diff options
author | Peter Zijlstra <peterz@infradead.org> | 2010-04-08 23:03:20 +0200 |
---|---|---|
committer | Ingo Molnar <mingo@elte.hu> | 2010-05-07 11:31:02 +0200 |
commit | ab608344bcbde4f55ec4cd911b686b0ce3eae076 (patch) | |
tree | ebd38efabfaab59d6de11a24143d70e1eec36fae /arch | |
parent | 2b0b5c6fe9b383f3cf35a0a6371c9d577bd523ff (diff) | |
download | linux-ab608344bcbde4f55ec4cd911b686b0ce3eae076.tar.gz linux-ab608344bcbde4f55ec4cd911b686b0ce3eae076.tar.bz2 linux-ab608344bcbde4f55ec4cd911b686b0ce3eae076.zip |
perf, x86: Improve the PEBS ABI
Rename perf_event_attr::precise to perf_event_attr::precise_ip and
widen it to 2 bits. This new field describes the required precision of
the PERF_SAMPLE_IP field:
0 - SAMPLE_IP can have arbitrary skid
1 - SAMPLE_IP must have constant skid
2 - SAMPLE_IP requested to have 0 skid
3 - SAMPLE_IP must have 0 skid
And modify the Intel PEBS code accordingly. The PEBS implementation
now supports up to precise_ip == 2, where we perform the IP fixup.
Also s/PERF_RECORD_MISC_EXACT/&_IP/ to clarify its meaning, this bit
should be set for each PERF_SAMPLE_IP field known to match the actual
instruction triggering the event.
This new scheme allows for a PEBS mode that uses the buffer for more
than a single event.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Stephane Eranian <eranian@google.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'arch')
-rw-r--r-- | arch/x86/kernel/cpu/perf_event.c | 17 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel.c | 4 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event_intel_ds.c | 12 |
3 files changed, 24 insertions, 9 deletions
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 4a3f1f2b9b91..27fa9eeed024 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -488,6 +488,21 @@ static int x86_setup_perfctr(struct perf_event *event) static int x86_pmu_hw_config(struct perf_event *event) { + if (event->attr.precise_ip) { + int precise = 0; + + /* Support for constant skid */ + if (x86_pmu.pebs) + precise++; + + /* Support for IP fixup */ + if (x86_pmu.lbr_nr) + precise++; + + if (event->attr.precise_ip > precise) + return -EOPNOTSUPP; + } + /* * Generate PMC IRQs: * (keep 'enabled' bit clear for now) @@ -1780,7 +1795,7 @@ unsigned long perf_misc_flags(struct pt_regs *regs) } if (regs->flags & PERF_EFLAGS_EXACT) - misc |= PERF_RECORD_MISC_EXACT; + misc |= PERF_RECORD_MISC_EXACT_IP; return misc; } diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index a4b56ac425cb..fdbc652d3feb 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -563,7 +563,7 @@ static void intel_pmu_disable_event(struct perf_event *event) x86_pmu_disable_event(event); - if (unlikely(event->attr.precise)) + if (unlikely(event->attr.precise_ip)) intel_pmu_pebs_disable(event); } @@ -615,7 +615,7 @@ static void intel_pmu_enable_event(struct perf_event *event) return; } - if (unlikely(event->attr.precise)) + if (unlikely(event->attr.precise_ip)) intel_pmu_pebs_enable(event); __x86_pmu_enable_event(hwc, ARCH_PERFMON_EVENTSEL_ENABLE); diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c index 35056f715e9e..18018d1311cd 100644 --- a/arch/x86/kernel/cpu/perf_event_intel_ds.c +++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c @@ -307,7 +307,7 @@ intel_pebs_constraints(struct perf_event *event) { struct event_constraint *c; - if (!event->attr.precise) + if (!event->attr.precise_ip) return NULL; if (x86_pmu.pebs_constraints) { @@ -330,7 +330,7 @@ static void intel_pmu_pebs_enable(struct perf_event *event) cpuc->pebs_enabled |= 1ULL << hwc->idx; WARN_ON_ONCE(cpuc->enabled); - if (x86_pmu.intel_cap.pebs_trap) + if (x86_pmu.intel_cap.pebs_trap && event->attr.precise_ip > 1) intel_pmu_lbr_enable(event); } @@ -345,7 +345,7 @@ static void intel_pmu_pebs_disable(struct perf_event *event) hwc->config |= ARCH_PERFMON_EVENTSEL_INT; - if (x86_pmu.intel_cap.pebs_trap) + if (x86_pmu.intel_cap.pebs_trap && event->attr.precise_ip > 1) intel_pmu_lbr_disable(event); } @@ -485,7 +485,7 @@ static void __intel_pmu_pebs_event(struct perf_event *event, regs.bp = pebs->bp; regs.sp = pebs->sp; - if (intel_pmu_pebs_fixup_ip(regs)) + if (event->attr.precise_ip > 1 && intel_pmu_pebs_fixup_ip(®s)) regs.flags |= PERF_EFLAGS_EXACT; else regs.flags &= ~PERF_EFLAGS_EXACT; @@ -518,7 +518,7 @@ static void intel_pmu_drain_pebs_core(struct pt_regs *iregs) WARN_ON_ONCE(!event); - if (!event->attr.precise) + if (!event->attr.precise_ip) return; n = top - at; @@ -570,7 +570,7 @@ static void intel_pmu_drain_pebs_nhm(struct pt_regs *iregs) WARN_ON_ONCE(!event); - if (!event->attr.precise) + if (!event->attr.precise_ip) continue; if (__test_and_set_bit(bit, (unsigned long *)&status)) |