| From | Andi Kleen <> | Subject | [PATCH] [210/275] perf, powerpc: Handle events that raise an exception without overflowing | Date | Wed, 30 Mar 2011 14:07:34 -0700 (PDT) |
| |
2.6.35-longterm review patch. If anyone has any objections, please let me know.
------------------ From: Anton Blanchard <anton@samba.org>
commit 0837e3242c73566fc1c0196b4ec61779c25ffc93 upstream.
Events on POWER7 can roll back if a speculative event doesn't eventually complete. Unfortunately in some rare cases they will raise a performance monitor exception. We need to catch this to ensure we reset the PMC. In all cases the PMC will be 256 or less cycles from overflow.
Signed-off-by: Anton Blanchard <anton@samba.org> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Andi Kleen <ak@linux.intel.com> LKML-Reference: <20110309143842.6c22845e@kryten> Signed-off-by: Ingo Molnar <mingo@elte.hu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
--- arch/powerpc/include/asm/reg.h | 1 + arch/powerpc/kernel/perf_event.c | 24 +++++++++++++++++++++++- 2 files changed, 24 insertions(+), 1 deletion(-)
Index: linux-2.6.35.y/arch/powerpc/include/asm/reg.h =================================================================== --- linux-2.6.35.y.orig/arch/powerpc/include/asm/reg.h 2011-03-29 22:50:34.616372673 -0700 +++ linux-2.6.35.y/arch/powerpc/include/asm/reg.h 2011-03-29 23:03:02.632232817 -0700 @@ -878,6 +878,7 @@ #define PV_970 0x0039 #define PV_POWER5 0x003A #define PV_POWER5p 0x003B +#define PV_POWER7 0x003F #define PV_970FX 0x003C #define PV_630 0x0040 #define PV_630p 0x0041 Index: linux-2.6.35.y/arch/powerpc/kernel/perf_event.c =================================================================== --- linux-2.6.35.y.orig/arch/powerpc/kernel/perf_event.c 2011-03-29 22:50:34.616372673 -0700 +++ linux-2.6.35.y/arch/powerpc/kernel/perf_event.c 2011-03-29 23:03:02.633232791 -0700 @@ -1230,6 +1230,28 @@ return ip; } +static bool pmc_overflow(unsigned long val) +{ + if ((int)val < 0) + return true; + + /* + * Events on POWER7 can roll back if a speculative event doesn't + * eventually complete. Unfortunately in some rare cases they will + * raise a performance monitor exception. We need to catch this to + * ensure we reset the PMC. In all cases the PMC will be 256 or less + * cycles from overflow. + * + * We only do this if the first pass fails to find any overflowing + * PMCs because a user might set a period of less than 256 and we + * don't want to mistakenly reset them. + */ + if (__is_processor(PV_POWER7) && ((0x80000000 - val) <= 256)) + return true; + + return false; +} + /* * Performance monitor interrupt stuff */ @@ -1277,7 +1299,7 @@ if (is_limited_pmc(i + 1)) continue; val = read_pmc(i + 1); - if ((int)val < 0) + if (pmc_overflow(val)) write_pmc(i + 1, 0); } }
|