linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jose <jose.sgarcia.aldonza@gmail.com>
To: linux-perf-users@vger.kernel.org
Subject: IPC sampling
Date: Fri, 3 Jun 2016 11:54:57 +0000 (UTC)	[thread overview]
Message-ID: <loom.20160603T134548-395@post.gmane.org> (raw)

Hi all,

I need lot of IPC values for a given program execution. I can use "perf stat":

perf stat -I 10 -e cycles,instructions <program>   

and I can calculate IPC from cycles and instructions.

           time             counts unit events
     0.100137436        163.924.770      cycles
     0.100137436         77.957.069      instructions
     0.200370555        250.248.467      cycles
     0.200370555        124.686.255      instructions
     0.300493502        260.316.390      cycles
     0.300493502        467.691.086      instructions
     0.400623447        260.334.219      cycles
     0.400623447        467.711.817      instructions
     0.500922368        389.205.220      cycles
     0.500922368        689.460.813      instructions
     0.601220080      3.229.694.697      cycles
     0.601220080      6.415.412.796      instructions
      .....

The problem is that the frequency is very limited for my purposes (interval
must be >= 100ms)

It seems that "perf record" can help me:

perf record -F 10000 -e cycles,instructions <ejecutable>    (-F allows
greater frequencies)

My problem now is that I don't know how to obtain "cycles" and
"instructions" from perf.data file. I have tried several options using "perf
report" and "perf script" but I don't achieve it.

Thanks

Jose

             reply	other threads:[~2016-06-03 12:00 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-03 11:54 Jose [this message]
2016-06-03 14:43 ` IPC sampling Andi Kleen
2016-06-03 15:23   ` Jose

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=loom.20160603T134548-395@post.gmane.org \
    --to=jose.sgarcia.aldonza@gmail.com \
    --cc=linux-perf-users@vger.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).