From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: Oleg Nesterov <oleg@redhat.com>
Cc: paulmck <paulmck@linux.ibm.com>,
Peter Zijlstra <peterz@infradead.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
linux-kernel <linux-kernel@vger.kernel.org>,
"Eric W. Biederman" <ebiederm@xmission.com>,
"Russell King, ARM Linux" <linux@armlinux.org.uk>,
Chris Metcalf <cmetcalf@ezchip.com>, Chris Lameter <cl@linux.com>,
Kirill Tkhai <tkhai@yandex.ru>, Mike Galbraith <efault@gmx.de>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@kernel.org>
Subject: Re: [RFC PATCH 1/2] Fix: sched/membarrier: p->mm->membarrier_state racy load
Date: Wed, 4 Sep 2019 11:24:30 -0400 (EDT) [thread overview]
Message-ID: <88198910.1581.1567610670849.JavaMail.zimbra@efficios.com> (raw)
In-Reply-To: <20190904105348.GA24568@redhat.com>
----- On Sep 4, 2019, at 6:53 AM, Oleg Nesterov oleg@redhat.com wrote:
> On 09/03, Mathieu Desnoyers wrote:
>>
>> @@ -1130,6 +1130,10 @@ struct task_struct {
>> unsigned long numa_pages_migrated;
>> #endif /* CONFIG_NUMA_BALANCING */
>>
>> +#ifdef CONFIG_MEMBARRIER
>> + atomic_t membarrier_state;
>> +#endif
>
> ...
>
>> +static inline void membarrier_prepare_task_switch(struct task_struct *t)
>> +{
>> + if (!t->mm)
>> + return;
>> + atomic_set(&t->membarrier_state,
>> + atomic_read(&t->mm->membarrier_state));
>> +}
>
> Why not
>
> rq->membarrier_state = next->mm ? t->mm->membarrier_state : 0;
>
> and
>
> if (cpu_rq(cpu)->membarrier_state & MEMBARRIER_STATE_GLOBAL_EXPEDITED) {
> ...
> }
>
> in membarrier_global_expedited() ? (I removed atomic_ to simplify)
>
> IOW, why this new member has to live in task_struct, not in rq?
As replied to Linus, if we copy the membarrier_state into the rq, we'd need
to ensure we have full memory barriers between:
prior user-space memory accesses / setting the runqueue membarrier state
and
setting the runqueue membarrier state / following user-space memory accesses
Because membarrier does not take any runqueue lock when it iterates over
runqueues.
I try to avoid putting too much memory barrier constraints on the scheduler
for membarrier, but if it's really the way forward it could be done.
And the basic question remains: it is acceptable performance-wise to load
mm->membarrier_state from sched switch prepare ?
Thanks,
Mathieu
--
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com
next prev parent reply other threads:[~2019-09-04 15:24 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-03 20:11 [RFC PATCH 1/2] Fix: sched/membarrier: p->mm->membarrier_state racy load Mathieu Desnoyers
2019-09-03 20:11 ` [RFC PATCH 2/2] Fix: sched/membarrier: private expedited registration check Mathieu Desnoyers
2019-09-03 20:24 ` [RFC PATCH 1/2] Fix: sched/membarrier: p->mm->membarrier_state racy load Peter Zijlstra
2019-09-03 20:36 ` Linus Torvalds
2019-09-04 15:19 ` Mathieu Desnoyers
2019-09-04 16:09 ` Peter Zijlstra
2019-09-04 17:12 ` Mathieu Desnoyers
2019-09-04 18:26 ` Peter Zijlstra
2019-09-06 0:51 ` Mathieu Desnoyers
2019-09-03 20:41 ` Mathieu Desnoyers
2019-09-04 11:28 ` Peter Zijlstra
2019-09-04 11:49 ` Peter Zijlstra
2019-09-04 15:26 ` Mathieu Desnoyers
2019-09-04 12:03 ` Oleg Nesterov
2019-09-04 12:43 ` Peter Zijlstra
2019-09-04 13:17 ` Oleg Nesterov
2019-09-03 20:27 ` Linus Torvalds
2019-09-03 20:53 ` Mathieu Desnoyers
2019-09-04 10:53 ` Oleg Nesterov
2019-09-04 11:39 ` Peter Zijlstra
2019-09-04 15:24 ` Mathieu Desnoyers [this message]
2019-09-04 11:11 ` Oleg Nesterov
2019-09-04 16:11 ` Mathieu Desnoyers
2019-09-08 13:46 ` Mathieu Desnoyers
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=88198910.1581.1567610670849.JavaMail.zimbra@efficios.com \
--to=mathieu.desnoyers@efficios.com \
--cc=cl@linux.com \
--cc=cmetcalf@ezchip.com \
--cc=ebiederm@xmission.com \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@armlinux.org.uk \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=paulmck@linux.ibm.com \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=tkhai@yandex.ru \
--cc=torvalds@linux-foundation.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).