From: Peter Zijlstra <peterz@infradead.org>
To: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Ingo Molnar <mingo@redhat.com>,
LKML <linux-kernel@vger.kernel.org>,
"Ma, Ling" <ling.ma@intel.com>,
"Zhang, Yanmin" <yanmin_zhang@linux.intel.com>,
"ego@in.ibm.com" <ego@in.ibm.com>,
"svaidy@linux.vnet.ibm.com" <svaidy@linux.vnet.ibm.com>
Subject: Re: change in sched cpu_power causing regressions with SCHED_MC
Date: Fri, 19 Feb 2010 20:47:55 +0100 [thread overview]
Message-ID: <1266608875.1529.749.camel@laptop> (raw)
In-Reply-To: <1266604594.2814.37.camel@sbs-t61.sc.intel.com>
On Fri, 2010-02-19 at 10:36 -0800, Suresh Siddha wrote:
> exec/fork balance is not broken. i.e., during exec/fork we balance the
> load equally among sockets/cores etc. What is broken is:
>
> a) In SMT case, once we end up in a situation where both the threads of
> the core are busy , with another core completely idle, load balance is
> not moving one of the threads to the idle core. This unbalanced
> situation can happen because of a previous wake-up decision and/or
> threads on other core went to sleep/died etc. Once we end up in this
> unbalanced situation, we continue in that state with out fixing it.
>
> b) Similar to "a", this is MC case where we end up four cores busy in
> one socket with other 4 cores in another socket completely idle. And
> this is the situation which we are trying to solve in this patch.
>
> In your above example, we test mostly fork/exec balance but not the
> above sleep/wakeup scenarios.
Ah, indeed. Let me extend my script to cover that.
The below script does indeed show a change, but the result still isn't
perfect, when I do ./show-loop 8, it starts 8 loops nicely spread over 2
sockets, the difference is that all 4 remaining would stay on socket 0,
the patched kernel gets 1 over to socket 1.
---
NR=$1; shift
cleanup()
{
killall loop
}
show_each_loop()
{
KILL=$1
ps -deo pid,sgi_p,cmd | grep loop | grep bash | while read pid cpu cmd; do
SOCKET=`cat /sys/devices/system/cpu/cpu${cpu}/topology/physical_package_id`
CORE=`cat /sys/devices/system/cpu/cpu${cpu}/topology/core_id`
SIBLINGS=`cat /sys/devices/system/cpu/cpu${cpu}/topology/thread_siblings_list`
printf "loop-%05d on CPU: %02d SOCKET: %02d CORE: %02d THREADS: ${SIBLINGS} " $pid $cpu $SOCKET $CORE
if [ $SOCKET -eq $KILL ]; then
kill $pid;
printf "(killed)"
fi
printf "\n"
done
}
trap cleanup SIGINT
echo "starting loops..."
for ((i=0; i<NR; i++)) ; do
./loop &
done
sleep 1;
echo "killing those on socket 1..."
echo ""
show_each_loop 1
echo ""
echo "watching load-balance work..."
echo ""
while sleep 1 ; do
show_each_loop -1 | sort | awk '{socket[$6]++; th[$8 + (256*$6)]++; print $0}
END { for (i in socket) { print "socket-" i ": " socket[i]; }
for (i in th) { if (th[i] > 1) { print "thread-" int(i/256)"/"(i%256) ": " th[i]; } } }'
echo ""
echo "-------------------"
echo ""
done
next prev parent reply other threads:[~2010-02-19 19:48 UTC|newest]
Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-13 1:14 [patch] sched: fix SMT scheduler regression in find_busiest_queue() Suresh Siddha
2010-02-13 1:31 ` change in sched cpu_power causing regressions with SCHED_MC Suresh Siddha
2010-02-13 10:36 ` Peter Zijlstra
2010-02-13 10:42 ` Peter Zijlstra
2010-02-13 18:37 ` Vaidyanathan Srinivasan
2010-02-13 18:49 ` Suresh Siddha
2010-02-13 18:39 ` Vaidyanathan Srinivasan
2010-02-19 2:16 ` Suresh Siddha
2010-02-19 12:32 ` Arun R Bharadwaj
2010-02-19 13:03 ` Vaidyanathan Srinivasan
2010-02-19 19:15 ` Suresh Siddha
2010-02-19 14:05 ` Peter Zijlstra
2010-02-19 18:36 ` Suresh Siddha
2010-02-19 19:47 ` Peter Zijlstra [this message]
2010-02-19 19:50 ` Suresh Siddha
2010-02-19 20:02 ` Peter Zijlstra
2010-02-20 1:13 ` Suresh Siddha
2010-02-22 18:50 ` Peter Zijlstra
2010-02-24 0:13 ` Suresh Siddha
2010-02-24 17:43 ` Peter Zijlstra
2010-02-24 19:31 ` Suresh Siddha
2010-02-26 10:24 ` [tip:sched/core] sched: Fix SCHED_MC regression caused by change in sched cpu_power tip-bot for Suresh Siddha
2010-02-26 14:55 ` tip-bot for Suresh Siddha
2010-02-19 19:52 ` change in sched cpu_power causing regressions with SCHED_MC Peter Zijlstra
2010-02-13 18:33 ` Vaidyanathan Srinivasan
2010-02-13 18:27 ` [patch] sched: fix SMT scheduler regression in find_busiest_queue() Vaidyanathan Srinivasan
2010-02-13 18:39 ` Suresh Siddha
2010-02-13 18:56 ` Vaidyanathan Srinivasan
2010-02-13 20:25 ` Vaidyanathan Srinivasan
2010-02-13 20:36 ` Vaidyanathan Srinivasan
2010-02-14 10:11 ` Peter Zijlstra
2010-02-15 12:35 ` Vaidyanathan Srinivasan
2010-02-15 13:00 ` Peter Zijlstra
2010-02-16 15:59 ` Vaidyanathan Srinivasan
2010-02-16 17:28 ` Peter Zijlstra
2010-02-16 18:25 ` Vaidyanathan Srinivasan
2010-02-16 18:46 ` Vaidyanathan Srinivasan
2010-02-16 18:48 ` Peter Zijlstra
2010-02-15 22:29 ` Peter Zijlstra
2010-02-16 14:16 ` [tip:sched/urgent] sched: Fix " tip-bot for Suresh Siddha
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=1266608875.1529.749.camel@laptop \
--to=peterz@infradead.org \
--cc=ego@in.ibm.com \
--cc=ling.ma@intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=suresh.b.siddha@intel.com \
--cc=svaidy@linux.vnet.ibm.com \
--cc=yanmin_zhang@linux.intel.com \
/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