From: Anton Blanchard <anton@samba.org>
To: Nicholas Berry <nikberry@med.umich.edu>
Cc: grendel@caudium.net, willy@w.ods.org, linux-kernel@vger.kernel.org
Subject: Re: Very high load on P4 machines with 2.4.28
Date: Wed, 5 Jan 2005 18:12:24 +1100 [thread overview]
Message-ID: <20050105071224.GL7335@krispykreme.ozlabs.ibm.com> (raw)
In-Reply-To: <s1dad55b.011@med-gwia-02a.med.umich.edu>
> Indeed. AIX (sorry) 5.3 on POWER5 explicitly disables SMT (IBM
> hyperthreading) if the load doesn't warrant it.
>
> (Now how about that for Linux?) :)
Its all there in ppc64 2.6 mainline:
for i in `seq 0 127`
do
echo XXX > /sys/devices/system/cpu/cpu$i/smt_snooze_delay
done
Will enable dynamic ST-SMT switching (XXX is the number of microseconds
we wait in the idle loop before sleeping the thread). Since the SMT
scheduler in 2.6 fills up primary threads first, we will keep SMT
disabled until we have enough work to do. This is probably what you are
referring to here.
To switch SMT off permanently on the fly:
for i in `seq 0 2 127`
do
echo 0 > /sys/devices_system/cpu/cpu$i/online
done
(ie hotplug cpu disable every second thread) and to switch SMT on again:
for i in `seq 0 2 127`
do
echo 1 > /sys/devices_system/cpu/cpu$i/online
done
Switching off SMT using either mode will give you a gain in performance
on some things, particularly single threaded stuff. The POWER5 chip
actually reconfigures itself on the fly and reallocates all the
resources to the one thread. eg on a single CPU, two thread box:
SMT enabled:
# ./lat_syscall -N 1
Simple syscall: 0.3360 microseconds
now disable SMT:
# echo 0 > /sys/devices/system/cpu/cpu1/online
# ./lat_syscall -N 1 null
Simple syscall: 0.2970 microseconds
Anton
next prev parent reply other threads:[~2005-01-05 7:13 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-01-04 22:41 Nicholas Berry
2005-01-04 23:05 ` Jesper Juhl
2005-01-05 0:02 ` Alan Cox
2005-01-05 1:21 ` Con Kolivas
2005-01-05 7:12 ` Anton Blanchard [this message]
-- strict thread matches above, loose matches on Subject: below --
2005-01-05 11:09 Indrek Kruusa
2005-01-05 14:03 ` Alan Cox
2005-01-04 19:58 Marek Habersack
2005-01-04 19:56 Marek Habersack
2005-01-04 22:03 ` Willy Tarreau
2005-01-04 23:07 ` Marek Habersack
2005-01-05 5:28 ` Willy Tarreau
2005-01-05 11:32 ` Marek Habersack
2005-01-04 22:05 ` Willy Tarreau
2005-01-04 23:09 ` Marek Habersack
2005-01-05 9:42 ` Marcelo Tosatti
2005-01-05 17:49 ` Marek Habersack
2005-01-06 18:56 ` Denis Vlasenko
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=20050105071224.GL7335@krispykreme.ozlabs.ibm.com \
--to=anton@samba.org \
--cc=grendel@caudium.net \
--cc=linux-kernel@vger.kernel.org \
--cc=nikberry@med.umich.edu \
--cc=willy@w.ods.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
all inboxes | Powered by JetHome®