From: Mike Galbraith <mgalbraith@suse.de>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
RT <linux-rt-users@vger.kernel.org>
Subject: Re: [RFC][PATCH 0/3] update to cpupri algorithm
Date: Sat, 30 Jul 2011 11:19:56 +0200 [thread overview]
Message-ID: <1312017596.8752.78.camel@marge.simson.net> (raw)
In-Reply-To: <20110729151340.419467771@goodmis.org>
On Fri, 2011-07-29 at 11:13 -0400, Steven Rostedt wrote:
> Hi Mike,
>
> Could you try this patch set out. Add the first patch and then
> run your tests. The first patch only adds benchmarking, and does not
> modify the scheduler algorithm.
>
> Do this:
>
> 1. apply first patch, build and boot
> 2. # mount -t debugfs nodev /sys/kernel/debug
> 3. # echo 0 > /sys/kernel/debug/cpupri; ./runtest; cat /sys/kernel/debug/cpupri > output
>
> The output will give you the contention of the vector locks in the
> cpupri algorithm.
>
> Then apply the second patch and do the same thing.
>
> Then apply the third patch and do the same thing.
>
> After that, could you send me the results of the output file for all
> three runs? The final patch should probably be the best overall
> results.
These patches are RFC, so here's my Comment. Steven rocks.
Below is what cyclictest has to say with the patches listed below
applied, and simulation model running on 56 isolated cores.
The model squeaks once in a while with jitter ever so slightly over the
30us squeak threshold, so I have yet more perturbation sources, but wrt
your patches, they are mana from heaven for largish boxen ;-)
patches/ipc-sem-Optimize-update_queue-for-bulk-wakeup-calls
patches/ipc-sem-Move-wake_up_process-out-of-the-spinlock-section
patches/ipc-sem-Cacheline-align-the-ipc-spinlock-for-semaphores
patches/ipc-sem-Update-description-of-the-implementation
patches/ipc-sem-Use-ERR_CAST
patches/ipc-sem-Bugfix-for-semop-not-reporting-successful-operation
patches/ipc-sys_semctl-Fix-kernel-stack-leakage
patches/cpupri-2-3-nolock.patch
patches/cpupri-3-3-atomic.patch
patches/posix-timers-RCU-conversion
cyclictest -Smqp 99 -D 30m; ./killit.sh
T: 0 ( 7362) P:99 I:1000 C:1799001 Min: 1 Act: 2 Avg: 1 Max: 19
T: 1 ( 7363) P:99 I:1500 C:1199333 Min: 1 Act: 2 Avg: 1 Max: 4
T: 2 ( 7364) P:99 I:2000 C: 899498 Min: 1 Act: 2 Avg: 1 Max: 4
T: 3 ( 7365) P:99 I:2500 C: 719597 Min: 1 Act: 2 Avg: 1 Max: 4
T: 4 ( 7366) P:99 I:3000 C: 599664 Min: 1 Act: 2 Avg: 1 Max: 4
T: 5 ( 7367) P:99 I:3500 C: 513997 Min: 1 Act: 2 Avg: 1 Max: 5
T: 6 ( 7368) P:99 I:4000 C: 449746 Min: 1 Act: 2 Avg: 1 Max: 4
T: 7 ( 7369) P:99 I:4500 C: 399774 Min: 1 Act: 2 Avg: 1 Max: 4
T: 8 ( 7370) P:99 I:5000 C: 359796 Min: 1 Act: 3 Avg: 2 Max: 5
T: 9 ( 7371) P:99 I:5500 C: 327087 Min: 1 Act: 3 Avg: 2 Max: 7
T:10 ( 7372) P:99 I:6000 C: 299829 Min: 1 Act: 3 Avg: 2 Max: 7
T:11 ( 7373) P:99 I:6500 C: 276765 Min: 1 Act: 3 Avg: 2 Max: 13
T:12 ( 7374) P:99 I:7000 C: 256995 Min: 1 Act: 3 Avg: 2 Max: 6
T:13 ( 7375) P:99 I:7500 C: 239862 Min: 1 Act: 3 Avg: 2 Max: 10
T:14 ( 7376) P:99 I:8000 C: 224870 Min: 1 Act: 2 Avg: 2 Max: 6
T:15 ( 7377) P:99 I:8500 C: 211642 Min: 1 Act: 4 Avg: 2 Max: 7
T:16 ( 7378) P:99 I:9000 C: 199884 Min: 1 Act: 3 Avg: 2 Max: 13
T:17 ( 7379) P:99 I:9500 C: 189364 Min: 1 Act: 4 Avg: 2 Max: 6
T:18 ( 7380) P:99 I:10000 C: 179895 Min: 1 Act: 3 Avg: 2 Max: 5
T:19 ( 7381) P:99 I:10500 C: 171328 Min: 1 Act: 3 Avg: 2 Max: 9
T:20 ( 7382) P:99 I:11000 C: 163540 Min: 1 Act: 4 Avg: 2 Max: 8
T:21 ( 7383) P:99 I:11500 C: 156430 Min: 2 Act: 3 Avg: 2 Max: 7
T:22 ( 7384) P:99 I:12000 C: 149912 Min: 2 Act: 3 Avg: 2 Max: 10
T:23 ( 7385) P:99 I:12500 C: 143915 Min: 2 Act: 3 Avg: 3 Max: 8
T:24 ( 7386) P:99 I:13000 C: 138380 Min: 2 Act: 4 Avg: 3 Max: 7
T:25 ( 7387) P:99 I:13500 C: 133254 Min: 2 Act: 2 Avg: 3 Max: 7
T:26 ( 7388) P:99 I:14000 C: 128495 Min: 1 Act: 3 Avg: 3 Max: 6
T:27 ( 7389) P:99 I:14500 C: 124064 Min: 1 Act: 4 Avg: 3 Max: 10
T:28 ( 7390) P:99 I:15000 C: 119928 Min: 2 Act: 4 Avg: 3 Max: 7
T:29 ( 7391) P:99 I:15500 C: 116059 Min: 2 Act: 3 Avg: 2 Max: 6
T:30 ( 7392) P:99 I:16000 C: 112432 Min: 2 Act: 4 Avg: 3 Max: 6
T:31 ( 7393) P:99 I:16500 C: 109025 Min: 1 Act: 3 Avg: 3 Max: 6
T:32 ( 7394) P:99 I:17000 C: 105818 Min: 2 Act: 4 Avg: 3 Max: 7
T:33 ( 7395) P:99 I:17500 C: 102795 Min: 2 Act: 4 Avg: 3 Max: 7
T:34 ( 7396) P:99 I:18000 C: 99939 Min: 1 Act: 3 Avg: 3 Max: 7
T:35 ( 7397) P:99 I:18500 C: 97238 Min: 2 Act: 4 Avg: 3 Max: 8
T:36 ( 7398) P:99 I:19000 C: 94679 Min: 2 Act: 3 Avg: 3 Max: 7
T:37 ( 7399) P:99 I:19500 C: 92251 Min: 1 Act: 3 Avg: 3 Max: 8
T:38 ( 7400) P:99 I:20000 C: 89945 Min: 2 Act: 4 Avg: 3 Max: 8
T:39 ( 7401) P:99 I:20500 C: 87751 Min: 2 Act: 4 Avg: 3 Max: 13
T:40 ( 7402) P:99 I:21000 C: 85661 Min: 2 Act: 3 Avg: 3 Max: 10
T:41 ( 7403) P:99 I:21500 C: 83669 Min: 1 Act: 3 Avg: 3 Max: 7
T:42 ( 7404) P:99 I:22000 C: 81767 Min: 2 Act: 3 Avg: 3 Max: 10
T:43 ( 7405) P:99 I:22500 C: 79950 Min: 2 Act: 4 Avg: 3 Max: 9
T:44 ( 7406) P:99 I:23000 C: 78212 Min: 1 Act: 2 Avg: 3 Max: 8
T:45 ( 7407) P:99 I:23500 C: 76548 Min: 1 Act: 3 Avg: 3 Max: 7
T:46 ( 7408) P:99 I:24000 C: 74953 Min: 1 Act: 3 Avg: 3 Max: 6
T:47 ( 7409) P:99 I:24500 C: 73423 Min: 1 Act: 3 Avg: 3 Max: 6
T:48 ( 7410) P:99 I:25000 C: 71955 Min: 2 Act: 4 Avg: 3 Max: 7
T:49 ( 7411) P:99 I:25500 C: 70544 Min: 1 Act: 3 Avg: 3 Max: 6
T:50 ( 7412) P:99 I:26000 C: 69187 Min: 2 Act: 4 Avg: 3 Max: 11
T:51 ( 7413) P:99 I:26500 C: 67881 Min: 1 Act: 3 Avg: 3 Max: 6
T:52 ( 7414) P:99 I:27000 C: 66624 Min: 2 Act: 3 Avg: 3 Max: 7
T:53 ( 7415) P:99 I:27500 C: 65413 Min: 1 Act: 4 Avg: 3 Max: 6
T:54 ( 7416) P:99 I:28000 C: 64245 Min: 2 Act: 4 Avg: 3 Max: 6
T:55 ( 7417) P:99 I:28500 C: 63118 Min: 2 Act: 3 Avg: 3 Max: 6
T:56 ( 7418) P:99 I:29000 C: 62029 Min: 2 Act: 3 Avg: 3 Max: 8
T:57 ( 7419) P:99 I:29500 C: 60978 Min: 2 Act: 3 Avg: 2 Max: 5
T:58 ( 7420) P:99 I:30000 C: 59961 Min: 2 Act: 3 Avg: 2 Max: 5
T:59 ( 7421) P:99 I:30500 C: 58978 Min: 2 Act: 3 Avg: 2 Max: 5
T:60 ( 7422) P:99 I:31000 C: 58027 Min: 2 Act: 3 Avg: 3 Max: 10
T:61 ( 7423) P:99 I:31500 C: 57106 Min: 2 Act: 3 Avg: 3 Max: 9
T:62 ( 7424) P:99 I:32000 C: 56213 Min: 2 Act: 2 Avg: 2 Max: 5
T:63 ( 7425) P:99 I:32500 C: 55349 Min: 2 Act: 5 Avg: 3 Max: 6
next prev parent reply other threads:[~2011-07-30 9:20 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-07-29 15:13 [RFC][PATCH 0/3] update to cpupri algorithm Steven Rostedt
2011-07-29 15:13 ` [RFC][PATCH 1/3] cpupri: Add profiling Steven Rostedt
2011-07-29 15:13 ` [RFC][PATCH 2/3] cpupri: Remove vector locks and read entire loop Steven Rostedt
2011-07-29 15:13 ` [RFC][PATCH 3/3] cpupri: Add atomic vector count to speed up loop Steven Rostedt
2011-07-29 15:38 ` [RFC][PATCH 0/3] update to cpupri algorithm Mike Galbraith
2011-07-29 18:24 ` Mike Galbraith
2011-07-30 7:12 ` Mike Galbraith
2011-07-30 8:16 ` Mike Galbraith
2011-07-30 9:19 ` Mike Galbraith [this message]
2011-08-01 14:18 ` Steven Rostedt
2011-08-01 18:41 ` Mike Galbraith
2011-08-01 18:54 ` Steven Rostedt
2011-08-01 19:06 ` Mike Galbraith
2011-08-02 8:59 ` Mike Galbraith
2011-08-02 8:46 ` Mike Galbraith
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=1312017596.8752.78.camel@marge.simson.net \
--to=mgalbraith@suse.de \
--cc=linux-rt-users@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.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 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.