linux/arch
Paul Mackerras 9fb1b36ca1 powerpc: Make sure IPI handlers see data written by IPI senders
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.

Cc: stable@vger.kernel.org # v3.0+
Reported-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
2012-09-05 16:05:22 +10:00
..
alpha alpha: Fix fall-out from disintegrating asm/system.h 2012-08-19 08:41:19 -07:00
arm arm-soc fixes for v3.6-rc3 2012-08-25 17:33:33 -07:00
avr32 ipc: use Kconfig options for __ARCH_WANT_[COMPAT_]IPC_PARSE_VERSION 2012-07-30 17:25:21 -07:00
blackfin Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k 2012-08-03 10:52:41 -07:00
c6x Enable atomic64 ops in C6X 2012-08-17 08:10:12 -07:00
cris ipc: use Kconfig options for __ARCH_WANT_[COMPAT_]IPC_PARSE_VERSION 2012-07-30 17:25:21 -07:00
frv Merge branch 'akpm' (Andrew's patch-bomb) 2012-07-30 17:25:34 -07:00
h8300 ipc: use Kconfig options for __ARCH_WANT_[COMPAT_]IPC_PARSE_VERSION 2012-07-30 17:25:21 -07:00
hexagon Merge branch 'trivial' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild 2012-07-30 11:24:53 -07:00
ia64 [IA64] defconfig: Remove CONFIG_MISC_DEVICES 2012-08-20 13:04:29 -07:00
m32r ipc: use Kconfig options for __ARCH_WANT_[COMPAT_]IPC_PARSE_VERSION 2012-07-30 17:25:21 -07:00
m68k m68k: select CONFIG_GENERIC_ATOMIC64 for all m68k CPU types 2012-08-17 10:04:24 +10:00
microblaze Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k 2012-08-03 10:52:41 -07:00
mips MIPS: pci-ar724x: avoid data bus error due to a missing PCIe module 2012-08-23 15:44:47 +02:00
mn10300 Merge branch 'akpm' (Andrew's patch-bomb) 2012-07-30 17:25:34 -07:00
openrisc Remove useless wrappers of asm-generic/rmap.h 2012-06-28 11:29:26 +02:00
parisc [PARISC] fix personality flag check in copy_thread() 2012-08-03 11:25:12 +01:00
powerpc powerpc: Make sure IPI handlers see data written by IPI senders 2012-09-05 16:05:22 +10:00
s390 s390/32: Don't clobber personality flags on exec 2012-08-30 16:28:07 +02:00
score new helper: signal_delivered() 2012-06-01 12:58:52 -04:00
sh Merge branches 'sh/urgent' and 'sh/gpiolib' into sh-latest 2012-08-09 13:21:13 +09:00
sparc sparc64: Be less verbose during vmemmap population. 2012-08-15 00:37:29 -07:00
tile memcg: rename config variables 2012-07-31 18:42:43 -07:00
um Merge branch 'for-linus-3.6-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml 2012-08-01 16:45:02 -07:00
unicore32 PCI changes for the 3.6 merge window: 2012-07-24 16:17:07 -07:00
x86 KVM: x86: fix KVM_GET_MSR for PV EOI 2012-08-27 18:03:05 -03:00
xtensa xtensa: select generic atomic64_t support 2012-07-31 18:42:39 -07:00
.gitignore
Kconfig ipc: use Kconfig options for __ARCH_WANT_[COMPAT_]IPC_PARSE_VERSION 2012-07-30 17:25:21 -07:00