From: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>
To: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>,
Suresh Siddha <suresh.b.siddha@intel.com>,
linux-kernel <linux-kernel@vger.kernel.org>,
Ingo Molnar <mingo@elte.hu>, Paul Turner <pjt@google.com>
Subject: Re: sched: Avoid SMT siblings in select_idle_sibling() if possible
Date: Sat, 25 Feb 2012 12:24:03 +0530 [thread overview]
Message-ID: <20120225065403.GB12313@linux.vnet.ibm.com> (raw)
In-Reply-To: <1329996064.7411.106.camel@marge.simpson.net>
* Mike Galbraith <efault@gmx.de> [2012-02-23 12:21:04]:
> Unpinned netperf TCP_RR and/or tbench pairs? Anything that's wakeup
> heavy should tell the tail.
Here are some tbench numbers:
Machine : 2 Intel Xeon X5650 (Westmere) CPUs (6 core/package)
Kernel : tip (HEAD at ebe97fa)
dbench : v4.0
One tbench server/client pair was run on same host 5 times (with
fs-cache being purged each time) and avg of 5 run for various cases
noted below:
Case A : HT enabled (24 logical CPUs)
Thr'put : 168.166 MB/s (SD_SHARE_PKG_RESOURCES + !SD_BALANCE_WAKE)
Thr'put : 169.564 MB/s (SD_SHARE_PKG_RESOURCES + SD_BALANCE_WAKE at mc/smt)
Thr'put : 173.151 MB/s (!SD_SHARE_PKG_RESOURCES + !SD_BALANCE_WAKE)
Case B : HT disabled (12 logical CPUs)
Thr'put : 167.977 MB/s (SD_SHARE_PKG_RESOURCES + !SD_BALANCE_WAKE)
Thr'put : 167.891 MB/s (SD_SHARE_PKG_RESOURCES + SD_BALANCE_WAKE at mc)
Thr'put : 173.801 MB/s (!SD_SHARE_PKG_RESOURCES + !SD_BALANCE_WAKE)
Observations:
a. ~3% improvement seen with SD_SHARE_PKG_RESOURCES disabled, which I guess
reflects the cost of waking to a cold L2 cache.
b. No degradation seen with SD_BALANCE_WAKE enabled at mc/smt domains
IMO we need to detect tbench type paired wakeups as synchronous case, in
which case blindly wakeup the task to cur_cpu (as cost of L2 cache miss
could outweight the cost of any reduced scheduling latencies).
IOW select_task_rq_fair() needs to be given better hint as to whether L2
cache has been made warm by someone (interrupt handler or a producer
task), in which case (consumer) task needs to be woken in the same L2 cache
domain (i.e on cur_cpu itself)?
- vatsa
next prev parent reply other threads:[~2012-02-25 6:54 UTC|newest]
Thread overview: 89+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-15 9:46 sched: Avoid SMT siblings in select_idle_sibling() if possible Peter Zijlstra
2011-11-16 1:14 ` Suresh Siddha
2011-11-16 9:24 ` Mike Galbraith
2011-11-16 18:37 ` Suresh Siddha
2011-11-17 1:59 ` Mike Galbraith
2011-11-17 15:38 ` Mike Galbraith
2011-11-17 15:56 ` Peter Zijlstra
2011-11-17 16:38 ` Mike Galbraith
2011-11-17 17:36 ` Suresh Siddha
2011-11-18 15:14 ` Mike Galbraith
2011-11-18 15:17 ` [patch 1/6] sched: use rt.nr_cpus_allowed to recover select_task_rq() cycles Mike Galbraith
2011-11-18 15:35 ` Peter Zijlstra
2011-11-18 17:34 ` Mike Galbraith
2011-11-22 14:17 ` Mike Galbraith
2011-11-22 14:18 ` [patch 1/7] " Mike Galbraith
2011-12-06 9:50 ` [tip:sched/core] sched: Use " tip-bot for Mike Galbraith
2011-11-22 14:20 ` [patch 2/7] sched: save some hrtick_start_fair cycles Mike Galbraith
2011-12-06 20:20 ` [tip:sched/core] sched: Save " tip-bot for Mike Galbraith
2011-11-22 14:21 ` [patch 3/7] sched: set skip_clock_update in yield_task_fair() Mike Galbraith
2011-11-23 11:53 ` Peter Zijlstra
2011-11-23 12:06 ` Mike Galbraith
2011-11-23 14:48 ` Peter Zijlstra
2011-11-24 3:50 ` Mike Galbraith
2011-11-24 10:12 ` Peter Zijlstra
2011-11-25 6:39 ` Mike Galbraith
2011-12-06 9:51 ` [tip:sched/core] sched: Set " tip-bot for Mike Galbraith
2011-11-22 14:22 ` [patch 4/7] sched: convert rq->avg_idle to rq->avg_event Mike Galbraith
2011-11-23 11:55 ` Peter Zijlstra
2011-11-23 12:09 ` Mike Galbraith
2011-11-23 12:27 ` Peter Zijlstra
2011-11-23 12:57 ` Mike Galbraith
2011-11-23 14:21 ` Mike Galbraith
2011-11-22 14:23 ` [patch 5/7] sched: ratelimit select_idle_sibling()for sync wakeups Mike Galbraith
2011-11-22 14:24 ` [patch 6/7] sched: use rq->avg_event to resurrect nohz ratelimiting Mike Galbraith
2011-11-23 11:57 ` Peter Zijlstra
2011-11-23 12:35 ` Mike Galbraith
2011-11-22 14:26 ` [patch 7/7] sched: only use TTWU_QUEUE when waker/wakee CPUs do not share top level cache Mike Galbraith
2011-11-23 12:08 ` Peter Zijlstra
2011-11-18 15:39 ` [patch 1/6] sched: use rt.nr_cpus_allowed to recover select_task_rq() cycles Hillf Danton
2011-11-18 15:18 ` [patch 2/6] sched: convert rq->avg_idle to rq->avg_event Mike Galbraith
2011-11-18 15:19 ` [patch 3/6] sched: use rq->avg_event to resurrect nohz ratelimiting Mike Galbraith
2011-11-18 15:36 ` Peter Zijlstra
2011-11-18 17:42 ` Mike Galbraith
2011-11-19 0:51 ` Van De Ven, Arjan
2011-11-19 4:15 ` Mike Galbraith
2011-11-18 15:20 ` [patch 4/6] sched: ratelimit select_idle_sibling()for sync wakeups Mike Galbraith
2011-11-18 15:22 ` [patch 5/6] sched: save some hrtick_start_fair cycles Mike Galbraith
2011-11-18 15:23 ` [patch 6/6] sched: set skip_clock_update in yield_task_fair() Mike Galbraith
2012-02-20 14:41 ` sched: Avoid SMT siblings in select_idle_sibling() if possible Peter Zijlstra
2012-02-20 15:03 ` Srivatsa Vaddagiri
2012-02-20 18:25 ` Mike Galbraith
2012-02-21 0:06 ` Srivatsa Vaddagiri
2012-02-21 6:37 ` Mike Galbraith
2012-02-21 8:09 ` Srivatsa Vaddagiri
2012-02-20 18:14 ` Mike Galbraith
2012-02-20 18:15 ` Peter Zijlstra
2012-02-20 19:07 ` Peter Zijlstra
2012-02-21 5:43 ` Mike Galbraith
2012-02-21 8:32 ` Srivatsa Vaddagiri
2012-02-21 9:21 ` Mike Galbraith
2012-02-21 10:37 ` Peter Zijlstra
2012-02-21 14:58 ` Srivatsa Vaddagiri
2012-02-23 10:49 ` Srivatsa Vaddagiri
2012-02-23 11:19 ` Ingo Molnar
2012-02-23 12:18 ` Srivatsa Vaddagiri
2012-02-23 11:20 ` Srivatsa Vaddagiri
2012-02-23 11:26 ` Ingo Molnar
2012-02-23 11:32 ` Srivatsa Vaddagiri
2012-02-23 16:17 ` Ingo Molnar
2012-02-23 11:21 ` Mike Galbraith
2012-02-25 6:54 ` Srivatsa Vaddagiri [this message]
2012-02-25 8:30 ` Mike Galbraith
2012-02-27 22:11 ` Suresh Siddha
2012-02-28 5:05 ` Mike Galbraith
2011-11-17 19:08 ` Suresh Siddha
2011-11-18 15:12 ` Peter Zijlstra
2011-11-18 15:26 ` Mike Galbraith
2011-12-06 9:49 ` [tip:sched/core] sched: Clean up domain traversal in select_idle_sibling() tip-bot for Suresh Siddha
2011-11-18 23:40 ` [tip:sched/core] sched: Avoid SMT siblings in select_idle_sibling() if possible tip-bot for Peter Zijlstra
[not found] <1329764866.2293.376.camhel@twins>
2012-03-05 15:24 ` Srivatsa Vaddagiri
2012-03-06 9:14 ` Ingo Molnar
2012-03-06 10:03 ` Srivatsa Vaddagiri
2012-03-22 15:32 ` Srivatsa Vaddagiri
2012-03-23 6:38 ` Mike Galbraith
2012-03-26 8:29 ` Peter Zijlstra
2012-03-26 8:36 ` Peter Zijlstra
2012-03-26 17:35 ` Srivatsa Vaddagiri
2012-03-26 18:06 ` Peter Zijlstra
2012-03-27 13:56 ` 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=20120225065403.GB12313@linux.vnet.ibm.com \
--to=vatsa@linux.vnet.ibm.com \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=suresh.b.siddha@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 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.