From: Eero Tamminen <eero.t.tamminen@intel.com> To: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>, igt-dev@lists.freedesktop.org Cc: Intel-gfx@lists.freedesktop.org, 3.14pi@ukr.net Subject: Re: [IGT 1/2] tools/intel_gpu_top: Add support for stdout logging Date: Tue, 12 Feb 2019 11:31:18 +0200 [thread overview] Message-ID: <97e1f448-9599-560f-c271-7db1394eb820@intel.com> (raw) In-Reply-To: <9cdcbecc-63a7-eb87-c7c3-4dc941553741@linux.intel.com> Hi, On 11.2.2019 11.49, Tvrtko Ursulin wrote: > On 08/02/2019 13:58, Eero Tamminen wrote: >> On 8.2.2019 14.03, Tvrtko Ursulin wrote: >>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com> >>> >>> Two new output modes are added: listing of text data to standard out (-l >>> on the command line), and dumping of JSON formatted records (-J), >>> also to >>> standard out. >>> >>> The first mode is selected automatically when non-interactive >>> standard out >>> is detected. >>> >>> Example of text output: >>> >>> Freq MHz IRQ RC6 Power IMC MiB/s RCS/0 >>> BCS/0 VCS/0 VCS/1 VECS/0 >>> req act /s % W rd wr % se wa % >>> se wa % se wa % se wa % se wa >>> 0 0 0 0 0.00 360 0 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 35 2 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 34 2 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 143 6 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 169 7 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 169 7 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >> >> Looks nice! >> >> If you add '#' to the start of the header lines, one could use >> something like the attached shell script to convert the saved output >> to SVG graphs with GnuPlot. >> >> Before including the script to igt, it would need to be modified to >> adapt to the number of engines, but maybe intel_gpu_top itself could >> generate the gnuplot control file when it exits, if given e.g. >> --gnuplot argument? > > Hello feature creep! :D > > I could add gnuplot output mode later, just to keep this stage simpler. > In that case I would prefer that this mode outputs a single file (or > stdout stream) which could be fed to gnuplot directly. Eg. intel_gpu_top > -g -o file.out && gnuplot file.out, or even, intel_gpu_top -g | gnuplot. > Quick googling shows it should be possible to embed the data in the > "control file", I didn't know that, but it seems simple. One just uses variable $name for the data part instead of headers, and uses that variable name instead of file name in plot commands after the data. New enough GnuPlot version is in all distros which have new enough kernel. > I am only not sure about the output filename in the > second example. But anyway, first example should definitely work. Gnuplot doesn't care about the control file name extension, and it can read it from stdin. Both of these seem to work fine: gnuplot - < plot.gnu gnuplot < plot.gnu - Eero _______________________________________________ Intel-gfx mailing list Intel-gfx@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/intel-gfx
WARNING: multiple messages have this Message-ID (diff)
From: Eero Tamminen via igt-dev <igt-dev@lists.freedesktop.org> To: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>, igt-dev@lists.freedesktop.org Cc: Intel-gfx@lists.freedesktop.org, 3.14pi@ukr.net, Tvrtko Ursulin <tvrtko.ursulin@intel.com> Subject: Re: [igt-dev] [IGT 1/2] tools/intel_gpu_top: Add support for stdout logging Date: Tue, 12 Feb 2019 11:31:18 +0200 [thread overview] Message-ID: <97e1f448-9599-560f-c271-7db1394eb820@intel.com> (raw) In-Reply-To: <9cdcbecc-63a7-eb87-c7c3-4dc941553741@linux.intel.com> Hi, On 11.2.2019 11.49, Tvrtko Ursulin wrote: > On 08/02/2019 13:58, Eero Tamminen wrote: >> On 8.2.2019 14.03, Tvrtko Ursulin wrote: >>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com> >>> >>> Two new output modes are added: listing of text data to standard out (-l >>> on the command line), and dumping of JSON formatted records (-J), >>> also to >>> standard out. >>> >>> The first mode is selected automatically when non-interactive >>> standard out >>> is detected. >>> >>> Example of text output: >>> >>> Freq MHz IRQ RC6 Power IMC MiB/s RCS/0 >>> BCS/0 VCS/0 VCS/1 VECS/0 >>> req act /s % W rd wr % se wa % >>> se wa % se wa % se wa % se wa >>> 0 0 0 0 0.00 360 0 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 35 2 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 34 2 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 143 6 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 169 7 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >>> 350 350 0 100 0.00 169 7 0.00 0 0 0.00 >>> 0 0 0.00 0 0 0.00 0 0 0.00 0 0 >> >> Looks nice! >> >> If you add '#' to the start of the header lines, one could use >> something like the attached shell script to convert the saved output >> to SVG graphs with GnuPlot. >> >> Before including the script to igt, it would need to be modified to >> adapt to the number of engines, but maybe intel_gpu_top itself could >> generate the gnuplot control file when it exits, if given e.g. >> --gnuplot argument? > > Hello feature creep! :D > > I could add gnuplot output mode later, just to keep this stage simpler. > In that case I would prefer that this mode outputs a single file (or > stdout stream) which could be fed to gnuplot directly. Eg. intel_gpu_top > -g -o file.out && gnuplot file.out, or even, intel_gpu_top -g | gnuplot. > Quick googling shows it should be possible to embed the data in the > "control file", I didn't know that, but it seems simple. One just uses variable $name for the data part instead of headers, and uses that variable name instead of file name in plot commands after the data. New enough GnuPlot version is in all distros which have new enough kernel. > I am only not sure about the output filename in the > second example. But anyway, first example should definitely work. Gnuplot doesn't care about the control file name extension, and it can read it from stdin. Both of these seem to work fine: gnuplot - < plot.gnu gnuplot < plot.gnu - Eero _______________________________________________ igt-dev mailing list igt-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/igt-dev
next prev parent reply other threads:[~2019-02-12 9:31 UTC|newest] Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top 2019-02-08 12:03 [IGT 1/2] tools/intel_gpu_top: Add support for stdout logging Tvrtko Ursulin 2019-02-08 12:03 ` [igt-dev] " Tvrtko Ursulin 2019-02-08 12:03 ` [IGT 2/2] tools/intel_gpu_top: Add file output capability Tvrtko Ursulin 2019-02-08 12:03 ` [igt-dev] " Tvrtko Ursulin 2019-02-08 12:14 ` Chris Wilson 2019-02-08 12:14 ` [igt-dev] " Chris Wilson 2019-02-08 12:10 ` [IGT 1/2] tools/intel_gpu_top: Add support for stdout logging Chris Wilson 2019-02-08 12:10 ` [igt-dev] " Chris Wilson 2019-02-08 12:29 ` [igt-dev] ✓ Fi.CI.BAT: success for series starting with [1/2] " Patchwork 2019-02-25 10:04 ` Tvrtko Ursulin 2019-02-08 13:58 ` [IGT 1/2] " Eero Tamminen 2019-02-08 13:58 ` [igt-dev] " Eero Tamminen 2019-02-11 9:49 ` Tvrtko Ursulin 2019-02-11 9:49 ` [igt-dev] " Tvrtko Ursulin 2019-02-12 9:31 ` Eero Tamminen [this message] 2019-02-12 9:31 ` Eero Tamminen via igt-dev 2019-02-08 14:11 ` [igt-dev] ✓ Fi.CI.IGT: success for series starting with [1/2] " Patchwork 2019-02-11 11:45 [IGT 1/2] " Tvrtko Ursulin 2019-02-11 12:21 ` Chris Wilson 2019-02-11 17:08 ` Tvrtko Ursulin
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=97e1f448-9599-560f-c271-7db1394eb820@intel.com \ --to=eero.t.tamminen@intel.com \ --cc=3.14pi@ukr.net \ --cc=Intel-gfx@lists.freedesktop.org \ --cc=igt-dev@lists.freedesktop.org \ --cc=tvrtko.ursulin@linux.intel.com \ /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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.