From: "Li, Aubrey" <aubrey.li@linux.intel.com>
To: Balbir Singh <bsingharora@gmail.com>,
"Joel Fernandes (Google)" <joel@joelfernandes.org>
Cc: Nishanth Aravamudan <naravamudan@digitalocean.com>,
Julien Desfossez <jdesfossez@digitalocean.com>,
Peter Zijlstra <peterz@infradead.org>,
Tim Chen <tim.c.chen@linux.intel.com>,
Vineeth Pillai <viremana@linux.microsoft.com>,
Aaron Lu <aaron.lwe@gmail.com>,
Aubrey Li <aubrey.intel@gmail.com>,
tglx@linutronix.de, linux-kernel@vger.kernel.org,
mingo@kernel.org, torvalds@linux-foundation.org,
fweisbec@gmail.com, keescook@chromium.org, kerrnel@google.com,
Phil Auld <pauld@redhat.com>,
Valentin Schneider <valentin.schneider@arm.com>,
Mel Gorman <mgorman@techsingularity.net>,
Pawan Gupta <pawan.kumar.gupta@linux.intel.com>,
Paolo Bonzini <pbonzini@redhat.com>,
vineeth@bitbyteword.org, Chen Yu <yu.c.chen@intel.com>,
Christian Brauner <christian.brauner@ubuntu.com>,
Agata Gruza <agata.gruza@intel.com>,
Antonio Gomez Iglesias <antonio.gomez.iglesias@intel.com>,
graf@amazon.com, konrad.wilk@oracle.com, dfaggioli@suse.com,
pjt@google.com, rostedt@goodmis.org, derkling@google.com,
benbjiang@tencent.com,
Alexandre Chartre <alexandre.chartre@oracle.com>,
James.Bottomley@hansenpartnership.com, OWeisse@umich.edu,
Dhaval Giani <dhaval.giani@oracle.com>,
Junaid Shahid <junaids@google.com>,
jsbarnes@google.com, chris.hyser@oracle.com,
Ben Segall <bsegall@google.com>, Josh Don <joshdon@google.com>,
Hao Luo <haoluo@google.com>,
Tom Lendacky <thomas.lendacky@amd.com>,
Aubrey Li <aubrey.li@intel.com>,
"Paul E. McKenney" <paulmck@kernel.org>,
Tim Chen <tim.c.chen@intel.com>
Subject: Re: [PATCH -tip 14/32] sched: migration changes for core scheduling
Date: Mon, 23 Nov 2020 12:36:10 +0800 [thread overview]
Message-ID: <0b2514ef-6cc3-c1a3-280b-5d9062c80a31@linux.intel.com> (raw)
In-Reply-To: <20201122235456.GF110669@balbir-desktop>
On 2020/11/23 7:54, Balbir Singh wrote:
> On Tue, Nov 17, 2020 at 06:19:44PM -0500, Joel Fernandes (Google) wrote:
>> From: Aubrey Li <aubrey.li@intel.com>
>>
>> - Don't migrate if there is a cookie mismatch
>> Load balance tries to move task from busiest CPU to the
>> destination CPU. When core scheduling is enabled, if the
>> task's cookie does not match with the destination CPU's
>> core cookie, this task will be skipped by this CPU. This
>> mitigates the forced idle time on the destination CPU.
>>
>> - Select cookie matched idle CPU
>> In the fast path of task wakeup, select the first cookie matched
>> idle CPU instead of the first idle CPU.
>>
>> - Find cookie matched idlest CPU
>> In the slow path of task wakeup, find the idlest CPU whose core
>> cookie matches with task's cookie
>>
>> - Don't migrate task if cookie not match
>> For the NUMA load balance, don't migrate task to the CPU whose
>> core cookie does not match with task's cookie
>>
>> Tested-by: Julien Desfossez <jdesfossez@digitalocean.com>
>> Signed-off-by: Aubrey Li <aubrey.li@linux.intel.com>
>> Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
>> Signed-off-by: Vineeth Remanan Pillai <viremana@linux.microsoft.com>
>> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org>
>> ---
>> kernel/sched/fair.c | 64 ++++++++++++++++++++++++++++++++++++++++----
>> kernel/sched/sched.h | 29 ++++++++++++++++++++
>> 2 files changed, 88 insertions(+), 5 deletions(-)
>>
>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>> index de82f88ba98c..ceb3906c9a8a 100644
>> --- a/kernel/sched/fair.c
>> +++ b/kernel/sched/fair.c
>> @@ -1921,6 +1921,15 @@ static void task_numa_find_cpu(struct task_numa_env *env,
>> if (!cpumask_test_cpu(cpu, env->p->cpus_ptr))
>> continue;
>>
>> +#ifdef CONFIG_SCHED_CORE
>> + /*
>> + * Skip this cpu if source task's cookie does not match
>> + * with CPU's core cookie.
>> + */
>> + if (!sched_core_cookie_match(cpu_rq(cpu), env->p))
>> + continue;
>> +#endif
>> +
>
> Any reason this is under an #ifdef? In sched_core_cookie_match() won't
> the check for sched_core_enabled() do the right thing even when
> CONFIG_SCHED_CORE is not enabed?>
Yes, sched_core_enabled works properly when CONFIG_SCHED_CORE is not
enabled. But when CONFIG_SCHED_CORE is not enabled, it does not make
sense to leave a core scheduler specific function here even at compile
time. Also, for the cases in hot path, this saves CPU cycles to avoid
a judgment.
>> env->dst_cpu = cpu;
>> if (task_numa_compare(env, taskimp, groupimp, maymove))
>> break;
>> @@ -5867,11 +5876,17 @@ find_idlest_group_cpu(struct sched_group *group, struct task_struct *p, int this
>>
>> /* Traverse only the allowed CPUs */
>> for_each_cpu_and(i, sched_group_span(group), p->cpus_ptr) {
>> + struct rq *rq = cpu_rq(i);
>> +
>> +#ifdef CONFIG_SCHED_CORE
>> + if (!sched_core_cookie_match(rq, p))
>> + continue;
>> +#endif
>> +
>> if (sched_idle_cpu(i))
>> return i;
>>
>> if (available_idle_cpu(i)) {
>> - struct rq *rq = cpu_rq(i);
>> struct cpuidle_state *idle = idle_get_state(rq);
>> if (idle && idle->exit_latency < min_exit_latency) {
>> /*
>> @@ -6129,8 +6144,18 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
>> for_each_cpu_wrap(cpu, cpus, target) {
>> if (!--nr)
>> return -1;
>> - if (available_idle_cpu(cpu) || sched_idle_cpu(cpu))
>> - break;
>> +
>> + if (available_idle_cpu(cpu) || sched_idle_cpu(cpu)) {
>> +#ifdef CONFIG_SCHED_CORE
>> + /*
>> + * If Core Scheduling is enabled, select this cpu
>> + * only if the process cookie matches core cookie.
>> + */
>> + if (sched_core_enabled(cpu_rq(cpu)) &&
>> + p->core_cookie == cpu_rq(cpu)->core->core_cookie)
>> +#endif
>> + break;
>> + }
>> }
>>
>> time = cpu_clock(this) - time;
>> @@ -7530,8 +7555,9 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env)
>> * We do not migrate tasks that are:
>> * 1) throttled_lb_pair, or
>> * 2) cannot be migrated to this CPU due to cpus_ptr, or
>> - * 3) running (obviously), or
>> - * 4) are cache-hot on their current CPU.
>> + * 3) task's cookie does not match with this CPU's core cookie
>> + * 4) running (obviously), or
>> + * 5) are cache-hot on their current CPU.
>> */
>> if (throttled_lb_pair(task_group(p), env->src_cpu, env->dst_cpu))
>> return 0;
>> @@ -7566,6 +7592,15 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env)
>> return 0;
>> }
>>
>> +#ifdef CONFIG_SCHED_CORE
>> + /*
>> + * Don't migrate task if the task's cookie does not match
>> + * with the destination CPU's core cookie.
>> + */
>> + if (!sched_core_cookie_match(cpu_rq(env->dst_cpu), p))
>> + return 0;
>> +#endif
>> +
>> /* Record that we found atleast one task that could run on dst_cpu */
>> env->flags &= ~LBF_ALL_PINNED;
>>
>> @@ -8792,6 +8827,25 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)
>> p->cpus_ptr))
>> continue;
>>
>> +#ifdef CONFIG_SCHED_CORE
>> + if (sched_core_enabled(cpu_rq(this_cpu))) {
>> + int i = 0;
>> + bool cookie_match = false;
>> +
>> + for_each_cpu(i, sched_group_span(group)) {
>> + struct rq *rq = cpu_rq(i);
>> +
>> + if (sched_core_cookie_match(rq, p)) {
>> + cookie_match = true;
>> + break;
>> + }
>> + }
>> + /* Skip over this group if no cookie matched */
>> + if (!cookie_match)
>> + continue;
>> + }
>> +#endif
>> +
>> local_group = cpumask_test_cpu(this_cpu,
>> sched_group_span(group));
>>
>> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
>> index e72942a9ee11..de553d39aa40 100644
>> --- a/kernel/sched/sched.h
>> +++ b/kernel/sched/sched.h
>> @@ -1135,6 +1135,35 @@ static inline raw_spinlock_t *rq_lockp(struct rq *rq)
>>
>> bool cfs_prio_less(struct task_struct *a, struct task_struct *b);
>>
>> +/*
>> + * Helper to check if the CPU's core cookie matches with the task's cookie
>> + * when core scheduling is enabled.
>> + * A special case is that the task's cookie always matches with CPU's core
>> + * cookie if the CPU is in an idle core.
>> + */
>> +static inline bool sched_core_cookie_match(struct rq *rq, struct task_struct *p)
>> +{
>> + bool idle_core = true;
>> + int cpu;
>> +
>> + /* Ignore cookie match if core scheduler is not enabled on the CPU. */
>> + if (!sched_core_enabled(rq))
>> + return true;
>> +
>> + for_each_cpu(cpu, cpu_smt_mask(cpu_of(rq))) {
>> + if (!available_idle_cpu(cpu)) {
>
> I was looking at this snippet and comparing this to is_core_idle(), the
> major difference is the check for vcpu_is_preempted(). Do we want to
> call the core as non idle if any vcpu was preempted on this CPU?
Yes, if there is a VCPU was preempted on this CPU, better not place task
on this core as the VCPU may be holding a spinlock and wants to be executed
again ASAP.
Thanks,
-Aubrey
next prev parent reply other threads:[~2020-11-23 4:36 UTC|newest]
Thread overview: 150+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-17 23:19 [PATCH -tip 00/32] Core scheduling (v9) Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 01/32] sched: Wrap rq::lock access Joel Fernandes (Google)
2020-11-19 23:31 ` Singh, Balbir
2020-11-20 16:55 ` Joel Fernandes
2020-11-22 8:52 ` Balbir Singh
2020-11-17 23:19 ` [PATCH -tip 02/32] sched: Introduce sched_class::pick_task() Joel Fernandes (Google)
2020-11-19 23:56 ` Singh, Balbir
2020-11-20 16:58 ` Joel Fernandes
2020-11-25 23:19 ` Balbir Singh
2020-11-25 16:28 ` Vincent Guittot
2020-11-26 9:07 ` Peter Zijlstra
2020-11-26 10:17 ` Vincent Guittot
2020-11-26 12:40 ` Peter Zijlstra
2020-11-17 23:19 ` [PATCH -tip 03/32] sched/fair: Fix pick_task_fair crashes due to empty rbtree Joel Fernandes (Google)
2020-11-20 10:15 ` Singh, Balbir
2020-11-20 18:11 ` Vineeth Pillai
2020-11-23 22:31 ` Balbir Singh
2020-11-24 8:31 ` Peter Zijlstra
2020-11-17 23:19 ` [PATCH -tip 04/32] sched: Core-wide rq->lock Joel Fernandes (Google)
2020-11-22 9:11 ` Balbir Singh
2020-11-24 8:16 ` Peter Zijlstra
2020-11-26 0:35 ` Balbir Singh
2020-11-17 23:19 ` [PATCH -tip 05/32] sched/fair: Add a few assertions Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 06/32] sched: Basic tracking of matching tasks Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 07/32] sched: Add core wide task selection and scheduling Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 08/32] sched/fair: Fix forced idle sibling starvation corner case Joel Fernandes (Google)
2020-11-22 10:35 ` Balbir Singh
2020-11-17 23:19 ` [PATCH -tip 09/32] sched/fair: Snapshot the min_vruntime of CPUs on force idle Joel Fernandes (Google)
2020-11-22 11:44 ` Balbir Singh
2020-11-23 12:31 ` Vineeth Pillai
2020-11-23 23:31 ` Balbir Singh
2020-11-24 9:09 ` Peter Zijlstra
2020-11-25 23:17 ` Balbir Singh
2020-11-26 8:23 ` Peter Zijlstra
2020-11-17 23:19 ` [PATCH -tip 10/32] sched: Fix priority inversion of cookied task with sibling Joel Fernandes (Google)
2020-11-22 22:41 ` Balbir Singh
2020-11-24 18:30 ` Joel Fernandes
2020-11-25 23:05 ` Balbir Singh
2020-11-26 8:29 ` Peter Zijlstra
2020-11-26 22:27 ` Balbir Singh
2020-12-01 17:49 ` Joel Fernandes
2020-11-17 23:19 ` [PATCH -tip 11/32] sched: Enqueue task into core queue only after vruntime is updated Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 12/32] sched: Simplify the core pick loop for optimized case Joel Fernandes (Google)
2020-11-24 12:04 ` Peter Zijlstra
2020-11-24 17:04 ` Joel Fernandes
2020-11-25 8:37 ` Peter Zijlstra
2020-11-17 23:19 ` [PATCH -tip 13/32] sched: Trivial forced-newidle balancer Joel Fernandes (Google)
2020-11-23 4:38 ` Balbir Singh
2020-11-23 15:07 ` Li, Aubrey
2020-11-23 23:35 ` Balbir Singh
2020-11-24 0:32 ` Li, Aubrey
2020-11-25 21:28 ` Balbir Singh
2020-11-17 23:19 ` [PATCH -tip 14/32] sched: migration changes for core scheduling Joel Fernandes (Google)
2020-11-22 23:54 ` Balbir Singh
2020-11-23 4:36 ` Li, Aubrey [this message]
2020-11-24 15:42 ` Peter Zijlstra
2020-11-25 3:12 ` Li, Aubrey
2020-11-25 22:57 ` Balbir Singh
2020-11-26 3:20 ` Li, Aubrey
2020-11-26 8:32 ` Balbir Singh
2020-11-26 9:26 ` Li, Aubrey
2020-11-30 9:33 ` Balbir Singh
2020-11-30 12:29 ` Li, Aubrey
2020-12-02 14:09 ` Li, Aubrey
2020-12-03 1:06 ` Li, Aubrey
2020-11-30 10:35 ` Vincent Guittot
2020-11-30 12:32 ` Li, Aubrey
2020-11-17 23:19 ` [PATCH -tip 15/32] sched: Improve snapshotting of min_vruntime for CGroups Joel Fernandes (Google)
2020-11-24 10:27 ` Peter Zijlstra
2020-11-24 17:07 ` Joel Fernandes
2020-11-25 8:41 ` Peter Zijlstra
2020-11-24 10:41 ` Peter Zijlstra
2020-11-17 23:19 ` [PATCH -tip 16/32] irq_work: Cleanup Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 17/32] arch/x86: Add a new TIF flag for untrusted tasks Joel Fernandes (Google)
2020-11-23 5:18 ` Balbir Singh
2020-11-17 23:19 ` [PATCH -tip 18/32] kernel/entry: Add support for core-wide protection of kernel-mode Joel Fernandes (Google)
2020-11-24 16:09 ` Peter Zijlstra
2020-11-24 17:52 ` Joel Fernandes
2020-11-25 9:37 ` Peter Zijlstra
2020-12-01 17:55 ` Joel Fernandes
2020-11-26 5:37 ` Balbir Singh
2020-11-17 23:19 ` [PATCH -tip 19/32] entry/idle: Enter and exit kernel protection during idle entry and exit Joel Fernandes (Google)
2020-11-24 16:13 ` Peter Zijlstra
2020-11-24 18:03 ` Joel Fernandes
2020-11-25 8:49 ` Peter Zijlstra
2020-12-01 18:24 ` Joel Fernandes
2020-11-17 23:19 ` [PATCH -tip 20/32] entry/kvm: Protect the kernel when entering from guest Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 21/32] sched: CGroup tagging interface for core scheduling Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 22/32] sched: Split the cookie and setup per-task cookie on fork Joel Fernandes (Google)
2020-11-25 11:07 ` Peter Zijlstra
2020-12-01 18:56 ` Joel Fernandes
2020-11-25 11:10 ` Peter Zijlstra
2020-12-01 19:20 ` Joel Fernandes
2020-12-01 19:34 ` Peter Zijlstra
2020-12-02 6:36 ` Josh Don
2020-12-02 7:54 ` Peter Zijlstra
2020-12-04 0:20 ` Josh Don
2020-12-06 17:49 ` Joel Fernandes
2020-11-25 11:11 ` Peter Zijlstra
2020-12-01 19:16 ` Joel Fernandes
2020-11-25 11:15 ` Peter Zijlstra
2020-12-01 19:11 ` Joel Fernandes
2020-12-01 19:20 ` Peter Zijlstra
2020-12-06 18:15 ` Joel Fernandes
2020-11-25 12:54 ` Peter Zijlstra
2020-12-01 18:38 ` Joel Fernandes
2020-11-25 13:03 ` Peter Zijlstra
2020-12-01 18:52 ` Joel Fernandes
2020-11-30 23:05 ` Balbir Singh
2020-11-17 23:19 ` [PATCH -tip 23/32] sched: Add a per-thread core scheduling interface Joel Fernandes (Google)
2020-11-25 13:08 ` Peter Zijlstra
2020-12-01 19:36 ` Joel Fernandes
2020-12-02 21:47 ` Chris Hyser
2020-12-02 23:13 ` chris hyser
2020-12-06 17:34 ` Joel Fernandes
2020-12-07 21:48 ` chris hyser
2020-12-09 18:52 ` Chris Hyser
2020-12-14 19:31 ` Joel Fernandes
2020-12-14 19:44 ` chris hyser
2020-12-14 23:25 ` Joel Fernandes
2020-12-15 14:56 ` chris hyser
2020-12-15 16:23 ` chris hyser
2020-12-15 18:13 ` Dhaval Giani
2020-12-16 0:35 ` Joel Fernandes
2020-11-17 23:19 ` [PATCH -tip 24/32] sched: Release references to the per-task cookie on exit Joel Fernandes (Google)
2020-11-25 13:03 ` Peter Zijlstra
2020-11-17 23:19 ` [PATCH -tip 25/32] sched: Refactor core cookie into struct Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 26/32] sched: Add a second-level tag for nested CGroup usecase Joel Fernandes (Google)
2020-11-25 13:42 ` Peter Zijlstra
2020-11-30 23:10 ` Balbir Singh
2020-12-01 20:08 ` Joel Fernandes
2020-12-02 6:18 ` Josh Don
2020-12-02 8:02 ` Peter Zijlstra
2020-12-02 18:53 ` Tejun Heo
2020-12-04 0:51 ` Josh Don
2020-12-04 15:45 ` Tejun Heo
2020-11-17 23:19 ` [PATCH -tip 27/32] sched/debug: Add CGroup node for printing group cookie if SCHED_DEBUG Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 28/32] kselftest: Add tests for core-sched interface Joel Fernandes (Google)
2020-11-17 23:19 ` [PATCH -tip 29/32] sched: Move core-scheduler interfacing code to a new file Joel Fernandes (Google)
2020-11-17 23:20 ` [PATCH -tip 30/32] Documentation: Add core scheduling documentation Joel Fernandes (Google)
2020-11-17 23:20 ` [PATCH -tip 31/32] sched: Add a coresched command line option Joel Fernandes (Google)
2020-11-19 23:39 ` Randy Dunlap
2020-11-25 13:45 ` Peter Zijlstra
2020-11-26 0:11 ` Balbir Singh
2020-11-17 23:20 ` [PATCH -tip 32/32] sched: Debug bits Joel Fernandes (Google)
2020-12-01 0:21 ` Balbir Singh
2021-01-15 15:10 ` Joel Fernandes
2020-11-24 11:48 ` [PATCH -tip 00/32] Core scheduling (v9) Vincent Guittot
2020-11-24 15:08 ` Joel Fernandes
2020-12-03 6:16 ` Ning, Hongyu
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=0b2514ef-6cc3-c1a3-280b-5d9062c80a31@linux.intel.com \
--to=aubrey.li@linux.intel.com \
--cc=James.Bottomley@hansenpartnership.com \
--cc=OWeisse@umich.edu \
--cc=aaron.lwe@gmail.com \
--cc=agata.gruza@intel.com \
--cc=alexandre.chartre@oracle.com \
--cc=antonio.gomez.iglesias@intel.com \
--cc=aubrey.intel@gmail.com \
--cc=aubrey.li@intel.com \
--cc=benbjiang@tencent.com \
--cc=bsegall@google.com \
--cc=bsingharora@gmail.com \
--cc=chris.hyser@oracle.com \
--cc=christian.brauner@ubuntu.com \
--cc=derkling@google.com \
--cc=dfaggioli@suse.com \
--cc=dhaval.giani@oracle.com \
--cc=fweisbec@gmail.com \
--cc=graf@amazon.com \
--cc=haoluo@google.com \
--cc=jdesfossez@digitalocean.com \
--cc=joel@joelfernandes.org \
--cc=joshdon@google.com \
--cc=jsbarnes@google.com \
--cc=junaids@google.com \
--cc=keescook@chromium.org \
--cc=kerrnel@google.com \
--cc=konrad.wilk@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@techsingularity.net \
--cc=mingo@kernel.org \
--cc=naravamudan@digitalocean.com \
--cc=pauld@redhat.com \
--cc=paulmck@kernel.org \
--cc=pawan.kumar.gupta@linux.intel.com \
--cc=pbonzini@redhat.com \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=thomas.lendacky@amd.com \
--cc=tim.c.chen@intel.com \
--cc=tim.c.chen@linux.intel.com \
--cc=torvalds@linux-foundation.org \
--cc=valentin.schneider@arm.com \
--cc=vineeth@bitbyteword.org \
--cc=viremana@linux.microsoft.com \
--cc=yu.c.chen@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;
as well as URLs for NNTP newsgroup(s).