From: Andi Kleen <andi@firstfloor.org>
To: Stephane Eranian <eranian@google.com>
Cc: Andi Kleen <andi@firstfloor.org>,
LKML <linux-kernel@vger.kernel.org>, x86 <x86@kernel.org>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Arnaldo Carvalho de Melo <acme@redhat.com>,
Andi Kleen <ak@linux.intel.com>
Subject: Re: [PATCH 18/31] perf, core: Add a concept of a weightened sample
Date: Fri, 28 Sep 2012 16:57:03 +0200 [thread overview]
Message-ID: <20120928145703.GR16230@one.firstfloor.org> (raw)
In-Reply-To: <CABPqkBQPVbobTgaV97E0qvOcfWnK8jMmoyp6NEuMn2+WHZ97HQ@mail.gmail.com>
> I came to the conclusion that yes we need something like a weight or cost
> as a generic way of reporting that in some modes the period is not really
> the right measure to evaluate the "cost" of an event.
I'm not fully sure if you're for or against it. I think
the patch is mostly orthogonal to what you're proposing
My main target is the TSX abort cost, the memory latencies
I just added as a bonus.
>
> I was testing my PEBS Load Latency patch this week, I came to that
> conclusion. The way perf report sorts samples based on aggregated
> periods per IP does not work for PEBS Load Latency (and possibly other
> modes). The sorting needs to be based on some cost that may be distinct
> from the period. By default, it would be the period, but for PEBS LL that
> would be the latency of the load at a specific IP. That would more reflect
> was is going on.
I originally folded the weight into nr_events, but in the end it turned
out it's fairly useful to expose both explicitely as sort keys.
In some cases you want the average weight, in others the total weight
(SUM(weight) * nr_events). I haven't tried to mess with the period
so far.
-Andi
next prev parent reply other threads:[~2012-09-28 14:57 UTC|newest]
Thread overview: 58+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-09-28 4:31 perf PMU support for Haswell Andi Kleen
2012-09-28 4:31 ` [PATCH 01/31] perf, x86: Add PEBSv2 record support Andi Kleen
2012-09-28 8:43 ` Peter Zijlstra
2012-09-28 8:54 ` Stephane Eranian
2012-09-28 9:28 ` Peter Zijlstra
2012-09-28 11:33 ` Stephane Eranian
2012-09-28 14:42 ` Andi Kleen
2012-09-28 4:31 ` [PATCH 02/31] perf, x86: Basic Haswell PMU support Andi Kleen
2012-09-28 9:05 ` Peter Zijlstra
2012-09-28 14:58 ` Andi Kleen
[not found] ` <CABPqkBQ90Crh+EpRQq0Y+xUvrj5vzrX_=SpJQyR4p8uFR_Hr=Q@mail.gmail.com>
2012-09-28 15:21 ` Peter Zijlstra
2012-09-28 15:23 ` Andi Kleen
2012-09-28 4:31 ` [PATCH 03/31] perf, x86: Basic Haswell PEBS support Andi Kleen
2012-09-28 8:50 ` Peter Zijlstra
2012-09-28 4:31 ` [PATCH 04/31] perf, core: Add generic intx/intx_checkpointed counter modifiers Andi Kleen
2012-09-28 9:02 ` Peter Zijlstra
2012-09-28 11:35 ` Stephane Eranian
2012-09-28 14:53 ` Andi Kleen
2012-09-28 15:19 ` Peter Zijlstra
2012-09-28 15:29 ` Andi Kleen
2012-09-28 15:36 ` Peter Zijlstra
2012-09-28 15:23 ` Peter Zijlstra
2012-09-28 15:37 ` Andi Kleen
2012-09-28 4:31 ` [PATCH 05/31] perf, tools: Add :c,:t event modifiers in perf tools Andi Kleen
2012-09-28 4:31 ` [PATCH 06/31] perf, tools: Add intx/intx_checkpoint to perf script and header printing Andi Kleen
2012-09-28 4:31 ` [PATCH 07/31] perf, x86: Implement the :t and :c qualifiers for Haswell Andi Kleen
2012-09-28 4:31 ` [PATCH 08/31] perf, x86: Report PEBS event in a raw format Andi Kleen
2012-09-28 8:54 ` Peter Zijlstra
2012-09-28 8:57 ` Stephane Eranian
2012-09-28 4:31 ` [PATCH 09/31] perf, kvm: Support :t and :c perf modifiers in KVM arch perfmon emulation Andi Kleen
2012-09-28 4:31 ` [PATCH 10/31] perf, x86: Support PERF_SAMPLE_ADDR on Haswell Andi Kleen
2012-09-28 4:31 ` [PATCH 11/31] perf, x86: Support Haswell v4 LBR format Andi Kleen
2012-09-28 4:31 ` [PATCH 12/31] perf, x86: Disable LBR recording for unknown LBR_FMT Andi Kleen
2012-09-28 4:31 ` [PATCH 13/31] perf, x86: Support LBR filtering by INTX/NOTX/ABORT Andi Kleen
2012-09-28 4:31 ` [PATCH 14/31] perf, tools: Add abort,notx,intx branch filter options to perf report -j Andi Kleen
2012-09-28 4:31 ` [PATCH 15/31] perf, tools: Support sorting by intx, abort branch flags Andi Kleen
2012-09-28 4:31 ` [PATCH 16/31] perf, x86: Support full width counting on Haswell Andi Kleen
2012-09-28 4:31 ` [PATCH 17/31] perf, x86: Avoid checkpointed counters causing excessive TSX aborts Andi Kleen
2012-09-28 4:31 ` [PATCH 18/31] perf, core: Add a concept of a weightened sample Andi Kleen
2012-09-28 9:06 ` Stephane Eranian
2012-09-28 14:57 ` Andi Kleen [this message]
2012-09-28 17:09 ` Stephane Eranian
2012-09-28 4:31 ` [PATCH 19/31] perf, x86: Support weight samples for PEBS Andi Kleen
2012-09-28 4:31 ` [PATCH 20/31] perf, tools: Add support for weight Andi Kleen
2012-09-28 4:31 ` [PATCH 21/31] perf, tools: Handle XBEGIN like a jump Andi Kleen
2012-09-28 4:31 ` [PATCH 22/31] perf, core: Define generic hardware transaction events Andi Kleen
2012-09-28 9:33 ` Peter Zijlstra
2012-09-28 4:31 ` [PATCH 23/31] perf, tools: Add support for generic transaction events to perf userspace Andi Kleen
2012-09-28 4:31 ` [PATCH 24/31] perf, x86: Add the Haswell implementation of the generic transaction events Andi Kleen
2012-09-28 4:31 ` [PATCH 25/31] perf, tools: Add perf stat --transaction Andi Kleen
2012-09-28 4:31 ` [PATCH 26/31] perf, x86: Support for printing PMU state on spurious PMIs Andi Kleen
2012-09-28 9:36 ` Peter Zijlstra
2012-09-28 11:39 ` Stephane Eranian
2012-09-28 4:31 ` [PATCH 27/31] perf, core: Add generic transaction flags Andi Kleen
2012-09-28 4:31 ` [PATCH 28/31] perf, x86: Add Haswell specific transaction flag reporting Andi Kleen
2012-09-28 4:31 ` [PATCH 29/31] perf, tools: Add support for record transaction flags Andi Kleen
2012-09-28 4:31 ` [PATCH 30/31] perf, tools: Point --sort documentation to --help Andi Kleen
2012-09-28 4:31 ` [PATCH 31/31] perf, tools: Add browser support for transaction flags 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=20120928145703.GR16230@one.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;
as well as URLs for NNTP newsgroup(s).