linux/arch
Don Zickus 90ed5b0fa5 perf/x86/p4: Block PMIs on init to prevent a stream of unkown NMIs
A bunch of unknown NMIs have popped up on a Pentium4 recently when booting
into a kdump kernel.  This was exposed because the watchdog timer went
from 60 seconds down to 10 seconds (increasing the ability to reproduce
this problem).

What is happening is on boot up of the second kernel (the kdump one),
the previous nmi_watchdogs were enabled on thread 0 and thread 1.  The
second kernel only initializes one cpu but the perf counter on thread 1
still counts.

Normally in a kdump scenario, the other cpus are blocking in an NMI loop,
but more importantly their local apics have the performance counters disabled
(iow LVTPC is masked).  So any counters that fire are masked and never get
through to the second kernel.

However, on a P4 the local apic is shared by both threads and thread1's PMI
(despite being configured to only interrupt thread1) will generate an NMI on
thread0.  Because thread0 knows nothing about this NMI, it is seen as an
unknown NMI.

This would be fine because it is a kdump kernel, strange things happen
what is the big deal about a single unknown NMI.

Unfortunately, the P4 comes with another quirk: clearing the overflow bit
to prevent a stream of NMIs.  This is the problem.

The kdump kernel can not execute because of the endless NMIs that happen.

To solve this, I instrumented the p4 perf init code, to walk all the counters
and zero them out (just like a normal reset would).

Now when the counters go off, they do not generate anything and no unknown
NMIs are seen.

I tested this on a P4 we have in our lab.  After two or three crashes, I could
normally reproduce the problem.  Now after 10 crashes, everything continues
to boot correctly.

Signed-off-by: Don Zickus <dzickus@redhat.com>
Cc: Dave Young <dyoung@redhat.com>
Cc: Vivek Goyal <vgoyal@redhat.com>
Cc: Cyrill Gorcunov <gorcunov@openvz.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20140120154115.GZ25953@redhat.com
[ Fixed a stylistic detail. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2014-02-09 13:20:35 +01:00
..
alpha alpha: fix broken network checksum 2014-01-31 09:21:55 -08:00
arc Merge branch 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild 2014-01-30 16:58:05 -08:00
arm ARM: SoC fixes for 3.14-rc1 2014-02-02 11:11:06 -08:00
arm64 arm64: defconfig: Expand default enabled features 2014-02-07 17:17:28 +00:00
avr32 Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2014-01-25 11:17:34 -08:00
blackfin Merge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media 2014-01-31 09:31:14 -08:00
c6x lib: Add missing arch generic-y entries for asm-generic/hash.h 2013-12-17 21:26:19 -05:00
cris CRIS correction for 3.14 2014-01-28 09:01:14 -08:00
frv Merge branch 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild 2014-01-30 16:58:05 -08:00
hexagon Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2014-01-25 11:17:34 -08:00
ia64 [IA64] Wire up new sched_setattr and sched_getattr syscalls 2014-01-28 09:52:53 -08:00
m32r Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2014-01-25 11:17:34 -08:00
m68k Merge branch 'for-3.14/core' of git://git.kernel.dk/linux-block 2014-01-30 11:19:05 -08:00
metag Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2014-01-25 11:17:34 -08:00
microblaze Microblaze patches for 3.14-rc1 2014-01-28 09:04:11 -08:00
mips MIPS: fpu.h: Fix build when CONFIG_BUG is not set 2014-02-06 13:42:43 +01:00
mn10300 Merge branch 'kbuild' of git://git.kernel.org/pub/scm/linux/kernel/git/mmarek/kbuild 2014-01-30 16:58:05 -08:00
openrisc OpenRISC updates for 3.14 2014-01-30 17:08:41 -08:00
parisc execve: use 'struct filename *' for executable name passing 2014-02-05 12:54:53 -08:00
powerpc Second batch of KVM updates. Some minor x86 fixes, 2014-01-31 08:37:32 -08:00
s390 Merge git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6 2014-02-05 15:51:42 -08:00
score Merge branch 'v4l_for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media 2014-01-31 09:31:14 -08:00
sh Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2014-01-25 11:17:34 -08:00
sparc sparc: Hook up sched_setattr and sched_getattr syscalls. 2014-01-29 00:45:06 -08:00
tile tile: remove compat_sys_lookup_dcookie declaration to fix compile error 2014-02-01 10:55:15 -08:00
um Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml 2014-01-26 11:06:16 -08:00
unicore32 arch/unicore32/kernel/early_printk.c:setup_early_printk: missing initialization 2014-01-27 21:02:39 -08:00
x86 perf/x86/p4: Block PMIs on init to prevent a stream of unkown NMIs 2014-02-09 13:20:35 +01:00
xtensa Merge branch 'for-3.14/core' of git://git.kernel.dk/linux-block 2014-01-30 11:19:05 -08:00
.gitignore
Kconfig stackprotector: Introduce CONFIG_CC_STACKPROTECTOR_STRONG 2013-12-20 09:38:40 +01:00