From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>,
Tvrtko Ursulin <tursulin@ursulin.net>,
igt-dev@lists.freedesktop.org
Cc: Intel-gfx@lists.freedesktop.org
Subject: Re: [igt-dev] [CI i-g-t] tests/perf_pmu: Avoid RT thread for accuracy test
Date: Mon, 26 Mar 2018 13:40:19 +0100 [thread overview]
Message-ID: <d596b0d7-14c5-8daf-8a9b-75895575e9a3@linux.intel.com> (raw)
In-Reply-To: <152206306117.4115.14045420465858687009@mail.alporthouse.com>
On 26/03/2018 12:17, Chris Wilson wrote:
> Quoting Tvrtko Ursulin (2018-03-26 11:57:58)
>> * No self-adjust - instead just report the achieved cycle and let the
>> parent check against it.
>
> Sniff, I was rather proud of our achievement. I had it in mind as a
> template for future autocalibration routines. Is it really useless
> overengineering, or worse broken?
It works fine I think, but the problem is I cannot locate a source of
systematic error which seems proportional to number of loop iterations.
:( After battling with trying to improve it for a couple days I decided
to try to see how the simpler approach will fare on the shards.
There's the tasklet delay, which made me think things could be better
without RT. And then polling on the spinner makes it worse in all cases
for me, however I fiddle with it. So again, I wanted to try the
simplification..
The version from this patch seems super stable on my system, but the 50%
case still has an apparent +.5-6% systematic error. Maybe on the shards
it will not be as stable..
Regards,
Tvrtko
_______________________________________________
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: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>,
Tvrtko Ursulin <tursulin@ursulin.net>,
igt-dev@lists.freedesktop.org
Cc: Intel-gfx@lists.freedesktop.org
Subject: Re: [igt-dev] [Intel-gfx] [CI i-g-t] tests/perf_pmu: Avoid RT thread for accuracy test
Date: Mon, 26 Mar 2018 13:40:19 +0100 [thread overview]
Message-ID: <d596b0d7-14c5-8daf-8a9b-75895575e9a3@linux.intel.com> (raw)
In-Reply-To: <152206306117.4115.14045420465858687009@mail.alporthouse.com>
On 26/03/2018 12:17, Chris Wilson wrote:
> Quoting Tvrtko Ursulin (2018-03-26 11:57:58)
>> * No self-adjust - instead just report the achieved cycle and let the
>> parent check against it.
>
> Sniff, I was rather proud of our achievement. I had it in mind as a
> template for future autocalibration routines. Is it really useless
> overengineering, or worse broken?
It works fine I think, but the problem is I cannot locate a source of
systematic error which seems proportional to number of loop iterations.
:( After battling with trying to improve it for a couple days I decided
to try to see how the simpler approach will fare on the shards.
There's the tasklet delay, which made me think things could be better
without RT. And then polling on the spinner makes it worse in all cases
for me, however I fiddle with it. So again, I wanted to try the
simplification..
The version from this patch seems super stable on my system, but the 50%
case still has an apparent +.5-6% systematic error. Maybe on the shards
it will not be as stable..
Regards,
Tvrtko
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev
next prev parent reply other threads:[~2018-03-26 12:40 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-26 10:57 [CI i-g-t] tests/perf_pmu: Avoid RT thread for accuracy test Tvrtko Ursulin
2018-03-26 10:57 ` [Intel-gfx] " Tvrtko Ursulin
2018-03-26 11:17 ` [igt-dev] " Chris Wilson
2018-03-26 11:17 ` Chris Wilson
2018-03-26 12:40 ` Tvrtko Ursulin [this message]
2018-03-26 12:40 ` [igt-dev] [Intel-gfx] " Tvrtko Ursulin
2018-03-26 11:23 ` [igt-dev] ✓ Fi.CI.BAT: success for " Patchwork
2018-03-26 13:04 ` [igt-dev] ✗ Fi.CI.IGT: warning " Patchwork
2018-03-27 14:31 ` [igt-dev] ✗ Fi.CI.BAT: failure " Patchwork
2018-03-27 17:08 ` [igt-dev] ✗ Fi.CI.BAT: warning " Patchwork
2018-03-28 9:22 ` [igt-dev] ✓ Fi.CI.BAT: success " Patchwork
2018-03-28 14:36 ` [igt-dev] ✓ Fi.CI.IGT: " Patchwork
2018-03-28 16:56 ` Tvrtko Ursulin
2018-03-28 17:10 ` Chris Wilson
2018-04-03 12:38 ` [PATCH i-g-t v2] " Tvrtko Ursulin
2018-04-03 12:38 ` [Intel-gfx] " Tvrtko Ursulin
2018-04-03 13:10 ` Chris Wilson
2018-04-03 13:10 ` [igt-dev] [Intel-gfx] " Chris Wilson
2018-04-03 16:09 ` Tvrtko Ursulin
2018-04-03 16:09 ` [Intel-gfx] " Tvrtko Ursulin
2018-04-03 16:24 ` Chris Wilson
2018-04-03 16:24 ` [igt-dev] [Intel-gfx] " Chris Wilson
2018-04-03 16:39 ` [PATCH i-g-t v3] " Tvrtko Ursulin
2018-04-03 16:39 ` [igt-dev] " Tvrtko Ursulin
2018-04-04 9:51 ` [PATCH i-g-t v4] " Tvrtko Ursulin
2018-04-04 9:51 ` [Intel-gfx] " Tvrtko Ursulin
2018-04-11 13:23 ` Chris Wilson
2018-04-11 13:23 ` [igt-dev] [Intel-gfx] " Chris Wilson
2018-04-11 13:52 ` Tvrtko Ursulin
2018-04-11 13:52 ` [igt-dev] [Intel-gfx] " Tvrtko Ursulin
2018-04-14 11:35 ` Chris Wilson
2018-04-14 11:35 ` [igt-dev] [Intel-gfx] " Chris Wilson
2018-04-16 9:55 ` Tvrtko Ursulin
2018-04-16 9:55 ` [igt-dev] [Intel-gfx] " Tvrtko Ursulin
2018-04-16 10:08 ` Chris Wilson
2018-04-16 10:08 ` [igt-dev] [Intel-gfx] " Chris Wilson
2018-04-03 14:23 ` [igt-dev] ✓ Fi.CI.BAT: success for tests/perf_pmu: Avoid RT thread for accuracy test (rev2) Patchwork
2018-04-03 16:41 ` [igt-dev] ✓ Fi.CI.IGT: " Patchwork
2018-04-03 17:15 ` [igt-dev] ✓ Fi.CI.BAT: success for tests/perf_pmu: Avoid RT thread for accuracy test (rev3) Patchwork
2018-04-03 18:33 ` [igt-dev] ✗ Fi.CI.IGT: failure " Patchwork
2018-04-04 11:13 ` Tvrtko Ursulin
2018-04-04 13:46 ` [igt-dev] ✓ Fi.CI.BAT: success for tests/perf_pmu: Avoid RT thread for accuracy test (rev4) Patchwork
2018-04-04 16:58 ` [igt-dev] ✓ Fi.CI.IGT: " Patchwork
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=d596b0d7-14c5-8daf-8a9b-75895575e9a3@linux.intel.com \
--to=tvrtko.ursulin@linux.intel.com \
--cc=Intel-gfx@lists.freedesktop.org \
--cc=chris@chris-wilson.co.uk \
--cc=igt-dev@lists.freedesktop.org \
--cc=tursulin@ursulin.net \
/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 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.