linux/arch/x86
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
..
boot x86, boot: Fix word-size assumptions in has_eflag() inline asm 2014-01-30 08:04:32 -08:00
configs x86, defconfig: Add DEVTMPFS and DEVTMPFS_MOUNT to *86*_defconfig 2013-11-04 20:01:55 -08:00
crypto crypto: aesni - fix build on x86 (32bit) 2014-01-15 11:36:34 +08:00
ia32 constify copy_siginfo_to_user{,32}() 2013-11-09 00:16:29 -05:00
include x86/nmi: Push duration printk() to irq context 2014-02-09 13:17:22 +01:00
kernel perf/x86/p4: Block PMIs on init to prevent a stream of unkown NMIs 2014-02-09 13:20:35 +01:00
kvm kvm: x86: move KVM_CAP_HYPERV_TIME outside #ifdef 2014-01-29 18:10:45 +01:00
lguest x86, asmlinkage, lguest: Fix C functions used by inline assembler 2014-01-29 22:17:17 -08:00
lib Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next 2014-01-25 11:17:34 -08:00
math-emu x86: math-emu: Drop already-disabled print of build date 2014-01-27 23:14:12 +01:00
mm Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2014-02-08 11:54:43 -08:00
net bpf: do not use reciprocal divide 2014-01-15 17:02:08 -08:00
oprofile perf: Fix arch_perf_out_copy_user default 2013-11-06 12:34:25 +01:00
pci ACPI and power management updates for 3.14-rc1 2014-01-24 15:51:02 -08:00
platform x86/efi: Allow mapping BGRT on x86-32 2014-02-05 23:39:34 +00:00
power
realmode Merge commit 'f4bcd8ccddb02833340652e9f46f5127828eb79d' into x86/build 2014-01-29 09:07:00 -08:00
syscalls sched: Add new scheduler syscalls to support an extended scheduling parameters ABI 2014-01-13 13:41:04 +01:00
tools Merge commit 'f4bcd8ccddb02833340652e9f46f5127828eb79d' into x86/build 2014-01-29 09:07:00 -08:00
um um, x86: Fix vDSO build 2014-01-12 16:47:31 +01:00
vdso Merge branch 'x86-cleanups-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2014-01-20 12:03:57 -08:00
video
xen Bug-fixes: 2014-02-05 16:01:11 -08:00
.gitignore
Kbuild
Kconfig * Avoid WARN_ON() when mapping BGRT on Baytrail (EFI 32-bit). 2014-02-07 11:27:30 -08:00
Kconfig.cpu
Kconfig.debug x86: Disable CONFIG_X86_DECODER_SELFTEST in allmod/allyesconfigs 2014-02-05 14:10:30 -08:00
Makefile x86, build: Build 16-bit code with -m16 where possible 2014-01-30 08:05:36 -08:00
Makefile_32.cpu
Makefile.um