2
0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-23 20:53:53 +08:00
linux-next/arch/x86
Peter Zijlstra e8a923cc1f perf/x86: Fix NMI measurements
OK, so what I'm actually seeing on my WSM is that sched/clock.c is
'broken' for the purpose we're using it for.

What triggered it is that my WSM-EP is broken :-(

  [    0.001000] tsc: Fast TSC calibration using PIT
  [    0.002000] tsc: Detected 2533.715 MHz processor
  [    0.500180] TSC synchronization [CPU#0 -> CPU#6]:
  [    0.505197] Measured 3 cycles TSC warp between CPUs, turning off TSC clock.
  [    0.004000] tsc: Marking TSC unstable due to check_tsc_sync_source failed

For some reason it consistently detects TSC skew, even though NHM+
should have a single clock domain for 'reasonable' systems.

This marks sched_clock_stable=0, which means that we do fancy stuff to
try and get a 'sane' clock. Part of this fancy stuff relies on the tick,
clearly that's gone when NOHZ=y. So for idle cpus time gets stuck, until
it either wakes up or gets kicked by another cpu.

While this is perfectly fine for the scheduler -- it only cares about
actually running stuff, and when we're running stuff we're obviously not
idle. This does somewhat break down for perf which can trigger events
just fine on an otherwise idle cpu.

So I've got NMIs get get 'measured' as taking ~1ms, which actually
don't last nearly that long:

          <idle>-0     [013] d.h.   886.311970: rcu_nmi_enter <-do_nmi
  ...
          <idle>-0     [013] d.h.   886.311997: perf_sample_event_took: HERE!!! : 1040990

So ftrace (which uses sched_clock(), not the fancy bits) only sees
~27us, but we measure ~1ms !!

Now since all this measurement stuff lives in x86 code, we can actually
fix it.

Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: mingo@kernel.org
Cc: dave.hansen@linux.intel.com
Cc: eranian@google.com
Cc: Don Zickus <dzickus@redhat.com>
Cc: jmario@redhat.com
Cc: acme@infradead.org
Link: http://lkml.kernel.org/r/20131017133350.GG3364@laptop.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2013-10-29 12:01:20 +01:00
..
boot Merge branch 'x86-kaslr-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2013-09-04 09:38:10 -07:00
configs x86, platform, kvm, kconfig: Turn existing .config's into KVM-capable configs 2013-05-28 12:11:32 +02:00
crypto Reinstate "crypto: crct10dif - Wrap crc_t10dif function all to use crypto transform framework" 2013-09-07 12:56:26 +10:00
ia32 Merge branch 'x86-smap-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2013-09-04 11:08:32 -07:00
include compiler/gcc4: Add quirk for 'asm goto' miscompilation bug 2013-10-11 07:39:14 +02:00
kernel perf/x86: Fix NMI measurements 2013-10-29 12:01:20 +01:00
kvm KVM: nVMX: fix shadow on EPT 2013-10-10 11:39:57 +02:00
lguest lguest: fix GPF in guest when using gdb. 2013-09-06 08:09:28 +09:30
lib Merge branch 'x86-smap-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2013-09-04 11:08:32 -07:00
math-emu
mm x86: finish user fault error path with fatal signal 2013-09-12 15:38:01 -07:00
net net: fix unsafe set_memory_rw from softirq 2013-10-07 15:16:45 -04:00
oprofile oprofilefs_create_...() do not need superblock argument 2013-09-03 22:52:48 -04:00
pci Revert "x86/PCI: MMCONFIG: Check earlier for MMCONFIG region at address zero" 2013-10-04 16:15:29 -06:00
platform x86, efi: Don't map Boot Services on i386 2013-09-18 14:42:33 +01:00
power x86, asmlinkage, power: Make various symbols used by the suspend asm code visible 2013-08-06 14:21:03 -07:00
realmode x86, relocs: Refactor the relocs tool to merge 32- and 64-bit ELF 2013-04-16 16:02:58 -07:00
syscalls unify compat fanotify_mark(2), switch to COMPAT_SYSCALL_DEFINE 2013-05-09 13:46:38 -04:00
tools Merge branch 'x86-asm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2013-09-04 08:39:38 -07:00
um um: prctl: Do not include linux/ptrace.h 2013-09-07 10:57:11 +02:00
vdso remove sched notifier for cross-cpu migrations 2013-07-18 12:29:30 +02:00
video
xen xen: Fix possible user space selector corruption 2013-10-10 14:39:37 +00:00
.gitignore
Kbuild
Kconfig x86/microcode: Correct Kconfig dependencies 2013-10-14 09:24:27 +02:00
Kconfig.cpu x86, 386 removal: Document Nx586 as a 386 and thus unsupported 2012-11-29 13:28:39 -08:00
Kconfig.debug Merge branch 'kconfig-diet' from Dave Hansen 2013-07-04 11:25:51 -07:00
Makefile x86, relocs: Move ELF relocation handling to C 2013-08-07 21:00:04 -07:00
Makefile_32.cpu x86, 386 removal: Remove CONFIG_M386 from Kconfig 2012-11-29 13:23:01 -08:00
Makefile.um