From: "Yan, Zheng" <zheng.z.yan@intel.com>
To: linux-kernel@vger.kernel.org
Cc: a.p.zijlstra@chello.nl, mingo@kernel.org, acme@infradead.org,
eranian@google.com, andi@firstfloor.org, "Yan,
Zheng" <zheng.z.yan@intel.com>
Subject: [PATCH v4 3/9] perf, x86: use the PEBS auto reload mechanism when possible
Date: Thu, 31 Jul 2014 14:44:58 +0800 [thread overview]
Message-ID: <1406789104-25863-4-git-send-email-zheng.z.yan@intel.com> (raw)
In-Reply-To: <1406789104-25863-1-git-send-email-zheng.z.yan@intel.com>
When a fixed period is specified, this patch make perf use the PEBS
auto reload mechanism. This makes normal profiling faster, because
it avoids one costly MSR write in the PMI handler.
Signef-off-by: Yan, Zheng <zheng.z.yan@intel.com>
---
arch/x86/kernel/cpu/perf_event.c | 15 +++++++++------
arch/x86/kernel/cpu/perf_event.h | 1 +
arch/x86/kernel/cpu/perf_event_intel_ds.c | 9 +++++++++
3 files changed, 19 insertions(+), 6 deletions(-)
diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 46f98f2..3311981 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -982,13 +982,16 @@ int x86_perf_event_set_period(struct perf_event *event)
per_cpu(pmc_prev_left[idx], smp_processor_id()) = left;
- /*
- * The hw event starts counting from this event offset,
- * mark it to be able to extra future deltas:
- */
- local64_set(&hwc->prev_count, (u64)-left);
+ if (!(hwc->flags & PERF_X86_EVENT_AUTO_RELOAD) ||
+ local64_read(&hwc->prev_count) != (u64)-left) {
+ /*
+ * The hw event starts counting from this event offset,
+ * mark it to be able to extra future deltas:
+ */
+ local64_set(&hwc->prev_count, (u64)-left);
- wrmsrl(hwc->event_base, (u64)(-left) & x86_pmu.cntval_mask);
+ wrmsrl(hwc->event_base, (u64)(-left) & x86_pmu.cntval_mask);
+ }
/*
* Due to erratum on certan cpu we need
diff --git a/arch/x86/kernel/cpu/perf_event.h b/arch/x86/kernel/cpu/perf_event.h
index d93a16b..3035930 100644
--- a/arch/x86/kernel/cpu/perf_event.h
+++ b/arch/x86/kernel/cpu/perf_event.h
@@ -69,6 +69,7 @@ struct event_constraint {
#define PERF_X86_EVENT_PEBS_ST 0x2 /* st data address sampling */
#define PERF_X86_EVENT_PEBS_ST_HSW 0x4 /* haswell style st data sampling */
#define PERF_X86_EVENT_COMMITTED 0x8 /* event passed commit_txn */
+#define PERF_X86_EVENT_AUTO_RELOAD 0x10 /* use PEBS auto-reload */
struct amd_nb {
int nb_id; /* NorthBridge id */
diff --git a/arch/x86/kernel/cpu/perf_event_intel_ds.c b/arch/x86/kernel/cpu/perf_event_intel_ds.c
index 696ade3..ecf8b5d6 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_ds.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_ds.c
@@ -716,6 +716,8 @@ void intel_pmu_pebs_enable(struct perf_event *event)
struct hw_perf_event *hwc = &event->hw;
hwc->config &= ~ARCH_PERFMON_EVENTSEL_INT;
+ if (!event->attr.freq)
+ hwc->flags |= PERF_X86_EVENT_AUTO_RELOAD;
cpuc->pebs_enabled |= 1ULL << hwc->idx;
@@ -723,6 +725,12 @@ void intel_pmu_pebs_enable(struct perf_event *event)
cpuc->pebs_enabled |= 1ULL << (hwc->idx + 32);
else if (event->hw.flags & PERF_X86_EVENT_PEBS_ST)
cpuc->pebs_enabled |= 1ULL << 63;
+
+ /* Use auto-reload if possible to save a MSR write in the PMI */
+ if (hwc->flags &PERF_X86_EVENT_AUTO_RELOAD) {
+ ds->pebs_event_reset[hwc->idx] =
+ (u64)-hwc->sample_period & x86_pmu.cntval_mask;
+ }
}
void intel_pmu_pebs_disable(struct perf_event *event)
@@ -741,6 +749,7 @@ void intel_pmu_pebs_disable(struct perf_event *event)
wrmsrl(MSR_IA32_PEBS_ENABLE, cpuc->pebs_enabled);
hwc->config |= ARCH_PERFMON_EVENTSEL_INT;
+ hwc->flags &= ~PERF_X86_EVENT_AUTO_RELOAD;
}
void intel_pmu_pebs_enable_all(void)
--
1.9.3
next prev parent reply other threads:[~2014-07-31 6:45 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-31 6:44 [PATCH v4 0/9] perf, x86: large PEBS interrupt threshold Yan, Zheng
2014-07-31 6:44 ` [PATCH v4 1/9] perf, core: introduce pmu context switch callback Yan, Zheng
2014-07-31 6:44 ` [PATCH v4 2/9] perf, x86: use context switch callback to flush LBR stack Yan, Zheng
2014-07-31 6:44 ` Yan, Zheng [this message]
2014-07-31 6:44 ` [PATCH v4 4/9] perf, x86: introduce setup_pebs_sample_data() Yan, Zheng
2014-07-31 6:45 ` [PATCH v4 5/9] perf, x86: large PEBS interrupt threshold Yan, Zheng
2014-07-31 6:45 ` [PATCH v4 6/9] perf, x86: handle multiple records in PEBS buffer Yan, Zheng
2014-07-31 6:45 ` [PATCH v4 7/9] perf, x86: drain PEBS buffer during context switch Yan, Zheng
2014-07-31 6:45 ` [PATCH v4 8/9] perf, x86: enlarge PEBS buffer Yan, Zheng
2014-07-31 6:45 ` [PATCH v4 9/9] tools, perf: Allow the user to disable time stamps Yan, Zheng
2014-07-31 14:45 ` Arnaldo Carvalho de Melo
2014-08-13 5:13 ` [tip:perf/core] perf record: Honour --no-time command line option tip-bot for Andi Kleen
2014-07-31 7:16 ` [PATCH v4 0/9] perf, x86: large PEBS interrupt threshold Ingo Molnar
2014-07-31 7:18 ` Yan, Zheng
2014-07-31 7:25 ` Ingo Molnar
2014-07-31 7:27 ` Peter Zijlstra
2014-07-31 7:36 ` Yan, Zheng
2014-07-31 7:49 ` Peter Zijlstra
2014-07-31 14:50 ` Andi Kleen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1406789104-25863-4-git-send-email-zheng.z.yan@intel.com \
--to=zheng.z.yan@intel.com \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@infradead.org \
--cc=andi@firstfloor.org \
--cc=eranian@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®