mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andi Kleen <andi@firstfloor.org>
To: a.p.zijlstra@chello.nl
Cc: x86@kernel.org, linux-kernel@vger.kernel.org, acme@redhat.com,
	eranian@google.com, Andi Kleen <ak@linux.intel.com>
Subject: [PATCH 31/34] perf, x86: Add a Haswell precise instructions event
Date: Thu, 18 Oct 2012 16:19:39 -0700	[thread overview]
Message-ID: <1350602382-12771-32-git-send-email-andi@firstfloor.org> (raw)
In-Reply-To: <1350602382-12771-1-git-send-email-andi@firstfloor.org>

From: Andi Kleen <ak@linux.intel.com>

Add a instructions-p event alias that uses the PDIR randomized instruction
retirement event. This is useful to avoid some systematic sampling shadow
problems. Normally PEBS sampling has a systematic shadow. With PDIR
enabled the hardware adds some randomization that statistically avoids
this problem. In this sense, it's more precise over a whole sampling
interval, but an individual sample can be less precise. But since we
sample overall it's a more precise event.

This could be used before using the explicit event code syntax, but it's easier
and more user friendly to use with an "instructions-p" alias. I expect
this will eventually become a common use case.

Right now for Haswell, will add to Ivy Bridge later too.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 arch/x86/kernel/cpu/perf_event_intel.c |    2 ++
 1 files changed, 2 insertions(+), 0 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index 94389a9..a734b6a 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -2085,6 +2085,7 @@ PMU_EVENT(cycles_t,       "cycles-t",       "event=0x3c,intx=1");
 PMU_EVENT(cycles_ct,      "cycles-ct",      "event=0x3c,intx=1,intx_cp=1");
 PMU_EVENT(insns_t,        "instructions-t", "event=0xc0,intx=1");
 PMU_EVENT(insns_ct,       "instructions-ct","event=0xc0,intx=1,intx_cp=1");
+PMU_EVENT(insns_prec,     "instructions-p", "event=0xc0,umask=0x01,precise=2");
 
 #define PMU_EVENT_PTR(x) &attr_ ## x .attr.attr
 
@@ -2107,6 +2108,7 @@ static struct attribute *hsw_events_attrs[] = {
 	PMU_EVENT_PTR(cycles_ct),
 	PMU_EVENT_PTR(insns_t),
 	PMU_EVENT_PTR(insns_ct),
+	PMU_EVENT_PTR(insns_prec),
 	NULL
 };
 
-- 
1.7.7.6


  parent reply	other threads:[~2012-10-18 23:24 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-18 23:19 perf PMU support for Haswell v3 Andi Kleen
2012-10-18 23:19 ` [PATCH 01/34] perf, x86: Add PEBSv2 record support Andi Kleen
2012-10-18 23:19 ` [PATCH 02/34] perf, x86: Basic Haswell PMU support v2 Andi Kleen
2012-10-18 23:19 ` [PATCH 03/34] perf, x86: Basic Haswell PEBS " Andi Kleen
2012-10-23 12:01   ` Peter Zijlstra
2012-10-23 13:31     ` Andi Kleen
2012-10-18 23:19 ` [PATCH 04/34] perf, x86: Support the TSX intx/intx_cp qualifiers Andi Kleen
2012-10-23 12:09   ` Peter Zijlstra
2012-10-23 13:33     ` Andi Kleen
2012-10-18 23:19 ` [PATCH 05/34] perf, x86: Report PEBS event in a raw format Andi Kleen
2012-10-23 12:31   ` Peter Zijlstra
2012-10-23 12:39     ` Stephane Eranian
2012-10-23 13:45       ` Andi Kleen
2012-10-23 13:58         ` Stephane Eranian
2012-10-23 14:11           ` Andi Kleen
2012-10-23 13:30     ` Andi Kleen
2012-10-23 13:50       ` Peter Zijlstra
2012-10-23 13:57         ` Andi Kleen
2012-10-24  7:50         ` Ingo Molnar
2012-10-18 23:19 ` [PATCH 06/34] perf, kvm: Support the intx/intx_cp modifiers in KVM arch perfmon emulation v2 Andi Kleen
2012-10-23 12:36   ` Peter Zijlstra
2012-10-23 12:53     ` Gleb Natapov
2012-10-23 13:35     ` Andi Kleen
2012-10-23 13:36       ` Andi Kleen
2012-10-23 13:05   ` [06/34] " Gleb Natapov
2012-10-23 13:20     ` Andi Kleen
2012-10-24 13:37       ` Gleb Natapov
2012-10-18 23:19 ` [PATCH 07/34] perf, x86: Support PERF_SAMPLE_ADDR on Haswell Andi Kleen
2012-10-18 23:19 ` [PATCH 08/34] perf, x86: Support Haswell v4 LBR format Andi Kleen
2012-10-23 12:37   ` Peter Zijlstra
2012-10-18 23:19 ` [PATCH 09/34] perf, x86: Disable LBR recording for unknown LBR_FMT Andi Kleen
2012-10-18 23:19 ` [PATCH 10/34] perf, x86: Support LBR filtering by INTX/NOTX/ABORT Andi Kleen
2012-10-18 23:19 ` [PATCH 11/34] perf, tools: Add abort,notx,intx branch filter options to perf report -j Andi Kleen
2012-10-23 12:44   ` Peter Zijlstra
2012-10-18 23:19 ` [PATCH 12/34] perf, tools: Support sorting by intx, abort branch flags Andi Kleen
2012-10-18 23:19 ` [PATCH 13/34] perf, x86: Support full width counting on Haswell Andi Kleen
2012-10-18 23:19 ` [PATCH 14/34] perf, x86: Avoid checkpointed counters causing excessive TSX aborts Andi Kleen
2012-10-23 13:03   ` Peter Zijlstra
2012-10-23 13:22     ` Andi Kleen
2012-10-23 13:05   ` Peter Zijlstra
2012-10-18 23:19 ` [PATCH 15/34] perf, core: Add a concept of a weightened sample Andi Kleen
2012-10-23 13:13   ` Peter Zijlstra
2012-10-23 13:19     ` Andi Kleen
2012-10-18 23:19 ` [PATCH 16/34] perf, x86: Support weight samples for PEBS Andi Kleen
2012-10-23 13:15   ` Peter Zijlstra
2012-10-23 14:05     ` Andi Kleen
2012-10-18 23:19 ` [PATCH 17/34] perf, tools: Add support for weight v2 Andi Kleen
2012-10-18 23:19 ` [PATCH 18/34] perf, tools: Handle XBEGIN like a jump Andi Kleen
2012-10-18 23:19 ` [PATCH 19/34] perf, x86: Support for printing PMU state on spurious PMIs v2 Andi Kleen
2012-10-18 23:19 ` [PATCH 20/34] perf, core: Add generic transaction flags Andi Kleen
2012-10-18 23:19 ` [PATCH 21/34] perf, x86: Add Haswell specific transaction flag reporting Andi Kleen
2012-10-18 23:19 ` [PATCH 22/34] perf, tools: Add support for record transaction flags Andi Kleen
2012-10-18 23:19 ` [PATCH 23/34] perf, tools: Point --sort documentation to --help Andi Kleen
2012-10-18 23:19 ` [PATCH 24/34] perf, tools: Add browser support for transaction flags Andi Kleen
2012-10-18 23:19 ` [PATCH 25/34] perf, tools: Move parse_events error printing to parse_events_options Andi Kleen
2012-10-18 23:19 ` [PATCH 26/34] perf, tools: Support events with - in the name Andi Kleen
2012-10-18 23:19 ` [PATCH 27/34] perf, x86: Report the arch perfmon events in sysfs Andi Kleen
2012-10-18 23:19 ` [PATCH 28/34] tools, perf: Add a precise event qualifier Andi Kleen
2012-10-18 23:19 ` [PATCH 29/34] perf, x86: Add Haswell TSX event aliases Andi Kleen
2012-10-18 23:19 ` [PATCH 30/34] perf, tools: Add perf stat --transaction v2 Andi Kleen
2012-10-18 23:19 ` Andi Kleen [this message]
2012-10-18 23:19 ` [PATCH 32/34] perf, tools: Support generic events as pmu event names v2 Andi Kleen
2012-10-18 23:19 ` [PATCH 33/34] perf, tools: Default to cpu// for events v2 Andi Kleen
2012-10-18 23:19 ` [PATCH 34/34] perf, tools: List kernel supplied event aliases in perf list v2 Andi Kleen
2012-10-19  9:13 ` perf PMU support for Haswell v3 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=1350602382-12771-32-git-send-email-andi@firstfloor.org \
    --to=andi@firstfloor.org \
    --cc=a.p.zijlstra@chello.nl \
    --cc=acme@redhat.com \
    --cc=ak@linux.intel.com \
    --cc=eranian@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=x86@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

Powered by JetHome