);
 
 	return irq;
 }
@@ -347,7 +351,6 @@ static struct hw_pci versatile_pci __initdata = {
 	.setup			= pci_versatile_setup,
 	.scan			= pci_versatile_scan_bus,
 	.preinit		= pci_versatile_preinit,
-	.postinit		= pci_versatile_postinit,
 };
 
 static int __init versatile_pci_init(void)

From d7353b25c855b3a8103647503deaa98f512bd439 Mon Sep 17 00:00:00 2001
From: Alessandro Zummo <azummo-armlinux@towertech.it>
Date: Wed, 22 Feb 2006 21:12:05 +0000
Subject: [PATCH 376/509] [ARM] 3342/1: NSLU2: Protect power button init
 routine with machine_is_nslu2()

Patch from Alessandro Zummo

The power button exit routine for the Linksys NSLU2 was not protected by
a machine_is_nslu2(). This patch fixes it.

Signed-off-by: Rod Whitby <rod@whitby.id.au>
Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp4xx/nslu2-power.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/arm/mach-ixp4xx/nslu2-power.c b/arch/arm/mach-ixp4xx/nslu2-power.c
index b0ad9e901f6e..d80c362bc539 100644
--- a/arch/arm/mach-ixp4xx/nslu2-power.c
+++ b/arch/arm/mach-ixp4xx/nslu2-power.c
@@ -77,6 +77,9 @@ static int __init nslu2_power_init(void)
 
 static void __exit nslu2_power_exit(void)
 {
+	if (!(machine_is_nslu2()))
+		return;
+
 	free_irq(NSLU2_RB_IRQ, NULL);
 	free_irq(NSLU2_PB_IRQ, NULL);
 }

From af898b8f602441a3bebe918a3b26adc92b30762e Mon Sep 17 00:00:00 2001
From: Alessandro Zummo <azummo-armlinux@towertech.it>
Date: Wed, 22 Feb 2006 21:12:06 +0000
Subject: [PATCH 377/509] [ARM] 3343/1: NAS100d: Fix incorrect I2C pin
 assignment

Patch from Alessandro Zummo

The I2C pin assignment for the Iomega NAS100d board was incorrect.  This
patch fixes it.  The correct assignment has now been tested using the
new RTC class and a new driver for the RTC on the NAS100d.

Signed-off-by: Rod Whitby <rod@whitby.id.au>
Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/asm-arm/arch-ixp4xx/nas100d.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/asm-arm/arch-ixp4xx/nas100d.h b/include/asm-arm/arch-ixp4xx/nas100d.h
index 51ac0180427c..84467a5190d0 100644
--- a/include/asm-arm/arch-ixp4xx/nas100d.h
+++ b/include/asm-arm/arch-ixp4xx/nas100d.h
@@ -19,8 +19,8 @@
 #error "Do not include this directly, instead #include <asm/hardware.h>"
 #endif
 
-#define NAS100D_SDA_PIN		6
-#define NAS100D_SCL_PIN		5
+#define NAS100D_SDA_PIN		5
+#define NAS100D_SCL_PIN		6
 
 /*
  * NAS100D PCI IRQs

From bc66d4496f106a8e3a936dc24c9965a2f9c13e50 Mon Sep 17 00:00:00 2001
From: Alessandro Zummo <azummo-armlinux@towertech.it>
Date: Wed, 22 Feb 2006 21:12:06 +0000
Subject: [PATCH 378/509] [ARM] 3344/1: NSLU2: beeper support

Patch from Alessandro Zummo

This patch adds support for the beeper
embedded in the NSLU2 to the machine setup code.

Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Rod Whitby <rod@whitby.id.au>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp4xx/nslu2-setup.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/arm/mach-ixp4xx/nslu2-setup.c b/arch/arm/mach-ixp4xx/nslu2-setup.c
index f260a9d34f70..55411f21d838 100644
--- a/arch/arm/mach-ixp4xx/nslu2-setup.c
+++ b/arch/arm/mach-ixp4xx/nslu2-setup.c
@@ -50,6 +50,12 @@ static struct platform_device nslu2_i2c_controller = {
 	.num_resources		= 0,
 };
 
+static struct platform_device nslu2_beeper = {
+	.name			= "ixp4xx-beeper",
+	.id			= NSLU2_GPIO_BUZZ,
+	.num_resources		= 0,
+};
+
 static struct resource nslu2_uart_resources[] = {
 	{
 		.start		= IXP4XX_UART1_BASE_PHYS,
@@ -97,6 +103,7 @@ static struct platform_device *nslu2_devices[] __initdata = {
 	&nslu2_i2c_controller,
 	&nslu2_flash,
 	&nslu2_uart,
+	&nslu2_beeper,
 };
 
 static void nslu2_power_off(void)

From 75d2f18088ded458f5bc4014b6c4e2d9651d41d4 Mon Sep 17 00:00:00 2001
From: Uli Luckas <u.luckas@road-gmbh.de>
Date: Wed, 22 Feb 2006 21:12:07 +0000
Subject: [PATCH 379/509] [ARM] 3345/1: Fix interday RTC alarms

Patch from Uli Luckas

This is a bugfix.

The comment in arch/arm/common/rtctime.c explains it:
* FIXME: for now, we just copy the alarm time because we're lazy (and
* is therefore buggy - setting a 10am alarm at 8pm will not result in
* the alarm triggering.)

This patch adds one day to the alarm iff the alarm wrapped beyond midnight and therefore appears to be in the past.

Signed-off-by: Uli Luckas <u.luckas@road-gmbh.de>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/common/rtctime.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/arch/arm/common/rtctime.c b/arch/arm/common/rtctime.c
index 48b1e19b131f..e851d86c212c 100644
--- a/arch/arm/common/rtctime.c
+++ b/arch/arm/common/rtctime.c
@@ -128,19 +128,27 @@ EXPORT_SYMBOL(rtc_tm_to_time);
 /*
  * Calculate the next alarm time given the requested alarm time mask
  * and the current time.
- *
- * FIXME: for now, we just copy the alarm time because we're lazy (and
- * is therefore buggy - setting a 10am alarm at 8pm will not result in
- * the alarm triggering.)
  */
 void rtc_next_alarm_time(struct rtc_time *next, struct rtc_time *now, struct rtc_time *alrm)
 {
+	unsigned long next_time;
+	unsigned long now_time;
+
 	next->tm_year = now->tm_year;
 	next->tm_mon = now->tm_mon;
 	next->tm_mday = now->tm_mday;
 	next->tm_hour = alrm->tm_hour;
 	next->tm_min = alrm->tm_min;
 	next->tm_sec = alrm->tm_sec;
+
+	rtc_tm_to_time(now, &now_time);
+	rtc_tm_to_time(next, &next_time);
+
+	if (next_time < now_time) {
+		/* Advance one day */
+		next_time += 60 * 60 * 24;
+		rtc_time_to_tm(next_time, next);
+	}
 }
 
 static inline int rtc_read_time(struct rtc_ops *ops, struct rtc_time *tm)

From 43cc19816b3fc5286258e6f5e43ef4ead458f9a3 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Wed, 22 Feb 2006 21:13:28 +0000
Subject: [PATCH 380/509] [ARM] CONFIG_CPU_MPCORE -> CONFIG_CPU_32v6K

CONFIG_CPU_MPCORE has never been a configuration symbol - it should
be CONFIG_CPU_32v6K.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/kernel/entry-armv.S | 2 +-
 arch/arm/mm/abort-ev6.S      | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
index 964cd717506b..ec48d70c6d8b 100644
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -566,7 +566,7 @@ ENTRY(__switch_to)
 	ldr	r6, [r2, #TI_CPU_DOMAIN]!
 #endif
 #if __LINUX_ARM_ARCH__ >= 6
-#ifdef CONFIG_CPU_MPCORE
+#ifdef CONFIG_CPU_32v6K
 	clrex
 #else
 	strex	r5, r4, [ip]			@ Clear exclusive monitor
diff --git a/arch/arm/mm/abort-ev6.S b/arch/arm/mm/abort-ev6.S
index dbd346033122..8a7f65ba14b7 100644
--- a/arch/arm/mm/abort-ev6.S
+++ b/arch/arm/mm/abort-ev6.S
@@ -20,7 +20,7 @@
  */
 	.align	5
 ENTRY(v6_early_abort)
-#ifdef CONFIG_CPU_MPCORE
+#ifdef CONFIG_CPU_32v6K
 	clrex
 #else
 	strex	r0, r1, [sp]			@ Clear the exclusive monitor

From df666b9c510fd27fd3b1afd9ddfa1eaa62ce12b3 Mon Sep 17 00:00:00 2001
From: Andrew Victor <andrew@sanpeople.com>
Date: Wed, 22 Feb 2006 21:23:35 +0000
Subject: [PATCH 381/509] [ARM] 3325/2: GPIO function to control multi-drive
 (open collector) capability

Patch from Andrew Victor

This patch adds the at91_set_multi_drive() function to enable/disable
the multi-drive (open collector) pin capability on the AT91RM9200
processor.

This is necessary to fix the UDC (USB Gadget) driver for the AT91RM9200
board as it will not allow the board reset line to be pulled low if the
pullup is not driven as an open collector output as the boards are wired
to the USB connector on both the DK/EK.

This version of the patch updates it to 2.6.16-rc4.
Orignal patch by Jeff Warren.

Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91rm9200/devices.c     |  4 +++-
 arch/arm/mach-at91rm9200/gpio.c        | 17 +++++++++++++++++
 include/asm-arm/arch-at91rm9200/gpio.h |  1 +
 3 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/arch/arm/mach-at91rm9200/devices.c b/arch/arm/mach-at91rm9200/devices.c
index 8df3e5245651..57eedd5beaf6 100644
--- a/arch/arm/mach-at91rm9200/devices.c
+++ b/arch/arm/mach-at91rm9200/devices.c
@@ -100,8 +100,10 @@ void __init at91_add_device_udc(struct at91_udc_data *data)
 		at91_set_gpio_input(data->vbus_pin, 0);
 		at91_set_deglitch(data->vbus_pin, 1);
 	}
-	if (data->pullup_pin)
+	if (data->pullup_pin) {
 		at91_set_gpio_output(data->pullup_pin, 0);
+		at91_set_multi_drive(data->pullup_pin, 1);
+	}
 
 	udc_data = *data;
 	platform_device_register(&at91rm9200_udc_device);
diff --git a/arch/arm/mach-at91rm9200/gpio.c b/arch/arm/mach-at91rm9200/gpio.c
index 2fd2ef583e4d..a9f718bf8ba8 100644
--- a/arch/arm/mach-at91rm9200/gpio.c
+++ b/arch/arm/mach-at91rm9200/gpio.c
@@ -159,6 +159,23 @@ int __init_or_module at91_set_deglitch(unsigned pin, int is_on)
 }
 EXPORT_SYMBOL(at91_set_deglitch);
 
+/*
+ * enable/disable the multi-driver; This is only valid for output and
+ * allows the output pin to run as an open collector output.
+ */
+int __init_or_module at91_set_multi_drive(unsigned pin, int is_on)
+{
+	void __iomem	*pio = pin_to_controller(pin);
+	unsigned	mask = pin_to_mask(pin);
+
+	if (!pio)
+		return -EINVAL;
+
+	__raw_writel(mask, pio + (is_on ? PIO_MDER : PIO_MDDR));
+	return 0;
+}
+EXPORT_SYMBOL(at91_set_multi_drive);
+
 /*--------------------------------------------------------------------------*/
 
 
diff --git a/include/asm-arm/arch-at91rm9200/gpio.h b/include/asm-arm/arch-at91rm9200/gpio.h
index 0f0a61e2f129..6176ab2dc417 100644
--- a/include/asm-arm/arch-at91rm9200/gpio.h
+++ b/include/asm-arm/arch-at91rm9200/gpio.h
@@ -183,6 +183,7 @@ extern int at91_set_B_periph(unsigned pin, int use_pullup);
 extern int at91_set_gpio_input(unsigned pin, int use_pullup);
 extern int at91_set_gpio_output(unsigned pin, int value);
 extern int at91_set_deglitch(unsigned pin, int is_on);
+extern int at91_set_multi_drive(unsigned pin, int is_on);
 
 /* callable at any time */
 extern int at91_set_gpio_value(unsigned pin, int value);

From 06e4479bd092eca4125e5507e7c22619a491dab3 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?M=E5rten=20Wikstr=F6m?= <marten.wikstrom@passito.se>
Date: Wed, 22 Feb 2006 22:27:23 +0000
Subject: [PATCH 382/509] [ARM] 3347/1: Bugfix for ixp4xx_set_irq_type()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Patch from Mårten Wikström

This patch fixes a bug in ixp4xx_set_irq_type() which leads to
GPIO being incorrectly set to both edge triggered for raising
as well as falling edge interrupt types. See the previous
discussion on patch 3312/1.

Signed-off-by: Mårten Wikström <marten.wikstrom@passito.se>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-ixp4xx/common.c | 20 +++++++++++++-------
 1 file changed, 13 insertions(+), 7 deletions(-)

diff --git a/arch/arm/mach-ixp4xx/common.c b/arch/arm/mach-ixp4xx/common.c
index 4bdc9d4526cd..fbadf3021b9e 100644
--- a/arch/arm/mach-ixp4xx/common.c
+++ b/arch/arm/mach-ixp4xx/common.c
@@ -111,24 +111,30 @@ static int ixp4xx_set_irq_type(unsigned int irq, unsigned int type)
 	if (line < 0)
 		return -EINVAL;
 
-	if (type & IRQT_BOTHEDGE) {
+	switch (type){
+	case IRQT_BOTHEDGE:
 		int_style = IXP4XX_GPIO_STYLE_TRANSITIONAL;
 		irq_type = IXP4XX_IRQ_EDGE;
-	} else  if (type & IRQT_RISING) {
+		break;
+	case IRQT_RISING:
 		int_style = IXP4XX_GPIO_STYLE_RISING_EDGE;
 		irq_type = IXP4XX_IRQ_EDGE;
-	} else if (type & IRQT_FALLING) {
+		break;
+	case IRQT_FALLING:
 		int_style = IXP4XX_GPIO_STYLE_FALLING_EDGE;
 		irq_type = IXP4XX_IRQ_EDGE;
-	} else if (type & IRQT_HIGH) {
+		break;
+	case IRQT_HIGH:
 		int_style = IXP4XX_GPIO_STYLE_ACTIVE_HIGH;
 		irq_type = IXP4XX_IRQ_LEVEL;
-	} else if (type & IRQT_LOW) {
+		break;
+	case IRQT_LOW:
 		int_style = IXP4XX_GPIO_STYLE_ACTIVE_LOW;
 		irq_type = IXP4XX_IRQ_LEVEL;
-	} else
+		break;
+	default:
 		return -EINVAL;
-
+	}
 	ixp4xx_config_irq(irq, irq_type);
 
 	if (line >= 8) {	/* pins 8-15 */

From a6ceda7457b2303dcb07d3c472b25d52bbdb5a29 Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Wed, 22 Feb 2006 14:35:52 -0800
Subject: [PATCH 383/509] [SCSI] esp: fix eh locking

esp_reset didn't get fixed when the EH locking changed.
->eh_bus_reset_handler is now called without the host lock held.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/scsi/esp.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/scsi/esp.c b/drivers/scsi/esp.c
index f6900538be90..87a8c3d2072c 100644
--- a/drivers/scsi/esp.c
+++ b/drivers/scsi/esp.c
@@ -2068,14 +2068,12 @@ static int esp_reset(struct scsi_cmnd *SCptr)
 {
 	struct esp *esp = (struct esp *) SCptr->device->host->hostdata;
 
+	spin_lock_irq(esp->ehost->host_lock);
 	(void) esp_do_resetbus(esp);
-
 	spin_unlock_irq(esp->ehost->host_lock);
 
 	wait_event(esp->reset_queue, (esp->resetting_bus == 0));
 
-	spin_lock_irq(esp->ehost->host_lock);
-
 	return SUCCESS;
 }
 

From 6cec2aed8686840906f6298391dc4fd04d9ba843 Mon Sep 17 00:00:00 2001
From: Steve French <smfltc@us.ibm.com>
Date: Wed, 22 Feb 2006 17:31:52 -0600
Subject: [PATCH 384/509] [PATCH] CIFS: CIFSSMBRead was returning an invalid
 pointer in buf on socket error

Thanks to Adrian Bunk for debugging the problem and to Shaggy for
helping find the solution.

Also added a fix for 64K pages we found in loosely-related testing

Signed-off-by: Dave Kleikamp <shaggy@austin.ibm.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/cifs/cifssmb.c | 7 ++++---
 fs/cifs/connect.c | 8 ++++----
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index 217323b0c896..b41e8b379652 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -1048,13 +1048,14 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon,
 			cifs_small_buf_release(iov[0].iov_base);
 		else if(resp_buf_type == CIFS_LARGE_BUFFER)
 			cifs_buf_release(iov[0].iov_base);
-	} else /* return buffer to caller to free */ /* BB FIXME how do we tell caller if it is not a large buffer */ {
-		*buf = iov[0].iov_base;
+	} else if(resp_buf_type != CIFS_NO_BUFFER) {
+		/* return buffer to caller to free */ 
+		*buf = iov[0].iov_base;		
 		if(resp_buf_type == CIFS_SMALL_BUFFER)
 			*pbuf_type = CIFS_SMALL_BUFFER;
 		else if(resp_buf_type == CIFS_LARGE_BUFFER)
 			*pbuf_type = CIFS_LARGE_BUFFER;
-	}
+	} /* else no valid buffer on return - leave as null */
 
 	/* Note: On -EAGAIN error only caller can retry on handle based calls
 		since file handle passed in no longer valid */
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index e488603fb1e7..ef5ae6f93c75 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -1795,10 +1795,10 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb,
 			   conjunction with 52K kvec constraint on arch with 4K
 			   page size  */
 
-		if(cifs_sb->rsize < PAGE_CACHE_SIZE) {
-			cifs_sb->rsize = PAGE_CACHE_SIZE; 
-			/* Windows ME does this */
-			cFYI(1,("Attempt to set readsize for mount to less than one page (4096)"));
+		if(cifs_sb->rsize < 2048) {
+			cifs_sb->rsize = 2048; 
+			/* Windows ME may prefer this */
+			cFYI(1,("readsize set to minimum 2048"));
 		}
 		cifs_sb->mnt_uid = volume_info.linux_uid;
 		cifs_sb->mnt_gid = volume_info.linux_gid;

From 977bdf06ca8dd7ed081fab8d30249d9e6b1c24d3 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 11:44:58 -0800
Subject: [PATCH 385/509] [PATCH] sky2: yukon-ec-u chipset initialization

Add more complete setup code for Yukon EC_U chipset.
Based on matching code in 8.31 code in SysKonnect vendor driver.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/sky2.c | 52 ++++++++++++++++++++++++++++++++--------
 drivers/net/sky2.h | 59 ++++++++++++++++++++++++++++++++++++++--------
 2 files changed, 91 insertions(+), 20 deletions(-)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index bfeba5b9cd7a..ce135b84a54c 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -232,7 +232,17 @@ static int sky2_set_power_state(struct sky2_hw *hw, pci_power_t state)
 			if (hw->ports > 1)
 				reg1 |= PCI_Y2_PHY2_COMA;
 		}
+
+		if (hw->chip_id == CHIP_ID_YUKON_EC_U) {
+			pci_write_config_dword(hw->pdev, PCI_DEV_REG3, 0);
+			pci_read_config_dword(hw->pdev, PCI_DEV_REG4, &reg1);
+			reg1 &= P_ASPM_CONTROL_MSK;
+			pci_write_config_dword(hw->pdev, PCI_DEV_REG4, reg1);
+			pci_write_config_dword(hw->pdev, PCI_DEV_REG5, 0);
+		}
+
 		pci_write_config_dword(hw->pdev, PCI_DEV_REG1, reg1);
+
 		break;
 
 	case PCI_D3hot:
@@ -463,16 +473,31 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port)
 		ledover |= PHY_M_LED_MO_RX(MO_LED_OFF);
 	}
 
-	gm_phy_write(hw, port, PHY_MARV_LED_CTRL, ledctrl);
+	if (hw->chip_id == CHIP_ID_YUKON_EC_U && hw->chip_rev >= 2) {
+		/* apply fixes in PHY AFE */
+		gm_phy_write(hw, port, 22, 255);
+		/* increase differential signal amplitude in 10BASE-T */
+		gm_phy_write(hw, port, 24, 0xaa99);
+		gm_phy_write(hw, port, 23, 0x2011);
+
+		/* fix for IEEE A/B Symmetry failure in 1000BASE-T */
+		gm_phy_write(hw, port, 24, 0xa204);
+		gm_phy_write(hw, port, 23, 0x2002);
+
+		/* set page register to 0 */
+		gm_phy_write(hw, port, 22, 0);
+	} else {
+		gm_phy_write(hw, port, PHY_MARV_LED_CTRL, ledctrl);
+
+		if (sky2->autoneg == AUTONEG_DISABLE || sky2->speed == SPEED_100) {
+			/* turn on 100 Mbps LED (LED_LINK100) */
+			ledover |= PHY_M_LED_MO_100(MO_LED_ON);
+		}
+
+		if (ledover)
+			gm_phy_write(hw, port, PHY_MARV_LED_OVER, ledover);
 
-	if (sky2->autoneg == AUTONEG_DISABLE || sky2->speed == SPEED_100) {
-		/* turn on 100 Mbps LED (LED_LINK100) */
-		ledover |= PHY_M_LED_MO_100(MO_LED_ON);
 	}
-
-	if (ledover)
-		gm_phy_write(hw, port, PHY_MARV_LED_OVER, ledover);
-
 	/* Enable phy interrupt on auto-negotiation complete (or link up) */
 	if (sky2->autoneg == AUTONEG_ENABLE)
 		gm_phy_write(hw, port, PHY_MARV_INT_MASK, PHY_M_IS_AN_COMPL);
@@ -953,6 +978,12 @@ static int sky2_rx_start(struct sky2_port *sky2)
 
 	sky2->rx_put = sky2->rx_next = 0;
 	sky2_qset(hw, rxq);
+
+	if (hw->chip_id == CHIP_ID_YUKON_EC_U && hw->chip_rev >= 2) {
+		/* MAC Rx RAM Read is controlled by hardware */
+		sky2_write32(hw, Q_ADDR(rxq, Q_F), F_M_RX_RAM_DIS);
+	}
+
 	sky2_prefetch_init(hw, rxq, sky2->rx_le_map, RX_LE_SIZE - 1);
 
 	rx_set_checksum(sky2);
@@ -1035,9 +1066,10 @@ static int sky2_up(struct net_device *dev)
 		    RB_RST_SET);
 
 	sky2_qset(hw, txqaddr[port]);
-	if (hw->chip_id == CHIP_ID_YUKON_EC_U)
-		sky2_write16(hw, Q_ADDR(txqaddr[port], Q_AL), 0x1a0);
 
+	/* Set almost empty threshold */
+	if (hw->chip_id == CHIP_ID_YUKON_EC_U && hw->chip_rev == 1)
+		sky2_write16(hw, Q_ADDR(txqaddr[port], Q_AL), 0x1a0);
 
 	sky2_prefetch_init(hw, txqaddr[port], sky2->tx_le_map,
 			   TX_RING_SIZE - 1);
diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h
index fd12c289a238..d1c71f50d8b9 100644
--- a/drivers/net/sky2.h
+++ b/drivers/net/sky2.h
@@ -5,14 +5,22 @@
 #define _SKY2_H
 
 /* PCI config registers */
-#define PCI_DEV_REG1	0x40
-#define PCI_DEV_REG2	0x44
-#define PCI_DEV_STATUS  0x7c
-#define PCI_OS_PCI_X    (1<<26)
+enum {
+	PCI_DEV_REG1	= 0x40,
+	PCI_DEV_REG2	= 0x44,
+	PCI_DEV_STATUS  = 0x7c,
+	PCI_DEV_REG3	= 0x80,
+	PCI_DEV_REG4	= 0x84,
+	PCI_DEV_REG5    = 0x88,
+};
 
-#define PEX_LNK_STAT	0xf2
-#define PEX_UNC_ERR_STAT 0x104
-#define PEX_DEV_CTRL	0xe8
+enum {
+	PEX_DEV_CAP	= 0xe4,
+	PEX_DEV_CTRL	= 0xe8,
+	PEX_DEV_STA	= 0xea,
+	PEX_LNK_STAT	= 0xf2,
+	PEX_UNC_ERR_STAT= 0x104,
+};
 
 /* Yukon-2 */
 enum pci_dev_reg_1 {
@@ -37,6 +45,25 @@ enum pci_dev_reg_2 {
 	PCI_USEDATA64	= 1<<0,		/* Use 64Bit Data bus ext */
 };
 
+/*	PCI_OUR_REG_4		32 bit	Our Register 4 (Yukon-ECU only) */
+enum pci_dev_reg_4 {
+					/* (Link Training & Status State Machine) */
+	P_TIMER_VALUE_MSK	= 0xffL<<16,	/* Bit 23..16:	Timer Value Mask */
+					/* (Active State Power Management) */
+	P_FORCE_ASPM_REQUEST	= 1<<15, /* Force ASPM Request (A1 only) */
+	P_ASPM_GPHY_LINK_DOWN	= 1<<14, /* GPHY Link Down (A1 only) */
+	P_ASPM_INT_FIFO_EMPTY	= 1<<13, /* Internal FIFO Empty (A1 only) */
+	P_ASPM_CLKRUN_REQUEST	= 1<<12, /* CLKRUN Request (A1 only) */
+
+	P_ASPM_FORCE_CLKREQ_ENA	= 1<<4,	/* Force CLKREQ Enable (A1b only) */
+	P_ASPM_CLKREQ_PAD_CTL	= 1<<3,	/* CLKREQ PAD Control (A1 only) */
+	P_ASPM_A1_MODE_SELECT	= 1<<2,	/* A1 Mode Select (A1 only) */
+	P_CLK_GATE_PEX_UNIT_ENA	= 1<<1,	/* Enable Gate PEX Unit Clock */
+	P_CLK_GATE_ROOT_COR_ENA	= 1<<0,	/* Enable Gate Root Core Clock */
+	P_ASPM_CONTROL_MSK	= P_FORCE_ASPM_REQUEST | P_ASPM_GPHY_LINK_DOWN
+				  | P_ASPM_CLKRUN_REQUEST | P_ASPM_INT_FIFO_EMPTY,
+};
+
 
 #define PCI_STATUS_ERROR_BITS (PCI_STATUS_DETECTED_PARITY | \
 			       PCI_STATUS_SIG_SYSTEM_ERROR | \
@@ -507,6 +534,16 @@ enum {
 };
 #define Q_ADDR(reg, offs) (B8_Q_REGS + (reg) + (offs))
 
+/*	Q_F				32 bit	Flag Register */
+enum {
+	F_ALM_FULL	= 1<<27, /* Rx FIFO: almost full */
+	F_EMPTY		= 1<<27, /* Tx FIFO: empty flag */
+	F_FIFO_EOF	= 1<<26, /* Tag (EOF Flag) bit in FIFO */
+	F_WM_REACHED	= 1<<25, /* Watermark reached */
+	F_M_RX_RAM_DIS	= 1<<24, /* MAC Rx RAM Read Port disable */
+	F_FIFO_LEVEL	= 0x1fL<<16, /* Bit 23..16:	# of Qwords in FIFO */
+	F_WATER_MARK	= 0x0007ffL, /* Bit 10.. 0:	Watermark */
+};
 
 /* Queue Prefetch Unit Offsets, use Y2_QADDR() to address (Yukon-2 only)*/
 enum {
@@ -909,10 +946,12 @@ enum {
 	PHY_BCOM_ID1_C0	= 0x6044,
 	PHY_BCOM_ID1_C5	= 0x6047,
 
-	PHY_MARV_ID1_B0	= 0x0C23, /* Yukon (PHY 88E1011) */
+	PHY_MARV_ID1_B0	= 0x0C23, /* Yukon 	(PHY 88E1011) */
 	PHY_MARV_ID1_B2	= 0x0C25, /* Yukon-Plus (PHY 88E1011) */
-	PHY_MARV_ID1_C2	= 0x0CC2, /* Yukon-EC (PHY 88E1111) */
-	PHY_MARV_ID1_Y2	= 0x0C91, /* Yukon-2 (PHY 88E1112) */
+	PHY_MARV_ID1_C2	= 0x0CC2, /* Yukon-EC	(PHY 88E1111) */
+	PHY_MARV_ID1_Y2	= 0x0C91, /* Yukon-2	(PHY 88E1112) */
+	PHY_MARV_ID1_FE = 0x0C83, /* Yukon-FE   (PHY 88E3082 Rev.A1) */
+	PHY_MARV_ID1_ECU= 0x0CB0, /* Yukon-ECU  (PHY 88E1149 Rev.B2?) */
 };
 
 /* Advertisement register bits */

From ff81fbbe321c3a468b6225c673ca57efa501fbed Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 11:44:59 -0800
Subject: [PATCH 386/509] [PATCH] sky2: limit coalescing values to ring size

Don't allow coalescing values to be bigger than the transmit ring.
Since if you set them that big, the interrupt never happens and driver
livelocks.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/sky2.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index ce135b84a54c..cf7fd28d6776 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -2884,11 +2884,11 @@ static int sky2_set_coalesce(struct net_device *dev,
 	    (ecmd->rx_coalesce_usecs_irq < tmin || ecmd->rx_coalesce_usecs_irq > tmax))
 		return -EINVAL;
 
-	if (ecmd->tx_max_coalesced_frames > 0xffff)
+	if (ecmd->tx_max_coalesced_frames >= TX_RING_SIZE-1)
 		return -EINVAL;
-	if (ecmd->rx_max_coalesced_frames > 0xff)
+	if (ecmd->rx_max_coalesced_frames > RX_MAX_PENDING)
 		return -EINVAL;
-	if (ecmd->rx_max_coalesced_frames_irq > 0xff)
+	if (ecmd->rx_max_coalesced_frames_irq >RX_MAX_PENDING)
 		return -EINVAL;
 
 	if (ecmd->tx_coalesce_usecs == 0)

From a8fd6266dafd564bae6758cb78c8c152e7d4115e Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 11:45:00 -0800
Subject: [PATCH 387/509] [PATCH] sky2: poke coalescing timer to fix hang

Need to restart the interrupt coalescing timer after clearing the interrupt,
to avoid races with interrupt timer and processing.

Patch from Carl-Daniel Halfinger

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/sky2.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index cf7fd28d6776..629809433cb3 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -1895,6 +1895,17 @@ static int sky2_poll(struct net_device *dev0, int *budget)
 
 	sky2_write32(hw, STAT_CTRL, SC_STAT_CLR_IRQ);
 
+	/*
+	 * Kick the STAT_LEV_TIMER_CTRL timer.
+	 * This fixes my hangs on Yukon-EC (0xb6) rev 1.
+	 * The if clause is there to start the timer only if it has been
+	 * configured correctly and not been disabled via ethtool.
+	 */
+	if (sky2_read8(hw, STAT_LEV_TIMER_CTRL) == TIM_START) {
+		sky2_write8(hw, STAT_LEV_TIMER_CTRL, TIM_STOP);
+		sky2_write8(hw, STAT_LEV_TIMER_CTRL, TIM_START);
+	}
+
 	hwidx = sky2_read16(hw, STAT_PUT_IDX);
 	BUG_ON(hwidx >= STATUS_RING_SIZE);
 	rmb();

From 9a6d343188f5f1e9537e700fc4139c2d905ff129 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 11:45:01 -0800
Subject: [PATCH 388/509] [PATCH] sky2: force early transmit status

Need to force a transmit coalesce timer restart after processing
transmit packets. Otherwise, can get transmit status after last
update and chip doesn't send the next one.

Can go with the chip defaults for coalescing timers, except for
Tx timer which needs to be bigger.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/sky2.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 629809433cb3..d913d3407e90 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -1988,13 +1988,12 @@ exit_loop:
 	sky2_tx_check(hw, 0, tx_done[0]);
 	sky2_tx_check(hw, 1, tx_done[1]);
 
-	if (likely(work_done < to_do)) {
-		/* need to restart TX timer */
-		if (is_ec_a1(hw)) {
-			sky2_write8(hw, STAT_TX_TIMER_CTRL, TIM_STOP);
-			sky2_write8(hw, STAT_TX_TIMER_CTRL, TIM_START);
-		}
+	if (sky2_read8(hw, STAT_TX_TIMER_CTRL) == TIM_START) {
+		sky2_write8(hw, STAT_TX_TIMER_CTRL, TIM_STOP);
+		sky2_write8(hw, STAT_TX_TIMER_CTRL, TIM_START);
+	}
 
+	if (likely(work_done < to_do)) {
 		netif_rx_complete(dev0);
 		hw->intr_mask |= Y2_IS_STAT_BMU;
 		sky2_write32(hw, B0_IMSK, hw->intr_mask);
@@ -2352,8 +2351,7 @@ static int sky2_reset(struct sky2_hw *hw)
 			sky2_write8(hw, STAT_FIFO_ISR_WM, 16);
 
 		sky2_write32(hw, STAT_TX_TIMER_INI, sky2_us2clk(hw, 1000));
-		sky2_write32(hw, STAT_LEV_TIMER_INI, sky2_us2clk(hw, 100));
-		sky2_write32(hw, STAT_ISR_TIMER_INI, sky2_us2clk(hw, 20));
+		sky2_write32(hw, STAT_ISR_TIMER_INI, sky2_us2clk(hw, 7));
 	}
 
 	/* enable status unit */

From 56a645cc1bc16ab33b33a3e0854a46c5d2c864f3 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 11:45:02 -0800
Subject: [PATCH 389/509] [PATCH] sky2: use device iomem to access PCI config

To avoid problems with PCI config access without ACPI (or busted ACPI tables),
use the device's window into PCI config space.

I know this probably will upset the purists, but I would rather have users
than ACPI testers. It also generates less code.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/sky2.c | 81 +++++++++++++++++++---------------------------
 drivers/net/sky2.h | 21 ++++++++++++
 2 files changed, 54 insertions(+), 48 deletions(-)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index d913d3407e90..fbbc85532c2e 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -195,11 +195,11 @@ static int sky2_set_power_state(struct sky2_hw *hw, pci_power_t state)
 	pr_debug("sky2_set_power_state %d\n", state);
 	sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
 
-	pci_read_config_word(hw->pdev, hw->pm_cap + PCI_PM_PMC, &power_control);
+	power_control = sky2_pci_read16(hw, hw->pm_cap + PCI_PM_PMC);
 	vaux = (sky2_read16(hw, B0_CTST) & Y2_VAUX_AVAIL) &&
 		(power_control & PCI_PM_CAP_PME_D3cold);
 
-	pci_read_config_word(hw->pdev, hw->pm_cap + PCI_PM_CTRL, &power_control);
+	power_control = sky2_pci_read16(hw, hw->pm_cap + PCI_PM_CTRL);
 
 	power_control |= PCI_PM_CTRL_PME_STATUS;
 	power_control &= ~(PCI_PM_CTRL_STATE_MASK);
@@ -223,7 +223,7 @@ static int sky2_set_power_state(struct sky2_hw *hw, pci_power_t state)
 			sky2_write8(hw, B2_Y2_CLK_GATE, 0);
 
 		/* Turn off phy power saving */
-		pci_read_config_dword(hw->pdev, PCI_DEV_REG1, &reg1);
+		reg1 = sky2_pci_read32(hw, PCI_DEV_REG1);
 		reg1 &= ~(PCI_Y2_PHY1_POWD | PCI_Y2_PHY2_POWD);
 
 		/* looks like this XL is back asswards .. */
@@ -234,26 +234,26 @@ static int sky2_set_power_state(struct sky2_hw *hw, pci_power_t state)
 		}
 
 		if (hw->chip_id == CHIP_ID_YUKON_EC_U) {
-			pci_write_config_dword(hw->pdev, PCI_DEV_REG3, 0);
-			pci_read_config_dword(hw->pdev, PCI_DEV_REG4, &reg1);
+			sky2_pci_write32(hw, PCI_DEV_REG3, 0);
+			reg1 = sky2_pci_read32(hw, PCI_DEV_REG4);
 			reg1 &= P_ASPM_CONTROL_MSK;
-			pci_write_config_dword(hw->pdev, PCI_DEV_REG4, reg1);
-			pci_write_config_dword(hw->pdev, PCI_DEV_REG5, 0);
+			sky2_pci_write32(hw, PCI_DEV_REG4, reg1);
+			sky2_pci_write32(hw, PCI_DEV_REG5, 0);
 		}
 
-		pci_write_config_dword(hw->pdev, PCI_DEV_REG1, reg1);
+		sky2_pci_write32(hw, PCI_DEV_REG1, reg1);
 
 		break;
 
 	case PCI_D3hot:
 	case PCI_D3cold:
 		/* Turn on phy power saving */
-		pci_read_config_dword(hw->pdev, PCI_DEV_REG1, &reg1);
+		reg1 = sky2_pci_read32(hw, PCI_DEV_REG1);
 		if (hw->chip_id == CHIP_ID_YUKON_XL && hw->chip_rev > 1)
 			reg1 &= ~(PCI_Y2_PHY1_POWD | PCI_Y2_PHY2_POWD);
 		else
 			reg1 |= (PCI_Y2_PHY1_POWD | PCI_Y2_PHY2_POWD);
-		pci_write_config_dword(hw->pdev, PCI_DEV_REG1, reg1);
+		sky2_pci_write32(hw, PCI_DEV_REG1, reg1);
 
 		if (hw->chip_id == CHIP_ID_YUKON_XL && hw->chip_rev > 1)
 			sky2_write8(hw, B2_Y2_CLK_GATE, 0);
@@ -275,7 +275,7 @@ static int sky2_set_power_state(struct sky2_hw *hw, pci_power_t state)
 		ret = -1;
 	}
 
-	pci_write_config_byte(hw->pdev, hw->pm_cap + PCI_PM_CTRL, power_control);
+	sky2_pci_write16(hw, hw->pm_cap + PCI_PM_CTRL, power_control);
 	sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_OFF);
 	return ret;
 }
@@ -2059,13 +2059,13 @@ static void sky2_hw_intr(struct sky2_hw *hw)
 	if (status & (Y2_IS_MST_ERR | Y2_IS_IRQ_STAT)) {
 		u16 pci_err;
 
-		pci_read_config_word(hw->pdev, PCI_STATUS, &pci_err);
+		pci_err = sky2_pci_read16(hw, PCI_STATUS);
 		if (net_ratelimit())
 			printk(KERN_ERR PFX "%s: pci hw error (0x%x)\n",
 			       pci_name(hw->pdev), pci_err);
 
 		sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
-		pci_write_config_word(hw->pdev, PCI_STATUS,
+		sky2_pci_write16(hw, PCI_STATUS,
 				      pci_err | PCI_STATUS_ERROR_BITS);
 		sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_OFF);
 	}
@@ -2074,7 +2074,7 @@ static void sky2_hw_intr(struct sky2_hw *hw)
 		/* PCI-Express uncorrectable Error occurred */
 		u32 pex_err;
 
-		pci_read_config_dword(hw->pdev, PEX_UNC_ERR_STAT, &pex_err);
+		pex_err = sky2_pci_read32(hw, PEX_UNC_ERR_STAT);
 
 		if (net_ratelimit())
 			printk(KERN_ERR PFX "%s: pci express error (0x%x)\n",
@@ -2082,7 +2082,7 @@ static void sky2_hw_intr(struct sky2_hw *hw)
 
 		/* clear the interrupt */
 		sky2_write32(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
-		pci_write_config_dword(hw->pdev, PEX_UNC_ERR_STAT,
+		sky2_pci_write32(hw, PEX_UNC_ERR_STAT,
 				       0xffffffffUL);
 		sky2_write32(hw, B2_TST_CTRL1, TST_CFG_WRITE_OFF);
 
@@ -2212,7 +2212,7 @@ static int sky2_reset(struct sky2_hw *hw)
 {
 	u16 status;
 	u8 t8, pmd_type;
-	int i, err;
+	int i;
 
 	sky2_write8(hw, B0_CTST, CS_RST_CLR);
 
@@ -2234,25 +2234,18 @@ static int sky2_reset(struct sky2_hw *hw)
 	sky2_write8(hw, B0_CTST, CS_RST_CLR);
 
 	/* clear PCI errors, if any */
-	err = pci_read_config_word(hw->pdev, PCI_STATUS, &status);
-	if (err)
-		goto pci_err;
+	status = sky2_pci_read16(hw, PCI_STATUS);
 
 	sky2_write8(hw, B2_TST_CTRL1, TST_CFG_WRITE_ON);
-	err = pci_write_config_word(hw->pdev, PCI_STATUS,
-				    status | PCI_STATUS_ERROR_BITS);
-	if (err)
-		goto pci_err;
+	sky2_pci_write16(hw, PCI_STATUS, status | PCI_STATUS_ERROR_BITS);
+
 
 	sky2_write8(hw, B0_CTST, CS_MRST_CLR);
 
 	/* clear any PEX errors */
-	if (pci_find_capability(hw->pdev, PCI_CAP_ID_EXP)) {
-		err = pci_write_config_dword(hw->pdev, PEX_UNC_ERR_STAT,
-						 0xffffffffUL);
-		if (err)
-			goto pci_err;
-	}
+	if (pci_find_capability(hw->pdev, PCI_CAP_ID_EXP)) 
+		sky2_pci_write32(hw, PEX_UNC_ERR_STAT, 0xffffffffUL);
+
 
 	pmd_type = sky2_read8(hw, B2_PMD_TYP);
 	hw->copper = !(pmd_type == 'L' || pmd_type == 'S');
@@ -2362,14 +2355,6 @@ static int sky2_reset(struct sky2_hw *hw)
 	sky2_write8(hw, STAT_ISR_TIMER_CTRL, TIM_START);
 
 	return 0;
-
-pci_err:
-	/* This is to catch a BIOS bug workaround where
-	 * mmconfig table doesn't have other buses.
-	 */
-	printk(KERN_ERR PFX "%s: can't access PCI config space\n",
-	       pci_name(hw->pdev));
-	return err;
 }
 
 static u32 sky2_supported_modes(const struct sky2_hw *hw)
@@ -3239,17 +3224,6 @@ static int __devinit sky2_probe(struct pci_dev *pdev,
 		}
 	}
 
-#ifdef __BIG_ENDIAN
-	/* byte swap descriptors in hardware */
-	{
-		u32 reg;
-
-		pci_read_config_dword(pdev, PCI_DEV_REG2, &reg);
-		reg |= PCI_REV_DESC;
-		pci_write_config_dword(pdev, PCI_DEV_REG2, reg);
-	}
-#endif
-
 	err = -ENOMEM;
 	hw = kzalloc(sizeof(*hw), GFP_KERNEL);
 	if (!hw) {
@@ -3268,6 +3242,17 @@ static int __devinit sky2_probe(struct pci_dev *pdev,
 	}
 	hw->pm_cap = pm_cap;
 
+#ifdef __BIG_ENDIAN
+	/* byte swap descriptors in hardware */
+	{
+		u32 reg;
+
+		reg = sky2_pci_read32(hw, PCI_DEV_REG2);
+		reg |= PCI_REV_DESC;
+		sky2_pci_write32(hw, PCI_DEV_REG2, reg);
+	}
+#endif
+
 	/* ring for status responses */
 	hw->st_le = pci_alloc_consistent(hw->pdev, STATUS_LE_BYTES,
 					 &hw->st_dma);
diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h
index d1c71f50d8b9..9e40766150b4 100644
--- a/drivers/net/sky2.h
+++ b/drivers/net/sky2.h
@@ -1951,4 +1951,25 @@ static inline void gma_set_addr(struct sky2_hw *hw, unsigned port, unsigned reg,
 	gma_write16(hw, port, reg+4,(u16) addr[2] | ((u16) addr[3] << 8));
 	gma_write16(hw, port, reg+8,(u16) addr[4] | ((u16) addr[5] << 8));
 }
+
+/* PCI config space access */
+static inline u32 sky2_pci_read32(const struct sky2_hw *hw, unsigned reg)
+{
+	return sky2_read32(hw, Y2_CFG_SPC + reg);
+}
+
+static inline u16 sky2_pci_read16(const struct sky2_hw *hw, unsigned reg)
+{
+	return sky2_read16(hw, Y2_CFG_SPC + reg);
+}
+
+static inline void sky2_pci_write32(struct sky2_hw *hw, unsigned reg, u32 val)
+{
+	sky2_write32(hw, Y2_CFG_SPC + reg, val);
+}
+
+static inline void sky2_pci_write16(struct sky2_hw *hw, unsigned reg, u16 val)
+{
+	sky2_write16(hw, Y2_CFG_SPC + reg, val);
+}
 #endif

From 791917deb63c6d8beb3f347ea0911371deff1624 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 11:45:03 -0800
Subject: [PATCH 390/509] [PATCH] sky2: close race on IRQ mask update.

Need to avoid race in updating IRQ mask.  This can probably be replaced
smarter use of the interrupt control registers (if/when chipset
docs are available).

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/sky2.c | 21 +++++++++++++++------
 drivers/net/sky2.h |  3 ++-
 2 files changed, 17 insertions(+), 7 deletions(-)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index fbbc85532c2e..ca8160d68229 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -1079,8 +1079,10 @@ static int sky2_up(struct net_device *dev)
 		goto err_out;
 
 	/* Enable interrupts from phy/mac for port */
+	spin_lock_irq(&hw->hw_lock);
 	hw->intr_mask |= (port == 0) ? Y2_IS_PORT_1 : Y2_IS_PORT_2;
 	sky2_write32(hw, B0_IMSK, hw->intr_mask);
+	spin_unlock_irq(&hw->hw_lock);
 	return 0;
 
 err_out:
@@ -1380,10 +1382,10 @@ static int sky2_down(struct net_device *dev)
 	netif_stop_queue(dev);
 
 	/* Disable port IRQ */
-	local_irq_disable();
+	spin_lock_irq(&hw->hw_lock);
 	hw->intr_mask &= ~((sky2->port == 0) ? Y2_IS_IRQ_PHY1 : Y2_IS_IRQ_PHY2);
 	sky2_write32(hw, B0_IMSK, hw->intr_mask);
-	local_irq_enable();
+	spin_unlock_irq(&hw->hw_lock);
 
 	flush_scheduled_work();
 
@@ -1665,10 +1667,10 @@ static void sky2_phy_task(void *arg)
 out:
 	up(&sky2->phy_sema);
 
-	local_irq_disable();
+	spin_lock_irq(&hw->hw_lock);
 	hw->intr_mask |= (sky2->port == 0) ? Y2_IS_IRQ_PHY1 : Y2_IS_IRQ_PHY2;
 	sky2_write32(hw, B0_IMSK, hw->intr_mask);
-	local_irq_enable();
+	spin_unlock_irq(&hw->hw_lock);
 }
 
 
@@ -1994,9 +1996,13 @@ exit_loop:
 	}
 
 	if (likely(work_done < to_do)) {
-		netif_rx_complete(dev0);
+		spin_lock_irq(&hw->hw_lock);
+		__netif_rx_complete(dev0);
+
 		hw->intr_mask |= Y2_IS_STAT_BMU;
 		sky2_write32(hw, B0_IMSK, hw->intr_mask);
+		spin_unlock_irq(&hw->hw_lock);
+
 		return 0;
 	} else {
 		*budget -= work_done;
@@ -2128,6 +2134,7 @@ static void sky2_phy_intr(struct sky2_hw *hw, unsigned port)
 
 	hw->intr_mask &= ~(port == 0 ? Y2_IS_IRQ_PHY1 : Y2_IS_IRQ_PHY2);
 	sky2_write32(hw, B0_IMSK, hw->intr_mask);
+
 	schedule_work(&sky2->phy_task);
 }
 
@@ -2141,6 +2148,7 @@ static irqreturn_t sky2_intr(int irq, void *dev_id, struct pt_regs *regs)
 	if (status == 0 || status == ~0)
 		return IRQ_NONE;
 
+	spin_lock(&hw->hw_lock);
 	if (status & Y2_IS_HW_ERR)
 		sky2_hw_intr(hw);
 
@@ -2169,7 +2177,7 @@ static irqreturn_t sky2_intr(int irq, void *dev_id, struct pt_regs *regs)
 
 	sky2_write32(hw, B0_Y2_SP_ICR, 2);
 
-	sky2_read32(hw, B0_IMSK);
+	spin_unlock(&hw->hw_lock);
 
 	return IRQ_HANDLED;
 }
@@ -3241,6 +3249,7 @@ static int __devinit sky2_probe(struct pci_dev *pdev,
 		goto err_out_free_hw;
 	}
 	hw->pm_cap = pm_cap;
+	spin_lock_init(&hw->hw_lock);
 
 #ifdef __BIG_ENDIAN
 	/* byte swap descriptors in hardware */
diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h
index 9e40766150b4..3edb98075e0a 100644
--- a/drivers/net/sky2.h
+++ b/drivers/net/sky2.h
@@ -1876,8 +1876,9 @@ struct sky2_port {
 struct sky2_hw {
 	void __iomem  	     *regs;
 	struct pci_dev	     *pdev;
-	u32		     intr_mask;
 	struct net_device    *dev[2];
+	spinlock_t	     hw_lock;
+	u32		     intr_mask;
 
 	int		     pm_cap;
 	int		     msi;

From bf637ec3ef4159da3dd156ecf6f6987d8c8c5dae Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Tue, 31 Jan 2006 00:13:06 -0500
Subject: [PATCH 391/509] sbp2: fix another deadlock after disconnection

If there were commands enqueued but not completed before an SBP-2 unit
was unplugged (or an attempt to reconnect failed), knodemgrd or any
process which tried to remove the device would sleep uninterruptibly
in blk_execute_rq().  Therefore make sure that all commands are
completed when sbp2 retreats.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Jody McIntyre <scjody@modernduck.com>
(cherry picked from 61daa34c132c5d4ed8630e2c46e9bf2f0c7b3428 commit)
---
 drivers/ieee1394/sbp2.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c
index c2c776fbda01..8963dd484eb9 100644
--- a/drivers/ieee1394/sbp2.c
+++ b/drivers/ieee1394/sbp2.c
@@ -643,9 +643,15 @@ static int sbp2_remove(struct device *dev)
 	if (!scsi_id)
 		return 0;
 
-	/* Trigger shutdown functions in scsi's highlevel. */
-	if (scsi_id->scsi_host)
+	if (scsi_id->scsi_host) {
+		/* Get rid of enqueued commands if there is no chance to
+		 * send them. */
+		if (!sbp2util_node_is_available(scsi_id))
+			sbp2scsi_complete_all_commands(scsi_id, DID_NO_CONNECT);
+		/* scsi_remove_device() will trigger shutdown functions of SCSI
+		 * highlevel drivers which would deadlock if blocked. */
 		scsi_unblock_requests(scsi_id->scsi_host);
+	}
 	sdev = scsi_id->sdev;
 	if (sdev) {
 		scsi_id->sdev = NULL;

From 35bdddb83f62978b5fad82a14fbfd78cc3a5a60c Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Tue, 31 Jan 2006 00:13:33 -0500
Subject: [PATCH 392/509] sbp2: variable status FIFO address (fix login
 timeout)

Let the ieee1394 core select a suitable 1394 address range for sbp2's
status FIFO instead of using a fixed range. Since the core only selects
addresses which are guaranteed to be out of the "physical range" as per
OHCI 1.1, this patch also fixes an old bug:

OHCI controllers which implement a writeable PhysicalUpperBound register
included sbp2's status FIFO in the physical range. That way sbp2 was
never notified of a succesful login and always failed after timeout.
Affected OHCI host adapters include ALi and Fujitsu controllers.

As another side effect of this patch, the status FIFO is no longer
located in a range for which OHCI chips perform "posted writes". Each
status write now requires a response subaction. But since large data
transfers involve only few status writes, there is no measurable
decrease of I/O throughput. What's more, the status FIFO is now safe
from potential host bus errors. Nevertheless, posted writes could be
re-enabled by extensions to the ARM features of the 1394 stack.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Jody McIntyre <scjody@modernduck.com>
(cherry picked from b2d38cccad4ef80d6b672b8f89aae5fe2907b113 commit)
---
 drivers/ieee1394/sbp2.c | 64 +++++++++++++++++++++++------------------
 drivers/ieee1394/sbp2.h | 64 +++++++++++++++--------------------------
 2 files changed, 59 insertions(+), 69 deletions(-)

diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c
index 8963dd484eb9..0672224fa109 100644
--- a/drivers/ieee1394/sbp2.c
+++ b/drivers/ieee1394/sbp2.c
@@ -748,11 +748,6 @@ static struct scsi_id_instance_data *sbp2_alloc_device(struct unit_directory *ud
 		hi->host = ud->ne->host;
 		INIT_LIST_HEAD(&hi->scsi_ids);
 
-		/* Register our sbp2 status address space... */
-		hpsb_register_addrspace(&sbp2_highlevel, ud->ne->host, &sbp2_ops,
-					SBP2_STATUS_FIFO_ADDRESS,
-					SBP2_STATUS_FIFO_ADDRESS +
-					SBP2_STATUS_FIFO_ENTRY_TO_OFFSET(SBP2_MAX_UDS_PER_NODE+1));
 #ifdef CONFIG_IEEE1394_SBP2_PHYS_DMA
 		/* Handle data movement if physical dma is not
 		 * enabled/supportedon host controller */
@@ -765,6 +760,18 @@ static struct scsi_id_instance_data *sbp2_alloc_device(struct unit_directory *ud
 
 	list_add_tail(&scsi_id->scsi_list, &hi->scsi_ids);
 
+	/* Register the status FIFO address range. We could use the same FIFO
+	 * for targets at different nodes. However we need different FIFOs per
+	 * target in order to support multi-unit devices. */
+	scsi_id->status_fifo_addr = hpsb_allocate_and_register_addrspace(
+			&sbp2_highlevel, ud->ne->host, &sbp2_ops,
+			sizeof(struct sbp2_status_block), sizeof(quadlet_t),
+			~0ULL, ~0ULL);
+	if (!scsi_id->status_fifo_addr) {
+		SBP2_ERR("failed to allocate status FIFO address range");
+		goto failed_alloc;
+	}
+
 	/* Register our host with the SCSI stack. */
 	scsi_host = scsi_host_alloc(&scsi_driver_template,
 				    sizeof(unsigned long));
@@ -1003,6 +1010,10 @@ static void sbp2_remove_device(struct scsi_id_instance_data *scsi_id)
 		SBP2_DMA_FREE("single query logins data");
 	}
 
+	if (scsi_id->status_fifo_addr)
+		hpsb_unregister_addrspace(&sbp2_highlevel, hi->host,
+			scsi_id->status_fifo_addr);
+
 	scsi_id->ud->device.driver_data = NULL;
 
 	SBP2_DEBUG("SBP-2 device removed, SCSI ID = %d", scsi_id->ud->id);
@@ -1081,11 +1092,10 @@ static int sbp2_query_logins(struct scsi_id_instance_data *scsi_id)
 		ORB_SET_QUERY_LOGINS_RESP_LENGTH(sizeof(struct sbp2_query_logins_response));
 	SBP2_DEBUG("sbp2_query_logins: reserved_resp_length initialized");
 
-	scsi_id->query_logins_orb->status_FIFO_lo = SBP2_STATUS_FIFO_ADDRESS_LO +
-						    SBP2_STATUS_FIFO_ENTRY_TO_OFFSET(scsi_id->ud->id);
-	scsi_id->query_logins_orb->status_FIFO_hi = (ORB_SET_NODE_ID(hi->host->node_id) |
-						     SBP2_STATUS_FIFO_ADDRESS_HI);
-	SBP2_DEBUG("sbp2_query_logins: status FIFO initialized");
+	scsi_id->query_logins_orb->status_fifo_hi =
+		ORB_SET_STATUS_FIFO_HI(scsi_id->status_fifo_addr, hi->host->node_id);
+	scsi_id->query_logins_orb->status_fifo_lo =
+		ORB_SET_STATUS_FIFO_LO(scsi_id->status_fifo_addr);
 
 	sbp2util_cpu_to_be32_buffer(scsi_id->query_logins_orb, sizeof(struct sbp2_query_logins_orb));
 
@@ -1190,11 +1200,10 @@ static int sbp2_login_device(struct scsi_id_instance_data *scsi_id)
 		ORB_SET_LOGIN_RESP_LENGTH(sizeof(struct sbp2_login_response));
 	SBP2_DEBUG("sbp2_login_device: passwd_resp_lengths initialized");
 
-	scsi_id->login_orb->status_FIFO_lo = SBP2_STATUS_FIFO_ADDRESS_LO +
-					     SBP2_STATUS_FIFO_ENTRY_TO_OFFSET(scsi_id->ud->id);
-	scsi_id->login_orb->status_FIFO_hi = (ORB_SET_NODE_ID(hi->host->node_id) |
-					      SBP2_STATUS_FIFO_ADDRESS_HI);
-	SBP2_DEBUG("sbp2_login_device: status FIFO initialized");
+	scsi_id->login_orb->status_fifo_hi =
+		ORB_SET_STATUS_FIFO_HI(scsi_id->status_fifo_addr, hi->host->node_id);
+	scsi_id->login_orb->status_fifo_lo =
+		ORB_SET_STATUS_FIFO_LO(scsi_id->status_fifo_addr);
 
 	/*
 	 * Byte swap ORB if necessary
@@ -1307,10 +1316,10 @@ static int sbp2_logout_device(struct scsi_id_instance_data *scsi_id)
 	scsi_id->logout_orb->login_ID_misc |= ORB_SET_NOTIFY(1);
 
 	scsi_id->logout_orb->reserved5 = 0x0;
-	scsi_id->logout_orb->status_FIFO_lo = SBP2_STATUS_FIFO_ADDRESS_LO +
-					      SBP2_STATUS_FIFO_ENTRY_TO_OFFSET(scsi_id->ud->id);
-	scsi_id->logout_orb->status_FIFO_hi = (ORB_SET_NODE_ID(hi->host->node_id) |
-					       SBP2_STATUS_FIFO_ADDRESS_HI);
+	scsi_id->logout_orb->status_fifo_hi =
+		ORB_SET_STATUS_FIFO_HI(scsi_id->status_fifo_addr, hi->host->node_id);
+	scsi_id->logout_orb->status_fifo_lo =
+		ORB_SET_STATUS_FIFO_LO(scsi_id->status_fifo_addr);
 
 	/*
 	 * Byte swap ORB if necessary
@@ -1372,10 +1381,10 @@ static int sbp2_reconnect_device(struct scsi_id_instance_data *scsi_id)
 	scsi_id->reconnect_orb->login_ID_misc |= ORB_SET_NOTIFY(1);
 
 	scsi_id->reconnect_orb->reserved5 = 0x0;
-	scsi_id->reconnect_orb->status_FIFO_lo = SBP2_STATUS_FIFO_ADDRESS_LO +
-						 SBP2_STATUS_FIFO_ENTRY_TO_OFFSET(scsi_id->ud->id);
-	scsi_id->reconnect_orb->status_FIFO_hi =
-		(ORB_SET_NODE_ID(hi->host->node_id) | SBP2_STATUS_FIFO_ADDRESS_HI);
+	scsi_id->reconnect_orb->status_fifo_hi =
+		ORB_SET_STATUS_FIFO_HI(scsi_id->status_fifo_addr, hi->host->node_id);
+	scsi_id->reconnect_orb->status_fifo_lo =
+		ORB_SET_STATUS_FIFO_LO(scsi_id->status_fifo_addr);
 
 	/*
 	 * Byte swap ORB if necessary
@@ -2112,7 +2121,6 @@ static int sbp2_handle_status_write(struct hpsb_host *host, int nodeid, int dest
 {
 	struct sbp2scsi_host_info *hi;
 	struct scsi_id_instance_data *scsi_id = NULL, *scsi_id_tmp;
-	u32 id;
 	struct scsi_cmnd *SCpnt = NULL;
 	u32 scsi_status = SBP2_SCSI_STATUS_GOOD;
 	struct sbp2_command_info *command;
@@ -2135,12 +2143,12 @@ static int sbp2_handle_status_write(struct hpsb_host *host, int nodeid, int dest
 	}
 
 	/*
-	 * Find our scsi_id structure by looking at the status fifo address written to by
-	 * the sbp2 device.
+	 * Find our scsi_id structure by looking at the status fifo address
+	 * written to by the sbp2 device.
 	 */
-	id = SBP2_STATUS_FIFO_OFFSET_TO_ENTRY((u32)(addr - SBP2_STATUS_FIFO_ADDRESS));
 	list_for_each_entry(scsi_id_tmp, &hi->scsi_ids, scsi_list) {
-		if (scsi_id_tmp->ne->nodeid == nodeid && scsi_id_tmp->ud->id == id) {
+		if (scsi_id_tmp->ne->nodeid == nodeid &&
+		    scsi_id_tmp->status_fifo_addr == addr) {
 			scsi_id = scsi_id_tmp;
 			break;
 		}
diff --git a/drivers/ieee1394/sbp2.h b/drivers/ieee1394/sbp2.h
index 900ea1d25e71..e2d357a9ea3a 100644
--- a/drivers/ieee1394/sbp2.h
+++ b/drivers/ieee1394/sbp2.h
@@ -33,15 +33,17 @@
 #define ORB_DIRECTION_NO_DATA_TRANSFER  0x2
 
 #define ORB_SET_NULL_PTR(value)			((value & 0x1) << 31)
-#define ORB_SET_NOTIFY(value)                   ((value & 0x1) << 31)
-#define ORB_SET_RQ_FMT(value)                   ((value & 0x3) << 29)	/* unused ? */
+#define ORB_SET_NOTIFY(value)			((value & 0x1) << 31)
+#define ORB_SET_RQ_FMT(value)			((value & 0x3) << 29)	/* unused ? */
 #define ORB_SET_NODE_ID(value)			((value & 0xffff) << 16)
-#define ORB_SET_DATA_SIZE(value)                (value & 0xffff)
-#define ORB_SET_PAGE_SIZE(value)                ((value & 0x7) << 16)
-#define ORB_SET_PAGE_TABLE_PRESENT(value)       ((value & 0x1) << 19)
-#define ORB_SET_MAX_PAYLOAD(value)              ((value & 0xf) << 20)
-#define ORB_SET_SPEED(value)                    ((value & 0x7) << 24)
-#define ORB_SET_DIRECTION(value)                ((value & 0x1) << 27)
+#define ORB_SET_STATUS_FIFO_HI(value, id)	(value >> 32 | ORB_SET_NODE_ID(id))
+#define ORB_SET_STATUS_FIFO_LO(value)		(value & 0xffffffff)
+#define ORB_SET_DATA_SIZE(value)		(value & 0xffff)
+#define ORB_SET_PAGE_SIZE(value)		((value & 0x7) << 16)
+#define ORB_SET_PAGE_TABLE_PRESENT(value)	((value & 0x1) << 19)
+#define ORB_SET_MAX_PAYLOAD(value)		((value & 0xf) << 20)
+#define ORB_SET_SPEED(value)			((value & 0x7) << 24)
+#define ORB_SET_DIRECTION(value)		((value & 0x1) << 27)
 
 struct sbp2_command_orb {
 	volatile u32 next_ORB_hi;
@@ -76,8 +78,8 @@ struct sbp2_login_orb {
 	u32 login_response_lo;
 	u32 lun_misc;
 	u32 passwd_resp_lengths;
-	u32 status_FIFO_hi;
-	u32 status_FIFO_lo;
+	u32 status_fifo_hi;
+	u32 status_fifo_lo;
 };
 
 #define RESPONSE_GET_LOGIN_ID(value)            (value & 0xffff)
@@ -102,8 +104,8 @@ struct sbp2_query_logins_orb {
 	u32 query_response_lo;
 	u32 lun_misc;
 	u32 reserved_resp_length;
-	u32 status_FIFO_hi;
-	u32 status_FIFO_lo;
+	u32 status_fifo_hi;
+	u32 status_fifo_lo;
 };
 
 #define RESPONSE_GET_MAX_LOGINS(value)          (value & 0xffff)
@@ -123,8 +125,8 @@ struct sbp2_reconnect_orb {
 	u32 reserved4;
 	u32 login_ID_misc;
 	u32 reserved5;
-	u32 status_FIFO_hi;
-	u32 status_FIFO_lo;
+	u32 status_fifo_hi;
+	u32 status_fifo_lo;
 };
 
 struct sbp2_logout_orb {
@@ -134,8 +136,8 @@ struct sbp2_logout_orb {
 	u32 reserved4;
 	u32 login_ID_misc;
 	u32 reserved5;
-	u32 status_FIFO_hi;
-	u32 status_FIFO_lo;
+	u32 status_fifo_hi;
+	u32 status_fifo_lo;
 };
 
 #define PAGE_TABLE_SET_SEGMENT_BASE_HI(value)   (value & 0xffff)
@@ -195,30 +197,6 @@ struct sbp2_status_block {
  * Miscellaneous SBP2 related config rom defines
  */
 
-/* The status fifo address definition below is used as a base for each
- * node, which a chunk seperately assigned to each unit directory in the
- * node.  For example, 0xfffe00000000ULL is used for the first sbp2 device
- * detected on node 0, 0xfffe00000020ULL for the next sbp2 device on node
- * 0, and so on.
- *
- * Note: We could use a single status fifo address for all sbp2 devices,
- * and figure out which sbp2 device the status belongs to by looking at
- * the source node id of the status write... but, using separate addresses
- * for each sbp2 unit directory allows for better code and the ability to
- * support multiple luns within a single 1394 node.
- *
- * Also note that we choose the address range below as it is a region
- * specified for write posting, where the ohci controller will
- * automatically send an ack_complete when the status is written by the
- * sbp2 device... saving a split transaction.   =)
- */
-#define SBP2_STATUS_FIFO_ADDRESS				0xfffe00000000ULL
-#define SBP2_STATUS_FIFO_ADDRESS_HI                             0xfffe
-#define SBP2_STATUS_FIFO_ADDRESS_LO                             0x0
-
-#define SBP2_STATUS_FIFO_ENTRY_TO_OFFSET(entry)			((entry) << 5)
-#define SBP2_STATUS_FIFO_OFFSET_TO_ENTRY(offset)		((offset) >> 5)
-
 #define SBP2_UNIT_DIRECTORY_OFFSET_KEY				0xd1
 #define SBP2_CSR_OFFSET_KEY					0x54
 #define SBP2_UNIT_SPEC_ID_KEY					0x12
@@ -258,7 +236,6 @@ struct sbp2_status_block {
  */
 
 #define SBP2_MAX_SG_ELEMENT_LENGTH	0xf000
-#define SBP2_MAX_UDS_PER_NODE		16	/* Maximum scsi devices per node */
 #define SBP2_MAX_SECTORS		255	/* Max sectors supported */
 #define SBP2_MAX_CMDS			8	/* This should be safe */
 
@@ -337,6 +314,11 @@ struct scsi_id_instance_data {
 	u32 sbp2_lun;
 	u32 sbp2_firmware_revision;
 
+	/*
+	 * Address for the device to write status blocks to
+	 */
+	u64 status_fifo_addr;
+
 	/*
 	 * Variable used for logins, reconnects, logouts, query logins
 	 */

From a80614d1adba903a1e5cb22bf14ebc640fc2ba4c Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Tue, 14 Feb 2006 22:04:19 -0500
Subject: [PATCH 393/509] sbp2: update 36byte inquiry workaround (fix
 compatibility regression)

Since about Linux 2.6.14, sbp2's inquiry workaround did not work anymore
due to changes in the SCSI layer. Update it to become effective again.
Testing one of the two known affected bridges has shown that skip_ms_page_8
is required as well.

Also, make force_inquiry_hack tunable via /sys/module/sbp2/parameters.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Jody McIntyre <scjody@modernduck.com>
(cherry picked from 99496037c6744fd938ffb8ccfc8fc91762322ff8 commit)
---
 drivers/ieee1394/sbp2.c | 42 +++++++++++++++++++----------------------
 1 file changed, 19 insertions(+), 23 deletions(-)

diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c
index 0672224fa109..eca92eb475a1 100644
--- a/drivers/ieee1394/sbp2.c
+++ b/drivers/ieee1394/sbp2.c
@@ -137,15 +137,15 @@ MODULE_PARM_DESC(exclusive_login, "Exclusive login to sbp2 device (default = 1)"
 /*
  * SCSI inquiry hack for really badly behaved sbp2 devices. Turn this on
  * if your sbp2 device is not properly handling the SCSI inquiry command.
- * This hack makes the inquiry look more like a typical MS Windows
- * inquiry.
+ * This hack makes the inquiry look more like a typical MS Windows inquiry
+ * by enforcing 36 byte inquiry and avoiding access to mode_sense page 8.
  *
  * If force_inquiry_hack=1 is required for your device to work,
  * please submit the logged sbp2_firmware_revision value of this device to
  * the linux1394-devel mailing list.
  */
 static int force_inquiry_hack;
-module_param(force_inquiry_hack, int, 0444);
+module_param(force_inquiry_hack, int, 0644);
 MODULE_PARM_DESC(force_inquiry_hack, "Force SCSI inquiry hack (default = 0)");
 
 /*
@@ -264,18 +264,17 @@ static struct hpsb_protocol_driver sbp2_driver = {
 	},
 };
 
-
-/* List of device firmware's that require a forced 36 byte inquiry.  */
+/*
+ * List of device firmwares that require the inquiry hack.
+ * Yields a few false positives but did not break other devices so far.
+ */
 static u32 sbp2_broken_inquiry_list[] = {
-	0x00002800,	/* Stefan Richter <richtest@bauwesen.tu-cottbus.de> */
+	0x00002800,	/* Stefan Richter <stefanr@s5r6.in-berlin.de> */
 			/* DViCO Momobay CX-1 */
 	0x00000200	/* Andreas Plesch <plesch@fas.harvard.edu> */
 			/* QPS Fire DVDBurner */
 };
 
-#define NUM_BROKEN_INQUIRY_DEVS \
-	(sizeof(sbp2_broken_inquiry_list)/sizeof(*sbp2_broken_inquiry_list))
-
 /**************************************
  * General utility functions
  **************************************/
@@ -1575,7 +1574,7 @@ static void sbp2_parse_unit_directory(struct scsi_id_instance_data *scsi_id,
 	/* Check for a blacklisted set of devices that require us to force
 	 * a 36 byte host inquiry. This can be overriden as a module param
 	 * (to force all hosts).  */
-	for (i = 0; i < NUM_BROKEN_INQUIRY_DEVS; i++) {
+	for (i = 0; i < ARRAY_SIZE(sbp2_broken_inquiry_list); i++) {
 		if ((firmware_revision & 0xffff00) ==
 				sbp2_broken_inquiry_list[i]) {
 			SBP2_WARN("Node " NODE_BUS_FMT ": Using 36byte inquiry workaround",
@@ -2021,18 +2020,6 @@ static int sbp2_send_command(struct scsi_id_instance_data *scsi_id,
 		return -EIO;
 	}
 
-	/*
-	 * The scsi stack sends down a request_bufflen which does not match the
-	 * length field in the scsi cdb. This causes some sbp2 devices to
-	 * reject this inquiry command. Fix the request_bufflen.
-	 */
-	if (*cmd == INQUIRY) {
-		if (force_inquiry_hack || scsi_id->workarounds & SBP2_BREAKAGE_INQUIRY_HACK)
-			request_bufflen = cmd[4] = 0x24;
-		else
-			request_bufflen = cmd[4];
-	}
-
 	/*
 	 * Now actually fill in the comamnd orb and sbp2 s/g list
 	 */
@@ -2489,7 +2476,16 @@ static void sbp2scsi_complete_command(struct scsi_id_instance_data *scsi_id,
 
 static int sbp2scsi_slave_alloc(struct scsi_device *sdev)
 {
-	((struct scsi_id_instance_data *)sdev->host->hostdata[0])->sdev = sdev;
+	struct scsi_id_instance_data *scsi_id =
+		(struct scsi_id_instance_data *)sdev->host->hostdata[0];
+
+	scsi_id->sdev = sdev;
+
+	if (force_inquiry_hack ||
+	    scsi_id->workarounds & SBP2_BREAKAGE_INQUIRY_HACK) {
+		sdev->inquiry_len = 36;
+		sdev->skip_ms_page_8 = 1;
+	}
 	return 0;
 }
 

From 85edae14e4ee5e68cf037e9e4bca7498ea16874d Mon Sep 17 00:00:00 2001
From: Michal Janusz Miroslaw <M.Miroslaw@elka.pw.edu.pl>
Date: Thu, 23 Feb 2006 09:49:35 +0000
Subject: [PATCH 394/509] [SERIAL] Trivial comment fix:
 include/linux/serial_reg.h

Trivial comment fix for include/linux/serial_reg.h

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 include/linux/serial_reg.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/linux/serial_reg.h b/include/linux/serial_reg.h
index 6a2bb955844b..3c8a6aa77415 100644
--- a/include/linux/serial_reg.h
+++ b/include/linux/serial_reg.h
@@ -247,10 +247,10 @@
 #define UART_CTR	0xFF
 
 /*
- * The 16C950 Additional Control Reigster
+ * The 16C950 Additional Control Register
  */
 #define UART_ACR_RXDIS	0x01	/* Receiver disable */
-#define UART_ACR_TXDIS	0x02	/* Receiver disable */
+#define UART_ACR_TXDIS	0x02	/* Transmitter disable */
 #define UART_ACR_DSRFC	0x04	/* DSR Flow Control */
 #define UART_ACR_TLENB	0x20	/* 950 trigger levels enable */
 #define UART_ACR_ICRRD	0x40	/* ICR Read enable */

From d856c66618f953fc3cd1e613226d5f098ad322c8 Mon Sep 17 00:00:00 2001
From: Russell King <rmk@dyn-67.arm.linux.org.uk>
Date: Thu, 23 Feb 2006 10:22:13 +0000
Subject: [PATCH 395/509] [SERIAL] Add comment about early_serial_setup()

early_serial_setup() must not be called after console initialisation.
Add a comment prior to the function explicitly stating this.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/serial/8250.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 244e8ff11977..7aca22c9976d 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -2326,6 +2326,12 @@ static struct uart_driver serial8250_reg = {
 	.cons			= SERIAL8250_CONSOLE,
 };
 
+/*
+ * early_serial_setup - early registration for 8250 ports
+ *
+ * Setup an 8250 port structure prior to console initialisation.  Use
+ * after console initialisation will cause undefined behaviour.
+ */
 int __init early_serial_setup(struct uart_port *port)
 {
 	if (port->line >= ARRAY_SIZE(serial8250_ports))

From bd6ef57e08d6cce32e93f9fc7b93d361b6a7884f Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Mon, 20 Feb 2006 19:07:31 +1100
Subject: [PATCH 396/509] [PATCH] powerpc: Initialise hvlpevent_queue.lock
 correctly

When I changed the hvlpevent_queue code to use a spinlock instead of a
custom atomic (719d1cd86780c156f954fc34f34481adac197aec) I didn't
initialise the lock anywhere, oops.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/iseries/lpevents.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/powerpc/platforms/iseries/lpevents.c b/arch/powerpc/platforms/iseries/lpevents.c
index 0b885300d1d1..8ca7b9396355 100644
--- a/arch/powerpc/platforms/iseries/lpevents.c
+++ b/arch/powerpc/platforms/iseries/lpevents.c
@@ -184,6 +184,8 @@ void setup_hvlpevent_queue(void)
 {
 	void *eventStack;
 
+	spin_lock_init(&hvlpevent_queue.lock);
+
 	/* Allocate a page for the Event Stack. */
 	eventStack = alloc_bootmem_pages(LpEventStackSize);
 	memset(eventStack, 0, LpEventStackSize);

From 5d06a99f543e734ceb53bbc9e550537be97f0c49 Mon Sep 17 00:00:00 2001
From: Francois Romieu <romieu@fr.zoreil.com>
Date: Thu, 23 Feb 2006 00:47:58 +0100
Subject: [PATCH 397/509] r8169: fix broken ring index handling in
 suspend/resume

rtl8169_hw_start() requires that the descriptor ring indexes be
set to zero. Let a deferred invocation of rtl8169_reset_task()
handle it. Enabling a few power management bits will not hurt
either.

suspend/resume is issued with irq on: the spinlock do not need
to save the irq flag.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
---
 drivers/net/r8169.c | 102 +++++++++++++++++++++++++-------------------
 1 file changed, 59 insertions(+), 43 deletions(-)

diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index 6e1018448eea..999fd6cef77e 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -287,6 +287,12 @@ enum RTL8169_register_content {
 	TxInterFrameGapShift = 24,
 	TxDMAShift = 8,	/* DMA burst value (0-7) is shift this many bits */
 
+	/* Config1 register p.24 */
+	PMEnable	= (1 << 0),	/* Power Management Enable */
+
+	/* Config5 register p.27 */
+	PMEStatus	= (1 << 0),	/* PME status can be reset by PCI RST# */
+
 	/* TBICSR p.28 */
 	TBIReset	= 0x80000000,
 	TBILoopback	= 0x40000000,
@@ -1442,6 +1448,11 @@ rtl8169_init_board(struct pci_dev *pdev, struct net_device **dev_out,
 	}
 	tp->chipset = i;
 
+	RTL_W8(Cfg9346, Cfg9346_Unlock);
+	RTL_W8(Config1, RTL_R8(Config1) | PMEnable);
+	RTL_W8(Config5, RTL_R8(Config5) & PMEStatus);
+	RTL_W8(Cfg9346, Cfg9346_Lock);
+
 	*ioaddr_out = ioaddr;
 	*dev_out = dev;
 out:
@@ -1612,49 +1623,6 @@ rtl8169_remove_one(struct pci_dev *pdev)
 	pci_set_drvdata(pdev, NULL);
 }
 
-#ifdef CONFIG_PM
-
-static int rtl8169_suspend(struct pci_dev *pdev, pm_message_t state)
-{
-	struct net_device *dev = pci_get_drvdata(pdev);
-	struct rtl8169_private *tp = netdev_priv(dev);
-	void __iomem *ioaddr = tp->mmio_addr;
-	unsigned long flags;
-
-	if (!netif_running(dev))
-		return 0;
-	
-	netif_device_detach(dev);
-	netif_stop_queue(dev);
-	spin_lock_irqsave(&tp->lock, flags);
-
-	/* Disable interrupts, stop Rx and Tx */
-	RTL_W16(IntrMask, 0);
-	RTL_W8(ChipCmd, 0);
-		
-	/* Update the error counts. */
-	tp->stats.rx_missed_errors += RTL_R32(RxMissed);
-	RTL_W32(RxMissed, 0);
-	spin_unlock_irqrestore(&tp->lock, flags);
-	
-	return 0;
-}
-
-static int rtl8169_resume(struct pci_dev *pdev)
-{
-	struct net_device *dev = pci_get_drvdata(pdev);
-
-	if (!netif_running(dev))
-	    return 0;
-
-	netif_device_attach(dev);
-	rtl8169_hw_start(dev);
-
-	return 0;
-}
-                                                                                
-#endif /* CONFIG_PM */
-
 static void rtl8169_set_rxbufsize(struct rtl8169_private *tp,
 				  struct net_device *dev)
 {
@@ -2700,6 +2668,54 @@ static struct net_device_stats *rtl8169_get_stats(struct net_device *dev)
 	return &tp->stats;
 }
 
+#ifdef CONFIG_PM
+
+static int rtl8169_suspend(struct pci_dev *pdev, pm_message_t state)
+{
+	struct net_device *dev = pci_get_drvdata(pdev);
+	struct rtl8169_private *tp = netdev_priv(dev);
+	void __iomem *ioaddr = tp->mmio_addr;
+
+	if (!netif_running(dev))
+		goto out;
+
+	netif_device_detach(dev);
+	netif_stop_queue(dev);
+
+	spin_lock_irq(&tp->lock);
+
+	rtl8169_asic_down(ioaddr);
+
+	tp->stats.rx_missed_errors += RTL_R32(RxMissed);
+	RTL_W32(RxMissed, 0);
+
+	spin_unlock_irq(&tp->lock);
+
+	pci_save_state(pdev);
+	pci_set_power_state(pdev, pci_choose_state(pdev, state));
+out:
+	return 0;
+}
+
+static int rtl8169_resume(struct pci_dev *pdev)
+{
+	struct net_device *dev = pci_get_drvdata(pdev);
+
+	if (!netif_running(dev))
+		goto out;
+
+	netif_device_attach(dev);
+
+	pci_set_power_state(pdev, PCI_D0);
+	pci_restore_state(pdev);
+
+	rtl8169_schedule_work(dev, rtl8169_reset_task);
+out:
+	return 0;
+}
+
+#endif /* CONFIG_PM */
+
 static struct pci_driver rtl8169_pci_driver = {
 	.name		= MODULENAME,
 	.id_table	= rtl8169_pci_tbl,

From 61a4dcc2f9b5c6861e7198b80dd73dd6e9247b7b Mon Sep 17 00:00:00 2001
From: Francois Romieu <romieu@fr.zoreil.com>
Date: Thu, 23 Feb 2006 00:55:25 +0100
Subject: [PATCH 398/509] r8169: enable wake on lan

Similar to 8139cp code but more inspired/lucky.

Signed-off-by: Francois Romieu <romieu@fr.zoreil.com>
---
 drivers/net/r8169.c | 87 +++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 87 insertions(+)

diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c
index 999fd6cef77e..8cc0d0bbdf50 100644
--- a/drivers/net/r8169.c
+++ b/drivers/net/r8169.c
@@ -290,7 +290,15 @@ enum RTL8169_register_content {
 	/* Config1 register p.24 */
 	PMEnable	= (1 << 0),	/* Power Management Enable */
 
+	/* Config3 register p.25 */
+	MagicPacket	= (1 << 5),	/* Wake up when receives a Magic Packet */
+	LinkUp		= (1 << 4),	/* Wake up when the cable connection is re-established */
+
 	/* Config5 register p.27 */
+	BWF		= (1 << 6),	/* Accept Broadcast wakeup frame */
+	MWF		= (1 << 5),	/* Accept Multicast wakeup frame */
+	UWF		= (1 << 4),	/* Accept Unicast wakeup frame */
+	LanWake		= (1 << 1),	/* LanWake enable/disable */
 	PMEStatus	= (1 << 0),	/* PME status can be reset by PCI RST# */
 
 	/* TBICSR p.28 */
@@ -439,6 +447,7 @@ struct rtl8169_private {
 	unsigned int (*phy_reset_pending)(void __iomem *);
 	unsigned int (*link_ok)(void __iomem *);
 	struct work_struct task;
+	unsigned wol_enabled : 1;
 };
 
 MODULE_AUTHOR("Realtek and the Linux r8169 crew <netdev@vger.kernel.org>");
@@ -613,6 +622,80 @@ static void rtl8169_link_option(int idx, u8 *autoneg, u16 *speed, u8 *duplex)
 	*duplex = p->duplex;
 }
 
+static void rtl8169_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+	struct rtl8169_private *tp = netdev_priv(dev);
+	void __iomem *ioaddr = tp->mmio_addr;
+	u8 options;
+
+	wol->wolopts = 0;
+
+#define WAKE_ANY (WAKE_PHY | WAKE_MAGIC | WAKE_UCAST | WAKE_BCAST | WAKE_MCAST)
+	wol->supported = WAKE_ANY;
+
+	spin_lock_irq(&tp->lock);
+
+	options = RTL_R8(Config1);
+	if (!(options & PMEnable))
+		goto out_unlock;
+
+	options = RTL_R8(Config3);
+	if (options & LinkUp)
+		wol->wolopts |= WAKE_PHY;
+	if (options & MagicPacket)
+		wol->wolopts |= WAKE_MAGIC;
+
+	options = RTL_R8(Config5);
+	if (options & UWF)
+		wol->wolopts |= WAKE_UCAST;
+	if (options & BWF)
+	        wol->wolopts |= WAKE_BCAST;
+	if (options & MWF)
+	        wol->wolopts |= WAKE_MCAST;
+
+out_unlock:
+	spin_unlock_irq(&tp->lock);
+}
+
+static int rtl8169_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol)
+{
+	struct rtl8169_private *tp = netdev_priv(dev);
+	void __iomem *ioaddr = tp->mmio_addr;
+	int i;
+	static struct {
+		u32 opt;
+		u16 reg;
+		u8  mask;
+	} cfg[] = {
+		{ WAKE_ANY,   Config1, PMEnable },
+		{ WAKE_PHY,   Config3, LinkUp },
+		{ WAKE_MAGIC, Config3, MagicPacket },
+		{ WAKE_UCAST, Config5, UWF },
+		{ WAKE_BCAST, Config5, BWF },
+		{ WAKE_MCAST, Config5, MWF },
+		{ WAKE_ANY,   Config5, LanWake }
+	};
+
+	spin_lock_irq(&tp->lock);
+
+	RTL_W8(Cfg9346, Cfg9346_Unlock);
+
+	for (i = 0; i < ARRAY_SIZE(cfg); i++) {
+		u8 options = RTL_R8(cfg[i].reg) & ~cfg[i].mask;
+		if (wol->wolopts & cfg[i].opt)
+			options |= cfg[i].mask;
+		RTL_W8(cfg[i].reg, options);
+	}
+
+	RTL_W8(Cfg9346, Cfg9346_Lock);
+
+	tp->wol_enabled = (wol->wolopts) ? 1 : 0;
+
+	spin_unlock_irq(&tp->lock);
+
+	return 0;
+}
+
 static void rtl8169_get_drvinfo(struct net_device *dev,
 				struct ethtool_drvinfo *info)
 {
@@ -1031,6 +1114,8 @@ static struct ethtool_ops rtl8169_ethtool_ops = {
 	.get_tso		= ethtool_op_get_tso,
 	.set_tso		= ethtool_op_set_tso,
 	.get_regs		= rtl8169_get_regs,
+	.get_wol		= rtl8169_get_wol,
+	.set_wol		= rtl8169_set_wol,
 	.get_strings		= rtl8169_get_strings,
 	.get_stats_count	= rtl8169_get_stats_count,
 	.get_ethtool_stats	= rtl8169_get_ethtool_stats,
@@ -2692,6 +2777,7 @@ static int rtl8169_suspend(struct pci_dev *pdev, pm_message_t state)
 	spin_unlock_irq(&tp->lock);
 
 	pci_save_state(pdev);
+	pci_enable_wake(pdev, pci_choose_state(pdev, state), tp->wol_enabled);
 	pci_set_power_state(pdev, pci_choose_state(pdev, state));
 out:
 	return 0;
@@ -2708,6 +2794,7 @@ static int rtl8169_resume(struct pci_dev *pdev)
 
 	pci_set_power_state(pdev, PCI_D0);
 	pci_restore_state(pdev);
+	pci_enable_wake(pdev, PCI_D0, 0);
 
 	rtl8169_schedule_work(dev, rtl8169_reset_task);
 out:

From a9cdab869ec343ccc601484fb535813e16c25f70 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 10:28:33 -0800
Subject: [PATCH 399/509] skge: NAPI/irq race fix

Fix a race in the receive NAPI, irq handling. The interrupt clear and the
start need to be separated.  Otherwise there is a window between the last
frame received and the NAPI done level handling.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/skge.c | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

diff --git a/drivers/net/skge.c b/drivers/net/skge.c
index 67fb19b8fde9..869c7cfb99a4 100644
--- a/drivers/net/skge.c
+++ b/drivers/net/skge.c
@@ -2678,8 +2678,7 @@ static int skge_poll(struct net_device *dev, int *budget)
 
 	/* restart receiver */
 	wmb();
-	skge_write8(hw, Q_ADDR(rxqaddr[skge->port], Q_CSR),
-		    CSR_START | CSR_IRQ_CL_F);
+	skge_write8(hw, Q_ADDR(rxqaddr[skge->port], Q_CSR), CSR_START);
 
 	*budget -= work_done;
 	dev->quota -= work_done;
@@ -2856,14 +2855,6 @@ static void skge_extirq(unsigned long data)
 	local_irq_enable();
 }
 
-static inline void skge_wakeup(struct net_device *dev)
-{
-	struct skge_port *skge = netdev_priv(dev);
-
-	prefetch(skge->rx_ring.to_clean);
-	netif_rx_schedule(dev);
-}
-
 static irqreturn_t skge_intr(int irq, void *dev_id, struct pt_regs *regs)
 {
 	struct skge_hw *hw = dev_id;
@@ -2874,13 +2865,15 @@ static irqreturn_t skge_intr(int irq, void *dev_id, struct pt_regs *regs)
 
 	status &= hw->intr_mask;
 	if (status & IS_R1_F) {
+		skge_write8(hw, Q_ADDR(Q_R1, Q_CSR), CSR_IRQ_CL_F);
 		hw->intr_mask &= ~IS_R1_F;
-		skge_wakeup(hw->dev[0]);
+		netif_rx_schedule(hw->dev[0]);
 	}
 
 	if (status & IS_R2_F) {
+		skge_write8(hw, Q_ADDR(Q_R2, Q_CSR), CSR_IRQ_CL_F);
 		hw->intr_mask &= ~IS_R2_F;
-		skge_wakeup(hw->dev[1]);
+		netif_rx_schedule(hw->dev[1]);
 	}
 
 	if (status & IS_XA1_F)

From 0781191cf69b7635e0d3ea55c6019e789d1936fa Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 10:28:34 -0800
Subject: [PATCH 400/509] skge: genesis phy initialzation

The SysKonnect Genesis based board would fail on initialization
with phy_read errors caused by not waiting for last phy write.

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/skge.c | 37 ++++++++++++++++++++++++++-----------
 1 file changed, 26 insertions(+), 11 deletions(-)

diff --git a/drivers/net/skge.c b/drivers/net/skge.c
index 869c7cfb99a4..af2e6782031b 100644
--- a/drivers/net/skge.c
+++ b/drivers/net/skge.c
@@ -879,13 +879,12 @@ static int __xm_phy_read(struct skge_hw *hw, int port, u16 reg, u16 *val)
 	int i;
 
 	xm_write16(hw, port, XM_PHY_ADDR, reg | hw->phy_addr);
-	xm_read16(hw, port, XM_PHY_DATA);
+	*val = xm_read16(hw, port, XM_PHY_DATA);
 
-	/* Need to wait for external PHY */
 	for (i = 0; i < PHY_RETRIES; i++) {
-		udelay(1);
 		if (xm_read16(hw, port, XM_MMU_CMD) & XM_MMU_PHY_RDY)
 			goto ready;
+		udelay(1);
 	}
 
 	return -ETIMEDOUT;
@@ -918,7 +917,12 @@ static int xm_phy_write(struct skge_hw *hw, int port, u16 reg, u16 val)
 
  ready:
 	xm_write16(hw, port, XM_PHY_DATA, val);
-	return 0;
+	for (i = 0; i < PHY_RETRIES; i++) {
+		if (!(xm_read16(hw, port, XM_MMU_CMD) & XM_MMU_PHY_BUSY))
+			return 0;
+		udelay(1);
+	}
+	return -ETIMEDOUT;
 }
 
 static void genesis_init(struct skge_hw *hw)
@@ -1168,13 +1172,17 @@ static void genesis_mac_init(struct skge_hw *hw, int port)
 	u32 r;
 	const u8 zero[6]  = { 0 };
 
-	/* Clear MIB counters */
-	xm_write16(hw, port, XM_STAT_CMD,
-			XM_SC_CLR_RXC | XM_SC_CLR_TXC);
-	/* Clear two times according to Errata #3 */
-	xm_write16(hw, port, XM_STAT_CMD,
-			XM_SC_CLR_RXC | XM_SC_CLR_TXC);
+	for (i = 0; i < 10; i++) {
+		skge_write16(hw, SK_REG(port, TX_MFF_CTRL1),
+			     MFF_SET_MAC_RST);
+		if (skge_read16(hw, SK_REG(port, TX_MFF_CTRL1)) & MFF_SET_MAC_RST)
+			goto reset_ok;
+		udelay(1);
+	}
 
+	printk(KERN_WARNING PFX "%s: genesis reset failed\n", dev->name);
+
+ reset_ok:
 	/* Unreset the XMAC. */
 	skge_write16(hw, SK_REG(port, TX_MFF_CTRL1), MFF_CLR_MAC_RST);
 
@@ -1191,7 +1199,7 @@ static void genesis_mac_init(struct skge_hw *hw, int port)
 		r |= GP_DIR_2|GP_IO_2;
 
 	skge_write32(hw, B2_GP_IO, r);
-	skge_read32(hw, B2_GP_IO);
+
 
 	/* Enable GMII interface */
 	xm_write16(hw, port, XM_HW_CFG, XM_HW_GMII_MD);
@@ -1205,6 +1213,13 @@ static void genesis_mac_init(struct skge_hw *hw, int port)
 	for (i = 1; i < 16; i++)
 		xm_outaddr(hw, port, XM_EXM(i), zero);
 
+	/* Clear MIB counters */
+	xm_write16(hw, port, XM_STAT_CMD,
+			XM_SC_CLR_RXC | XM_SC_CLR_TXC);
+	/* Clear two times according to Errata #3 */
+	xm_write16(hw, port, XM_STAT_CMD,
+			XM_SC_CLR_RXC | XM_SC_CLR_TXC);
+
 	/* configure Rx High Water Mark (XM_RX_HI_WM) */
 	xm_write16(hw, port, XM_RX_HI_WM, 1450);
 

From 80dd857daca1cf541b10118991569470d62c1d38 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@osdl.org>
Date: Wed, 22 Feb 2006 10:28:35 -0800
Subject: [PATCH 401/509] skge: protect interrupt mask

There is a race between updating the irq mask and setting it
which can be triggered on SMP with a bad cable.
Similar patch from Ingo Molnar and Thomas Gleixner

Signed-off-by: Stephen Hemminger <shemminger@osdl.org>
---
 drivers/net/skge.c | 21 ++++++++++++++-------
 drivers/net/skge.h |  1 +
 2 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/drivers/net/skge.c b/drivers/net/skge.c
index af2e6782031b..25e028b7ce48 100644
--- a/drivers/net/skge.c
+++ b/drivers/net/skge.c
@@ -2185,8 +2185,10 @@ static int skge_up(struct net_device *dev)
 	skge->tx_avail = skge->tx_ring.count - 1;
 
 	/* Enable IRQ from port */
+	spin_lock_irq(&hw->hw_lock);
 	hw->intr_mask |= portirqmask[port];
 	skge_write32(hw, B0_IMSK, hw->intr_mask);
+	spin_unlock_irq(&hw->hw_lock);
 
 	/* Initialize MAC */
 	spin_lock_bh(&hw->phy_lock);
@@ -2244,8 +2246,10 @@ static int skge_down(struct net_device *dev)
 	else
 		yukon_stop(skge);
 
+	spin_lock_irq(&hw->hw_lock);
 	hw->intr_mask &= ~portirqmask[skge->port];
 	skge_write32(hw, B0_IMSK, hw->intr_mask);
+	spin_unlock_irq(&hw->hw_lock);
 
 	/* Stop transmitter */
 	skge_write8(hw, Q_ADDR(txqaddr[port], Q_CSR), CSR_STOP);
@@ -2701,10 +2705,11 @@ static int skge_poll(struct net_device *dev, int *budget)
 	if (work_done >=  to_do)
 		return 1; /* not done */
 
-	netif_rx_complete(dev);
-	hw->intr_mask |= portirqmask[skge->port];
-	skge_write32(hw, B0_IMSK, hw->intr_mask);
-	skge_read32(hw, B0_IMSK);
+	spin_lock_irq(&hw->hw_lock);
+	__netif_rx_complete(dev);
+  	hw->intr_mask |= portirqmask[skge->port];
+  	skge_write32(hw, B0_IMSK, hw->intr_mask);
+ 	spin_unlock_irq(&hw->hw_lock);
 
 	return 0;
 }
@@ -2864,10 +2869,10 @@ static void skge_extirq(unsigned long data)
 	}
 	spin_unlock(&hw->phy_lock);
 
-	local_irq_disable();
+	spin_lock_irq(&hw->hw_lock);
 	hw->intr_mask |= IS_EXT_REG;
 	skge_write32(hw, B0_IMSK, hw->intr_mask);
-	local_irq_enable();
+	spin_unlock_irq(&hw->hw_lock);
 }
 
 static irqreturn_t skge_intr(int irq, void *dev_id, struct pt_regs *regs)
@@ -2878,7 +2883,7 @@ static irqreturn_t skge_intr(int irq, void *dev_id, struct pt_regs *regs)
 	if (status == 0 || status == ~0) /* hotplug or shared irq */
 		return IRQ_NONE;
 
-	status &= hw->intr_mask;
+	spin_lock(&hw->hw_lock);
 	if (status & IS_R1_F) {
 		skge_write8(hw, Q_ADDR(Q_R1, Q_CSR), CSR_IRQ_CL_F);
 		hw->intr_mask &= ~IS_R1_F;
@@ -2930,6 +2935,7 @@ static irqreturn_t skge_intr(int irq, void *dev_id, struct pt_regs *regs)
 	}
 
 	skge_write32(hw, B0_IMSK, hw->intr_mask);
+	spin_unlock(&hw->hw_lock);
 
 	return IRQ_HANDLED;
 }
@@ -3298,6 +3304,7 @@ static int __devinit skge_probe(struct pci_dev *pdev,
 
 	hw->pdev = pdev;
 	spin_lock_init(&hw->phy_lock);
+	spin_lock_init(&hw->hw_lock);
 	tasklet_init(&hw->ext_tasklet, skge_extirq, (unsigned long) hw);
 
 	hw->regs = ioremap_nocache(pci_resource_start(pdev, 0), 0x4000);
diff --git a/drivers/net/skge.h b/drivers/net/skge.h
index 2efdacc290e5..941f12a333b6 100644
--- a/drivers/net/skge.h
+++ b/drivers/net/skge.h
@@ -2402,6 +2402,7 @@ struct skge_hw {
 
 	struct tasklet_struct ext_tasklet;
 	spinlock_t	     phy_lock;
+	spinlock_t	     hw_lock;
 };
 
 enum {

From 42cf93cd464e0df3c85d298c647411bae6d99e6e Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Tue, 21 Feb 2006 13:37:35 -0800
Subject: [PATCH 402/509] [NETFILTER]: Fix bridge netfilter related in
 xfrm_lookup

The bridge-netfilter code attaches a fake dst_entry with dst->ops == NULL
to purely bridged packets. When these packets are SNATed and a policy
lookup is done, xfrm_lookup crashes because it tries to dereference
dst->ops.

Change xfrm_lookup not to dereference dst->ops before checking for the
DST_NOXFRM flag and set this flag in the fake dst_entry.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/br_netfilter.c | 1 +
 net/xfrm/xfrm_policy.c    | 7 ++++---
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index 6bb0c7eb1ef0..e060aad8624d 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -90,6 +90,7 @@ static struct rtable __fake_rtable = {
 			.dev			= &__fake_net_device,
 			.path			= &__fake_rtable.u.dst,
 			.metrics		= {[RTAX_MTU - 1] = 1500},
+			.flags			= DST_NOXFRM,
 		}
 	},
 	.rt_flags	= 0,
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 5e6b05ac1260..8206025d8e46 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -782,7 +782,7 @@ int xfrm_lookup(struct dst_entry **dst_p, struct flowi *fl,
 	int nx = 0;
 	int err;
 	u32 genid;
-	u16 family = dst_orig->ops->family;
+	u16 family;
 	u8 dir = policy_to_flow_dir(XFRM_POLICY_OUT);
 	u32 sk_sid = security_sk_sid(sk, fl, dir);
 restart:
@@ -796,13 +796,14 @@ restart:
 		if ((dst_orig->flags & DST_NOXFRM) || !xfrm_policy_list[XFRM_POLICY_OUT])
 			return 0;
 
-		policy = flow_cache_lookup(fl, sk_sid, family, dir,
-					   xfrm_policy_lookup);
+		policy = flow_cache_lookup(fl, sk_sid, dst_orig->ops->family,
+					   dir, xfrm_policy_lookup);
 	}
 
 	if (!policy)
 		return 0;
 
+	family = dst_orig->ops->family;
 	policy->curlft.use_time = (unsigned long)xtime.tv_sec;
 
 	switch (policy->action) {

From 85259878499d6c428cba191bb4e415a250dcd75a Mon Sep 17 00:00:00 2001
From: Suresh Bhogavilli <sbhogavilli@verisign.com>
Date: Tue, 21 Feb 2006 13:42:22 -0800
Subject: [PATCH 403/509] [IPV4]: Fix garbage collection of multipath route
 entries

When garbage collecting route cache entries of multipath routes
in rt_garbage_collect(), entries were deleted from the hash bucket
'i' while holding a spin lock on bucket 'k' resulting in a system
hang.  Delete entries, if any, from bucket 'k' instead.

Signed-off-by: Suresh Bhogavilli <sbhogavilli@verisign.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/route.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index d82c242ea704..fca5fe0cf94a 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -835,7 +835,7 @@ static int rt_garbage_collect(void)
 					int r;
 
 					rthp = rt_remove_balanced_route(
-						&rt_hash_table[i].chain,
+						&rt_hash_table[k].chain,
 						rth,
 						&r);
 					goal -= r;

From 21380b81ef8699179b535e197a95b891a7badac7 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Wed, 22 Feb 2006 14:47:13 -0800
Subject: [PATCH 404/509] [XFRM]: Eliminate refcounting confusion by creating
 __xfrm_state_put().

We often just do an atomic_dec(&x->refcnt) on an xfrm_state object
because we know there is more than 1 reference remaining and thus
we can elide the heavier xfrm_state_put() call.

Do this behind an inline function called __xfrm_state_put() so that is
more obvious and also to allow us to more cleanly add refcount
debugging later.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h    | 5 +++++
 net/key/af_key.c      | 2 +-
 net/xfrm/xfrm_state.c | 8 ++++----
 net/xfrm/xfrm_user.c  | 2 +-
 4 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index d6111a2f0a23..004e645f3e18 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -403,6 +403,11 @@ unsigned xfrm_spi_hash(xfrm_address_t *addr, u32 spi, u8 proto, unsigned short f
 
 extern void __xfrm_state_destroy(struct xfrm_state *);
 
+static inline void __xfrm_state_put(struct xfrm_state *x)
+{
+	atomic_dec(&x->refcnt);
+}
+
 static inline void xfrm_state_put(struct xfrm_state *x)
 {
 	if (atomic_dec_and_test(&x->refcnt))
diff --git a/net/key/af_key.c b/net/key/af_key.c
index ae86d237a456..b2d4d1dd2116 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -1423,7 +1423,7 @@ static int pfkey_add(struct sock *sk, struct sk_buff *skb, struct sadb_msg *hdr,
 
 	if (err < 0) {
 		x->km.state = XFRM_STATE_DEAD;
-		xfrm_state_put(x);
+		__xfrm_state_put(x);
 		goto out;
 	}
 
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index e12d0be5f976..c656cbaf35e8 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -220,14 +220,14 @@ static int __xfrm_state_delete(struct xfrm_state *x)
 		x->km.state = XFRM_STATE_DEAD;
 		spin_lock(&xfrm_state_lock);
 		list_del(&x->bydst);
-		atomic_dec(&x->refcnt);
+		__xfrm_state_put(x);
 		if (x->id.spi) {
 			list_del(&x->byspi);
-			atomic_dec(&x->refcnt);
+			__xfrm_state_put(x);
 		}
 		spin_unlock(&xfrm_state_lock);
 		if (del_timer(&x->timer))
-			atomic_dec(&x->refcnt);
+			__xfrm_state_put(x);
 
 		/* The number two in this test is the reference
 		 * mentioned in the comment below plus the reference
@@ -243,7 +243,7 @@ static int __xfrm_state_delete(struct xfrm_state *x)
 		 * The xfrm_state_alloc call gives a reference, and that
 		 * is what we are dropping here.
 		 */
-		atomic_dec(&x->refcnt);
+		__xfrm_state_put(x);
 		err = 0;
 	}
 
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index ac87a09ba83e..7de17559249a 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -345,7 +345,7 @@ static int xfrm_add_sa(struct sk_buff *skb, struct nlmsghdr *nlh, void **xfrma)
 
 	if (err < 0) {
 		x->km.state = XFRM_STATE_DEAD;
-		xfrm_state_put(x);
+		__xfrm_state_put(x);
 		goto out;
 	}
 

From f8d0e3f11593928ac3f968c378a44e80b04488c9 Mon Sep 17 00:00:00 2001
From: Jamal Hadi Salim <hadi@cyberus.ca>
Date: Thu, 23 Feb 2006 16:18:01 -0800
Subject: [PATCH 405/509] [NET] ethernet: Fix first packet goes out with MAC
 00:00:00:00:00:00

When you turn off ARP on a netdevice then the first packet always goes
out with a dstMAC of all zeroes. This is because the first packet is
used to resolve ARP entries. Even though the ARP entry may be resolved
(I tried by setting a static ARP entry for a host i was pinging from),
it gets overwritten by virtue of having the netdevice disabling ARP.

Subsequent packets go out fine with correct dstMAC address (which may
be why people have ignored reporting this issue).

To cut the story short:

the culprit code is in net/ethernet/eth.c::eth_header()

----
        /*
         *      Anyway, the loopback-device should never use this
function...
         */

        if (dev->flags & (IFF_LOOPBACK|IFF_NOARP))
        {
                memset(eth->h_dest, 0, dev->addr_len);
                return ETH_HLEN;
        }

	if(daddr)
        {
                memcpy(eth->h_dest,daddr,dev->addr_len);
                return ETH_HLEN;
        }

----

Note how the h_dest is being reset when device has IFF_NOARP.

As a note:
All devices including loopback pass a daddr. loopback in fact passes
a 0 all the time ;->
This means i can delete the check totaly or i can remove the IFF_NOARP

Alexey says:
--------------------
I think, it was me who did this crap. It was so long ago I do not remember
why it was made.

I remember some troubles with dummy device. It tried to resolve
addresses, apparently, without success and generated errors instead of
blackholing. I think the problem was eventually solved at neighbour
level.

After some thinking I suspect the deletion of this chunk could change
behaviour of some parts which do not use neighbour cache f.e. packet
socket.

I think safer approach would be to move this chunk after if (daddr).
And the possibility to remove this completely could be analyzed later.
--------------------

Patch updated with Alexey's safer suggestions.

Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Acked-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ethernet/eth.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/net/ethernet/eth.c b/net/ethernet/eth.c
index 9890fd97e538..c971f14712ec 100644
--- a/net/ethernet/eth.c
+++ b/net/ethernet/eth.c
@@ -95,6 +95,12 @@ int eth_header(struct sk_buff *skb, struct net_device *dev, unsigned short type,
 		saddr = dev->dev_addr;
 	memcpy(eth->h_source,saddr,dev->addr_len);
 
+	if(daddr)
+	{
+		memcpy(eth->h_dest,daddr,dev->addr_len);
+		return ETH_HLEN;
+	}
+	
 	/*
 	 *	Anyway, the loopback-device should never use this function... 
 	 */
@@ -105,12 +111,6 @@ int eth_header(struct sk_buff *skb, struct net_device *dev, unsigned short type,
 		return ETH_HLEN;
 	}
 	
-	if(daddr)
-	{
-		memcpy(eth->h_dest,daddr,dev->addr_len);
-		return ETH_HLEN;
-	}
-	
 	return -ETH_HLEN;
 }
 

From 4da3089f2b582b21e1374ccc6df722d4361eb915 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Thu, 23 Feb 2006 16:19:26 -0800
Subject: [PATCH 406/509] [IPSEC]: Use TOS when doing tunnel lookups

We should use the TOS because it's one of the routing keys.  It also
means that we update the correct routing cache entry when PMTU occurs.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/xfrm4_policy.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 45f7ae58f2c0..f285bbf296e2 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -35,6 +35,7 @@ __xfrm4_find_bundle(struct flowi *fl, struct xfrm_policy *policy)
 		if (xdst->u.rt.fl.oif == fl->oif &&	/*XXX*/
 		    xdst->u.rt.fl.fl4_dst == fl->fl4_dst &&
 	    	    xdst->u.rt.fl.fl4_src == fl->fl4_src &&
+	    	    xdst->u.rt.fl.fl4_tos == fl->fl4_tos &&
 		    xfrm_bundle_ok(xdst, fl, AF_INET)) {
 			dst_clone(dst);
 			break;
@@ -61,7 +62,8 @@ __xfrm4_bundle_create(struct xfrm_policy *policy, struct xfrm_state **xfrm, int
 		.nl_u = {
 			.ip4_u = {
 				.saddr = local,
-				.daddr = remote
+				.daddr = remote,
+				.tos = fl->fl4_tos
 			}
 		}
 	};
@@ -230,6 +232,7 @@ _decode_session4(struct sk_buff *skb, struct flowi *fl)
 	fl->proto = iph->protocol;
 	fl->fl4_dst = iph->daddr;
 	fl->fl4_src = iph->saddr;
+	fl->fl4_tos = iph->tos;
 }
 
 static inline int xfrm4_garbage_collect(void)

From 35eaa31e5d6b0653c11b5661572152295b45b7a7 Mon Sep 17 00:00:00 2001
From: Richard Lucassen <spamtrap@lucassen.org>
Date: Thu, 23 Feb 2006 16:23:51 -0800
Subject: [PATCH 407/509] [NET]: Increase default IFB device count.

The most usable number of ifb devices is 2. Change the default to 2.

Signed-off-by: Richard Lucassen <spamtrap@lucassen.org>
Signed-off-by: Jamal Hadi Salim <hadi@cyberus.ca>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/ifb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c
index 1b699259b4ec..31fb2d75dc44 100644
--- a/drivers/net/ifb.c
+++ b/drivers/net/ifb.c
@@ -57,7 +57,7 @@ struct ifb_private {
 	struct sk_buff_head     tq;
 };
 
-static int numifbs = 1;
+static int numifbs = 2;
 
 static void ri_tasklet(unsigned long dev);
 static int ifb_xmit(struct sk_buff *skb, struct net_device *dev);

From 337a7128dbe68ebe7627b6f954cb32d30d7b11c6 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Tue, 21 Feb 2006 17:22:55 +1100
Subject: [PATCH 408/509] [PATCH] powerpc: Only calculate htab_size in one
 place for kexec

For kexec we need to know the size of the MMU hash table.

Currently we calculate the size once in the htab code, and then twice more in
the kexec code, once using htab_hash_mask and once using ppc64_pft_size.
On some machines the ppc64_pft_size calculation is broken because
ppc64_pft_size is not set.

So we need to fix the second calculation, but better still we should just
calculate the size once and use it everywhere else.

Tested on Power5 LPAR, Power4 non-LPAR and Power3.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/machine_kexec_64.c | 10 +++-------
 arch/powerpc/mm/hash_utils_64.c        |  3 ++-
 include/asm-powerpc/mmu.h              |  1 +
 3 files changed, 6 insertions(+), 8 deletions(-)

diff --git a/arch/powerpc/kernel/machine_kexec_64.c b/arch/powerpc/kernel/machine_kexec_64.c
index d6431440c54f..ee166c586642 100644
--- a/arch/powerpc/kernel/machine_kexec_64.c
+++ b/arch/powerpc/kernel/machine_kexec_64.c
@@ -26,8 +26,6 @@
 #include <asm/prom.h>
 #include <asm/smp.h>
 
-#define HASH_GROUP_SIZE 0x80	/* size of each hash group, asm/mmu.h */
-
 int default_machine_kexec_prepare(struct kimage *image)
 {
 	int i;
@@ -61,7 +59,7 @@ int default_machine_kexec_prepare(struct kimage *image)
 	 */
 	if (htab_address) {
 		low = __pa(htab_address);
-		high = low + (htab_hash_mask + 1) * HASH_GROUP_SIZE;
+		high = low + htab_size_bytes;
 
 		for (i = 0; i < image->nr_segments; i++) {
 			begin = image->segment[i].mem;
@@ -294,7 +292,7 @@ void default_machine_kexec(struct kimage *image)
 }
 
 /* Values we need to export to the second kernel via the device tree. */
-static unsigned long htab_base, htab_size, kernel_end;
+static unsigned long htab_base, kernel_end;
 
 static struct property htab_base_prop = {
 	.name = "linux,htab-base",
@@ -305,7 +303,7 @@ static struct property htab_base_prop = {
 static struct property htab_size_prop = {
 	.name = "linux,htab-size",
 	.length = sizeof(unsigned long),
-	.value = (unsigned char *)&htab_size,
+	.value = (unsigned char *)&htab_size_bytes,
 };
 
 static struct property kernel_end_prop = {
@@ -331,8 +329,6 @@ static void __init export_htab_values(void)
 
 	htab_base = __pa(htab_address);
 	prom_add_property(node, &htab_base_prop);
-
-	htab_size = 1UL << ppc64_pft_size;
 	prom_add_property(node, &htab_size_prop);
 
  out:
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index 149351a84b94..b1f614c612dd 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -88,6 +88,7 @@ static unsigned long _SDR1;
 struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT];
 
 hpte_t *htab_address;
+unsigned long htab_size_bytes;
 unsigned long htab_hash_mask;
 int mmu_linear_psize = MMU_PAGE_4K;
 int mmu_virtual_psize = MMU_PAGE_4K;
@@ -399,7 +400,7 @@ void create_section_mapping(unsigned long start, unsigned long end)
 
 void __init htab_initialize(void)
 {
-	unsigned long table, htab_size_bytes;
+	unsigned long table;
 	unsigned long pteg_count;
 	unsigned long mode_rw;
 	unsigned long base = 0, size = 0;
diff --git a/include/asm-powerpc/mmu.h b/include/asm-powerpc/mmu.h
index d096d9e76ad7..b0b9a3f8cdc2 100644
--- a/include/asm-powerpc/mmu.h
+++ b/include/asm-powerpc/mmu.h
@@ -112,6 +112,7 @@ typedef struct {
 } hpte_t;
 
 extern hpte_t *htab_address;
+extern unsigned long htab_size_bytes;
 extern unsigned long htab_hash_mask;
 
 /*

From c57914a4f24322a8f7ef06a8e2fca5f0b2c98878 Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Tue, 21 Feb 2006 21:06:41 +0100
Subject: [PATCH 409/509] [PATCH] ppc: fix adb breakage in xmon

Fix up xmon compilation after the last change.
Remove lots of dead code, all the pmac and chrp support is in arch/powerpc

Signed-off-by: Olaf Hering <olh@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/ppc/xmon/adb.c   | 212 ------------------------------------------
 arch/ppc/xmon/start.c | 169 +--------------------------------
 arch/ppc/xmon/xmon.c  | 108 ---------------------
 3 files changed, 3 insertions(+), 486 deletions(-)
 delete mode 100644 arch/ppc/xmon/adb.c

diff --git a/arch/ppc/xmon/adb.c b/arch/ppc/xmon/adb.c
deleted file mode 100644
index e91384dcccac..000000000000
--- a/arch/ppc/xmon/adb.c
+++ /dev/null
@@ -1,212 +0,0 @@
-/*
- * Copyright (C) 1996 Paul Mackerras.
- */
-#include "nonstdio.h"
-#include "privinst.h"
-
-#define scanhex	xmon_scanhex
-#define skipbl	xmon_skipbl
-
-#define ADB_B		(*(volatile unsigned char *)0xf3016000)
-#define ADB_SR		(*(volatile unsigned char *)0xf3017400)
-#define ADB_ACR		(*(volatile unsigned char *)0xf3017600)
-#define ADB_IFR		(*(volatile unsigned char *)0xf3017a00)
-
-static inline void eieio(void) { asm volatile ("eieio" : :); }
-
-#define N_ADB_LOG	1000
-struct adb_log {
-    unsigned char b;
-    unsigned char ifr;
-    unsigned char acr;
-    unsigned int time;
-} adb_log[N_ADB_LOG];
-int n_adb_log;
-
-void
-init_adb_log(void)
-{
-    adb_log[0].b = ADB_B;
-    adb_log[0].ifr = ADB_IFR;
-    adb_log[0].acr = ADB_ACR;
-    adb_log[0].time = get_dec();
-    n_adb_log = 0;
-}
-
-void
-dump_adb_log(void)
-{
-    unsigned t, t0;
-    struct adb_log *ap;
-    int i;
-
-    ap = adb_log;
-    t0 = ap->time;
-    for (i = 0; i <= n_adb_log; ++i, ++ap) {
-	t = t0 - ap->time;
-	printf("b=%x ifr=%x acr=%x at %d.%.7d\n", ap->b, ap->ifr, ap->acr,
-	       t / 1000000000, (t % 1000000000) / 100);
-    }
-}
-
-void
-adb_chklog(void)
-{
-    struct adb_log *ap = &adb_log[n_adb_log + 1];
-
-    ap->b = ADB_B;
-    ap->ifr = ADB_IFR;
-    ap->acr = ADB_ACR;
-    if (ap->b != ap[-1].b || (ap->ifr & 4) != (ap[-1].ifr & 4)
-	|| ap->acr != ap[-1].acr) {
-	ap->time = get_dec();
-	++n_adb_log;
-    }
-}
-
-int
-adb_bitwait(int bmask, int bval, int fmask, int fval)
-{
-    int i;
-    struct adb_log *ap;
-
-    for (i = 10000; i > 0; --i) {
-	adb_chklog();
-	ap = &adb_log[n_adb_log];
-	if ((ap->b & bmask) == bval && (ap->ifr & fmask) == fval)
-	    return 0;
-    }
-    return -1;
-}
-
-int
-adb_wait(void)
-{
-    if (adb_bitwait(0, 0, 4, 4) < 0) {
-	printf("adb: ready wait timeout\n");
-	return -1;
-    }
-    return 0;
-}
-
-void
-adb_readin(void)
-{
-    int i, j;
-    unsigned char d[64];
-
-    if (ADB_B & 8) {
-	printf("ADB_B: %x\n", ADB_B);
-	return;
-    }
-    i = 0;
-    adb_wait();
-    j = ADB_SR;
-    eieio();
-    ADB_B &= ~0x20;
-    eieio();
-    for (;;) {
-	if (adb_wait() < 0)
-	    break;
-	d[i++] = ADB_SR;
-	eieio();
-	if (ADB_B & 8)
-	    break;
-	ADB_B ^= 0x10;
-	eieio();
-    }
-    ADB_B |= 0x30;
-    if (adb_wait() == 0)
-	j = ADB_SR;
-    for (j = 0; j < i; ++j)
-	printf("%.2x ", d[j]);
-    printf("\n");
-}
-
-int
-adb_write(unsigned char *d, int i)
-{
-    int j;
-    unsigned x;
-
-    if ((ADB_B & 8) == 0) {
-	printf("r: ");
-	adb_readin();
-    }
-    for (;;) {
-	ADB_ACR = 0x1c;
-	eieio();
-	ADB_SR = d[0];
-	eieio();
-	ADB_B &= ~0x20;
-	eieio();
-	if (ADB_B & 8)
-	    break;
-	ADB_ACR = 0xc;
-	eieio();
-	ADB_B |= 0x20;
-	eieio();
-	adb_readin();
-    }
-    adb_wait();
-    for (j = 1; j < i; ++j) {
-	ADB_SR = d[j];
-	eieio();
-	ADB_B ^= 0x10;
-	eieio();
-	if (adb_wait() < 0)
-	    break;
-    }
-    ADB_ACR = 0xc;
-    eieio();
-    x = ADB_SR;
-    eieio();
-    ADB_B |= 0x30;
-    return j;
-}
-
-void
-adbcmds(void)
-{
-    char cmd;
-    unsigned rtcu, rtcl, dec, pdec, x;
-    int i, j;
-    unsigned char d[64];
-
-    cmd = skipbl();
-    switch (cmd) {
-    case 't':
-	for (;;) {
-	    rtcl = get_rtcl();
-	    rtcu = get_rtcu();
-	    dec = get_dec();
-	    printf("rtc u=%u l=%u dec=%x (%d = %d.%.7d)\n",
-		   rtcu, rtcl, dec, pdec - dec, (pdec - dec) / 1000000000,
-		   ((pdec - dec) % 1000000000) / 100);
-	    pdec = dec;
-	    if (cmd == 'x')
-		break;
-	    while (xmon_read(stdin, &cmd, 1) != 1)
-		;
-	}
-	break;
-    case 'r':
-	init_adb_log();
-	while (adb_bitwait(8, 0, 0, 0) == 0)
-	    adb_readin();
-	break;
-    case 'w':
-	i = 0;
-	while (scanhex(&x))
-	    d[i++] = x;
-	init_adb_log();
-	j = adb_write(d, i);
-	printf("sent %d bytes\n", j);
-	while (adb_bitwait(8, 0, 0, 0) == 0)
-	    adb_readin();
-	break;
-    case 'l':
-	dump_adb_log();
-	break;
-    }
-}
diff --git a/arch/ppc/xmon/start.c b/arch/ppc/xmon/start.c
index 484f5bb1aa3e..ff86b2d814cb 100644
--- a/arch/ppc/xmon/start.c
+++ b/arch/ppc/xmon/start.c
@@ -6,16 +6,11 @@
 #include <asm/machdep.h>
 #include <asm/io.h>
 #include <asm/page.h>
-#include <linux/adb.h>
-#include <linux/pmu.h>
-#include <linux/cuda.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>
 #include <linux/sysrq.h>
 #include <linux/bitops.h>
 #include <asm/xmon.h>
-#include <asm/prom.h>
-#include <asm/bootx.h>
 #include <asm/machdep.h>
 #include <asm/errno.h>
 #include <asm/processor.h>
@@ -26,9 +21,7 @@ static volatile unsigned char *sccc, *sccd;
 unsigned int TXRDY, RXRDY, DLAB;
 static int xmon_expect(const char *str, unsigned int timeout);
 
-static int use_screen;
 static int via_modem;
-static int xmon_use_sccb;
 
 #define TB_SPEED	25000000
 
@@ -46,47 +39,6 @@ void buf_access(void)
 		sccd[3] &= ~DLAB;	/* reset DLAB */
 }
 
-extern int adb_init(void);
-
-#ifdef CONFIG_PPC_CHRP
-/*
- * This looks in the "ranges" property for the primary PCI host bridge
- * to find the physical address of the start of PCI/ISA I/O space.
- * It is basically a cut-down version of pci_process_bridge_OF_ranges.
- */
-static unsigned long chrp_find_phys_io_base(void)
-{
-	struct device_node *node;
-	unsigned int *ranges;
-	unsigned long base = CHRP_ISA_IO_BASE;
-	int rlen = 0;
-	int np;
-
-	node = find_devices("isa");
-	if (node != NULL) {
-		node = node->parent;
-		if (node == NULL || node->type == NULL
-		    || strcmp(node->type, "pci") != 0)
-			node = NULL;
-	}
-	if (node == NULL)
-		node = find_devices("pci");
-	if (node == NULL)
-		return base;
-
-	ranges = (unsigned int *) get_property(node, "ranges", &rlen);
-	np = prom_n_addr_cells(node) + 5;
-	while ((rlen -= np * sizeof(unsigned int)) >= 0) {
-		if ((ranges[0] >> 24) == 1 && ranges[2] == 0) {
-			/* I/O space starting at 0, grab the phys base */
-			base = ranges[np - 3];
-			break;
-		}
-		ranges += np;
-	}
-	return base;
-}
-#endif /* CONFIG_PPC_CHRP */
 
 #ifdef CONFIG_MAGIC_SYSRQ
 static void sysrq_handle_xmon(int key, struct pt_regs *regs,
@@ -109,22 +61,6 @@ xmon_map_scc(void)
 #ifdef CONFIG_PPC_MULTIPLATFORM
 	volatile unsigned char *base;
 
-#ifdef CONFIG_PPC_CHRP
-	base = (volatile unsigned char *) isa_io_base;
-	if (_machine == _MACH_chrp)
-		base = (volatile unsigned char *)
-			ioremap(chrp_find_phys_io_base(), 0x1000);
-
-	sccc = base + 0x3fd;
-	sccd = base + 0x3f8;
-	if (xmon_use_sccb) {
-		sccc -= 0x100;
-		sccd -= 0x100;
-	}
-	TXRDY = 0x20;
-	RXRDY = 1;
-	DLAB = 0x80;
-#endif /* CONFIG_PPC_CHRP */
 #elif defined(CONFIG_GEMINI)
 	/* should already be mapped by the kernel boot */
 	sccc = (volatile unsigned char *) 0xffeffb0d;
@@ -143,7 +79,7 @@ xmon_map_scc(void)
 	register_sysrq_key('x', &sysrq_xmon_op);
 }
 
-static int scc_initialized = 0;
+static int scc_initialized;
 
 void xmon_init_scc(void);
 
@@ -163,14 +99,6 @@ xmon_write(void *handle, void *ptr, int nb)
 			break;
 #endif
 
-#ifdef CONFIG_BOOTX_TEXT
-	if (use_screen) {
-		/* write it on the screen */
-		for (i = 0; i < nb; ++i)
-			btext_drawchar(*p++);
-		goto out;
-	}
-#endif
 	if (!scc_initialized)
 		xmon_init_scc();
 	ct = 0;
@@ -190,7 +118,6 @@ xmon_write(void *handle, void *ptr, int nb)
 		eieio();
 	}
 
- out:
 #ifdef CONFIG_SMP
 	if (!locked)
 		clear_bit(0, &xmon_write_lock);
@@ -199,65 +126,7 @@ xmon_write(void *handle, void *ptr, int nb)
 }
 
 int xmon_wants_key;
-int xmon_adb_keycode;
 
-#ifdef CONFIG_BOOTX_TEXT
-static int xmon_adb_shiftstate;
-
-static unsigned char xmon_keytab[128] =
-	"asdfhgzxcv\000bqwer"				/* 0x00 - 0x0f */
-	"yt123465=97-80]o"				/* 0x10 - 0x1f */
-	"u[ip\rlj'k;\\,/nm."				/* 0x20 - 0x2f */
-	"\t `\177\0\033\0\0\0\0\0\0\0\0\0\0"		/* 0x30 - 0x3f */
-	"\0.\0*\0+\0\0\0\0\0/\r\0-\0"			/* 0x40 - 0x4f */
-	"\0\0000123456789\0\0\0";			/* 0x50 - 0x5f */
-
-static unsigned char xmon_shift_keytab[128] =
-	"ASDFHGZXCV\000BQWER"				/* 0x00 - 0x0f */
-	"YT!@#$^%+(&_*)}O"				/* 0x10 - 0x1f */
-	"U{IP\rLJ\"K:|<?NM>"				/* 0x20 - 0x2f */
-	"\t ~\177\0\033\0\0\0\0\0\0\0\0\0\0"		/* 0x30 - 0x3f */
-	"\0.\0*\0+\0\0\0\0\0/\r\0-\0"			/* 0x40 - 0x4f */
-	"\0\0000123456789\0\0\0";			/* 0x50 - 0x5f */
-
-static int
-xmon_get_adb_key(void)
-{
-	int k, t, on;
-
-	xmon_wants_key = 1;
-	for (;;) {
-		xmon_adb_keycode = -1;
-		t = 0;
-		on = 0;
-		do {
-			if (--t < 0) {
-				on = 1 - on;
-				btext_drawchar(on? 0xdb: 0x20);
-				btext_drawchar('\b');
-				t = 200000;
-			}
-			do_poll_adb();
-		} while (xmon_adb_keycode == -1);
-		k = xmon_adb_keycode;
-		if (on)
-			btext_drawstring(" \b");
-
-		/* test for shift keys */
-		if ((k & 0x7f) == 0x38 || (k & 0x7f) == 0x7b) {
-			xmon_adb_shiftstate = (k & 0x80) == 0;
-			continue;
-		}
-		if (k >= 0x80)
-			continue;	/* ignore up transitions */
-		k = (xmon_adb_shiftstate? xmon_shift_keytab: xmon_keytab)[k];
-		if (k != 0)
-			break;
-	}
-	xmon_wants_key = 0;
-	return k;
-}
-#endif /* CONFIG_BOOTX_TEXT */
 
 int
 xmon_read(void *handle, void *ptr, int nb)
@@ -265,18 +134,11 @@ xmon_read(void *handle, void *ptr, int nb)
     char *p = ptr;
     int i;
 
-#ifdef CONFIG_BOOTX_TEXT
-    if (use_screen) {
-	for (i = 0; i < nb; ++i)
-	    *p++ = xmon_get_adb_key();
-	return i;
-    }
-#endif
     if (!scc_initialized)
 	xmon_init_scc();
     for (i = 0; i < nb; ++i) {
 	while ((*sccc & RXRDY) == 0)
-	    do_poll_adb();
+	    ;
 	buf_access();
 	*p++ = *sccd;
     }
@@ -287,7 +149,7 @@ int
 xmon_read_poll(void)
 {
 	if ((*sccc & RXRDY) == 0) {
-		do_poll_adb();
+		;
 		return -1;
 	}
 	buf_access();
@@ -297,15 +159,6 @@ xmon_read_poll(void)
 void
 xmon_init_scc(void)
 {
-	if ( _machine == _MACH_chrp )
-	{
-		sccd[3] = 0x83; eieio();	/* LCR = 8N1 + DLAB */
-		sccd[0] = 12; eieio();		/* DLL = 9600 baud */
-		sccd[1] = 0; eieio();
-		sccd[2] = 0; eieio();		/* FCR = 0 */
-		sccd[3] = 3; eieio();		/* LCR = 8N1 */
-		sccd[1] = 0; eieio();		/* IER = 0 */
-	}
 	scc_initialized = 1;
 	if (via_modem) {
 		for (;;) {
@@ -321,22 +174,6 @@ xmon_init_scc(void)
 	}
 }
 
-#if 0
-extern int (*prom_entry)(void *);
-
-int
-xmon_exit(void)
-{
-    struct prom_args {
-	char *service;
-    } args;
-
-    for (;;) {
-	args.service = "exit";
-	(*prom_entry)(&args);
-    }
-}
-#endif
 
 void *xmon_stdin;
 void *xmon_stdout;
diff --git a/arch/ppc/xmon/xmon.c b/arch/ppc/xmon/xmon.c
index bdaf6597b4c2..06fa44b5c647 100644
--- a/arch/ppc/xmon/xmon.c
+++ b/arch/ppc/xmon/xmon.c
@@ -12,8 +12,6 @@
 #include <linux/kallsyms.h>
 #include <asm/ptrace.h>
 #include <asm/string.h>
-#include <asm/prom.h>
-#include <asm/bootx.h>
 #include <asm/machdep.h>
 #include <asm/xmon.h>
 #include "nonstdio.h"
@@ -101,9 +99,6 @@ void cacheflush(void);
 static void cpu_cmd(void);
 #endif /* CONFIG_SMP */
 static void csum(void);
-#ifdef CONFIG_BOOTX_TEXT
-static void vidcmds(void);
-#endif
 static void bootcmds(void);
 static void proccall(void);
 static void printtime(void);
@@ -522,11 +517,6 @@ cmds(struct pt_regs *excp)
 			cpu_cmd();
 			break;
 #endif /* CONFIG_SMP */
-#ifdef CONFIG_BOOTX_TEXT
-		case 'v':
-			vidcmds();
-			break;
-#endif
 		case 'z':
 			bootcmds();
 			break;
@@ -618,43 +608,6 @@ static void cpu_cmd(void)
 }
 #endif /* CONFIG_SMP */
 
-#ifdef CONFIG_BOOTX_TEXT
-extern boot_infos_t disp_bi;
-
-static void vidcmds(void)
-{
-	int c = inchar();
-	unsigned int val, w;
-	extern int boot_text_mapped;
-
-	if (!boot_text_mapped)
-		return;
-	if (c != '\n' && scanhex(&val)) {
-		switch (c) {
-		case 'd':
-			w = disp_bi.dispDeviceRowBytes
-				/ (disp_bi.dispDeviceDepth >> 3);
-			disp_bi.dispDeviceDepth = val;
-			disp_bi.dispDeviceRowBytes = w * (val >> 3);
-			return;
-		case 'p':
-			disp_bi.dispDeviceRowBytes = val;
-			return;
-		case 'w':
-			disp_bi.dispDeviceRect[2] = val;
-			return;
-		case 'h':
-			disp_bi.dispDeviceRect[3] = val;
-			return;
-		}
-	}
-	printf("W = %d (0x%x) H = %d (0x%x) D = %d (0x%x) P = %d (0x%x)\n",
-	       disp_bi.dispDeviceRect[2], disp_bi.dispDeviceRect[2],
-	       disp_bi.dispDeviceRect[3], disp_bi.dispDeviceRect[3],
-	       disp_bi.dispDeviceDepth, disp_bi.dispDeviceDepth,
-	       disp_bi.dispDeviceRowBytes, disp_bi.dispDeviceRowBytes);
-}
-#endif /* CONFIG_BOOTX_TEXT */
 
 static unsigned short fcstab[256] = {
 	0x0000, 0x1189, 0x2312, 0x329b, 0x4624, 0x57ad, 0x6536, 0x74bf,
@@ -1020,7 +973,6 @@ dump_hash_table(void)
 }
 #else
 
-#ifndef CONFIG_PPC64BRIDGE
 static void
 dump_hash_table_seg(unsigned seg, unsigned start, unsigned end)
 {
@@ -1079,66 +1031,6 @@ dump_hash_table_seg(unsigned seg, unsigned start, unsigned end)
 		printf(" ... %x\n", last_va);
 }
 
-#else /* CONFIG_PPC64BRIDGE */
-static void
-dump_hash_table_seg(unsigned seg, unsigned start, unsigned end)
-{
-	extern void *Hash;
-	extern unsigned long Hash_size;
-	unsigned *htab = Hash;
-	unsigned hsize = Hash_size;
-	unsigned v, hmask, va, last_va;
-	int found, last_found, i;
-	unsigned *hg, w1, last_w2, last_va0;
-
-	last_found = 0;
-	hmask = hsize / 128 - 1;
-	va = start;
-	start = (start >> 12) & 0xffff;
-	end = (end >> 12) & 0xffff;
-	for (v = start; v < end; ++v) {
-		found = 0;
-		hg = htab + (((v ^ seg) & hmask) * 32);
-		w1 = 1 | (seg << 12) | ((v & 0xf800) >> 4);
-		for (i = 0; i < 8; ++i, hg += 4) {
-			if (hg[1] == w1) {
-				found = 1;
-				break;
-			}
-		}
-		if (!found) {
-			w1 ^= 2;
-			hg = htab + ((~(v ^ seg) & hmask) * 32);
-			for (i = 0; i < 8; ++i, hg += 4) {
-				if (hg[1] == w1) {
-					found = 1;
-					break;
-				}
-			}
-		}
-		if (!(last_found && found && (hg[3] & ~0x180) == last_w2 + 4096)) {
-			if (last_found) {
-				if (last_va != last_va0)
-					printf(" ... %x", last_va);
-				printf("\n");
-			}
-			if (found) {
-				printf("%x to %x", va, hg[3]);
-				last_va0 = va;
-			}
-			last_found = found;
-		}
-		if (found) {
-			last_w2 = hg[3] & ~0x180;
-			last_va = va;
-		}
-		va += 4096;
-	}
-	if (last_found)
-		printf(" ... %x\n", last_va);
-}
-#endif /* CONFIG_PPC64BRIDGE */
-
 static unsigned hash_ctx;
 static unsigned hash_start;
 static unsigned hash_end;

From 01aaed9d43d5fff1ddb4c8de859f87ed7ee3608a Mon Sep 17 00:00:00 2001
From: Haren Myneni <haren@us.ibm.com>
Date: Tue, 7 Feb 2006 15:47:03 -0800
Subject: [PATCH 410/509] [PATCH] powerpc: Trivial fix to set the proper
 timeout value for kdump

The panic CPU is waiting forever due to some large timeout value if some
CPU is not responding to an IPI.
This patch fixes the problem - the maximum waiting period will be
10 seconds and then the kdump boot will go ahead.

Signed-off-by: Haren Myneni <haren@us.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/crash.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c
index 8c21d378f5d2..778f22fd85d2 100644
--- a/arch/powerpc/kernel/crash.c
+++ b/arch/powerpc/kernel/crash.c
@@ -134,8 +134,10 @@ static void crash_kexec_prepare_cpus(void)
 	 * the crash CPU will send an IPI and wait for other CPUs to
 	 * respond. If not, proceed the kexec boot even though we failed to
 	 * capture other CPU states.
+	 * Delay of at least 10 seconds.
 	 */
-	msecs = 1000000;
+	printk(KERN_ALERT "Sending IPI to other cpus...\n");
+	msecs = 10000;
 	while ((atomic_read(&waiting_for_crash_ipi) > 0) && (--msecs > 0)) {
 		barrier();
 		mdelay(1);

From f1434a4854407a262d194411245eb9ee66221f90 Mon Sep 17 00:00:00 2001
From: Alan Curry <pacman@TheWorld.com>
Date: Wed, 22 Feb 2006 01:42:37 -0500
Subject: [PATCH 411/509] [PATCH] powerpc: fix altivec_unavailable_exception
 Oopses

altivec_unavailable_exception is called without setting r3... it looks like
the r3 that actually gets passed in as struct pt_regs *regs is the
undisturbed value of r3 at the time the altivec instruction was encountered.
The user actually gets to choose the pt_regs printed in the Oops!

This fixes the oops by passing the correct pt_regs pointer to
altivec_unavailable_exception.

Signed-off-by: Alan Curry <pacman@TheWorld.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/head_32.S | 1 +
 arch/ppc/kernel/head.S        | 1 +
 2 files changed, 2 insertions(+)

diff --git a/arch/powerpc/kernel/head_32.S b/arch/powerpc/kernel/head_32.S
index 03b25f9359f8..a0579e859b21 100644
--- a/arch/powerpc/kernel/head_32.S
+++ b/arch/powerpc/kernel/head_32.S
@@ -714,6 +714,7 @@ AltiVecUnavailable:
 #ifdef CONFIG_ALTIVEC
 	bne	load_up_altivec		/* if from user, just load it up */
 #endif /* CONFIG_ALTIVEC */
+	addi	r3,r1,STACK_FRAME_OVERHEAD
 	EXC_XFER_EE_LITE(0xf20, altivec_unavailable_exception)
 
 PerformanceMonitor:
diff --git a/arch/ppc/kernel/head.S b/arch/ppc/kernel/head.S
index c5a890dca9cf..53ea845fb911 100644
--- a/arch/ppc/kernel/head.S
+++ b/arch/ppc/kernel/head.S
@@ -751,6 +751,7 @@ AltiVecUnavailable:
 #ifdef CONFIG_ALTIVEC
 	bne	load_up_altivec		/* if from user, just load it up */
 #endif /* CONFIG_ALTIVEC */
+	addi	r3,r1,STACK_FRAME_OVERHEAD
 	EXC_XFER_EE_LITE(0xf20, altivec_unavailable_exception)
 
 #ifdef CONFIG_PPC64BRIDGE

From 1775dbbcd02cab0c41329dd2cec5b69c7fafd13f Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Wed, 22 Feb 2006 09:46:02 -0600
Subject: [PATCH 412/509] [PATCH] powerpc: Enable coherency for all pages on
 83xx to fix PCI data corruption

On the 83xx platform to ensure the PCI inbound memory is handled properly we
have to turn on coherency for all pages in the MMU.  Otherwise we see
corruption if inbound "prefetching/streaming" is enabled on the PCI controller.

Signed-off-by: Randy Vinson <rvinson@mvista.com>
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/cputable.h | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h
index 64210549f56b..90d005bb4d1c 100644
--- a/include/asm-powerpc/cputable.h
+++ b/include/asm-powerpc/cputable.h
@@ -159,9 +159,11 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 #endif
 
 /* We need to mark all pages as being coherent if we're SMP or we
- * have a 74[45]x and an MPC107 host bridge.
+ * have a 74[45]x and an MPC107 host bridge. Also 83xx requires
+ * it for PCI "streaming/prefetch" to work properly.
  */
-#if defined(CONFIG_SMP) || defined(CONFIG_MPC10X_BRIDGE)
+#if defined(CONFIG_SMP) || defined(CONFIG_MPC10X_BRIDGE) \
+	|| defined(CONFIG_PPC_83xx)
 #define CPU_FTR_COMMON                  CPU_FTR_NEED_COHERENT
 #else
 #define CPU_FTR_COMMON                  0
@@ -277,7 +279,8 @@ enum {
 	CPU_FTRS_G2_LE = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_MAYBE_CAN_DOZE |
 	    CPU_FTR_USE_TB | CPU_FTR_MAYBE_CAN_NAP | CPU_FTR_HAS_HIGH_BATS,
 	CPU_FTRS_E300 = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_MAYBE_CAN_DOZE |
-	    CPU_FTR_USE_TB | CPU_FTR_MAYBE_CAN_NAP | CPU_FTR_HAS_HIGH_BATS,
+	    CPU_FTR_USE_TB | CPU_FTR_MAYBE_CAN_NAP | CPU_FTR_HAS_HIGH_BATS |
+	    CPU_FTR_COMMON,
 	CPU_FTRS_CLASSIC32 = CPU_FTR_COMMON | CPU_FTR_SPLIT_ID_CACHE |
 	    CPU_FTR_USE_TB | CPU_FTR_HPTE_TABLE,
 	CPU_FTRS_POWER3_32 = CPU_FTR_COMMON | CPU_FTR_SPLIT_ID_CACHE |

From 611ae59c62e688792cd1e6a68b9dc0f68d0e0dff Mon Sep 17 00:00:00 2001
From: Kelly Daly <kelly@au1.ibm.com>
Date: Thu, 23 Feb 2006 14:32:59 +1100
Subject: [PATCH 413/509] [PATCH] powerpc: disable OProfile for iSeries

Disable OProfile in Kconfig for iSeries to prevent hangs.  OProfile
was not originally intended to work with legacy iSeries.

Signed-off-by: Kelly Daly <kelly@au.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/oprofile/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/powerpc/oprofile/Kconfig b/arch/powerpc/oprofile/Kconfig
index eb2dece76a54..d03c0e5ca870 100644
--- a/arch/powerpc/oprofile/Kconfig
+++ b/arch/powerpc/oprofile/Kconfig
@@ -1,4 +1,5 @@
 config PROFILING
+	depends on !PPC_ISERIES
 	bool "Profiling support (EXPERIMENTAL)"
 	help
 	  Say Y here to enable the extended profiling support mechanisms used

From 47f78a49206b7f9b0d283ba46a2a5a6ee1796472 Mon Sep 17 00:00:00 2001
From: R Sharada <sharada@in.ibm.com>
Date: Wed, 22 Feb 2006 21:43:08 +0530
Subject: [PATCH 414/509] [PATCH] powerpc64: fix spinlock recursion in
 native_hpte_clear

native_hpte_clear has a spinlock recursion problem with the native_tlbie_lock
being called twice, once in native_hpte_clear() and once within tlbie().
Fix the problem by changing the call to tlbie() in native_hpte_clear() to
__tlbie(). It still supports only 4k pages for now.

Signed-off-by: R Sharada <sharada@in.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/mm/hash_native_64.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/mm/hash_native_64.c b/arch/powerpc/mm/hash_native_64.c
index d96bcfe4c6f6..33654d1b1b43 100644
--- a/arch/powerpc/mm/hash_native_64.c
+++ b/arch/powerpc/mm/hash_native_64.c
@@ -403,12 +403,17 @@ static void native_hpte_clear(void)
 		 */
 		hpte_v = hptep->v;
 
+		/*
+		 * Call __tlbie() here rather than tlbie() since we
+		 * already hold the native_tlbie_lock.
+		 */
 		if (hpte_v & HPTE_V_VALID) {
 			hptep->v = 0;
-			tlbie(slot2va(hpte_v, slot), MMU_PAGE_4K, 0);
+			__tlbie(slot2va(hpte_v, slot), MMU_PAGE_4K);
 		}
 	}
 
+	asm volatile("eieio; tlbsync; ptesync":::"memory");
 	spin_unlock(&native_tlbie_lock);
 	local_irq_restore(flags);
 }

From cb2c9b2741346eb23b177187a51ff5abf08295bd Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Mon, 13 Feb 2006 14:48:35 +1100
Subject: [PATCH 415/509] [PATCH] powerpc: Fix runlatch performance issues

The runlatch SPR can take a lot of time to write. My original runlatch
code would set it on every exception entry even though most of the time
this was not required. It would also continually set it in the idle
loop, which is an issue on an SMT capable processor.

Now we cache the runlatch value in a threadinfo bit, and only check for
it in decrementer and hardware interrupt exceptions as well as the idle
loop. Boot on POWER3, POWER5 and iseries, and compile tested on pmac32.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/head_64.S          | 12 ++--------
 arch/powerpc/kernel/process.c          | 32 ++++++++++++++++++++++++++
 arch/powerpc/platforms/iseries/setup.c |  1 +
 include/asm-powerpc/reg.h              | 31 ++-----------------------
 include/asm-powerpc/thread_info.h      |  4 ++--
 5 files changed, 39 insertions(+), 41 deletions(-)

diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 2b03a09fe5e9..bb845eed0e98 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -319,7 +319,6 @@ exception_marker:
 label##_pSeries:					\
 	HMT_MEDIUM;					\
 	mtspr	SPRN_SPRG1,r13;		/* save r13 */	\
-	RUNLATCH_ON(r13);				\
 	EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, label##_common)
 
 #define STD_EXCEPTION_ISERIES(n, label, area)		\
@@ -327,7 +326,6 @@ label##_pSeries:					\
 label##_iSeries:					\
 	HMT_MEDIUM;					\
 	mtspr	SPRN_SPRG1,r13;		/* save r13 */	\
-	RUNLATCH_ON(r13);				\
 	EXCEPTION_PROLOG_ISERIES_1(area);		\
 	EXCEPTION_PROLOG_ISERIES_2;			\
 	b	label##_common
@@ -337,7 +335,6 @@ label##_iSeries:					\
 label##_iSeries:							\
 	HMT_MEDIUM;							\
 	mtspr	SPRN_SPRG1,r13;		/* save r13 */			\
-	RUNLATCH_ON(r13);						\
 	EXCEPTION_PROLOG_ISERIES_1(PACA_EXGEN);				\
 	lbz	r10,PACAPROCENABLED(r13);				\
 	cmpwi	0,r10,0;						\
@@ -390,6 +387,7 @@ label##_common:						\
 label##_common:						\
 	EXCEPTION_PROLOG_COMMON(trap, PACA_EXGEN);	\
 	DISABLE_INTS;					\
+	bl	.ppc64_runlatch_on;			\
 	addi	r3,r1,STACK_FRAME_OVERHEAD;		\
 	bl	hdlr;					\
 	b	.ret_from_except_lite
@@ -407,7 +405,6 @@ __start_interrupts:
 _machine_check_pSeries:
 	HMT_MEDIUM
 	mtspr	SPRN_SPRG1,r13		/* save r13 */
-	RUNLATCH_ON(r13)
 	EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common)
 
 	. = 0x300
@@ -434,7 +431,6 @@ END_FTR_SECTION_IFCLR(CPU_FTR_SLB)
 data_access_slb_pSeries:
 	HMT_MEDIUM
 	mtspr	SPRN_SPRG1,r13
-	RUNLATCH_ON(r13)
 	mfspr	r13,SPRN_SPRG3		/* get paca address into r13 */
 	std	r3,PACA_EXSLB+EX_R3(r13)
 	mfspr	r3,SPRN_DAR
@@ -460,7 +456,6 @@ data_access_slb_pSeries:
 instruction_access_slb_pSeries:
 	HMT_MEDIUM
 	mtspr	SPRN_SPRG1,r13
-	RUNLATCH_ON(r13)
 	mfspr	r13,SPRN_SPRG3		/* get paca address into r13 */
 	std	r3,PACA_EXSLB+EX_R3(r13)
 	mfspr	r3,SPRN_SRR0		/* SRR0 is faulting address */
@@ -491,7 +486,6 @@ instruction_access_slb_pSeries:
 	.globl	system_call_pSeries
 system_call_pSeries:
 	HMT_MEDIUM
-	RUNLATCH_ON(r9)
 	mr	r9,r13
 	mfmsr	r10
 	mfspr	r13,SPRN_SPRG3
@@ -575,7 +569,6 @@ slb_miss_user_pseries:
 system_reset_fwnmi:
 	HMT_MEDIUM
 	mtspr	SPRN_SPRG1,r13		/* save r13 */
-	RUNLATCH_ON(r13)
 	EXCEPTION_PROLOG_PSERIES(PACA_EXGEN, system_reset_common)
 
 	.globl machine_check_fwnmi
@@ -583,7 +576,6 @@ system_reset_fwnmi:
 machine_check_fwnmi:
 	HMT_MEDIUM
 	mtspr	SPRN_SPRG1,r13		/* save r13 */
-	RUNLATCH_ON(r13)
 	EXCEPTION_PROLOG_PSERIES(PACA_EXMC, machine_check_common)
 
 #ifdef CONFIG_PPC_ISERIES
@@ -894,7 +886,6 @@ unrecov_fer:
 	.align	7
 	.globl data_access_common
 data_access_common:
-	RUNLATCH_ON(r10)		/* It wont fit in the 0x300 handler */
 	mfspr	r10,SPRN_DAR
 	std	r10,PACA_EXGEN+EX_DAR(r13)
 	mfspr	r10,SPRN_DSISR
@@ -1042,6 +1033,7 @@ hardware_interrupt_common:
 	EXCEPTION_PROLOG_COMMON(0x500, PACA_EXGEN)
 hardware_interrupt_entry:
 	DISABLE_INTS
+	bl	.ppc64_runlatch_on
 	addi	r3,r1,STACK_FRAME_OVERHEAD
 	bl	.do_IRQ
 	b	.ret_from_except_lite
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index 57703994a063..c225cf154bfe 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -888,3 +888,35 @@ void dump_stack(void)
 	show_stack(current, NULL);
 }
 EXPORT_SYMBOL(dump_stack);
+
+#ifdef CONFIG_PPC64
+void ppc64_runlatch_on(void)
+{
+	unsigned long ctrl;
+
+	if (cpu_has_feature(CPU_FTR_CTRL) && !test_thread_flag(TIF_RUNLATCH)) {
+		HMT_medium();
+
+		ctrl = mfspr(SPRN_CTRLF);
+		ctrl |= CTRL_RUNLATCH;
+		mtspr(SPRN_CTRLT, ctrl);
+
+		set_thread_flag(TIF_RUNLATCH);
+	}
+}
+
+void ppc64_runlatch_off(void)
+{
+	unsigned long ctrl;
+
+	if (cpu_has_feature(CPU_FTR_CTRL) && test_thread_flag(TIF_RUNLATCH)) {
+		HMT_medium();
+
+		clear_thread_flag(TIF_RUNLATCH);
+
+		ctrl = mfspr(SPRN_CTRLF);
+		ctrl &= ~CTRL_RUNLATCH;
+		mtspr(SPRN_CTRLT, ctrl);
+	}
+}
+#endif
diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c
index 3f8790146b00..3ecc4a652d82 100644
--- a/arch/powerpc/platforms/iseries/setup.c
+++ b/arch/powerpc/platforms/iseries/setup.c
@@ -648,6 +648,7 @@ static void yield_shared_processor(void)
 	 * here and let the timer_interrupt code sort out the actual time.
 	 */
 	get_lppaca()->int_dword.fields.decr_int = 1;
+	ppc64_runlatch_on();
 	process_iSeries_events();
 }
 
diff --git a/include/asm-powerpc/reg.h b/include/asm-powerpc/reg.h
index 12ecc9b9f285..72bfe3af0460 100644
--- a/include/asm-powerpc/reg.h
+++ b/include/asm-powerpc/reg.h
@@ -615,27 +615,9 @@
 #define proc_trap()	asm volatile("trap")
 
 #ifdef CONFIG_PPC64
-static inline void ppc64_runlatch_on(void)
-{
-	unsigned long ctrl;
 
-	if (cpu_has_feature(CPU_FTR_CTRL)) {
-		ctrl = mfspr(SPRN_CTRLF);
-		ctrl |= CTRL_RUNLATCH;
-		mtspr(SPRN_CTRLT, ctrl);
-	}
-}
-
-static inline void ppc64_runlatch_off(void)
-{
-	unsigned long ctrl;
-
-	if (cpu_has_feature(CPU_FTR_CTRL)) {
-		ctrl = mfspr(SPRN_CTRLF);
-		ctrl &= ~CTRL_RUNLATCH;
-		mtspr(SPRN_CTRLT, ctrl);
-	}
-}
+extern void ppc64_runlatch_on(void);
+extern void ppc64_runlatch_off(void);
 
 extern unsigned long scom970_read(unsigned int address);
 extern void scom970_write(unsigned int address, unsigned long value);
@@ -645,15 +627,6 @@ extern void scom970_write(unsigned int address, unsigned long value);
 #define __get_SP()	({unsigned long sp; \
 			asm volatile("mr %0,1": "=r" (sp)); sp;})
 
-#else /* __ASSEMBLY__ */
-
-#define RUNLATCH_ON(REG)			\
-BEGIN_FTR_SECTION				\
-	mfspr	(REG),SPRN_CTRLF;		\
-	ori	(REG),(REG),CTRL_RUNLATCH;	\
-	mtspr	SPRN_CTRLT,(REG);		\
-END_FTR_SECTION_IFSET(CPU_FTR_CTRL)
-
 #endif /* __ASSEMBLY__ */
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_REG_H */
diff --git a/include/asm-powerpc/thread_info.h b/include/asm-powerpc/thread_info.h
index c044ec16a879..237fc2b72974 100644
--- a/include/asm-powerpc/thread_info.h
+++ b/include/asm-powerpc/thread_info.h
@@ -113,7 +113,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_POLLING_NRFLAG	4	/* true if poll_idle() is polling
 					   TIF_NEED_RESCHED */
 #define TIF_32BIT		5	/* 32 bit binary */
-/* #define SPARE		6 */
+#define TIF_RUNLATCH		6	/* Is the runlatch enabled? */
 #define TIF_ABI_PENDING		7	/* 32/64 bit switch needed */
 #define TIF_SYSCALL_AUDIT	8	/* syscall auditing active */
 #define TIF_SINGLESTEP		9	/* singlestepping active */
@@ -131,7 +131,7 @@ static inline struct thread_info *current_thread_info(void)
 #define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
 #define _TIF_32BIT		(1<<TIF_32BIT)
-/* #define _SPARE		(1<<SPARE) */
+#define _TIF_RUNLATCH		(1<<TIF_RUNLATCH)
 #define _TIF_ABI_PENDING	(1<<TIF_ABI_PENDING)
 #define _TIF_SYSCALL_AUDIT	(1<<TIF_SYSCALL_AUDIT)
 #define _TIF_SINGLESTEP		(1<<TIF_SINGLESTEP)

From f1870f772c5e884862b4dd8f1ec2147247dda0ef Mon Sep 17 00:00:00 2001
From: Anton Blanchard <anton@samba.org>
Date: Mon, 13 Feb 2006 18:11:13 +1100
Subject: [PATCH 416/509] [PATCH] powerpc64: remove broken/bitrotted HMT
 support

HMT support is currently broken and needs to be reworked to play nicely
with the SMT scheduler. Remove the bit rotten bits for the time being.

I also updated an incorrect comment, we enter __secondary_hold with the
physical cpu id in r3.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/head_64.S          | 97 ++------------------------
 arch/powerpc/kernel/prom_init.c        | 38 ----------
 arch/powerpc/platforms/pseries/Kconfig |  7 --
 3 files changed, 4 insertions(+), 138 deletions(-)

diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index bb845eed0e98..11f2cd5af7dc 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -139,7 +139,7 @@ _GLOBAL(__secondary_hold)
 	ori	r24,r24,MSR_RI
 	mtmsrd	r24			/* RI on */
 
-	/* Grab our linux cpu number */
+	/* Grab our physical cpu number */
 	mr	r24,r3
 
 	/* Tell the master cpu we're here */
@@ -153,11 +153,7 @@ _GLOBAL(__secondary_hold)
 	cmpdi	0,r4,1
 	bne	100b
 
-#ifdef CONFIG_HMT
-	SET_REG_IMMEDIATE(r4, .hmt_init)
-	mtctr	r4
-	bctr
-#elif defined(CONFIG_SMP) || defined(CONFIG_KEXEC)
+#if defined(CONFIG_SMP) || defined(CONFIG_KEXEC)
 	LOAD_REG_IMMEDIATE(r4, .pSeries_secondary_smp_init)
 	mtctr	r4
 	mr	r3,r24
@@ -1808,22 +1804,6 @@ _STATIC(start_here_multiplatform)
 	ori	r6,r6,MSR_RI
 	mtmsrd	r6			/* RI on */
 
-#ifdef CONFIG_HMT
-	/* Start up the second thread on cpu 0 */
-	mfspr	r3,SPRN_PVR
-	srwi	r3,r3,16
-	cmpwi	r3,0x34			/* Pulsar  */
-	beq	90f
-	cmpwi	r3,0x36			/* Icestar */
-	beq	90f
-	cmpwi	r3,0x37			/* SStar   */
-	beq	90f
-	b	91f			/* HMT not supported */
-90:	li	r3,0
-	bl	.hmt_start_secondary
-91:
-#endif
-
 	/* The following gets the stack and TOC set up with the regs */
 	/* pointing to the real addr of the kernel stack.  This is   */
 	/* all done to support the C function call below which sets  */
@@ -1937,77 +1917,8 @@ _STATIC(start_here_common)
 
 	bl .start_kernel
 
-_GLOBAL(hmt_init)
-#ifdef CONFIG_HMT
-	LOAD_REG_IMMEDIATE(r5, hmt_thread_data)
-	mfspr	r7,SPRN_PVR
-	srwi	r7,r7,16
-	cmpwi	r7,0x34			/* Pulsar  */
-	beq	90f
-	cmpwi	r7,0x36			/* Icestar */
-	beq	91f
-	cmpwi	r7,0x37			/* SStar   */
-	beq	91f
-	b	101f
-90:	mfspr	r6,SPRN_PIR
-	andi.	r6,r6,0x1f
-	b	92f
-91:	mfspr	r6,SPRN_PIR
-	andi.	r6,r6,0x3ff
-92:	sldi	r4,r24,3
-	stwx	r6,r5,r4
-	bl	.hmt_start_secondary
-	b	101f
-
-__hmt_secondary_hold:
-	LOAD_REG_IMMEDIATE(r5, hmt_thread_data)
-	clrldi	r5,r5,4
-	li	r7,0
-	mfspr	r6,SPRN_PIR
-	mfspr	r8,SPRN_PVR
-	srwi	r8,r8,16
-	cmpwi	r8,0x34
-	bne	93f
-	andi.	r6,r6,0x1f
-	b	103f
-93:	andi.	r6,r6,0x3f
-
-103:	lwzx	r8,r5,r7
-	cmpw	r8,r6
-	beq	104f
-	addi	r7,r7,8
-	b	103b
-
-104:	addi	r7,r7,4
-	lwzx	r9,r5,r7
-	mr	r24,r9
-101:
-#endif
-	mr	r3,r24
-	b	.pSeries_secondary_smp_init
-
-#ifdef CONFIG_HMT
-_GLOBAL(hmt_start_secondary)
-	LOAD_REG_IMMEDIATE(r4,__hmt_secondary_hold)
-	clrldi	r4,r4,4
-	mtspr	SPRN_NIADORM, r4
-	mfspr	r4, SPRN_MSRDORM
-	li	r5, -65
-	and	r4, r4, r5
-	mtspr	SPRN_MSRDORM, r4
-	lis	r4,0xffef
-	ori	r4,r4,0x7403
-	mtspr	SPRN_TSC, r4
-	li	r4,0x1f4
-	mtspr	SPRN_TST, r4
-	mfspr	r4, SPRN_HID0
-	ori	r4, r4, 0x1
-	mtspr	SPRN_HID0, r4
-	mfspr	r4, SPRN_CTRLF
-	oris	r4, r4, 0x40
-	mtspr	SPRN_CTRLT, r4
-	blr
-#endif
+	/* Not reached */
+	BUG_OPCODE
 
 /*
  * We put a few things here that have to be page-aligned.
diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index ec7153f4d47c..d34fe537400e 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -205,14 +205,6 @@ static cell_t __initdata regbuf[1024];
 
 #define MAX_CPU_THREADS 2
 
-/* TO GO */
-#ifdef CONFIG_HMT
-struct {
-	unsigned int pir;
-	unsigned int threadid;
-} hmt_thread_data[NR_CPUS];
-#endif /* CONFIG_HMT */
-
 /*
  * Error results ... some OF calls will return "-1" on error, some
  * will return 0, some will return either. To simplify, here are
@@ -1319,10 +1311,6 @@ static void __init prom_hold_cpus(void)
 	 */
 	*spinloop = 0;
 
-#ifdef CONFIG_HMT
-	for (i = 0; i < NR_CPUS; i++)
-		RELOC(hmt_thread_data)[i].pir = 0xdeadbeef;
-#endif
 	/* look for cpus */
 	for (node = 0; prom_next_node(&node); ) {
 		type[0] = 0;
@@ -1389,32 +1377,6 @@ static void __init prom_hold_cpus(void)
 		/* Reserve cpu #s for secondary threads.   They start later. */
 		cpuid += cpu_threads;
 	}
-#ifdef CONFIG_HMT
-	/* Only enable HMT on processors that provide support. */
-	if (__is_processor(PV_PULSAR) || 
-	    __is_processor(PV_ICESTAR) ||
-	    __is_processor(PV_SSTAR)) {
-		prom_printf("    starting secondary threads\n");
-
-		for (i = 0; i < NR_CPUS; i += 2) {
-			if (!cpu_online(i))
-				continue;
-
-			if (i == 0) {
-				unsigned long pir = mfspr(SPRN_PIR);
-				if (__is_processor(PV_PULSAR)) {
-					RELOC(hmt_thread_data)[i].pir = 
-						pir & 0x1f;
-				} else {
-					RELOC(hmt_thread_data)[i].pir = 
-						pir & 0x3ff;
-				}
-			}
-		}
-	} else {
-		prom_printf("Processor is not HMT capable\n");
-	}
-#endif
 
 	if (cpuid > NR_CPUS)
 		prom_printf("WARNING: maximum CPUs (" __stringify(NR_CPUS)
diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
index e3fc3407bb1f..4e5c8f8d869d 100644
--- a/arch/powerpc/platforms/pseries/Kconfig
+++ b/arch/powerpc/platforms/pseries/Kconfig
@@ -9,13 +9,6 @@ config PPC_SPLPAR
 	  processors, that is, which share physical processors between
 	  two or more partitions.
 
-config HMT
-	bool "Hardware multithreading"
-	depends on SMP && PPC_PSERIES && BROKEN
-	help
-	  This option enables hardware multithreading on RS64 cpus.
-	  pSeries systems p620 and p660 have such a cpu type.
-
 config EEH
 	bool "PCI Extended Error Handling (EEH)" if EMBEDDED
 	depends on PPC_PSERIES

From 7c375b9aba4cabaeb1ca8f82807d40fd0a37103a Mon Sep 17 00:00:00 2001
From: Olof Johansson <olof@lixom.net>
Date: Sun, 12 Feb 2006 17:30:31 -0600
Subject: [PATCH 417/509] [PATCH] powerpc: Update {g5,pseries,ppc64}_defconfig

Update defconfigs for g5, pseries and generic ppc64. Default choices
for everything, with the following exceptions:

 * Enable WINDFARM_PM112 on g5 and ppc64.
 * Increase CONFIG_NR_CPUS to 4 in g5_defconfig
 * CONFIG_TIGON3=y instead of =m in g5_defconfig

Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/configs/g5_defconfig      | 174 ++++++++++---------------
 arch/powerpc/configs/ppc64_defconfig   | 173 ++++++++++--------------
 arch/powerpc/configs/pseries_defconfig | 127 +++++++-----------
 3 files changed, 189 insertions(+), 285 deletions(-)

diff --git a/arch/powerpc/configs/g5_defconfig b/arch/powerpc/configs/g5_defconfig
index d6fed3f56580..2c3fd2007676 100644
--- a/arch/powerpc/configs/g5_defconfig
+++ b/arch/powerpc/configs/g5_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.15-rc5
-# Tue Dec 20 15:59:30 2005
+# Linux kernel version: 2.6.16-rc2
+# Fri Feb 10 17:33:08 2006
 #
 CONFIG_PPC64=y
 CONFIG_64BIT=y
@@ -16,6 +16,10 @@ CONFIG_COMPAT=y
 CONFIG_SYSVIPC_COMPAT=y
 CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
 CONFIG_ARCH_MAY_HAVE_PC_FDC=y
+CONFIG_PPC_OF=y
+# CONFIG_PPC_UDBG_16550 is not set
+CONFIG_GENERIC_TBSYNC=y
+# CONFIG_DEFAULT_UIMAGE is not set
 
 #
 # Processor support
@@ -26,13 +30,12 @@ CONFIG_PPC_FPU=y
 CONFIG_ALTIVEC=y
 CONFIG_PPC_STD_MMU=y
 CONFIG_SMP=y
-CONFIG_NR_CPUS=2
+CONFIG_NR_CPUS=4
 
 #
 # Code maturity level options
 #
 CONFIG_EXPERIMENTAL=y
-CONFIG_CLEAN_COMPILE=y
 CONFIG_LOCK_KERNEL=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
 
@@ -47,8 +50,6 @@ CONFIG_POSIX_MQUEUE=y
 # CONFIG_BSD_PROCESS_ACCT is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_HOTPLUG=y
-CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 # CONFIG_CPUSETS is not set
@@ -58,8 +59,10 @@ CONFIG_CC_OPTIMIZE_FOR_SIZE=y
 CONFIG_KALLSYMS=y
 # CONFIG_KALLSYMS_ALL is not set
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
 CONFIG_PRINTK=y
 CONFIG_BUG=y
+CONFIG_ELF_CORE=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -68,8 +71,10 @@ CONFIG_CC_ALIGN_FUNCTIONS=0
 CONFIG_CC_ALIGN_LABELS=0
 CONFIG_CC_ALIGN_LOOPS=0
 CONFIG_CC_ALIGN_JUMPS=0
+CONFIG_SLAB=y
 # CONFIG_TINY_SHMEM is not set
 CONFIG_BASE_SMALL=0
+# CONFIG_SLOB is not set
 
 #
 # Loadable module support
@@ -112,13 +117,12 @@ CONFIG_PPC_PMAC=y
 CONFIG_PPC_PMAC64=y
 # CONFIG_PPC_MAPLE is not set
 # CONFIG_PPC_CELL is not set
-CONFIG_PPC_OF=y
 CONFIG_U3_DART=y
 CONFIG_MPIC=y
 # CONFIG_PPC_RTAS is not set
 # CONFIG_MMIO_NVRAM is not set
+CONFIG_MPIC_BROKEN_U3=y
 # CONFIG_PPC_MPC106 is not set
-CONFIG_GENERIC_TBSYNC=y
 CONFIG_CPU_FREQ=y
 CONFIG_CPU_FREQ_TABLE=y
 # CONFIG_CPU_FREQ_DEBUG is not set
@@ -151,6 +155,7 @@ CONFIG_FORCE_MAX_ZONEORDER=13
 CONFIG_IOMMU_VMERGE=y
 # CONFIG_HOTPLUG_CPU is not set
 CONFIG_KEXEC=y
+# CONFIG_CRASH_DUMP is not set
 CONFIG_IRQ_ALL_CPUS=y
 # CONFIG_NUMA is not set
 CONFIG_ARCH_SELECT_MEMORY_MODEL=y
@@ -202,6 +207,7 @@ CONFIG_NET=y
 #
 # Networking options
 #
+# CONFIG_NETDEBUG is not set
 CONFIG_PACKET=y
 # CONFIG_PACKET_MMAP is not set
 CONFIG_UNIX=y
@@ -239,6 +245,7 @@ CONFIG_NETFILTER=y
 # Core Netfilter Configuration
 #
 # CONFIG_NETFILTER_NETLINK is not set
+# CONFIG_NETFILTER_XTABLES is not set
 
 #
 # IP: Netfilter Configuration
@@ -255,65 +262,6 @@ CONFIG_IP_NF_TFTP=m
 CONFIG_IP_NF_AMANDA=m
 # CONFIG_IP_NF_PPTP is not set
 CONFIG_IP_NF_QUEUE=m
-CONFIG_IP_NF_IPTABLES=m
-CONFIG_IP_NF_MATCH_LIMIT=m
-CONFIG_IP_NF_MATCH_IPRANGE=m
-CONFIG_IP_NF_MATCH_MAC=m
-CONFIG_IP_NF_MATCH_PKTTYPE=m
-CONFIG_IP_NF_MATCH_MARK=m
-CONFIG_IP_NF_MATCH_MULTIPORT=m
-CONFIG_IP_NF_MATCH_TOS=m
-CONFIG_IP_NF_MATCH_RECENT=m
-CONFIG_IP_NF_MATCH_ECN=m
-CONFIG_IP_NF_MATCH_DSCP=m
-CONFIG_IP_NF_MATCH_AH_ESP=m
-CONFIG_IP_NF_MATCH_LENGTH=m
-CONFIG_IP_NF_MATCH_TTL=m
-CONFIG_IP_NF_MATCH_TCPMSS=m
-CONFIG_IP_NF_MATCH_HELPER=m
-CONFIG_IP_NF_MATCH_STATE=m
-CONFIG_IP_NF_MATCH_CONNTRACK=m
-CONFIG_IP_NF_MATCH_OWNER=m
-CONFIG_IP_NF_MATCH_ADDRTYPE=m
-CONFIG_IP_NF_MATCH_REALM=m
-CONFIG_IP_NF_MATCH_SCTP=m
-# CONFIG_IP_NF_MATCH_DCCP is not set
-CONFIG_IP_NF_MATCH_COMMENT=m
-CONFIG_IP_NF_MATCH_CONNMARK=m
-CONFIG_IP_NF_MATCH_CONNBYTES=m
-CONFIG_IP_NF_MATCH_HASHLIMIT=m
-CONFIG_IP_NF_MATCH_STRING=m
-CONFIG_IP_NF_FILTER=m
-CONFIG_IP_NF_TARGET_REJECT=m
-CONFIG_IP_NF_TARGET_LOG=m
-CONFIG_IP_NF_TARGET_ULOG=m
-CONFIG_IP_NF_TARGET_TCPMSS=m
-CONFIG_IP_NF_TARGET_NFQUEUE=m
-CONFIG_IP_NF_NAT=m
-CONFIG_IP_NF_NAT_NEEDED=y
-CONFIG_IP_NF_TARGET_MASQUERADE=m
-CONFIG_IP_NF_TARGET_REDIRECT=m
-CONFIG_IP_NF_TARGET_NETMAP=m
-CONFIG_IP_NF_TARGET_SAME=m
-CONFIG_IP_NF_NAT_SNMP_BASIC=m
-CONFIG_IP_NF_NAT_IRC=m
-CONFIG_IP_NF_NAT_FTP=m
-CONFIG_IP_NF_NAT_TFTP=m
-CONFIG_IP_NF_NAT_AMANDA=m
-CONFIG_IP_NF_MANGLE=m
-CONFIG_IP_NF_TARGET_TOS=m
-CONFIG_IP_NF_TARGET_ECN=m
-CONFIG_IP_NF_TARGET_DSCP=m
-CONFIG_IP_NF_TARGET_MARK=m
-CONFIG_IP_NF_TARGET_CLASSIFY=m
-CONFIG_IP_NF_TARGET_TTL=m
-CONFIG_IP_NF_TARGET_CONNMARK=m
-CONFIG_IP_NF_TARGET_CLUSTERIP=m
-CONFIG_IP_NF_RAW=m
-CONFIG_IP_NF_TARGET_NOTRACK=m
-CONFIG_IP_NF_ARPTABLES=m
-CONFIG_IP_NF_ARPFILTER=m
-CONFIG_IP_NF_ARP_MANGLE=m
 
 #
 # DCCP Configuration (EXPERIMENTAL)
@@ -324,6 +272,11 @@ CONFIG_IP_NF_ARP_MANGLE=m
 # SCTP Configuration (EXPERIMENTAL)
 #
 # CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
 # CONFIG_ATM is not set
 # CONFIG_BRIDGE is not set
 # CONFIG_VLAN_8021Q is not set
@@ -342,7 +295,6 @@ CONFIG_LLC=y
 # QoS and/or fair queueing
 #
 # CONFIG_NET_SCHED is not set
-CONFIG_NET_CLS_ROUTE=y
 
 #
 # Network testing
@@ -545,13 +497,7 @@ CONFIG_SCSI_SATA_SVW=y
 # CONFIG_SCSI_IPR is not set
 # CONFIG_SCSI_QLOGIC_FC is not set
 # CONFIG_SCSI_QLOGIC_1280 is not set
-CONFIG_SCSI_QLA2XXX=y
-# CONFIG_SCSI_QLA21XX is not set
-# CONFIG_SCSI_QLA22XX is not set
-# CONFIG_SCSI_QLA2300 is not set
-# CONFIG_SCSI_QLA2322 is not set
-# CONFIG_SCSI_QLA6312 is not set
-# CONFIG_SCSI_QLA24XX is not set
+# CONFIG_SCSI_QLA_FC is not set
 # CONFIG_SCSI_LPFC is not set
 # CONFIG_SCSI_DC395x is not set
 # CONFIG_SCSI_DC390T is not set
@@ -614,7 +560,6 @@ CONFIG_IEEE1394_SBP2=m
 CONFIG_IEEE1394_ETH1394=m
 CONFIG_IEEE1394_DV1394=m
 CONFIG_IEEE1394_RAWIO=y
-# CONFIG_IEEE1394_CMP is not set
 
 #
 # I2O device support
@@ -630,6 +575,7 @@ CONFIG_THERM_PM72=y
 CONFIG_WINDFARM=y
 CONFIG_WINDFARM_PM81=y
 CONFIG_WINDFARM_PM91=y
+CONFIG_WINDFARM_PM112=y
 
 #
 # Network device support
@@ -682,8 +628,9 @@ CONFIG_E1000=y
 # CONFIG_R8169 is not set
 # CONFIG_SIS190 is not set
 # CONFIG_SKGE is not set
+# CONFIG_SKY2 is not set
 # CONFIG_SK98LIN is not set
-CONFIG_TIGON3=m
+CONFIG_TIGON3=y
 # CONFIG_BNX2 is not set
 # CONFIG_MV643XX_ETH is not set
 
@@ -861,8 +808,7 @@ CONFIG_I2C_ALGOBIT=y
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
 # CONFIG_I2C_PIIX4 is not set
-CONFIG_I2C_KEYWEST=y
-CONFIG_I2C_PMAC_SMU=y
+CONFIG_I2C_POWERMAC=y
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
 # CONFIG_I2C_PROSAVAGE is not set
@@ -894,6 +840,12 @@ CONFIG_I2C_PMAC_SMU=y
 # CONFIG_I2C_DEBUG_BUS is not set
 # CONFIG_I2C_DEBUG_CHIP is not set
 
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+
 #
 # Dallas's 1-wire bus
 #
@@ -961,7 +913,6 @@ CONFIG_FB_RADEON_I2C=y
 # CONFIG_FB_KYRO is not set
 # CONFIG_FB_3DFX is not set
 # CONFIG_FB_VOODOO1 is not set
-# CONFIG_FB_CYBLA is not set
 # CONFIG_FB_TRIDENT is not set
 # CONFIG_FB_VIRTUAL is not set
 
@@ -1008,9 +959,10 @@ CONFIG_SND_OSSEMUL=y
 CONFIG_SND_MIXER_OSS=m
 CONFIG_SND_PCM_OSS=m
 CONFIG_SND_SEQUENCER_OSS=y
+# CONFIG_SND_DYNAMIC_MINORS is not set
+CONFIG_SND_SUPPORT_OLD_API=y
 # CONFIG_SND_VERBOSE_PRINTK is not set
 # CONFIG_SND_DEBUG is not set
-CONFIG_SND_GENERIC_DRIVER=y
 
 #
 # Generic devices
@@ -1024,6 +976,8 @@ CONFIG_SND_GENERIC_DRIVER=y
 #
 # PCI devices
 #
+# CONFIG_SND_AD1889 is not set
+# CONFIG_SND_ALS4000 is not set
 # CONFIG_SND_ALI5451 is not set
 # CONFIG_SND_ATIIXP is not set
 # CONFIG_SND_ATIIXP_MODEM is not set
@@ -1032,39 +986,38 @@ CONFIG_SND_GENERIC_DRIVER=y
 # CONFIG_SND_AU8830 is not set
 # CONFIG_SND_AZT3328 is not set
 # CONFIG_SND_BT87X is not set
-# CONFIG_SND_CS46XX is not set
+# CONFIG_SND_CA0106 is not set
+# CONFIG_SND_CMIPCI is not set
 # CONFIG_SND_CS4281 is not set
+# CONFIG_SND_CS46XX is not set
 # CONFIG_SND_EMU10K1 is not set
 # CONFIG_SND_EMU10K1X is not set
-# CONFIG_SND_CA0106 is not set
-# CONFIG_SND_KORG1212 is not set
-# CONFIG_SND_MIXART is not set
-# CONFIG_SND_NM256 is not set
-# CONFIG_SND_RME32 is not set
-# CONFIG_SND_RME96 is not set
-# CONFIG_SND_RME9652 is not set
-# CONFIG_SND_HDSP is not set
-# CONFIG_SND_HDSPM is not set
-# CONFIG_SND_TRIDENT is not set
-# CONFIG_SND_YMFPCI is not set
-# CONFIG_SND_AD1889 is not set
-# CONFIG_SND_ALS4000 is not set
-# CONFIG_SND_CMIPCI is not set
 # CONFIG_SND_ENS1370 is not set
 # CONFIG_SND_ENS1371 is not set
 # CONFIG_SND_ES1938 is not set
 # CONFIG_SND_ES1968 is not set
-# CONFIG_SND_MAESTRO3 is not set
 # CONFIG_SND_FM801 is not set
+# CONFIG_SND_HDA_INTEL is not set
+# CONFIG_SND_HDSP is not set
+# CONFIG_SND_HDSPM is not set
 # CONFIG_SND_ICE1712 is not set
 # CONFIG_SND_ICE1724 is not set
 # CONFIG_SND_INTEL8X0 is not set
 # CONFIG_SND_INTEL8X0M is not set
+# CONFIG_SND_KORG1212 is not set
+# CONFIG_SND_MAESTRO3 is not set
+# CONFIG_SND_MIXART is not set
+# CONFIG_SND_NM256 is not set
+# CONFIG_SND_PCXHR is not set
+# CONFIG_SND_RME32 is not set
+# CONFIG_SND_RME96 is not set
+# CONFIG_SND_RME9652 is not set
 # CONFIG_SND_SONICVIBES is not set
+# CONFIG_SND_TRIDENT is not set
 # CONFIG_SND_VIA82XX is not set
 # CONFIG_SND_VIA82XX_MODEM is not set
 # CONFIG_SND_VX222 is not set
-# CONFIG_SND_HDA_INTEL is not set
+# CONFIG_SND_YMFPCI is not set
 
 #
 # ALSA PowerMac devices
@@ -1136,13 +1089,16 @@ CONFIG_USB_STORAGE_DPCM=y
 CONFIG_USB_STORAGE_SDDR09=y
 CONFIG_USB_STORAGE_SDDR55=y
 CONFIG_USB_STORAGE_JUMPSHOT=y
+# CONFIG_USB_STORAGE_ALAUDA is not set
 # CONFIG_USB_STORAGE_ONETOUCH is not set
+# CONFIG_USB_LIBUSUAL is not set
 
 #
 # USB Input Devices
 #
 CONFIG_USB_HID=y
 CONFIG_USB_HIDINPUT=y
+# CONFIG_USB_HIDINPUT_POWERBOOK is not set
 CONFIG_HID_FF=y
 CONFIG_HID_PID=y
 CONFIG_LOGITECH_FF=y
@@ -1159,6 +1115,7 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_YEALINK is not set
 # CONFIG_USB_XPAD is not set
 # CONFIG_USB_ATI_REMOTE is not set
+# CONFIG_USB_ATI_REMOTE2 is not set
 # CONFIG_USB_KEYSPAN_REMOTE is not set
 # CONFIG_USB_APPLETOUCH is not set
 
@@ -1207,6 +1164,7 @@ CONFIG_USB_SERIAL_GENERIC=y
 # CONFIG_USB_SERIAL_AIRPRIME is not set
 # CONFIG_USB_SERIAL_ANYDATA is not set
 CONFIG_USB_SERIAL_BELKIN=m
+# CONFIG_USB_SERIAL_WHITEHEAT is not set
 CONFIG_USB_SERIAL_DIGI_ACCELEPORT=m
 # CONFIG_USB_SERIAL_CP2101 is not set
 CONFIG_USB_SERIAL_CYPRESS_M8=m
@@ -1287,6 +1245,10 @@ CONFIG_USB_EZUSB=y
 # SN Devices
 #
 
+#
+# EDAC - error detection and reporting (RAS)
+#
+
 #
 # File systems
 #
@@ -1317,6 +1279,7 @@ CONFIG_XFS_EXPORT=y
 CONFIG_XFS_SECURITY=y
 CONFIG_XFS_POSIX_ACL=y
 # CONFIG_XFS_RT is not set
+# CONFIG_OCFS2_FS is not set
 # CONFIG_MINIX_FS is not set
 # CONFIG_ROMFS_FS is not set
 CONFIG_INOTIFY=y
@@ -1357,6 +1320,7 @@ CONFIG_HUGETLBFS=y
 CONFIG_HUGETLB_PAGE=y
 CONFIG_RAMFS=y
 # CONFIG_RELAYFS_FS is not set
+# CONFIG_CONFIGFS_FS is not set
 
 #
 # Miscellaneous filesystems
@@ -1426,6 +1390,7 @@ CONFIG_MSDOS_PARTITION=y
 # CONFIG_SGI_PARTITION is not set
 # CONFIG_ULTRIX_PARTITION is not set
 # CONFIG_SUN_PARTITION is not set
+# CONFIG_KARMA_PARTITION is not set
 # CONFIG_EFI_PARTITION is not set
 
 #
@@ -1481,10 +1446,6 @@ CONFIG_CRC32=y
 CONFIG_LIBCRC32C=m
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=m
-CONFIG_TEXTSEARCH=y
-CONFIG_TEXTSEARCH_KMP=m
-CONFIG_TEXTSEARCH_BM=m
-CONFIG_TEXTSEARCH_FSM=m
 
 #
 # Instrumentation Support
@@ -1497,24 +1458,31 @@ CONFIG_OPROFILE=y
 # Kernel hacking
 #
 # CONFIG_PRINTK_TIME is not set
-CONFIG_DEBUG_KERNEL=y
 CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_KERNEL=y
 CONFIG_LOG_BUF_SHIFT=17
 CONFIG_DETECT_SOFTLOCKUP=y
 # CONFIG_SCHEDSTATS is not set
 # CONFIG_DEBUG_SLAB is not set
+CONFIG_DEBUG_MUTEXES=y
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
 CONFIG_DEBUG_FS=y
 # CONFIG_DEBUG_VM is not set
+CONFIG_FORCED_INLINING=y
 # CONFIG_RCU_TORTURE_TEST is not set
 # CONFIG_DEBUG_STACKOVERFLOW is not set
 # CONFIG_DEBUG_STACK_USAGE is not set
 # CONFIG_DEBUGGER is not set
 CONFIG_IRQSTACKS=y
 CONFIG_BOOTX_TEXT=y
+# CONFIG_PPC_EARLY_DEBUG_LPAR is not set
+# CONFIG_PPC_EARLY_DEBUG_G5 is not set
+# CONFIG_PPC_EARLY_DEBUG_RTAS is not set
+# CONFIG_PPC_EARLY_DEBUG_MAPLE is not set
+# CONFIG_PPC_EARLY_DEBUG_ISERIES is not set
 
 #
 # Security options
diff --git a/arch/powerpc/configs/ppc64_defconfig b/arch/powerpc/configs/ppc64_defconfig
index 6f6c6bed1aa5..0362a70aa97c 100644
--- a/arch/powerpc/configs/ppc64_defconfig
+++ b/arch/powerpc/configs/ppc64_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.15-rc5
-# Tue Dec 20 15:59:38 2005
+# Linux kernel version: 2.6.16-rc2
+# Fri Feb 10 17:32:14 2006
 #
 CONFIG_PPC64=y
 CONFIG_64BIT=y
@@ -16,6 +16,10 @@ CONFIG_COMPAT=y
 CONFIG_SYSVIPC_COMPAT=y
 CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
 CONFIG_ARCH_MAY_HAVE_PC_FDC=y
+CONFIG_PPC_OF=y
+CONFIG_PPC_UDBG_16550=y
+CONFIG_GENERIC_TBSYNC=y
+# CONFIG_DEFAULT_UIMAGE is not set
 
 #
 # Processor support
@@ -33,7 +37,6 @@ CONFIG_NR_CPUS=32
 # Code maturity level options
 #
 CONFIG_EXPERIMENTAL=y
-CONFIG_CLEAN_COMPILE=y
 CONFIG_LOCK_KERNEL=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
 
@@ -48,8 +51,6 @@ CONFIG_POSIX_MQUEUE=y
 # CONFIG_BSD_PROCESS_ACCT is not set
 CONFIG_SYSCTL=y
 # CONFIG_AUDIT is not set
-CONFIG_HOTPLUG=y
-CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 CONFIG_CPUSETS=y
@@ -59,8 +60,10 @@ CONFIG_CC_OPTIMIZE_FOR_SIZE=y
 CONFIG_KALLSYMS=y
 CONFIG_KALLSYMS_ALL=y
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
 CONFIG_PRINTK=y
 CONFIG_BUG=y
+CONFIG_ELF_CORE=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -69,8 +72,10 @@ CONFIG_CC_ALIGN_FUNCTIONS=0
 CONFIG_CC_ALIGN_LABELS=0
 CONFIG_CC_ALIGN_LOOPS=0
 CONFIG_CC_ALIGN_JUMPS=0
+CONFIG_SLAB=y
 # CONFIG_TINY_SHMEM is not set
 CONFIG_BASE_SMALL=0
+# CONFIG_SLOB is not set
 
 #
 # Loadable module support
@@ -113,7 +118,6 @@ CONFIG_PPC_PMAC=y
 CONFIG_PPC_PMAC64=y
 CONFIG_PPC_MAPLE=y
 # CONFIG_PPC_CELL is not set
-CONFIG_PPC_OF=y
 CONFIG_XICS=y
 CONFIG_U3_DART=y
 CONFIG_MPIC=y
@@ -124,8 +128,8 @@ CONFIG_RTAS_FLASH=m
 # CONFIG_MMIO_NVRAM is not set
 CONFIG_MPIC_BROKEN_U3=y
 CONFIG_IBMVIO=y
+# CONFIG_IBMEBUS is not set
 # CONFIG_PPC_MPC106 is not set
-CONFIG_GENERIC_TBSYNC=y
 CONFIG_CPU_FREQ=y
 CONFIG_CPU_FREQ_TABLE=y
 # CONFIG_CPU_FREQ_DEBUG is not set
@@ -158,6 +162,7 @@ CONFIG_FORCE_MAX_ZONEORDER=13
 CONFIG_IOMMU_VMERGE=y
 CONFIG_HOTPLUG_CPU=y
 CONFIG_KEXEC=y
+# CONFIG_CRASH_DUMP is not set
 CONFIG_IRQ_ALL_CPUS=y
 CONFIG_PPC_SPLPAR=y
 CONFIG_EEH=y
@@ -178,6 +183,7 @@ CONFIG_HAVE_MEMORY_PRESENT=y
 CONFIG_SPARSEMEM_EXTREME=y
 # CONFIG_MEMORY_HOTPLUG is not set
 CONFIG_SPLIT_PTLOCK_CPUS=4
+CONFIG_MIGRATION=y
 # CONFIG_PPC_64K_PAGES is not set
 # CONFIG_SCHED_SMT is not set
 CONFIG_PROC_DEVICETREE=y
@@ -221,6 +227,7 @@ CONFIG_NET=y
 #
 # Networking options
 #
+# CONFIG_NETDEBUG is not set
 CONFIG_PACKET=y
 # CONFIG_PACKET_MMAP is not set
 CONFIG_UNIX=y
@@ -260,6 +267,7 @@ CONFIG_NETFILTER=y
 CONFIG_NETFILTER_NETLINK=y
 CONFIG_NETFILTER_NETLINK_QUEUE=m
 CONFIG_NETFILTER_NETLINK_LOG=m
+# CONFIG_NETFILTER_XTABLES is not set
 
 #
 # IP: Netfilter Configuration
@@ -277,65 +285,6 @@ CONFIG_IP_NF_TFTP=m
 CONFIG_IP_NF_AMANDA=m
 # CONFIG_IP_NF_PPTP is not set
 CONFIG_IP_NF_QUEUE=m
-CONFIG_IP_NF_IPTABLES=m
-CONFIG_IP_NF_MATCH_LIMIT=m
-CONFIG_IP_NF_MATCH_IPRANGE=m
-CONFIG_IP_NF_MATCH_MAC=m
-CONFIG_IP_NF_MATCH_PKTTYPE=m
-CONFIG_IP_NF_MATCH_MARK=m
-CONFIG_IP_NF_MATCH_MULTIPORT=m
-CONFIG_IP_NF_MATCH_TOS=m
-CONFIG_IP_NF_MATCH_RECENT=m
-CONFIG_IP_NF_MATCH_ECN=m
-CONFIG_IP_NF_MATCH_DSCP=m
-CONFIG_IP_NF_MATCH_AH_ESP=m
-CONFIG_IP_NF_MATCH_LENGTH=m
-CONFIG_IP_NF_MATCH_TTL=m
-CONFIG_IP_NF_MATCH_TCPMSS=m
-CONFIG_IP_NF_MATCH_HELPER=m
-CONFIG_IP_NF_MATCH_STATE=m
-CONFIG_IP_NF_MATCH_CONNTRACK=m
-CONFIG_IP_NF_MATCH_OWNER=m
-CONFIG_IP_NF_MATCH_ADDRTYPE=m
-CONFIG_IP_NF_MATCH_REALM=m
-CONFIG_IP_NF_MATCH_SCTP=m
-CONFIG_IP_NF_MATCH_DCCP=m
-CONFIG_IP_NF_MATCH_COMMENT=m
-CONFIG_IP_NF_MATCH_CONNMARK=m
-CONFIG_IP_NF_MATCH_CONNBYTES=m
-CONFIG_IP_NF_MATCH_HASHLIMIT=m
-CONFIG_IP_NF_MATCH_STRING=m
-CONFIG_IP_NF_FILTER=m
-CONFIG_IP_NF_TARGET_REJECT=m
-CONFIG_IP_NF_TARGET_LOG=m
-CONFIG_IP_NF_TARGET_ULOG=m
-CONFIG_IP_NF_TARGET_TCPMSS=m
-CONFIG_IP_NF_TARGET_NFQUEUE=m
-CONFIG_IP_NF_NAT=m
-CONFIG_IP_NF_NAT_NEEDED=y
-CONFIG_IP_NF_TARGET_MASQUERADE=m
-CONFIG_IP_NF_TARGET_REDIRECT=m
-CONFIG_IP_NF_TARGET_NETMAP=m
-CONFIG_IP_NF_TARGET_SAME=m
-CONFIG_IP_NF_NAT_SNMP_BASIC=m
-CONFIG_IP_NF_NAT_IRC=m
-CONFIG_IP_NF_NAT_FTP=m
-CONFIG_IP_NF_NAT_TFTP=m
-CONFIG_IP_NF_NAT_AMANDA=m
-CONFIG_IP_NF_MANGLE=m
-CONFIG_IP_NF_TARGET_TOS=m
-CONFIG_IP_NF_TARGET_ECN=m
-CONFIG_IP_NF_TARGET_DSCP=m
-CONFIG_IP_NF_TARGET_MARK=m
-CONFIG_IP_NF_TARGET_CLASSIFY=m
-CONFIG_IP_NF_TARGET_TTL=m
-CONFIG_IP_NF_TARGET_CONNMARK=m
-CONFIG_IP_NF_TARGET_CLUSTERIP=m
-CONFIG_IP_NF_RAW=m
-CONFIG_IP_NF_TARGET_NOTRACK=m
-CONFIG_IP_NF_ARPTABLES=m
-CONFIG_IP_NF_ARPFILTER=m
-CONFIG_IP_NF_ARP_MANGLE=m
 
 #
 # DCCP Configuration (EXPERIMENTAL)
@@ -346,6 +295,11 @@ CONFIG_IP_NF_ARP_MANGLE=m
 # SCTP Configuration (EXPERIMENTAL)
 #
 # CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
 # CONFIG_ATM is not set
 # CONFIG_BRIDGE is not set
 # CONFIG_VLAN_8021Q is not set
@@ -364,7 +318,6 @@ CONFIG_LLC=y
 # QoS and/or fair queueing
 #
 # CONFIG_NET_SCHED is not set
-CONFIG_NET_CLS_ROUTE=y
 
 #
 # Network testing
@@ -572,13 +525,7 @@ CONFIG_SCSI_IPR_TRACE=y
 CONFIG_SCSI_IPR_DUMP=y
 # CONFIG_SCSI_QLOGIC_FC is not set
 # CONFIG_SCSI_QLOGIC_1280 is not set
-CONFIG_SCSI_QLA2XXX=y
-CONFIG_SCSI_QLA21XX=m
-CONFIG_SCSI_QLA22XX=m
-CONFIG_SCSI_QLA2300=m
-CONFIG_SCSI_QLA2322=m
-CONFIG_SCSI_QLA6312=m
-CONFIG_SCSI_QLA24XX=m
+# CONFIG_SCSI_QLA_FC is not set
 CONFIG_SCSI_LPFC=m
 # CONFIG_SCSI_DC395x is not set
 # CONFIG_SCSI_DC390T is not set
@@ -642,8 +589,6 @@ CONFIG_IEEE1394_SBP2=m
 CONFIG_IEEE1394_ETH1394=m
 CONFIG_IEEE1394_DV1394=m
 CONFIG_IEEE1394_RAWIO=y
-CONFIG_IEEE1394_CMP=m
-CONFIG_IEEE1394_AMDTP=m
 
 #
 # I2O device support
@@ -659,6 +604,7 @@ CONFIG_THERM_PM72=y
 CONFIG_WINDFARM=y
 CONFIG_WINDFARM_PM81=y
 CONFIG_WINDFARM_PM91=y
+CONFIG_WINDFARM_PM112=y
 
 #
 # Network device support
@@ -731,6 +677,7 @@ CONFIG_E1000=y
 # CONFIG_R8169 is not set
 # CONFIG_SIS190 is not set
 # CONFIG_SKGE is not set
+# CONFIG_SKY2 is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 CONFIG_TIGON3=y
@@ -853,6 +800,7 @@ CONFIG_HW_CONSOLE=y
 CONFIG_SERIAL_8250=y
 CONFIG_SERIAL_8250_CONSOLE=y
 CONFIG_SERIAL_8250_NR_UARTS=4
+CONFIG_SERIAL_8250_RUNTIME_UARTS=4
 # CONFIG_SERIAL_8250_EXTENDED is not set
 
 #
@@ -880,6 +828,7 @@ CONFIG_HVCS=m
 # CONFIG_WATCHDOG is not set
 # CONFIG_RTC is not set
 CONFIG_GEN_RTC=y
+# CONFIG_GEN_RTC_X is not set
 # CONFIG_DTLK is not set
 # CONFIG_R3964 is not set
 # CONFIG_APPLICOM is not set
@@ -923,8 +872,7 @@ CONFIG_I2C_AMD8111=y
 # CONFIG_I2C_I801 is not set
 # CONFIG_I2C_I810 is not set
 # CONFIG_I2C_PIIX4 is not set
-CONFIG_I2C_KEYWEST=y
-CONFIG_I2C_PMAC_SMU=y
+CONFIG_I2C_POWERMAC=y
 # CONFIG_I2C_NFORCE2 is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
 # CONFIG_I2C_PROSAVAGE is not set
@@ -956,6 +904,12 @@ CONFIG_I2C_PMAC_SMU=y
 # CONFIG_I2C_DEBUG_BUS is not set
 # CONFIG_I2C_DEBUG_CHIP is not set
 
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+
 #
 # Dallas's 1-wire bus
 #
@@ -1028,7 +982,6 @@ CONFIG_FB_RADEON_I2C=y
 # CONFIG_FB_KYRO is not set
 # CONFIG_FB_3DFX is not set
 # CONFIG_FB_VOODOO1 is not set
-# CONFIG_FB_CYBLA is not set
 # CONFIG_FB_TRIDENT is not set
 # CONFIG_FB_VIRTUAL is not set
 
@@ -1073,9 +1026,10 @@ CONFIG_SND_OSSEMUL=y
 CONFIG_SND_MIXER_OSS=m
 CONFIG_SND_PCM_OSS=m
 CONFIG_SND_SEQUENCER_OSS=y
+# CONFIG_SND_DYNAMIC_MINORS is not set
+CONFIG_SND_SUPPORT_OLD_API=y
 # CONFIG_SND_VERBOSE_PRINTK is not set
 # CONFIG_SND_DEBUG is not set
-CONFIG_SND_GENERIC_DRIVER=y
 
 #
 # Generic devices
@@ -1089,6 +1043,8 @@ CONFIG_SND_GENERIC_DRIVER=y
 #
 # PCI devices
 #
+# CONFIG_SND_AD1889 is not set
+# CONFIG_SND_ALS4000 is not set
 # CONFIG_SND_ALI5451 is not set
 # CONFIG_SND_ATIIXP is not set
 # CONFIG_SND_ATIIXP_MODEM is not set
@@ -1097,39 +1053,38 @@ CONFIG_SND_GENERIC_DRIVER=y
 # CONFIG_SND_AU8830 is not set
 # CONFIG_SND_AZT3328 is not set
 # CONFIG_SND_BT87X is not set
-# CONFIG_SND_CS46XX is not set
+# CONFIG_SND_CA0106 is not set
+# CONFIG_SND_CMIPCI is not set
 # CONFIG_SND_CS4281 is not set
+# CONFIG_SND_CS46XX is not set
 # CONFIG_SND_EMU10K1 is not set
 # CONFIG_SND_EMU10K1X is not set
-# CONFIG_SND_CA0106 is not set
-# CONFIG_SND_KORG1212 is not set
-# CONFIG_SND_MIXART is not set
-# CONFIG_SND_NM256 is not set
-# CONFIG_SND_RME32 is not set
-# CONFIG_SND_RME96 is not set
-# CONFIG_SND_RME9652 is not set
-# CONFIG_SND_HDSP is not set
-# CONFIG_SND_HDSPM is not set
-# CONFIG_SND_TRIDENT is not set
-# CONFIG_SND_YMFPCI is not set
-# CONFIG_SND_AD1889 is not set
-# CONFIG_SND_ALS4000 is not set
-# CONFIG_SND_CMIPCI is not set
 # CONFIG_SND_ENS1370 is not set
 # CONFIG_SND_ENS1371 is not set
 # CONFIG_SND_ES1938 is not set
 # CONFIG_SND_ES1968 is not set
-# CONFIG_SND_MAESTRO3 is not set
 # CONFIG_SND_FM801 is not set
+# CONFIG_SND_HDA_INTEL is not set
+# CONFIG_SND_HDSP is not set
+# CONFIG_SND_HDSPM is not set
 # CONFIG_SND_ICE1712 is not set
 # CONFIG_SND_ICE1724 is not set
 # CONFIG_SND_INTEL8X0 is not set
 # CONFIG_SND_INTEL8X0M is not set
+# CONFIG_SND_KORG1212 is not set
+# CONFIG_SND_MAESTRO3 is not set
+# CONFIG_SND_MIXART is not set
+# CONFIG_SND_NM256 is not set
+# CONFIG_SND_PCXHR is not set
+# CONFIG_SND_RME32 is not set
+# CONFIG_SND_RME96 is not set
+# CONFIG_SND_RME9652 is not set
 # CONFIG_SND_SONICVIBES is not set
+# CONFIG_SND_TRIDENT is not set
 # CONFIG_SND_VIA82XX is not set
 # CONFIG_SND_VIA82XX_MODEM is not set
 # CONFIG_SND_VX222 is not set
-# CONFIG_SND_HDA_INTEL is not set
+# CONFIG_SND_YMFPCI is not set
 
 #
 # ALSA PowerMac devices
@@ -1201,13 +1156,16 @@ CONFIG_USB_STORAGE=m
 # CONFIG_USB_STORAGE_SDDR09 is not set
 # CONFIG_USB_STORAGE_SDDR55 is not set
 # CONFIG_USB_STORAGE_JUMPSHOT is not set
+# CONFIG_USB_STORAGE_ALAUDA is not set
 # CONFIG_USB_STORAGE_ONETOUCH is not set
+# CONFIG_USB_LIBUSUAL is not set
 
 #
 # USB Input Devices
 #
 CONFIG_USB_HID=y
 CONFIG_USB_HIDINPUT=y
+# CONFIG_USB_HIDINPUT_POWERBOOK is not set
 # CONFIG_HID_FF is not set
 CONFIG_USB_HIDDEV=y
 # CONFIG_USB_AIPTEK is not set
@@ -1221,6 +1179,7 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_YEALINK is not set
 # CONFIG_USB_XPAD is not set
 # CONFIG_USB_ATI_REMOTE is not set
+# CONFIG_USB_ATI_REMOTE2 is not set
 # CONFIG_USB_KEYSPAN_REMOTE is not set
 # CONFIG_USB_APPLETOUCH is not set
 
@@ -1306,6 +1265,10 @@ CONFIG_INFINIBAND_IPOIB=m
 # SN Devices
 #
 
+#
+# EDAC - error detection and reporting (RAS)
+#
+
 #
 # File systems
 #
@@ -1340,6 +1303,7 @@ CONFIG_XFS_EXPORT=y
 CONFIG_XFS_SECURITY=y
 CONFIG_XFS_POSIX_ACL=y
 # CONFIG_XFS_RT is not set
+# CONFIG_OCFS2_FS is not set
 # CONFIG_MINIX_FS is not set
 # CONFIG_ROMFS_FS is not set
 CONFIG_INOTIFY=y
@@ -1379,6 +1343,7 @@ CONFIG_HUGETLBFS=y
 CONFIG_HUGETLB_PAGE=y
 CONFIG_RAMFS=y
 # CONFIG_RELAYFS_FS is not set
+# CONFIG_CONFIGFS_FS is not set
 
 #
 # Miscellaneous filesystems
@@ -1449,6 +1414,7 @@ CONFIG_MSDOS_PARTITION=y
 # CONFIG_SGI_PARTITION is not set
 # CONFIG_ULTRIX_PARTITION is not set
 # CONFIG_SUN_PARTITION is not set
+# CONFIG_KARMA_PARTITION is not set
 # CONFIG_EFI_PARTITION is not set
 
 #
@@ -1504,10 +1470,6 @@ CONFIG_CRC32=y
 CONFIG_LIBCRC32C=m
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=m
-CONFIG_TEXTSEARCH=y
-CONFIG_TEXTSEARCH_KMP=m
-CONFIG_TEXTSEARCH_BM=m
-CONFIG_TEXTSEARCH_FSM=m
 
 #
 # Instrumentation Support
@@ -1520,18 +1482,20 @@ CONFIG_OPROFILE=y
 # Kernel hacking
 #
 # CONFIG_PRINTK_TIME is not set
-CONFIG_DEBUG_KERNEL=y
 CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_KERNEL=y
 CONFIG_LOG_BUF_SHIFT=17
 CONFIG_DETECT_SOFTLOCKUP=y
 # CONFIG_SCHEDSTATS is not set
 # CONFIG_DEBUG_SLAB is not set
+CONFIG_DEBUG_MUTEXES=y
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
 CONFIG_DEBUG_FS=y
 # CONFIG_DEBUG_VM is not set
+CONFIG_FORCED_INLINING=y
 # CONFIG_RCU_TORTURE_TEST is not set
 CONFIG_DEBUG_STACKOVERFLOW=y
 CONFIG_DEBUG_STACK_USAGE=y
@@ -1540,6 +1504,11 @@ CONFIG_XMON=y
 # CONFIG_XMON_DEFAULT is not set
 CONFIG_IRQSTACKS=y
 CONFIG_BOOTX_TEXT=y
+# CONFIG_PPC_EARLY_DEBUG_LPAR is not set
+# CONFIG_PPC_EARLY_DEBUG_G5 is not set
+# CONFIG_PPC_EARLY_DEBUG_RTAS is not set
+# CONFIG_PPC_EARLY_DEBUG_MAPLE is not set
+# CONFIG_PPC_EARLY_DEBUG_ISERIES is not set
 
 #
 # Security options
diff --git a/arch/powerpc/configs/pseries_defconfig b/arch/powerpc/configs/pseries_defconfig
index aa9893a1f6e8..daaf038a1faa 100644
--- a/arch/powerpc/configs/pseries_defconfig
+++ b/arch/powerpc/configs/pseries_defconfig
@@ -1,7 +1,7 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.15-rc5
-# Tue Dec 20 15:59:40 2005
+# Linux kernel version: 2.6.16-rc2
+# Fri Feb 10 17:33:32 2006
 #
 CONFIG_PPC64=y
 CONFIG_64BIT=y
@@ -16,6 +16,10 @@ CONFIG_COMPAT=y
 CONFIG_SYSVIPC_COMPAT=y
 CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER=y
 CONFIG_ARCH_MAY_HAVE_PC_FDC=y
+CONFIG_PPC_OF=y
+CONFIG_PPC_UDBG_16550=y
+# CONFIG_GENERIC_TBSYNC is not set
+# CONFIG_DEFAULT_UIMAGE is not set
 
 #
 # Processor support
@@ -33,7 +37,6 @@ CONFIG_NR_CPUS=128
 # Code maturity level options
 #
 CONFIG_EXPERIMENTAL=y
-CONFIG_CLEAN_COMPILE=y
 CONFIG_LOCK_KERNEL=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
 
@@ -49,8 +52,6 @@ CONFIG_POSIX_MQUEUE=y
 CONFIG_SYSCTL=y
 CONFIG_AUDIT=y
 CONFIG_AUDITSYSCALL=y
-CONFIG_HOTPLUG=y
-CONFIG_KOBJECT_UEVENT=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 CONFIG_CPUSETS=y
@@ -60,8 +61,10 @@ CONFIG_CC_OPTIMIZE_FOR_SIZE=y
 CONFIG_KALLSYMS=y
 CONFIG_KALLSYMS_ALL=y
 # CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
 CONFIG_PRINTK=y
 CONFIG_BUG=y
+CONFIG_ELF_CORE=y
 CONFIG_BASE_FULL=y
 CONFIG_FUTEX=y
 CONFIG_EPOLL=y
@@ -70,8 +73,10 @@ CONFIG_CC_ALIGN_FUNCTIONS=0
 CONFIG_CC_ALIGN_LABELS=0
 CONFIG_CC_ALIGN_LOOPS=0
 CONFIG_CC_ALIGN_JUMPS=0
+CONFIG_SLAB=y
 # CONFIG_TINY_SHMEM is not set
 CONFIG_BASE_SMALL=0
+# CONFIG_SLOB is not set
 
 #
 # Loadable module support
@@ -113,7 +118,6 @@ CONFIG_PPC_PSERIES=y
 # CONFIG_PPC_PMAC is not set
 # CONFIG_PPC_MAPLE is not set
 # CONFIG_PPC_CELL is not set
-CONFIG_PPC_OF=y
 CONFIG_XICS=y
 # CONFIG_U3_DART is not set
 CONFIG_MPIC=y
@@ -123,8 +127,8 @@ CONFIG_RTAS_PROC=y
 CONFIG_RTAS_FLASH=m
 # CONFIG_MMIO_NVRAM is not set
 CONFIG_IBMVIO=y
+# CONFIG_IBMEBUS is not set
 # CONFIG_PPC_MPC106 is not set
-# CONFIG_GENERIC_TBSYNC is not set
 # CONFIG_CPU_FREQ is not set
 # CONFIG_WANT_EARLY_SERIAL is not set
 
@@ -145,6 +149,7 @@ CONFIG_FORCE_MAX_ZONEORDER=13
 CONFIG_IOMMU_VMERGE=y
 CONFIG_HOTPLUG_CPU=y
 CONFIG_KEXEC=y
+# CONFIG_CRASH_DUMP is not set
 CONFIG_IRQ_ALL_CPUS=y
 CONFIG_PPC_SPLPAR=y
 CONFIG_EEH=y
@@ -165,6 +170,7 @@ CONFIG_HAVE_MEMORY_PRESENT=y
 CONFIG_SPARSEMEM_EXTREME=y
 # CONFIG_MEMORY_HOTPLUG is not set
 CONFIG_SPLIT_PTLOCK_CPUS=4
+CONFIG_MIGRATION=y
 CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID=y
 # CONFIG_PPC_64K_PAGES is not set
 CONFIG_SCHED_SMT=y
@@ -209,6 +215,7 @@ CONFIG_NET=y
 #
 # Networking options
 #
+# CONFIG_NETDEBUG is not set
 CONFIG_PACKET=y
 # CONFIG_PACKET_MMAP is not set
 CONFIG_UNIX=y
@@ -248,6 +255,7 @@ CONFIG_NETFILTER=y
 CONFIG_NETFILTER_NETLINK=y
 CONFIG_NETFILTER_NETLINK_QUEUE=m
 CONFIG_NETFILTER_NETLINK_LOG=m
+# CONFIG_NETFILTER_XTABLES is not set
 
 #
 # IP: Netfilter Configuration
@@ -265,65 +273,6 @@ CONFIG_IP_NF_TFTP=m
 CONFIG_IP_NF_AMANDA=m
 # CONFIG_IP_NF_PPTP is not set
 CONFIG_IP_NF_QUEUE=m
-CONFIG_IP_NF_IPTABLES=m
-CONFIG_IP_NF_MATCH_LIMIT=m
-CONFIG_IP_NF_MATCH_IPRANGE=m
-CONFIG_IP_NF_MATCH_MAC=m
-CONFIG_IP_NF_MATCH_PKTTYPE=m
-CONFIG_IP_NF_MATCH_MARK=m
-CONFIG_IP_NF_MATCH_MULTIPORT=m
-CONFIG_IP_NF_MATCH_TOS=m
-CONFIG_IP_NF_MATCH_RECENT=m
-CONFIG_IP_NF_MATCH_ECN=m
-CONFIG_IP_NF_MATCH_DSCP=m
-CONFIG_IP_NF_MATCH_AH_ESP=m
-CONFIG_IP_NF_MATCH_LENGTH=m
-CONFIG_IP_NF_MATCH_TTL=m
-CONFIG_IP_NF_MATCH_TCPMSS=m
-CONFIG_IP_NF_MATCH_HELPER=m
-CONFIG_IP_NF_MATCH_STATE=m
-CONFIG_IP_NF_MATCH_CONNTRACK=m
-CONFIG_IP_NF_MATCH_OWNER=m
-CONFIG_IP_NF_MATCH_ADDRTYPE=m
-CONFIG_IP_NF_MATCH_REALM=m
-CONFIG_IP_NF_MATCH_SCTP=m
-# CONFIG_IP_NF_MATCH_DCCP is not set
-CONFIG_IP_NF_MATCH_COMMENT=m
-CONFIG_IP_NF_MATCH_CONNMARK=m
-CONFIG_IP_NF_MATCH_CONNBYTES=m
-CONFIG_IP_NF_MATCH_HASHLIMIT=m
-CONFIG_IP_NF_MATCH_STRING=m
-CONFIG_IP_NF_FILTER=m
-CONFIG_IP_NF_TARGET_REJECT=m
-CONFIG_IP_NF_TARGET_LOG=m
-CONFIG_IP_NF_TARGET_ULOG=m
-CONFIG_IP_NF_TARGET_TCPMSS=m
-CONFIG_IP_NF_TARGET_NFQUEUE=m
-CONFIG_IP_NF_NAT=m
-CONFIG_IP_NF_NAT_NEEDED=y
-CONFIG_IP_NF_TARGET_MASQUERADE=m
-CONFIG_IP_NF_TARGET_REDIRECT=m
-CONFIG_IP_NF_TARGET_NETMAP=m
-CONFIG_IP_NF_TARGET_SAME=m
-CONFIG_IP_NF_NAT_SNMP_BASIC=m
-CONFIG_IP_NF_NAT_IRC=m
-CONFIG_IP_NF_NAT_FTP=m
-CONFIG_IP_NF_NAT_TFTP=m
-CONFIG_IP_NF_NAT_AMANDA=m
-CONFIG_IP_NF_MANGLE=m
-CONFIG_IP_NF_TARGET_TOS=m
-CONFIG_IP_NF_TARGET_ECN=m
-CONFIG_IP_NF_TARGET_DSCP=m
-CONFIG_IP_NF_TARGET_MARK=m
-CONFIG_IP_NF_TARGET_CLASSIFY=m
-CONFIG_IP_NF_TARGET_TTL=m
-CONFIG_IP_NF_TARGET_CONNMARK=m
-CONFIG_IP_NF_TARGET_CLUSTERIP=m
-CONFIG_IP_NF_RAW=m
-CONFIG_IP_NF_TARGET_NOTRACK=m
-CONFIG_IP_NF_ARPTABLES=m
-CONFIG_IP_NF_ARPFILTER=m
-CONFIG_IP_NF_ARP_MANGLE=m
 
 #
 # DCCP Configuration (EXPERIMENTAL)
@@ -334,6 +283,11 @@ CONFIG_IP_NF_ARP_MANGLE=m
 # SCTP Configuration (EXPERIMENTAL)
 #
 # CONFIG_IP_SCTP is not set
+
+#
+# TIPC Configuration (EXPERIMENTAL)
+#
+# CONFIG_TIPC is not set
 # CONFIG_ATM is not set
 # CONFIG_BRIDGE is not set
 # CONFIG_VLAN_8021Q is not set
@@ -352,7 +306,6 @@ CONFIG_LLC=y
 # QoS and/or fair queueing
 #
 # CONFIG_NET_SCHED is not set
-CONFIG_NET_CLS_ROUTE=y
 
 #
 # Network testing
@@ -550,13 +503,7 @@ CONFIG_SCSI_IPR_TRACE=y
 CONFIG_SCSI_IPR_DUMP=y
 # CONFIG_SCSI_QLOGIC_FC is not set
 # CONFIG_SCSI_QLOGIC_1280 is not set
-CONFIG_SCSI_QLA2XXX=y
-CONFIG_SCSI_QLA21XX=m
-CONFIG_SCSI_QLA22XX=m
-CONFIG_SCSI_QLA2300=m
-CONFIG_SCSI_QLA2322=m
-CONFIG_SCSI_QLA6312=m
-CONFIG_SCSI_QLA24XX=m
+# CONFIG_SCSI_QLA_FC is not set
 CONFIG_SCSI_LPFC=m
 # CONFIG_SCSI_DC395x is not set
 # CONFIG_SCSI_DC390T is not set
@@ -678,6 +625,7 @@ CONFIG_E1000=y
 # CONFIG_R8169 is not set
 # CONFIG_SIS190 is not set
 # CONFIG_SKGE is not set
+# CONFIG_SKY2 is not set
 # CONFIG_SK98LIN is not set
 # CONFIG_VIA_VELOCITY is not set
 CONFIG_TIGON3=y
@@ -803,6 +751,7 @@ CONFIG_HW_CONSOLE=y
 CONFIG_SERIAL_8250=y
 CONFIG_SERIAL_8250_CONSOLE=y
 CONFIG_SERIAL_8250_NR_UARTS=4
+CONFIG_SERIAL_8250_RUNTIME_UARTS=4
 # CONFIG_SERIAL_8250_EXTENDED is not set
 
 #
@@ -908,6 +857,12 @@ CONFIG_I2C_ALGOBIT=y
 # CONFIG_I2C_DEBUG_BUS is not set
 # CONFIG_I2C_DEBUG_CHIP is not set
 
+#
+# SPI support
+#
+# CONFIG_SPI is not set
+# CONFIG_SPI_MASTER is not set
+
 #
 # Dallas's 1-wire bus
 #
@@ -976,7 +931,6 @@ CONFIG_FB_RADEON_I2C=y
 # CONFIG_FB_KYRO is not set
 # CONFIG_FB_3DFX is not set
 # CONFIG_FB_VOODOO1 is not set
-# CONFIG_FB_CYBLA is not set
 # CONFIG_FB_TRIDENT is not set
 # CONFIG_FB_VIRTUAL is not set
 
@@ -1061,12 +1015,15 @@ CONFIG_USB_STORAGE=y
 # CONFIG_USB_STORAGE_SDDR09 is not set
 # CONFIG_USB_STORAGE_SDDR55 is not set
 # CONFIG_USB_STORAGE_JUMPSHOT is not set
+# CONFIG_USB_STORAGE_ALAUDA is not set
+# CONFIG_USB_LIBUSUAL is not set
 
 #
 # USB Input Devices
 #
 CONFIG_USB_HID=y
 CONFIG_USB_HIDINPUT=y
+# CONFIG_USB_HIDINPUT_POWERBOOK is not set
 # CONFIG_HID_FF is not set
 CONFIG_USB_HIDDEV=y
 # CONFIG_USB_AIPTEK is not set
@@ -1080,6 +1037,7 @@ CONFIG_USB_HIDDEV=y
 # CONFIG_USB_YEALINK is not set
 # CONFIG_USB_XPAD is not set
 # CONFIG_USB_ATI_REMOTE is not set
+# CONFIG_USB_ATI_REMOTE2 is not set
 # CONFIG_USB_KEYSPAN_REMOTE is not set
 # CONFIG_USB_APPLETOUCH is not set
 
@@ -1166,6 +1124,10 @@ CONFIG_INFINIBAND_IPOIB=m
 # SN Devices
 #
 
+#
+# EDAC - error detection and reporting (RAS)
+#
+
 #
 # File systems
 #
@@ -1200,6 +1162,7 @@ CONFIG_XFS_EXPORT=y
 CONFIG_XFS_SECURITY=y
 CONFIG_XFS_POSIX_ACL=y
 # CONFIG_XFS_RT is not set
+# CONFIG_OCFS2_FS is not set
 # CONFIG_MINIX_FS is not set
 # CONFIG_ROMFS_FS is not set
 CONFIG_INOTIFY=y
@@ -1240,6 +1203,7 @@ CONFIG_HUGETLBFS=y
 CONFIG_HUGETLB_PAGE=y
 CONFIG_RAMFS=y
 # CONFIG_RELAYFS_FS is not set
+# CONFIG_CONFIGFS_FS is not set
 
 #
 # Miscellaneous filesystems
@@ -1351,10 +1315,6 @@ CONFIG_CRC32=y
 CONFIG_LIBCRC32C=m
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=m
-CONFIG_TEXTSEARCH=y
-CONFIG_TEXTSEARCH_KMP=m
-CONFIG_TEXTSEARCH_BM=m
-CONFIG_TEXTSEARCH_FSM=m
 
 #
 # Instrumentation Support
@@ -1367,18 +1327,20 @@ CONFIG_OPROFILE=y
 # Kernel hacking
 #
 # CONFIG_PRINTK_TIME is not set
-CONFIG_DEBUG_KERNEL=y
 CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_KERNEL=y
 CONFIG_LOG_BUF_SHIFT=17
 CONFIG_DETECT_SOFTLOCKUP=y
 # CONFIG_SCHEDSTATS is not set
 # CONFIG_DEBUG_SLAB is not set
+CONFIG_DEBUG_MUTEXES=y
 # CONFIG_DEBUG_SPINLOCK is not set
 # CONFIG_DEBUG_SPINLOCK_SLEEP is not set
 # CONFIG_DEBUG_KOBJECT is not set
 # CONFIG_DEBUG_INFO is not set
 CONFIG_DEBUG_FS=y
 # CONFIG_DEBUG_VM is not set
+CONFIG_FORCED_INLINING=y
 # CONFIG_RCU_TORTURE_TEST is not set
 CONFIG_DEBUG_STACKOVERFLOW=y
 CONFIG_DEBUG_STACK_USAGE=y
@@ -1387,6 +1349,11 @@ CONFIG_XMON=y
 CONFIG_XMON_DEFAULT=y
 CONFIG_IRQSTACKS=y
 # CONFIG_BOOTX_TEXT is not set
+# CONFIG_PPC_EARLY_DEBUG_LPAR is not set
+# CONFIG_PPC_EARLY_DEBUG_G5 is not set
+# CONFIG_PPC_EARLY_DEBUG_RTAS is not set
+# CONFIG_PPC_EARLY_DEBUG_MAPLE is not set
+# CONFIG_PPC_EARLY_DEBUG_ISERIES is not set
 
 #
 # Security options

From 72b138198cd6307c679b35d677ed64105b94ab48 Mon Sep 17 00:00:00 2001
From: Segher Boessenkool <segher@kernel.crashing.org>
Date: Fri, 17 Feb 2006 11:25:42 +0100
Subject: [PATCH 418/509] [PATCH] powerpc: Fix some MPIC + HT APIC buglets

Do disable, not enable, the HT APIC IRQ in the function that is
supposed to.
Enable the MPIC IRQ before enabling the downstream APIC IRQ, avoids
potentially losing an interrupt.

Signed-off-by: Segher Boessenkool <segher@kernel.crashing.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/sysdev/mpic.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/arch/powerpc/sysdev/mpic.c b/arch/powerpc/sysdev/mpic.c
index 4f26304d0263..7dcdfcb3c984 100644
--- a/arch/powerpc/sysdev/mpic.c
+++ b/arch/powerpc/sysdev/mpic.c
@@ -234,7 +234,7 @@ static void mpic_shutdown_ht_interrupt(struct mpic *mpic, unsigned int source,
 	spin_lock_irqsave(&mpic->fixup_lock, flags);
 	writeb(0x10 + 2 * fixup->index, fixup->base + 2);
 	tmp = readl(fixup->base + 4);
-	tmp &= ~1U;
+	tmp |= 1;
 	writel(tmp, fixup->base + 4);
 	spin_unlock_irqrestore(&mpic->fixup_lock, flags);
 }
@@ -446,14 +446,15 @@ static unsigned int mpic_startup_irq(unsigned int irq)
 #ifdef CONFIG_MPIC_BROKEN_U3
 	struct mpic *mpic = mpic_from_irq(irq);
 	unsigned int src = irq - mpic->irq_offset;
-
-	if (mpic_is_ht_interrupt(mpic, src))
-		mpic_startup_ht_interrupt(mpic, src, irq_desc[irq].status);
-
 #endif /* CONFIG_MPIC_BROKEN_U3 */
 
 	mpic_enable_irq(irq);
 
+#ifdef CONFIG_MPIC_BROKEN_U3
+	if (mpic_is_ht_interrupt(mpic, src))
+		mpic_startup_ht_interrupt(mpic, src, irq_desc[irq].status);
+#endif /* CONFIG_MPIC_BROKEN_U3 */
+
 	return 0;
 }
 

From 4558f417f49595337b7e9cc3e92bc0856c588ac1 Mon Sep 17 00:00:00 2001
From: Segher Boessenkool <segher@kernel.crashing.org>
Date: Fri, 17 Feb 2006 11:30:30 +0100
Subject: [PATCH 419/509] [PATCH] powerpc: Don't re-assign PCI resources on
 Maple

Maple firmware does not need PCI resource allocation, and in fact, it
can cause problems in some strange cases.

Signed-off-by: Segher Boessenkool <segher@kernel.crashing.org>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/maple/pci.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/platforms/maple/pci.c b/arch/powerpc/platforms/maple/pci.c
index 7d4099a34f92..85d6c93659cc 100644
--- a/arch/powerpc/platforms/maple/pci.c
+++ b/arch/powerpc/platforms/maple/pci.c
@@ -435,8 +435,8 @@ void __init maple_pci_init(void)
 			PCI_DN(np)->busno = 0xf0;
 	}
 
-	/* Tell pci.c to use the common resource allocation mecanism */
-	pci_probe_only = 0;
+	/* Tell pci.c to not change any resource allocations.  */
+	pci_probe_only = 1;
 	
 	/* Allow all IO */
 	io_page_mask = -1;

From fb5c594c2acc441f0d2d8f457484a0e0e9285db3 Mon Sep 17 00:00:00 2001
From: Michal Ostrowski <mostrows@watson.ibm.com>
Date: Sat, 18 Feb 2006 09:29:59 -0500
Subject: [PATCH 420/509] [PATCH] Fix race condition in hvc console.

tty_schedule_flip() would schedule a thread that would call flush_to_ldisc().
If tty_buffer_request_room() gets called prior to that thread running --
which is likely in this loop in hvc_poll(), it would set the active flag
in the tty buffer and consequently flush_to_ldisc() would ignore it.

The result is that input on the hvc console is not processed.

This fix calls tty_flip_buffer_push (and flags the tty as
"low_latency").  The push to the ldisc thus happens synchronously.

Signed-off-by: Michal Ostrowski <mostrows@watson.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 drivers/char/hvc_console.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c
index 1994a92d4733..f65b2e14a485 100644
--- a/drivers/char/hvc_console.c
+++ b/drivers/char/hvc_console.c
@@ -335,6 +335,8 @@ static int hvc_open(struct tty_struct *tty, struct file * filp)
 	} /* else count == 0 */
 
 	tty->driver_data = hp;
+	tty->low_latency = 1; /* Makes flushes to ldisc synchronous. */
+
 	hp->tty = tty;
 	/* Save for request_irq outside of spin_lock. */
 	irq = hp->irq;
@@ -633,9 +635,6 @@ static int hvc_poll(struct hvc_struct *hp)
 			tty_insert_flip_char(tty, buf[i], 0);
 		}
 
-		if (count)
-			tty_schedule_flip(tty);
-
 		/*
 		 * Account for the total amount read in one loop, and if above
 		 * 64 bytes, we do a quick schedule loop to let the tty grok
@@ -656,6 +655,9 @@ static int hvc_poll(struct hvc_struct *hp)
  bail:
 	spin_unlock_irqrestore(&hp->lock, flags);
 
+	if (read_total)
+		tty_flip_buffer_push(tty);
+	
 	return poll_mask;
 }
 

From c6622f63db86fcbd41bf6fe05ddf2e00c1e51ced Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@samba.org>
Date: Fri, 24 Feb 2006 10:06:59 +1100
Subject: [PATCH 421/509] powerpc: Implement accurate task and CPU time
 accounting

This implements accurate task and cpu time accounting for 64-bit
powerpc kernels.  Instead of accounting a whole jiffy of time to a
task on a timer interrupt because that task happened to be running at
the time, we now account time in units of timebase ticks according to
the actual time spent by the task in user mode and kernel mode.  We
also count the time spent processing hardware and software interrupts
accurately.  This is conditional on CONFIG_VIRT_CPU_ACCOUNTING.  If
that is not set, we do tick-based approximate accounting as before.

To get this accurate information, we read either the PURR (processor
utilization of resources register) on POWER5 machines, or the timebase
on other machines on

* each entry to the kernel from usermode
* each exit to usermode
* transitions between process context, hard irq context and soft irq
  context in kernel mode
* context switches.

On POWER5 systems with shared-processor logical partitioning we also
read both the PURR and the timebase at each timer interrupt and
context switch in order to determine how much time has been taken by
the hypervisor to run other partitions ("steal" time).  Unfortunately,
since we need values of the PURR on both threads at the same time to
accurately calculate the steal time, and since we can only calculate
steal time on a per-core basis, the apportioning of the steal time
between idle time (time which we ceded to the hypervisor in the idle
loop) and actual stolen time is somewhat approximate at the moment.

This is all based quite heavily on what s390 does, and it uses the
generic interfaces that were added by the s390 developers,
i.e. account_system_time(), account_user_time(), etc.

This patch doesn't add any new interfaces between the kernel and
userspace, and doesn't change the units in which time is reported to
userspace by things such as /proc/stat, /proc/<pid>/stat, getrusage(),
times(), etc.  Internally the various task and cpu times are stored in
timebase units, but they are converted to USER_HZ units (1/100th of a
second) when reported to userspace.  Some precision is therefore lost
but there should not be any accumulating error, since the internal
accumulation is at full precision.

Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/Kconfig              |  15 ++
 arch/powerpc/kernel/asm-offsets.c |   3 +
 arch/powerpc/kernel/entry_64.S    |   7 +-
 arch/powerpc/kernel/head_64.S     |   9 ++
 arch/powerpc/kernel/irq.c         |  30 +++-
 arch/powerpc/kernel/process.c     |   7 +-
 arch/powerpc/kernel/smp.c         |   4 +-
 arch/powerpc/kernel/time.c        | 236 +++++++++++++++++++++++++++++-
 include/asm-powerpc/cputable.h    |   4 +-
 include/asm-powerpc/cputime.h     | 202 +++++++++++++++++++++++++
 include/asm-powerpc/irq.h         |   6 +-
 include/asm-powerpc/paca.h        |   5 +
 include/asm-powerpc/ppc_asm.h     |  42 ++++++
 include/asm-powerpc/system.h      |   4 +
 include/asm-powerpc/time.h        |  15 ++
 include/asm-ppc/time.h            |   5 +
 16 files changed, 577 insertions(+), 17 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index fb0dcb994b84..d112aed2999b 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -250,6 +250,21 @@ config PPC_STD_MMU_32
 	def_bool y
 	depends on PPC_STD_MMU && PPC32
 
+config VIRT_CPU_ACCOUNTING
+	bool "Deterministic task and CPU time accounting"
+	depends on PPC64
+	default y
+	help
+	  Select this option to enable more accurate task and CPU time
+	  accounting.  This is done by reading a CPU counter on each
+	  kernel entry and exit and on transitions within the kernel
+	  between system, softirq and hardirq state, so there is a
+	  small performance impact.  This also enables accounting of
+	  stolen time on logically-partitioned systems running on
+	  IBM POWER5-based machines.
+
+	  If in doubt, say Y here.
+
 config SMP
 	depends on PPC_STD_MMU
 	bool "Symmetric multi-processing support"
diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
index 840aad43a98b..18810ac55bcc 100644
--- a/arch/powerpc/kernel/asm-offsets.c
+++ b/arch/powerpc/kernel/asm-offsets.c
@@ -137,6 +137,9 @@ int main(void)
 	DEFINE(PACAEMERGSP, offsetof(struct paca_struct, emergency_sp));
 	DEFINE(PACALPPACAPTR, offsetof(struct paca_struct, lppaca_ptr));
 	DEFINE(PACAHWCPUID, offsetof(struct paca_struct, hw_cpu_id));
+	DEFINE(PACA_STARTPURR, offsetof(struct paca_struct, startpurr));
+	DEFINE(PACA_USER_TIME, offsetof(struct paca_struct, user_time));
+	DEFINE(PACA_SYSTEM_TIME, offsetof(struct paca_struct, system_time));
 
 	DEFINE(LPPACASRR0, offsetof(struct lppaca, saved_srr0));
 	DEFINE(LPPACASRR1, offsetof(struct lppaca, saved_srr1));
diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
index 79a0c910f0d8..8f606c1889fa 100644
--- a/arch/powerpc/kernel/entry_64.S
+++ b/arch/powerpc/kernel/entry_64.S
@@ -61,6 +61,7 @@ system_call_common:
 	std	r12,_MSR(r1)
 	std	r0,GPR0(r1)
 	std	r10,GPR1(r1)
+	ACCOUNT_CPU_USER_ENTRY(r10, r11)
 	std	r2,GPR2(r1)
 	std	r3,GPR3(r1)
 	std	r4,GPR4(r1)
@@ -168,8 +169,9 @@ syscall_error_cont:
 	stdcx.	r0,0,r1			/* to clear the reservation */
 	andi.	r6,r8,MSR_PR
 	ld	r4,_LINK(r1)
-	beq-	1f			/* only restore r13 if */
-	ld	r13,GPR13(r1)		/* returning to usermode */
+	beq-	1f
+	ACCOUNT_CPU_USER_EXIT(r11, r12)
+	ld	r13,GPR13(r1)	/* only restore r13 if returning to usermode */
 1:	ld	r2,GPR2(r1)
 	li	r12,MSR_RI
 	andc	r11,r10,r12
@@ -536,6 +538,7 @@ restore:
 	 * userspace
 	 */
 	beq	1f
+	ACCOUNT_CPU_USER_EXIT(r3, r4)
 	REST_GPR(13, r1)
 1:
 	ld	r3,_CTR(r1)
diff --git a/arch/powerpc/kernel/head_64.S b/arch/powerpc/kernel/head_64.S
index 2b21ec499285..be3ae7733577 100644
--- a/arch/powerpc/kernel/head_64.S
+++ b/arch/powerpc/kernel/head_64.S
@@ -277,6 +277,7 @@ exception_marker:
 	std	r10,0(r1);		/* make stack chain pointer	*/ \
 	std	r0,GPR0(r1);		/* save r0 in stackframe	*/ \
 	std	r10,GPR1(r1);		/* save r1 in stackframe	*/ \
+	ACCOUNT_CPU_USER_ENTRY(r9, r10);				   \
 	std	r2,GPR2(r1);		/* save r2 in stackframe	*/ \
 	SAVE_4GPRS(3, r1);		/* save r3 - r6 in stackframe	*/ \
 	SAVE_2GPRS(7, r1);		/* save r7, r8 in stackframe	*/ \
@@ -844,6 +845,14 @@ fast_exception_return:
 	ld	r11,_NIP(r1)
 	andi.	r3,r12,MSR_RI		/* check if RI is set */
 	beq-	unrecov_fer
+
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING
+	andi.	r3,r12,MSR_PR
+	beq	2f
+	ACCOUNT_CPU_USER_EXIT(r3, r4)
+2:
+#endif
+
 	ld	r3,_CCR(r1)
 	ld	r4,_LINK(r1)
 	ld	r5,_CTR(r1)
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index edb2b00edbd2..24dc8117b822 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -369,6 +369,7 @@ unsigned int real_irq_to_virt_slowpath(unsigned int real_irq)
 	return NO_IRQ;
 
 }
+#endif /* CONFIG_PPC64 */
 
 #ifdef CONFIG_IRQSTACKS
 struct thread_info *softirq_ctx[NR_CPUS];
@@ -392,10 +393,24 @@ void irq_ctx_init(void)
 	}
 }
 
+static inline void do_softirq_onstack(void)
+{
+	struct thread_info *curtp, *irqtp;
+
+	curtp = current_thread_info();
+	irqtp = softirq_ctx[smp_processor_id()];
+	irqtp->task = curtp->task;
+	call_do_softirq(irqtp);
+	irqtp->task = NULL;
+}
+
+#else
+#define do_softirq_onstack()	__do_softirq()
+#endif /* CONFIG_IRQSTACKS */
+
 void do_softirq(void)
 {
 	unsigned long flags;
-	struct thread_info *curtp, *irqtp;
 
 	if (in_interrupt())
 		return;
@@ -403,19 +418,18 @@ void do_softirq(void)
 	local_irq_save(flags);
 
 	if (local_softirq_pending()) {
-		curtp = current_thread_info();
-		irqtp = softirq_ctx[smp_processor_id()];
-		irqtp->task = curtp->task;
-		call_do_softirq(irqtp);
-		irqtp->task = NULL;
+		account_system_vtime(current);
+		local_bh_disable();
+		do_softirq_onstack();
+		account_system_vtime(current);
+		__local_bh_enable();
 	}
 
 	local_irq_restore(flags);
 }
 EXPORT_SYMBOL(do_softirq);
 
-#endif /* CONFIG_IRQSTACKS */
-
+#ifdef CONFIG_PPC64
 static int __init setup_noirqdistrib(char *str)
 {
 	distribute_irqs = 0;
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index dd774c3c9302..1770a066c217 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -45,9 +45,9 @@
 #include <asm/mmu.h>
 #include <asm/prom.h>
 #include <asm/machdep.h>
+#include <asm/time.h>
 #ifdef CONFIG_PPC64
 #include <asm/firmware.h>
-#include <asm/time.h>
 #endif
 
 extern unsigned long _get_SP(void);
@@ -328,6 +328,11 @@ struct task_struct *__switch_to(struct task_struct *prev,
 #endif
 
 	local_irq_save(flags);
+
+	account_system_vtime(current);
+	account_process_vtime(current);
+	calculate_steal_time();
+
 	last = _switch(old_thread, new_thread);
 
 	local_irq_restore(flags);
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index 13595a64f013..805eaedbc308 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -541,7 +541,7 @@ int __devinit start_secondary(void *unused)
 		smp_ops->take_timebase();
 
 	if (system_state > SYSTEM_BOOTING)
-		per_cpu(last_jiffy, cpu) = get_tb();
+		snapshot_timebase();
 
 	spin_lock(&call_lock);
 	cpu_set(cpu, cpu_online_map);
@@ -573,6 +573,8 @@ void __init smp_cpus_done(unsigned int max_cpus)
 
 	set_cpus_allowed(current, old_mask);
 
+	snapshot_timebases();
+
 	dump_numa_cpu_topology();
 }
 
diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 2a7ddc579379..0b34db28916f 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -51,6 +51,7 @@
 #include <linux/percpu.h>
 #include <linux/rtc.h>
 #include <linux/jiffies.h>
+#include <linux/posix-timers.h>
 
 #include <asm/io.h>
 #include <asm/processor.h>
@@ -135,6 +136,220 @@ unsigned long tb_last_stamp;
  */
 DEFINE_PER_CPU(unsigned long, last_jiffy);
 
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING
+/*
+ * Factors for converting from cputime_t (timebase ticks) to
+ * jiffies, milliseconds, seconds, and clock_t (1/USER_HZ seconds).
+ * These are all stored as 0.64 fixed-point binary fractions.
+ */
+u64 __cputime_jiffies_factor;
+u64 __cputime_msec_factor;
+u64 __cputime_sec_factor;
+u64 __cputime_clockt_factor;
+
+static void calc_cputime_factors(void)
+{
+	struct div_result res;
+
+	div128_by_32(HZ, 0, tb_ticks_per_sec, &res);
+	__cputime_jiffies_factor = res.result_low;
+	div128_by_32(1000, 0, tb_ticks_per_sec, &res);
+	__cputime_msec_factor = res.result_low;
+	div128_by_32(1, 0, tb_ticks_per_sec, &res);
+	__cputime_sec_factor = res.result_low;
+	div128_by_32(USER_HZ, 0, tb_ticks_per_sec, &res);
+	__cputime_clockt_factor = res.result_low;
+}
+
+/*
+ * Read the PURR on systems that have it, otherwise the timebase.
+ */
+static u64 read_purr(void)
+{
+	if (cpu_has_feature(CPU_FTR_PURR))
+		return mfspr(SPRN_PURR);
+	return mftb();
+}
+
+/*
+ * Account time for a transition between system, hard irq
+ * or soft irq state.
+ */
+void account_system_vtime(struct task_struct *tsk)
+{
+	u64 now, delta;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	now = read_purr();
+	delta = now - get_paca()->startpurr;
+	get_paca()->startpurr = now;
+	if (!in_interrupt()) {
+		delta += get_paca()->system_time;
+		get_paca()->system_time = 0;
+	}
+	account_system_time(tsk, 0, delta);
+	local_irq_restore(flags);
+}
+
+/*
+ * Transfer the user and system times accumulated in the paca
+ * by the exception entry and exit code to the generic process
+ * user and system time records.
+ * Must be called with interrupts disabled.
+ */
+void account_process_vtime(struct task_struct *tsk)
+{
+	cputime_t utime;
+
+	utime = get_paca()->user_time;
+	get_paca()->user_time = 0;
+	account_user_time(tsk, utime);
+}
+
+static void account_process_time(struct pt_regs *regs)
+{
+	int cpu = smp_processor_id();
+
+	account_process_vtime(current);
+	run_local_timers();
+	if (rcu_pending(cpu))
+		rcu_check_callbacks(cpu, user_mode(regs));
+	scheduler_tick();
+ 	run_posix_cpu_timers(current);
+}
+
+#ifdef CONFIG_PPC_SPLPAR
+/*
+ * Stuff for accounting stolen time.
+ */
+struct cpu_purr_data {
+	int	initialized;			/* thread is running */
+	u64	tb0;			/* timebase at origin time */
+	u64	purr0;			/* PURR at origin time */
+	u64	tb;			/* last TB value read */
+	u64	purr;			/* last PURR value read */
+	u64	stolen;			/* stolen time so far */
+	spinlock_t lock;
+};
+
+static DEFINE_PER_CPU(struct cpu_purr_data, cpu_purr_data);
+
+static void snapshot_tb_and_purr(void *data)
+{
+	struct cpu_purr_data *p = &__get_cpu_var(cpu_purr_data);
+
+	p->tb0 = mftb();
+	p->purr0 = mfspr(SPRN_PURR);
+	p->tb = p->tb0;
+	p->purr = 0;
+	wmb();
+	p->initialized = 1;
+}
+
+/*
+ * Called during boot when all cpus have come up.
+ */
+void snapshot_timebases(void)
+{
+	int cpu;
+
+	if (!cpu_has_feature(CPU_FTR_PURR))
+		return;
+	for_each_cpu(cpu)
+		spin_lock_init(&per_cpu(cpu_purr_data, cpu).lock);
+	on_each_cpu(snapshot_tb_and_purr, NULL, 0, 1);
+}
+
+void calculate_steal_time(void)
+{
+	u64 tb, purr, t0;
+	s64 stolen;
+	struct cpu_purr_data *p0, *pme, *phim;
+	int cpu;
+
+	if (!cpu_has_feature(CPU_FTR_PURR))
+		return;
+	cpu = smp_processor_id();
+	pme = &per_cpu(cpu_purr_data, cpu);
+	if (!pme->initialized)
+		return;		/* this can happen in early boot */
+	p0 = &per_cpu(cpu_purr_data, cpu & ~1);
+	phim = &per_cpu(cpu_purr_data, cpu ^ 1);
+	spin_lock(&p0->lock);
+	tb = mftb();
+	purr = mfspr(SPRN_PURR) - pme->purr0;
+	if (!phim->initialized || !cpu_online(cpu ^ 1)) {
+		stolen = (tb - pme->tb) - (purr - pme->purr);
+	} else {
+		t0 = pme->tb0;
+		if (phim->tb0 < t0)
+			t0 = phim->tb0;
+		stolen = phim->tb - t0 - phim->purr - purr - p0->stolen;
+	}
+	if (stolen > 0) {
+		account_steal_time(current, stolen);
+		p0->stolen += stolen;
+	}
+	pme->tb = tb;
+	pme->purr = purr;
+	spin_unlock(&p0->lock);
+}
+
+/*
+ * Must be called before the cpu is added to the online map when
+ * a cpu is being brought up at runtime.
+ */
+static void snapshot_purr(void)
+{
+	int cpu;
+	u64 purr;
+	struct cpu_purr_data *p0, *pme, *phim;
+	unsigned long flags;
+
+	if (!cpu_has_feature(CPU_FTR_PURR))
+		return;
+	cpu = smp_processor_id();
+	pme = &per_cpu(cpu_purr_data, cpu);
+	p0 = &per_cpu(cpu_purr_data, cpu & ~1);
+	phim = &per_cpu(cpu_purr_data, cpu ^ 1);
+	spin_lock_irqsave(&p0->lock, flags);
+	pme->tb = pme->tb0 = mftb();
+	purr = mfspr(SPRN_PURR);
+	if (!phim->initialized) {
+		pme->purr = 0;
+		pme->purr0 = purr;
+	} else {
+		/* set p->purr and p->purr0 for no change in p0->stolen */
+		pme->purr = phim->tb - phim->tb0 - phim->purr - p0->stolen;
+		pme->purr0 = purr - pme->purr;
+	}
+	pme->initialized = 1;
+	spin_unlock_irqrestore(&p0->lock, flags);
+}
+
+#endif /* CONFIG_PPC_SPLPAR */
+
+#else /* ! CONFIG_VIRT_CPU_ACCOUNTING */
+#define calc_cputime_factors()
+#define account_process_time(regs)	update_process_times(user_mode(regs))
+#define calculate_steal_time()		do { } while (0)
+#endif
+
+#if !(defined(CONFIG_VIRT_CPU_ACCOUNTING) && defined(CONFIG_PPC_SPLPAR))
+#define snapshot_purr()			do { } while (0)
+#endif
+
+/*
+ * Called when a cpu comes up after the system has finished booting,
+ * i.e. as a result of a hotplug cpu action.
+ */
+void snapshot_timebase(void)
+{
+	__get_cpu_var(last_jiffy) = get_tb();
+	snapshot_purr();
+}
+
 void __delay(unsigned long loops)
 {
 	unsigned long start;
@@ -382,6 +597,7 @@ static void iSeries_tb_recal(void)
 						new_tb_ticks_per_jiffy, sign, tick_diff );
 				tb_ticks_per_jiffy = new_tb_ticks_per_jiffy;
 				tb_ticks_per_sec   = new_tb_ticks_per_sec;
+				calc_cputime_factors();
 				div128_by_32( XSEC_PER_SEC, 0, tb_ticks_per_sec, &divres );
 				do_gtod.tb_ticks_per_sec = tb_ticks_per_sec;
 				tb_to_xs = divres.result_low;
@@ -430,6 +646,7 @@ void timer_interrupt(struct pt_regs * regs)
 	irq_enter();
 
 	profile_tick(CPU_PROFILING, regs);
+	calculate_steal_time();
 
 #ifdef CONFIG_PPC_ISERIES
 	get_lppaca()->int_dword.fields.decr_int = 0;
@@ -451,7 +668,7 @@ void timer_interrupt(struct pt_regs * regs)
 		 * is the case.
 		 */
 		if (!cpu_is_offline(cpu))
-			update_process_times(user_mode(regs));
+			account_process_time(regs);
 
 		/*
 		 * No need to check whether cpu is offline here; boot_cpuid
@@ -508,13 +725,27 @@ void wakeup_decrementer(void)
 void __init smp_space_timers(unsigned int max_cpus)
 {
 	int i;
+	unsigned long half = tb_ticks_per_jiffy / 2;
 	unsigned long offset = tb_ticks_per_jiffy / max_cpus;
 	unsigned long previous_tb = per_cpu(last_jiffy, boot_cpuid);
 
 	/* make sure tb > per_cpu(last_jiffy, cpu) for all cpus always */
 	previous_tb -= tb_ticks_per_jiffy;
+	/*
+	 * The stolen time calculation for POWER5 shared-processor LPAR
+	 * systems works better if the two threads' timebase interrupts
+	 * are staggered by half a jiffy with respect to each other.
+	 */
 	for_each_cpu(i) {
-		if (i != boot_cpuid) {
+		if (i == boot_cpuid)
+			continue;
+		if (i == (boot_cpuid ^ 1))
+			per_cpu(last_jiffy, i) =
+				per_cpu(last_jiffy, boot_cpuid) - half;
+		else if (i & 1)
+			per_cpu(last_jiffy, i) =
+				per_cpu(last_jiffy, i ^ 1) + half;
+		else {
 			previous_tb += offset;
 			per_cpu(last_jiffy, i) = previous_tb;
 		}
@@ -706,6 +937,7 @@ void __init time_init(void)
 	tb_ticks_per_sec = ppc_tb_freq;
 	tb_ticks_per_usec = ppc_tb_freq / 1000000;
 	tb_to_us = mulhwu_scale_factor(ppc_tb_freq, 1000000);
+	calc_cputime_factors();
 
 	/*
 	 * Calculate the length of each tick in ns.  It will not be
diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h
index 90d005bb4d1c..99d12ff6346c 100644
--- a/include/asm-powerpc/cputable.h
+++ b/include/asm-powerpc/cputable.h
@@ -117,6 +117,7 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 #define CPU_FTR_MMCRA_SIHV		ASM_CONST(0x0000080000000000)
 #define CPU_FTR_CI_LARGE_PAGE		ASM_CONST(0x0000100000000000)
 #define CPU_FTR_PAUSE_ZERO		ASM_CONST(0x0000200000000000)
+#define CPU_FTR_PURR			ASM_CONST(0x0000400000000000)
 #else
 /* ensure on 32b processors the flags are available for compiling but
  * don't do anything */
@@ -132,6 +133,7 @@ extern void do_cpu_ftr_fixups(unsigned long offset);
 #define CPU_FTR_LOCKLESS_TLBIE		ASM_CONST(0x0)
 #define CPU_FTR_MMCRA_SIHV		ASM_CONST(0x0)
 #define CPU_FTR_CI_LARGE_PAGE		ASM_CONST(0x0)
+#define CPU_FTR_PURR			ASM_CONST(0x0)
 #endif
 
 #ifndef __ASSEMBLY__
@@ -316,7 +318,7 @@ enum {
 	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 |
 	    CPU_FTR_MMCRA | CPU_FTR_SMT |
 	    CPU_FTR_COHERENT_ICACHE | CPU_FTR_LOCKLESS_TLBIE |
-	    CPU_FTR_MMCRA_SIHV,
+	    CPU_FTR_MMCRA_SIHV | CPU_FTR_PURR,
 	CPU_FTRS_CELL = CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB |
 	    CPU_FTR_HPTE_TABLE | CPU_FTR_PPCAS_ARCH_V2 |
 	    CPU_FTR_ALTIVEC_COMP | CPU_FTR_MMCRA | CPU_FTR_SMT |
diff --git a/include/asm-powerpc/cputime.h b/include/asm-powerpc/cputime.h
index 6d68ad7e0ea3..a21185d47883 100644
--- a/include/asm-powerpc/cputime.h
+++ b/include/asm-powerpc/cputime.h
@@ -1 +1,203 @@
+/*
+ * Definitions for measuring cputime on powerpc machines.
+ *
+ * Copyright (C) 2006 Paul Mackerras, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * If we have CONFIG_VIRT_CPU_ACCOUNTING, we measure cpu time in
+ * the same units as the timebase.  Otherwise we measure cpu time
+ * in jiffies using the generic definitions.
+ */
+
+#ifndef __POWERPC_CPUTIME_H
+#define __POWERPC_CPUTIME_H
+
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
 #include <asm-generic/cputime.h>
+#else
+
+#include <linux/types.h>
+#include <linux/time.h>
+#include <asm/div64.h>
+#include <asm/time.h>
+#include <asm/param.h>
+
+typedef u64 cputime_t;
+typedef u64 cputime64_t;
+
+#define cputime_zero			((cputime_t)0)
+#define cputime_max			((~((cputime_t)0) >> 1) - 1)
+#define cputime_add(__a, __b)		((__a) +  (__b))
+#define cputime_sub(__a, __b)		((__a) -  (__b))
+#define cputime_div(__a, __n)		((__a) /  (__n))
+#define cputime_halve(__a)		((__a) >> 1)
+#define cputime_eq(__a, __b)		((__a) == (__b))
+#define cputime_gt(__a, __b)		((__a) >  (__b))
+#define cputime_ge(__a, __b)		((__a) >= (__b))
+#define cputime_lt(__a, __b)		((__a) <  (__b))
+#define cputime_le(__a, __b)		((__a) <= (__b))
+
+#define cputime64_zero			((cputime64_t)0)
+#define cputime64_add(__a, __b)		((__a) + (__b))
+#define cputime_to_cputime64(__ct)	(__ct)
+
+#ifdef __KERNEL__
+
+/*
+ * Convert cputime <-> jiffies
+ */
+extern u64 __cputime_jiffies_factor;
+
+static inline unsigned long cputime_to_jiffies(const cputime_t ct)
+{
+	return mulhdu(ct, __cputime_jiffies_factor);
+}
+
+static inline cputime_t jiffies_to_cputime(const unsigned long jif)
+{
+	cputime_t ct;
+	unsigned long sec;
+
+	/* have to be a little careful about overflow */
+	ct = jif % HZ;
+	sec = jif / HZ;
+	if (ct) {
+		ct *= tb_ticks_per_sec;
+		do_div(ct, HZ);
+	}
+	if (sec)
+		ct += (cputime_t) sec * tb_ticks_per_sec;
+	return ct;
+}
+
+static inline u64 cputime64_to_jiffies64(const cputime_t ct)
+{
+	return mulhdu(ct, __cputime_jiffies_factor);
+}
+
+/*
+ * Convert cputime <-> milliseconds
+ */
+extern u64 __cputime_msec_factor;
+
+static inline unsigned long cputime_to_msecs(const cputime_t ct)
+{
+	return mulhdu(ct, __cputime_msec_factor);
+}
+
+static inline cputime_t msecs_to_cputime(const unsigned long ms)
+{
+	cputime_t ct;
+	unsigned long sec;
+
+	/* have to be a little careful about overflow */
+	ct = ms % 1000;
+	sec = ms / 1000;
+	if (ct) {
+		ct *= tb_ticks_per_sec;
+		do_div(ct, 1000);
+	}
+	if (sec)
+		ct += (cputime_t) sec * tb_ticks_per_sec;
+	return ct;
+}
+
+/*
+ * Convert cputime <-> seconds
+ */
+extern u64 __cputime_sec_factor;
+
+static inline unsigned long cputime_to_secs(const cputime_t ct)
+{
+	return mulhdu(ct, __cputime_sec_factor);
+}
+
+static inline cputime_t secs_to_cputime(const unsigned long sec)
+{
+	return (cputime_t) sec * tb_ticks_per_sec;
+}
+
+/*
+ * Convert cputime <-> timespec
+ */
+static inline void cputime_to_timespec(const cputime_t ct, struct timespec *p)
+{
+	u64 x = ct;
+	unsigned int frac;
+
+	frac = do_div(x, tb_ticks_per_sec);
+	p->tv_sec = x;
+	x = (u64) frac * 1000000000;
+	do_div(x, tb_ticks_per_sec);
+	p->tv_nsec = x;
+}
+
+static inline cputime_t timespec_to_cputime(const struct timespec *p)
+{
+	cputime_t ct;
+
+	ct = (u64) p->tv_nsec * tb_ticks_per_sec;
+	do_div(ct, 1000000000);
+	return ct + (u64) p->tv_sec * tb_ticks_per_sec;
+}
+
+/*
+ * Convert cputime <-> timeval
+ */
+static inline void cputime_to_timeval(const cputime_t ct, struct timeval *p)
+{
+	u64 x = ct;
+	unsigned int frac;
+
+	frac = do_div(x, tb_ticks_per_sec);
+	p->tv_sec = x;
+	x = (u64) frac * 1000000;
+	do_div(x, tb_ticks_per_sec);
+	p->tv_usec = x;
+}
+
+static inline cputime_t timeval_to_cputime(const struct timeval *p)
+{
+	cputime_t ct;
+
+	ct = (u64) p->tv_usec * tb_ticks_per_sec;
+	do_div(ct, 1000000);
+	return ct + (u64) p->tv_sec * tb_ticks_per_sec;
+}
+
+/*
+ * Convert cputime <-> clock_t (units of 1/USER_HZ seconds)
+ */
+extern u64 __cputime_clockt_factor;
+
+static inline unsigned long cputime_to_clock_t(const cputime_t ct)
+{
+	return mulhdu(ct, __cputime_clockt_factor);
+}
+
+static inline cputime_t clock_t_to_cputime(const unsigned long clk)
+{
+	cputime_t ct;
+	unsigned long sec;
+
+	/* have to be a little careful about overflow */
+	ct = clk % USER_HZ;
+	sec = clk / USER_HZ;
+	if (ct) {
+		ct *= tb_ticks_per_sec;
+		do_div(ct, USER_HZ);
+	}
+	if (sec)
+		ct += (cputime_t) sec * tb_ticks_per_sec;
+	return ct;
+}
+
+#define cputime64_to_clock_t(ct)	cputime_to_clock_t((cputime_t)(ct))
+
+#endif /* __KERNEL__ */
+#endif /* CONFIG_VIRT_CPU_ACCOUNTING */
+#endif /* __POWERPC_CPUTIME_H */
diff --git a/include/asm-powerpc/irq.h b/include/asm-powerpc/irq.h
index 8eb7e857ec4c..51f87d9993b6 100644
--- a/include/asm-powerpc/irq.h
+++ b/include/asm-powerpc/irq.h
@@ -479,6 +479,10 @@ extern int distribute_irqs;
 struct irqaction;
 struct pt_regs;
 
+#define __ARCH_HAS_DO_SOFTIRQ
+
+extern void __do_softirq(void);
+
 #ifdef CONFIG_IRQSTACKS
 /*
  * Per-cpu stacks for handling hard and soft interrupts.
@@ -491,8 +495,6 @@ extern void call_do_softirq(struct thread_info *tp);
 extern int call___do_IRQ(int irq, struct pt_regs *regs,
 		struct thread_info *tp);
 
-#define __ARCH_HAS_DO_SOFTIRQ
-
 #else
 #define irq_ctx_init()
 
diff --git a/include/asm-powerpc/paca.h b/include/asm-powerpc/paca.h
index ec94b51074fc..4465b95ebef0 100644
--- a/include/asm-powerpc/paca.h
+++ b/include/asm-powerpc/paca.h
@@ -96,6 +96,11 @@ struct paca_struct {
 	u64 saved_r1;			/* r1 save for RTAS calls */
 	u64 saved_msr;			/* MSR saved here by enter_rtas */
 	u8 proc_enabled;		/* irq soft-enable flag */
+
+	/* Stuff for accurate time accounting */
+	u64 user_time;			/* accumulated usermode TB ticks */
+	u64 system_time;		/* accumulated system TB ticks */
+	u64 startpurr;			/* PURR/TB value snapshot */
 };
 
 extern struct paca_struct paca[];
diff --git a/include/asm-powerpc/ppc_asm.h b/include/asm-powerpc/ppc_asm.h
index ab8688d39024..dd1c0a913d5f 100644
--- a/include/asm-powerpc/ppc_asm.h
+++ b/include/asm-powerpc/ppc_asm.h
@@ -14,6 +14,48 @@
 
 #define SZL			(BITS_PER_LONG/8)
 
+/*
+ * Stuff for accurate CPU time accounting.
+ * These macros handle transitions between user and system state
+ * in exception entry and exit and accumulate time to the
+ * user_time and system_time fields in the paca.
+ */
+
+#ifndef CONFIG_VIRT_CPU_ACCOUNTING
+#define ACCOUNT_CPU_USER_ENTRY(ra, rb)
+#define ACCOUNT_CPU_USER_EXIT(ra, rb)
+#else
+#define ACCOUNT_CPU_USER_ENTRY(ra, rb)					\
+	beq	2f;			/* if from kernel mode */	\
+BEGIN_FTR_SECTION;							\
+	mfspr	ra,SPRN_PURR;		/* get processor util. reg */	\
+END_FTR_SECTION_IFSET(CPU_FTR_PURR);					\
+BEGIN_FTR_SECTION;							\
+	mftb	ra;			/* or get TB if no PURR */	\
+END_FTR_SECTION_IFCLR(CPU_FTR_PURR);					\
+	ld	rb,PACA_STARTPURR(r13);				\
+	std	ra,PACA_STARTPURR(r13);					\
+	subf	rb,rb,ra;		/* subtract start value */	\
+	ld	ra,PACA_USER_TIME(r13);					\
+	add	ra,ra,rb;		/* add on to user time */	\
+	std	ra,PACA_USER_TIME(r13);					\
+2:
+
+#define ACCOUNT_CPU_USER_EXIT(ra, rb)					\
+BEGIN_FTR_SECTION;							\
+	mfspr	ra,SPRN_PURR;		/* get processor util. reg */	\
+END_FTR_SECTION_IFSET(CPU_FTR_PURR);					\
+BEGIN_FTR_SECTION;							\
+	mftb	ra;			/* or get TB if no PURR */	\
+END_FTR_SECTION_IFCLR(CPU_FTR_PURR);					\
+	ld	rb,PACA_STARTPURR(r13);				\
+	std	ra,PACA_STARTPURR(r13);					\
+	subf	rb,rb,ra;		/* subtract start value */	\
+	ld	ra,PACA_SYSTEM_TIME(r13);				\
+	add	ra,ra,rb;		/* add on to user time */	\
+	std	ra,PACA_SYSTEM_TIME(r13);
+#endif
+
 /*
  * Macros for storing registers into and loading registers from
  * exception frames.
diff --git a/include/asm-powerpc/system.h b/include/asm-powerpc/system.h
index d9bf53653b10..41b7a5b3d701 100644
--- a/include/asm-powerpc/system.h
+++ b/include/asm-powerpc/system.h
@@ -424,5 +424,9 @@ static inline void create_function_call(unsigned long addr, void * func)
 	create_branch(addr, func_addr, BRANCH_SET_LINK);
 }
 
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING
+extern void account_system_vtime(struct task_struct *);
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_SYSTEM_H */
diff --git a/include/asm-powerpc/time.h b/include/asm-powerpc/time.h
index baddc9ab57ad..912118db13ae 100644
--- a/include/asm-powerpc/time.h
+++ b/include/asm-powerpc/time.h
@@ -41,6 +41,7 @@ extern time_t last_rtc_update;
 
 extern void generic_calibrate_decr(void);
 extern void wakeup_decrementer(void);
+extern void snapshot_timebase(void);
 
 /* Some sane defaults: 125 MHz timebase, 1GHz processor */
 extern unsigned long ppc_proc_freq;
@@ -221,5 +222,19 @@ struct cpu_usage {
 
 DECLARE_PER_CPU(struct cpu_usage, cpu_usage_array);
 
+#ifdef CONFIG_VIRT_CPU_ACCOUNTING
+extern void account_process_vtime(struct task_struct *tsk);
+#else
+#define account_process_vtime(tsk)		do { } while (0)
+#endif
+
+#if defined(CONFIG_VIRT_CPU_ACCOUNTING) && defined(CONFIG_PPC_SPLPAR)
+extern void calculate_steal_time(void);
+extern void snapshot_timebases(void);
+#else
+#define calculate_steal_time()			do { } while (0)
+#define snapshot_timebases()			do { } while (0)
+#endif
+
 #endif /* __KERNEL__ */
 #endif /* __PPC64_TIME_H */
diff --git a/include/asm-ppc/time.h b/include/asm-ppc/time.h
index 321fb75b5f22..c86112323c9f 100644
--- a/include/asm-ppc/time.h
+++ b/include/asm-ppc/time.h
@@ -153,5 +153,10 @@ extern __inline__ unsigned binary_tbl(void) {
 ({unsigned z; asm ("mulhwu %0,%1,%2" : "=r" (z) : "r" (x), "r" (y)); z;})
 
 unsigned mulhwu_scale_factor(unsigned, unsigned);
+
+#define account_process_vtime(tsk)		do { } while (0)
+#define calculate_steal_time()			do { } while (0)
+#define snapshot_timebases()			do { } while (0)
+
 #endif /* __ASM_TIME_H__ */
 #endif /* __KERNEL__ */

From 20f4eb3e502d68b12224577ebcd2cd50cc6e14e4 Mon Sep 17 00:00:00 2001
From: David Gibson <david@gibson.dropbear.id.au>
Date: Mon, 20 Feb 2006 14:05:56 +1100
Subject: [PATCH 422/509] [PATCH] powerpc: Fixup for STRICT_MM_TYPECHECKS

Currently ARCH=powerpc will not compile when STRICT_MM_TYPECHECKS is
turned on and CONFIG_64K_PAGES is turned off.  This corrects the
problem.

Signed-off-by: David Gibson <dwg@au1.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/pgtable-4k.h | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/include/asm-powerpc/pgtable-4k.h b/include/asm-powerpc/pgtable-4k.h
index e9590c06ad92..35f92813464c 100644
--- a/include/asm-powerpc/pgtable-4k.h
+++ b/include/asm-powerpc/pgtable-4k.h
@@ -62,9 +62,14 @@
 /* shift to put page number into pte */
 #define PTE_RPN_SHIFT	(17)
 
-#define __real_pte(e,p)		((real_pte_t)(e))
-#define __rpte_to_pte(r)	(r)
-#define __rpte_to_hidx(r,index)	(pte_val((r)) >> 12)
+#ifdef STRICT_MM_TYPECHECKS
+#define __real_pte(e,p)		((real_pte_t){(e)})
+#define __rpte_to_pte(r)	((r).pte)
+#else
+#define __real_pte(e,p)		(e)
+#define __rpte_to_pte(r)	(__pte(r))
+#endif
+#define __rpte_to_hidx(r,index)	(pte_val(__rpte_to_pte(r)) >> 12)
 
 #define pte_iterate_hashed_subpages(rpte, psize, va, index, shift)       \
 	do {							         \

From 4f629d7db32decbadaab2abfa4d021fee94990ef Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 20 Feb 2006 10:40:28 +0100
Subject: [PATCH 423/509] [PATCH] powerpc: newline for ISYNC_ON_SMP

Add a newline at the end of the ISYNC_ON_SMP string.
Needed for a subsequent patch.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/synch.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-powerpc/synch.h b/include/asm-powerpc/synch.h
index c90d9d9aae72..2cda3c38a9fa 100644
--- a/include/asm-powerpc/synch.h
+++ b/include/asm-powerpc/synch.h
@@ -15,7 +15,7 @@
 #endif
 
 #ifdef CONFIG_SMP
-#define ISYNC_ON_SMP	"\n\tisync"
+#define ISYNC_ON_SMP	"\n\tisync\n"
 #define LWSYNC_ON_SMP	__stringify(LWSYNC) "\n"
 #else
 #define ISYNC_ON_SMP

From f055affb89f587a03f3411c3fd49ef31295c3d48 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Mon, 20 Feb 2006 10:41:40 +0100
Subject: [PATCH 424/509] [PATCH] powerpc: native atomic_add_unless

Do atomic_add_unless natively instead of using cmpxchg.
Improved register allocation idea from Joel Schopp.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 include/asm-powerpc/atomic.h | 38 +++++++++++++++++++++++-------------
 1 file changed, 24 insertions(+), 14 deletions(-)

diff --git a/include/asm-powerpc/atomic.h b/include/asm-powerpc/atomic.h
index 147a38dcc766..bb3c0ab7e667 100644
--- a/include/asm-powerpc/atomic.h
+++ b/include/asm-powerpc/atomic.h
@@ -8,6 +8,7 @@
 typedef struct { volatile int counter; } atomic_t;
 
 #ifdef __KERNEL__
+#include <linux/compiler.h>
 #include <asm/synch.h>
 #include <asm/asm-compat.h>
 
@@ -176,20 +177,29 @@ static __inline__ int atomic_dec_return(atomic_t *v)
  * Atomically adds @a to @v, so long as it was not @u.
  * Returns non-zero if @v was not @u, and zero otherwise.
  */
-#define atomic_add_unless(v, a, u)			\
-({							\
-	int c, old;					\
-	c = atomic_read(v);				\
-	for (;;) {					\
-		if (unlikely(c == (u)))			\
-			break;				\
-		old = atomic_cmpxchg((v), c, c + (a));	\
-		if (likely(old == c))			\
-			break;				\
-		c = old;				\
-	}						\
-	c != (u);					\
-})
+static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
+{
+	int t;
+
+	__asm__ __volatile__ (
+	LWSYNC_ON_SMP
+"1:	lwarx	%0,0,%1		# atomic_add_unless\n\
+	cmpw	0,%0,%3 \n\
+	beq-	2f \n\
+	add	%0,%2,%0 \n"
+	PPC405_ERR77(0,%2)
+"	stwcx.	%0,0,%1 \n\
+	bne-	1b \n"
+	ISYNC_ON_SMP
+"	subf	%0,%2,%0 \n\
+2:"
+	: "=&r" (t)
+	: "r" (&v->counter), "r" (a), "r" (u)
+	: "cc", "memory");
+
+	return t != u;
+}
+
 #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
 
 #define atomic_sub_and_test(a, v)	(atomic_sub_return((a), (v)) == 0)

From ad9f6713ae59f319ed676c2d014a7756b62f1c51 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@stusta.de>
Date: Sun, 5 Feb 2006 00:37:47 +0100
Subject: [PATCH 425/509] [PATCH] drivers/net/tlan.c: #ifdef CONFIG_PCI the PCI
 specific code

drivers/net/tlan.c compiles with CONFIG_PCI=n only with a warning and
due to the dead code elimination of gcc.

Additionally, this fixes the only compile error I found with
CONFIG_PCI=n and the gcc -Werror-implicit-function-declaration
flag on i386.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/net/tlan.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/tlan.c b/drivers/net/tlan.c
index c2506b56a186..12076f8f942c 100644
--- a/drivers/net/tlan.c
+++ b/drivers/net/tlan.c
@@ -536,6 +536,7 @@ static int __devinit TLan_probe1(struct pci_dev *pdev,
 	u16		   device_id;
 	int		   reg, rc = -ENODEV;
 
+#ifdef CONFIG_PCI
 	if (pdev) {
 		rc = pci_enable_device(pdev);
 		if (rc)
@@ -547,6 +548,7 @@ static int __devinit TLan_probe1(struct pci_dev *pdev,
 			goto err_out;
 		}
 	}
+#endif  /*  CONFIG_PCI  */
 
 	dev = alloc_etherdev(sizeof(TLanPrivateInfo));
 	if (dev == NULL) {

From 3672b638ec1d5b1020ea27986060b830f09c96c1 Mon Sep 17 00:00:00 2001
From: Anton Altaparmakov <aia21@cantab.net>
Date: Fri, 24 Feb 2006 09:55:07 +0000
Subject: [PATCH 426/509] NTFS: - Cope with attribute list attribute having
 invalid flags. 	Windows copes with this and even chkdsk does not
 detect or fix this 	so we have to cope with it, too.  Thanks to Pawel Kot
 for reporting 	the problem.       - Miscellaneous updates to layout.h.

Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
---
 fs/ntfs/ChangeLog | 28 ++++++++++++++++++---------
 fs/ntfs/inode.c   | 49 +++++++++++++++++++++++++++++++++++++----------
 fs/ntfs/layout.h  | 25 +++++++++++++++++-------
 3 files changed, 76 insertions(+), 26 deletions(-)

diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog
index 02f44094bda9..4a62201e8441 100644
--- a/fs/ntfs/ChangeLog
+++ b/fs/ntfs/ChangeLog
@@ -1,9 +1,9 @@
 ToDo/Notes:
 	- Find and fix bugs.
 	- The only places in the kernel where a file is resized are
-	  ntfs_file_write*() and ntfs_truncate() for both of which i_sem is
+	  ntfs_file_write*() and ntfs_truncate() for both of which i_mutex is
 	  held.  Just have to be careful in read-/writepage and other helpers
-	  not running under i_sem that we play nice...  Also need to be careful
+	  not running under i_mutex that we play nice.  Also need to be careful
 	  with initialized_size extension in ntfs_file_write*() and writepage.
 	  UPDATE: The only things that need to be checked are the compressed
 	  write and the other attribute resize/write cases like index
@@ -19,6 +19,16 @@ ToDo/Notes:
 	- Enable the code for setting the NT4 compatibility flag when we start
 	  making NTFS 1.2 specific modifications.
 
+2.1.26 - Minor bug fixes and updates.
+
+	- We have struct kmem_cache now so use it instead of the typedef
+	  kmem_cache_t.  (Pekka Enberg)
+	- Miscellaneous updates to layout.h.
+	- Cope with attribute list attribute having invalid flags.  Windows
+	  copes with this and even chkdsk does not detect or fix this so we
+	  have to cope with it, too.  Thanks to Pawel Kot for reporting the
+	  problem.
+
 2.1.25 - (Almost) fully implement write(2) and truncate(2).
 
 	- Change ntfs_map_runlist_nolock(), ntfs_attr_find_vcn_nolock() and
@@ -373,7 +383,7 @@ ToDo/Notes:
 	  single one of them had an mst error.  (Thanks to Ken MacFerrin for
 	  the bug report.)
 	- Fix error handling in fs/ntfs/quota.c::ntfs_mark_quotas_out_of_date()
-	  where we failed to release i_sem on the $Quota/$Q attribute inode.
+	  where we failed to release i_mutex on the $Quota/$Q attribute inode.
 	- Fix bug in handling of bad inodes in fs/ntfs/namei.c::ntfs_lookup().
 	- Add mapping of unmapped buffers to all remaining code paths, i.e.
 	  fs/ntfs/aops.c::ntfs_write_mst_block(), mft.c::ntfs_sync_mft_mirror(),
@@ -874,7 +884,7 @@ ToDo/Notes:
 	  clusters. (Philipp Thomas)
 	- attrib.c::load_attribute_list(): Fix bug when initialized_size is a
 	  multiple of the block_size but not the cluster size. (Szabolcs
-	  Szakacsits <szaka@sienet.hu>)
+	  Szakacsits)
 
 2.1.2 - Important bug fixes aleviating the hangs in statfs.
 
@@ -884,7 +894,7 @@ ToDo/Notes:
 
 	- Add handling for initialized_size != data_size in compressed files.
 	- Reduce function local stack usage from 0x3d4 bytes to just noise in
-	  fs/ntfs/upcase.c. (Randy Dunlap <rdunlap@xenotime.net>)
+	  fs/ntfs/upcase.c. (Randy Dunlap)
 	- Remove compiler warnings for newer gcc.
 	- Pages are no longer kmapped by mm/filemap.c::generic_file_write()
 	  around calls to ->{prepare,commit}_write.  Adapt NTFS appropriately
@@ -1201,11 +1211,11 @@ ToDo/Notes:
 	  the kernel. We probably want a kernel generic init_address_space()
 	  function...
 	- Drop BKL from ntfs_readdir() after consultation with Al Viro. The
-	  only caller of ->readdir() is vfs_readdir() which holds i_sem during
-	  the call, and i_sem is sufficient protection against changes in the
-	  directory inode (including ->i_size).
+	  only caller of ->readdir() is vfs_readdir() which holds i_mutex
+	  during the call, and i_mutex is sufficient protection against changes
+	  in the directory inode (including ->i_size).
 	- Use generic_file_llseek() for directories (as opposed to
-	  default_llseek()) as this downs i_sem instead of the BKL which is
+	  default_llseek()) as this downs i_mutex instead of the BKL which is
 	  what we now need for exclusion against ->f_pos changes considering we
 	  no longer take the BKL in ntfs_readdir().
 
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index ea1bd3feea1b..55263b7de9c0 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -677,13 +677,28 @@ static int ntfs_read_locked_inode(struct inode *vi)
 		ntfs_debug("Attribute list found in inode 0x%lx.", vi->i_ino);
 		NInoSetAttrList(ni);
 		a = ctx->attr;
-		if (a->flags & ATTR_IS_ENCRYPTED ||
-				a->flags & ATTR_COMPRESSION_MASK ||
-				a->flags & ATTR_IS_SPARSE) {
+		if (a->flags & ATTR_COMPRESSION_MASK) {
 			ntfs_error(vi->i_sb, "Attribute list attribute is "
-					"compressed/encrypted/sparse.");
+					"compressed.");
 			goto unm_err_out;
 		}
+		if (a->flags & ATTR_IS_ENCRYPTED ||
+				a->flags & ATTR_IS_SPARSE) {
+			if (a->non_resident) {
+				ntfs_error(vi->i_sb, "Non-resident attribute "
+						"list attribute is encrypted/"
+						"sparse.");
+				goto unm_err_out;
+			}
+			ntfs_warning(vi->i_sb, "Resident attribute list "
+					"attribute in inode 0x%lx is marked "
+					"encrypted/sparse which is not true.  "
+					"However, Windows allows this and "
+					"chkdsk does not detect or correct it "
+					"so we will just ignore the invalid "
+					"flags and pretend they are not set.",
+					vi->i_ino);
+		}
 		/* Now allocate memory for the attribute list. */
 		ni->attr_list_size = (u32)ntfs_attr_size(a);
 		ni->attr_list = ntfs_malloc_nofs(ni->attr_list_size);
@@ -1809,19 +1824,33 @@ int ntfs_read_inode_mount(struct inode *vi)
 	} else /* if (!err) */ {
 		ATTR_LIST_ENTRY *al_entry, *next_al_entry;
 		u8 *al_end;
+		static const char *es = "  Not allowed.  $MFT is corrupt.  "
+				"You should run chkdsk.";
 
 		ntfs_debug("Attribute list attribute found in $MFT.");
 		NInoSetAttrList(ni);
 		a = ctx->attr;
-		if (a->flags & ATTR_IS_ENCRYPTED ||
-				a->flags & ATTR_COMPRESSION_MASK ||
-				a->flags & ATTR_IS_SPARSE) {
+		if (a->flags & ATTR_COMPRESSION_MASK) {
 			ntfs_error(sb, "Attribute list attribute is "
-					"compressed/encrypted/sparse. Not "
-					"allowed. $MFT is corrupt. You should "
-					"run chkdsk.");
+					"compressed.%s", es);
 			goto put_err_out;
 		}
+		if (a->flags & ATTR_IS_ENCRYPTED ||
+				a->flags & ATTR_IS_SPARSE) {
+			if (a->non_resident) {
+				ntfs_error(sb, "Non-resident attribute list "
+						"attribute is encrypted/"
+						"sparse.%s", es);
+				goto put_err_out;
+			}
+			ntfs_warning(sb, "Resident attribute list attribute "
+					"in $MFT system file is marked "
+					"encrypted/sparse which is not true.  "
+					"However, Windows allows this and "
+					"chkdsk does not detect or correct it "
+					"so we will just ignore the invalid "
+					"flags and pretend they are not set.");
+		}
 		/* Now allocate memory for the attribute list. */
 		ni->attr_list_size = (u32)ntfs_attr_size(a);
 		ni->attr_list = ntfs_malloc_nofs(ni->attr_list_size);
diff --git a/fs/ntfs/layout.h b/fs/ntfs/layout.h
index f5678d5d7919..bb408d4dcbb0 100644
--- a/fs/ntfs/layout.h
+++ b/fs/ntfs/layout.h
@@ -838,15 +838,19 @@ enum {
 	   F_A_DEVICE, F_A_DIRECTORY, F_A_SPARSE_FILE, F_A_REPARSE_POINT,
 	   F_A_COMPRESSED, and F_A_ENCRYPTED and preserves the rest.  This mask
 	   is used to to obtain all flags that are valid for setting. */
-
 	/*
-	 * The following flags are only present in the FILE_NAME attribute (in
+	 * The following flag is only present in the FILE_NAME attribute (in
 	 * the field file_attributes).
 	 */
 	FILE_ATTR_DUP_FILE_NAME_INDEX_PRESENT	= const_cpu_to_le32(0x10000000),
 	/* Note, this is a copy of the corresponding bit from the mft record,
 	   telling us whether this is a directory or not, i.e. whether it has
 	   an index root attribute or not. */
+	/*
+	 * The following flag is present both in the STANDARD_INFORMATION
+	 * attribute and in the FILE_NAME attribute (in the field
+	 * file_attributes).
+	 */
 	FILE_ATTR_DUP_VIEW_INDEX_PRESENT	= const_cpu_to_le32(0x20000000),
 	/* Note, this is a copy of the corresponding bit from the mft record,
 	   telling us whether this file has a view index present (eg. object id
@@ -1071,9 +1075,15 @@ typedef struct {
 					   modified. */
 /* 20*/	sle64 last_access_time;		/* Time this mft record was last
 					   accessed. */
-/* 28*/	sle64 allocated_size;		/* Byte size of allocated space for the
-					   data attribute. NOTE: Is a multiple
-					   of the cluster size. */
+/* 28*/	sle64 allocated_size;		/* Byte size of on-disk allocated space
+					   for the data attribute.  So for
+					   normal $DATA, this is the
+					   allocated_size from the unnamed
+					   $DATA attribute and for compressed
+					   and/or sparse $DATA, this is the
+					   compressed_size from the unnamed
+					   $DATA attribute.  NOTE: This is a
+					   multiple of the cluster size. */
 /* 30*/	sle64 data_size;		/* Byte size of actual data in data
 					   attribute. */
 /* 38*/	FILE_ATTR_FLAGS file_attributes;	/* Flags describing the file. */
@@ -1904,12 +1914,13 @@ enum {
 	VOLUME_DELETE_USN_UNDERWAY	= const_cpu_to_le16(0x0010),
 	VOLUME_REPAIR_OBJECT_ID		= const_cpu_to_le16(0x0020),
 
+	VOLUME_CHKDSK_UNDERWAY		= const_cpu_to_le16(0x4000),
 	VOLUME_MODIFIED_BY_CHKDSK	= const_cpu_to_le16(0x8000),
 
-	VOLUME_FLAGS_MASK		= const_cpu_to_le16(0x803f),
+	VOLUME_FLAGS_MASK		= const_cpu_to_le16(0xc03f),
 
 	/* To make our life easier when checking if we must mount read-only. */
-	VOLUME_MUST_MOUNT_RO_MASK	= const_cpu_to_le16(0x8027),
+	VOLUME_MUST_MOUNT_RO_MASK	= const_cpu_to_le16(0xc027),
 } __attribute__ ((__packed__));
 
 typedef le16 VOLUME_FLAGS;

From 78af34f03d33d2ba179c9d35685860170b94a285 Mon Sep 17 00:00:00 2001
From: Anton Altaparmakov <aia21@cantab.net>
Date: Fri, 24 Feb 2006 10:32:33 +0000
Subject: [PATCH 427/509] NTFS: Implement support for sector sizes above 512
 bytes (up to the maximum       supported by NTFS which is 4096 bytes).

---
 fs/ntfs/ChangeLog |   6 ++
 fs/ntfs/aops.c    |  18 +++---
 fs/ntfs/file.c    |   8 +--
 fs/ntfs/mft.c     |   8 +--
 fs/ntfs/super.c   | 151 ++++++++++++++++++++++++++++++----------------
 5 files changed, 121 insertions(+), 70 deletions(-)

diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog
index 4a62201e8441..e66b4ac2fade 100644
--- a/fs/ntfs/ChangeLog
+++ b/fs/ntfs/ChangeLog
@@ -21,8 +21,14 @@ ToDo/Notes:
 
 2.1.26 - Minor bug fixes and updates.
 
+	- Fix a potential overflow in file.c where a cast to s64 was missing in
+	  a left shift of a page index.
+	- The struct inode has had its i_sem semaphore changed to a mutex named
+	  i_mutex.
 	- We have struct kmem_cache now so use it instead of the typedef
 	  kmem_cache_t.  (Pekka Enberg)
+	- Implement support for sector sizes above 512 bytes (up to the maximum
+	  supported by NTFS which is 4096 bytes).
 	- Miscellaneous updates to layout.h.
 	- Cope with attribute list attribute having invalid flags.  Windows
 	  copes with this and even chkdsk does not detect or fix this so we
diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
index 1c0a4315876a..7e361da770b3 100644
--- a/fs/ntfs/aops.c
+++ b/fs/ntfs/aops.c
@@ -2,7 +2,7 @@
  * aops.c - NTFS kernel address space operations and page cache handling.
  *	    Part of the Linux-NTFS project.
  *
- * Copyright (c) 2001-2005 Anton Altaparmakov
+ * Copyright (c) 2001-2006 Anton Altaparmakov
  * Copyright (c) 2002 Richard Russon
  *
  * This program/include file is free software; you can redistribute it and/or
@@ -200,8 +200,8 @@ static int ntfs_read_block(struct page *page)
 	/* $MFT/$DATA must have its complete runlist in memory at all times. */
 	BUG_ON(!ni->runlist.rl && !ni->mft_no && !NInoAttr(ni));
 
-	blocksize_bits = VFS_I(ni)->i_blkbits;
-	blocksize = 1 << blocksize_bits;
+	blocksize = vol->sb->s_blocksize;
+	blocksize_bits = vol->sb->s_blocksize_bits;
 
 	if (!page_has_buffers(page)) {
 		create_empty_buffers(page, blocksize, 0);
@@ -569,10 +569,8 @@ static int ntfs_write_block(struct page *page, struct writeback_control *wbc)
 
 	BUG_ON(!NInoNonResident(ni));
 	BUG_ON(NInoMstProtected(ni));
-
-	blocksize_bits = vi->i_blkbits;
-	blocksize = 1 << blocksize_bits;
-
+	blocksize = vol->sb->s_blocksize;
+	blocksize_bits = vol->sb->s_blocksize_bits;
 	if (!page_has_buffers(page)) {
 		BUG_ON(!PageUptodate(page));
 		create_empty_buffers(page, blocksize,
@@ -949,8 +947,8 @@ static int ntfs_write_mst_block(struct page *page,
 	 */
 	BUG_ON(!(is_mft || S_ISDIR(vi->i_mode) ||
 			(NInoAttr(ni) && ni->type == AT_INDEX_ALLOCATION)));
-	bh_size_bits = vi->i_blkbits;
-	bh_size = 1 << bh_size_bits;
+	bh_size = vol->sb->s_blocksize;
+	bh_size_bits = vol->sb->s_blocksize_bits;
 	max_bhs = PAGE_CACHE_SIZE / bh_size;
 	BUG_ON(!max_bhs);
 	BUG_ON(max_bhs > MAX_BUF_PER_PAGE);
@@ -1596,7 +1594,7 @@ void mark_ntfs_record_dirty(struct page *page, const unsigned int ofs) {
 
 	BUG_ON(!PageUptodate(page));
 	end = ofs + ni->itype.index.block_size;
-	bh_size = 1 << VFS_I(ni)->i_blkbits;
+	bh_size = VFS_I(ni)->i_sb->s_blocksize;
 	spin_lock(&mapping->private_lock);
 	if (unlikely(!page_has_buffers(page))) {
 		spin_unlock(&mapping->private_lock);
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index 3a119a87686a..5027d3d1b3fe 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -1,7 +1,7 @@
 /*
  * file.c - NTFS kernel file operations.  Part of the Linux-NTFS project.
  *
- * Copyright (c) 2001-2005 Anton Altaparmakov
+ * Copyright (c) 2001-2006 Anton Altaparmakov
  *
  * This program/include file is free software; you can redistribute it and/or
  * modify it under the terms of the GNU General Public License as published
@@ -529,8 +529,8 @@ static int ntfs_prepare_pages_for_non_resident_write(struct page **pages,
 			"index 0x%lx, nr_pages 0x%x, pos 0x%llx, bytes 0x%zx.",
 			vi->i_ino, ni->type, pages[0]->index, nr_pages,
 			(long long)pos, bytes);
-	blocksize_bits = vi->i_blkbits;
-	blocksize = 1 << blocksize_bits;
+	blocksize = vol->sb->s_blocksize;
+	blocksize_bits = vol->sb->s_blocksize_bits;
 	u = 0;
 	do {
 		struct page *page = pages[u];
@@ -1525,7 +1525,7 @@ static inline int ntfs_commit_pages_after_non_resident_write(
 
 	vi = pages[0]->mapping->host;
 	ni = NTFS_I(vi);
-	blocksize = 1 << vi->i_blkbits;
+	blocksize = vi->i_sb->s_blocksize;
 	end = pos + bytes;
 	u = 0;
 	do {
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index 0c65cbb8c5cf..6499aafc2258 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -1,7 +1,7 @@
 /**
  * mft.c - NTFS kernel mft record operations. Part of the Linux-NTFS project.
  *
- * Copyright (c) 2001-2005 Anton Altaparmakov
+ * Copyright (c) 2001-2006 Anton Altaparmakov
  * Copyright (c) 2002 Richard Russon
  *
  * This program/include file is free software; you can redistribute it and/or
@@ -473,7 +473,7 @@ int ntfs_sync_mft_mirror(ntfs_volume *vol, const unsigned long mft_no,
 	runlist_element *rl;
 	unsigned int block_start, block_end, m_start, m_end, page_ofs;
 	int i_bhs, nr_bhs, err = 0;
-	unsigned char blocksize_bits = vol->mftmirr_ino->i_blkbits;
+	unsigned char blocksize_bits = vol->sb->s_blocksize_bits;
 
 	ntfs_debug("Entering for inode 0x%lx.", mft_no);
 	BUG_ON(!max_bhs);
@@ -672,8 +672,8 @@ int write_mft_record_nolock(ntfs_inode *ni, MFT_RECORD *m, int sync)
 {
 	ntfs_volume *vol = ni->vol;
 	struct page *page = ni->page;
-	unsigned char blocksize_bits = vol->mft_ino->i_blkbits;
-	unsigned int blocksize = 1 << blocksize_bits;
+	unsigned int blocksize = vol->sb->s_blocksize;
+	unsigned char blocksize_bits = vol->sb->s_blocksize_bits;
 	int max_bhs = vol->mft_record_size / blocksize;
 	struct buffer_head *bhs[max_bhs];
 	struct buffer_head *bh, *head;
diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c
index e9c0d80dfab1..489f7049146b 100644
--- a/fs/ntfs/super.c
+++ b/fs/ntfs/super.c
@@ -1,7 +1,7 @@
 /*
  * super.c - NTFS kernel super block handling. Part of the Linux-NTFS project.
  *
- * Copyright (c) 2001-2005 Anton Altaparmakov
+ * Copyright (c) 2001-2006 Anton Altaparmakov
  * Copyright (c) 2001,2002 Richard Russon
  *
  * This program/include file is free software; you can redistribute it and/or
@@ -22,6 +22,7 @@
 
 #include <linux/stddef.h>
 #include <linux/init.h>
+#include <linux/slab.h>
 #include <linux/string.h>
 #include <linux/spinlock.h>
 #include <linux/blkdev.h>	/* For bdev_hardsect_size(). */
@@ -641,7 +642,7 @@ static struct buffer_head *read_ntfs_boot_sector(struct super_block *sb,
 {
 	const char *read_err_str = "Unable to read %s boot sector.";
 	struct buffer_head *bh_primary, *bh_backup;
-	long nr_blocks = NTFS_SB(sb)->nr_blocks;
+	sector_t nr_blocks = NTFS_SB(sb)->nr_blocks;
 
 	/* Try to read primary boot sector. */
 	if ((bh_primary = sb_bread(sb, 0))) {
@@ -688,13 +689,18 @@ hotfix_primary_boot_sector:
 		/*
 		 * If we managed to read sector zero and the volume is not
 		 * read-only, copy the found, valid backup boot sector to the
-		 * primary boot sector.
+		 * primary boot sector.  Note we only copy the actual boot
+		 * sector structure, not the actual whole device sector as that
+		 * may be bigger and would potentially damage the $Boot system
+		 * file (FIXME: Would be nice to know if the backup boot sector
+		 * on a large sector device contains the whole boot loader or
+		 * just the first 512 bytes).
 		 */
 		if (!(sb->s_flags & MS_RDONLY)) {
 			ntfs_warning(sb, "Hot-fix: Recovering invalid primary "
 					"boot sector from backup copy.");
 			memcpy(bh_primary->b_data, bh_backup->b_data,
-					sb->s_blocksize);
+					NTFS_BLOCK_SIZE);
 			mark_buffer_dirty(bh_primary);
 			sync_dirty_buffer(bh_primary);
 			if (buffer_uptodate(bh_primary)) {
@@ -733,9 +739,13 @@ static BOOL parse_ntfs_boot_sector(ntfs_volume *vol, const NTFS_BOOT_SECTOR *b)
 			vol->sector_size);
 	ntfs_debug("vol->sector_size_bits = %i (0x%x)", vol->sector_size_bits,
 			vol->sector_size_bits);
-	if (vol->sector_size != vol->sb->s_blocksize)
-		ntfs_warning(vol->sb, "The boot sector indicates a sector size "
-				"different from the device sector size.");
+	if (vol->sector_size < vol->sb->s_blocksize) {
+		ntfs_error(vol->sb, "Sector size (%i) is smaller than the "
+				"device block size (%lu).  This is not "
+				"supported.  Sorry.", vol->sector_size,
+				vol->sb->s_blocksize);
+		return FALSE;
+	}
 	ntfs_debug("sectors_per_cluster = 0x%x", b->bpb.sectors_per_cluster);
 	sectors_per_cluster_bits = ffs(b->bpb.sectors_per_cluster) - 1;
 	ntfs_debug("sectors_per_cluster_bits = 0x%x",
@@ -748,16 +758,11 @@ static BOOL parse_ntfs_boot_sector(ntfs_volume *vol, const NTFS_BOOT_SECTOR *b)
 	ntfs_debug("vol->cluster_size = %i (0x%x)", vol->cluster_size,
 			vol->cluster_size);
 	ntfs_debug("vol->cluster_size_mask = 0x%x", vol->cluster_size_mask);
-	ntfs_debug("vol->cluster_size_bits = %i (0x%x)",
-			vol->cluster_size_bits, vol->cluster_size_bits);
-	if (vol->sector_size > vol->cluster_size) {
-		ntfs_error(vol->sb, "Sector sizes above the cluster size are "
-				"not supported.  Sorry.");
-		return FALSE;
-	}
-	if (vol->sb->s_blocksize > vol->cluster_size) {
-		ntfs_error(vol->sb, "Cluster sizes smaller than the device "
-				"sector size are not supported.  Sorry.");
+	ntfs_debug("vol->cluster_size_bits = %i", vol->cluster_size_bits);
+	if (vol->cluster_size < vol->sector_size) {
+		ntfs_error(vol->sb, "Cluster size (%i) is smaller than the "
+				"sector size (%i).  This is not supported.  "
+				"Sorry.", vol->cluster_size, vol->sector_size);
 		return FALSE;
 	}
 	clusters_per_mft_record = b->clusters_per_mft_record;
@@ -786,11 +791,18 @@ static BOOL parse_ntfs_boot_sector(ntfs_volume *vol, const NTFS_BOOT_SECTOR *b)
 	 * we store $MFT/$DATA, the table of mft records in the page cache.
 	 */
 	if (vol->mft_record_size > PAGE_CACHE_SIZE) {
-		ntfs_error(vol->sb, "Mft record size %i (0x%x) exceeds the "
-				"page cache size on your system %lu (0x%lx).  "
+		ntfs_error(vol->sb, "Mft record size (%i) exceeds the "
+				"PAGE_CACHE_SIZE on your system (%lu).  "
 				"This is not supported.  Sorry.",
-				vol->mft_record_size, vol->mft_record_size,
-				PAGE_CACHE_SIZE, PAGE_CACHE_SIZE);
+				vol->mft_record_size, PAGE_CACHE_SIZE);
+		return FALSE;
+	}
+	/* We cannot support mft record sizes below the sector size. */
+	if (vol->mft_record_size < vol->sector_size) {
+		ntfs_error(vol->sb, "Mft record size (%i) is smaller than the "
+				"sector size (%i).  This is not supported.  "
+				"Sorry.", vol->mft_record_size,
+				vol->sector_size);
 		return FALSE;
 	}
 	clusters_per_index_record = b->clusters_per_index_record;
@@ -816,6 +828,14 @@ static BOOL parse_ntfs_boot_sector(ntfs_volume *vol, const NTFS_BOOT_SECTOR *b)
 	ntfs_debug("vol->index_record_size_bits = %i (0x%x)",
 			vol->index_record_size_bits,
 			vol->index_record_size_bits);
+	/* We cannot support index record sizes below the sector size. */
+	if (vol->index_record_size < vol->sector_size) {
+		ntfs_error(vol->sb, "Index record size (%i) is smaller than "
+				"the sector size (%i).  This is not "
+				"supported.  Sorry.", vol->index_record_size,
+				vol->sector_size);
+		return FALSE;
+	}
 	/*
 	 * Get the size of the volume in clusters and check for 64-bit-ness.
 	 * Windows currently only uses 32 bits to save the clusters so we do
@@ -845,15 +865,18 @@ static BOOL parse_ntfs_boot_sector(ntfs_volume *vol, const NTFS_BOOT_SECTOR *b)
 	}
 	ll = sle64_to_cpu(b->mft_lcn);
 	if (ll >= vol->nr_clusters) {
-		ntfs_error(vol->sb, "MFT LCN is beyond end of volume.  Weird.");
+		ntfs_error(vol->sb, "MFT LCN (%lli, 0x%llx) is beyond end of "
+				"volume.  Weird.", (unsigned long long)ll,
+				(unsigned long long)ll);
 		return FALSE;
 	}
 	vol->mft_lcn = ll;
 	ntfs_debug("vol->mft_lcn = 0x%llx", (long long)vol->mft_lcn);
 	ll = sle64_to_cpu(b->mftmirr_lcn);
 	if (ll >= vol->nr_clusters) {
-		ntfs_error(vol->sb, "MFTMirr LCN is beyond end of volume.  "
-				"Weird.");
+		ntfs_error(vol->sb, "MFTMirr LCN (%lli, 0x%llx) is beyond end "
+				"of volume.  Weird.", (unsigned long long)ll,
+				(unsigned long long)ll);
 		return FALSE;
 	}
 	vol->mftmirr_lcn = ll;
@@ -2685,7 +2708,7 @@ static int ntfs_fill_super(struct super_block *sb, void *opt, const int silent)
 	ntfs_volume *vol;
 	struct buffer_head *bh;
 	struct inode *tmp_ino;
-	int result;
+	int blocksize, result;
 
 	ntfs_debug("Entering.");
 #ifndef NTFS_RW
@@ -2724,60 +2747,85 @@ static int ntfs_fill_super(struct super_block *sb, void *opt, const int silent)
 	if (!parse_options(vol, (char*)opt))
 		goto err_out_now;
 
+	/* We support sector sizes up to the PAGE_CACHE_SIZE. */
+	if (bdev_hardsect_size(sb->s_bdev) > PAGE_CACHE_SIZE) {
+		if (!silent)
+			ntfs_error(sb, "Device has unsupported sector size "
+					"(%i).  The maximum supported sector "
+					"size on this architecture is %lu "
+					"bytes.",
+					bdev_hardsect_size(sb->s_bdev),
+					PAGE_CACHE_SIZE);
+		goto err_out_now;
+	}
 	/*
-	 * TODO: Fail safety check. In the future we should really be able to
-	 * cope with this being the case, but for now just bail out.
+	 * Setup the device access block size to NTFS_BLOCK_SIZE or the hard
+	 * sector size, whichever is bigger.
 	 */
-	if (bdev_hardsect_size(sb->s_bdev) > NTFS_BLOCK_SIZE) {
+	blocksize = sb_min_blocksize(sb, NTFS_BLOCK_SIZE);
+	if (blocksize < NTFS_BLOCK_SIZE) {
 		if (!silent)
-			ntfs_error(sb, "Device has unsupported hardsect_size.");
+			ntfs_error(sb, "Unable to set device block size.");
 		goto err_out_now;
 	}
-
-	/* Setup the device access block size to NTFS_BLOCK_SIZE. */
-	if (sb_set_blocksize(sb, NTFS_BLOCK_SIZE) != NTFS_BLOCK_SIZE) {
+	BUG_ON(blocksize != sb->s_blocksize);
+	ntfs_debug("Set device block size to %i bytes (block size bits %i).",
+			blocksize, sb->s_blocksize_bits);
+	/* Determine the size of the device in units of block_size bytes. */
+	if (!i_size_read(sb->s_bdev->bd_inode)) {
 		if (!silent)
-			ntfs_error(sb, "Unable to set block size.");
+			ntfs_error(sb, "Unable to determine device size.");
 		goto err_out_now;
 	}
-
-	/* Get the size of the device in units of NTFS_BLOCK_SIZE bytes. */
 	vol->nr_blocks = i_size_read(sb->s_bdev->bd_inode) >>
-			NTFS_BLOCK_SIZE_BITS;
-
+			sb->s_blocksize_bits;
 	/* Read the boot sector and return unlocked buffer head to it. */
 	if (!(bh = read_ntfs_boot_sector(sb, silent))) {
 		if (!silent)
 			ntfs_error(sb, "Not an NTFS volume.");
 		goto err_out_now;
 	}
-
 	/*
-	 * Extract the data from the boot sector and setup the ntfs super block
+	 * Extract the data from the boot sector and setup the ntfs volume
 	 * using it.
 	 */
 	result = parse_ntfs_boot_sector(vol, (NTFS_BOOT_SECTOR*)bh->b_data);
-
-	/* Initialize the cluster and mft allocators. */
-	ntfs_setup_allocators(vol);
-
 	brelse(bh);
-
 	if (!result) {
 		if (!silent)
 			ntfs_error(sb, "Unsupported NTFS filesystem.");
 		goto err_out_now;
 	}
-
 	/*
-	 * TODO: When we start coping with sector sizes different from
-	 * NTFS_BLOCK_SIZE, we now probably need to set the blocksize of the
-	 * device (probably to NTFS_BLOCK_SIZE).
+	 * If the boot sector indicates a sector size bigger than the current
+	 * device block size, switch the device block size to the sector size.
+	 * TODO: It may be possible to support this case even when the set
+	 * below fails, we would just be breaking up the i/o for each sector
+	 * into multiple blocks for i/o purposes but otherwise it should just
+	 * work.  However it is safer to leave disabled until someone hits this
+	 * error message and then we can get them to try it without the setting
+	 * so we know for sure that it works.
 	 */
-
+	if (vol->sector_size > blocksize) {
+		blocksize = sb_set_blocksize(sb, vol->sector_size);
+		if (blocksize != vol->sector_size) {
+			if (!silent)
+				ntfs_error(sb, "Unable to set device block "
+						"size to sector size (%i).",
+						vol->sector_size);
+			goto err_out_now;
+		}
+		BUG_ON(blocksize != sb->s_blocksize);
+		vol->nr_blocks = i_size_read(sb->s_bdev->bd_inode) >>
+				sb->s_blocksize_bits;
+		ntfs_debug("Changed device block size to %i bytes (block size "
+				"bits %i) to match volume sector size.",
+				blocksize, sb->s_blocksize_bits);
+	}
+	/* Initialize the cluster and mft allocators. */
+	ntfs_setup_allocators(vol);
 	/* Setup remaining fields in the super block. */
 	sb->s_magic = NTFS_SB_MAGIC;
-
 	/*
 	 * Ntfs allows 63 bits for the file size, i.e. correct would be:
 	 *	sb->s_maxbytes = ~0ULL >> 1;
@@ -2787,9 +2835,8 @@ static int ntfs_fill_super(struct super_block *sb, void *opt, const int silent)
 	 * without overflowing the index or to 2^63 - 1, whichever is smaller.
 	 */
 	sb->s_maxbytes = MAX_LFS_FILESIZE;
-
+	/* Ntfs measures time in 100ns intervals. */
 	sb->s_time_gran = 100;
-
 	/*
 	 * Now load the metadata required for the page cache and our address
 	 * space operations to function. We do this by setting up a specialised

From 1cf3109ffb26a6ea572fd02436bd10458b4b2187 Mon Sep 17 00:00:00 2001
From: Anton Altaparmakov <aia21@cantab.net>
Date: Fri, 24 Feb 2006 10:48:14 +0000
Subject: [PATCH 428/509] NTFS: Do more detailed reporting of why we cannot
 mount read-write by      special casing the VOLUME_MODIFIED_BY_CHKDSK flag.

Signed-off-by: Anton Altaparmakov <aia21@cantab.net>
---
 Documentation/filesystems/ntfs.txt |  6 ++++++
 fs/ntfs/ChangeLog                  |  2 ++
 fs/ntfs/Makefile                   |  2 +-
 fs/ntfs/super.c                    | 34 ++++++++++++++++++++++++------
 fs/ntfs/upcase.c                   | 10 ++++-----
 fs/ntfs/volume.h                   | 28 ++++++++++++------------
 6 files changed, 53 insertions(+), 29 deletions(-)

diff --git a/Documentation/filesystems/ntfs.txt b/Documentation/filesystems/ntfs.txt
index 614de3124901..251168587899 100644
--- a/Documentation/filesystems/ntfs.txt
+++ b/Documentation/filesystems/ntfs.txt
@@ -457,6 +457,12 @@ ChangeLog
 
 Note, a technical ChangeLog aimed at kernel hackers is in fs/ntfs/ChangeLog.
 
+2.1.26:
+	- Implement support for sector sizes above 512 bytes (up to the maximum
+	  supported by NTFS which is 4096 bytes).
+	- Enhance support for NTFS volumes which were supported by Windows but
+	  not by Linux due to invalid attribute list attribute flags.
+	- A few minor updates and bug fixes.
 2.1.25:
 	- Write support is now extended with write(2) being able to both
 	  overwrite existing file data and to extend files.  Also, if a write
diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog
index e66b4ac2fade..9d8ffa89e2c2 100644
--- a/fs/ntfs/ChangeLog
+++ b/fs/ntfs/ChangeLog
@@ -29,6 +29,8 @@ ToDo/Notes:
 	  kmem_cache_t.  (Pekka Enberg)
 	- Implement support for sector sizes above 512 bytes (up to the maximum
 	  supported by NTFS which is 4096 bytes).
+	- Do more detailed reporting of why we cannot mount read-write by
+	  special casing the VOLUME_MODIFIED_BY_CHKDSK flag.
 	- Miscellaneous updates to layout.h.
 	- Cope with attribute list attribute having invalid flags.  Windows
 	  copes with this and even chkdsk does not detect or fix this so we
diff --git a/fs/ntfs/Makefile b/fs/ntfs/Makefile
index d0d45d1c853a..d95fac7fdeb6 100644
--- a/fs/ntfs/Makefile
+++ b/fs/ntfs/Makefile
@@ -6,7 +6,7 @@ ntfs-objs := aops.o attrib.o collate.o compress.o debug.o dir.o file.o \
 	     index.o inode.o mft.o mst.o namei.o runlist.o super.o sysctl.o \
 	     unistr.o upcase.o
 
-EXTRA_CFLAGS = -DNTFS_VERSION=\"2.1.25\"
+EXTRA_CFLAGS = -DNTFS_VERSION=\"2.1.26\"
 
 ifeq ($(CONFIG_NTFS_DEBUG),y)
 EXTRA_CFLAGS += -DDEBUG
diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c
index 489f7049146b..368a8ec10668 100644
--- a/fs/ntfs/super.c
+++ b/fs/ntfs/super.c
@@ -472,9 +472,16 @@ static int ntfs_remount(struct super_block *sb, int *flags, char *opt)
 			ntfs_error(sb, "Volume is dirty and read-only%s", es);
 			return -EROFS;
 		}
+		if (vol->vol_flags & VOLUME_MODIFIED_BY_CHKDSK) {
+			ntfs_error(sb, "Volume has been modified by chkdsk "
+					"and is read-only%s", es);
+			return -EROFS;
+		}
 		if (vol->vol_flags & VOLUME_MUST_MOUNT_RO_MASK) {
-			ntfs_error(sb, "Volume has unsupported flags set and "
-					"is read-only%s", es);
+			ntfs_error(sb, "Volume has unsupported flags set "
+					"(0x%x) and is read-only%s",
+					(unsigned)le16_to_cpu(vol->vol_flags),
+					es);
 			return -EROFS;
 		}
 		if (ntfs_set_volume_flags(vol, VOLUME_IS_DIRTY)) {
@@ -1845,11 +1852,24 @@ get_ctx_vol_failed:
 	/* Make sure that no unsupported volume flags are set. */
 	if (vol->vol_flags & VOLUME_MUST_MOUNT_RO_MASK) {
 		static const char *es1a = "Volume is dirty";
-		static const char *es1b = "Volume has unsupported flags set";
-		static const char *es2 = ".  Run chkdsk and mount in Windows.";
-		const char *es1;
-		
-		es1 = vol->vol_flags & VOLUME_IS_DIRTY ? es1a : es1b;
+		static const char *es1b = "Volume has been modified by chkdsk";
+		static const char *es1c = "Volume has unsupported flags set";
+		static const char *es2a = ".  Run chkdsk and mount in Windows.";
+		static const char *es2b = ".  Mount in Windows.";
+		const char *es1, *es2;
+
+		es2 = es2a;
+		if (vol->vol_flags & VOLUME_IS_DIRTY)
+			es1 = es1a;
+		else if (vol->vol_flags & VOLUME_MODIFIED_BY_CHKDSK) {
+			es1 = es1b;
+			es2 = es2b;
+		} else {
+			es1 = es1c;
+			ntfs_warning(sb, "Unsupported volume flags 0x%x "
+					"encountered.",
+					(unsigned)le16_to_cpu(vol->vol_flags));
+		}
 		/* If a read-write mount, convert it to a read-only mount. */
 		if (!(sb->s_flags & MS_RDONLY)) {
 			if (!(vol->on_errors & (ON_ERRORS_REMOUNT_RO |
diff --git a/fs/ntfs/upcase.c b/fs/ntfs/upcase.c
index 879cdf1d5bd3..9101807dc81a 100644
--- a/fs/ntfs/upcase.c
+++ b/fs/ntfs/upcase.c
@@ -3,10 +3,7 @@
  *	      Part of the Linux-NTFS project.
  *
  * Copyright (c) 2001 Richard Russon <ntfs@flatcap.org>
- * Copyright (c) 2001-2004 Anton Altaparmakov
- *
- * Modified for mkntfs inclusion 9 June 2001 by Anton Altaparmakov.
- * Modified for kernel inclusion 10 September 2001 by Anton Altparmakov.
+ * Copyright (c) 2001-2006 Anton Altaparmakov
  *
  * This program is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the Free
@@ -75,12 +72,13 @@ ntfschar *generate_default_upcase(void)
 	if (!uc)
 		return uc;
 	memset(uc, 0, default_upcase_len * sizeof(ntfschar));
+	/* Generate the little endian Unicode upcase table used by ntfs. */
 	for (i = 0; i < default_upcase_len; i++)
 		uc[i] = cpu_to_le16(i);
 	for (r = 0; uc_run_table[r][0]; r++)
 		for (i = uc_run_table[r][0]; i < uc_run_table[r][1]; i++)
-			uc[i] = cpu_to_le16((le16_to_cpu(uc[i]) +
-					uc_run_table[r][2]));
+			uc[i] = cpu_to_le16(le16_to_cpu(uc[i]) +
+					uc_run_table[r][2]);
 	for (r = 0; uc_dup_table[r][0]; r++)
 		for (i = uc_dup_table[r][0]; i < uc_dup_table[r][1]; i += 2)
 			uc[i + 1] = cpu_to_le16(le16_to_cpu(uc[i + 1]) - 1);
diff --git a/fs/ntfs/volume.h b/fs/ntfs/volume.h
index 375cd20a9f61..406ab55dfb32 100644
--- a/fs/ntfs/volume.h
+++ b/fs/ntfs/volume.h
@@ -2,7 +2,7 @@
  * volume.h - Defines for volume structures in NTFS Linux kernel driver. Part
  *	      of the Linux-NTFS project.
  *
- * Copyright (c) 2001-2005 Anton Altaparmakov
+ * Copyright (c) 2001-2006 Anton Altaparmakov
  * Copyright (c) 2002 Richard Russon
  *
  * This program/include file is free software; you can redistribute it and/or
@@ -41,10 +41,8 @@ typedef struct {
 	 * structure has stabilized... (AIA)
 	 */
 	/* Device specifics. */
-	struct super_block *sb;		/* Pointer back to the super_block,
-					   so we don't have to get the offset
-					   every time. */
-	LCN nr_blocks;			/* Number of NTFS_BLOCK_SIZE bytes
+	struct super_block *sb;		/* Pointer back to the super_block. */
+	LCN nr_blocks;			/* Number of sb->s_blocksize bytes
 					   sized blocks on the device. */
 	/* Configuration provided by user at mount time. */
 	unsigned long flags;		/* Miscellaneous flags, see below. */
@@ -141,8 +139,8 @@ typedef enum {
 	NV_ShowSystemFiles,	/* 1: Return system files in ntfs_readdir(). */
 	NV_CaseSensitive,	/* 1: Treat file names as case sensitive and
 				      create filenames in the POSIX namespace.
-				      Otherwise be case insensitive and create
-				      file names in WIN32 namespace. */
+				      Otherwise be case insensitive but still
+				      create file names in POSIX namespace. */
 	NV_LogFileEmpty,	/* 1: $LogFile journal is empty. */
 	NV_QuotaOutOfDate,	/* 1: $Quota is out of date. */
 	NV_UsnJrnlStamped,	/* 1: $UsnJrnl has been stamped. */
@@ -153,7 +151,7 @@ typedef enum {
  * Macro tricks to expand the NVolFoo(), NVolSetFoo(), and NVolClearFoo()
  * functions.
  */
-#define NVOL_FNS(flag)					\
+#define DEFINE_NVOL_BIT_OPS(flag)					\
 static inline int NVol##flag(ntfs_volume *vol)		\
 {							\
 	return test_bit(NV_##flag, &(vol)->flags);	\
@@ -168,12 +166,12 @@ static inline void NVolClear##flag(ntfs_volume *vol)	\
 }
 
 /* Emit the ntfs volume bitops functions. */
-NVOL_FNS(Errors)
-NVOL_FNS(ShowSystemFiles)
-NVOL_FNS(CaseSensitive)
-NVOL_FNS(LogFileEmpty)
-NVOL_FNS(QuotaOutOfDate)
-NVOL_FNS(UsnJrnlStamped)
-NVOL_FNS(SparseEnabled)
+DEFINE_NVOL_BIT_OPS(Errors)
+DEFINE_NVOL_BIT_OPS(ShowSystemFiles)
+DEFINE_NVOL_BIT_OPS(CaseSensitive)
+DEFINE_NVOL_BIT_OPS(LogFileEmpty)
+DEFINE_NVOL_BIT_OPS(QuotaOutOfDate)
+DEFINE_NVOL_BIT_OPS(UsnJrnlStamped)
+DEFINE_NVOL_BIT_OPS(SparseEnabled)
 
 #endif /* _LINUX_NTFS_VOLUME_H */

From 0c0888908dec145aaaa40d8a49d34913573f5a27 Mon Sep 17 00:00:00 2001
From: Hugo Santos <hsantos@av.it.pt>
Date: Fri, 24 Feb 2006 13:16:25 -0800
Subject: [PATCH 429/509] [IPV6] ip6_tunnel: release cached dst on change of
 tunnel params

The included patch fixes ip6_tunnel to release the cached dst entry
when the tunnel parameters (such as tunnel endpoints) are changed so
they are used immediatly for the next encapsulated packets.

Signed-off-by: Hugo Santos <hsantos@av.it.pt>
Acked-by: Ville Nuorvala <vnuorval@tcs.hut.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/ip6_tunnel.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index faea8a120ee2..48597538db3f 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -884,6 +884,7 @@ ip6ip6_tnl_change(struct ip6_tnl *t, struct ip6_tnl_parm *p)
 	t->parms.encap_limit = p->encap_limit;
 	t->parms.flowinfo = p->flowinfo;
 	t->parms.link = p->link;
+	ip6_tnl_dst_reset(t);
 	ip6ip6_tnl_link_config(t);
 	return 0;
 }

From d91675f9c7f5752e8657df1e1d926bd6a624434f Mon Sep 17 00:00:00 2001
From: YOSHIFUJI Hideaki <yosufuji@linux-ipv6.org>
Date: Fri, 24 Feb 2006 13:18:33 -0800
Subject: [PATCH 430/509] [IPV6]: Do not ignore IPV6_MTU socket option.

Based on patch by Hoerdt Mickael <hoerdt@clarinet.u-strasbg.fr>.

Signed-off-by: YOSHIFUJI Hideaki <yosufuji@linux-ipv6.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv6/ip6_output.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index efa3e72cfcfa..f999edd846a9 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -494,6 +494,7 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 	struct net_device *dev;
 	struct sk_buff *frag;
 	struct rt6_info *rt = (struct rt6_info*)skb->dst;
+	struct ipv6_pinfo *np = skb->sk ? inet6_sk(skb->sk) : NULL;
 	struct ipv6hdr *tmp_hdr;
 	struct frag_hdr *fh;
 	unsigned int mtu, hlen, left, len;
@@ -505,7 +506,12 @@ static int ip6_fragment(struct sk_buff *skb, int (*output)(struct sk_buff *))
 	hlen = ip6_find_1stfragopt(skb, &prevhdr);
 	nexthdr = *prevhdr;
 
-	mtu = dst_mtu(&rt->u.dst) - hlen - sizeof(struct frag_hdr);
+	mtu = dst_mtu(&rt->u.dst);
+	if (np && np->frag_size < mtu) {
+		if (np->frag_size)
+			mtu = np->frag_size;
+	}
+	mtu -= hlen + sizeof(struct frag_hdr);
 
 	if (skb_shinfo(skb)->frag_list) {
 		int first_len = skb_pagelen(skb);
@@ -882,7 +888,12 @@ int ip6_append_data(struct sock *sk, int getfrag(void *from, char *to,
 		inet->cork.fl = *fl;
 		np->cork.hop_limit = hlimit;
 		np->cork.tclass = tclass;
-		inet->cork.fragsize = mtu = dst_mtu(rt->u.dst.path);
+		mtu = dst_mtu(rt->u.dst.path);
+		if (np && np->frag_size < mtu) {
+			if (np->frag_size)
+				mtu = np->frag_size;
+		}
+		inet->cork.fragsize = mtu;
 		if (dst_allfrag(rt->u.dst.path))
 			inet->cork.flags |= IPCORK_ALLFRAG;
 		inet->cork.length = 0;

From abbea7187296a7fb316a55f2319438c2bf881f0a Mon Sep 17 00:00:00 2001
From: Andrew Victor <andrew@sanpeople.com>
Date: Fri, 24 Feb 2006 22:27:50 +0000
Subject: [PATCH 431/509] [ARM] 3348/1: Disable GPIO interrupts

Patch from Andrew Victor

disable_irq() lazily disables the interrupt, so the IRQ is only disabled
once the interrupt occurs again.  The GPIO interrupt handler therefore
must first check disable_depth to see if the IRQ needs to be disabled.

Orignal patch by Bill Gatliff.

Signed-off-by: Andrew Victor <andrew@sanpeople.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-at91rm9200/gpio.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-at91rm9200/gpio.c b/arch/arm/mach-at91rm9200/gpio.c
index a9f718bf8ba8..0e396feec468 100644
--- a/arch/arm/mach-at91rm9200/gpio.c
+++ b/arch/arm/mach-at91rm9200/gpio.c
@@ -274,8 +274,18 @@ static void gpio_irq_handler(unsigned irq, struct irqdesc *desc, struct pt_regs
 		gpio = &irq_desc[pin];
 
 		while (isr) {
-			if (isr & 1)
-				gpio->handle(pin, gpio, regs);
+			if (isr & 1) {
+				if (unlikely(gpio->disable_depth)) {
+					/*
+					 * The core ARM interrupt handler lazily disables IRQs so
+					 * another IRQ must be generated before it actually gets
+					 * here to be disabled on the GPIO controller.
+					 */
+					gpio_irq_mask(pin);
+				}
+				else
+					gpio->handle(pin, gpio, regs);
+			}
 			pin++;
 			gpio++;
 			isr >>= 1;

From cde05cf2145b0aa06dd61277060bfba5d38acb0b Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Fri, 24 Feb 2006 13:03:50 -0800
Subject: [PATCH 432/509] [PATCH] m32r: enable asm code optimization

Add -O2 option to AFLAGS to enable asm code optimization for m32r.

On m32r gas, "-m32r2 -O" option enables assembler's parallel code
generation optimization for M32R2 ISA as a default.  So, "-no-parallel"
option is required explicitly for a cpu core with single instuction
issuing, for example, VDEC2.

Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/m32r/Makefile | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/m32r/Makefile b/arch/m32r/Makefile
index 983d438b14b6..4b3c90ba926c 100644
--- a/arch/m32r/Makefile
+++ b/arch/m32r/Makefile
@@ -12,14 +12,14 @@ CFLAGS_MODULE += -mmodel=large
 
 ifdef CONFIG_CHIP_VDEC2
 cflags-$(CONFIG_ISA_M32R2)	+= -DNO_FPU -Wa,-bitinst
-aflags-$(CONFIG_ISA_M32R2)	+= -DNO_FPU -Wa,-bitinst
+aflags-$(CONFIG_ISA_M32R2)	+= -DNO_FPU -O2 -Wa,-bitinst -Wa,-no-parallel
 else
 cflags-$(CONFIG_ISA_M32R2)	+= -DNO_FPU -m32r2
-aflags-$(CONFIG_ISA_M32R2)	+= -DNO_FPU -m32r2
+aflags-$(CONFIG_ISA_M32R2)	+= -DNO_FPU -m32r2 -O2
 endif
 
 cflags-$(CONFIG_ISA_M32R)	+= -DNO_FPU
-aflags-$(CONFIG_ISA_M32R)	+= -DNO_FPU -Wa,-no-bitinst
+aflags-$(CONFIG_ISA_M32R)	+= -DNO_FPU -O2 -Wa,-no-bitinst
 
 CFLAGS += $(cflags-y)
 AFLAGS += $(aflags-y)

From 6ced13cdcab440931b87829b0f2d0dedacfb3f2d Mon Sep 17 00:00:00 2001
From: Hirokazu Takata <takata@linux-m32r.org>
Date: Fri, 24 Feb 2006 13:03:51 -0800
Subject: [PATCH 433/509] [PATCH] m32r: fix and update for gcc-4.0

Fix and update for gcc-4.0.

- arch/m32r/kernel/signal.c:
  Change type of the 8th parameter of sys_rt_sigsuspend() from
  'struct pt_regs' to 'struct pt_regs *'.
  This functions make use of the 'regs' parameter to return status value,
  but gcc-4.0 optimizes and removes it as a dead code.
  Functions, sys_sigaltstack() and sys_rt_sigreturn(), have also modified.

- arch/m32r/lib/usercopy.c, include/asm-m32r/uaccess.h:
  Add early-clobber constraints('&') to output values of asm statements;
  these constraints seems to be required for gcc-4.0 register assignment.

Signed-off-by: Hayato Fujiwara <fujiwara@linux-m32r.org>
Signed-off-by: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/m32r/kernel/signal.c  | 24 ++++++++++--------------
 arch/m32r/lib/usercopy.c   |  4 ++--
 include/asm-m32r/uaccess.h |  8 ++++----
 3 files changed, 16 insertions(+), 20 deletions(-)

diff --git a/arch/m32r/kernel/signal.c b/arch/m32r/kernel/signal.c
index 71763f7a1d19..cb33097fefc4 100644
--- a/arch/m32r/kernel/signal.c
+++ b/arch/m32r/kernel/signal.c
@@ -36,7 +36,7 @@ int do_signal(struct pt_regs *, sigset_t *);
 asmlinkage int
 sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
 		  unsigned long r2, unsigned long r3, unsigned long r4,
-		  unsigned long r5, unsigned long r6, struct pt_regs regs)
+		  unsigned long r5, unsigned long r6, struct pt_regs *regs)
 {
 	sigset_t saveset, newset;
 
@@ -54,21 +54,21 @@ sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	regs.r0 = -EINTR;
+	regs->r0 = -EINTR;
 	while (1) {
 		current->state = TASK_INTERRUPTIBLE;
 		schedule();
-		if (do_signal(&regs, &saveset))
-			return regs.r0;
+		if (do_signal(regs, &saveset))
+			return regs->r0;
 	}
 }
 
 asmlinkage int
 sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
 		unsigned long r2, unsigned long r3, unsigned long r4,
-		unsigned long r5, unsigned long r6, struct pt_regs regs)
+		unsigned long r5, unsigned long r6, struct pt_regs *regs)
 {
-	return do_sigaltstack(uss, uoss, regs.spu);
+	return do_sigaltstack(uss, uoss, regs->spu);
 }
 
 
@@ -140,11 +140,10 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc,
 asmlinkage int
 sys_rt_sigreturn(unsigned long r0, unsigned long r1,
 		 unsigned long r2, unsigned long r3, unsigned long r4,
-		 unsigned long r5, unsigned long r6, struct pt_regs regs)
+		 unsigned long r5, unsigned long r6, struct pt_regs *regs)
 {
-	struct rt_sigframe __user *frame = (struct rt_sigframe __user *)regs.spu;
+	struct rt_sigframe __user *frame = (struct rt_sigframe __user *)regs->spu;
 	sigset_t set;
-	stack_t st;
 	int result;
 
 	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
@@ -158,14 +157,11 @@ sys_rt_sigreturn(unsigned long r0, unsigned long r1,
 	recalc_sigpending();
 	spin_unlock_irq(&current->sighand->siglock);
 
-	if (restore_sigcontext(&regs, &frame->uc.uc_mcontext, &result))
+	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &result))
 		goto badframe;
 
-	if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
+	if (do_sigaltstack(&frame->uc.uc_stack, NULL, regs->spu) == -EFAULT)
 		goto badframe;
-	/* It is more difficult to avoid calling this function than to
-	   call it and ignore errors.  */
-	do_sigaltstack(&st, NULL, regs.spu);
 
 	return result;
 
diff --git a/arch/m32r/lib/usercopy.c b/arch/m32r/lib/usercopy.c
index ce16bbe26a52..2d1dd2106c4d 100644
--- a/arch/m32r/lib/usercopy.c
+++ b/arch/m32r/lib/usercopy.c
@@ -64,7 +64,7 @@ do {									\
 		"	.balign 4\n"					\
 		"	.long 0b,3b\n"					\
 		".previous"						\
-		: "=r"(res), "=r"(count), "=&r" (__d0), "=&r" (__d1),	\
+		: "=&r"(res), "=&r"(count), "=&r" (__d0), "=&r" (__d1),	\
 		  "=&r" (__d2)						\
 		: "i"(-EFAULT), "0"(count), "1"(count), "3"(src), 	\
 		  "4"(dst)						\
@@ -101,7 +101,7 @@ do {									\
 		"	.balign 4\n"					\
 		"	.long 0b,3b\n"					\
 		".previous"						\
-		: "=r"(res), "=r"(count), "=&r" (__d0), "=&r" (__d1),	\
+		: "=&r"(res), "=&r"(count), "=&r" (__d0), "=&r" (__d1),	\
 		  "=&r" (__d2)						\
 		: "i"(-EFAULT), "0"(count), "1"(count), "3"(src),	\
 		  "4"(dst)						\
diff --git a/include/asm-m32r/uaccess.h b/include/asm-m32r/uaccess.h
index 0da7c47d2f01..e8ae61956a51 100644
--- a/include/asm-m32r/uaccess.h
+++ b/include/asm-m32r/uaccess.h
@@ -328,7 +328,7 @@ extern void __put_user_bad(void);
                 "       .long 1b,4b\n"                                  \
                 "       .long 2b,4b\n"                                  \
                 ".previous"                                             \
-                : "=r"(err)                                             \
+                : "=&r"(err)                                             \
                 : "r"(x), "r"(addr), "i"(-EFAULT), "0"(err)		\
                 : "r14", "memory")
 
@@ -353,7 +353,7 @@ extern void __put_user_bad(void);
 		"	.long 1b,4b\n"					\
 		"	.long 2b,4b\n"					\
 		".previous"						\
-		: "=r"(err)						\
+		: "=&r"(err)						\
 		: "r"(x), "r"(addr), "i"(-EFAULT), "0"(err)		\
 		: "r14", "memory")
 #else
@@ -398,7 +398,7 @@ struct __large_struct { unsigned long buf[100]; };
 		"	.balign 4\n"					\
 		"	.long 1b,3b\n"					\
 		".previous"						\
-		: "=r"(err)						\
+		: "=&r"(err)						\
 		: "r"(x), "r"(addr), "i"(-EFAULT), "0"(err)		\
 		: "r14", "memory")
 
@@ -442,7 +442,7 @@ do {									\
 		"	.balign 4\n"					\
 		"	.long 1b,3b\n"					\
 		".previous"						\
-		: "=r"(err), "=&r"(x)					\
+		: "=&r"(err), "=&r"(x)					\
 		: "r"(addr), "i"(-EFAULT), "0"(err)			\
 		: "r14", "memory")
 

From 6f595cffedc09da3f5bed13afc86aac64e67c4d7 Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@keyaccess.nl>
Date: Fri, 24 Feb 2006 13:03:51 -0800
Subject: [PATCH 434/509] [PATCH] snd-cs4236 typo fix

I noticed on 2.6.16-rc4 that my MPU-401 wasn't functional, due to a simple
copy & paste error in sound/isa/cs423x/cs4236.c.

Acked-by: Takashi Iwai <tiwai@suse.de>
Cc: Jaroslav Kysela <perex@suse.cz>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 sound/isa/cs423x/cs4236.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/isa/cs423x/cs4236.c b/sound/isa/cs423x/cs4236.c
index 4fa431040564..99a42138bea0 100644
--- a/sound/isa/cs423x/cs4236.c
+++ b/sound/isa/cs423x/cs4236.c
@@ -414,7 +414,7 @@ static int __devinit snd_card_cs423x_pnpc(int dev, struct snd_card_cs4236 *acard
 	}
 	/* MPU initialization */
 	if (acard->mpu && mpu_port[dev] > 0) {
-		if (snd_cs423x_pnp_init_mpu(dev, acard->ctrl, cfg) < 0)
+		if (snd_cs423x_pnp_init_mpu(dev, acard->mpu, cfg) < 0)
 			goto error;
 	}
 	kfree(cfg);

From b1a3aa209161af3b7ca8ce8eae89b15faa96f612 Mon Sep 17 00:00:00 2001
From: Takashi Iwai <tiwai@suse.de>
Date: Fri, 24 Feb 2006 13:03:52 -0800
Subject: [PATCH 435/509] [PATCH] alsa: fix bogus snd_device_free() in
 opl3-oss.c

Remove snd_device_free() for an opl3-oss instance which should have been
released.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Cc: Jaroslav Kysela <perex@suse.cz>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 sound/drivers/opl3/opl3_oss.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/sound/drivers/opl3/opl3_oss.c b/sound/drivers/opl3/opl3_oss.c
index 31f1f2e25aa0..0345ae647681 100644
--- a/sound/drivers/opl3/opl3_oss.c
+++ b/sound/drivers/opl3/opl3_oss.c
@@ -146,7 +146,7 @@ void snd_opl3_init_seq_oss(struct snd_opl3 *opl3, char *name)
 void snd_opl3_free_seq_oss(struct snd_opl3 *opl3)
 {
 	if (opl3->oss_seq_dev) {
-		snd_device_free(opl3->card, opl3->oss_seq_dev);
+		/* The instance should have been released in prior */
 		opl3->oss_seq_dev = NULL;
 	}
 }

From 31bc5a33346b6dd35be219d1416449e0064e9123 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Fri, 24 Feb 2006 13:03:53 -0800
Subject: [PATCH 436/509] [PATCH] uml: correct error messages in COW driver

Improve some error messages in the COW driver, and say V3, not V2, when
talking about V3 format.  Also resync with our userspace code utility a bit
more.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/cow_sys.h  |  2 +-
 arch/um/drivers/cow_user.c | 21 ++++++++++++---------
 2 files changed, 13 insertions(+), 10 deletions(-)

diff --git a/arch/um/drivers/cow_sys.h b/arch/um/drivers/cow_sys.h
index c83fc5d68936..df25263d2ad0 100644
--- a/arch/um/drivers/cow_sys.h
+++ b/arch/um/drivers/cow_sys.h
@@ -33,7 +33,7 @@ static inline int cow_file_size(char *file, unsigned long long *size_out)
 	return(os_file_size(file, size_out));
 }
 
-static inline int cow_write_file(int fd, char *buf, int size)
+static inline int cow_write_file(int fd, void *buf, int size)
 {
 	return(os_write_file(fd, buf, size));
 }
diff --git a/arch/um/drivers/cow_user.c b/arch/um/drivers/cow_user.c
index fbe2217db5dd..d1c86bc377bd 100644
--- a/arch/um/drivers/cow_user.c
+++ b/arch/um/drivers/cow_user.c
@@ -176,7 +176,7 @@ int write_cow_header(char *cow_file, int fd, char *backing_file,
 	err = -ENOMEM;
 	header = cow_malloc(sizeof(*header));
 	if(header == NULL){
-		cow_printf("Failed to allocate COW V3 header\n");
+		cow_printf("write_cow_header - failed to allocate COW V3 header\n");
 		goto out;
 	}
 	header->magic = htonl(COW_MAGIC);
@@ -196,15 +196,17 @@ int write_cow_header(char *cow_file, int fd, char *backing_file,
 
 	err = os_file_modtime(header->backing_file, &modtime);
 	if(err < 0){
-		cow_printf("Backing file '%s' mtime request failed, "
-			   "err = %d\n", header->backing_file, -err);
+		cow_printf("write_cow_header - backing file '%s' mtime "
+			   "request failed, err = %d\n", header->backing_file,
+			   -err);
 		goto out_free;
 	}
 
 	err = cow_file_size(header->backing_file, size);
 	if(err < 0){
-		cow_printf("Couldn't get size of backing file '%s', "
-			   "err = %d\n", header->backing_file, -err);
+		cow_printf("write_cow_header - couldn't get size of "
+			   "backing file '%s', err = %d\n",
+			   header->backing_file, -err);
 		goto out_free;
 	}
 
@@ -214,10 +216,11 @@ int write_cow_header(char *cow_file, int fd, char *backing_file,
 	header->alignment = htonl(alignment);
 	header->cow_format = COW_BITMAP;
 
-	err = os_write_file(fd, header, sizeof(*header));
+	err = cow_write_file(fd, header, sizeof(*header));
 	if(err != sizeof(*header)){
-		cow_printf("Write of header to new COW file '%s' failed, "
-			   "err = %d\n", cow_file, -err);
+		cow_printf("write_cow_header - write of header to "
+			   "new COW file '%s' failed, err = %d\n", cow_file,
+			   -err);
 		goto out_free;
 	}
 	err = 0;
@@ -299,7 +302,7 @@ int read_cow_header(int (*reader)(__u64, char *, int, void *), void *arg,
 	}
 	else if(version == 3){
 		if(n < sizeof(header->v3)){
-			cow_printf("read_cow_header - failed to read V2 "
+			cow_printf("read_cow_header - failed to read V3 "
 				   "header\n");
 			goto out;
 		}

From 07f4e2c61c76e8b543c0a2589063aea85c15fb25 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Fri, 24 Feb 2006 13:03:55 -0800
Subject: [PATCH 437/509] [PATCH] uml: fix usage of kernel_errno in place of
 errno

To avoid conflicts, in kernel files errno is expanded to kernel_errno, to
distinguish it from glibc errno.  In this case, the code wants to use the libc
errno but the kernel one is used; in the other usage, we return errno in place
of -errno in case of an error.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/os.h       |  3 +++
 arch/um/os-Linux/process.c | 16 ++++++++++++++++
 arch/um/sys-i386/ldt.c     |  9 +++------
 3 files changed, 22 insertions(+), 6 deletions(-)

diff --git a/arch/um/include/os.h b/arch/um/include/os.h
index eb1710b81255..2a1c64d8d0bf 100644
--- a/arch/um/include/os.h
+++ b/arch/um/include/os.h
@@ -179,8 +179,11 @@ extern void os_stop_process(int pid);
 extern void os_kill_process(int pid, int reap_child);
 extern void os_kill_ptraced_process(int pid, int reap_child);
 extern void os_usr1_process(int pid);
+extern long os_ptrace_ldt(long pid, long addr, long data);
+
 extern int os_getpid(void);
 extern int os_getpgrp(void);
+
 extern void init_new_thread_stack(void *sig_stack, void (*usr1_handler)(int));
 extern void init_new_thread_signals(int altstack);
 extern int run_kernel_thread(int (*fn)(void *), void *arg, void **jmp_ptr);
diff --git a/arch/um/os-Linux/process.c b/arch/um/os-Linux/process.c
index 7f5e2dac2a35..d261888f39c4 100644
--- a/arch/um/os-Linux/process.c
+++ b/arch/um/os-Linux/process.c
@@ -19,6 +19,7 @@
 #include "irq_user.h"
 #include "kern_util.h"
 #include "longjmp.h"
+#include "skas_ptrace.h"
 
 #define ARBITRARY_ADDR -1
 #define FAILURE_PID    -1
@@ -100,6 +101,21 @@ void os_kill_process(int pid, int reap_child)
 		
 }
 
+/* This is here uniquely to have access to the userspace errno, i.e. the one
+ * used by ptrace in case of error.
+ */
+
+long os_ptrace_ldt(long pid, long addr, long data)
+{
+	int ret;
+
+	ret = ptrace(PTRACE_LDT, pid, addr, data);
+
+	if (ret < 0)
+		return -errno;
+	return ret;
+}
+
 /* Kill off a ptraced child by all means available.  kill it normally first,
  * then PTRACE_KILL it, then PTRACE_CONT it in case it's in a run state from
  * which it can't exit directly.
diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c
index 1fa09a79a10b..fe0877b3509c 100644
--- a/arch/um/sys-i386/ldt.c
+++ b/arch/um/sys-i386/ldt.c
@@ -107,7 +107,7 @@ long write_ldt_entry(struct mm_id * mm_idp, int func, struct user_desc * desc,
 		 * So we need to switch child's mm into our userspace, then
 		 * later switch back.
 		 *
-		 * Note: I'm unshure: should interrupts be disabled here?
+		 * Note: I'm unsure: should interrupts be disabled here?
 		 */
 		if(!current->active_mm || current->active_mm == &init_mm ||
 		   mm_idp != &current->active_mm->context.skas.id)
@@ -129,9 +129,7 @@ long write_ldt_entry(struct mm_id * mm_idp, int func, struct user_desc * desc,
 			pid = userspace_pid[cpu];
 		}
 
-		res = ptrace(PTRACE_LDT, pid, 0, (unsigned long) &ldt_op);
-		if(res)
-			res = errno;
+		res = os_ptrace_ldt(pid, 0, (unsigned long) &ldt_op);
 
 		if(proc_mm)
 			put_cpu();
@@ -181,8 +179,7 @@ static long read_ldt_from_host(void __user * ptr, unsigned long bytecount)
 	 */
 
 	cpu = get_cpu();
-	res = ptrace(PTRACE_LDT, userspace_pid[cpu], 0,
-		     (unsigned long) &ptrace_ldt);
+	res = os_ptrace_ldt(userspace_pid[cpu], 0, (unsigned long) &ptrace_ldt);
 	put_cpu();
 	if(res < 0)
 		goto out;

From 635dd50b7dc69b698e8808ff2802a6cfc31385a8 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Fri, 24 Feb 2006 13:03:55 -0800
Subject: [PATCH 438/509] [PATCH] uml: fix ((unused)) attribute

Use __attribute_used__ instead of __attribute__ ((unused)).  This will help
with GCC > 3.2.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/include/init.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/um/include/init.h b/arch/um/include/init.h
index cbd79a8d213d..d4de7c0120ce 100644
--- a/arch/um/include/init.h
+++ b/arch/um/include/init.h
@@ -122,7 +122,7 @@ extern struct uml_param __uml_setup_start, __uml_setup_end;
 
 #define __exitcall(fn) static exitcall_t __exitcall_##fn __exit_call = fn
 
-#define __init_call __attribute__ ((unused,__section__ (".initcall.init")))
+#define __init_call	__attribute_used__ __attribute__ ((__section__ (".initcall.init")))
 
 #endif
 

From dc1561ac019ff7b6f75c5175abd2ec65c8dbd581 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Fri, 24 Feb 2006 13:03:56 -0800
Subject: [PATCH 439/509] [PATCH] uml: os_connect_socket error path fixup

Fix an fd leak and a return of -1 instead of -errno in the error path - this
showed up in intensive testing of HPPFS, the os_connect_socket user.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/os-Linux/file.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/arch/um/os-Linux/file.c b/arch/um/os-Linux/file.c
index f55773c819e6..3bd10deea280 100644
--- a/arch/um/os-Linux/file.c
+++ b/arch/um/os-Linux/file.c
@@ -272,14 +272,23 @@ int os_connect_socket(char *name)
 	snprintf(sock.sun_path, sizeof(sock.sun_path), "%s", name);
 
 	fd = socket(AF_UNIX, SOCK_STREAM, 0);
-	if(fd < 0)
-		return(fd);
+	if(fd < 0) {
+		err = -errno;
+		goto out;
+	}
 
 	err = connect(fd, (struct sockaddr *) &sock, sizeof(sock));
-	if(err)
-		return(-errno);
+	if(err) {
+		err = -errno;
+		goto out_close;
+	}
 
-	return(fd);
+	return fd;
+
+out_close:
+	close(fd);
+out:
+	return err;
 }
 
 void os_close_file(int fd)

From f462e8f913bdc7a28ce55508d0c045a0c445b157 Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Fri, 24 Feb 2006 13:03:57 -0800
Subject: [PATCH 440/509] [PATCH] uml: better error reporting for read_output

Do precise error handling: print precise error messages, distinguishing short
reads and read errors.  This functions fails frequently enough for me so I
bothered doing this fix.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/net_user.c | 34 ++++++++++++++++++++++------------
 1 file changed, 22 insertions(+), 12 deletions(-)

diff --git a/arch/um/drivers/net_user.c b/arch/um/drivers/net_user.c
index 098fa65981ab..0e2f06187ea7 100644
--- a/arch/um/drivers/net_user.c
+++ b/arch/um/drivers/net_user.c
@@ -47,10 +47,12 @@ void tap_check_ips(char *gate_addr, unsigned char *eth_addr)
 	}
 }
 
+/* Do reliable error handling as this fails frequently enough. */
 void read_output(int fd, char *output, int len)
 {
-	int remain, n, actual;
+	int remain, ret, expected;
 	char c;
+	char *str;
 
 	if(output == NULL){
 		output = &c;
@@ -58,23 +60,31 @@ void read_output(int fd, char *output, int len)
 	}
 		
 	*output = '\0';
-	n = os_read_file(fd, &remain, sizeof(remain));
-	if(n != sizeof(remain)){
-		printk("read_output - read of length failed, err = %d\n", -n);
-		return;
+	ret = os_read_file(fd, &remain, sizeof(remain));
+
+	if (ret != sizeof(remain)) {
+		expected = sizeof(remain);
+		str = "length";
+		goto err;
 	}
 
 	while(remain != 0){
-		n = (remain < len) ? remain : len;
-		actual = os_read_file(fd, output, n);
-		if(actual != n){
-			printk("read_output - read of data failed, "
-			       "err = %d\n", -actual);
-			return;
+		expected = (remain < len) ? remain : len;
+		ret = os_read_file(fd, output, expected);
+		if (ret != expected) {
+			str = "data";
+			goto err;
 		}
-		remain -= actual;
+		remain -= ret;
 	}
+
 	return;
+
+err:
+	if (ret < 0)
+		printk("read_output - read of %s failed, errno = %d\n", str, -ret);
+	else
+		printk("read_output - read of %s failed, read only %d of %d bytes\n", str, ret, expected);
 }
 
 int net_read(int fd, void *buf, int len)

From fe1db50c7222c67466e41241bc7ef17b469bcf1d Mon Sep 17 00:00:00 2001
From: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Date: Fri, 24 Feb 2006 13:03:58 -0800
Subject: [PATCH 441/509] [PATCH] uml: tidying COW code

Improve (especially for coherence) some prototypes, and return code of
init_cow_file in error case - for a short write return -EINVAL, otherwise
return the error we got!

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/um/drivers/cow.h      | 2 +-
 arch/um/drivers/cow_sys.h  | 4 ++--
 arch/um/drivers/cow_user.c | 3 ++-
 arch/um/drivers/ubd_kern.c | 2 +-
 4 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/arch/um/drivers/cow.h b/arch/um/drivers/cow.h
index dc36b222100b..04e3958266e0 100644
--- a/arch/um/drivers/cow.h
+++ b/arch/um/drivers/cow.h
@@ -46,7 +46,7 @@ extern int file_reader(__u64 offset, char *buf, int len, void *arg);
 extern int read_cow_header(int (*reader)(__u64, char *, int, void *),
 			   void *arg, __u32 *version_out,
 			   char **backing_file_out, time_t *mtime_out,
-			   unsigned long long *size_out, int *sectorsize_out,
+			   __u64 *size_out, int *sectorsize_out,
 			   __u32 *align_out, int *bitmap_offset_out);
 
 extern int write_cow_header(char *cow_file, int fd, char *backing_file,
diff --git a/arch/um/drivers/cow_sys.h b/arch/um/drivers/cow_sys.h
index df25263d2ad0..94de4ead4f7a 100644
--- a/arch/um/drivers/cow_sys.h
+++ b/arch/um/drivers/cow_sys.h
@@ -23,12 +23,12 @@ static inline char *cow_strdup(char *str)
 	return(uml_strdup(str));
 }
 
-static inline int cow_seek_file(int fd, unsigned long long offset)
+static inline int cow_seek_file(int fd, __u64 offset)
 {
 	return(os_seek_file(fd, offset));
 }
 
-static inline int cow_file_size(char *file, unsigned long long *size_out)
+static inline int cow_file_size(char *file, __u64 *size_out)
 {
 	return(os_file_size(file, size_out));
 }
diff --git a/arch/um/drivers/cow_user.c b/arch/um/drivers/cow_user.c
index d1c86bc377bd..61951b721268 100644
--- a/arch/um/drivers/cow_user.c
+++ b/arch/um/drivers/cow_user.c
@@ -362,7 +362,8 @@ int init_cow_file(int fd, char *cow_file, char *backing_file, int sectorsize,
 	if(err != sizeof(zero)){
 		cow_printf("Write of bitmap to new COW file '%s' failed, "
 			   "err = %d\n", cow_file, -err);
-		err = -EINVAL;
+		if (err >= 0)
+			err = -EINVAL;
 		goto out;
 	}
 
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index 101efd26d467..fa617e0719ab 100644
--- a/arch/um/drivers/ubd_kern.c
+++ b/arch/um/drivers/ubd_kern.c
@@ -1135,7 +1135,7 @@ static int path_requires_switch(char *from_cmdline, char *from_cow, char *cow)
 static int backing_file_mismatch(char *file, __u64 size, time_t mtime)
 {
 	unsigned long modtime;
-	long long actual;
+	unsigned long long actual;
 	int err;
 
 	err = os_file_modtime(file, &modtime);

From d1521260f57d70d0ba86d2a309ec1ce7979be2fc Mon Sep 17 00:00:00 2001
From: Samuel Thibault <samuel.thibault@ens-lyon.org>
Date: Fri, 24 Feb 2006 13:03:59 -0800
Subject: [PATCH 442/509] [PATCH] vgacon: no vertical resizing on EGA

EGA boards suck: they mostly have write-only registers.  This is
particularly problematic for the overflow register: for being able to write
to it, we would have to handle vertical sync & such too, which (I'd say)
would potentially break a lot of configurations.  Instead, just disabling
vertical resize for EGA boards is just nice enough (horizontal resize still
works).

Fixes http://bugzilla.kernel.org/show_bug.cgi?id=6106

Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Cc: Rafal Olearski <olearski@mail2.kim.net.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/console/vgacon.c | 67 ++++++++++++++++++----------------
 1 file changed, 35 insertions(+), 32 deletions(-)

diff --git a/drivers/video/console/vgacon.c b/drivers/video/console/vgacon.c
index 12d9329d1408..5a86978537d2 100644
--- a/drivers/video/console/vgacon.c
+++ b/drivers/video/console/vgacon.c
@@ -509,57 +509,60 @@ static int vgacon_doresize(struct vc_data *c,
 {
 	unsigned long flags;
 	unsigned int scanlines = height * c->vc_font.height;
-	u8 scanlines_lo, r7, vsync_end, mode, max_scan;
+	u8 scanlines_lo = 0, r7 = 0, vsync_end = 0, mode, max_scan;
 
 	spin_lock_irqsave(&vga_lock, flags);
 
-	outb_p(VGA_CRTC_MAX_SCAN, vga_video_port_reg);
-	max_scan = inb_p(vga_video_port_val);
-
-	if (max_scan & 0x80)
-		scanlines <<= 1;
-
 	vgacon_xres = width * VGA_FONTWIDTH;
 	vgacon_yres = height * c->vc_font.height;
-	outb_p(VGA_CRTC_MODE, vga_video_port_reg);
-	mode = inb_p(vga_video_port_val);
+	if (vga_video_type >= VIDEO_TYPE_VGAC) {
+		outb_p(VGA_CRTC_MAX_SCAN, vga_video_port_reg);
+		max_scan = inb_p(vga_video_port_val);
 
-	if (mode & 0x04)
-		scanlines >>= 1;
+		if (max_scan & 0x80)
+			scanlines <<= 1;
 
-	scanlines -= 1;
-	scanlines_lo = scanlines & 0xff;
+		outb_p(VGA_CRTC_MODE, vga_video_port_reg);
+		mode = inb_p(vga_video_port_val);
 
-	outb_p(VGA_CRTC_OVERFLOW, vga_video_port_reg);
-	r7 = inb_p(vga_video_port_val) & ~0x42;
+		if (mode & 0x04)
+			scanlines >>= 1;
 
-	if (scanlines & 0x100)
-		r7 |= 0x02;
-	if (scanlines & 0x200)
-		r7 |= 0x40;
+		scanlines -= 1;
+		scanlines_lo = scanlines & 0xff;
 
-	/* deprotect registers */
-	outb_p(VGA_CRTC_V_SYNC_END, vga_video_port_reg);
-	vsync_end = inb_p(vga_video_port_val);
-	outb_p(VGA_CRTC_V_SYNC_END, vga_video_port_reg);
-	outb_p(vsync_end & ~0x80, vga_video_port_val);
+		outb_p(VGA_CRTC_OVERFLOW, vga_video_port_reg);
+		r7 = inb_p(vga_video_port_val) & ~0x42;
+
+		if (scanlines & 0x100)
+			r7 |= 0x02;
+		if (scanlines & 0x200)
+			r7 |= 0x40;
+
+		/* deprotect registers */
+		outb_p(VGA_CRTC_V_SYNC_END, vga_video_port_reg);
+		vsync_end = inb_p(vga_video_port_val);
+		outb_p(VGA_CRTC_V_SYNC_END, vga_video_port_reg);
+		outb_p(vsync_end & ~0x80, vga_video_port_val);
+	}
 
 	outb_p(VGA_CRTC_H_DISP, vga_video_port_reg);
 	outb_p(width - 1, vga_video_port_val);
 	outb_p(VGA_CRTC_OFFSET, vga_video_port_reg);
 	outb_p(width >> 1, vga_video_port_val);
 
-	outb_p(VGA_CRTC_V_DISP_END, vga_video_port_reg);
-	outb_p(scanlines_lo, vga_video_port_val);
-	outb_p(VGA_CRTC_OVERFLOW, vga_video_port_reg);
-	outb_p(r7,vga_video_port_val);
+	if (vga_video_type >= VIDEO_TYPE_VGAC) {
+		outb_p(VGA_CRTC_V_DISP_END, vga_video_port_reg);
+		outb_p(scanlines_lo, vga_video_port_val);
+		outb_p(VGA_CRTC_OVERFLOW, vga_video_port_reg);
+		outb_p(r7,vga_video_port_val);
 
-	/* reprotect registers */
-	outb_p(VGA_CRTC_V_SYNC_END, vga_video_port_reg);
-	outb_p(vsync_end, vga_video_port_val);
+		/* reprotect registers */
+		outb_p(VGA_CRTC_V_SYNC_END, vga_video_port_reg);
+		outb_p(vsync_end, vga_video_port_val);
+	}
 
 	spin_unlock_irqrestore(&vga_lock, flags);
-
 	return 0;
 }
 

From 124d90be62343f71bbb7a6b4a907b5584181e6d5 Mon Sep 17 00:00:00 2001
From: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Date: Fri, 24 Feb 2006 13:04:08 -0800
Subject: [PATCH 443/509] [PATCH] Kprobes causes NX protection fault on i686
 SMP

Fix a problem seen on i686 machine with NX support where the instruction
could not be single stepped because of NX bit set on the memory pages
allocated by kprobes module.  This patch provides allocation of instruction
solt so that the processor can execute the instruction from that location
similar to x86_64 architecture.  Thanks to Bibo and Masami for testing this
patch.

Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/kprobes.c | 16 ++++++++++++++--
 include/asm-i386/kprobes.h |  7 +++++--
 2 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index 6483eeb1a4e8..694a13997637 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -58,6 +58,11 @@ static inline int is_IF_modifier(kprobe_opcode_t opcode)
 
 int __kprobes arch_prepare_kprobe(struct kprobe *p)
 {
+	/* insn: must be on special executable page on i386. */
+	p->ainsn.insn = get_insn_slot();
+	if (!p->ainsn.insn)
+		return -ENOMEM;
+
 	memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t));
 	p->opcode = *p->addr;
 	return 0;
@@ -77,6 +82,13 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
 			   (unsigned long) p->addr + sizeof(kprobe_opcode_t));
 }
 
+void __kprobes arch_remove_kprobe(struct kprobe *p)
+{
+	down(&kprobe_mutex);
+	free_insn_slot(p->ainsn.insn);
+	up(&kprobe_mutex);
+}
+
 static inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
 {
 	kcb->prev_kprobe.kp = kprobe_running();
@@ -111,7 +123,7 @@ static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 	if (p->opcode == BREAKPOINT_INSTRUCTION)
 		regs->eip = (unsigned long)p->addr;
 	else
-		regs->eip = (unsigned long)&p->ainsn.insn;
+		regs->eip = (unsigned long)p->ainsn.insn;
 }
 
 /* Called with kretprobe_lock held */
@@ -351,7 +363,7 @@ static void __kprobes resume_execution(struct kprobe *p,
 {
 	unsigned long *tos = (unsigned long *)&regs->esp;
 	unsigned long next_eip = 0;
-	unsigned long copy_eip = (unsigned long)&p->ainsn.insn;
+	unsigned long copy_eip = (unsigned long)p->ainsn.insn;
 	unsigned long orig_eip = (unsigned long)p->addr;
 
 	switch (p->ainsn.insn[0]) {
diff --git a/include/asm-i386/kprobes.h b/include/asm-i386/kprobes.h
index 27cac050a60e..a0d2d74a7dda 100644
--- a/include/asm-i386/kprobes.h
+++ b/include/asm-i386/kprobes.h
@@ -27,6 +27,9 @@
 #include <linux/types.h>
 #include <linux/ptrace.h>
 
+#define  __ARCH_WANT_KPROBES_INSN_SLOT
+
+struct kprobe;
 struct pt_regs;
 
 typedef u8 kprobe_opcode_t;
@@ -40,14 +43,14 @@ typedef u8 kprobe_opcode_t;
 
 #define JPROBE_ENTRY(pentry)	(kprobe_opcode_t *)pentry
 #define ARCH_SUPPORTS_KRETPROBES
-#define arch_remove_kprobe(p)	do {} while (0)
 
+void arch_remove_kprobe(struct kprobe *p);
 void kretprobe_trampoline(void);
 
 /* Architecture specific copy of original instruction*/
 struct arch_specific_insn {
 	/* copy of the original instruction */
-	kprobe_opcode_t insn[MAX_INSN_SIZE];
+	kprobe_opcode_t *insn;
 };
 
 struct prev_kprobe {

From c314b6f1fa462acdb89323c75c597eeaae056e7c Mon Sep 17 00:00:00 2001
From: Simon Vogl <simon.vogl@researchstudios.at>
Date: Fri, 24 Feb 2006 13:04:09 -0800
Subject: [PATCH 444/509] [PATCH] cfi: init wait queue in chip struct

Fix a kernel oops for Intel P30 flashes, where the wait queue head was not
initialized for the flchip struct, which in turn caused a crash at the
first read operation.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/mtd/chips/cfi_cmdset_0001.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index ded2c33f5b85..1c074d63ff3a 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -408,6 +408,7 @@ struct mtd_info *cfi_cmdset_0001(struct map_info *map, int primary)
 		cfi->chips[i].buffer_write_time = 1<<cfi->cfiq->BufWriteTimeoutTyp;
 		cfi->chips[i].erase_time = 1<<cfi->cfiq->BlockEraseTimeoutTyp;
 		cfi->chips[i].ref_point_counter = 0;
+		init_waitqueue_head(&(cfi->chips[i].wq));
 	}
 
 	map->fldrv = &cfi_intelext_chipdrv;

From 8d5c822b2920be9016806f61fd552d2301cfa2fc Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Fri, 24 Feb 2006 13:04:10 -0800
Subject: [PATCH 445/509] [PATCH] voyager: fix boot panic by adding topology
 export

It looks like I can't get away without exporting topology functions from
voyager any longer, so add them to the voyager subarchitecture.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mach-voyager/voyager_basic.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/arch/i386/mach-voyager/voyager_basic.c b/arch/i386/mach-voyager/voyager_basic.c
index aa49a33a572c..6761d294f260 100644
--- a/arch/i386/mach-voyager/voyager_basic.c
+++ b/arch/i386/mach-voyager/voyager_basic.c
@@ -23,6 +23,9 @@
 #include <linux/delay.h>
 #include <linux/reboot.h>
 #include <linux/sysrq.h>
+#include <linux/smp.h>
+#include <linux/nodemask.h>
+#include <asm/cpu.h>
 #include <asm/io.h>
 #include <asm/voyager.h>
 #include <asm/vic.h>
@@ -329,3 +332,15 @@ void machine_power_off(void)
 		pm_power_off();
 }
 
+static struct i386_cpu cpu_devices[NR_CPUS];
+
+static int __init topology_init(void)
+{
+	int i;
+
+	for_each_present_cpu(i)
+		register_cpu(&cpu_devices[i].cpu, i, NULL);
+	return 0;
+}
+
+subsys_initcall(topology_init);

From f68a106f224c21148c5264a429fac149dc7ad0ac Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Fri, 24 Feb 2006 13:04:11 -0800
Subject: [PATCH 446/509] [PATCH] voyager: fix the cpu_possible_map to make
 voyager boot again

Right at the moment (thanks to a patch from Andrew), cpu_possible_map on
voyager is CPU_MASK_NONE, which means the machine always thinks it has no
CPUs.  Fix that by doing an early initialisation of the cpu_possible_map
from the cpu_phys_present_map.

(akpm: we aim to please)

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/mach-voyager/voyager_smp.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/i386/mach-voyager/voyager_smp.c b/arch/i386/mach-voyager/voyager_smp.c
index 6e4c3baef6cc..8165626a5c30 100644
--- a/arch/i386/mach-voyager/voyager_smp.c
+++ b/arch/i386/mach-voyager/voyager_smp.c
@@ -402,6 +402,7 @@ find_smp_config(void)
 	cpus_addr(phys_cpu_present_map)[0] |= voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK + 1) << 8;
 	cpus_addr(phys_cpu_present_map)[0] |= voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK + 2) << 16;
 	cpus_addr(phys_cpu_present_map)[0] |= voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK + 3) << 24;
+	cpu_possible_map = phys_cpu_present_map;
 	printk("VOYAGER SMP: phys_cpu_present_map = 0x%lx\n", cpus_addr(phys_cpu_present_map)[0]);
 	/* Here we set up the VIC to enable SMP */
 	/* enable the CPIs by writing the base vector to their register */

From 1e275d406bf6b88e4de6925cf594b64bb2ec49bc Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Fri, 24 Feb 2006 13:04:12 -0800
Subject: [PATCH 447/509] [PATCH] page migration: Fix MPOL_INTERLEAVE behavior
 for migration via mbind()

migrate_pages_to() allocates a list of new pages on the intended target
node or with the intended policy and then uses the list of new pages as
targets for the migration of a list of pages out of place.

When the pages are allocated it is not clear which of the out of place
pages will be moved to the new pages.  So we cannot specify an address as
needed by alloc_page_vma().  This causes problem for MPOL_INTERLEAVE which
will currently allocate the pages on the first node of the set.  If mbind
is used with vma that has the policy of MPOL_INTERLEAVE then the
interleaving of pages may be destroyed.

This patch fixes that by generating a fake address for each alloc_page_vma
which will result is a distribution of pages as prescribed by
MPOL_INTERLEAVE.

Lee also noted that the sequence of nodes for the new pages seems to be
inverted.  So we also invert the way the lists of pages for migration are
build.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Looks-ok-to: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/mempolicy.c | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 880831bd3003..67af4cea1e23 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -552,7 +552,7 @@ static void migrate_page_add(struct page *page, struct list_head *pagelist,
 	 */
 	if ((flags & MPOL_MF_MOVE_ALL) || page_mapcount(page) == 1) {
 		if (isolate_lru_page(page))
-			list_add(&page->lru, pagelist);
+			list_add_tail(&page->lru, pagelist);
 	}
 }
 
@@ -569,6 +569,7 @@ static int migrate_pages_to(struct list_head *pagelist,
 	LIST_HEAD(moved);
 	LIST_HEAD(failed);
 	int err = 0;
+	unsigned long offset = 0;
 	int nr_pages;
 	struct page *page;
 	struct list_head *p;
@@ -576,8 +577,21 @@ static int migrate_pages_to(struct list_head *pagelist,
 redo:
 	nr_pages = 0;
 	list_for_each(p, pagelist) {
-		if (vma)
-			page = alloc_page_vma(GFP_HIGHUSER, vma, vma->vm_start);
+		if (vma) {
+			/*
+			 * The address passed to alloc_page_vma is used to
+			 * generate the proper interleave behavior. We fake
+			 * the address here by an increasing offset in order
+			 * to get the proper distribution of pages.
+			 *
+			 * No decision has been made as to which page
+			 * a certain old page is moved to so we cannot
+			 * specify the correct address.
+			 */
+			page = alloc_page_vma(GFP_HIGHUSER, vma,
+					offset + vma->vm_start);
+			offset += PAGE_SIZE;
+		}
 		else
 			page = alloc_pages_node(dest, GFP_HIGHUSER, 0);
 
@@ -585,7 +599,7 @@ redo:
 			err = -ENOMEM;
 			goto out;
 		}
-		list_add(&page->lru, &newlist);
+		list_add_tail(&page->lru, &newlist);
 		nr_pages++;
 		if (nr_pages > MIGRATE_CHUNK_SIZE)
 			break;

From 2b932f6cf052920fb3a6281499e08209b08f5086 Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Fri, 24 Feb 2006 13:04:14 -0800
Subject: [PATCH 448/509] [PATCH] x86: fix broken SMP boot sequence

Recent GDT changes broke the SMP boot sequence if the booting CPU is
numbered anything other than zero.  There's also a subtle source of error
in that the boot time CPU now uses cpu_gdt_table (which is actually the GDT
for booting CPUs in head.S).  This patch fixes both problems by making GDT
descriptors themselves allocated from a per_cpu area and switching to them
in cpu_init(), which now means that cpu_gdt_table is exclusively used for
booting CPUs again.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Cc: Zachary Amsden <zach@vmware.com>
Cc: Matt Tolentino <metolent@snoqualmie.dp.intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/cpu/common.c | 32 ++++++++++++++++++++++++++++----
 arch/i386/kernel/efi.c        | 12 +++++++-----
 arch/i386/kernel/head.S       |  2 --
 arch/i386/kernel/i386_ksyms.c |  2 --
 arch/i386/kernel/smpboot.c    |  6 ------
 include/asm-i386/desc.h       |  6 ++++--
 6 files changed, 39 insertions(+), 21 deletions(-)

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 7eb9213734a3..4ecd4b326ded 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -4,6 +4,7 @@
 #include <linux/smp.h>
 #include <linux/module.h>
 #include <linux/percpu.h>
+#include <linux/bootmem.h>
 #include <asm/semaphore.h>
 #include <asm/processor.h>
 #include <asm/i387.h>
@@ -18,6 +19,9 @@
 
 #include "cpu.h"
 
+DEFINE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
+EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr);
+
 DEFINE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]);
 EXPORT_PER_CPU_SYMBOL(cpu_16bit_stack);
 
@@ -571,8 +575,9 @@ void __devinit cpu_init(void)
 	int cpu = smp_processor_id();
 	struct tss_struct * t = &per_cpu(init_tss, cpu);
 	struct thread_struct *thread = &current->thread;
-	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
+	struct desc_struct *gdt;
 	__u32 stk16_off = (__u32)&per_cpu(cpu_16bit_stack, cpu);
+	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
 
 	if (cpu_test_and_set(cpu, cpu_initialized)) {
 		printk(KERN_WARNING "CPU#%d already initialized!\n", cpu);
@@ -589,6 +594,25 @@ void __devinit cpu_init(void)
 		set_in_cr4(X86_CR4_TSD);
 	}
 
+	/*
+	 * This is a horrible hack to allocate the GDT.  The problem
+	 * is that cpu_init() is called really early for the boot CPU
+	 * (and hence needs bootmem) but much later for the secondary
+	 * CPUs, when bootmem will have gone away
+	 */
+	if (NODE_DATA(0)->bdata->node_bootmem_map) {
+		gdt = (struct desc_struct *)alloc_bootmem_pages(PAGE_SIZE);
+		/* alloc_bootmem_pages panics on failure, so no check */
+		memset(gdt, 0, PAGE_SIZE);
+	} else {
+		gdt = (struct desc_struct *)get_zeroed_page(GFP_KERNEL);
+		if (unlikely(!gdt)) {
+			printk(KERN_CRIT "CPU%d failed to allocate GDT\n", cpu);
+			for (;;)
+				local_irq_enable();
+		}
+	}
+
 	/*
 	 * Initialize the per-CPU GDT with the boot GDT,
 	 * and set up the GDT descriptor:
@@ -601,10 +625,10 @@ void __devinit cpu_init(void)
 		((((__u64)stk16_off) << 32) & 0xff00000000000000ULL) |
 		(CPU_16BIT_STACK_SIZE - 1);
 
-	cpu_gdt_descr[cpu].size = GDT_SIZE - 1;
- 	cpu_gdt_descr[cpu].address = (unsigned long)gdt;
+	cpu_gdt_descr->size = GDT_SIZE - 1;
+ 	cpu_gdt_descr->address = (unsigned long)gdt;
 
-	load_gdt(&cpu_gdt_descr[cpu]);
+	load_gdt(cpu_gdt_descr);
 	load_idt(&idt_descr);
 
 	/*
diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c
index ecad519fd395..e3e42fd62401 100644
--- a/arch/i386/kernel/efi.c
+++ b/arch/i386/kernel/efi.c
@@ -103,17 +103,19 @@ static void efi_call_phys_prelog(void)
 	 */
 	local_flush_tlb();
 
-	cpu_gdt_descr[0].address = __pa(cpu_gdt_descr[0].address);
-	load_gdt((struct Xgt_desc_struct *) __pa(&cpu_gdt_descr[0]));
+	per_cpu(cpu_gdt_descr, 0).address =
+				 __pa(per_cpu(cpu_gdt_descr, 0).address);
+	load_gdt((struct Xgt_desc_struct *)__pa(&per_cpu(cpu_gdt_descr, 0)));
 }
 
 static void efi_call_phys_epilog(void)
 {
 	unsigned long cr4;
 
-	cpu_gdt_descr[0].address =
-		(unsigned long) __va(cpu_gdt_descr[0].address);
-	load_gdt(&cpu_gdt_descr[0]);
+	per_cpu(cpu_gdt_descr, 0).address =
+			(unsigned long)__va(per_cpu(cpu_gdt_descr, 0).address);
+	load_gdt((struct Xgt_desc_struct *)__va(&per_cpu(cpu_gdt_descr, 0)));
+
 	cr4 = read_cr4();
 
 	if (cr4 & X86_CR4_PSE) {
diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index 2bee6499edd9..e0b7c632efbc 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -534,5 +534,3 @@ ENTRY(cpu_gdt_table)
 	.quad 0x0000000000000000	/* 0xf0 - unused */
 	.quad 0x0000000000000000	/* 0xf8 - GDT entry 31: double-fault TSS */
 
-	/* Be sure this is zeroed to avoid false validations in Xen */
-	.fill PAGE_SIZE_asm / 8 - GDT_ENTRIES,8,0
diff --git a/arch/i386/kernel/i386_ksyms.c b/arch/i386/kernel/i386_ksyms.c
index 3999bec50c33..055325056a74 100644
--- a/arch/i386/kernel/i386_ksyms.c
+++ b/arch/i386/kernel/i386_ksyms.c
@@ -3,8 +3,6 @@
 #include <asm/checksum.h>
 #include <asm/desc.h>
 
-EXPORT_SYMBOL_GPL(cpu_gdt_descr);
-
 EXPORT_SYMBOL(__down_failed);
 EXPORT_SYMBOL(__down_failed_interruptible);
 EXPORT_SYMBOL(__down_failed_trylock);
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index fb00ab7b7612..eba7f53f8b4a 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -898,12 +898,6 @@ static int __devinit do_boot_cpu(int apicid, int cpu)
 	unsigned long start_eip;
 	unsigned short nmi_high = 0, nmi_low = 0;
 
-	if (!cpu_gdt_descr[cpu].address &&
-	    !(cpu_gdt_descr[cpu].address = get_zeroed_page(GFP_KERNEL))) {
-		printk("Failed to allocate GDT for CPU %d\n", cpu);
-		return 1;
-	}
-
 	++cpucount;
 
 	/*
diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h
index 494e73bca095..89b8b82c82b3 100644
--- a/include/asm-i386/desc.h
+++ b/include/asm-i386/desc.h
@@ -24,11 +24,13 @@ struct Xgt_desc_struct {
 	unsigned short pad;
 } __attribute__ ((packed));
 
-extern struct Xgt_desc_struct idt_descr, cpu_gdt_descr[NR_CPUS];
+extern struct Xgt_desc_struct idt_descr;
+DECLARE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
+
 
 static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 {
-	return ((struct desc_struct *)cpu_gdt_descr[cpu].address);
+	return (struct desc_struct *)per_cpu(cpu_gdt_descr, cpu).address;
 }
 
 #define load_TR_desc() __asm__ __volatile__("ltr %w0"::"q" (GDT_ENTRY_TSS*8))

From 68b06deb2b343c040485a9fc6c813577bf6d5cf5 Mon Sep 17 00:00:00 2001
From: Kaj-Michael Lang <milang@tal.org>
Date: Fri, 24 Feb 2006 13:04:15 -0800
Subject: [PATCH 449/509] [PATCH] gbefb: IP32 gbefb depth change fix

The gbefb driver does not update the framebuffer layers visual setting when
depth is changed with fbset, resulting in strange colors (very dark blue in
16-bit, almost black in 24-bit).

Signed-off-by: Kaj-Michael Lang <milang@tal.org>
Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Antonino Daplas <adaplas@pol.net>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/gbefb.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/video/gbefb.c b/drivers/video/gbefb.c
index c9a7cdf6d543..5e25b9860196 100644
--- a/drivers/video/gbefb.c
+++ b/drivers/video/gbefb.c
@@ -656,12 +656,15 @@ static int gbefb_set_par(struct fb_info *info)
 	switch (bytesPerPixel) {
 	case 1:
 		SET_GBE_FIELD(WID, TYP, val, GBE_CMODE_I8);
+		info->fix.visual = FB_VISUAL_PSEUDOCOLOR;
 		break;
 	case 2:
 		SET_GBE_FIELD(WID, TYP, val, GBE_CMODE_ARGB5);
+		info->fix.visual = FB_VISUAL_TRUECOLOR;
 		break;
 	case 4:
 		SET_GBE_FIELD(WID, TYP, val, GBE_CMODE_RGB8);
+		info->fix.visual = FB_VISUAL_TRUECOLOR;
 		break;
 	}
 	SET_GBE_FIELD(WID, BUF, val, GBE_BMODE_BOTH);

From 80c410dc14f7783411b4becf083069d69daaa4a7 Mon Sep 17 00:00:00 2001
From: Martin Michlmayr <tbm@cyrius.com>
Date: Fri, 24 Feb 2006 13:04:16 -0800
Subject: [PATCH 450/509] [PATCH] gbefb: Set default of FB_GBE_MEM to 4 MB

Allocating more than 4 MB memory for the GBE (SGI O2) framebuffer completely
breakfs gbefb support at the moment.  According to comments on #mipslinux,
more than 4 MB has never worked correctly in Linux.  Therefore, the default
should be 4 MB.

Signed-off-by: Martin Michlmayr <tbm@cyrius.com>
Signed-off-by: Antonino Daplas <adaplas@pol.net>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index e64ed16bd42f..f5079c78ba4e 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -520,7 +520,7 @@ config FB_GBE
 config FB_GBE_MEM
 	int "Video memory size in MB"
 	depends on FB_GBE
-	default 8
+	default 4
 	help
 	  This is the amount of memory reserved for the framebuffer,
 	  which can be any value between 1MB and 8MB.

From cacfc8cf4ed6e05a0d9a8bd17ab85536abd0f6c5 Mon Sep 17 00:00:00 2001
From: Freddy Spierenburg <freddy@dusktilldawn.nl>
Date: Fri, 24 Feb 2006 13:04:17 -0800
Subject: [PATCH 451/509] [PATCH] au1100fb: replaced io_remap_page_range() with
 io_remap_pfn_range()

Replaced the no longer existing io_remap_page_range() routine with the
io_remap_pfn_range() routine.  Did not have a chance yet to test the
functionality of the driver, but at least the kernel compiles cleanly again.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/au1100fb.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/video/au1100fb.c b/drivers/video/au1100fb.c
index 2406899f1207..3d04b2def0f1 100644
--- a/drivers/video/au1100fb.c
+++ b/drivers/video/au1100fb.c
@@ -49,6 +49,7 @@
 #include <linux/interrupt.h>
 #include <linux/ctype.h>
 #include <linux/dma-mapping.h>
+#include <linux/platform_device.h>
 
 #include <asm/mach-au1x00/au1000.h>
 
@@ -406,7 +407,7 @@ int au1100fb_fb_mmap(struct fb_info *fbi, struct vm_area_struct *vma)
 
 	vma->vm_flags |= VM_IO;
 
-	if (io_remap_page_range(vma, vma->vm_start, off,
+	if (io_remap_pfn_range(vma, vma->vm_start, off >> PAGE_SHIFT,
 				vma->vm_end - vma->vm_start,
 				vma->vm_page_prot)) {
 		return -EAGAIN;

From ee713059d4922e4ee17700496d9eb3b95b1ab836 Mon Sep 17 00:00:00 2001
From: "Antonino A. Daplas" <adaplas@gmail.com>
Date: Fri, 24 Feb 2006 13:04:20 -0800
Subject: [PATCH 452/509] [PATCH] Fix pseudo_palette setup in
 asiliantfb_setcolreg()

The setcolreg function will attempt to write 24 color entries to the
pseudo_pallette.  However, the pseudo_palette has only space for 16 entries.

Thanks to Atsushi Nemoto for reporting this bug.

Signed-off-by: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/video/asiliantfb.c | 21 +++++++++------------
 1 file changed, 9 insertions(+), 12 deletions(-)

diff --git a/drivers/video/asiliantfb.c b/drivers/video/asiliantfb.c
index 69f75547865d..c924d81f7978 100644
--- a/drivers/video/asiliantfb.c
+++ b/drivers/video/asiliantfb.c
@@ -322,32 +322,29 @@ static int asiliantfb_setcolreg(u_int regno, u_int red, u_int green, u_int blue,
 	writeb(green, mmio_base + 0x791);
 	writeb(blue, mmio_base + 0x791);
 
-	switch(p->var.bits_per_pixel) {
-	case 15:
-		if (regno < 16) {
+	if (regno < 16) {
+		switch(p->var.red.offset) {
+		case 10: /* RGB 555 */
 			((u32 *)(p->pseudo_palette))[regno] =
 				((red & 0xf8) << 7) |
 				((green & 0xf8) << 2) |
 				((blue & 0xf8) >> 3);
-		}
-		break;
-	case 16:
-		if (regno < 16) {
+			break;
+		case 11: /* RGB 565 */
 			((u32 *)(p->pseudo_palette))[regno] =
 				((red & 0xf8) << 8) |
 				((green & 0xfc) << 3) |
 				((blue & 0xf8) >> 3);
-		}
-		break;
-	case 24:
-		if (regno < 24) {
+			break;
+		case 16: /* RGB 888 */
 			((u32 *)(p->pseudo_palette))[regno] =
 				(red << 16)  |
 				(green << 8) |
 				(blue);
+			break;
 		}
-		break;
 	}
+
 	return 0;
 }
 

From c04030e16dbea2f7581f82cc6688695927f6ac5b Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Fri, 24 Feb 2006 13:04:21 -0800
Subject: [PATCH 453/509] [PATCH] flags parameter for linkat

I'm currently at the POSIX meeting and one thing covered was the
incompatibility of Linux's link() with the POSIX definition.  The name.
Linux does not follow symlinks, POSIX requires it does.

Even if somebody thinks this is a good default behavior we cannot change this
because it would break the ABI.  But the fact remains that some application
might want this behavior.

We have one chance to help implementing this without breaking the behavior.
 For this we could use the new linkat interface which would need a new
flags parameter.  If the new parameter is AT_SYMLINK_FOLLOW the new
behavior could be invoked.

I do not want to introduce such a patch now.  But we could add the
parameter now, just don't use it.  The patch below would do this.  Can we
get this late patch applied before the release more or less fixes the
syscall API?

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/mips/kernel/scall32-o32.S    | 2 +-
 arch/s390/kernel/compat_wrapper.S | 1 +
 fs/namei.c                        | 8 ++++++--
 include/linux/syscalls.h          | 2 +-
 4 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index d83e033dbc87..2f2dc54b2e26 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -626,7 +626,7 @@ einval:	li	v0, -EINVAL
 	sys	sys_fstatat64		4
 	sys	sys_unlinkat		3
 	sys	sys_renameat		4	/* 4295 */
-	sys	sys_linkat		4
+	sys	sys_linkat		5
 	sys	sys_symlinkat		3
 	sys	sys_readlinkat		4
 	sys	sys_fchmodat		3
diff --git a/arch/s390/kernel/compat_wrapper.S b/arch/s390/kernel/compat_wrapper.S
index 615964cca15f..50e80138e7ad 100644
--- a/arch/s390/kernel/compat_wrapper.S
+++ b/arch/s390/kernel/compat_wrapper.S
@@ -1552,6 +1552,7 @@ sys_linkat_wrapper:
 	llgtr	%r3,%r3			# const char *
 	lgfr	%r4,%r4			# int
 	llgtr	%r5,%r5			# const char *
+	lgfr	%r6,%r6			# int
 	jg	sys_linkat
 
 	.globl sys_symlinkat_wrapper
diff --git a/fs/namei.c b/fs/namei.c
index e28de846c591..557dcf395ca1 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2224,13 +2224,17 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de
  * and other special files.  --ADM
  */
 asmlinkage long sys_linkat(int olddfd, const char __user *oldname,
-			   int newdfd, const char __user *newname)
+			   int newdfd, const char __user *newname,
+			   int flags)
 {
 	struct dentry *new_dentry;
 	struct nameidata nd, old_nd;
 	int error;
 	char * to;
 
+	if (flags != 0)
+		return -EINVAL;
+
 	to = getname(newname);
 	if (IS_ERR(to))
 		return PTR_ERR(to);
@@ -2263,7 +2267,7 @@ exit:
 
 asmlinkage long sys_link(const char __user *oldname, const char __user *newname)
 {
-	return sys_linkat(AT_FDCWD, oldname, AT_FDCWD, newname);
+	return sys_linkat(AT_FDCWD, oldname, AT_FDCWD, newname, 0);
 }
 
 /*
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index d73501ba7e44..b9ea44ac0ddb 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -543,7 +543,7 @@ asmlinkage long sys_unlinkat(int dfd, const char __user * pathname, int flag);
 asmlinkage long sys_symlinkat(const char __user * oldname,
 			      int newdfd, const char __user * newname);
 asmlinkage long sys_linkat(int olddfd, const char __user *oldname,
-			   int newdfd, const char __user *newname);
+			   int newdfd, const char __user *newname, int flags);
 asmlinkage long sys_renameat(int olddfd, const char __user * oldname,
 			     int newdfd, const char __user * newname);
 asmlinkage long sys_futimesat(int dfd, char __user *filename,

From d4f7796e9b387e471ab0e8ed4e0c2bd616b3c193 Mon Sep 17 00:00:00 2001
From: Christoph Lameter <clameter@engr.sgi.com>
Date: Fri, 24 Feb 2006 13:04:22 -0800
Subject: [PATCH 454/509] [PATCH] vmscan: fix zone_reclaim

- PF_SWAPWRITE needs to be set for RECLAIM_SWAP to be able to write
  out pages to swap. Currently RECLAIM_SWAP may not do that.

- remove setting nr_reclaimed pages after slab reclaim since the slab shrinking
  code does not use that and the nr_reclaimed pages is just right for the
  intended follow up action.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 mm/vmscan.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/mm/vmscan.c b/mm/vmscan.c
index 1838c15ca4fd..b0af7593d01e 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1908,7 +1908,12 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 		sc.swap_cluster_max = SWAP_CLUSTER_MAX;
 
 	cond_resched();
-	p->flags |= PF_MEMALLOC;
+	/*
+	 * We need to be able to allocate from the reserves for RECLAIM_SWAP
+	 * and we also need to be able to write out pages for RECLAIM_WRITE
+	 * and RECLAIM_SWAP.
+	 */
+	p->flags |= PF_MEMALLOC | PF_SWAPWRITE;
 	reclaim_state.reclaimed_slab = 0;
 	p->reclaim_state = &reclaim_state;
 
@@ -1932,11 +1937,10 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order)
 		 * a long time.
 		 */
 		shrink_slab(sc.nr_scanned, gfp_mask, order);
-		sc.nr_reclaimed = 1;    /* Avoid getting the off node timeout */
 	}
 
 	p->reclaim_state = NULL;
-	current->flags &= ~PF_MEMALLOC;
+	current->flags &= ~(PF_MEMALLOC | PF_SWAPWRITE);
 
 	if (sc.nr_reclaimed == 0)
 		zone->last_unsuccessful_zone_reclaim = jiffies;

From 8dde0509e74ff6044cf1788c917a22facce9f68d Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@osdl.org>
Date: Fri, 24 Feb 2006 13:04:23 -0800
Subject: [PATCH 455/509] [PATCH] ramfs: update dir mtime and ctime

Phil Marek <philipp.marek@bmlv.gv.at> points out that ramfs forgets to update
a directory's mtime and ctime when it is modified.

Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/ramfs/inode.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c
index c66bd5e4c05c..cde5d48994ae 100644
--- a/fs/ramfs/inode.c
+++ b/fs/ramfs/inode.c
@@ -27,6 +27,7 @@
 #include <linux/fs.h>
 #include <linux/pagemap.h>
 #include <linux/highmem.h>
+#include <linux/time.h>
 #include <linux/init.h>
 #include <linux/string.h>
 #include <linux/smp_lock.h>
@@ -104,6 +105,7 @@ ramfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev)
 		d_instantiate(dentry, inode);
 		dget(dentry);	/* Extra count - pin the dentry in core */
 		error = 0;
+		dir->i_mtime = dir->i_ctime = CURRENT_TIME;
 	}
 	return error;
 }

From d9dde59ba03095e526640988c0fedd75e93bc8b7 Mon Sep 17 00:00:00 2001
From: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Date: Fri, 24 Feb 2006 13:04:24 -0800
Subject: [PATCH 456/509] [PATCH] dm: missing bdput/thaw_bdev at removal

Need to unfreeze and release bdev otherwise the bdev inode with
inconsistent state is reused later and cause problem.

Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Acked-by: Alasdair G Kergon <agk@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/dm.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index e9adeb9d172f..dad9e403d59d 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -849,6 +849,10 @@ static struct mapped_device *alloc_dev(unsigned int minor, int persistent)
 
 static void free_dev(struct mapped_device *md)
 {
+	if (md->suspended_bdev) {
+		thaw_bdev(md->suspended_bdev, NULL);
+		bdput(md->suspended_bdev);
+	}
 	free_minor(md->disk->first_minor);
 	mempool_destroy(md->tio_pool);
 	mempool_destroy(md->io_pool);

From 63d94e482df769f31e8b1097f06c3a3fba7bced4 Mon Sep 17 00:00:00 2001
From: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Date: Fri, 24 Feb 2006 13:04:25 -0800
Subject: [PATCH 457/509] [PATCH] dm: free minor after unlink gendisk

Minor number should be freed after del_gendisk().  Otherwise, there could
be a window where 2 registered gendisk has same minor number.

Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Acked-by: Alasdair G Kergon <agk@redhat.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/md/dm.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index dad9e403d59d..745ca1f67b14 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -849,14 +849,16 @@ static struct mapped_device *alloc_dev(unsigned int minor, int persistent)
 
 static void free_dev(struct mapped_device *md)
 {
+	unsigned int minor = md->disk->first_minor;
+
 	if (md->suspended_bdev) {
 		thaw_bdev(md->suspended_bdev, NULL);
 		bdput(md->suspended_bdev);
 	}
-	free_minor(md->disk->first_minor);
 	mempool_destroy(md->tio_pool);
 	mempool_destroy(md->io_pool);
 	del_gendisk(md->disk);
+	free_minor(minor);
 	put_disk(md->disk);
 	blk_put_queue(md->queue);
 	kfree(md);

From ad329b1519c0091806046b0e49ab073ea590dc11 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Fri, 24 Feb 2006 13:04:26 -0800
Subject: [PATCH 458/509] [PATCH] tmpfs: recommend remount for mpol

akpm points out that switching to a non-NUMA kernel could be irritating
if mounting tmpfs fails on an mpol option: tmpfs.txt recommend remount.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/filesystems/tmpfs.txt | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/Documentation/filesystems/tmpfs.txt b/Documentation/filesystems/tmpfs.txt
index 8a155418c705..1773106976a2 100644
--- a/Documentation/filesystems/tmpfs.txt
+++ b/Documentation/filesystems/tmpfs.txt
@@ -92,6 +92,15 @@ NodeList format is a comma-separated list of decimal numbers and ranges,
 a range being two hyphen-separated decimal numbers, the smallest and
 largest node numbers in the range.  For example, mpol=bind:0-3,5,7,9-15
 
+Note that trying to mount a tmpfs with an mpol option will fail if the
+running kernel does not support NUMA; and will fail if its nodelist
+specifies a node >= MAX_NUMNODES.  If your system relies on that tmpfs
+being mounted, but from time to time runs a kernel built without NUMA
+capability (perhaps a safe recovery kernel), or configured to support
+fewer nodes, then it is advisable to omit the mpol option from automatic
+mount options.  It can be added later, when the tmpfs is already mounted
+on MountPoint, by 'mount -o remount,mpol=Policy:NodeList MountPoint'.
+
 
 To specify the initial root directory you can use the following mount
 options:

From 9c869edac591977314323a4eaad5f7633fca684f Mon Sep 17 00:00:00 2001
From: Zachary Amsden <zach@vmware.com>
Date: Fri, 24 Feb 2006 13:04:27 -0800
Subject: [PATCH 459/509] [PATCH] Fix topology.c location

When compiling a non-default subarch, topology.c is missing from the kernel
build.  This causes builds with CONFIG_HOTPLUG_CPU to fail.  In addition,
on Intel processors with cpuid level > 4, it causes intel_cacheinfo.c to
reference uninitialized data that should have been set up by the initcall
in topology.c which calls register_cpu.  This causes a kernel panic on boot
on newer Intel processors.  Moving topology.c to arch/i386/kernel fixes
both of these problems.

Thanks to Dan Hecht for finding and fixing this problem.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Dan Hecht <dhect@vmware.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/Makefile                     | 2 +-
 arch/i386/{mach-default => kernel}/topology.c | 6 +++---
 arch/i386/mach-default/Makefile               | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)
 rename arch/i386/{mach-default => kernel}/topology.c (94%)

diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index 60c3f76dfca4..53bb9a79e274 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -7,7 +7,7 @@ extra-y := head.o init_task.o vmlinux.lds
 obj-y	:= process.o semaphore.o signal.o entry.o traps.o irq.o \
 		ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_i386.o \
 		pci-dma.o i386_ksyms.o i387.o dmi_scan.o bootflag.o \
-		quirks.o i8237.o
+		quirks.o i8237.o topology.o
 
 obj-y				+= cpu/
 obj-y				+= timers/
diff --git a/arch/i386/mach-default/topology.c b/arch/i386/kernel/topology.c
similarity index 94%
rename from arch/i386/mach-default/topology.c
rename to arch/i386/kernel/topology.c
index b64314069e78..67a0e1baa28b 100644
--- a/arch/i386/mach-default/topology.c
+++ b/arch/i386/kernel/topology.c
@@ -1,12 +1,12 @@
 /*
- * arch/i386/mach-generic/topology.c - Populate driverfs with topology information
+ * arch/i386/kernel/topology.c - Populate driverfs with topology information
  *
  * Written by: Matthew Dobson, IBM Corporation
  * Original Code: Paul Dorwin, IBM Corporation, Patrick Mochel, OSDL
  *
  * Copyright (C) 2002, IBM Corp.
  *
- * All rights reserved.          
+ * All rights reserved.
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
@@ -34,7 +34,7 @@ static struct i386_cpu cpu_devices[NR_CPUS];
 
 int arch_register_cpu(int num){
 	struct node *parent = NULL;
-	
+
 #ifdef CONFIG_NUMA
 	int node = cpu_to_node(num);
 	if (node_online(node))
diff --git a/arch/i386/mach-default/Makefile b/arch/i386/mach-default/Makefile
index e95bb0237921..012fe34459e6 100644
--- a/arch/i386/mach-default/Makefile
+++ b/arch/i386/mach-default/Makefile
@@ -2,4 +2,4 @@
 # Makefile for the linux kernel.
 #
 
-obj-y				:= setup.o topology.o
+obj-y				:= setup.o

From 329dda083e496bc5ffbb4b1973243bd8a9420e24 Mon Sep 17 00:00:00 2001
From: Kumar Gala <galak@kernel.crashing.org>
Date: Fri, 24 Feb 2006 10:54:52 -0600
Subject: [PATCH 460/509] [PATCH] powerpc: Fix mem= cmdline handling on
 arch/powerpc for !MULTIPLATFORM

mem= command line option was being ignored in arch/powerpc if we were not
a CONFIG_MULTIPLATFORM (which is handled via prom_init stub). The initial
command line extraction and parsing needed to be moved earlier in the boot
process and have code to actual parse mem= and do something about it.

Also, fixed a compile warning in the file.

Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Acked-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/powerpc/kernel/prom.c | 54 +++++++++++++++++++++++++++-----------
 1 file changed, 38 insertions(+), 16 deletions(-)

diff --git a/arch/powerpc/kernel/prom.c b/arch/powerpc/kernel/prom.c
index 294832a7e0a6..6dbd21726770 100644
--- a/arch/powerpc/kernel/prom.c
+++ b/arch/powerpc/kernel/prom.c
@@ -816,8 +816,6 @@ void __init unflatten_device_tree(void)
 {
 	unsigned long start, mem, size;
 	struct device_node **allnextp = &allnodes;
-	char *p = NULL;
-	int l = 0;
 
 	DBG(" -> unflatten_device_tree()\n");
 
@@ -857,19 +855,6 @@ void __init unflatten_device_tree(void)
 	if (of_chosen == NULL)
 		of_chosen = of_find_node_by_path("/chosen@0");
 
-	/* Retreive command line */
-	if (of_chosen != NULL) {
-		p = (char *)get_property(of_chosen, "bootargs", &l);
-		if (p != NULL && l > 0)
-			strlcpy(cmd_line, p, min(l, COMMAND_LINE_SIZE));
-	}
-#ifdef CONFIG_CMDLINE
-	if (l == 0 || (l == 1 && (*p) == 0))
-		strlcpy(cmd_line, CONFIG_CMDLINE, COMMAND_LINE_SIZE);
-#endif /* CONFIG_CMDLINE */
-
-	DBG("Command line is: %s\n", cmd_line);
-
 	DBG(" <- unflatten_device_tree()\n");
 }
 
@@ -940,6 +925,8 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
 {
 	u32 *prop;
 	unsigned long *lprop;
+	unsigned long l;
+	char *p;
 
 	DBG("search \"chosen\", depth: %d, uname: %s\n", depth, uname);
 
@@ -1004,6 +991,41 @@ static int __init early_init_dt_scan_chosen(unsigned long node,
                crashk_res.end = crashk_res.start + *lprop - 1;
 #endif
 
+	/* Retreive command line */
+ 	p = of_get_flat_dt_prop(node, "bootargs", &l);
+	if (p != NULL && l > 0)
+		strlcpy(cmd_line, p, min((int)l, COMMAND_LINE_SIZE));
+
+#ifdef CONFIG_CMDLINE
+	if (l == 0 || (l == 1 && (*p) == 0))
+		strlcpy(cmd_line, CONFIG_CMDLINE, COMMAND_LINE_SIZE);
+#endif /* CONFIG_CMDLINE */
+
+	DBG("Command line is: %s\n", cmd_line);
+
+	if (strstr(cmd_line, "mem=")) {
+		char *p, *q;
+		unsigned long maxmem = 0;
+
+		for (q = cmd_line; (p = strstr(q, "mem=")) != 0; ) {
+			q = p + 4;
+			if (p > cmd_line && p[-1] != ' ')
+				continue;
+			maxmem = simple_strtoul(q, &q, 0);
+			if (*q == 'k' || *q == 'K') {
+				maxmem <<= 10;
+				++q;
+			} else if (*q == 'm' || *q == 'M') {
+				maxmem <<= 20;
+				++q;
+			} else if (*q == 'g' || *q == 'G') {
+				maxmem <<= 30;
+				++q;
+			}
+		}
+		memory_limit = maxmem;
+	}
+
 	/* break now */
 	return 1;
 }
@@ -1124,7 +1146,7 @@ static void __init early_reserve_mem(void)
 			size_32 = *(reserve_map_32++);
 			if (size_32 == 0)
 				break;
-			DBG("reserving: %lx -> %lx\n", base_32, size_32);
+			DBG("reserving: %x -> %x\n", base_32, size_32);
 			lmb_reserve(base_32, size_32);
 		}
 		return;

From a0124d780d06db711e8a92135d774940588a27da Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Sat, 25 Feb 2006 03:55:38 -0500
Subject: [PATCH 461/509] [PATCH] x86-64: react to new topology.c location

Commit 9c869edac591977314323a4eaad5f7633fca684f moved the i386 topology.c
file. That change broke x86-64 compiles, as it uses the same file.

Signed-off-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile
index 72fe60c20d39..a098a11e7755 100644
--- a/arch/x86_64/kernel/Makefile
+++ b/arch/x86_64/kernel/Makefile
@@ -43,7 +43,7 @@ CFLAGS_vsyscall.o		:= $(PROFILING) -g0
 
 bootflag-y			+= ../../i386/kernel/bootflag.o
 cpuid-$(subst m,y,$(CONFIG_X86_CPUID))  += ../../i386/kernel/cpuid.o
-topology-y                     += ../../i386/mach-default/topology.o
+topology-y                     += ../../i386/kernel/topology.o
 microcode-$(subst m,y,$(CONFIG_MICROCODE))  += ../../i386/kernel/microcode.o
 intel_cacheinfo-y		+= ../../i386/kernel/cpu/intel_cacheinfo.o
 quirks-y			+= ../../i386/kernel/quirks.o

From 0ee304d5802dc62746f13f12d4cb4ec4ed285f66 Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Sat, 25 Feb 2006 13:52:30 +0900
Subject: [PATCH 462/509] [PATCH] sata_sil: add board ID for 3512

3512 is slightly different from 3112 errata-wise.  Differentiate it.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/scsi/sata_sil.c | 21 +++++++++++++++++----
 1 file changed, 17 insertions(+), 4 deletions(-)

diff --git a/drivers/scsi/sata_sil.c b/drivers/scsi/sata_sil.c
index 17f74d3c10e7..510c2e0bd90e 100644
--- a/drivers/scsi/sata_sil.c
+++ b/drivers/scsi/sata_sil.c
@@ -53,7 +53,8 @@ enum {
 
 	sil_3112		= 0,
 	sil_3112_m15w		= 1,
-	sil_3114		= 2,
+	sil_3512		= 2,
+	sil_3114		= 3,
 
 	SIL_FIFO_R0		= 0x40,
 	SIL_FIFO_W0		= 0x41,
@@ -90,7 +91,7 @@ static void sil_post_set_mode (struct ata_port *ap);
 static const struct pci_device_id sil_pci_tbl[] = {
 	{ 0x1095, 0x3112, PCI_ANY_ID, PCI_ANY_ID, 0, 0, sil_3112_m15w },
 	{ 0x1095, 0x0240, PCI_ANY_ID, PCI_ANY_ID, 0, 0, sil_3112_m15w },
-	{ 0x1095, 0x3512, PCI_ANY_ID, PCI_ANY_ID, 0, 0, sil_3112 },
+	{ 0x1095, 0x3512, PCI_ANY_ID, PCI_ANY_ID, 0, 0, sil_3512 },
 	{ 0x1095, 0x3114, PCI_ANY_ID, PCI_ANY_ID, 0, 0, sil_3114 },
 	{ 0x1002, 0x436e, PCI_ANY_ID, PCI_ANY_ID, 0, 0, sil_3112_m15w },
 	{ 0x1002, 0x4379, PCI_ANY_ID, PCI_ANY_ID, 0, 0, sil_3112_m15w },
@@ -185,7 +186,8 @@ static const struct ata_port_info sil_port_info[] = {
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
 		.port_ops	= &sil_ops,
-	}, /* sil_3112_15w - keep it sync'd w/ sil_3112 */
+	},
+	/* sil_3112_15w - keep it sync'd w/ sil_3112 */
 	{
 		.sht		= &sil_sht,
 		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
@@ -195,7 +197,18 @@ static const struct ata_port_info sil_port_info[] = {
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
 		.port_ops	= &sil_ops,
-	}, /* sil_3114 */
+	},
+	/* sil_3512 */
+	{
+		.sht		= &sil_sht,
+		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
+				  ATA_FLAG_SRST | ATA_FLAG_MMIO,
+		.pio_mask	= 0x1f,			/* pio0-4 */
+		.mwdma_mask	= 0x07,			/* mwdma0-2 */
+		.udma_mask	= 0x3f,			/* udma0-5 */
+		.port_ops	= &sil_ops,
+	},
+	/* sil_3114 */
 	{
 		.sht		= &sil_sht,
 		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |

From e4e10e3e7995f5bd481d2720bf30d3a661d110ca Mon Sep 17 00:00:00 2001
From: Tejun Heo <htejun@gmail.com>
Date: Sat, 25 Feb 2006 13:52:30 +0900
Subject: [PATCH 463/509] [PATCH] sata_sil: implement R_ERR on DMA activate FIS
 errata fix

Silicon Image has disclosed a new sil3114/3152 errata and workaround
which causes the controller to return R_ERR on DMA activate FIS if the
FIS is received while the next PRD is being fetched.  This patch
implements the workaround.

This errata results in lock up and doesn't trigger if m15w workaround
is in effect.  We stopped applying m15w to 3512 and 3114 in 2.6.14-rc1
which makes 3512/3114 lock up with some drives on all kernel versions
since 2.6.14-rc1 upto now (2.6.16-rc4).  This patch should fix the
regression.

Signed-off-by: Tejun Heo <htejun@gmail.com>
Signed-off-by: Jeff Garzik <jeff@garzik.org>
---
 drivers/scsi/sata_sil.c | 33 +++++++++++++++++++++++++++------
 1 file changed, 27 insertions(+), 6 deletions(-)

diff --git a/drivers/scsi/sata_sil.c b/drivers/scsi/sata_sil.c
index 510c2e0bd90e..9face3c6aa21 100644
--- a/drivers/scsi/sata_sil.c
+++ b/drivers/scsi/sata_sil.c
@@ -49,6 +49,7 @@
 #define DRV_VERSION	"0.9"
 
 enum {
+	SIL_FLAG_RERR_ON_DMA_ACT = (1 << 29),
 	SIL_FLAG_MOD15WRITE	= (1 << 30),
 
 	sil_3112		= 0,
@@ -202,7 +203,8 @@ static const struct ata_port_info sil_port_info[] = {
 	{
 		.sht		= &sil_sht,
 		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
-				  ATA_FLAG_SRST | ATA_FLAG_MMIO,
+				  ATA_FLAG_SRST | ATA_FLAG_MMIO |
+				  SIL_FLAG_RERR_ON_DMA_ACT,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
@@ -212,7 +214,8 @@ static const struct ata_port_info sil_port_info[] = {
 	{
 		.sht		= &sil_sht,
 		.host_flags	= ATA_FLAG_SATA | ATA_FLAG_NO_LEGACY |
-				  ATA_FLAG_SRST | ATA_FLAG_MMIO,
+				  ATA_FLAG_SRST | ATA_FLAG_MMIO |
+				  SIL_FLAG_RERR_ON_DMA_ACT,
 		.pio_mask	= 0x1f,			/* pio0-4 */
 		.mwdma_mask	= 0x07,			/* mwdma0-2 */
 		.udma_mask	= 0x3f,			/* udma0-5 */
@@ -229,12 +232,13 @@ static const struct {
 	unsigned long scr;	/* SATA control register block */
 	unsigned long sien;	/* SATA Interrupt Enable register */
 	unsigned long xfer_mode;/* data transfer mode register */
+	unsigned long sfis_cfg;	/* SATA FIS reception config register */
 } sil_port[] = {
 	/* port 0 ... */
-	{ 0x80, 0x8A, 0x00, 0x100, 0x148, 0xb4 },
-	{ 0xC0, 0xCA, 0x08, 0x180, 0x1c8, 0xf4 },
-	{ 0x280, 0x28A, 0x200, 0x300, 0x348, 0x2b4 },
-	{ 0x2C0, 0x2CA, 0x208, 0x380, 0x3c8, 0x2f4 },
+	{ 0x80, 0x8A, 0x00, 0x100, 0x148, 0xb4, 0x14c },
+	{ 0xC0, 0xCA, 0x08, 0x180, 0x1c8, 0xf4, 0x1cc },
+	{ 0x280, 0x28A, 0x200, 0x300, 0x348, 0x2b4, 0x34c },
+	{ 0x2C0, 0x2CA, 0x208, 0x380, 0x3c8, 0x2f4, 0x3cc },
 	/* ... port 3 */
 };
 
@@ -484,6 +488,23 @@ static int sil_init_one (struct pci_dev *pdev, const struct pci_device_id *ent)
 		dev_printk(KERN_WARNING, &pdev->dev,
 			 "cache line size not set.  Driver may not function\n");
 
+	/* Apply R_ERR on DMA activate FIS errata workaround */
+	if (probe_ent->host_flags & SIL_FLAG_RERR_ON_DMA_ACT) {
+		int cnt;
+
+		for (i = 0, cnt = 0; i < probe_ent->n_ports; i++) {
+			tmp = readl(mmio_base + sil_port[i].sfis_cfg);
+			if ((tmp & 0x3) != 0x01)
+				continue;
+			if (!cnt)
+				dev_printk(KERN_INFO, &pdev->dev,
+					   "Applying R_ERR on DMA activate "
+					   "FIS errata fix\n");
+			writel(tmp & ~0x3, mmio_base + sil_port[i].sfis_cfg);
+			cnt++;
+		}
+	}
+
 	if (ent->driver_data == sil_3114) {
 		irq_mask = SIL_MASK_4PORT;
 

From 7bef4b397874eee4484457040e8a1013361d7758 Mon Sep 17 00:00:00 2001
From: Daniele Venzano <venza@brownhat.org>
Date: Sat, 25 Feb 2006 17:01:09 -0500
Subject: [PATCH 464/509] Fix Wake on LAN support in sis900

Fix two bugs in the WoL implementation of sis900. The first causes
hangs on some system on driver load, the second causes troubles
when disabling WoL support.  Both fixes are one liner and really
simple.

Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Daniele Venzano <venza@brownhat.org>
---
 drivers/net/sis900.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/sis900.c b/drivers/net/sis900.c
index 3d95fa20cd88..7a952fe60be2 100644
--- a/drivers/net/sis900.c
+++ b/drivers/net/sis900.c
@@ -540,7 +540,7 @@ static int __devinit sis900_probe(struct pci_dev *pci_dev,
 	printk("%2.2x.\n", net_dev->dev_addr[i]);
 
 	/* Detect Wake on Lan support */
-	ret = inl(CFGPMC & PMESP);
+	ret = (inl(net_dev->base_addr + CFGPMC) & PMESP) >> 27;
 	if (netif_msg_probe(sis_priv) && (ret & PME_D3C) == 0)
 		printk(KERN_INFO "%s: Wake on LAN only available from suspend to RAM.", net_dev->name);
 
@@ -2040,7 +2040,7 @@ static int sis900_set_wol(struct net_device *net_dev, struct ethtool_wolinfo *wo
 
 	if (wol->wolopts == 0) {
 		pci_read_config_dword(sis_priv->pci_dev, CFGPMCSR, &cfgpmcsr);
-		cfgpmcsr |= ~PME_EN;
+		cfgpmcsr &= ~PME_EN;
 		pci_write_config_dword(sis_priv->pci_dev, CFGPMCSR, cfgpmcsr);
 		outl(pmctrl_bits, pmctrl_addr);
 		if (netif_msg_wol(sis_priv))

From 489708007785389941a89fa06aedc5ec53303c96 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@ftp.linux.org.uk>
Date: Sun, 26 Feb 2006 08:34:10 -0600
Subject: [PATCH 465/509] [PATCH] sd: fix memory corruption with broken mode
 page headers

There's a problem in sd where we blindly believe the length of the
headers and block descriptors.  Some devices return insane values for
these and cause our length to end up greater than the actual buffer
size, so check to make sure.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

Also removed the buffer size magic number (512) and added DPOFUA of
zero to the defaults

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/scsi/sd.c | 20 +++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 930db398d107..9d9872347f56 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -89,6 +89,11 @@
 #define SD_MAX_RETRIES		5
 #define SD_PASSTHROUGH_RETRIES	1
 
+/*
+ * Size of the initial data buffer for mode and read capacity data
+ */
+#define SD_BUF_SIZE		512
+
 static void scsi_disk_release(struct kref *kref);
 
 struct scsi_disk {
@@ -1239,7 +1244,7 @@ sd_do_mode_sense(struct scsi_device *sdp, int dbd, int modepage,
 
 /*
  * read write protect setting, if possible - called only in sd_revalidate_disk()
- * called with buffer of length 512
+ * called with buffer of length SD_BUF_SIZE
  */
 static void
 sd_read_write_protect_flag(struct scsi_disk *sdkp, char *diskname,
@@ -1297,7 +1302,7 @@ sd_read_write_protect_flag(struct scsi_disk *sdkp, char *diskname,
 
 /*
  * sd_read_cache_type - called only from sd_revalidate_disk()
- * called with buffer of length 512
+ * called with buffer of length SD_BUF_SIZE
  */
 static void
 sd_read_cache_type(struct scsi_disk *sdkp, char *diskname,
@@ -1342,6 +1347,8 @@ sd_read_cache_type(struct scsi_disk *sdkp, char *diskname,
 
 	/* Take headers and block descriptors into account */
 	len += data.header_length + data.block_descriptor_length;
+	if (len > SD_BUF_SIZE)
+		goto bad_sense;
 
 	/* Get the data */
 	res = sd_do_mode_sense(sdp, dbd, modepage, buffer, len, &data, &sshdr);
@@ -1354,6 +1361,12 @@ sd_read_cache_type(struct scsi_disk *sdkp, char *diskname,
 		int ct = 0;
 		int offset = data.header_length + data.block_descriptor_length;
 
+		if (offset >= SD_BUF_SIZE - 2) {
+			printk(KERN_ERR "%s: malformed MODE SENSE response",
+				diskname);
+			goto defaults;
+		}
+
 		if ((buffer[offset] & 0x3f) != modepage) {
 			printk(KERN_ERR "%s: got wrong page\n", diskname);
 			goto defaults;
@@ -1398,6 +1411,7 @@ defaults:
 	       diskname);
 	sdkp->WCE = 0;
 	sdkp->RCD = 0;
+	sdkp->DPOFUA = 0;
 }
 
 /**
@@ -1421,7 +1435,7 @@ static int sd_revalidate_disk(struct gendisk *disk)
 	if (!scsi_device_online(sdp))
 		goto out;
 
-	buffer = kmalloc(512, GFP_KERNEL | __GFP_DMA);
+	buffer = kmalloc(SD_BUF_SIZE, GFP_KERNEL | __GFP_DMA);
 	if (!buffer) {
 		printk(KERN_WARNING "(sd_revalidate_disk:) Memory allocation "
 		       "failure.\n");

From 04a3d311c01d3ad287750c5c8d03fa614475af91 Mon Sep 17 00:00:00 2001
From: Marc Zyngier <maz@misterjones.org>
Date: Sun, 26 Feb 2006 12:02:56 +0100
Subject: [PATCH 466/509] [PATCH] Fix Specialix SI probing

As the (probably) last user of a Specialix SI board, I noticed that
recent kernels would fail to probe the sucker.  Quick investigation
indicate a few missing braces...

I left the double probing in place, as it looks like it's been here
forever.

Signed-off-by: Marc Zyngier <maz@misterjones.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/sx.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index c2490e270f1f..588e75ec1630 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -2173,15 +2173,17 @@ static int probe_si (struct sx_board *board)
 	if ( IS_SI1_BOARD(board)) {
 		/* This should be an SI1 board, which has this
 		   location writable... */
-		if (read_sx_byte (board, SI2_ISA_ID_BASE) != 0x10)
+		if (read_sx_byte (board, SI2_ISA_ID_BASE) != 0x10) {
 			func_exit ();
 			return 0; 
+		}
 	} else {
 		/* This should be an SI2 board, which has the bottom
 		   3 bits non-writable... */
-		if (read_sx_byte (board, SI2_ISA_ID_BASE) == 0x10)
+		if (read_sx_byte (board, SI2_ISA_ID_BASE) == 0x10) {
 			func_exit ();
 			return 0; 
+		}
 	}
 
 	/* Now we're pretty much convinced that there is an SI board here, 
@@ -2192,15 +2194,17 @@ static int probe_si (struct sx_board *board)
 	if ( IS_SI1_BOARD(board)) {
 		/* This should be an SI1 board, which has this
 		   location writable... */
-		if (read_sx_byte (board, SI2_ISA_ID_BASE) != 0x10)
+		if (read_sx_byte (board, SI2_ISA_ID_BASE) != 0x10) {
 			func_exit();
 			return 0; 
+		}
 	} else {
 		/* This should be an SI2 board, which has the bottom
 		   3 bits non-writable... */
-		if (read_sx_byte (board, SI2_ISA_ID_BASE) == 0x10)
+		if (read_sx_byte (board, SI2_ISA_ID_BASE) == 0x10) {
 			func_exit ();
 			return 0; 
+		}
 	}
 
 	printheader ();

From 60b08c67220cf6faef7410ac6adba23a8a743bf7 Mon Sep 17 00:00:00 2001
From: Jon Mason <jdmason@us.ibm.com>
Date: Sun, 26 Feb 2006 04:18:22 +0100
Subject: [PATCH 467/509] [PATCH] x86_64: no_iommu removal in pci-gart.c

In previous versions of pci-gart.c, no_iommu was used to determine if IOMMU was
disabled in the GART DMA mapping functions.  This changed in 2.6.16 and now
gart_xxx() functions are only called if gart is enabled.  Therefore, uses of
no_iommu in the GART code are no longer necessary and can be removed.

Also, it removes double deceleration of no_iommu and force_iommu in pci.h and
proto.h, by removing the deceleration in pci.h.

Lastly, end_pfn off by one error.

Tested (along with patch 1/2) on dual opteron with gart enabled, iommu=soft,
and iommu=off.

Signed-off-by: Jon Mason <jdmason@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/aperture.c |  2 +-
 arch/x86_64/kernel/pci-gart.c | 22 +++++-----------------
 include/asm-x86_64/pci.h      |  2 --
 3 files changed, 6 insertions(+), 20 deletions(-)

diff --git a/arch/x86_64/kernel/aperture.c b/arch/x86_64/kernel/aperture.c
index e4e2b7d01f89..a0f955b9995f 100644
--- a/arch/x86_64/kernel/aperture.c
+++ b/arch/x86_64/kernel/aperture.c
@@ -248,7 +248,7 @@ void __init iommu_hole_init(void)
 		/* Got the aperture from the AGP bridge */
 	} else if (swiotlb && !valid_agp) {
 		/* Do nothing */
-	} else if ((!no_iommu && end_pfn >= MAX_DMA32_PFN) ||
+	} else if ((!no_iommu && end_pfn > MAX_DMA32_PFN) ||
 		   force_iommu ||
 		   valid_agp ||
 		   fallback_aper_force) { 
diff --git a/arch/x86_64/kernel/pci-gart.c b/arch/x86_64/kernel/pci-gart.c
index dd0718dc178b..0c3f052ba6ce 100644
--- a/arch/x86_64/kernel/pci-gart.c
+++ b/arch/x86_64/kernel/pci-gart.c
@@ -228,11 +228,6 @@ static inline int need_iommu(struct device *dev, unsigned long addr, size_t size
 	int mmu = high;
 	if (force_iommu) 
 		mmu = 1; 
-	if (no_iommu) { 
-		if (high) 
-			panic("PCI-DMA: high address but no IOMMU.\n"); 
-		mmu = 0; 
-	} 	
 	return mmu; 
 }
 
@@ -241,11 +236,6 @@ static inline int nonforced_iommu(struct device *dev, unsigned long addr, size_t
 	u64 mask = *dev->dma_mask;
 	int high = addr + size >= mask;
 	int mmu = high;
-	if (no_iommu) { 
-		if (high) 
-			panic("PCI-DMA: high address but no IOMMU.\n"); 
-		mmu = 0; 
-	} 	
 	return mmu; 
 }
 
@@ -310,7 +300,7 @@ void gart_unmap_sg(struct device *dev, struct scatterlist *sg, int nents, int di
 
 	for (i = 0; i < nents; i++) {
 		struct scatterlist *s = &sg[i];
-		if (!s->dma_length)
+		if (!s->dma_length || !s->length)
 			break;
 		dma_unmap_single(dev, s->dma_address, s->dma_length, dir);
 	}
@@ -364,6 +354,7 @@ static int __dma_map_cont(struct scatterlist *sg, int start, int stopat,
 		
 		BUG_ON(i > start && s->offset);
 		if (i == start) {
+			*sout = *s; 
 			sout->dma_address = iommu_bus_base;
 			sout->dma_address += iommu_page*PAGE_SIZE + s->offset;
 			sout->dma_length = s->length;
@@ -390,6 +381,7 @@ static inline int dma_map_cont(struct scatterlist *sg, int start, int stopat,
 {
 	if (!need) { 
 		BUG_ON(stopat - start != 1);
+		*sout = sg[start]; 
 		sout->dma_length = sg[start].length; 
 		return 0;
 	} 
@@ -632,17 +624,13 @@ static int __init pci_iommu_init(void)
 		(agp_copy_info(agp_bridge, &info) < 0);
 #endif	
 
-	if (swiotlb) { 
-		no_iommu = 1;
+	if (swiotlb)
 		return -1; 
-	} 
-	
+
 	if (no_iommu ||
 	    (!force_iommu && end_pfn <= MAX_DMA32_PFN) ||
 	    !iommu_aperture ||
 	    (no_agp && init_k8_gatt(&info) < 0)) {
-		no_iommu = 1;
-		no_iommu_init();
 		printk(KERN_INFO "PCI-DMA: Disabling IOMMU.\n");
 		if (end_pfn > MAX_DMA32_PFN) {
 			printk(KERN_ERR "WARNING more than 4GB of memory "
diff --git a/include/asm-x86_64/pci.h b/include/asm-x86_64/pci.h
index fd03e15d7ea6..8a05af264d18 100644
--- a/include/asm-x86_64/pci.h
+++ b/include/asm-x86_64/pci.h
@@ -19,8 +19,6 @@ extern unsigned int pcibios_assign_all_busses(void);
 #endif
 #define pcibios_scan_all_fns(a, b)	0
 
-extern int no_iommu, force_iommu;
-
 extern unsigned long pci_mem_start;
 #define PCIBIOS_MIN_IO		0x1000
 #define PCIBIOS_MIN_MEM		(pci_mem_start)

From f83f2b5fbab4585f4de4523c7879d60e3f85a248 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Sun, 26 Feb 2006 04:18:25 +0100
Subject: [PATCH 468/509] [PATCH] x86_64: fix USER_PTRS_PER_PGD

The value, while currently unused in the native kernel, was off by one.

Signed-Off-By: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 include/asm-x86_64/pgtable.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h
index 8fbf4dd72115..715fd94cf577 100644
--- a/include/asm-x86_64/pgtable.h
+++ b/include/asm-x86_64/pgtable.h
@@ -131,7 +131,7 @@ static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long
 #define PGDIR_SIZE	(1UL << PGDIR_SHIFT)
 #define PGDIR_MASK	(~(PGDIR_SIZE-1))
 
-#define USER_PTRS_PER_PGD	(TASK_SIZE/PGDIR_SIZE)
+#define USER_PTRS_PER_PGD	((TASK_SIZE-1)/PGDIR_SIZE+1)
 #define FIRST_USER_ADDRESS	0
 
 #ifndef __ASSEMBLY__

From 5342fba5412cead88b61ead07168615dbeba1ee3 Mon Sep 17 00:00:00 2001
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Sun, 26 Feb 2006 04:18:28 +0100
Subject: [PATCH 469/509] [PATCH] x86_64: Check for bad elf entry address.

Fixes a local DOS on Intel systems that lead to an endless
recursive fault.  AMD machines don't seem to be affected.

Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/binfmt_elf.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 1b117a441298..c2eac2a50bd2 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -938,6 +938,11 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 		kfree(elf_interpreter);
 	} else {
 		elf_entry = loc->elf_ex.e_entry;
+		if (BAD_ADDR(elf_entry)) {
+			send_sig(SIGSEGV, current, 0);
+			retval = -ENOEXEC; /* Nobody gets to see this, but.. */
+			goto out_free_dentry;
+		}
 	}
 
 	kfree(elf_phdata);

From 13a229abc25640813f1480c0478dfc6bdbc1c19e Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sun, 26 Feb 2006 04:18:31 +0100
Subject: [PATCH 470/509] [PATCH] x86_64: Only do the clustered systems have
 unsynchronized TSC assumption on IBM systems

Big Unisys systems have multiple clusters too, but they have an
synchronized TSC.

I'm using the SMBIOS to check for vendor == IBM.

Cc: Chris McDermott <lcm@us.ibm.com>
Cc: "Protasevich, Natalie" <Natalie.Protasevich@unisys.com>

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/acpi/boot.c |  3 ---
 arch/x86_64/kernel/apic.c    |  9 ++++++++-
 include/asm-x86_64/acpi.h    | 14 ++++++++++++++
 3 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 79577f0ace98..8309a7b2cd63 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -44,9 +44,6 @@ extern void __init clustered_apic_check(void);
 extern int gsi_irq_sharing(int gsi);
 #include <asm/proto.h>
 
-static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 0; }
-
-
 #else				/* X86 */
 
 #ifdef	CONFIG_X86_LOCAL_APIC
diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index e5b14c57eaa0..d70605eda333 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -962,12 +962,14 @@ void smp_apic_timer_interrupt(struct pt_regs *regs)
 	irq_exit();
 }
 
+int __initdata unsync_tsc_on_multicluster;
+
 /*
  * oem_force_hpet_timer -- force HPET mode for some boxes.
  *
  * Thus far, the major user of this is IBM's Summit2 series:
  *
- * Clustered boxes may have unsynced TSC problems if they are
+ * Some clustered boxes may have unsynced TSC problems if they are
  * multi-chassis. Use available data to take a good guess.
  * If in doubt, go HPET.
  */
@@ -977,6 +979,11 @@ __cpuinit int oem_force_hpet_timer(void)
 	unsigned id;
 	DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS);
 
+	/* Only do this check on IBM machines - big Unisys systems
+	   use multiple clusters too, but have synchronized TSC */
+	if (!unsync_tsc_on_multicluster)
+		return 0;
+
 	bitmap_zero(clustermap, NUM_APIC_CLUSTERS);
 
 	for (i = 0; i < NR_CPUS; i++) {
diff --git a/include/asm-x86_64/acpi.h b/include/asm-x86_64/acpi.h
index aa1c7b2e438c..e2b9923189a0 100644
--- a/include/asm-x86_64/acpi.h
+++ b/include/asm-x86_64/acpi.h
@@ -164,6 +164,20 @@ extern u8 x86_acpiid_to_apicid[];
 
 extern int acpi_skip_timer_override;
 
+extern int unsync_tsc_on_multicluster;
+
+static inline int acpi_madt_oem_check(char *oem, char *productid) 
+{ 
+	/* Copied from i386. Probably has too many entries. */
+	if (!strncmp(oem, "IBM ENSW", 8) && 
+        	(!strncmp(productid, "VIGIL SMP", 9) 
+             		|| !strncmp(productid, "EXA", 3)
+			|| !strncmp(productid, "RUTHLESS SMP", 12))) {
+		unsync_tsc_on_multicluster = 1;
+        }
+        return 0;
+}
+
 #endif /*__KERNEL__*/
 
 #endif /*_ASM_ACPI_H*/

From 6070f9ec6b03cc46cd0242523326f7a296f47c1c Mon Sep 17 00:00:00 2001
From: Andreas Deresch <aderesch@fs.tum.de>
Date: Sun, 26 Feb 2006 04:18:34 +0100
Subject: [PATCH 471/509] [PATCH] i386: Handle non existing APICs without
 panicing

[description from AK]

This fixes booting in APIC mode on some ACER laptops. x86-64
did a similar change some time ago.

See http://bugzilla.kernel.org/show_bug.cgi?id=4700 for details

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/io_apic.c |  6 ++++--
 arch/i386/kernel/mpparse.c | 10 ++++++++--
 2 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index f2dd218d88cb..235822b3f41b 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -2566,8 +2566,10 @@ int __init io_apic_get_unique_id (int ioapic, int apic_id)
 		spin_unlock_irqrestore(&ioapic_lock, flags);
 
 		/* Sanity check */
-		if (reg_00.bits.ID != apic_id)
-			panic("IOAPIC[%d]: Unable change apic_id!\n", ioapic);
+		if (reg_00.bits.ID != apic_id) {
+			printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic);
+			return -1;
+		}
 	}
 
 	apic_printk(APIC_VERBOSE, KERN_INFO
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index e7609abf3796..e6e2f43db85e 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -915,6 +915,7 @@ void __init mp_register_ioapic (
 	u32			gsi_base)
 {
 	int			idx = 0;
+	int			tmpid;
 
 	if (nr_ioapics >= MAX_IO_APICS) {
 		printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
@@ -935,9 +936,14 @@ void __init mp_register_ioapic (
 
 	set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
 	if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && (boot_cpu_data.x86 < 15))
-		mp_ioapics[idx].mpc_apicid = io_apic_get_unique_id(idx, id);
+		tmpid = io_apic_get_unique_id(idx, id);
 	else
-		mp_ioapics[idx].mpc_apicid = id;
+		tmpid = id;
+	if (tmpid == -1) {
+		nr_ioapics--;
+		return;
+	}
+	mp_ioapics[idx].mpc_apicid = tmpid;
 	mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
 	
 	/* 

From e78256b8f3e2850ad55c2d69e1429e6c2607afd3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sun, 26 Feb 2006 04:18:37 +0100
Subject: [PATCH 472/509] [PATCH] x86-64/i386: Use common X86_PM_TIMER option
 and make it EMBEDDED

This makes x86-64 use the common X86_PM_TIMER Kconfig entry in drivers/acpi

And since PM timer is needed for correct timing on a lot of systems
now (e.g. AMD dual cores) and we often get bug reports from people
who forgot to set it make it depend on CONFIG_EMBEDDED. x86-64 had
this change before and it's a good thing.

I also fixed the description slightly to make this more clear.

Cc: len.brown@intel.com

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/Kconfig  | 15 ---------------
 drivers/acpi/Kconfig |  8 +++-----
 2 files changed, 3 insertions(+), 20 deletions(-)

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 2f9deca31cc9..babc31b3ef12 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -354,21 +354,6 @@ config HPET_TIMER
 	  as it is off-chip.  You can find the HPET spec at
 	  <http://www.intel.com/hardwaredesign/hpetspec.htm>.
 
-config X86_PM_TIMER
-	bool "PM timer" if EMBEDDED
-	depends on ACPI
-	default y
-	help
-	  Support the ACPI PM timer for time keeping. This is slow,
-	  but is useful on some chipsets without HPET on systems with more
-	  than one CPU. On a single processor or single socket multi core
-	  system it is normally not required.
-	  When the PM timer is active 64bit vsyscalls are disabled
-	  and should not be enabled (/proc/sys/kernel/vsyscall64 should
-	  not be changed).
-	  The kernel selects the PM timer only as a last resort, so it is
-	  useful to enable just in case.
-
 config HPET_EMULATE_RTC
 	bool "Provide RTC interrupt"
 	depends on HPET_TIMER && RTC=y
diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
index 0cce28c4025b..9dc2fbe6efa7 100644
--- a/drivers/acpi/Kconfig
+++ b/drivers/acpi/Kconfig
@@ -285,9 +285,8 @@ config ACPI_SYSTEM
 	  dump your ACPI DSDT table using /proc/acpi/dsdt.
 
 config X86_PM_TIMER
-	bool "Power Management Timer Support"
+	bool "Power Management Timer Support" if EMBEDDED
 	depends on X86
-	depends on !X86_64
 	default y
 	help
 	  The Power Management Timer is available on all ACPI-capable,
@@ -298,9 +297,8 @@ config X86_PM_TIMER
 	  voltage scaling, unlike the commonly used Time Stamp Counter
 	  (TSC) timing source.
 
-	  So, if you see messages like 'Losing too many ticks!' in the
-	  kernel logs, and/or you are using this on a notebook which
-	  does not yet have an HPET, you should say "Y" here.
+	  You should nearly always say Y here because many modern
+	  systems require this timer. 
 
 config ACPI_CONTAINER
 	tristate "ACPI0004,PNP0A05 and PNP0A06 Container Driver (EXPERIMENTAL)"

From 1f9921539208f6d88f600a801e333d718e4a13ff Mon Sep 17 00:00:00 2001
From: Chris McDermott <lcm@us.ibm.com>
Date: Sun, 26 Feb 2006 04:18:40 +0100
Subject: [PATCH 473/509] [PATCH] x86_64: Fix NMI watchdog on x460

[description from AK]

Old check for the IO-APIC watchdog during the timer check was wrong -
it obviously should only drop into this if the IO-APIC watchdog is used.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/io_apic.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index 2585c1d92b26..ffab8a756664 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -1850,7 +1850,7 @@ static inline void check_timer(void)
 	}
 	printk(" failed.\n");
 
-	if (nmi_watchdog) {
+	if (nmi_watchdog == NMI_IO_APIC) {
 		printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n");
 		nmi_watchdog = 0;
 	}

From 2eb1bdbad89b19c99f8ac1de1492cdabbff6b3d3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sun, 26 Feb 2006 04:18:43 +0100
Subject: [PATCH 474/509] [PATCH] x86_64: Disable ACPI blacklist by year for
 now on x86-64

ACPI is initialized very early on x86-64, before the DMI code is
initialized.  This means it would often discover a 0 year and then turn
off ACPI because it thought the BIOS was too old.  Some systems don't
boot without ACPI so this was a problem.

I have a full fix by adding new very early DMI detection, but it needs
more testing before it can be merged.  For 2.6.16 let's just turn the
check off.  It never made much sense anyways because there are no x86-64
systems older than 2002 or so and they generally all have working ACPI.

Cc: len.brown@intel.com

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/acpi/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
index 9dc2fbe6efa7..33e2ca847a26 100644
--- a/drivers/acpi/Kconfig
+++ b/drivers/acpi/Kconfig
@@ -247,7 +247,7 @@ config ACPI_CUSTOM_DSDT_FILE
 	  Enter the full path name to the file wich includes the AmlCode declaration.
 
 config ACPI_BLACKLIST_YEAR
-	int "Disable ACPI for systems before Jan 1st this year" if X86
+	int "Disable ACPI for systems before Jan 1st this year" if X86_32
 	default 0
 	help
 	  enter a 4-digit year, eg. 2001 to disable ACPI by default

From e2c0388866dc12bef56b178b958f9b778fe6c687 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sun, 26 Feb 2006 04:18:46 +0100
Subject: [PATCH 475/509] [PATCH] x86_64: Fix the additional_cpus=.. option

It didn't set up the CPU possible map early enough, so the
option didn't actually work.

Noticed by Heiko Carstens

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/setup.c   | 6 ++++++
 arch/x86_64/kernel/smpboot.c | 2 +-
 include/asm-x86_64/proto.h   | 1 +
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 9435ab7d6fb8..5de7eaf5d97c 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -423,6 +423,12 @@ static __init void parse_cmdline_early (char ** cmdline_p)
 		else if(!memcmp(from, "elfcorehdr=", 11))
 			elfcorehdr_addr = memparse(from+11, &from);
 #endif
+
+#ifdef CONFIG_SMP
+		else if (!memcmp(from, "additional_cpus=", 16))
+			setup_additional_cpus(from+16);
+#endif
+
 	next_char:
 		c = *(from++);
 		if (!c)
diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index 67e4e28f4df8..b82eb86e4f5d 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1244,7 +1244,7 @@ void __cpu_die(unsigned int cpu)
  	printk(KERN_ERR "CPU %u didn't die...\n", cpu);
 }
 
-static __init int setup_additional_cpus(char *s)
+__init int setup_additional_cpus(char *s)
 {
 	return get_option(&s, &additional_cpus);
 }
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index eca3f2d633db..8bdcbd0aa03f 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -134,6 +134,7 @@ extern int force_iommu;
 
 extern int reboot_force;
 extern int notsc_setup(char *);
+extern int setup_additional_cpus(char *);
 
 extern void smp_local_timer_interrupt(struct pt_regs * regs);
 

From e8b917775b572bc27de105f1317c2de4335db5b3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sun, 26 Feb 2006 04:18:49 +0100
Subject: [PATCH 476/509] [PATCH] x86_64: Move the SMP time selection earlier

SMP time selection originally ran after all CPUs were brought up because
it needed to know the number of CPUs to decide if it needs an MP safe
timer or not.

This is not needed anymore because we know present CPUs early.

This fixes a couple of problems:
 - apicmaintimer didn't always work because it relied on state that was
   set up time_init_gtod too late.
 - The output for the used timer in early kernel log was misleading
   because time_init_gtod could actually change it later.  Now always
   print the final timer choice

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/smpboot.c |  2 --
 arch/x86_64/kernel/time.c    | 22 +++++++++++-----------
 include/asm-x86_64/proto.h   |  1 -
 3 files changed, 11 insertions(+), 14 deletions(-)

diff --git a/arch/x86_64/kernel/smpboot.c b/arch/x86_64/kernel/smpboot.c
index b82eb86e4f5d..66e98659d077 100644
--- a/arch/x86_64/kernel/smpboot.c
+++ b/arch/x86_64/kernel/smpboot.c
@@ -1152,8 +1152,6 @@ void __init smp_cpus_done(unsigned int max_cpus)
 	setup_ioapic_dest();
 #endif
 
-	time_init_gtod();
-
 	check_nmi_watchdog();
 }
 
diff --git a/arch/x86_64/kernel/time.c b/arch/x86_64/kernel/time.c
index 67841d11ed1f..3080f84bf7b7 100644
--- a/arch/x86_64/kernel/time.c
+++ b/arch/x86_64/kernel/time.c
@@ -48,6 +48,8 @@ static void cpufreq_delayed_get(void);
 extern void i8254_timer_resume(void);
 extern int using_apic_timer;
 
+static char *time_init_gtod(void);
+
 DEFINE_SPINLOCK(rtc_lock);
 DEFINE_SPINLOCK(i8253_lock);
 
@@ -901,6 +903,7 @@ static struct irqaction irq0 = {
 void __init time_init(void)
 {
 	char *timename;
+	char *gtod;
 
 #ifdef HPET_HACK_ENABLE_DANGEROUS
         if (!vxtime.hpet_address) {
@@ -945,21 +948,19 @@ void __init time_init(void)
 		timename = "PIT";
 	}
 
-	printk(KERN_INFO "time.c: Using %ld.%06ld MHz %s timer.\n",
-	       vxtime_hz / 1000000, vxtime_hz % 1000000, timename);
+	vxtime.mode = VXTIME_TSC;
+	gtod = time_init_gtod();
+
+	printk(KERN_INFO "time.c: Using %ld.%06ld MHz WALL %s GTOD %s timer.\n",
+	       vxtime_hz / 1000000, vxtime_hz % 1000000, timename, gtod);
 	printk(KERN_INFO "time.c: Detected %d.%03d MHz processor.\n",
 		cpu_khz / 1000, cpu_khz % 1000);
-	vxtime.mode = VXTIME_TSC;
 	vxtime.quot = (1000000L << 32) / vxtime_hz;
 	vxtime.tsc_quot = (1000L << 32) / cpu_khz;
 	vxtime.last_tsc = get_cycles_sync();
 	setup_irq(0, &irq0);
 
 	set_cyc2ns_scale(cpu_khz);
-
-#ifndef CONFIG_SMP
-	time_init_gtod();
-#endif
 }
 
 /*
@@ -981,9 +982,9 @@ __cpuinit int unsynchronized_tsc(void)
 }
 
 /*
- * Decide after all CPUs are booted what mode gettimeofday should use.
+ * Decide what mode gettimeofday should use.
  */
-void __init time_init_gtod(void)
+__init static char *time_init_gtod(void)
 {
 	char *timetype;
 
@@ -1011,8 +1012,7 @@ void __init time_init_gtod(void)
 		timetype = hpet_use_timer ? "HPET/TSC" : "PIT/TSC";
 		vxtime.mode = VXTIME_TSC;
 	}
-
-	printk(KERN_INFO "time.c: Using %s based timekeeping.\n", timetype);
+	return timetype;
 }
 
 __setup("report_lost_ticks", time_setup);
diff --git a/include/asm-x86_64/proto.h b/include/asm-x86_64/proto.h
index 8bdcbd0aa03f..3ba8fd45fcb3 100644
--- a/include/asm-x86_64/proto.h
+++ b/include/asm-x86_64/proto.h
@@ -39,7 +39,6 @@ extern void config_acpi_tables(void);
 extern void ia32_syscall(void);
 extern void iommu_hole_init(void);
 
-extern void time_init_gtod(void);
 extern int pmtimer_mark_offset(void);
 extern void pmtimer_resume(void);
 extern void pmtimer_wait(unsigned);

From ab9b32ee626e9b6df4ce2560a70ae15e62423cf4 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sun, 26 Feb 2006 04:18:52 +0100
Subject: [PATCH 477/509] [PATCH] x86_64: Better ATI timer fix

The previous experiment for using apicmaintimer on ATI systems didn't
work out very well.  In particular laptops with C2/C3 support often
don't let it tick during idle, which makes it useless.  There were also
some other bugs that made the apicmaintimer often not used at all.

I tried some other experiments - running timer over RTC and some other
things but they didn't really work well neither.

I rechecked the specs now and it turns out this simple change is
actually enough to avoid the double ticks on the ATI systems.  We just
turn off IRQ 0 in the 8254 and only route it directly using the IO-APIC.

I tested it on a few ATI systems and it worked there.  In fact it worked
on all chipsets (NVidia, Intel, AMD, ATI) I tried it on.

According to the ACPI spec routing should always work through the
IO-APIC so I think it's the correct thing to do anyways (and most of the
old gunk in check_timer should be thrown away for x86-64).

But for 2.6.16 it's best to do a fairly minimal change:
 - Use the known to be working everywhere-but-ATI IRQ0 both over 8254
   and IO-APIC setup everywhere
 - Except on ATI disable IRQ0 in the 8254
 - Remove the code to select apicmaintimer on ATI chipsets
 - Add some boot options to allow to override this (just paranoia)

In 2.6.17 I hope to switch the default over to this for everybody.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 Documentation/x86_64/boot-options.txt |  4 +++
 arch/x86_64/kernel/io_apic.c          | 46 +++++++++++++++++----------
 2 files changed, 33 insertions(+), 17 deletions(-)

diff --git a/Documentation/x86_64/boot-options.txt b/Documentation/x86_64/boot-options.txt
index 153740f460a6..1921353259ae 100644
--- a/Documentation/x86_64/boot-options.txt
+++ b/Documentation/x86_64/boot-options.txt
@@ -52,6 +52,10 @@ APICs
 		 apicmaintimer. Useful when your PIT timer is totally
 		 broken.
 
+   disable_8254_timer / enable_8254_timer
+		 Enable interrupt 0 timer routing over the 8254 in addition to over
+	         the IO-APIC. The kernel tries to set a sensible default.
+
 Early Console
 
    syntax: earlyprintk=vga
diff --git a/arch/x86_64/kernel/io_apic.c b/arch/x86_64/kernel/io_apic.c
index ffab8a756664..ffed464e6b12 100644
--- a/arch/x86_64/kernel/io_apic.c
+++ b/arch/x86_64/kernel/io_apic.c
@@ -50,6 +50,8 @@ static int no_timer_check;
 
 int disable_timer_pin_1 __initdata;
 
+int timer_over_8254 __initdata = 1;
+
 /* Where if anywhere is the i8259 connect in external int mode */
 static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
 
@@ -251,6 +253,20 @@ static int __init enable_ioapic_setup(char *str)
 __setup("noapic", disable_ioapic_setup);
 __setup("apic", enable_ioapic_setup);
 
+static int __init setup_disable_8254_timer(char *s)
+{
+	timer_over_8254 = -1;
+	return 1;
+}
+static int __init setup_enable_8254_timer(char *s)
+{
+	timer_over_8254 = 2;
+	return 1;
+}
+
+__setup("disable_8254_timer", setup_disable_8254_timer);
+__setup("enable_8254_timer", setup_enable_8254_timer);
+
 #include <asm/pci-direct.h>
 #include <linux/pci_ids.h>
 #include <linux/pci.h>
@@ -309,27 +325,20 @@ void __init check_ioapic(void)
 #endif
 					/* RED-PEN skip them on mptables too? */
 					return;
+
+				/* This should be actually default, but
+				   for 2.6.16 let's do it for ATI only where
+				   it's really needed. */
 				case PCI_VENDOR_ID_ATI:
-					if (apic_runs_main_timer != 0)
-						break;
-#ifdef CONFIG_ACPI
-					/* Don't do this for laptops right
-					   right now because their timer
-					   doesn't necessarily tick in C2/3 */
-					if (acpi_fadt.revision >= 3 &&
-			(acpi_fadt.plvl2_lat + acpi_fadt.plvl3_lat) < 1100) {
-						printk(KERN_INFO
-"ATI board detected, but seems to be a laptop. Timer might be shakey, sorry\n");
-						break;
-					}
-#endif					
+					if (timer_over_8254 == 1) {	
+						timer_over_8254 = 0;	
 					printk(KERN_INFO
-	     "ATI board detected. Using APIC/PM timer.\n");
-					apic_runs_main_timer = 1;
-					nohpet = 1;
+		"ATI board detected. Disabling timer routing over 8254.\n");
+					}	
 					return;
 				} 
 
+
 				/* No multi-function device? */
 				type = read_pci_config_byte(num,slot,func,
 							    PCI_HEADER_TYPE);
@@ -1773,6 +1782,8 @@ static inline void unlock_ExtINT_logic(void)
  * a wide range of boards and BIOS bugs.  Fortunately only the timer IRQ
  * is so screwy.  Thanks to Brian Perkins for testing/hacking this beast
  * fanatically on his truly buggy board.
+ *
+ * FIXME: really need to revamp this for modern platforms only.
  */
 static inline void check_timer(void)
 {
@@ -1795,7 +1806,8 @@ static inline void check_timer(void)
 	 */
 	apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
 	init_8259A(1);
-	enable_8259A_irq(0);
+	if (timer_over_8254 > 0)
+		enable_8259A_irq(0);
 
 	pin1  = find_isa_irq_pin(0, mp_INT);
 	apic1 = find_isa_irq_apic(0, mp_INT);

From fc5870f66279fabedc9dbba7c28451bbb8f47778 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Sun, 26 Feb 2006 04:18:55 +0100
Subject: [PATCH 478/509] [PATCH] x86_64: Fix ioctl compat code for /dev/rtc

RTC_IRQP_SET/RTC_EPOCH_SET don't take a pointer to an argument, but the
argument itself.  This actually simplifies the code and makes it work.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 fs/compat_ioctl.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 057e60217fc5..537ac70edfe5 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -2531,18 +2531,9 @@ static int rtc_ioctl(unsigned fd, unsigned cmd, unsigned long arg)
 		val32 = kval;
 		return put_user(val32, (unsigned int __user *)arg);
 	case RTC_IRQP_SET32:
+		return sys_ioctl(fd, RTC_IRQP_SET, arg); 
 	case RTC_EPOCH_SET32:
-		ret = get_user(val32, (unsigned int __user *)arg);
-		if (ret)
-			return ret;
-		kval = val32;
-
-		set_fs(KERNEL_DS);
-		ret = sys_ioctl(fd, (cmd == RTC_IRQP_SET32) ?
-				RTC_IRQP_SET : RTC_EPOCH_SET,
-				(unsigned long)&kval);
-		set_fs(oldfs);
-		return ret;
+		return sys_ioctl(fd, RTC_EPOCH_SET, arg);
 	default:
 		/* unreached */
 		return -ENOIOCTLCMD;

From d51761233d9e3be4cdf10f7482a50463bbd78c78 Mon Sep 17 00:00:00 2001
From: Brian Magnuson <magnuson@rcn.com>
Date: Mon, 27 Feb 2006 04:02:04 +0100
Subject: [PATCH 479/509] [PATCH] fix build on x86_64 with !CONFIG_HOTPLUG_CPU

The commit e2c0388866dc12bef56b178b958f9b778fe6c687 added
setup_additional_cpus to setup.c but this is only defined if
CONFIG_HOTPLUG_CPU is set.  This patch changes the #ifdef to reflect that.

Signed-off-by: Brian Magnuson <magnuson@rcn.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/x86_64/kernel/setup.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c
index 5de7eaf5d97c..aa55e3cec665 100644
--- a/arch/x86_64/kernel/setup.c
+++ b/arch/x86_64/kernel/setup.c
@@ -424,7 +424,7 @@ static __init void parse_cmdline_early (char ** cmdline_p)
 			elfcorehdr_addr = memparse(from+11, &from);
 #endif
 
-#ifdef CONFIG_SMP
+#ifdef CONFIG_HOTPLUG_CPU
 		else if (!memcmp(from, "additional_cpus=", 16))
 			setup_additional_cpus(from+16);
 #endif

From e18f9b4be430189d79a01a75734bf7cfdc22cc3f Mon Sep 17 00:00:00 2001
From: James Bottomley <James.Bottomley@SteelEye.com>
Date: Sun, 26 Feb 2006 17:07:45 -0600
Subject: [PATCH 480/509] [PATCH] fix voyager after topology.c move

Commit 9c869edac591977314323a4eaad5f7633fca684f broke voyager again
rather subtly because it already had its own topology exporting
functions, so now each CPU gets registered twice.

I think we can actually use the generic ones, so I don't propose
reverting it.  The attached should eliminate the voyager topology
functions in favour of the generic ones.

I also added a define to ensure voyager is never hotplug CPU (we don't
have the support in the SMP harness).

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig                      |  2 +-
 arch/i386/mach-voyager/voyager_basic.c | 14 --------------
 2 files changed, 1 insertion(+), 15 deletions(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 0afec8566e7b..af411596a318 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -733,7 +733,7 @@ config PHYSICAL_START
 
 config HOTPLUG_CPU
 	bool "Support for hot-pluggable CPUs (EXPERIMENTAL)"
-	depends on SMP && HOTPLUG && EXPERIMENTAL
+	depends on SMP && HOTPLUG && EXPERIMENTAL && !X86_VOYAGER
 	---help---
 	  Say Y here to experiment with turning CPUs off and on.  CPUs
 	  can be controlled through /sys/devices/system/cpu.
diff --git a/arch/i386/mach-voyager/voyager_basic.c b/arch/i386/mach-voyager/voyager_basic.c
index 6761d294f260..b584060ec004 100644
--- a/arch/i386/mach-voyager/voyager_basic.c
+++ b/arch/i386/mach-voyager/voyager_basic.c
@@ -25,7 +25,6 @@
 #include <linux/sysrq.h>
 #include <linux/smp.h>
 #include <linux/nodemask.h>
-#include <asm/cpu.h>
 #include <asm/io.h>
 #include <asm/voyager.h>
 #include <asm/vic.h>
@@ -331,16 +330,3 @@ void machine_power_off(void)
 	if (pm_power_off)
 		pm_power_off();
 }
-
-static struct i386_cpu cpu_devices[NR_CPUS];
-
-static int __init topology_init(void)
-{
-	int i;
-
-	for_each_present_cpu(i)
-		register_cpu(&cpu_devices[i].cpu, i, NULL);
-	return 0;
-}
-
-subsys_initcall(topology_init);

From 043df59eb3798c094e6ba47136f3d3b34a6791a7 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Sat, 25 Feb 2006 12:15:31 -0800
Subject: [PATCH 481/509] [SPARC64]: Implement futex_atomic_op_inuser().

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/asm-sparc64/futex.h | 88 +++++++++++++++++++++++++++++++++++--
 1 file changed, 84 insertions(+), 4 deletions(-)

diff --git a/include/asm-sparc64/futex.h b/include/asm-sparc64/futex.h
index 6a332a9f099c..0caf60147e97 100644
--- a/include/asm-sparc64/futex.h
+++ b/include/asm-sparc64/futex.h
@@ -1,6 +1,86 @@
-#ifndef _ASM_FUTEX_H
-#define _ASM_FUTEX_H
+#ifndef _SPARC64_FUTEX_H
+#define _SPARC64_FUTEX_H
 
-#include <asm-generic/futex.h>
+#include <linux/futex.h>
+#include <asm/errno.h>
+#include <asm/system.h>
+#include <asm/uaccess.h>
 
-#endif
+#define __futex_cas_op(insn, ret, oldval, uaddr, oparg)	\
+	__asm__ __volatile__(				\
+	"\n1:	lduwa	[%3] %%asi, %2\n"		\
+	"	" insn "\n"				\
+	"2:	casa	[%3] %%asi, %2, %1\n"		\
+	"	cmp	%2, %1\n"			\
+	"	bne,pn	%%icc, 1b\n"			\
+	"	 mov	0, %0\n"			\
+	"3:\n"						\
+	"	.section .fixup,#alloc,#execinstr\n"	\
+	"	.align	4\n"				\
+	"4:	ba	3b\n"				\
+	"	 mov	%5, %0\n"			\
+	"	.previous\n"				\
+	"	.section __ex_table,#alloc\n"		\
+	"	.align	4\n"				\
+	"	.word	1b, 4b\n"			\
+	"	.word	2b, 4b\n"			\
+	"	.previous\n"				\
+	: "=&r" (ret), "=&r" (oldval), "=&r" (tem)	\
+	: "r" (uaddr), "r" (oparg), "i" (-EFAULT)	\
+	: "memory")
+
+static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr)
+{
+	int op = (encoded_op >> 28) & 7;
+	int cmp = (encoded_op >> 24) & 15;
+	int oparg = (encoded_op << 8) >> 20;
+	int cmparg = (encoded_op << 20) >> 20;
+	int oldval = 0, ret, tem;
+
+	if (unlikely(!access_ok(VERIFY_WRITE, uaddr, sizeof(int))))
+		return -EFAULT;
+	if (unlikely((((unsigned long) uaddr) & 0x3UL)))
+		return -EINVAL;
+
+	if (encoded_op & (FUTEX_OP_OPARG_SHIFT << 28))
+		oparg = 1 << oparg;
+
+	inc_preempt_count();
+
+	switch (op) {
+	case FUTEX_OP_SET:
+		__futex_cas_op("mov\t%4, %1", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ADD:
+		__futex_cas_op("add\t%2, %4, %1", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_OR:
+		__futex_cas_op("or\t%2, %4, %1", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_ANDN:
+		__futex_cas_op("and\t%2, %4, %1", ret, oldval, uaddr, oparg);
+		break;
+	case FUTEX_OP_XOR:
+		__futex_cas_op("xor\t%2, %4, %1", ret, oldval, uaddr, oparg);
+		break;
+	default:
+		ret = -ENOSYS;
+	}
+
+	dec_preempt_count();
+
+	if (!ret) {
+		switch (cmp) {
+		case FUTEX_OP_CMP_EQ: ret = (oldval == cmparg); break;
+		case FUTEX_OP_CMP_NE: ret = (oldval != cmparg); break;
+		case FUTEX_OP_CMP_LT: ret = (oldval < cmparg); break;
+		case FUTEX_OP_CMP_GE: ret = (oldval >= cmparg); break;
+		case FUTEX_OP_CMP_LE: ret = (oldval <= cmparg); break;
+		case FUTEX_OP_CMP_GT: ret = (oldval > cmparg); break;
+		default: ret = -ENOSYS;
+		}
+	}
+	return ret;
+}
+
+#endif /* !(_SPARC64_FUTEX_H) */

From 7abea9214585823f7f19d91872d7c6f8874bef9a Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@sunset.davemloft.net>
Date: Sat, 25 Feb 2006 13:39:56 -0800
Subject: [PATCH 482/509] [SPARC64]: Make cpu_present_map available earlier.

The change to kernel/sched.c's init code to use for_each_cpu()
requires that the cpu_possible_map be setup much earlier.

Set it up via setup_arch(), constrained to NR_CPUS, and later
constrain it to max_cpus in smp_prepare_cpus().

This fixes SMP booting on sparc64.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/kernel/setup.c |  2 ++
 arch/sparc64/kernel/smp.c   | 28 +++++++++++++++++++---------
 include/asm-sparc64/smp.h   |  6 ++++++
 3 files changed, 27 insertions(+), 9 deletions(-)

diff --git a/arch/sparc64/kernel/setup.c b/arch/sparc64/kernel/setup.c
index 054461e6946d..158bd31e15b7 100644
--- a/arch/sparc64/kernel/setup.c
+++ b/arch/sparc64/kernel/setup.c
@@ -542,6 +542,8 @@ void __init setup_arch(char **cmdline_p)
 	}
 #endif
 
+	smp_setup_cpu_possible_map();
+
 	paging_init();
 }
 
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index 1fb6323e65a4..1f7ad8a69052 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -1079,18 +1079,12 @@ int setup_profiling_timer(unsigned int multiplier)
 	return 0;
 }
 
+/* Constrain the number of cpus to max_cpus.  */
 void __init smp_prepare_cpus(unsigned int max_cpus)
 {
-	int instance, mid;
-
-	instance = 0;
-	while (!cpu_find_by_instance(instance, NULL, &mid)) {
-		if (mid < max_cpus)
-			cpu_set(mid, phys_cpu_present_map);
-		instance++;
-	}
-
 	if (num_possible_cpus() > max_cpus) {
+		int instance, mid;
+
 		instance = 0;
 		while (!cpu_find_by_instance(instance, NULL, &mid)) {
 			if (mid != boot_cpu_id) {
@@ -1105,6 +1099,22 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	smp_store_cpu_info(boot_cpu_id);
 }
 
+/* Set this up early so that things like the scheduler can init
+ * properly.  We use the same cpu mask for both the present and
+ * possible cpu map.
+ */
+void __init smp_setup_cpu_possible_map(void)
+{
+	int instance, mid;
+
+	instance = 0;
+	while (!cpu_find_by_instance(instance, NULL, &mid)) {
+		if (mid < NR_CPUS)
+			cpu_set(mid, phys_cpu_present_map);
+		instance++;
+	}
+}
+
 void __devinit smp_prepare_boot_cpu(void)
 {
 	if (hard_smp_processor_id() >= NR_CPUS) {
diff --git a/include/asm-sparc64/smp.h b/include/asm-sparc64/smp.h
index 110a2de89123..473edb2603ec 100644
--- a/include/asm-sparc64/smp.h
+++ b/include/asm-sparc64/smp.h
@@ -66,8 +66,14 @@ static __inline__ int hard_smp_processor_id(void)
 
 #define raw_smp_processor_id() (current_thread_info()->cpu)
 
+extern void smp_setup_cpu_possible_map(void);
+
 #endif /* !(__ASSEMBLY__) */
 
+#else
+
+#define smp_setup_cpu_possible_map() do { } while (0)
+
 #endif /* !(CONFIG_SMP) */
 
 #define NO_PROC_ID		0xFF

From add2b6fdae9d7fc251c229e76252f731917094c4 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@evo.osdl.org>
Date: Sun, 26 Feb 2006 20:24:40 -0800
Subject: [PATCH 483/509] Make Kprobes depend on modules

Commit 9ec4b1f356b3bad928ae8e2aa9caebfa737d52df made kprobes not compile
without module support, so just make that clear in the Kconfig file.

Also, since it's marked EXPERIMENTAL, make that dependency explicit too.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/Kconfig    | 1 +
 arch/ia64/Kconfig    | 1 +
 arch/powerpc/Kconfig | 2 +-
 arch/sparc64/Kconfig | 1 +
 arch/x86_64/Kconfig  | 1 +
 5 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index af411596a318..5b1a7d46d1d9 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -1060,6 +1060,7 @@ source "arch/i386/oprofile/Kconfig"
 
 config KPROBES
 	bool "Kprobes (EXPERIMENTAL)"
+	depends on EXPERIMENTAL && MODULES
 	help
 	  Kprobes allows you to trap at almost any kernel address and
 	  execute a callback function.  register_kprobe() establishes
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 845cd0902a50..a85ea9d37f05 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -453,6 +453,7 @@ source "arch/ia64/oprofile/Kconfig"
 
 config KPROBES
 	bool "Kprobes (EXPERIMENTAL)"
+	depends on EXPERIMENTAL && MODULES
 	help
 	  Kprobes allows you to trap at almost any kernel address and
 	  execute a callback function.  register_kprobe() establishes
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 80d114a3a837..a834f9e0bbb3 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -963,7 +963,7 @@ source "arch/powerpc/oprofile/Kconfig"
 
 config KPROBES
 	bool "Kprobes (EXPERIMENTAL)"
-	depends on PPC64
+	depends on PPC64 && EXPERIMENTAL && MODULES
 	help
 	  Kprobes allows you to trap at almost any kernel address and
 	  execute a callback function.  register_kprobe() establishes
diff --git a/arch/sparc64/Kconfig b/arch/sparc64/Kconfig
index ab733be9af08..4c0a50a76554 100644
--- a/arch/sparc64/Kconfig
+++ b/arch/sparc64/Kconfig
@@ -383,6 +383,7 @@ source "arch/sparc64/oprofile/Kconfig"
 
 config KPROBES
 	bool "Kprobes (EXPERIMENTAL)"
+	depends on EXPERIMENTAL && MODULES
 	help
 	  Kprobes allows you to trap at almost any kernel address and
 	  execute a callback function.  register_kprobe() establishes
diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index babc31b3ef12..e18eb79bf855 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -577,6 +577,7 @@ source "arch/x86_64/oprofile/Kconfig"
 
 config KPROBES
 	bool "Kprobes (EXPERIMENTAL)"
+	depends on EXPERIMENTAL && MODULES
 	help
 	  Kprobes allows you to trap at almost any kernel address and
 	  execute a callback function.  register_kprobe() establishes

From 2cf82c0256b198ae28c465f2c4d7c12c836ea5ea Mon Sep 17 00:00:00 2001
From: Paul Mackerras <paulus@samba.org>
Date: Mon, 27 Feb 2006 15:41:47 +1100
Subject: [PATCH 484/509] powerpc: Export variables used in conversions to/from
 cputime_t

The inline cputime_to_foo and foo_to_cputime conversion functions in
include/asm-powerpc/cputime.h refer to 5 variables, which need to be
exported if those functions are to be usable from modules.

Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/time.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/powerpc/kernel/time.c b/arch/powerpc/kernel/time.c
index 0b34db28916f..4f20a5f15d49 100644
--- a/arch/powerpc/kernel/time.c
+++ b/arch/powerpc/kernel/time.c
@@ -99,6 +99,7 @@ unsigned long tb_ticks_per_jiffy;
 unsigned long tb_ticks_per_usec = 100; /* sane default */
 EXPORT_SYMBOL(tb_ticks_per_usec);
 unsigned long tb_ticks_per_sec;
+EXPORT_SYMBOL(tb_ticks_per_sec);	/* for cputime_t conversions */
 u64 tb_to_xs;
 unsigned tb_to_us;
 
@@ -143,9 +144,13 @@ DEFINE_PER_CPU(unsigned long, last_jiffy);
  * These are all stored as 0.64 fixed-point binary fractions.
  */
 u64 __cputime_jiffies_factor;
+EXPORT_SYMBOL(__cputime_jiffies_factor);
 u64 __cputime_msec_factor;
+EXPORT_SYMBOL(__cputime_msec_factor);
 u64 __cputime_sec_factor;
+EXPORT_SYMBOL(__cputime_sec_factor);
 u64 __cputime_clockt_factor;
+EXPORT_SYMBOL(__cputime_clockt_factor);
 
 static void calc_cputime_factors(void)
 {

From b9a33cebac70d6f67a769ce8d4078fee2b254ada Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@g5.osdl.org>
Date: Sun, 26 Feb 2006 21:09:35 -0800
Subject: [PATCH 485/509] Linux v2.6.16-rc5

---
 Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Makefile b/Makefile
index 77a448c8e776..46eea76bc570 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 16
-EXTRAVERSION =-rc4
+EXTRAVERSION =-rc5
 NAME=Sliding Snow Leopard
 
 # *DOCUMENTATION*

From d904ffd6e0c3ab7631b07c21c9f9234a2e72d1f3 Mon Sep 17 00:00:00 2001
From: Marc Zyngier <maz@misterjones.org>
Date: Mon, 27 Feb 2006 12:08:00 +0100
Subject: [PATCH 486/509] [PATCH] Fix Specialix SX corruption

With the latest kernels, I experienced some strange corruption, some
'*****' being randomly inserted in the character flow, like this:

	ashes:~#
	ashes:~#
	a*******shes:~#
	ashes:~#
	ashes:~#

Further investigation shows that the problem was introduced during
Alan's "TTY layer buffering revamp" patch, the amount of data to be
copied being reduced after buffer allocation.  Moving the count fixup
around solves the problem.

Signed-off-by: Marc Zyngier <maz@misterjones.org>
Approved-by: Rogier Wolff <R.E.Wolff@BitWizard.nl>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 drivers/char/sx.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index 588e75ec1630..a6b4f02bdceb 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -1095,17 +1095,17 @@ static inline void sx_receive_chars (struct sx_port *port)
 
 		sx_dprintk (SX_DEBUG_RECEIVE, "rxop=%d, c = %d.\n", rx_op, c); 
 
+		/* Don't copy past the end of the hardware receive buffer */
+		if (rx_op + c > 0x100) c = 0x100 - rx_op;
+
+		sx_dprintk (SX_DEBUG_RECEIVE, "c = %d.\n", c);
+
 		/* Don't copy more bytes than there is room for in the buffer */
 
 		c = tty_prepare_flip_string(tty, &rp, c);
 
 		sx_dprintk (SX_DEBUG_RECEIVE, "c = %d.\n", c); 
 
-		/* Don't copy past the end of the hardware receive buffer */
-		if (rx_op + c > 0x100) c = 0x100 - rx_op;
-
-		sx_dprintk (SX_DEBUG_RECEIVE, "c = %d.\n", c);
-
 		/* If for one reason or another, we can't copy more data, we're done! */
 		if (c == 0) break;
 

From 3e6cb2d38a9c9758170813497a860c64543643d5 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 21 Feb 2006 18:32:14 +0000
Subject: [PATCH 487/509] [MIPS] Use "=R" constraint to avoid compiler errors
 in cmpxchg().

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/system.h | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/include/asm-mips/system.h b/include/asm-mips/system.h
index e8e5d4143377..ddae9bae31af 100644
--- a/include/asm-mips/system.h
+++ b/include/asm-mips/system.h
@@ -322,7 +322,7 @@ static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old,
 #endif
 		"2:							\n"
 		"	.set	pop					\n"
-		: "=&r" (retval), "=m" (*m)
+		: "=&r" (retval), "=R" (*m)
 		: "R" (*m), "Jr" (old), "Jr" (new)
 		: "memory");
 	} else if (cpu_has_llsc) {
@@ -342,7 +342,7 @@ static inline unsigned long __cmpxchg_u32(volatile int * m, unsigned long old,
 #endif
 		"2:							\n"
 		"	.set	pop					\n"
-		: "=&r" (retval), "=m" (*m)
+		: "=&r" (retval), "=R" (*m)
 		: "R" (*m), "Jr" (old), "Jr" (new)
 		: "memory");
 	} else {
@@ -379,7 +379,7 @@ static inline unsigned long __cmpxchg_u64(volatile int * m, unsigned long old,
 #endif
 		"2:							\n"
 		"	.set	pop					\n"
-		: "=&r" (retval), "=m" (*m)
+		: "=&r" (retval), "=R" (*m)
 		: "R" (*m), "Jr" (old), "Jr" (new)
 		: "memory");
 	} else if (cpu_has_llsc) {
@@ -397,7 +397,7 @@ static inline unsigned long __cmpxchg_u64(volatile int * m, unsigned long old,
 #endif
 		"2:							\n"
 		"	.set	pop					\n"
-		: "=&r" (retval), "=m" (*m)
+		: "=&r" (retval), "=R" (*m)
 		: "R" (*m), "Jr" (old), "Jr" (new)
 		: "memory");
 	} else {

From 9b6695a8adfe0916e81ddd810a5b9db3eb8b0e46 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 23 Feb 2006 12:23:27 +0000
Subject: [PATCH 488/509] [MIPS] SMP: Fix initialization order bug.

A recent change requires cpu_possible_map to be initialized before
smp_sched_init() but most MIPS platforms were initializing their
processors in the prom_prepare_cpus callback of smp_prepare_cpus.  The
simple fix of calling prom_prepare_cpus from one of the earlier SMP
initialization hooks doesn't work well either since IPIs may require
init_IRQ() to have completed, so bit the bullet and split
prom_prepare_cpus into two initialization functions, plat_smp_setup
which is called early from setup_arch and plat_prepare_cpus called where
prom_prepare_cpus used to be called.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/setup.c            |  3 +++
 arch/mips/kernel/smp.c              |  2 +-
 arch/mips/kernel/smp_mt.c           | 13 +++++++------
 arch/mips/pmc-sierra/yosemite/smp.c | 24 ++++++------------------
 arch/mips/sgi-ip27/ip27-smp.c       |  7 ++++++-
 arch/mips/sibyte/cfe/smp.c          | 10 +++++++---
 include/asm-mips/smp.h              | 11 +++++++++--
 7 files changed, 39 insertions(+), 31 deletions(-)

diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index d86affa21278..d9293c558e41 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -540,6 +540,9 @@ void __init setup_arch(char **cmdline_p)
 	sparse_init();
 	paging_init();
 	resource_init();
+#ifdef CONFIG_SMP
+	plat_smp_setup();
+#endif
 }
 
 int __init fpu_disable(char *s)
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 5e189862e523..06ed90752424 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -236,7 +236,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	init_new_context(current, &init_mm);
 	current_thread_info()->cpu = 0;
 	smp_tune_scheduling();
-	prom_prepare_cpus(max_cpus);
+	plat_prepare_cpus(max_cpus);
 }
 
 /* preload SMP state for boot cpu */
diff --git a/arch/mips/kernel/smp_mt.c b/arch/mips/kernel/smp_mt.c
index c930364830d0..993b8bf56aaf 100644
--- a/arch/mips/kernel/smp_mt.c
+++ b/arch/mips/kernel/smp_mt.c
@@ -143,7 +143,7 @@ static struct irqaction irq_call = {
  * Make sure all CPU's are in a sensible state before we boot any of the
  * secondarys
  */
-void prom_prepare_cpus(unsigned int max_cpus)
+void plat_smp_setup(void)
 {
 	unsigned long val;
 	int i, num;
@@ -179,11 +179,9 @@ void prom_prepare_cpus(unsigned int max_cpus)
 				write_vpe_c0_vpeconf0(tmp);
 
 				/* Record this as available CPU */
-				if (i < max_cpus) {
-					cpu_set(i, phys_cpu_present_map);
-					__cpu_number_map[i]	= ++num;
-					__cpu_logical_map[num]	= i;
-				}
+				cpu_set(i, phys_cpu_present_map);
+				__cpu_number_map[i]	= ++num;
+				__cpu_logical_map[num]	= i;
 			}
 
 			/* disable multi-threading with TC's */
@@ -241,7 +239,10 @@ void prom_prepare_cpus(unsigned int max_cpus)
 		set_vi_handler (MIPS_CPU_IPI_RESCHED_IRQ, ipi_resched_dispatch);
 		set_vi_handler (MIPS_CPU_IPI_CALL_IRQ, ipi_call_dispatch);
 	}
+}
 
+void __init plat_prepare_cpus(unsigned int max_cpus)
+{
 	cpu_ipi_resched_irq = MIPSCPU_INT_BASE + MIPS_CPU_IPI_RESCHED_IRQ;
 	cpu_ipi_call_irq = MIPSCPU_INT_BASE + MIPS_CPU_IPI_CALL_IRQ;
 
diff --git a/arch/mips/pmc-sierra/yosemite/smp.c b/arch/mips/pmc-sierra/yosemite/smp.c
index 7f8fda962190..c197311e15d3 100644
--- a/arch/mips/pmc-sierra/yosemite/smp.c
+++ b/arch/mips/pmc-sierra/yosemite/smp.c
@@ -50,37 +50,25 @@ void __init prom_grab_secondary(void)
  * We don't want to start the secondary CPU yet nor do we have a nice probing
  * feature in PMON so we just assume presence of the secondary core.
  */
-static char maxcpus_string[] __initdata =
-	KERN_WARNING "max_cpus set to 0; using 1 instead\n";
-
-void __init prom_prepare_cpus(unsigned int max_cpus)
+void __init plat_smp_setup(void)
 {
-	int enabled = 0, i;
-
-	if (max_cpus == 0) {
-		printk(maxcpus_string);
-		max_cpus = 1;
-	}
+	int i;
 
 	cpus_clear(phys_cpu_present_map);
 
 	for (i = 0; i < 2; i++) {
-		if (i == max_cpus)
-			break;
-
-		/*
-		 * The boot CPU
-		 */
 		cpu_set(i, phys_cpu_present_map);
 		__cpu_number_map[i]	= i;
 		__cpu_logical_map[i]	= i;
-		enabled++;
 	}
+}
 
+void __init plat_prepare_cpus(unsigned int max_cpus)
+{
 	/*
 	 * Be paranoid.  Enable the IPI only if we're really about to go SMP.
 	 */
-	if (enabled > 1)
+	if (cpus_weight(cpu_possible_map))
 		set_c0_status(STATUSF_IP5);
 }
 
diff --git a/arch/mips/sgi-ip27/ip27-smp.c b/arch/mips/sgi-ip27/ip27-smp.c
index dbef3f6b5650..09fa7f5216f0 100644
--- a/arch/mips/sgi-ip27/ip27-smp.c
+++ b/arch/mips/sgi-ip27/ip27-smp.c
@@ -140,7 +140,7 @@ static __init void intr_clear_all(nasid_t nasid)
 		REMOTE_HUB_CLR_INTR(nasid, i);
 }
 
-void __init prom_prepare_cpus(unsigned int max_cpus)
+void __init plat_smp_setup(void)
 {
 	cnodeid_t	cnode;
 
@@ -161,6 +161,11 @@ void __init prom_prepare_cpus(unsigned int max_cpus)
 	alloc_cpupda(0, 0);
 }
 
+void __init plat_prepare_cpus(unsigned int max_cpus)
+{
+	/* We already did everything necessary earlier */
+}
+
 /*
  * Launch a slave into smp_bootstrap().  It doesn't take an argument, and we
  * set sp to the kernel stack of the newly created idle process, gp to the proc
diff --git a/arch/mips/sibyte/cfe/smp.c b/arch/mips/sibyte/cfe/smp.c
index 4477af3d8074..eab20e2db323 100644
--- a/arch/mips/sibyte/cfe/smp.c
+++ b/arch/mips/sibyte/cfe/smp.c
@@ -31,7 +31,7 @@
  *
  * Common setup before any secondaries are started
  */
-void __init prom_prepare_cpus(unsigned int max_cpus)
+void __init plat_smp_setup(void)
 {
 	int i, num;
 
@@ -40,14 +40,18 @@ void __init prom_prepare_cpus(unsigned int max_cpus)
 	__cpu_number_map[0] = 0;
 	__cpu_logical_map[0] = 0;
 
-	for (i=1, num=0; i<NR_CPUS; i++) {
+	for (i = 1, num = 0; i < NR_CPUS; i++) {
 		if (cfe_cpu_stop(i) == 0) {
 			cpu_set(i, phys_cpu_present_map);
 			__cpu_number_map[i] = ++num;
 			__cpu_logical_map[num] = i;
 		}
 	}
-	printk("Detected %i available secondary CPU(s)\n", num);
+	printk(KERN_INFO "Detected %i available secondary CPU(s)\n", num);
+}
+
+void __init plat_prepare_cpus(unsigned int max_cpus)
+{
 }
 
 /*
diff --git a/include/asm-mips/smp.h b/include/asm-mips/smp.h
index 5618f1e12f40..75c6fe7c2126 100644
--- a/include/asm-mips/smp.h
+++ b/include/asm-mips/smp.h
@@ -58,7 +58,9 @@ static inline int num_booting_cpus(void)
 	return cpus_weight(cpu_callout_map);
 }
 
-/* These are defined by the board-specific code. */
+/*
+ * These are defined by the board-specific code.
+ */
 
 /*
  * Cause the function described by call_data to be executed on the passed
@@ -79,7 +81,12 @@ extern void prom_boot_secondary(int cpu, struct task_struct *idle);
 extern void prom_init_secondary(void);
 
 /*
- * Detect available CPUs, populate phys_cpu_present_map before smp_init
+ * Populate cpu_possible_map before smp_init, called from setup_arch.
+ */
+extern void plat_smp_setup(void);
+
+/*
+ * Called after init_IRQ but before __cpu_up.
  */
 extern void prom_prepare_cpus(unsigned int max_cpus);
 

From 92f22c183cd669c8575767fede8fe43bb4f7bce9 Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Thu, 23 Feb 2006 14:10:53 +0000
Subject: [PATCH 489/509] [MIPS] Fix atomic*_sub_if_positive return value.

Reported and initial fix by Thomas Koeller <thomas.koeller@baslerweb.com>,
rewritten by me.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 include/asm-mips/atomic.h | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/include/asm-mips/atomic.h b/include/asm-mips/atomic.h
index 654b97d3e13a..2c8b853376c9 100644
--- a/include/asm-mips/atomic.h
+++ b/include/asm-mips/atomic.h
@@ -250,7 +250,10 @@ static __inline__ int atomic_sub_if_positive(int i, atomic_t * v)
 		"	subu	%0, %1, %3				\n"
 		"	bltz	%0, 1f					\n"
 		"	sc	%0, %2					\n"
+		"	.set	noreorder				\n"
 		"	beqzl	%0, 1b					\n"
+		"	 subu	%0, %1, %3				\n"
+		"	.set	reorder					\n"
 		"	sync						\n"
 		"1:							\n"
 		"	.set	mips0					\n"
@@ -266,7 +269,10 @@ static __inline__ int atomic_sub_if_positive(int i, atomic_t * v)
 		"	subu	%0, %1, %3				\n"
 		"	bltz	%0, 1f					\n"
 		"	sc	%0, %2					\n"
+		"	.set	noreorder				\n"
 		"	beqz	%0, 1b					\n"
+		"	 subu	%0, %1, %3				\n"
+		"	.set	reorder					\n"
 		"	sync						\n"
 		"1:							\n"
 		"	.set	mips0					\n"
@@ -598,7 +604,10 @@ static __inline__ long atomic64_sub_if_positive(long i, atomic64_t * v)
 		"	dsubu	%0, %1, %3				\n"
 		"	bltz	%0, 1f					\n"
 		"	scd	%0, %2					\n"
+		"	.set	noreorder				\n"
 		"	beqzl	%0, 1b					\n"
+		"	 dsubu	%0, %1, %3				\n"
+		"	.set	reorder					\n"
 		"	sync						\n"
 		"1:							\n"
 		"	.set	mips0					\n"
@@ -614,7 +623,10 @@ static __inline__ long atomic64_sub_if_positive(long i, atomic64_t * v)
 		"	dsubu	%0, %1, %3				\n"
 		"	bltz	%0, 1f					\n"
 		"	scd	%0, %2					\n"
+		"	.set	noreorder				\n"
 		"	beqz	%0, 1b					\n"
+		"	 dsubu	%0, %1, %3				\n"
+		"	.set	reorder					\n"
 		"	sync						\n"
 		"1:							\n"
 		"	.set	mips0					\n"

From 2fd628fe25e1f3d07996b0dab728ea0702f81306 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Tue, 21 Feb 2006 15:59:00 +0900
Subject: [PATCH 490/509] [MIPS] Use generic compat routines for readdir,
 getdents

Not just cleanup but also fixes O32 readdir(2) emulation.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
---
 arch/mips/kernel/linux32.c     | 54 ----------------------------------
 arch/mips/kernel/scall64-n32.S |  2 +-
 arch/mips/kernel/scall64-o32.S |  4 +--
 3 files changed, 3 insertions(+), 57 deletions(-)

diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c
index 5f68b220c26d..e00e5f6e7fdd 100644
--- a/arch/mips/kernel/linux32.c
+++ b/arch/mips/kernel/linux32.c
@@ -161,60 +161,6 @@ out:
 	return error;
 }
 
-struct dirent32 {
-	unsigned int	d_ino;
-	unsigned int	d_off;
-	unsigned short	d_reclen;
-	char		d_name[NAME_MAX + 1];
-};
-
-static void
-xlate_dirent(void *dirent64, void *dirent32, long n)
-{
-	long off;
-	struct dirent *dirp;
-	struct dirent32 *dirp32;
-
-	off = 0;
-	while (off < n) {
-		dirp = (struct dirent *)(dirent64 + off);
-		dirp32 = (struct dirent32 *)(dirent32 + off);
-		off += dirp->d_reclen;
-		dirp32->d_ino = dirp->d_ino;
-		dirp32->d_off = (unsigned int)dirp->d_off;
-		dirp32->d_reclen = dirp->d_reclen;
-		strncpy(dirp32->d_name, dirp->d_name, dirp->d_reclen - ((3 * 4) + 2));
-	}
-	return;
-}
-
-asmlinkage long
-sys32_getdents(unsigned int fd, void * dirent32, unsigned int count)
-{
-	long n;
-	void *dirent64;
-
-	dirent64 = (void *)((unsigned long)(dirent32 + (sizeof(long) - 1)) & ~(sizeof(long) - 1));
-	if ((n = sys_getdents(fd, dirent64, count - (dirent64 - dirent32))) < 0)
-		return(n);
-	xlate_dirent(dirent64, dirent32, n);
-	return(n);
-}
-
-asmlinkage int old_readdir(unsigned int fd, void * dirent, unsigned int count);
-
-asmlinkage int
-sys32_readdir(unsigned int fd, void * dirent32, unsigned int count)
-{
-	int n;
-	struct dirent dirent64;
-
-	if ((n = old_readdir(fd, &dirent64, count)) < 0)
-		return(n);
-	xlate_dirent(&dirent64, dirent32, dirent64.d_reclen);
-	return(n);
-}
-
 asmlinkage int
 sys32_waitpid(compat_pid_t pid, unsigned int *stat_addr, int options)
 {
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index d87b5446fa13..02c8267e45e7 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -195,7 +195,7 @@ EXPORT(sysn32_call_table)
 	PTR	sys_fdatasync
 	PTR	sys_truncate
 	PTR	sys_ftruncate			/* 6075 */
-	PTR	sys32_getdents
+	PTR	compat_sys_getdents
 	PTR	sys_getcwd
 	PTR	sys_chdir
 	PTR	sys_fchdir
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index 5b0414018c9a..797e0d874889 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -293,7 +293,7 @@ sys_call_table:
 	PTR	sys_uselib
 	PTR	sys_swapon
 	PTR	sys_reboot
-	PTR	sys32_readdir
+	PTR	compat_sys_old_readdir
 	PTR	old_mmap			/* 4090 */
 	PTR	sys_munmap
 	PTR	sys_truncate
@@ -345,7 +345,7 @@ sys_call_table:
 	PTR	sys_setfsuid
 	PTR	sys_setfsgid
 	PTR	sys32_llseek			/* 4140 */
-	PTR	sys32_getdents
+	PTR	compat_sys_getdents
 	PTR	compat_sys_select
 	PTR	sys_flock
 	PTR	sys_msync

From 051d3cbd96909b2fe6b5038e7bbe77f41356db05 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Mon, 27 Feb 2006 12:51:27 -0800
Subject: [PATCH 491/509] [TG3]: Fix Sun tg3 variant detection.

Some Sun parts don't have PCI_VENDOR_ID_SUN in the subsystem
vendor ID.  So add another fallback test, which is the name
of the OBP firmware device tree node.  If it's a Sun part we'll
get "network", else it will be named "ethernet".

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/tg3.c | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c
index e7dc653d5bd6..e8e92c853e89 100644
--- a/drivers/net/tg3.c
+++ b/drivers/net/tg3.c
@@ -9408,6 +9408,15 @@ static int __devinit tg3_is_sun_570X(struct tg3 *tp)
 			return 0;
 		if (venid == PCI_VENDOR_ID_SUN)
 			return 1;
+
+		/* TG3 chips onboard the SunBlade-2500 don't have the
+		 * subsystem-vendor-id set to PCI_VENDOR_ID_SUN but they
+		 * are distinguishable from non-Sun variants by being
+		 * named "network" by the firmware.  Non-Sun cards will
+		 * show up as being named "ethernet".
+		 */
+		if (!strcmp(pcp->prom_name, "network"))
+			return 1;
 	}
 	return 0;
 }

From 4bf05eceecf2efb4c883e9e9b17825682e7330dd Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 27 Feb 2006 13:00:01 -0800
Subject: [PATCH 492/509] [IPSEC] esp: Kill unnecessary block and indentation

We used to keep sg on the stack which is why the extra block was useful.
We've long since stopped doing that so let's kill the block and save
some indentation.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/esp4.c | 101 ++++++++++++++++++++++++------------------------
 1 file changed, 50 insertions(+), 51 deletions(-)

diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 73bfcae8af9c..3f47419cb9c5 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -150,6 +150,10 @@ static int esp_input(struct xfrm_state *x, struct xfrm_decap_state *decap, struc
 	int elen = skb->len - sizeof(struct ip_esp_hdr) - esp->conf.ivlen - alen;
 	int nfrags;
 	int encap_len = 0;
+	u8 nexthdr[2];
+	struct scatterlist *sg;
+	u8 workbuf[60];
+	int padlen;
 
 	if (!pskb_may_pull(skb, sizeof(struct ip_esp_hdr)))
 		goto out;
@@ -185,60 +189,55 @@ static int esp_input(struct xfrm_state *x, struct xfrm_decap_state *decap, struc
 	if (esp->conf.ivlen)
 		crypto_cipher_set_iv(esp->conf.tfm, esph->enc_data, crypto_tfm_alg_ivsize(esp->conf.tfm));
 
-        {
-		u8 nexthdr[2];
-		struct scatterlist *sg = &esp->sgbuf[0];
-		u8 workbuf[60];
-		int padlen;
+	sg = &esp->sgbuf[0];
 
-		if (unlikely(nfrags > ESP_NUM_FAST_SG)) {
-			sg = kmalloc(sizeof(struct scatterlist)*nfrags, GFP_ATOMIC);
-			if (!sg)
-				goto out;
-		}
-		skb_to_sgvec(skb, sg, sizeof(struct ip_esp_hdr) + esp->conf.ivlen, elen);
-		crypto_cipher_decrypt(esp->conf.tfm, sg, sg, elen);
-		if (unlikely(sg != &esp->sgbuf[0]))
-			kfree(sg);
-
-		if (skb_copy_bits(skb, skb->len-alen-2, nexthdr, 2))
-			BUG();
-
-		padlen = nexthdr[0];
-		if (padlen+2 >= elen)
+	if (unlikely(nfrags > ESP_NUM_FAST_SG)) {
+		sg = kmalloc(sizeof(struct scatterlist)*nfrags, GFP_ATOMIC);
+		if (!sg)
 			goto out;
-
-		/* ... check padding bits here. Silly. :-) */ 
-
-		if (x->encap && decap && decap->decap_type) {
-			struct esp_decap_data *encap_data;
-			struct udphdr *uh = (struct udphdr *) (iph+1);
-
-			encap_data = (struct esp_decap_data *) (decap->decap_data);
-			encap_data->proto = 0;
-
-			switch (decap->decap_type) {
-			case UDP_ENCAP_ESPINUDP:
-			case UDP_ENCAP_ESPINUDP_NON_IKE:
-				encap_data->proto = AF_INET;
-				encap_data->saddr.a4 = iph->saddr;
-				encap_data->sport = uh->source;
-				encap_len = (void*)esph - (void*)uh;
-				break;
-
-			default:
-				goto out;
-			}
-		}
-
-		iph->protocol = nexthdr[1];
-		pskb_trim(skb, skb->len - alen - padlen - 2);
-		memcpy(workbuf, skb->nh.raw, iph->ihl*4);
-		skb->h.raw = skb_pull(skb, sizeof(struct ip_esp_hdr) + esp->conf.ivlen);
-		skb->nh.raw += encap_len + sizeof(struct ip_esp_hdr) + esp->conf.ivlen;
-		memcpy(skb->nh.raw, workbuf, iph->ihl*4);
-		skb->nh.iph->tot_len = htons(skb->len);
 	}
+	skb_to_sgvec(skb, sg, sizeof(struct ip_esp_hdr) + esp->conf.ivlen, elen);
+	crypto_cipher_decrypt(esp->conf.tfm, sg, sg, elen);
+	if (unlikely(sg != &esp->sgbuf[0]))
+		kfree(sg);
+
+	if (skb_copy_bits(skb, skb->len-alen-2, nexthdr, 2))
+		BUG();
+
+	padlen = nexthdr[0];
+	if (padlen+2 >= elen)
+		goto out;
+
+	/* ... check padding bits here. Silly. :-) */ 
+
+	if (x->encap && decap && decap->decap_type) {
+		struct esp_decap_data *encap_data;
+		struct udphdr *uh = (struct udphdr *) (iph+1);
+
+		encap_data = (struct esp_decap_data *) (decap->decap_data);
+		encap_data->proto = 0;
+
+		switch (decap->decap_type) {
+		case UDP_ENCAP_ESPINUDP:
+		case UDP_ENCAP_ESPINUDP_NON_IKE:
+			encap_data->proto = AF_INET;
+			encap_data->saddr.a4 = iph->saddr;
+			encap_data->sport = uh->source;
+			encap_len = (void*)esph - (void*)uh;
+			break;
+
+		default:
+			goto out;
+		}
+	}
+
+	iph->protocol = nexthdr[1];
+	pskb_trim(skb, skb->len - alen - padlen - 2);
+	memcpy(workbuf, skb->nh.raw, iph->ihl*4);
+	skb->h.raw = skb_pull(skb, sizeof(struct ip_esp_hdr) + esp->conf.ivlen);
+	skb->nh.raw += encap_len + sizeof(struct ip_esp_hdr) + esp->conf.ivlen;
+	memcpy(skb->nh.raw, workbuf, iph->ihl*4);
+	skb->nh.iph->tot_len = htons(skb->len);
 
 	return 0;
 

From 752c1f4c78fe86d0fd6497387f763306b0d8fc53 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Mon, 27 Feb 2006 13:00:40 -0800
Subject: [PATCH 493/509] [IPSEC]: Kill post_input hook and do NAT-T in
 esp_input directly

The only reason post_input exists at all is that it gives us the
potential to adjust the checksums incrementally in future which
we ought to do.

However, after thinking about it for a bit we can adjust the
checksums without using this post_input stuff at all.  The crucial
point is that only the inner-most NAT-T SA needs to be considered
when adjusting checksums.  What's more, the checksum adjustment
comes down to a single u32 due to the linearity of IP checksums.

We just happen to have a spare u32 lying around in our skb structure :)
When ip_summed is set to CHECKSUM_NONE on input, the value of skb->csum
is currently unused.  All we have to do is to make that the checksum
adjustment and voila, there goes all the post_input and decap structures!

I've left in the decap data structures for now since it's intricately
woven into the sec_path stuff.  We can kill them later too.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/xfrm.h     |   1 -
 net/ipv4/esp4.c        | 128 ++++++++++++-----------------------------
 net/xfrm/xfrm_policy.c |   7 ---
 3 files changed, 38 insertions(+), 98 deletions(-)

diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 004e645f3e18..8d362c49b8a9 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -233,7 +233,6 @@ struct xfrm_type
 	int			(*init_state)(struct xfrm_state *x);
 	void			(*destructor)(struct xfrm_state *);
 	int			(*input)(struct xfrm_state *, struct xfrm_decap_state *, struct sk_buff *skb);
-	int			(*post_input)(struct xfrm_state *, struct xfrm_decap_state *, struct sk_buff *skb);
 	int			(*output)(struct xfrm_state *, struct sk_buff *pskb);
 	/* Estimate maximal size of result of transformation of a dgram */
 	u32			(*get_max_size)(struct xfrm_state *, int size);
diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
index 3f47419cb9c5..09590f356086 100644
--- a/net/ipv4/esp4.c
+++ b/net/ipv4/esp4.c
@@ -12,13 +12,6 @@
 #include <net/protocol.h>
 #include <net/udp.h>
 
-/* decapsulation data for use when post-processing */
-struct esp_decap_data {
-	xfrm_address_t	saddr;
-	__u16		sport;
-	__u8		proto;
-};
-
 static int esp_output(struct xfrm_state *x, struct sk_buff *skb)
 {
 	int err;
@@ -210,25 +203,47 @@ static int esp_input(struct xfrm_state *x, struct xfrm_decap_state *decap, struc
 
 	/* ... check padding bits here. Silly. :-) */ 
 
-	if (x->encap && decap && decap->decap_type) {
-		struct esp_decap_data *encap_data;
-		struct udphdr *uh = (struct udphdr *) (iph+1);
+	if (x->encap) {
+		struct xfrm_encap_tmpl *encap = x->encap;
+		struct udphdr *uh;
 
-		encap_data = (struct esp_decap_data *) (decap->decap_data);
-		encap_data->proto = 0;
-
-		switch (decap->decap_type) {
-		case UDP_ENCAP_ESPINUDP:
-		case UDP_ENCAP_ESPINUDP_NON_IKE:
-			encap_data->proto = AF_INET;
-			encap_data->saddr.a4 = iph->saddr;
-			encap_data->sport = uh->source;
-			encap_len = (void*)esph - (void*)uh;
-			break;
-
-		default:
+		if (encap->encap_type != decap->decap_type)
 			goto out;
+
+		uh = (struct udphdr *)(iph + 1);
+		encap_len = (void*)esph - (void*)uh;
+
+		/*
+		 * 1) if the NAT-T peer's IP or port changed then
+		 *    advertize the change to the keying daemon.
+		 *    This is an inbound SA, so just compare
+		 *    SRC ports.
+		 */
+		if (iph->saddr != x->props.saddr.a4 ||
+		    uh->source != encap->encap_sport) {
+			xfrm_address_t ipaddr;
+
+			ipaddr.a4 = iph->saddr;
+			km_new_mapping(x, &ipaddr, uh->source);
+				
+			/* XXX: perhaps add an extra
+			 * policy check here, to see
+			 * if we should allow or
+			 * reject a packet from a
+			 * different source
+			 * address/port.
+			 */
 		}
+	
+		/*
+		 * 2) ignore UDP/TCP checksums in case
+		 *    of NAT-T in Transport Mode, or
+		 *    perform other post-processing fixes
+		 *    as per draft-ietf-ipsec-udp-encaps-06,
+		 *    section 3.1.2
+		 */
+		if (!x->props.mode)
+			skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}
 
 	iph->protocol = nexthdr[1];
@@ -245,63 +260,6 @@ out:
 	return -EINVAL;
 }
 
-static int esp_post_input(struct xfrm_state *x, struct xfrm_decap_state *decap, struct sk_buff *skb)
-{
-  
-	if (x->encap) {
-		struct xfrm_encap_tmpl *encap;
-		struct esp_decap_data *decap_data;
-
-		encap = x->encap;
-		decap_data = (struct esp_decap_data *)(decap->decap_data);
-
-		/* first, make sure that the decap type == the encap type */
-		if (encap->encap_type != decap->decap_type)
-			return -EINVAL;
-
-		switch (encap->encap_type) {
-		default:
-		case UDP_ENCAP_ESPINUDP:
-		case UDP_ENCAP_ESPINUDP_NON_IKE:
-			/*
-			 * 1) if the NAT-T peer's IP or port changed then
-			 *    advertize the change to the keying daemon.
-			 *    This is an inbound SA, so just compare
-			 *    SRC ports.
-			 */
-			if (decap_data->proto == AF_INET &&
-			    (decap_data->saddr.a4 != x->props.saddr.a4 ||
-			     decap_data->sport != encap->encap_sport)) {
-				xfrm_address_t ipaddr;
-
-				ipaddr.a4 = decap_data->saddr.a4;
-				km_new_mapping(x, &ipaddr, decap_data->sport);
-					
-				/* XXX: perhaps add an extra
-				 * policy check here, to see
-				 * if we should allow or
-				 * reject a packet from a
-				 * different source
-				 * address/port.
-				 */
-			}
-		
-			/*
-			 * 2) ignore UDP/TCP checksums in case
-			 *    of NAT-T in Transport Mode, or
-			 *    perform other post-processing fixes
-			 *    as per * draft-ietf-ipsec-udp-encaps-06,
-			 *    section 3.1.2
-			 */
-			if (!x->props.mode)
-				skb->ip_summed = CHECKSUM_UNNECESSARY;
-
-			break;
-		}
-	}
-	return 0;
-}
-
 static u32 esp4_get_max_size(struct xfrm_state *x, int mtu)
 {
 	struct esp_data *esp = x->data;
@@ -457,7 +415,6 @@ static struct xfrm_type esp_type =
 	.destructor	= esp_destroy,
 	.get_max_size	= esp4_get_max_size,
 	.input		= esp_input,
-	.post_input	= esp_post_input,
 	.output		= esp_output
 };
 
@@ -469,15 +426,6 @@ static struct net_protocol esp4_protocol = {
 
 static int __init esp4_init(void)
 {
-	struct xfrm_decap_state decap;
-
-	if (sizeof(struct esp_decap_data)  >
-	    sizeof(decap.decap_data)) {
-		extern void decap_data_too_small(void);
-
-		decap_data_too_small();
-	}
-
 	if (xfrm_register_type(&esp_type, AF_INET) < 0) {
 		printk(KERN_INFO "ip esp init: can't add xfrm type\n");
 		return -EAGAIN;
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 8206025d8e46..ae62054a9fc4 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -996,13 +996,6 @@ int __xfrm_policy_check(struct sock *sk, int dir, struct sk_buff *skb,
 			struct sec_decap_state *xvec = &(skb->sp->x[i]);
 			if (!xfrm_selector_match(&xvec->xvec->sel, &fl, family))
 				return 0;
-
-			/* If there is a post_input processor, try running it */
-			if (xvec->xvec->type->post_input &&
-			    (xvec->xvec->type->post_input)(xvec->xvec,
-							   &(xvec->decap),
-							   skb) != 0)
-				return 0;
 		}
 	}
 

From e02f7d1603c955126e88cc08149509d00be25cb9 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 27 Feb 2006 13:02:52 -0800
Subject: [PATCH 494/509] [NETFILTER]: nf_queue: don't copy registered rerouter
 data

Use the registered data structure instead of copying it.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/nf_queue.c | 28 +++++++++-------------------
 1 file changed, 9 insertions(+), 19 deletions(-)

diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index d3a4f30a7f22..24ad41e6601b 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -16,7 +16,7 @@
  * for queueing and must reinject all packets it receives, no matter what.
  */
 static struct nf_queue_handler *queue_handler[NPROTO];
-static struct nf_queue_rerouter *queue_rerouter;
+static struct nf_queue_rerouter *queue_rerouter[NPROTO];
 
 static DEFINE_RWLOCK(queue_handler_lock);
 
@@ -64,7 +64,7 @@ int nf_register_queue_rerouter(int pf, struct nf_queue_rerouter *rer)
 		return -EINVAL;
 
 	write_lock_bh(&queue_handler_lock);
-	memcpy(&queue_rerouter[pf], rer, sizeof(queue_rerouter[pf]));
+	queue_rerouter[pf] = rer;
 	write_unlock_bh(&queue_handler_lock);
 
 	return 0;
@@ -77,7 +77,7 @@ int nf_unregister_queue_rerouter(int pf)
 		return -EINVAL;
 
 	write_lock_bh(&queue_handler_lock);
-	memset(&queue_rerouter[pf], 0, sizeof(queue_rerouter[pf]));
+	queue_rerouter[pf] = NULL;
 	write_unlock_bh(&queue_handler_lock);
 	return 0;
 }
@@ -123,7 +123,7 @@ int nf_queue(struct sk_buff **skb,
 		return 1;
 	}
 
-	info = kmalloc(sizeof(*info)+queue_rerouter[pf].rer_size, GFP_ATOMIC);
+	info = kmalloc(sizeof(*info)+queue_rerouter[pf]->rer_size, GFP_ATOMIC);
 	if (!info) {
 		if (net_ratelimit())
 			printk(KERN_ERR "OOM queueing packet %p\n",
@@ -155,14 +155,14 @@ int nf_queue(struct sk_buff **skb,
 		if (physoutdev) dev_hold(physoutdev);
 	}
 #endif
-	if (queue_rerouter[pf].save)
-		queue_rerouter[pf].save(*skb, info);
+	if (queue_rerouter[pf]->save)
+		queue_rerouter[pf]->save(*skb, info);
 
 	status = queue_handler[pf]->outfn(*skb, info, queuenum,
 					  queue_handler[pf]->data);
 
-	if (status >= 0 && queue_rerouter[pf].reroute)
-		status = queue_rerouter[pf].reroute(skb, info);
+	if (status >= 0 && queue_rerouter[pf]->reroute)
+		status = queue_rerouter[pf]->reroute(skb, info);
 
 	read_unlock(&queue_handler_lock);
 
@@ -322,22 +322,12 @@ int __init netfilter_queue_init(void)
 {
 #ifdef CONFIG_PROC_FS
 	struct proc_dir_entry *pde;
-#endif
-	queue_rerouter = kmalloc(NPROTO * sizeof(struct nf_queue_rerouter),
-				 GFP_KERNEL);
-	if (!queue_rerouter)
-		return -ENOMEM;
 
-#ifdef CONFIG_PROC_FS
 	pde = create_proc_entry("nf_queue", S_IRUGO, proc_net_netfilter);
-	if (!pde) {
-		kfree(queue_rerouter);
+	if (!pde)
 		return -1;
-	}
 	pde->proc_fops = &nfqueue_file_ops;
 #endif
-	memset(queue_rerouter, 0, NPROTO * sizeof(struct nf_queue_rerouter));
-
 	return 0;
 }
 

From f92f871989c97a24d284ac60b0f880222ddf87ac Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 27 Feb 2006 13:03:10 -0800
Subject: [PATCH 495/509] [NETFILTER]: nf_queue: check if rerouter is present
 before using it

Every rerouter needs to provide a save and a reroute function, we don't
need to check for them. But we do need to check if a rerouter is registered
at all for the current family, with bridging for example packets of
unregistered families can hit nf_queue.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/nf_queue.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index 24ad41e6601b..1fc7152fba8f 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -155,13 +155,13 @@ int nf_queue(struct sk_buff **skb,
 		if (physoutdev) dev_hold(physoutdev);
 	}
 #endif
-	if (queue_rerouter[pf]->save)
+	if (queue_rerouter[pf])
 		queue_rerouter[pf]->save(*skb, info);
 
 	status = queue_handler[pf]->outfn(*skb, info, queuenum,
 					  queue_handler[pf]->data);
 
-	if (status >= 0 && queue_rerouter[pf]->reroute)
+	if (status >= 0 && queue_rerouter[pf])
 		status = queue_rerouter[pf]->reroute(skb, info);
 
 	read_unlock(&queue_handler_lock);

From 7a11b9848ae27e571f219fab5541bd84700f0d68 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 27 Feb 2006 13:03:24 -0800
Subject: [PATCH 496/509] [NETFILTER]: nf_queue: fix rerouting after packet
 mangling

Packets should be rerouted when they come back from userspace, not before.
Also move the queue_rerouters to RCU to avoid taking the queue_handler_lock
for each reinjected packet.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/nf_queue.c | 22 +++++++++++++++-------
 1 file changed, 15 insertions(+), 7 deletions(-)

diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index 1fc7152fba8f..c61f7237237f 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -6,6 +6,7 @@
 #include <linux/skbuff.h>
 #include <linux/netfilter.h>
 #include <linux/seq_file.h>
+#include <linux/rcupdate.h>
 #include <net/protocol.h>
 
 #include "nf_internals.h"
@@ -64,7 +65,7 @@ int nf_register_queue_rerouter(int pf, struct nf_queue_rerouter *rer)
 		return -EINVAL;
 
 	write_lock_bh(&queue_handler_lock);
-	queue_rerouter[pf] = rer;
+	rcu_assign_pointer(queue_rerouter[pf], rer);
 	write_unlock_bh(&queue_handler_lock);
 
 	return 0;
@@ -77,8 +78,9 @@ int nf_unregister_queue_rerouter(int pf)
 		return -EINVAL;
 
 	write_lock_bh(&queue_handler_lock);
-	queue_rerouter[pf] = NULL;
+	rcu_assign_pointer(queue_rerouter[pf], NULL);
 	write_unlock_bh(&queue_handler_lock);
+	synchronize_rcu();
 	return 0;
 }
 EXPORT_SYMBOL_GPL(nf_unregister_queue_rerouter);
@@ -114,6 +116,7 @@ int nf_queue(struct sk_buff **skb,
 	struct net_device *physindev = NULL;
 	struct net_device *physoutdev = NULL;
 #endif
+	struct nf_queue_rerouter *rerouter;
 
 	/* QUEUE == DROP if noone is waiting, to be safe. */
 	read_lock(&queue_handler_lock);
@@ -155,15 +158,13 @@ int nf_queue(struct sk_buff **skb,
 		if (physoutdev) dev_hold(physoutdev);
 	}
 #endif
-	if (queue_rerouter[pf])
-		queue_rerouter[pf]->save(*skb, info);
+	rerouter = rcu_dereference(queue_rerouter[pf]);
+	if (rerouter)
+		rerouter->save(*skb, info);
 
 	status = queue_handler[pf]->outfn(*skb, info, queuenum,
 					  queue_handler[pf]->data);
 
-	if (status >= 0 && queue_rerouter[pf])
-		status = queue_rerouter[pf]->reroute(skb, info);
-
 	read_unlock(&queue_handler_lock);
 
 	if (status < 0) {
@@ -189,6 +190,7 @@ void nf_reinject(struct sk_buff *skb, struct nf_info *info,
 {
 	struct list_head *elem = &info->elem->list;
 	struct list_head *i;
+	struct nf_queue_rerouter *rerouter;
 
 	rcu_read_lock();
 
@@ -225,6 +227,12 @@ void nf_reinject(struct sk_buff *skb, struct nf_info *info,
 		verdict = NF_ACCEPT;
 	}
 
+	if (verdict == NF_ACCEPT) {
+		rerouter = rcu_dereference(queue_rerouter[info->pf]);
+		if (rerouter && rerouter->reroute(&skb, info) < 0)
+			verdict = NF_DROP;
+	}
+
 	if (verdict == NF_ACCEPT) {
 	next_hook:
 		verdict = nf_iterate(&nf_hooks[info->pf][info->hook],

From e121e9ecb08c3a9843243f461290869ff08be900 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 27 Feb 2006 13:03:39 -0800
Subject: [PATCH 497/509] [NETFILTER]: nf_queue: remove unnecessary check for
 outfn

The only point of registering a queue handler is to provide an outfn,
so there is no need to check for it.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/nf_queue.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index c61f7237237f..913df7dcbada 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -120,7 +120,7 @@ int nf_queue(struct sk_buff **skb,
 
 	/* QUEUE == DROP if noone is waiting, to be safe. */
 	read_lock(&queue_handler_lock);
-	if (!queue_handler[pf] || !queue_handler[pf]->outfn) {
+	if (!queue_handler[pf]) {
 		read_unlock(&queue_handler_lock);
 		kfree_skb(*skb);
 		return 1;

From 45fe4dc08cbf9510074b97a16606366c1d405f4d Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 27 Feb 2006 13:03:55 -0800
Subject: [PATCH 498/509] [NETFILTER]: nf_queue: fix end-of-list check

The comparison wants to find out if the last list iteration reached the
end of the list. It needs to compare the iterator with the list head to
do this, not the element it is looking for.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/nf_queue.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index 913df7dcbada..d9f0d7ef103b 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -214,7 +214,7 @@ void nf_reinject(struct sk_buff *skb, struct nf_info *info,
   			break;
   	}
   
-	if (elem == &nf_hooks[info->pf][info->hook]) {
+	if (i == &nf_hooks[info->pf][info->hook]) {
 		/* The module which sent it to userspace is gone. */
 		NFDEBUG("%s: module disappeared, dropping packet.\n",
 			__FUNCTION__);

From bafac2a512bf4fd2ce7520f3976ce8aab4435f74 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Mon, 27 Feb 2006 13:04:17 -0800
Subject: [PATCH 499/509] [NETFILTER]: Restore {ipt,ip6t,ebt}_LOG compatibility

The nfnetlink_log infrastructure changes broke compatiblity of the LOG
targets. They currently use whatever log backend was registered first,
which means that if ipt_ULOG was loaded first, no messages will be printed
to the ring buffer anymore.

Restore compatiblity by using the old log functions by default and only use
the nf_log backend if the user explicitly said so.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/netfilter_bridge/ebt_log.h | 1 +
 include/linux/netfilter_ipv4/ipt_LOG.h   | 3 ++-
 include/linux/netfilter_ipv6/ip6t_LOG.h  | 3 ++-
 net/bridge/netfilter/ebt_log.c           | 7 ++++++-
 net/ipv4/netfilter/ipt_LOG.c             | 7 ++++++-
 net/ipv6/netfilter/ip6t_LOG.c            | 7 ++++++-
 6 files changed, 23 insertions(+), 5 deletions(-)

diff --git a/include/linux/netfilter_bridge/ebt_log.h b/include/linux/netfilter_bridge/ebt_log.h
index 358fbc84fb59..96e231ae7554 100644
--- a/include/linux/netfilter_bridge/ebt_log.h
+++ b/include/linux/netfilter_bridge/ebt_log.h
@@ -3,6 +3,7 @@
 
 #define EBT_LOG_IP 0x01 /* if the frame is made by ip, log the ip information */
 #define EBT_LOG_ARP 0x02
+#define EBT_LOG_NFLOG 0x04
 #define EBT_LOG_MASK (EBT_LOG_IP | EBT_LOG_ARP)
 #define EBT_LOG_PREFIX_SIZE 30
 #define EBT_LOG_WATCHER "log"
diff --git a/include/linux/netfilter_ipv4/ipt_LOG.h b/include/linux/netfilter_ipv4/ipt_LOG.h
index 22d16177319b..892f9a33fea8 100644
--- a/include/linux/netfilter_ipv4/ipt_LOG.h
+++ b/include/linux/netfilter_ipv4/ipt_LOG.h
@@ -6,7 +6,8 @@
 #define IPT_LOG_TCPOPT		0x02	/* Log TCP options */
 #define IPT_LOG_IPOPT		0x04	/* Log IP options */
 #define IPT_LOG_UID		0x08	/* Log UID owning local socket */
-#define IPT_LOG_MASK		0x0f
+#define IPT_LOG_NFLOG		0x10	/* Log using nf_log backend */
+#define IPT_LOG_MASK		0x1f
 
 struct ipt_log_info {
 	unsigned char level;
diff --git a/include/linux/netfilter_ipv6/ip6t_LOG.h b/include/linux/netfilter_ipv6/ip6t_LOG.h
index 9008ff5c40ae..060c1a1c6c60 100644
--- a/include/linux/netfilter_ipv6/ip6t_LOG.h
+++ b/include/linux/netfilter_ipv6/ip6t_LOG.h
@@ -6,7 +6,8 @@
 #define IP6T_LOG_TCPOPT		0x02	/* Log TCP options */
 #define IP6T_LOG_IPOPT		0x04	/* Log IP options */
 #define IP6T_LOG_UID		0x08	/* Log UID owning local socket */
-#define IP6T_LOG_MASK		0x0f
+#define IP6T_LOG_NFLOG		0x10	/* Log using nf_log backend */
+#define IP6T_LOG_MASK		0x1f
 
 struct ip6t_log_info {
 	unsigned char level;
diff --git a/net/bridge/netfilter/ebt_log.c b/net/bridge/netfilter/ebt_log.c
index 0128fbbe2328..288ff1d4ccc4 100644
--- a/net/bridge/netfilter/ebt_log.c
+++ b/net/bridge/netfilter/ebt_log.c
@@ -166,7 +166,12 @@ static void ebt_log(const struct sk_buff *skb, unsigned int hooknr,
 	li.u.log.level = info->loglevel;
 	li.u.log.logflags = info->bitmask;
 
-	nf_log_packet(PF_BRIDGE, hooknr, skb, in, out, &li, info->prefix);
+	if (info->bitmask & EBT_LOG_NFLOG)
+		nf_log_packet(PF_BRIDGE, hooknr, skb, in, out, &li,
+		              info->prefix);
+	else
+		ebt_log_packet(PF_BRIDGE, hooknr, skb, in, out, &li,
+		               info->prefix);
 }
 
 static struct ebt_watcher log =
diff --git a/net/ipv4/netfilter/ipt_LOG.c b/net/ipv4/netfilter/ipt_LOG.c
index 6606ddb66a29..cc27545ff97f 100644
--- a/net/ipv4/netfilter/ipt_LOG.c
+++ b/net/ipv4/netfilter/ipt_LOG.c
@@ -425,7 +425,12 @@ ipt_log_target(struct sk_buff **pskb,
 	li.u.log.level = loginfo->level;
 	li.u.log.logflags = loginfo->logflags;
 
-	nf_log_packet(PF_INET, hooknum, *pskb, in, out, &li, loginfo->prefix);
+	if (loginfo->logflags & IPT_LOG_NFLOG)
+		nf_log_packet(PF_INET, hooknum, *pskb, in, out, &li,
+		              loginfo->prefix);
+	else
+		ipt_log_packet(PF_INET, hooknum, *pskb, in, out, &li,
+		               loginfo->prefix);
 
 	return IPT_CONTINUE;
 }
diff --git a/net/ipv6/netfilter/ip6t_LOG.c b/net/ipv6/netfilter/ip6t_LOG.c
index 77c725832dec..6b930efa9fb9 100644
--- a/net/ipv6/netfilter/ip6t_LOG.c
+++ b/net/ipv6/netfilter/ip6t_LOG.c
@@ -436,7 +436,12 @@ ip6t_log_target(struct sk_buff **pskb,
 	li.u.log.level = loginfo->level;
 	li.u.log.logflags = loginfo->logflags;
 
-	nf_log_packet(PF_INET6, hooknum, *pskb, in, out, &li, loginfo->prefix);
+	if (loginfo->logflags & IP6T_LOG_NFLOG)
+		nf_log_packet(PF_INET6, hooknum, *pskb, in, out, &li,
+		              loginfo->prefix);
+	else
+		ip6t_log_packet(PF_INET6, hooknum, *pskb, in, out, &li,
+		                loginfo->prefix);
 
 	return IP6T_CONTINUE;
 }

From ba13c98405ba44d37d148376c6615e6c8babbfdc Mon Sep 17 00:00:00 2001
From: Arnaldo Carvalho de Melo <acme@mandriva.com>
Date: Mon, 27 Feb 2006 13:30:43 -0800
Subject: [PATCH 500/509] [REQSK]: Don't reset rskq_defer_accept in
 reqsk_queue_alloc

In 295f7324ff8d9ea58b4d3ec93b1aaa1d80e048a9 I moved defer_accept from
tcp_sock to request_queue and mistakingly reset it at reqsl_queue_alloc, causing
calls to setsockopt(TCP_DEFER_ACCEPT ) to be lost after bind, the fix is to
remove the zeroing of rskq_defer_accept from reqsl_queue_alloc.

Thanks to Alexandra N. Kossovsky <Alexandra.Kossovsky@oktetlabs.ru> for
reporting and testing the suggested fix.

Signed-off-by: Arnaldo Carvalho de Melo <acme@mandriva.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/request_sock.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/net/core/request_sock.c b/net/core/request_sock.c
index b8203de5ff07..98f0fc923f91 100644
--- a/net/core/request_sock.c
+++ b/net/core/request_sock.c
@@ -52,7 +52,6 @@ int reqsk_queue_alloc(struct request_sock_queue *queue,
 	get_random_bytes(&lopt->hash_rnd, sizeof(lopt->hash_rnd));
 	rwlock_init(&queue->syn_wait_lock);
 	queue->rskq_accept_head = queue->rskq_accept_head = NULL;
-	queue->rskq_defer_accept = 0;
 	lopt->nr_table_entries = nr_table_entries;
 
 	write_lock_bh(&queue->syn_wait_lock);

From 07ff2fa8fcb3d9207f1c16e5acf9086d5731ed8b Mon Sep 17 00:00:00 2001
From: Nathan Scott <nathans@sgi.com>
Date: Tue, 28 Feb 2006 12:29:51 +1100
Subject: [PATCH 501/509] [XFS] Fix a realtime allocator regression introduced
 by an old iget race fix.  Noticed by Roger Willcocks.

SGI-PV: 949821
SGI-Modid: xfs-linux-melb:xfs-kern:25257a

Signed-off-by: Nathan Scott <nathans@sgi.com>
---
 fs/xfs/xfs_rtalloc.c | 29 ++++++++++++++---------------
 1 file changed, 14 insertions(+), 15 deletions(-)

diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 06fc061c50fc..5b413946b1c5 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -130,7 +130,8 @@ xfs_growfs_rt_alloc(
 		/*
 		 * Lock the inode.
 		 */
-		if ((error = xfs_trans_iget(mp, tp, ino, 0, XFS_ILOCK_EXCL, &ip)))
+		if ((error = xfs_trans_iget(mp, tp, ino, 0,
+						XFS_ILOCK_EXCL, &ip)))
 			goto error_exit;
 		XFS_BMAP_INIT(&flist, &firstblock);
 		/*
@@ -170,8 +171,8 @@ xfs_growfs_rt_alloc(
 			/*
 			 * Lock the bitmap inode.
 			 */
-			if ((error = xfs_trans_iget(mp, tp, ino, 0, XFS_ILOCK_EXCL,
-					&ip)))
+			if ((error = xfs_trans_iget(mp, tp, ino, 0,
+							XFS_ILOCK_EXCL, &ip)))
 				goto error_exit;
 			/*
 			 * Get a buffer for the block.
@@ -2023,8 +2024,8 @@ xfs_growfs_rt(
 		/*
 		 * Lock out other callers by grabbing the bitmap inode lock.
 		 */
-		if ((error = xfs_trans_iget(mp, tp, 0, mp->m_sb.sb_rbmino,
-				XFS_ILOCK_EXCL, &ip)))
+		if ((error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rbmino, 0,
+						XFS_ILOCK_EXCL, &ip)))
 			goto error_exit;
 		ASSERT(ip == mp->m_rbmip);
 		/*
@@ -2037,8 +2038,8 @@ xfs_growfs_rt(
 		/*
 		 * Get the summary inode into the transaction.
 		 */
-		if ((error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rsumino,
-				0, XFS_ILOCK_EXCL, &ip)))
+		if ((error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rsumino, 0,
+						XFS_ILOCK_EXCL, &ip)))
 			goto error_exit;
 		ASSERT(ip == mp->m_rsumip);
 		/*
@@ -2158,10 +2159,9 @@ xfs_rtallocate_extent(
 	/*
 	 * Lock out other callers by grabbing the bitmap inode lock.
 	 */
-	error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rbmino, 0, XFS_ILOCK_EXCL, &ip);
-	if (error) {
+	if ((error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rbmino, 0,
+					XFS_ILOCK_EXCL, &ip)))
 		return error;
-	}
 	sumbp = NULL;
 	/*
 	 * Allocate by size, or near another block, or exactly at some block.
@@ -2221,10 +2221,9 @@ xfs_rtfree_extent(
 	/*
 	 * Synchronize by locking the bitmap inode.
 	 */
-	error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rbmino, 0, XFS_ILOCK_EXCL, &ip);
-	if (error) {
+	if ((error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rbmino, 0,
+					XFS_ILOCK_EXCL, &ip)))
 		return error;
-	}
 #if defined(__KERNEL__) && defined(DEBUG)
 	/*
 	 * Check to see that this whole range is currently allocated.
@@ -2365,8 +2364,8 @@ xfs_rtpick_extent(
 	__uint64_t	seq;		/* sequence number of file creation */
 	__uint64_t	*seqp;		/* pointer to seqno in inode */
 
-	error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rbmino, 0, XFS_ILOCK_EXCL, &ip);
-	if (error)
+	if ((error = xfs_trans_iget(mp, tp, mp->m_sb.sb_rbmino, 0,
+					XFS_ILOCK_EXCL, &ip)))
 		return error;
 	ASSERT(ip == mp->m_rbmip);
 	seqp = (__uint64_t *)&ip->i_d.di_atime;

From dae81d4774ecbeb7d24bb9a6a4db9f9baee54d85 Mon Sep 17 00:00:00 2001
From: Nathan Scott <nathans@sgi.com>
Date: Tue, 28 Feb 2006 12:30:13 +1100
Subject: [PATCH 502/509] [XFS] Reduce stack use during quota mounts (caused a
 panic).  This regressed recently via the fix for inherited quota inode
 attributes.

SGI-PV: 947312
SGI-Modid: xfs-linux-melb:xfs-kern:25318a

Signed-off-by: Nathan Scott <nathans@sgi.com>
---
 fs/xfs/quota/xfs_qm.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index 53a00fb217fa..7c0e39dc6189 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -68,6 +68,9 @@ kmem_zone_t	*qm_dqzone;
 kmem_zone_t	*qm_dqtrxzone;
 STATIC kmem_shaker_t	xfs_qm_shaker;
 
+STATIC cred_t	xfs_zerocr;
+STATIC xfs_inode_t	xfs_zeroino;
+
 STATIC void	xfs_qm_list_init(xfs_dqlist_t *, char *, int);
 STATIC void	xfs_qm_list_destroy(xfs_dqlist_t *);
 
@@ -1393,8 +1396,6 @@ xfs_qm_qino_alloc(
 	xfs_trans_t	*tp;
 	int		error;
 	unsigned long	s;
-	cred_t		zerocr;
-	xfs_inode_t	zeroino;
 	int		committed;
 
 	tp = xfs_trans_alloc(mp, XFS_TRANS_QM_QINOCREATE);
@@ -1406,11 +1407,9 @@ xfs_qm_qino_alloc(
 		xfs_trans_cancel(tp, 0);
 		return error;
 	}
-	memset(&zerocr, 0, sizeof(zerocr));
-	memset(&zeroino, 0, sizeof(zeroino));
 
-	if ((error = xfs_dir_ialloc(&tp, &zeroino, S_IFREG, 1, 0,
-				   &zerocr, 0, 1, ip, &committed))) {
+	if ((error = xfs_dir_ialloc(&tp, &xfs_zeroino, S_IFREG, 1, 0,
+				   &xfs_zerocr, 0, 1, ip, &committed))) {
 		xfs_trans_cancel(tp, XFS_TRANS_RELEASE_LOG_RES |
 				 XFS_TRANS_ABORT);
 		return error;

From 2353e8e9b6ae29aad77935f21735a30f5cc419b4 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <sandeen@sgi.com>
Date: Tue, 28 Feb 2006 12:30:30 +1100
Subject: [PATCH 503/509] [XFS] Don't map non-uptodate buffers in
 xfs_probe_cluster; also fixes obscure corruption case

SGI-PV: 942658
SGI-Modid: xfs-linux-melb:xfs-kern:207119a

Signed-off-by: Eric Sandeen <sandeen@sgi.com>
Signed-off-by: Nathan Scott <nathans@sgi.com>
---
 fs/xfs/linux-2.6/xfs_aops.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 8f2beec526cf..74d8be87f983 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -540,7 +540,7 @@ xfs_probe_cluster(
 
 	/* First sum forwards in this page */
 	do {
-		if (mapped != buffer_mapped(bh))
+		if (!buffer_uptodate(bh) || (mapped != buffer_mapped(bh)))
 			return total;
 		total += bh->b_size;
 	} while ((bh = bh->b_this_page) != head);

From c3c013a2c218cdede2d2e73df01ed4f813538941 Mon Sep 17 00:00:00 2001
From: Jeff Garzik <jeff@garzik.org>
Date: Mon, 27 Feb 2006 22:31:19 -0500
Subject: [PATCH 504/509] [libata] Disable FUA

Until problems are sorted.
---
 drivers/scsi/libata-core.c | 4 ++++
 drivers/scsi/libata-scsi.c | 2 ++
 drivers/scsi/libata.h      | 1 +
 3 files changed, 7 insertions(+)

diff --git a/drivers/scsi/libata-core.c b/drivers/scsi/libata-core.c
index 5f1d7580218d..4f91b0dc572b 100644
--- a/drivers/scsi/libata-core.c
+++ b/drivers/scsi/libata-core.c
@@ -82,6 +82,10 @@ int atapi_enabled = 0;
 module_param(atapi_enabled, int, 0444);
 MODULE_PARM_DESC(atapi_enabled, "Enable discovery of ATAPI devices (0=off, 1=on)");
 
+int libata_fua = 0;
+module_param_named(fua, libata_fua, int, 0444);
+MODULE_PARM_DESC(fua, "FUA support (0=off, 1=on)");
+
 MODULE_AUTHOR("Jeff Garzik");
 MODULE_DESCRIPTION("Library module for ATA devices");
 MODULE_LICENSE("GPL");
diff --git a/drivers/scsi/libata-scsi.c b/drivers/scsi/libata-scsi.c
index 07b1e7cc61df..59503c9ccac9 100644
--- a/drivers/scsi/libata-scsi.c
+++ b/drivers/scsi/libata-scsi.c
@@ -1708,6 +1708,8 @@ static int ata_dev_supports_fua(u16 *id)
 {
 	unsigned char model[41], fw[9];
 
+	if (!libata_fua)
+		return 0;
 	if (!ata_id_has_fua(id))
 		return 0;
 
diff --git a/drivers/scsi/libata.h b/drivers/scsi/libata.h
index e03ce48b7b4b..fddaf479a544 100644
--- a/drivers/scsi/libata.h
+++ b/drivers/scsi/libata.h
@@ -41,6 +41,7 @@ struct ata_scsi_args {
 
 /* libata-core.c */
 extern int atapi_enabled;
+extern int libata_fua;
 extern struct ata_queued_cmd *ata_qc_new_init(struct ata_port *ap,
 				      struct ata_device *dev);
 extern int ata_rwcmd_protocol(struct ata_queued_cmd *qc);

From 637029c6cb5efcbaa3d5831af4c1972bdd629779 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@g5.osdl.org>
Date: Mon, 27 Feb 2006 20:41:56 -0800
Subject: [PATCH 505/509] Revert "[PATCH] x86_64: Only do the clustered systems
 have unsynchronized TSC assumption on IBM systems"

This reverts commit 13a229abc25640813f1480c0478dfc6bdbc1c19e.

Quoth Andi:
  "After some consideration and feedback from various people it turns
   out this wasn't that good an idea.  It has some problems and needs
   more work.  Since it was only an optimization anyways it's best to
   just back it out again for now."

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
---
 arch/i386/kernel/acpi/boot.c |  3 +++
 arch/x86_64/kernel/apic.c    |  9 +--------
 include/asm-x86_64/acpi.h    | 14 --------------
 3 files changed, 4 insertions(+), 22 deletions(-)

diff --git a/arch/i386/kernel/acpi/boot.c b/arch/i386/kernel/acpi/boot.c
index 8309a7b2cd63..79577f0ace98 100644
--- a/arch/i386/kernel/acpi/boot.c
+++ b/arch/i386/kernel/acpi/boot.c
@@ -44,6 +44,9 @@ extern void __init clustered_apic_check(void);
 extern int gsi_irq_sharing(int gsi);
 #include <asm/proto.h>
 
+static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 0; }
+
+
 #else				/* X86 */
 
 #ifdef	CONFIG_X86_LOCAL_APIC
diff --git a/arch/x86_64/kernel/apic.c b/arch/x86_64/kernel/apic.c
index d70605eda333..e5b14c57eaa0 100644
--- a/arch/x86_64/kernel/apic.c
+++ b/arch/x86_64/kernel/apic.c
@@ -962,14 +962,12 @@ void smp_apic_timer_interrupt(struct pt_regs *regs)
 	irq_exit();
 }
 
-int __initdata unsync_tsc_on_multicluster;
-
 /*
  * oem_force_hpet_timer -- force HPET mode for some boxes.
  *
  * Thus far, the major user of this is IBM's Summit2 series:
  *
- * Some clustered boxes may have unsynced TSC problems if they are
+ * Clustered boxes may have unsynced TSC problems if they are
  * multi-chassis. Use available data to take a good guess.
  * If in doubt, go HPET.
  */
@@ -979,11 +977,6 @@ __cpuinit int oem_force_hpet_timer(void)
 	unsigned id;
 	DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS);
 
-	/* Only do this check on IBM machines - big Unisys systems
-	   use multiple clusters too, but have synchronized TSC */
-	if (!unsync_tsc_on_multicluster)
-		return 0;
-
 	bitmap_zero(clustermap, NUM_APIC_CLUSTERS);
 
 	for (i = 0; i < NR_CPUS; i++) {
diff --git a/include/asm-x86_64/acpi.h b/include/asm-x86_64/acpi.h
index e2b9923189a0..aa1c7b2e438c 100644
--- a/include/asm-x86_64/acpi.h
+++ b/include/asm-x86_64/acpi.h
@@ -164,20 +164,6 @@ extern u8 x86_acpiid_to_apicid[];
 
 extern int acpi_skip_timer_override;
 
-extern int unsync_tsc_on_multicluster;
-
-static inline int acpi_madt_oem_check(char *oem, char *productid) 
-{ 
-	/* Copied from i386. Probably has too many entries. */
-	if (!strncmp(oem, "IBM ENSW", 8) && 
-        	(!strncmp(productid, "VIGIL SMP", 9) 
-             		|| !strncmp(productid, "EXA", 3)
-			|| !strncmp(productid, "RUTHLESS SMP", 12))) {
-		unsync_tsc_on_multicluster = 1;
-        }
-        return 0;
-}
-
 #endif /*__KERNEL__*/
 
 #endif /*_ASM_ACPI_H*/

From 827c1a6c1a5dcb2902fecfb648f9af6a532934eb Mon Sep 17 00:00:00 2001
From: John Rose <johnrose@austin.ibm.com>
Date: Fri, 24 Feb 2006 11:34:23 -0600
Subject: [PATCH 506/509] [PATCH] powerpc: fix dynamic PCI probe regression

Some hotplug driver functions were migrated to the kernel for use by EEH
in commit 2bf6a8fa21570f37fd1789610da30f70a05ac5e3.

Previously, the PCI Hotplug module had been changed to use the new
OFDT-based PCI probe when appropriate:
5fa80fcdca9d20d30c9ecec30d4dbff4ed93a5c6

When rpaphp_pci_config_slot() was moved from the rpaphp driver to the
new kernel function pcibios_add_pci_devices(), the OFDT-based probe
stuff was dropped.  This patch restores it.

Signed-off-by: John Rose <johnrose@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/pseries/eeh.c       | 14 +++++++++
 arch/powerpc/platforms/pseries/pci_dlpar.c | 36 ++++++++++++----------
 include/asm-powerpc/eeh.h                  |  7 +++--
 3 files changed, 38 insertions(+), 19 deletions(-)

diff --git a/arch/powerpc/platforms/pseries/eeh.c b/arch/powerpc/platforms/pseries/eeh.c
index 83578313ee7e..2ab9dcdfb415 100644
--- a/arch/powerpc/platforms/pseries/eeh.c
+++ b/arch/powerpc/platforms/pseries/eeh.c
@@ -893,6 +893,20 @@ void eeh_add_device_tree_early(struct device_node *dn)
 }
 EXPORT_SYMBOL_GPL(eeh_add_device_tree_early);
 
+void eeh_add_device_tree_late(struct pci_bus *bus)
+{
+	struct pci_dev *dev;
+
+	list_for_each_entry(dev, &bus->devices, bus_list) {
+ 		eeh_add_device_late(dev);
+ 		if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE) {
+ 			struct pci_bus *subbus = dev->subordinate;
+ 			if (subbus)
+ 				eeh_add_device_tree_late(subbus);
+ 		}
+	}
+}
+
 /**
  * eeh_add_device_late - perform EEH initialization for the indicated pci device
  * @dev: pci device for which to set up EEH
diff --git a/arch/powerpc/platforms/pseries/pci_dlpar.c b/arch/powerpc/platforms/pseries/pci_dlpar.c
index bdaa8aabdaa6..f3bad900bbcf 100644
--- a/arch/powerpc/platforms/pseries/pci_dlpar.c
+++ b/arch/powerpc/platforms/pseries/pci_dlpar.c
@@ -106,6 +106,8 @@ pcibios_fixup_new_pci_devices(struct pci_bus *bus, int fix_bus)
 			}
 		}
 	}
+
+	eeh_add_device_tree_late(bus);
 }
 EXPORT_SYMBOL_GPL(pcibios_fixup_new_pci_devices);
 
@@ -114,7 +116,6 @@ pcibios_pci_config_bridge(struct pci_dev *dev)
 {
 	u8 sec_busno;
 	struct pci_bus *child_bus;
-	struct pci_dev *child_dev;
 
 	/* Get busno of downstream bus */
 	pci_read_config_byte(dev, PCI_SECONDARY_BUS, &sec_busno);
@@ -129,10 +130,6 @@ pcibios_pci_config_bridge(struct pci_dev *dev)
 
 	pci_scan_child_bus(child_bus);
 
-	list_for_each_entry(child_dev, &child_bus->devices, bus_list) {
-		eeh_add_device_late(child_dev);
-	}
-
 	/* Fixup new pci devices without touching bus struct */
 	pcibios_fixup_new_pci_devices(child_bus, 0);
 
@@ -160,18 +157,25 @@ pcibios_add_pci_devices(struct pci_bus * bus)
 
 	eeh_add_device_tree_early(dn);
 
-	/* pci_scan_slot should find all children */
-	slotno = PCI_SLOT(PCI_DN(dn->child)->devfn);
-	num = pci_scan_slot(bus, PCI_DEVFN(slotno, 0));
-	if (num) {
-		pcibios_fixup_new_pci_devices(bus, 1);
-		pci_bus_add_devices(bus);
-	}
+	if (_machine == PLATFORM_PSERIES_LPAR) {
+		/* use ofdt-based probe */
+		of_scan_bus(dn, bus);
+		if (!list_empty(&bus->devices)) {
+			pcibios_fixup_new_pci_devices(bus, 0);
+			pci_bus_add_devices(bus);
+		}
+	} else {
+		/* use legacy probe */
+		slotno = PCI_SLOT(PCI_DN(dn->child)->devfn);
+		num = pci_scan_slot(bus, PCI_DEVFN(slotno, 0));
+		if (num) {
+			pcibios_fixup_new_pci_devices(bus, 1);
+			pci_bus_add_devices(bus);
+		}
 
-	list_for_each_entry(dev, &bus->devices, bus_list) {
-		eeh_add_device_late (dev);
-		if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE)
-			pcibios_pci_config_bridge(dev);
+		list_for_each_entry(dev, &bus->devices, bus_list)
+			if (dev->hdr_type == PCI_HEADER_TYPE_BRIDGE)
+				pcibios_pci_config_bridge(dev);
 	}
 }
 EXPORT_SYMBOL_GPL(pcibios_add_pci_devices);
diff --git a/include/asm-powerpc/eeh.h b/include/asm-powerpc/eeh.h
index b263fb2fa6e4..7dfb408fe2ca 100644
--- a/include/asm-powerpc/eeh.h
+++ b/include/asm-powerpc/eeh.h
@@ -27,6 +27,7 @@
 #include <linux/string.h>
 
 struct pci_dev;
+struct pci_bus;
 struct device_node;
 
 #ifdef CONFIG_EEH
@@ -61,7 +62,7 @@ void __init pci_addr_cache_build(void);
  */
 void eeh_add_device_early(struct device_node *);
 void eeh_add_device_tree_early(struct device_node *);
-void eeh_add_device_late(struct pci_dev *);
+void eeh_add_device_tree_late(struct pci_bus *);
 
 /**
  * eeh_remove_device - undo EEH setup for the indicated pci device
@@ -116,12 +117,12 @@ static inline void pci_addr_cache_build(void) { }
 
 static inline void eeh_add_device_early(struct device_node *dn) { }
 
-static inline void eeh_add_device_late(struct pci_dev *dev) { }
-
 static inline void eeh_remove_device(struct pci_dev *dev) { }
 
 static inline void eeh_add_device_tree_early(struct device_node *dn) { }
 
+static inline void eeh_add_device_tree_late(struct pci_bus *bus) { }
+
 static inline void eeh_remove_bus_device(struct pci_dev *dev) { }
 #define EEH_POSSIBLE_ERROR(val, type) (0)
 #define EEH_IO_ERROR_VALUE(size) (-1UL)

From 634473db86502b6444c3cebd279a06e0b8737527 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Sun, 26 Feb 2006 08:09:00 +1100
Subject: [PATCH 507/509] [PATCH] powerpc: vdso 64bits gettimeofday bug

A bug in the assembly code of the vdso can cause gettimeofday() to hang
or to return incorrect results. The wrong register was used to test for
pending updates of the calibration variables and to create a dependency
for subsequent loads. This fixes it.

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/kernel/vdso64/gettimeofday.S | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/kernel/vdso64/gettimeofday.S b/arch/powerpc/kernel/vdso64/gettimeofday.S
index ccaeda5136d1..4ee871f1cadb 100644
--- a/arch/powerpc/kernel/vdso64/gettimeofday.S
+++ b/arch/powerpc/kernel/vdso64/gettimeofday.S
@@ -225,9 +225,9 @@ V_FUNCTION_BEGIN(__do_get_xsec)
   .cfi_startproc
 	/* check for update count & load values */
 1:	ld	r8,CFG_TB_UPDATE_COUNT(r3)
-	andi.	r0,r4,1			/* pending update ? loop */
+	andi.	r0,r8,1			/* pending update ? loop */
 	bne-	1b
-	xor	r0,r4,r4		/* create dependency */
+	xor	r0,r8,r8		/* create dependency */
 	add	r3,r3,r0
 
 	/* Get TB & offset it */

From 273d2803817c9e050e8d6c3c271db7d61f2fb259 Mon Sep 17 00:00:00 2001
From: Olaf Hering <olh@suse.de>
Date: Mon, 27 Feb 2006 15:52:59 +0100
Subject: [PATCH 508/509] [PATCH] powerpc: fix NULL pointer in
 handle_eeh_events

This patch fixes a crash in handle_eeh_events,
but ethtool -t still doesnt work right.

...
pepino:~ # cpu 0x3: Vector: 300 (Data Access) at [c00000005192bbe0]
    pc: c00000000004a380: .handle_eeh_events+0xe0/0x23c
    lr: c00000000004a374: .handle_eeh_events+0xd4/0x23c
    sp: c00000005192be60
   msr: 9000000000009032
   dar: 268
 dsisr: 40000000
  current = 0xc0000001fe7bf1a0
  paca    = 0xc00000000048b280
    pid   = 16322, comm = eehd
enter ? for help
[c00000005192bf00] c00000000004a808 .eeh_event_handler+0xcc/0x130
[c00000005192bf90] c000000000025e00 .kernel_thread+0x4c/0x68

...

(none):/# /usr/sbin/ethtool -i eth0
driver: e100
version: 3.5.10-k2-NAPI
firmware-version: N/A
bus-info: 0000:21:01.0
(none):/# /usr/sbin/ethtool -t eth0
Call Trace:
[C00000000F8DEFF0] [C00000000000F270] .show_stack+0x74/0x1b4 (unreliable)
[C00000000F8DF0A0] [C000000000049D04] .eeh_dn_check_failure+0x290/0x2d8
[C00000000F8DF150] [C000000000049E58] .eeh_check_failure+0x10c/0x138
[C00000000F8DF1E0] [C0000000002DFDB0] .e100_hw_reset+0x70/0xf4
[C00000000F8DF270] [C0000000002E1BBC] .e100_hw_init+0x2c/0x260
[C00000000F8DF310] [C0000000002E2464] .e100_loopback_test+0x8c/0x220
[C00000000F8DF3C0] [C0000000002E28DC] .e100_diag_test+0xdc/0x16c
[C00000000F8DF490] [C000000000420BE0] .dev_ethtool+0xf24/0x14f8
[C00000000F8DF8F0] [C00000000041F4A8] .dev_ioctl+0x5cc/0x740
[C00000000F8DFA20] [C00000000040FEFC] .sock_ioctl+0x3d0/0x404
[C00000000F8DFAC0] [C0000000000D513C] .do_ioctl+0x68/0x108
[C00000000F8DFB50] [C0000000000D56B0] .vfs_ioctl+0x4d4/0x510
[C00000000F8DFC10] [C0000000000D5740] .sys_ioctl+0x54/0x94
[C00000000F8DFCC0] [C0000000000FB6EC] .ethtool_ioctl+0x11c/0x150
[C00000000F8DFD60] [C0000000000F7E40] .compat_sys_ioctl+0x338/0x3bc
[C00000000F8DFE30] [C00000000000871C] syscall_exit+0x0/0x40
EEH: Detected PCI bus error on device 0000:21:01.0
EEH: This PCI device has failed 1 times since last reboot: <NULL> -

modprobe: FATAL: Could not load /lib/modules/2.6.16-rc4-git7/modules.dep: No such file or directory

Cannot get strings: No such device
(none):/#
(none):/# EEH: Unable to configure device bridge (-3) for /pci@400000000110/pci@2,2

(none):/# Call Trace:
[C00000000FA17940] [C00000000000F270] .show_stack+0x74/0x1b4 (unreliable)
[C00000000FA179F0] [C000000000049D04] .eeh_dn_check_failure+0x290/0x2d8
[C00000000FA17AA0] [C00000000001E114] .rtas_read_config+0x120/0x154
[C00000000FA17B40] [C000000000049664] .early_enable_eeh+0x274/0x2bc
[C00000000FA17C00] [C000000000049708] .eeh_add_device_early+0x5c/0x6c
[C00000000FA17C90] [C000000000049748] .eeh_add_device_tree_early+0x30/0x5c
[C00000000FA17D20] [C000000000046568] .pcibios_add_pci_devices+0x8c/0x1f8
[C00000000FA17DD0] [C00000000004A528] .eeh_reset_device+0xe0/0x110
[C00000000FA17E60] [C00000000004A698] .handle_eeh_events+0x140/0x250
[C00000000FA17F00] [C00000000004AC7C] .eeh_event_handler+0xe8/0x140
[C00000000FA17F90] [C000000000025784] .kernel_thread+0x4c/0x68
EEH: Detected PCI bus error on device <NULL>
EEH: This PCI device has failed 1 times since last reboot: <NULL> -
EEH: Unable to configure device bridge (-3) for /pci@400000000110/pci@2,2
Call Trace:
[C00000000FA17940] [C00000000000F270] .show_stack+0x74/0x1b4 (unreliable)
[C00000000FA179F0] [C000000000049D04] .eeh_dn_check_failure+0x290/0x2d8
[C00000000FA17AA0] [C00000000001E114] .rtas_read_config+0x120/0x154
[C00000000FA17B40] [C000000000049664] .early_enable_eeh+0x274/0x2bc
[C00000000FA17C00] [C000000000049708] .eeh_add_device_early+0x5c/0x6c
[C00000000FA17C90] [C000000000049748] .eeh_add_device_tree_early+0x30/0x5c
[C00000000FA17D20] [C000000000046568] .pcibios_add_pci_devices+0x8c/0x1f8
[C00000000FA17DD0] [C00000000004A528] .eeh_reset_device+0xe0/0x110
[C00000000FA17E60] [C00000000004A698] .handle_eeh_events+0x140/0x250
[C00000000FA17F00] [C00000000004AC7C] .eeh_event_handler+0xe8/0x140
[C00000000FA17F90] [C000000000025784] .kernel_thread+0x4c/0x68
EEH: Detected PCI bus error on device <NULL>
EEH: This PCI device has failed 1 times since last reboot: <NULL> -
EEH: Unable to configure device bridge (-3) for /pci@400000000110/pci@2,2
Call Trace:
[C00000000FA17940] [C00000000000F270] .show_stack+0x74/0x1b4 (unreliable)
[C00000000FA179F0] [C000000000049D04] .eeh_dn_check_failure+0x290/0x2d8
[C00000000FA17AA0] [C00000000001E114] .rtas_read_config+0x120/0x154
[C00000000FA17B40] [C000000000049664] .early_enable_eeh+0x274/0x2bc
[C00000000FA17C00] [C000000000049708] .eeh_add_device_early+0x5c/0x6c
[C00000000FA17C90] [C000000000049748] .eeh_add_device_tree_early+0x30/0x5c
[C00000000FA17D20] [C000000000046568] .pcibios_add_pci_devices+0x8c/0x1f8
[C00000000FA17DD0] [C00000000004A528] .eeh_reset_device+0xe0/0x110
[C00000000FA17E60] [C00000000004A698] .handle_eeh_events+0x140/0x250
[C00000000FA17F00] [C00000000004AC7C] .eeh_event_handler+0xe8/0x140
[C00000000FA17F90] [C000000000025784] .kernel_thread+0x4c/0x68
EEH: Detected PCI bus error on device <NULL>
and so on

Signed-off-by: Olaf Hering <olh@suse.de>
Acked-by: Linas Vepstas <linas@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/platforms/pseries/eeh_driver.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/platforms/pseries/eeh_driver.c b/arch/powerpc/platforms/pseries/eeh_driver.c
index e3cbba49fd6e..b811d5ff92fe 100644
--- a/arch/powerpc/platforms/pseries/eeh_driver.c
+++ b/arch/powerpc/platforms/pseries/eeh_driver.c
@@ -37,7 +37,7 @@
 
 static inline const char * pcid_name (struct pci_dev *pdev)
 {
-	if (pdev->dev.driver)
+	if (pdev && pdev->dev.driver)
 		return pdev->dev.driver->name;
 	return "";
 }

From 56ec6462af9cba56a04439154e5768672d6f390f Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Tue, 28 Feb 2006 14:54:26 +1100
Subject: [PATCH 509/509] [PATCH] powerpc/iseries: Fix double phys_to_abs bug
 in htab_bolt_mapping

Before the merge I updated create_pte_mapping() to work for iSeries, by
calling iSeries_hpte_bolt_or_insert. (4c55130b2aa93370f1bf52d2304394e91cf8ee39)

Later we changed iSeries_hpte_insert to cope with the bolting case, and called
that instead from create_pte_mapping() (which was renamed to htab_bolt_mapping)
(3c726f8dee6f55e96475574e9f645327e461884c).

Unfortunately that change introduced a subtle bug, where we pass an absolute
address to iSeries_hpte_insert() where it expects a physical address. This
leads to us calling phys_to_abs() twice on the physical address, which is
seriously bogus.

This only causes a problem if the absolute address from the first translation
can be looked up again in the chunk_map, which depends on the size and layout
of memory. I've seen it fail on one box, but not others.

The minimal fix is to pass the physical address to iSeries_hpte_insert(). For
2.6.17 we should make phys_to_abs() BUG if we try to double-translate an
address.

Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Paul Mackerras <paulus@samba.org>
---
 arch/powerpc/mm/hash_utils_64.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index b1f614c612dd..e9d589eefc14 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -169,7 +169,7 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
 #ifdef CONFIG_PPC_ISERIES
 		if (_machine == PLATFORM_ISERIES_LPAR)
 			ret = iSeries_hpte_insert(hpteg, va,
-						  virt_to_abs(paddr),
+						  __pa(vaddr),
 						  tmp_mode,
 						  HPTE_V_BOLTED,
 						  psize);