locking/mcs: Allow architecture specific asm files to be used for contended case

This patch allows each architecture to add its specific assembly optimized
arch_mcs_spin_lock_contended and arch_mcs_spinlock_uncontended for
MCS lock and unlock functions.

Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Scott J Norton <scott.norton@hp.com>
Cc: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
Cc: AswinChandramouleeswaran <aswin@hp.com>
Cc: George Spelvin <linux@horizon.com>
Cc: Rik vanRiel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: MichelLespinasse <walken@google.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Alex Shi <alex.shi@linaro.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: "Figo.zhang" <figo1802@gmail.com>
Cc: "Paul E.McKenney" <paulmck@linux.vnet.ibm.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Waiman Long <waiman.long@hp.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Matthew R Wilcox <matthew.r.wilcox@intel.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1390347382.3138.67.camel@schen9-DESK
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Git-repo: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
Git-commit: ddf1d169c0a489d498c1799a7043904a43b0c159
[joonwoop@codeaurora.org: Resolve merge conflicts; we don't have changes
for arch other than ARM/ARM64]
Signed-off-by: Joonwoo Park <joonwoop@codeaurora.org>

arch/alpha/include/asm/Kbuild

@@ -2,4 +2,5 @@
 
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += mcs_spinlock.h
 generic-y += trace_clock.h

arch/arc/include/asm/Kbuild

@@ -20,6 +20,7 @@ generic-y += kmap_types.h
 generic-y += kvm_para.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += msgbuf.h
 generic-y += param.h

arch/arm/include/asm/Kbuild

@@ -13,6 +13,7 @@ generic-y += irq_regs.h
 generic-y += kdebug.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += msgbuf.h
 generic-y += param.h
 generic-y += parport.h

arch/arm64/include/asm/Kbuild

@@ -21,6 +21,7 @@ generic-y += kmap_types.h
 generic-y += kvm_para.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += msgbuf.h
 generic-y += mutex.h

arch/avr32/include/asm/Kbuild

@@ -1,5 +1,6 @@
 
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += mcs_spinlock.h
 generic-y += param.h
 generic-y += trace_clock.h

arch/blackfin/include/asm/Kbuild

@@ -19,6 +19,7 @@ generic-y += kmap_types.h
 generic-y += kvm_para.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += msgbuf.h
 generic-y += mutex.h

arch/c6x/include/asm/Kbuild

@@ -24,6 +24,7 @@ generic-y += irq_regs.h
 generic-y += kdebug.h
 generic-y += kmap_types.h
 generic-y += local.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += mmu.h
 generic-y += mmu_context.h

arch/cris/include/asm/Kbuild

@@ -5,5 +5,6 @@ header-y += arch-v32/
 
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += mcs_spinlock.h
 generic-y += module.h
 generic-y += trace_clock.h

arch/frv/include/asm/Kbuild

@@ -1,4 +1,5 @@
 
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += mcs_spinlock.h
 generic-y += trace_clock.h

arch/hexagon/include/asm/Kbuild

@@ -25,6 +25,7 @@ generic-y += kdebug.h
 generic-y += kmap_types.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += msgbuf.h
 generic-y += pci.h

arch/ia64/include/asm/Kbuild

@@ -2,4 +2,5 @@
 generic-y += clkdev.h
 generic-y += exec.h
 generic-y += kvm_para.h
+generic-y += mcs_spinlock.h
 generic-y += trace_clock.h

arch/m32r/include/asm/Kbuild

@@ -1,5 +1,6 @@
 
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += mcs_spinlock.h
 generic-y += module.h
 generic-y += trace_clock.h

arch/m68k/include/asm/Kbuild

@@ -15,6 +15,7 @@ generic-y += kmap_types.h
 generic-y += kvm_para.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += mutex.h
 generic-y += percpu.h

arch/metag/include/asm/Kbuild

@@ -23,6 +23,7 @@ generic-y += kmap_types.h
 generic-y += kvm_para.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += msgbuf.h
 generic-y += mutex.h
 generic-y += param.h

arch/microblaze/include/asm/Kbuild

@@ -1,5 +1,6 @@
 
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += mcs_spinlock.h
 generic-y += syscalls.h
 generic-y += trace_clock.h

arch/mips/include/asm/Kbuild

@@ -1,2 +1,3 @@
 # MIPS headers
+generic-y += mcs_spinlock.h
 generic-y += trace_clock.h

arch/mn10300/include/asm/Kbuild

@@ -1,4 +1,6 @@
 
 generic-y += clkdev.h
 generic-y += exec.h
+generic-y += mcs_spinlock.h
+>>>>>>> ddf1d16... locking/mcs: Allow architecture specific asm files to be used for contended case
 generic-y += trace_clock.h

arch/openrisc/include/asm/Kbuild

@@ -34,6 +34,7 @@ generic-y += kdebug.h
 generic-y += kmap_types.h
 generic-y += kvm_para.h
 generic-y += local.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += module.h
 generic-y += msgbuf.h

arch/parisc/include/asm/Kbuild

@@ -11,6 +11,7 @@ generic-y += kdebug.h
 generic-y += kvm_para.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += mutex.h
 generic-y += param.h
 generic-y += percpu.h

arch/powerpc/include/asm/Kbuild

@@ -1,4 +1,5 @@
 
 generic-y += clkdev.h
+generic-y += mcs_spinlock.h
 generic-y += rwsem.h
 generic-y += trace_clock.h

arch/s390/include/asm/Kbuild

@@ -1,4 +1,5 @@
 
 
 generic-y += clkdev.h
+generic-y += mcs_spinlock.h
 generic-y += trace_clock.h

arch/score/include/asm/Kbuild

@@ -2,4 +2,5 @@
 header-y +=
 
 generic-y += clkdev.h
+generic-y += mcs_spinlock.h
 generic-y += trace_clock.h

arch/sh/include/asm/Kbuild

@@ -14,6 +14,7 @@ generic-y += irq_regs.h
 generic-y += kvm_para.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += msgbuf.h
 generic-y += param.h

arch/sparc/include/asm/Kbuild

@@ -10,6 +10,7 @@ generic-y += irq_regs.h
 generic-y += linkage.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += module.h
 generic-y += mutex.h
 generic-y += serial.h

arch/tile/include/asm/Kbuild

@@ -17,6 +17,7 @@ generic-y += ipcbuf.h
 generic-y += irq_regs.h
 generic-y += kdebug.h
 generic-y += local.h
+generic-y += mcs_spinlock.h
 generic-y += msgbuf.h
 generic-y += mutex.h
 generic-y += param.h

arch/um/include/asm/Kbuild

@@ -13,6 +13,7 @@ generic-y += hw_irq.h
 generic-y += io.h
 generic-y += irq_regs.h
 generic-y += kdebug.h
+generic-y += mcs_spinlock.h
 generic-y += mutex.h
 generic-y += param.h
 generic-y += pci.h

arch/unicore32/include/asm/Kbuild

@@ -24,6 +24,7 @@ generic-y += irq_regs.h
 generic-y += kdebug.h
 generic-y += kmap_types.h
 generic-y += local.h
+generic-y += mcs_spinlock.h
 generic-y += mman.h
 generic-y += module.h
 generic-y += msgbuf.h

arch/x86/include/asm/Kbuild

@@ -5,3 +5,4 @@ genhdr-y += unistd_64.h
 genhdr-y += unistd_x32.h
 
 generic-y += clkdev.h
+generic-y += mcs_spinlock.h

arch/xtensa/include/asm/Kbuild

@@ -18,6 +18,7 @@ generic-y += kvm_para.h
 generic-y += linkage.h
 generic-y += local.h
 generic-y += local64.h
+generic-y += mcs_spinlock.h
 generic-y += percpu.h
 generic-y += resource.h
 generic-y += scatterlist.h

include/asm-generic/mcs_spinlock.h

@@ -0,0 +1,13 @@
+#ifndef __ASM_MCS_SPINLOCK_H
+#define __ASM_MCS_SPINLOCK_H
+
+/*
+ * Architectures can define their own:
+ *
+ *   arch_mcs_spin_lock_contended(l)
+ *   arch_mcs_spin_unlock_contended(l)
+ *
+ * See kernel/locking/mcs_spinlock.c.
+ */
+
+#endif /* __ASM_MCS_SPINLOCK_H */

include/linux/mcs_spinlock.h

@@ -0,0 +1,114 @@
+/*
+ * MCS lock defines
+ *
+ * This file contains the main data structure and API definitions of MCS lock.
+ *
+ * The MCS lock (proposed by Mellor-Crummey and Scott) is a simple spin-lock
+ * with the desirable properties of being fair, and with each cpu trying
+ * to acquire the lock spinning on a local variable.
+ * It avoids expensive cache bouncings that common test-and-set spin-lock
+ * implementations incur.
+ */
+#ifndef __LINUX_MCS_SPINLOCK_H
+#define __LINUX_MCS_SPINLOCK_H
+
+#include <asm/mcs_spinlock.h>
+
+struct mcs_spinlock {
+	struct mcs_spinlock *next;
+	int locked; /* 1 if lock acquired */
+};
+
+#ifndef arch_mcs_spin_lock_contended
+/*
+ * Using smp_load_acquire() provides a memory barrier that ensures
+ * subsequent operations happen after the lock is acquired.
+ */
+#define arch_mcs_spin_lock_contended(l)					\
+do {									\
+	while (!(smp_load_acquire(l)))					\
+		arch_mutex_cpu_relax();					\
+} while (0)
+#endif
+
+#ifndef arch_mcs_spin_unlock_contended
+/*
+ * smp_store_release() provides a memory barrier to ensure all
+ * operations in the critical section has been completed before
+ * unlocking.
+ */
+#define arch_mcs_spin_unlock_contended(l)				\
+	smp_store_release((l), 1)
+#endif
+
+/*
+ * Note: the smp_load_acquire/smp_store_release pair is not
+ * sufficient to form a full memory barrier across
+ * cpus for many architectures (except x86) for mcs_unlock and mcs_lock.
+ * For applications that need a full barrier across multiple cpus
+ * with mcs_unlock and mcs_lock pair, smp_mb__after_unlock_lock() should be
+ * used after mcs_lock.
+ */
+
+/*
+ * In order to acquire the lock, the caller should declare a local node and
+ * pass a reference of the node to this function in addition to the lock.
+ * If the lock has already been acquired, then this will proceed to spin
+ * on this node->locked until the previous lock holder sets the node->locked
+ * in mcs_spin_unlock().
+ *
+ * We don't inline mcs_spin_lock() so that perf can correctly account for the
+ * time spent in this lock function.
+ */
+static inline
+void mcs_spin_lock(struct mcs_spinlock **lock, struct mcs_spinlock *node)
+{
+	struct mcs_spinlock *prev;
+
+	/* Init node */
+	node->locked = 0;
+	node->next   = NULL;
+
+	prev = xchg(lock, node);
+	if (likely(prev == NULL)) {
+		/*
+		 * Lock acquired, don't need to set node->locked to 1. Threads
+		 * only spin on its own node->locked value for lock acquisition.
+		 * However, since this thread can immediately acquire the lock
+		 * and does not proceed to spin on its own node->locked, this
+		 * value won't be used. If a debug mode is needed to
+		 * audit lock status, then set node->locked value here.
+		 */
+		return;
+	}
+	ACCESS_ONCE(prev->next) = node;
+
+	/* Wait until the lock holder passes the lock down. */
+	arch_mcs_spin_lock_contended(&node->locked);
+}
+
+/*
+ * Releases the lock. The caller should pass in the corresponding node that
+ * was used to acquire the lock.
+ */
+static inline
+void mcs_spin_unlock(struct mcs_spinlock **lock, struct mcs_spinlock *node)
+{
+	struct mcs_spinlock *next = ACCESS_ONCE(node->next);
+
+	if (likely(!next)) {
+		/*
+		 * Release the lock by setting it to NULL
+		 */
+		if (likely(cmpxchg(lock, node, NULL) == node))
+			return;
+		/* Wait until the next pointer is set */
+		while (!(next = ACCESS_ONCE(node->next)))
+			arch_mutex_cpu_relax();
+	}
+
+	/* Pass lock to next waiter. */
+	arch_mcs_spin_unlock_contended(&next->locked);
+}
+
+#endif /* __LINUX_MCS_SPINLOCK_H */