From: Will Deacon <will@kernel.org> To: Valentin Schneider <valentin.schneider@arm.com> Cc: linux-arm-kernel@lists.infradead.org, linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org, Catalin Marinas <catalin.marinas@arm.com>, Marc Zyngier <maz@kernel.org>, Greg Kroah-Hartman <gregkh@linuxfoundation.org>, Peter Zijlstra <peterz@infradead.org>, Morten Rasmussen <morten.rasmussen@arm.com>, Qais Yousef <qais.yousef@arm.com>, Suren Baghdasaryan <surenb@google.com>, Quentin Perret <qperret@google.com>, Tejun Heo <tj@kernel.org>, Johannes Weiner <hannes@cmpxchg.org>, Ingo Molnar <mingo@redhat.com>, Juri Lelli <juri.lelli@redhat.com>, Vincent Guittot <vincent.guittot@linaro.org>, "Rafael J. Wysocki" <rjw@rjwysocki.net>, Dietmar Eggemann <dietmar.eggemann@arm.com>, Daniel Bristot de Oliveira <bristot@redhat.com>, kernel-team@android.com Subject: Re: [PATCH v7 01/22] sched: Favour predetermined active CPU as migration destination Date: Wed, 26 May 2021 17:03:18 +0100 [thread overview] Message-ID: <20210526160317.GB19691@willie-the-truck> (raw) In-Reply-To: <877djlhhmb.mognet@arm.com> On Wed, May 26, 2021 at 12:14:20PM +0100, Valentin Schneider wrote: > On 25/05/21 16:14, Will Deacon wrote: > > diff --git a/kernel/sched/core.c b/kernel/sched/core.c > > index 5226cc26a095..1702a60d178d 100644 > > --- a/kernel/sched/core.c > > +++ b/kernel/sched/core.c > > @@ -1869,6 +1869,7 @@ static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf, > > struct migration_arg { > > struct task_struct *task; > > int dest_cpu; > > + const struct cpumask *dest_mask; > > struct set_affinity_pending *pending; > > }; > > > > @@ -1917,6 +1918,7 @@ static int migration_cpu_stop(void *data) > > struct set_affinity_pending *pending = arg->pending; > > struct task_struct *p = arg->task; > > int dest_cpu = arg->dest_cpu; > > + const struct cpumask *dest_mask = arg->dest_mask; > > struct rq *rq = this_rq(); > > bool complete = false; > > struct rq_flags rf; > > @@ -1956,12 +1958,8 @@ static int migration_cpu_stop(void *data) > > complete = true; > > } > > > > - if (dest_cpu < 0) { > > - if (cpumask_test_cpu(task_cpu(p), &p->cpus_mask)) > > - goto out; > > - > > - dest_cpu = cpumask_any_distribute(&p->cpus_mask); > > - } > > + if (dest_mask && (cpumask_test_cpu(task_cpu(p), dest_mask))) > > + goto out; > > > > IIRC the reason we deferred the pick to migration_cpu_stop() was because of > those insane races involving multiple SCA calls the likes of: > > p->cpus_mask = [0, 1]; p on CPU0 > > CPUx CPUy CPU0 > > SCA(p, [2]) > __do_set_cpus_allowed(); > queue migration_cpu_stop() > SCA(p, [3]) > __do_set_cpus_allowed(); > migration_cpu_stop() > > The stopper needs to use the latest cpumask set by the second SCA despite > having an arg->pending set up by the first SCA. Doesn't this break here? Yes, well spotted. I was so caught up with the hotplug race that I didn't even consider a straightforward SCA race. Hurumph. > I'm not sure I've paged back in all of the subtleties laying in ambush > here, but what about the below? I can't break it, but I'm also not very familiar with this code. Please can you post it as a proper patch so that I drop this from my series? Thanks, Will
WARNING: multiple messages have this Message-ID (diff)
From: Will Deacon <will@kernel.org> To: Valentin Schneider <valentin.schneider@arm.com> Cc: linux-arm-kernel@lists.infradead.org, linux-arch@vger.kernel.org, linux-kernel@vger.kernel.org, Catalin Marinas <catalin.marinas@arm.com>, Marc Zyngier <maz@kernel.org>, Greg Kroah-Hartman <gregkh@linuxfoundation.org>, Peter Zijlstra <peterz@infradead.org>, Morten Rasmussen <morten.rasmussen@arm.com>, Qais Yousef <qais.yousef@arm.com>, Suren Baghdasaryan <surenb@google.com>, Quentin Perret <qperret@google.com>, Tejun Heo <tj@kernel.org>, Johannes Weiner <hannes@cmpxchg.org>, Ingo Molnar <mingo@redhat.com>, Juri Lelli <juri.lelli@redhat.com>, Vincent Guittot <vincent.guittot@linaro.org>, "Rafael J. Wysocki" <rjw@rjwysocki.net>, Dietmar Eggemann <dietmar.eggemann@arm.com>, Daniel Bristot de Oliveira <bristot@redhat.com>, kernel-team@android.com Subject: Re: [PATCH v7 01/22] sched: Favour predetermined active CPU as migration destination Date: Wed, 26 May 2021 17:03:18 +0100 [thread overview] Message-ID: <20210526160317.GB19691@willie-the-truck> (raw) In-Reply-To: <877djlhhmb.mognet@arm.com> On Wed, May 26, 2021 at 12:14:20PM +0100, Valentin Schneider wrote: > On 25/05/21 16:14, Will Deacon wrote: > > diff --git a/kernel/sched/core.c b/kernel/sched/core.c > > index 5226cc26a095..1702a60d178d 100644 > > --- a/kernel/sched/core.c > > +++ b/kernel/sched/core.c > > @@ -1869,6 +1869,7 @@ static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf, > > struct migration_arg { > > struct task_struct *task; > > int dest_cpu; > > + const struct cpumask *dest_mask; > > struct set_affinity_pending *pending; > > }; > > > > @@ -1917,6 +1918,7 @@ static int migration_cpu_stop(void *data) > > struct set_affinity_pending *pending = arg->pending; > > struct task_struct *p = arg->task; > > int dest_cpu = arg->dest_cpu; > > + const struct cpumask *dest_mask = arg->dest_mask; > > struct rq *rq = this_rq(); > > bool complete = false; > > struct rq_flags rf; > > @@ -1956,12 +1958,8 @@ static int migration_cpu_stop(void *data) > > complete = true; > > } > > > > - if (dest_cpu < 0) { > > - if (cpumask_test_cpu(task_cpu(p), &p->cpus_mask)) > > - goto out; > > - > > - dest_cpu = cpumask_any_distribute(&p->cpus_mask); > > - } > > + if (dest_mask && (cpumask_test_cpu(task_cpu(p), dest_mask))) > > + goto out; > > > > IIRC the reason we deferred the pick to migration_cpu_stop() was because of > those insane races involving multiple SCA calls the likes of: > > p->cpus_mask = [0, 1]; p on CPU0 > > CPUx CPUy CPU0 > > SCA(p, [2]) > __do_set_cpus_allowed(); > queue migration_cpu_stop() > SCA(p, [3]) > __do_set_cpus_allowed(); > migration_cpu_stop() > > The stopper needs to use the latest cpumask set by the second SCA despite > having an arg->pending set up by the first SCA. Doesn't this break here? Yes, well spotted. I was so caught up with the hotplug race that I didn't even consider a straightforward SCA race. Hurumph. > I'm not sure I've paged back in all of the subtleties laying in ambush > here, but what about the below? I can't break it, but I'm also not very familiar with this code. Please can you post it as a proper patch so that I drop this from my series? Thanks, Will _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2021-05-26 16:03 UTC|newest] Thread overview: 114+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-05-25 15:14 [PATCH v7 00/22] Add support for 32-bit tasks on asymmetric AArch32 systems Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 01/22] sched: Favour predetermined active CPU as migration destination Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-26 11:14 ` Valentin Schneider 2021-05-26 11:14 ` Valentin Schneider 2021-05-26 12:32 ` Peter Zijlstra 2021-05-26 12:32 ` Peter Zijlstra 2021-05-26 12:36 ` Valentin Schneider 2021-05-26 12:36 ` Valentin Schneider 2021-05-26 16:03 ` Will Deacon [this message] 2021-05-26 16:03 ` Will Deacon 2021-05-26 17:46 ` Valentin Schneider 2021-05-26 17:46 ` Valentin Schneider 2021-05-25 15:14 ` [PATCH v7 02/22] arm64: cpuinfo: Split AArch32 registers out into a separate struct Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 03/22] arm64: Allow mismatched 32-bit EL0 support Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 04/22] KVM: arm64: Kill 32-bit vCPUs on systems with mismatched " Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 05/22] arm64: Kill 32-bit applications scheduled on 64-bit-only CPUs Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 06/22] arm64: Advertise CPUs capable of running 32-bit applications in sysfs Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 07/22] sched: Introduce task_cpu_possible_mask() to limit fallback rq selection Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 08/22] cpuset: Don't use the cpu_possible_mask as a last resort for cgroup v1 Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-26 15:02 ` Peter Zijlstra 2021-05-26 15:02 ` Peter Zijlstra 2021-05-26 16:07 ` Will Deacon 2021-05-26 16:07 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 09/22] cpuset: Honour task_cpu_possible_mask() in guarantee_online_cpus() Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 10/22] sched: Reject CPU affinity changes based on task_cpu_possible_mask() Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-26 15:15 ` Peter Zijlstra 2021-05-26 15:15 ` Peter Zijlstra 2021-05-26 16:12 ` Will Deacon 2021-05-26 16:12 ` Will Deacon 2021-05-26 17:56 ` Peter Zijlstra 2021-05-26 17:56 ` Peter Zijlstra 2021-05-26 18:59 ` Will Deacon 2021-05-26 18:59 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 11/22] sched: Introduce task_struct::user_cpus_ptr to track requested affinity Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 12/22] sched: Split the guts of sched_setaffinity() into a helper function Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 13/22] sched: Allow task CPU affinity to be restricted on asymmetric systems Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-26 16:20 ` Peter Zijlstra 2021-05-26 16:20 ` Peter Zijlstra 2021-05-26 16:35 ` Will Deacon 2021-05-26 16:35 ` Will Deacon 2021-05-26 16:30 ` Peter Zijlstra 2021-05-26 16:30 ` Peter Zijlstra 2021-05-26 17:02 ` Will Deacon 2021-05-26 17:02 ` Will Deacon 2021-05-27 7:56 ` Peter Zijlstra 2021-05-27 7:56 ` Peter Zijlstra 2021-05-25 15:14 ` [PATCH v7 14/22] sched: Introduce task_cpus_dl_admissible() to check proposed affinity Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 15/22] freezer: Add frozen_or_skipped() helper function Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 16/22] sched: Defer wakeup in ttwu() for unschedulable frozen tasks Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-27 14:10 ` Peter Zijlstra 2021-05-27 14:10 ` Peter Zijlstra 2021-05-27 14:31 ` Peter Zijlstra 2021-05-27 14:31 ` Peter Zijlstra 2021-05-27 14:44 ` Will Deacon 2021-05-27 14:44 ` Will Deacon 2021-05-27 14:55 ` Peter Zijlstra 2021-05-27 14:55 ` Peter Zijlstra 2021-05-27 14:50 ` Peter Zijlstra 2021-05-27 14:50 ` Peter Zijlstra 2021-05-28 10:49 ` Peter Zijlstra 2021-05-28 10:49 ` Peter Zijlstra 2021-05-27 14:36 ` Will Deacon 2021-05-27 14:36 ` Will Deacon 2021-06-01 8:21 ` [RFC][PATCH] freezer,sched: Rewrite core freezer logic Peter Zijlstra 2021-06-01 8:21 ` Peter Zijlstra 2021-06-01 11:27 ` Peter Zijlstra 2021-06-01 11:27 ` Peter Zijlstra 2021-06-02 12:54 ` Will Deacon 2021-06-02 12:54 ` Will Deacon 2021-06-03 10:35 ` Peter Zijlstra 2021-06-03 10:35 ` Peter Zijlstra 2021-06-03 10:58 ` Will Deacon 2021-06-03 10:58 ` Will Deacon 2021-06-03 11:26 ` Peter Zijlstra 2021-06-03 11:26 ` Peter Zijlstra 2021-06-03 11:36 ` Will Deacon 2021-06-03 11:36 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 17/22] arm64: Implement task_cpu_possible_mask() Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 18/22] arm64: exec: Adjust affinity for compat tasks with mismatched 32-bit EL0 Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 19/22] arm64: Prevent offlining first CPU with 32-bit EL0 on mismatched system Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 20/22] arm64: Hook up cmdline parameter to allow mismatched 32-bit EL0 Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 21/22] arm64: Remove logic to kill 32-bit tasks on 64-bit-only cores Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 15:14 ` [PATCH v7 22/22] Documentation: arm64: describe asymmetric 32-bit support Will Deacon 2021-05-25 15:14 ` Will Deacon 2021-05-25 17:13 ` Marc Zyngier 2021-05-25 17:13 ` Marc Zyngier 2021-05-25 17:27 ` Will Deacon 2021-05-25 17:27 ` Will Deacon 2021-05-25 18:11 ` Marc Zyngier 2021-05-25 18:11 ` Marc Zyngier 2021-05-26 16:00 ` Will Deacon 2021-05-26 16:00 ` Will Deacon
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=20210526160317.GB19691@willie-the-truck \ --to=will@kernel.org \ --cc=bristot@redhat.com \ --cc=catalin.marinas@arm.com \ --cc=dietmar.eggemann@arm.com \ --cc=gregkh@linuxfoundation.org \ --cc=hannes@cmpxchg.org \ --cc=juri.lelli@redhat.com \ --cc=kernel-team@android.com \ --cc=linux-arch@vger.kernel.org \ --cc=linux-arm-kernel@lists.infradead.org \ --cc=linux-kernel@vger.kernel.org \ --cc=maz@kernel.org \ --cc=mingo@redhat.com \ --cc=morten.rasmussen@arm.com \ --cc=peterz@infradead.org \ --cc=qais.yousef@arm.com \ --cc=qperret@google.com \ --cc=rjw@rjwysocki.net \ --cc=surenb@google.com \ --cc=tj@kernel.org \ --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: linkBe 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.