From: Con Kolivas <kernel@kolivas.org>
To: Con Kolivas <kernel@kolivas.org>
Cc: "Jack O'Quin" <joq@io.com>, Alexander Nyberg <alexn@dsv.su.se>,
Ingo Molnar <mingo@elte.hu>, linux <linux-kernel@vger.kernel.org>,
CK Kernel <ck@vds.kolivas.org>, Rui Nuno Capela <rncbc@rncbc.org>
Subject: Re: [PATCH]sched: Isochronous class v2 for unprivileged soft rt scheduling
Date: Wed, 26 Jan 2005 01:38:05 +1100 [thread overview]
Message-ID: <41F659CD.9070904@kolivas.org> (raw)
In-Reply-To: <41F64410.4000702@kolivas.org>
[-- Attachment #1: Type: text/plain, Size: 2864 bytes --]
Con Kolivas wrote:
> There were numerous bugs in the SCHED_ISO design prior to now, so it
> really was not performing as expected. What is most interesting is that
> the DSP load goes to much higher levels now if xruns are avoided and
> stay at those high levels. If I push the cpu load too much so that they
> get transiently throttled from SCHED_ISO, after the Xrun the dsp load
> drops to half. Is this expected behaviour?
>
> Anyway the next patch works well in my environment. Jack, while I
> realise you're getting the results you want from Ingo's dropped
> privilege, dropped cpu limit patch I would appreciate you testing this
> patch. It is not clear yet what direction we will take, but even if we
> dont do this, it would be nice just because of the effort on my part.
>
> This version of the patch has full priority support and both ISO_RR and
> ISO_FIFO.
>
> This is the patch to apply to 2.6.11-rc2-mm1:
> http://ck.kolivas.org/patches/SCHED_ISO/2.6.11-rc2-mm1/2.6.11-rc2-mm1-iso-prio-fifo.diff
Just for completeness, benchmarks:
logs and pretty pictures:
http://ck.kolivas.org/patches/SCHED_ISO/iso3-benchmarks/
SCHED_ISO:
Total seconds ran . . . . . . : 300
Number of clients . . . . . . : 10
Ports per client . . . . . . : 4
Frames per buffer . . . . . . : 64
Number of runs . . . . . . . :( 3)
Timeout Count . . . . . . . . :( 0)
XRUN Count . . . . . . . . . : 0
Delay Count (>spare time) . . : 0
Delay Count (>1000 usecs) . . : 0
Delay Maximum . . . . . . . . : 150 usecs
Cycle Maximum . . . . . . . . : 725 usecs
Average DSP Load. . . . . . . : 32.3 %
Average CPU System Load . . . : 6.0 %
Average CPU User Load . . . . : 33.6 %
Average CPU Nice Load . . . . : 0.0 %
Average CPU I/O Wait Load . . : 0.1 %
Average CPU IRQ Load . . . . : 0.1 %
Average CPU Soft-IRQ Load . . : 0.0 %
Average Interrupt Rate . . . : 1758.9 /sec
Average Context-Switch Rate . : 9208.7 /sec
and SCHED_ISO in the presence of continuous compile:
Total seconds ran . . . . . . : 300
Number of clients . . . . . . : 10
Ports per client . . . . . . : 4
Frames per buffer . . . . . . : 64
Number of runs . . . . . . . :( 3)
Timeout Count . . . . . . . . :( 0)
XRUN Count . . . . . . . . . : 0
Delay Count (>spare time) . . : 0
Delay Count (>1000 usecs) . . : 0
Delay Maximum . . . . . . . . : 375 usecs
Cycle Maximum . . . . . . . . : 726 usecs
Average DSP Load. . . . . . . : 35.8 %
Average CPU System Load . . . : 15.1 %
Average CPU User Load . . . . : 82.9 %
Average CPU Nice Load . . . . : 0.0 %
Average CPU I/O Wait Load . . : 1.8 %
Average CPU IRQ Load . . . . : 0.2 %
Average CPU Soft-IRQ Load . . : 0.0 %
Average Interrupt Rate . . . : 1772.6 /sec
Average Context-Switch Rate . : 9565.2 /sec
Cheers,
Con
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 256 bytes --]
next prev parent reply other threads:[~2005-01-25 14:39 UTC|newest]
Thread overview: 198+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-01-19 22:39 [PATCH]sched: Isochronous class v2 for unprivileged soft rt scheduling Con Kolivas
2005-01-20 0:16 ` utz lehmann
2005-01-20 0:33 ` Con Kolivas
2005-01-20 4:26 ` utz lehmann
2005-01-20 5:55 ` Con Kolivas
2005-01-20 17:54 ` Alexander Nyberg
2005-01-20 20:27 ` Con Kolivas
2005-01-20 0:53 ` Con Kolivas
2005-01-20 1:32 ` Jack O'Quin
2005-01-20 2:06 ` Con Kolivas
2005-01-20 2:45 ` Jack O'Quin
2005-01-20 4:07 ` Con Kolivas
2005-01-20 4:57 ` Jack O'Quin
2005-01-20 5:05 ` Gene Heskett
2005-01-20 5:59 ` Con Kolivas
2005-01-20 6:35 ` Con Kolivas
2005-01-20 15:19 ` Jack O'Quin
2005-01-20 15:42 ` Paul Davis
2005-01-20 16:47 ` Jack O'Quin
2005-01-20 17:25 ` Ingo Molnar
2005-01-22 0:09 ` Jack O'Quin
2005-01-22 16:54 ` Ingo Molnar
2005-01-22 21:23 ` Jack O'Quin
2005-01-23 2:06 ` Nick Piggin
2005-01-23 2:58 ` Chris Wright
2005-01-24 8:59 ` Ingo Molnar
2005-01-24 9:55 ` Paolo Ciarrocchi
2005-01-24 10:29 ` Nick Piggin
2005-01-24 10:46 ` Ingo Molnar
2005-01-24 12:58 ` [patch, 2.6.11-rc2] sched: /proc/sys/kernel/rt_cpu_limit tunable Ingo Molnar
2005-01-24 13:34 ` Ingo Molnar
2005-01-24 13:53 ` Con Kolivas
2005-01-24 14:01 ` [ck] " Con Kolivas
[not found] ` <87k6q2umla.fsf@sulphur.joq.us>
2005-01-25 6:28 ` Nick Piggin
2005-01-25 14:12 ` Ingo Molnar
2005-01-25 8:37 ` Ingo Molnar
2005-01-25 21:36 ` Jack O'Quin
2005-01-25 21:49 ` Ingo Molnar
2005-01-25 21:55 ` Chris Wright
2005-01-25 21:57 ` Ingo Molnar
2005-01-25 22:03 ` Chris Wright
2005-01-25 22:08 ` Ingo Molnar
2005-01-25 22:16 ` Chris Wright
2005-01-25 22:44 ` Bill Rugolsky Jr.
2005-01-26 5:12 ` Jack O'Quin
2005-01-26 7:27 ` Ingo Molnar
2005-01-26 11:02 ` [patch, 2.6.11-rc2] sched: RLIMIT_RT_CPU feature, -D7 Ingo Molnar
2005-01-25 13:56 ` [patch, 2.6.11-rc2] sched: RLIMIT_RT_CPU_RATIO feature Ingo Molnar
2005-01-25 14:06 ` Con Kolivas
2005-01-25 22:18 ` Peter Williams
2005-01-25 22:26 ` Peter Williams
2005-01-26 10:08 ` [patch, 2.6.11-rc2] sched: RLIMIT_RT_CPU feature, -D7 Ingo Molnar
2005-01-26 14:22 ` Jack O'Quin
2005-01-26 16:18 ` [ck] " Cal
2005-01-26 16:29 ` Cal
2005-01-26 16:41 ` Jack O'Quin
2005-01-26 17:57 ` Cal
2005-01-26 18:57 ` Jack O'Quin
2005-01-27 2:03 ` Cal
2005-01-27 8:51 ` [patch, 2.6.11-rc2] sched: RLIMIT_RT_CPU feature, -D8 Ingo Molnar
2005-01-27 12:48 ` Cal
2005-01-27 16:31 ` Mike Galbraith
2005-01-26 21:28 ` [patch, 2.6.11-rc2] sched: RLIMIT_RT_CPU feature, -D7 Peter Williams
2005-01-26 21:44 ` Peter Williams
2005-01-26 9:20 ` [patch, 2.6.11-rc2] sched: RLIMIT_RT_CPU_RATIO feature Ingo Molnar
2005-01-31 23:03 ` Peter Williams
2005-02-01 10:11 ` [patch] sys_setpriority() euid semantics fix Ingo Molnar
2005-02-01 21:46 ` Peter Williams
2005-01-26 5:24 ` [patch, 2.6.11-rc2] sched: RLIMIT_RT_CPU_RATIO feature Jack O'Quin
2005-01-26 7:04 ` Ingo Molnar
2005-01-26 22:27 ` Jack O'Quin
2005-01-26 23:29 ` Nick Piggin
2005-01-27 2:31 ` Jack O'Quin
2005-01-27 3:26 ` Nick Piggin
2005-01-27 5:15 ` Jack O'Quin
2005-01-27 5:54 ` Nick Piggin
2005-01-27 8:35 ` Ingo Molnar
2005-01-27 8:59 ` Ingo Molnar
2005-01-27 11:35 ` Ingo Molnar
2005-02-02 5:10 ` Jack O'Quin
2005-02-02 11:10 ` Bill Huey
2005-02-02 16:44 ` Jack O'Quin
2005-02-02 21:14 ` Bill Huey
2005-02-02 21:20 ` Bill Huey
2005-02-02 21:21 ` Ingo Molnar
2005-02-02 21:34 ` Bill Huey
2005-02-02 22:59 ` Paul Davis
2005-02-03 2:46 ` Bill Huey
2005-02-03 14:20 ` Paul Davis
2005-02-03 20:19 ` Con Kolivas
2005-02-03 20:47 ` Ingo Molnar
2005-02-03 21:15 ` Paul Davis
2005-02-03 21:28 ` Ingo Molnar
2005-02-03 21:41 ` Paul Davis
2005-02-03 21:59 ` Ingo Molnar
2005-02-03 22:24 ` Paul Davis
2005-02-03 22:26 ` Ingo Molnar
2005-02-04 0:36 ` Tristan Wibberley
2005-02-03 21:48 ` Peter Williams
2005-02-04 16:41 ` Jack O'Quin
2005-02-04 21:38 ` Peter Williams
2005-02-03 21:41 ` Ingo Molnar
2005-02-03 23:01 ` Bill Huey
2005-02-11 21:27 ` Lee Revell
2005-02-02 21:54 ` Peter Williams
2005-02-02 23:03 ` Paul Davis
2005-02-02 23:46 ` Peter Williams
2005-02-03 1:13 ` Jack O'Quin
2005-02-03 3:10 ` Peter Williams
2005-02-03 3:56 ` Jack O'Quin
2005-02-03 21:36 ` Ingo Molnar
2005-02-04 0:35 ` Chris Wright
2005-02-04 17:21 ` Jack O'Quin
2005-02-03 2:54 ` Bill Huey
2005-02-03 3:25 ` Peter Williams
2005-02-02 11:37 ` Ingo Molnar
2005-02-02 16:01 ` Jack O'Quin
2005-02-02 18:59 ` Lee Revell
2005-02-02 19:31 ` Jack O'Quin
2005-02-02 20:29 ` Ingo Molnar
2005-02-02 22:45 ` Jack O'Quin
2005-02-02 20:17 ` Ingo Molnar
2005-01-27 20:01 ` Lee Revell
2005-01-28 6:38 ` Ingo Molnar
2005-01-28 8:09 ` Jack O'Quin
2005-01-28 8:08 ` Ingo Molnar
2005-01-28 8:35 ` Jack O'Quin
2005-01-28 8:40 ` Ingo Molnar
2005-01-28 9:01 ` Jack O'Quin
2005-01-28 9:11 ` Ingo Molnar
2005-01-29 0:44 ` Lee Revell
2005-01-28 9:51 ` Mike Galbraith
2005-01-28 22:16 ` Peter Williams
2005-01-28 22:19 ` Ingo Molnar
2005-01-29 7:02 ` Jack O'Quin
2005-01-31 22:29 ` Bill Davidsen
2005-02-01 0:39 ` Bill Huey
2005-01-25 5:16 ` [PATCH]sched: Isochronous class v2 for unprivileged soft rt scheduling Jack O'Quin
2005-01-25 15:09 ` Ingo Molnar
2005-01-23 20:48 ` Jack O'Quin
2005-01-23 22:57 ` Con Kolivas
2005-01-24 1:06 ` Jack O'Quin
2005-01-24 1:09 ` Con Kolivas
2005-01-24 4:45 ` Jack O'Quin
2005-01-24 4:53 ` Jack O'Quin
2005-01-24 6:28 ` Jack O'Quin
2005-01-24 6:35 ` Con Kolivas
2005-01-24 6:57 ` Jack O'Quin
2005-01-24 22:58 ` Con Kolivas
2005-01-25 3:55 ` Con Kolivas
2005-01-25 13:05 ` Con Kolivas
2005-01-25 14:38 ` Con Kolivas [this message]
2005-01-25 18:36 ` Jack O'Quin
2005-01-25 20:52 ` Rui Nuno Capela
2005-01-24 21:46 ` Con Kolivas
2005-01-23 7:38 ` Jack O'Quin
2005-01-23 7:41 ` Con Kolivas
2005-01-24 6:30 ` Jack O'Quin
2005-01-24 20:55 ` Ingo Molnar
2005-01-20 21:59 ` Peter Chubb
2005-01-21 0:30 ` Jack O'Quin
2005-01-22 14:06 ` Paul Davis
2005-01-20 17:49 ` ross
2005-01-20 19:07 ` Lee Revell
2005-01-20 23:17 ` Con Kolivas
2005-01-21 7:48 ` Ingo Molnar
2005-02-07 3:09 ` Werner Almesberger
2005-02-07 3:27 ` Jack O'Quin
2005-02-07 3:27 ` Con Kolivas
2005-01-20 9:06 ` Rui Nuno Capela
2005-01-20 17:09 ` Rui Nuno Capela
2005-01-20 19:32 ` Jack O'Quin
2005-01-21 9:18 ` Rui Nuno Capela
2005-01-21 16:23 ` Con Kolivas
2005-01-21 16:40 ` Jack O'Quin
2005-01-22 0:06 ` Con Kolivas
2005-01-22 6:18 ` Jack O'Quin
2005-01-22 6:19 ` Con Kolivas
2005-01-22 6:48 ` Con Kolivas
2005-01-22 6:50 ` Con Kolivas
2005-01-22 7:09 ` Con Kolivas
2005-01-22 20:22 ` Jack O'Quin
2005-01-23 1:02 ` Con Kolivas
2005-01-23 3:02 ` Jack O'Quin
2005-01-23 4:29 ` Con Kolivas
2005-01-23 4:46 ` Jack O'Quin
2005-01-23 4:50 ` Con Kolivas
2005-01-23 7:37 ` Mike Galbraith
2005-01-23 13:57 ` Paul Davis
2005-01-23 1:31 ` Con Kolivas
2005-01-23 1:41 ` Paul Davis
2005-01-23 1:56 ` Con Kolivas
2005-01-23 4:50 ` Jack O'Quin
2005-01-21 23:30 ` utz lehmann
2005-01-21 23:48 ` Con Kolivas
2005-01-22 0:28 ` utz lehmann
2005-01-22 3:52 ` Con Kolivas
2005-01-22 6:15 ` Jack O'Quin
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=41F659CD.9070904@kolivas.org \
--to=kernel@kolivas.org \
--cc=alexn@dsv.su.se \
--cc=ck@vds.kolivas.org \
--cc=joq@io.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=rncbc@rncbc.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).