From: Petr Vorel <pvorel@suse.cz>
To: Cyril Hrubis <chrubis@suse.cz>
Cc: joe.liu@mediatek.com, ltp@lists.linux.it
Subject: Re: [LTP] [PATCH v3] sched: starvation: Autocallibrate the timeout
Date: Tue, 30 Jul 2024 10:49:24 +0200 [thread overview]
Message-ID: <20240730084924.GC1327610@pevik> (raw)
In-Reply-To: <ZqiRaf0bPloFV3LI@yuki>
Hi,
> Hi!
> > tst_tmpdir.c:316: TINFO: Using /tmp/LTP_starv8seE as tmpdir (tmpfs filesystem)
> > tst_test.c:1806: TINFO: LTP version: 20240524
> > tst_test.c:1650: TINFO: Timeout per run is 0h 00m 30s
> > starvation.c:71: TINFO: Setting affinity to CPU 0
> > tst_test.c:1658: TINFO: Updating max runtime to 0h 04m 00s
> > tst_test.c:1650: TINFO: Timeout per run is 0h 04m 30s
> > starvation.c:117: TPASS: wait_for_pid(child_pid) passed
> > => test runs ~ 13s - 19s on aarch64, ppc64le and x86_64. Therefore not sure if
> > 04m max runtime is good.
OK, now testing with your patchset :)
https://github.com/pevik/ltp/commit/0392e5142c5e8cb1d38dec9916a9f0118cf27afe
All tests are on KVM.
* aarch64 6.10.1-1-default (8s)
tst_test.c:1650: TINFO: Timeout per run is 0h 00m 30s
starvation.c:52: TINFO: CPU did 120000000 loops in 42654us
tst_test.c:1658: TINFO: Updating max runtime to 0h 00m 42s
tst_test.c:1650: TINFO: Timeout per run is 0h 01m 12s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* ppc64le 6.10.1-1-default (9s)
...
tst_test.c:1650: TINFO: Timeout per run is 0h 00m 30s
starvation.c:52: TINFO: CPU did 120000000 loops in 98081us
tst_test.c:1658: TINFO: Updating max runtime to 0h 01m 38s
=> FYI HERE IS DIFFERENCE over 1 min
tst_test.c:1650: TINFO: Timeout per run is 0h 02m 08s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* x86_64 6.10.1-1-default (10s)
tst_test.c:1650: TINFO: Timeout per run is 0h 00m 30s
starvation.c:52: TINFO: CPU did 120000000 loops in 41908us
tst_test.c:1658: TINFO: Updating max runtime to 0h 00m 41s
tst_test.c:1650: TINFO: Timeout per run is 0h 01m 11s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
Some SLES
* s390x 6.4 (6s)
starvation.c:52: TINFO: CPU did 120000000 loops in 53599us
tst_test.c:1659: TINFO: Updating max runtime to 0h 00m 53s
tst_test.c:1652: TINFO: Timeout per run is 0h 01m 23s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* s390x 6.4 (6s)
starvation.c:52: TINFO: CPU did 120000000 loops in 53599us
tst_test.c:1659: TINFO: Updating max runtime to 0h 00m 53s
tst_test.c:1652: TINFO: Timeout per run is 0h 01m 23s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* x86_64 5.14.21 (5s)
starvation.c:52: TINFO: CPU did 120000000 loops in 75176us
tst_test.c:1659: TINFO: Updating max runtime to 0h 01m 15s
tst_test.c:1652: TINFO: Timeout per run is 0h 01m 45s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* aarch64 5.14.21 (9s)
starvation.c:52: TINFO: CPU did 120000000 loops in 42404us
tst_test.c:1659: TINFO: Updating max runtime to 0h 00m 42s
tst_test.c:1652: TINFO: Timeout per run is 0h 01m 12s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* ppc64le 5.14.21 (16s)
starvation.c:52: TINFO: CPU did 120000000 loops in 147223us
tst_test.c:1659: TINFO: Updating max runtime to 0h 02m 27s
tst_test.c:1652: TINFO: Timeout per run is 0h 02m 57s
=> FYI HERE IS DIFFERENCE 2 min
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* aarch64 5.3.18 (8s)
starvation.c:52: TINFO: CPU did 120000000 loops in 42382us
tst_test.c:1659: TINFO: Updating max runtime to 0h 00m 42s
tst_test.c:1652: TINFO: Timeout per run is 0h 01m 12s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* ppc64le 5.3.18 (50s)
starvation.c:52: TINFO: CPU did 120000000 loops in 140564us
tst_test.c:1659: TINFO: Updating max runtime to 0h 02m 20s
tst_test.c:1652: TINFO: Timeout per run is 0h 02m 50s
=> FYI HERE IS DIFFERENCE 2 min
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* aarch64 4.12.14 (12s)
starvation.c:52: TINFO: CPU did 120000000 loops in 55491us
tst_test.c:1659: TINFO: Updating max runtime to 0h 00m 55s
tst_test.c:1652: TINFO: Timeout per run is 0h 01m 25s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* ppc64le 4.12.14 (1m 9s FYI HIGHEST RUNTIME)
starvation.c:52: TINFO: CPU did 120000000 loops in 147102us
tst_test.c:1659: TINFO: Updating max runtime to 0h 02m 27s
tst_test.c:1652: TINFO: Timeout per run is 0h 02m 57s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* s390x 4.12.14 (5s)
starvation.c:52: TINFO: CPU did 120000000 loops in 49973us
tst_test.c:1659: TINFO: Updating max runtime to 0h 00m 49s
tst_test.c:1652: TINFO: Timeout per run is 0h 01m 19s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
* x86_64 4.4.180 (10s)
starvation.c:52: TINFO: CPU did 120000000 loops in 37042us
tst_test.c:1659: TINFO: Updating max runtime to 0h 00m 37s
tst_test.c:1652: TINFO: Timeout per run is 0h 01m 07s
starvation.c:150: TPASS: Haven't reproduced scheduller starvation.
Kind regards,
Petr
--
Mailing list info: https://lists.linux.it/listinfo/ltp
next prev parent reply other threads:[~2024-07-30 8:49 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-07-22 14:54 [LTP] [PATCH v3] sched: starvation: Autocallibrate the timeout Cyril Hrubis
2024-07-29 20:51 ` Petr Vorel
2024-07-30 7:08 ` Cyril Hrubis
2024-07-30 7:33 ` Petr Vorel
2024-07-30 8:49 ` Petr Vorel [this message]
2024-07-30 8:55 ` Cyril Hrubis
2024-08-01 12:15 ` Anders Roxell
2024-08-28 13:38 ` Cyril Hrubis
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=20240730084924.GC1327610@pevik \
--to=pvorel@suse.cz \
--cc=chrubis@suse.cz \
--cc=joe.liu@mediatek.com \
--cc=ltp@lists.linux.it \
/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.