From: Steven Sistare <steven.sistare@oracle.com>
To: Vincent Guittot <vincent.guittot@linaro.org>
Cc: Ingo Molnar <mingo@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
subhra.mazumdar@oracle.com,
Dhaval Giani <dhaval.giani@oracle.com>,
daniel.m.jordan@oracle.com, pavel.tatashin@microsoft.com,
Matt Fleming <matt@codeblueprint.co.uk>,
Mike Galbraith <umgwanakikbuti@gmail.com>,
Rik van Riel <riel@redhat.com>, Josef Bacik <jbacik@fb.com>,
Juri Lelli <juri.lelli@redhat.com>,
Valentin Schneider <valentin.schneider@arm.com>,
Quentin Perret <quentin.perret@arm.com>,
linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v4 00/10] steal tasks to improve CPU utilization
Date: Mon, 10 Dec 2018 11:29:08 -0500 [thread overview]
Message-ID: <df0197ac-d4eb-437e-1176-dbc9b23a94bb@oracle.com> (raw)
In-Reply-To: <CAKfTPtBuAsbF2r8wSigM=ckpKCqj0hcHbi+y3ZAaC7dqYcFY2g@mail.gmail.com>
On 12/10/2018 11:10 AM, Vincent Guittot wrote:
> Hi Steven,
>
> On Thu, 6 Dec 2018 at 22:38, Steve Sistare <steven.sistare@oracle.com> wrote:
>>
>> When a CPU has no more CFS tasks to run, and idle_balance() fails to
>> find a task, then attempt to steal a task from an overloaded CPU in the
>> same LLC. Maintain and use a bitmap of overloaded CPUs to efficiently
>> identify candidates. To minimize search time, steal the first migratable
>> task that is found when the bitmap is traversed. For fairness, search
>> for migratable tasks on an overloaded CPU in order of next to run.
>>
>> This simple stealing yields a higher CPU utilization than idle_balance()
>> alone, because the search is cheap, so it may be called every time the CPU
>> is about to go idle. idle_balance() does more work because it searches
>> widely for the busiest queue, so to limit its CPU consumption, it declines
>> to search if the system is too busy. Simple stealing does not offload the
>> globally busiest queue, but it is much better than running nothing at all.
>>
>> The bitmap of overloaded CPUs is a new type of sparse bitmap, designed to
>> reduce cache contention vs the usual bitmap when many threads concurrently
>> set, clear, and visit elements.
>>
>> Patch 1 defines the sparsemask type and its operations.
>>
>> Patches 2, 3, and 4 implement the bitmap of overloaded CPUs.
>>
>> Patches 5 and 6 refactor existing code for a cleaner merge of later
>> patches.
>>
>> Patches 7 and 8 implement task stealing using the overloaded CPUs bitmap.
>>
>> Patch 9 disables stealing on systems with more than 2 NUMA nodes for the
>> time being because of performance regressions that are not due to stealing
>> per-se. See the patch description for details.
>>
>> Patch 10 adds schedstats for comparing the new behavior to the old, and
>> provided as a convenience for developers only, not for integration.
>>
>> The patch series is based on kernel 4.20.0-rc1. It compiles, boots, and
>> runs with/without each of CONFIG_SCHED_SMT, CONFIG_SMP, CONFIG_SCHED_DEBUG,
>> and CONFIG_PREEMPT. It runs without error with CONFIG_DEBUG_PREEMPT +
>> CONFIG_SLUB_DEBUG + CONFIG_DEBUG_PAGEALLOC + CONFIG_DEBUG_MUTEXES +
>> CONFIG_DEBUG_SPINLOCK + CONFIG_DEBUG_ATOMIC_SLEEP. CPU hot plug and CPU
>> bandwidth control were tested.
>>
>> Stealing improves utilization with only a modest CPU overhead in scheduler
>> code. In the following experiment, hackbench is run with varying numbers
>> of groups (40 tasks per group), and the delta in /proc/schedstat is shown
>> for each run, averaged per CPU, augmented with these non-standard stats:
>>
>> %find - percent of time spent in old and new functions that search for
>> idle CPUs and tasks to steal and set the overloaded CPUs bitmap.
>>
>> steal - number of times a task is stolen from another CPU.
>>
>> X6-2: 1 socket * 10 cores * 2 hyperthreads = 20 CPUs
>> Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz
>> hackbench <grps> process 100000
>> sched_wakeup_granularity_ns=15000000
>>
>> baseline
>> grps time %busy slice sched idle wake %find steal
>> 1 8.084 75.02 0.10 105476 46291 59183 0.31 0
>> 2 13.892 85.33 0.10 190225 70958 119264 0.45 0
>> 3 19.668 89.04 0.10 263896 87047 176850 0.49 0
>> 4 25.279 91.28 0.10 322171 94691 227474 0.51 0
>> 8 47.832 94.86 0.09 630636 144141 486322 0.56 0
>>
>> new
>> grps time %busy slice sched idle wake %find steal %speedup
>> 1 5.938 96.80 0.24 31255 7190 24061 0.63 7433 36.1
>> 2 11.491 99.23 0.16 74097 4578 69512 0.84 19463 20.9
>> 3 16.987 99.66 0.15 115824 1985 113826 0.77 24707 15.8
>> 4 22.504 99.80 0.14 167188 2385 164786 0.75 29353 12.3
>> 8 44.441 99.86 0.11 389153 1616 387401 0.67 38190 7.6
>>
>> Elapsed time improves by 8 to 36%, and CPU busy utilization is up
>> by 5 to 22% hitting 99% for 2 or more groups (80 or more tasks).
>> The cost is at most 0.4% more find time.
>
> I have run some hackbench tests on my hikey arm64 octo cores with your
> patchset. My original intent was to send a tested-by but I have some
> performances regressions.
> This hikey is the smp one and not the asymetric hikey960 that Valentin
> used for his tests
> The sched domain topology is
> domain-0: span=0-3 level=MC and domain-0: span=4-7 level=MC
> domain-1: span=0-7 level=DIE
>
> I have run 12 times hackbench -g $j -P -l 2000 with j equals to 1 2 3 4 8
>
> grps time
> 1 1.396
> 2 2.699
> 3 3.617
> 4 4.498
> 8 7.721
>
> Then after disabling STEAL in sched_feature with echo NO_STEAL >
> /sys/kernel/debug/sched_features , the results become:
> grps time
> 1 1.217
> 2 1.973
> 3 2.855
> 4 3.932
> 8 7.674
>
> I haven't looked in details about some possible reasons of such
> difference yet and haven't collected the stats that you added with
> patch 10.
> Have you got a script to collect and post process them ?
>
> Regards,
> Vincent
Thanks Vincent. What is the value of /proc/sys/kernel/sched_wakeup_granularity_ns?
Try 15000000. Your 8-core system is heavily overloaded with 40 * groups tasks,
and I suspect preemptions are killing performance.
I have a python script to post-process schedstat files, but it does many things
and is large and I am not ready to share it. I can write a short bash script if
that would help.
- Steve
next prev parent reply other threads:[~2018-12-10 16:29 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-06 21:28 [PATCH v4 00/10] steal tasks to improve CPU utilization Steve Sistare
2018-12-06 21:28 ` [PATCH v4 01/10] sched: Provide sparsemask, a reduced contention bitmap Steve Sistare
2019-01-31 19:18 ` Tim Chen
2018-12-06 21:28 ` [PATCH v4 02/10] sched/topology: Provide hooks to allocate data shared per LLC Steve Sistare
2018-12-06 21:28 ` [PATCH v4 03/10] sched/topology: Provide cfs_overload_cpus bitmap Steve Sistare
2018-12-07 20:20 ` Valentin Schneider
2018-12-07 22:35 ` Steven Sistare
2018-12-08 18:33 ` Valentin Schneider
2018-12-06 21:28 ` [PATCH v4 04/10] sched/fair: Dynamically update cfs_overload_cpus Steve Sistare
2018-12-07 20:20 ` Valentin Schneider
2018-12-07 22:35 ` Steven Sistare
2018-12-08 18:47 ` Valentin Schneider
2018-12-06 21:28 ` [PATCH v4 05/10] sched/fair: Hoist idle_stamp up from idle_balance Steve Sistare
2018-12-06 21:28 ` [PATCH v4 06/10] sched/fair: Generalize the detach_task interface Steve Sistare
2018-12-06 21:28 ` [PATCH v4 07/10] sched/fair: Provide can_migrate_task_llc Steve Sistare
2018-12-06 21:28 ` [PATCH v4 08/10] sched/fair: Steal work from an overloaded CPU when CPU goes idle Steve Sistare
2018-12-07 20:21 ` Valentin Schneider
2018-12-07 22:36 ` Steven Sistare
2018-12-08 18:39 ` Valentin Schneider
2018-12-06 21:28 ` [PATCH v4 09/10] sched/fair: disable stealing if too many NUMA nodes Steve Sistare
2018-12-07 11:43 ` Valentin Schneider
2018-12-07 13:37 ` Steven Sistare
2018-12-06 21:28 ` [PATCH v4 10/10] sched/fair: Provide idle search schedstats Steve Sistare
2018-12-07 11:56 ` Valentin Schneider
2018-12-07 13:45 ` Steven Sistare
2018-12-24 12:25 ` Rick Lindsley
2019-01-14 17:04 ` Steven Sistare
2018-12-07 20:30 ` [PATCH v4 00/10] steal tasks to improve CPU utilization Valentin Schneider
2018-12-07 22:36 ` Steven Sistare
2019-02-01 15:07 ` Valentin Schneider
2018-12-10 16:10 ` Vincent Guittot
2018-12-10 16:29 ` Steven Sistare [this message]
2018-12-10 16:33 ` Vincent Guittot
2018-12-10 17:08 ` Vincent Guittot
2018-12-10 17:20 ` Steven Sistare
2018-12-10 17:06 ` Valentin Schneider
2019-01-04 13:44 ` Shijith Thotton
2019-01-14 16:55 ` Steven Sistare
2019-01-31 17:16 ` Dhaval Giani
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=df0197ac-d4eb-437e-1176-dbc9b23a94bb@oracle.com \
--to=steven.sistare@oracle.com \
--cc=daniel.m.jordan@oracle.com \
--cc=dhaval.giani@oracle.com \
--cc=jbacik@fb.com \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=matt@codeblueprint.co.uk \
--cc=mingo@redhat.com \
--cc=pavel.tatashin@microsoft.com \
--cc=peterz@infradead.org \
--cc=quentin.perret@arm.com \
--cc=riel@redhat.com \
--cc=subhra.mazumdar@oracle.com \
--cc=umgwanakikbuti@gmail.com \
--cc=valentin.schneider@arm.com \
--cc=vincent.guittot@linaro.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;
as well as URLs for NNTP newsgroup(s).