From mboxrd@z Thu Jan 1 00:00:00 1970 From: Vincent Guittot Date: Wed, 21 Oct 2020 12:25:32 +0000 Subject: Re: [PATCH] sched/fair: check for idle core Message-Id: <20201021122532.GA30733@vingu-book> List-Id: References: <1603211879-1064-1-git-send-email-Julia.Lawall@inria.fr> <20201021112038.GC32041@suse.de> In-Reply-To: <20201021112038.GC32041@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: Mel Gorman Cc: Julia Lawall , Ingo Molnar , kernel-janitors@vger.kernel.org, Peter Zijlstra , Juri Lelli , Dietmar Eggemann , Steven Rostedt , Ben Segall , Daniel Bristot de Oliveira , linux-kernel@vger.kernel.org, Valentin Schneider , Gilles.Muller@inria.fr Le mercredi 21 oct. 2020 =E0 12:20:38 (+0100), Mel Gorman a =E9crit : > On Tue, Oct 20, 2020 at 06:37:59PM +0200, Julia Lawall wrote: > > On a thread wakeup, the change [1] from runnable load average to load > > average for comparing candidate cores means that recent short-running > > daemons on the core where a thread ran previously can be considered to > > have a higher load than the core performing the wakeup, even when the > > core where the thread ran previously is currently idle. This can > > cause a thread to migrate, taking the place of some other thread that > > is about to wake up, and so on. To avoid unnecessary migrations, > > extend wake_affine_idle to check whether the core where the thread > > previously ran is currently idle, and if so return that core as the > > target. > >=20 > > [1] commit 11f10e5420f6ce ("sched/fair: Use load instead of runnable > > load in wakeup path") > >=20 > > This particularly has an impact when using passive (intel_cpufreq) > > power management, where kworkers run every 0.004 seconds on all cores, > > increasing the likelihood that an idle core will be considered to have > > a load. > >=20 > > The following numbers were obtained with the benchmarking tool > > hyperfine (https://github.com/sharkdp/hyperfine) on the NAS parallel > > benchmarks (https://www.nas.nasa.gov/publications/npb.html). The > > tests were run on an 80-core Intel(R) Xeon(R) CPU E7-8870 v4 @ > > 2.10GHz. Active (intel_pstate) and passive (intel_cpufreq) power > > management were used. Times are in seconds. All experiments use all > > 160 hardware threads. > >=20 > > v5.9/active v5.9+patch/active > > bt.C.c 24.725724+-0.962340 23.349608+-1.607214 > > lu.C.x 29.105952+-4.804203 25.249052+-5.561617 > > sp.C.x 31.220696+-1.831335 30.227760+-2.429792 > > ua.C.x 26.606118+-1.767384 25.778367+-1.263850 > >=20 > > v5.9/passive v5.9+patch/passive > > bt.C.c 25.330360+-1.028316 23.544036+-1.020189 > > lu.C.x 35.872659+-4.872090 23.719295+-3.883848 > > sp.C.x 32.141310+-2.289541 29.125363+-0.872300 > > ua.C.x 29.024597+-1.667049 25.728888+-1.539772 > >=20 > > On the smaller data sets (A and B) and on the other NAS benchmarks > > there is no impact on performance. > >=20 > > Signed-off-by: Julia Lawall >=20 > I suspect that the benefit of this patch is due to avoiding the overhead > of wake_affine_weight() check because the following check exists in > select_idle_sibling >=20 > /* > * If the previous CPU is cache affine and idle, don't be stupid: > */ > if (prev !=3D target && cpus_share_cache(prev, target) && > (available_idle_cpu(prev) || sched_idle_cpu(prev))) > return prev; >=20 > Still, the concept makes some sense to avoid wake_affine_weight but look > at the earlier part of wake_affine_idle() >=20 > if (available_idle_cpu(this_cpu) && cpus_share_cache(this_cpu, pr= ev_cpu)) > return available_idle_cpu(prev_cpu) ? prev_cpu : this_cpu; >=20 > This thing is almost completely useless because this_cpu is only going to > be idle if it's a wakeup from interrupt context when the CPU was otherwise > idle *but* it takes care to only use the CPU if this and prev share LLC. >=20 > The patch as it stands may leave a task on a remote node when it should > have been pulled local to the waker because prev happened to be idle. This > is not guaranteed because a node could have multiple LLCs and prev is > still appropriate but that's a different problem entirely and requires > much deeper surgery. Still, not pulling a task from a remote node is > a change in expected behaviour. While it's possible that NUMA domains > will not even reach this path, it depends on the NUMA distance as can > be seen in sd_init() for the setting of SD_WAKE_AFFINE so I think the > cpus_share_cache check is necessary. >=20 > I think it would be more appropriate to rework that block that checks > this_cpu to instead check if the CPUs share cache first and then return o= ne > of them (preference to prev based on the comment above it about avoiding > a migration) if either one is idle. >=20 > I see Vincent already agreed with the patch so I could be wrong. Vincent, > did I miss something stupid? This patch fixes the problem that we don't favor anymore the prev_cpu when = it is idle since commit 11f10e5420f6ce because load is not null when cpu is idle whereas run= nable_load was And this is important because this will=A0then decide in which LLC we will = looks for a cpu >=20 > --=20 > Mel Gorman > SUSE Labs