Piteball/android_kernel_google_msm
1
0
Fork 0
mirror of https://github.com/followmsi/android_kernel_google_msm.git synced 2024-11-06 23:17:41 +00:00
Code Issues Projects Releases Wiki Activity
android_kernel_google_msm/arch/powerpc/kernel
History
Paul Mackerras 242fa18ad8 powerpc: Make sure IPI handlers see data written by IPI senders 
commit 9fb1b36ca1 upstream.

We have been observing hangs, both of KVM guest vcpu tasks and more
generally, where a process that is woken doesn't properly wake up and
continue to run, but instead sticks in TASK_WAKING state.  This
happens because the update of rq->wake_list in ttwu_queue_remote()
is not ordered with the update of ipi_message in
smp_muxed_ipi_message_pass(), and the reading of rq->wake_list in
scheduler_ipi() is not ordered with the reading of ipi_message in
smp_ipi_demux().  Thus it is possible for the IPI receiver not to see
the updated rq->wake_list and therefore conclude that there is nothing
for it to do.

In order to make sure that anything done before smp_send_reschedule()
is ordered before anything done in the resulting call to scheduler_ipi(),
this adds barriers in smp_muxed_message_pass() and smp_ipi_demux().
The barrier in smp_muxed_message_pass() is a full barrier to ensure that
there is a full ordering between the smp_send_reschedule() caller and
scheduler_ipi().  In smp_ipi_demux(), we use xchg() rather than
xchg_local() because xchg() includes release and acquire barriers.
Using xchg() rather than xchg_local() makes sense given that
ipi_message is not just accessed locally.

This moves the barrier between setting the message and calling the
cause_ipi() function into the individual cause_ipi implementations.
Most of them -- those that used outb, out_8 or similar -- already had
a full barrier because out_8 etc. include a sync before the MMIO
store.  This adds an explicit barrier in the two remaining cases.

These changes made no measurable difference to the speed of IPIs as
measured using a simple ping-pong latency test across two CPUs on
different cores of a POWER7 machine.

The analysis of the reason why processes were not waking up properly
is due to Milton Miller.

Reported-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2012-09-14 10:00:22 -07:00
..
vdso32
vdso64
.gitignore
align.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
asm-offsets.c powerpc: Restore correct DSCR in context switch 2012-09-14 10:00:22 -07:00
audit.c
btext.c powerpc: various straight conversions from module.h --> export.h 2011-10-31 19:30:44 -04:00
cacheinfo.c cpu: convert 'cpu' and 'machinecheck' sysdev_class to a regular subsystem 2011-12-21 14:29:42 -08:00
cacheinfo.h
clock.c powerpc: various straight conversions from module.h --> export.h 2011-10-31 19:30:44 -04:00
compat_audit.c
cpu_setup_6xx.S
cpu_setup_44x.S
cpu_setup_a2.S powerpc/book3e: Add ICSWX/ACOP support to Book3e cores like A2 2011-11-25 14:11:28 +11:00
cpu_setup_fsl_booke.S
cpu_setup_pa6t.S
cpu_setup_power7.S
cpu_setup_ppc970.S
cputable.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
crash.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
crash_dump.c powerpc: Rename mapping based RELOCATABLE to DYNAMIC_MEMSTART for BookE 2011-12-20 10:20:19 -05:00
dbell.c powerpc: Make sure IPI handlers see data written by IPI senders 2012-09-14 10:00:22 -07:00
dma-iommu.c PowerPC: adapt for dma_map_ops changes 2012-03-28 16:36:33 +02:00
dma-swiotlb.c PowerPC: adapt for dma_map_ops changes 2012-03-28 16:36:33 +02:00
dma.c PowerPC: adapt for dma_map_ops changes 2012-03-28 16:36:33 +02:00
entry_32.S powerpc: Fix page fault with lockdep regression 2012-04-10 17:21:35 +10:00
entry_64.S powerpc: Restore correct DSCR in context switch 2012-09-14 10:00:22 -07:00
exceptions-64e.S powerpc: Rework lazy-interrupt handling 2012-03-09 13:25:06 +11:00
exceptions-64s.S powerpc/irq: Make alignment & program interrupt behave the same 2012-05-09 09:42:33 +10:00
fadump.c powerpc: Fix fallout from system.h split up 2012-04-02 14:00:03 +10:00
firmware.c powerpc: fix implicit use of cache.h in kernel/firmware.c 2011-10-31 19:30:41 -04:00
fpu.S
fsl_booke_entry_mapping.S
ftrace.c powerpc/ftrace: Fix assembly trampoline register usage 2012-08-09 08:31:27 -07:00
head_8xx.S powerpc: Call do_page_fault() with interrupts off 2012-03-09 10:55:08 +11:00
head_32.S powerpc: Call do_page_fault() with interrupts off 2012-03-09 10:55:08 +11:00
head_40x.S powerpc: Call do_page_fault() with interrupts off 2012-03-09 10:55:08 +11:00
head_44x.S powerpc/44x: Enable CONFIG_RELOCATABLE for PPC44x 2011-12-20 10:21:57 -05:00
head_64.S powerpc: Rework lazy-interrupt handling 2012-03-09 13:25:06 +11:00
head_booke.h powerpc: Call do_page_fault() with interrupts off 2012-03-09 10:55:08 +11:00
head_fsl_booke.S powerpc: Call do_page_fault() with interrupts off 2012-03-09 10:55:08 +11:00
hw_breakpoint.c powerpc: remove non-required uses of include <linux/module.h> 2011-10-31 19:30:44 -04:00
ibmebus.c Merge branch 'for-linus' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping 2012-04-04 17:13:43 -07:00
idle.c powerpc: Fix broken cpu_idle_wait() implementation 2012-06-01 15:18:16 +08:00
idle_6xx.S
idle_book3e.S powerpc: Rework lazy-interrupt handling 2012-03-09 13:25:06 +11:00
idle_e500.S powerpc: e500mc: Fix: use CONFIG_PPC_E500MC in idle_e500.S 2011-10-11 23:23:24 -05:00
idle_power4.S powerpc: Fix power4/970 idle code regression with lockdep 2012-03-16 09:28:17 +11:00
idle_power7.S powerpc: Rework lazy-interrupt handling 2012-03-09 13:25:06 +11:00
init_task.c powerpc: various straight conversions from module.h --> export.h 2011-10-31 19:30:44 -04:00
io-workarounds.c powerpc: io-workarounds.c was implicitly getting init_mm 2011-10-31 19:30:37 -04:00
io.c powerpc: various straight conversions from module.h --> export.h 2011-10-31 19:30:44 -04:00
iomap.c powerpc: switch to GENERIC_PCI_IOMAP 2011-11-28 21:13:18 +02:00
iommu.c fadump: Register for firmware assisted dump. 2012-02-23 10:50:01 +11:00
irq.c powerpc: Fix build of some debug irq code 2012-07-16 09:03:49 -07:00
isa-bridge.c powerpc: Remove FW_FEATURE ISERIES from arch code 2012-03-21 11:16:11 +11:00
jump_label.c powerpc: Fix build breakage in jump_label.c 2011-11-16 14:25:19 +11:00
kgdb.c powerpc: Fix fallout from system.h split up 2012-04-02 14:00:03 +10:00
kprobes.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
kvm.c KVM: PPC: Avoid patching paravirt template code 2012-03-05 14:52:34 +02:00
kvm_emul.S KVM: PPC: Avoid patching paravirt template code 2012-03-05 14:52:34 +02:00
l2cr_6xx.S
legacy_serial.c powerpc: Fix build on some non-freescale platforms 2012-01-25 13:33:22 +11:00
lparcfg.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
machine_kexec.c powerpc/irqdomain: Fix broken NR_IRQ references 2012-04-30 10:45:26 +10:00
machine_kexec_32.c treewide: convert uses of ATTRIB_NORETURN to __noreturn 2012-01-12 20:13:03 -08:00
machine_kexec_64.c treewide: convert uses of ATTRIB_NORETURN to __noreturn 2012-01-12 20:13:03 -08:00
Makefile powerpc/perf: Move perf core & PMU code into a subdirectory 2012-02-23 10:50:04 +11:00
misc.S powerpc: Remove legacy iSeries bits from assembly files 2012-03-09 10:54:59 +11:00
misc_32.S powerpc/4xx: Fix typos in kexec config dependencies 2011-11-16 14:47:54 +11:00
misc_64.S
module.c powerpc: remove non-required uses of include <linux/module.h> 2011-10-31 19:30:44 -04:00
module_32.c powerpc: Fix kernel panic during kernel module load 2012-06-17 11:21:22 -07:00
module_64.c
msi.c
nvram_64.c
of_platform.c powerpc/eeh: Introduce EEH device 2012-03-09 11:39:29 +11:00
paca.c powerpc: Remove FW_FEATURE ISERIES from arch code 2012-03-21 11:16:11 +11:00
pci-common.c Merge branch 'linux-next' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci 2012-03-23 14:02:12 -07:00
pci_32.c powerpc/PCI: compute I/O space bus-to-resource offset consistently 2012-03-20 10:41:44 -07:00
pci_64.c powerpc/PCI: compute I/O space bus-to-resource offset consistently 2012-03-20 10:41:44 -07:00
pci_dn.c powerpc/powernv: PCI support for p7IOC under OPAL v2 2011-11-25 14:53:15 +11:00
pci_of_scan.c powerpc/PCI: convert devtree bus addresses to resource 2012-03-20 10:41:51 -07:00
pmc.c bug.h: add include of it to various implicit C users 2012-02-29 17:15:08 -05:00
ppc32.h
ppc_ksyms.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
ppc_save_regs.S
proc_powerpc.c
process.c powerpc: Fix DSCR inheritance in copy_thread() 2012-09-14 10:00:22 -07:00
prom.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
prom_init.c Disintegrate and delete asm/system.h 2012-03-28 15:58:21 -07:00
prom_init_check.sh powerpc/powernv: Get kernel command line accross OPAL takeover 2011-09-20 16:09:48 +10:00
prom_parse.c powerpc: remove non-required uses of include <linux/module.h> 2011-10-31 19:30:44 -04:00
ptrace.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
ptrace32.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
reloc_32.S powerpc: Process dynamic relocations for kernel 2011-12-20 10:21:08 -05:00
reloc_64.S
rtas-proc.c
rtas-rtc.c
rtas.c Disintegrate and delete asm/system.h 2012-03-28 15:58:21 -07:00
rtas_flash.c powerpc/pseries: Cancel RTAS event scan before firmware flash 2011-11-25 14:11:29 +11:00
rtas_pci.c Merge branch 'linux-next' of git://git.kernel.org/pub/scm/linux/kernel/git/jbarnes/pci 2012-03-23 14:02:12 -07:00
rtasd.c powerpc/pseries: Cancel RTAS event scan before firmware flash 2011-11-25 14:11:29 +11:00
setup-common.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
setup.h
setup_32.c powerpc: fix build when CONFIG_BOOKE_WDT is enabled 2012-04-19 14:55:43 -05:00
setup_64.c Disintegrate and delete asm/system.h 2012-03-28 15:58:21 -07:00
signal.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
signal.h powerpc: Fix various issues with return to userspace 2012-02-22 16:48:53 +11:00
signal_32.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
signal_64.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
smp-tbsync.c atomic: use <linux/atomic.h> 2011-07-26 16:49:47 -07:00
smp.c powerpc: Make sure IPI handlers see data written by IPI senders 2012-09-14 10:00:22 -07:00
softemu8xx.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
stacktrace.c powerpc: various straight conversions from module.h --> export.h 2011-10-31 19:30:44 -04:00
suspend.c
swsusp.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
swsusp_32.S
swsusp_64.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
swsusp_asm64.S
swsusp_booke.S
sys_ppc32.c Disintegrate asm/system.h for PowerPC 2012-03-28 18:30:02 +01:00
syscalls.c
sysfs.c powerpc: Keep thread.dscr and thread.dscr_inherit in sync 2012-09-14 10:00:22 -07:00
systbl.S
systbl_chk.c
systbl_chk.sh
tau_6xx.c
time.c powerpc/time: Sanity check of decrementer expiration is necessary 2012-06-17 11:21:22 -07:00
traps.c powerpc: Keep thread.dscr and thread.dscr_inherit in sync 2012-09-14 10:00:22 -07:00
udbg.c powerpc: Random little legacy iSeries removal tidy ups 2012-03-28 11:33:23 +11:00
udbg_16550.c
vdso.c Disintegrate and delete asm/system.h 2012-03-28 15:58:21 -07:00
vecemu.c
vector.S
vio.c Merge branch 'for-linus' of git://git.linaro.org/people/mszyprowski/linux-dma-mapping 2012-04-04 17:13:43 -07:00
vmlinux.lds.S powerpc: Remove legacy iSeries bits from assembly files 2012-03-09 10:54:59 +11:00