linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Balbir Singh <bsingharora@gmail.com>,
	LKML <linux-kernel@vger.kernel.org>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Nicholas Piggin <nicholas.piggin@gmail.com>
Subject: Re: [RFC][PATCH] Fix a race between rwsem and the scheduler
Date: Tue, 30 Aug 2016 15:04:27 +0200	[thread overview]
Message-ID: <20160830130426.GA17795@redhat.com> (raw)
In-Reply-To: <20160830121937.GQ10138@twins.programming.kicks-ass.net>

On 08/30, Peter Zijlstra wrote:
>
> 	/*
> 	 * Ensure we load p->on_rq _after_ p->state, otherwise it would
> 	 * be possible to, falsely, observe p->on_rq == 0 and get stuck
> 	 * in smp_cond_load_acquire() below.
> 	 *
> 	 * sched_ttwu_pending()			try_to_wake_up()
> 	 *   [S] p->on_rq = 1;			[L] P->state
> 	 *       UNLOCK rq->lock
> 	 *
> 	 * schedule()				RMB
> 	 *       LOCK rq->lock
> 	 *       UNLOCK rq->lock
> 	 *
> 	 * [task p]
> 	 *   [S] p->state = UNINTERRUPTIBLE	[L] p->on_rq
> 	 *
> 	 * Pairs with the UNLOCK+LOCK on rq->lock from the
> 	 * last wakeup of our task and the schedule that got our task
> 	 * current.
> 	 */

Confused... how this connects to UNLOCK+LOCK on rq->lock? A LOAD can
leak into the critical section.

But context switch should imply mb() we can rely on?

Oleg.

  reply	other threads:[~2016-08-30 13:05 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-30  8:49 [RFC][PATCH] Fix a race between rwsem and the scheduler Balbir Singh
2016-08-30  9:13 ` Nicholas Piggin
2016-08-30 12:19 ` Peter Zijlstra
2016-08-30 13:04   ` Oleg Nesterov [this message]
2016-08-30 14:13     ` Peter Zijlstra
2016-08-30 16:57       ` Oleg Nesterov
2016-08-30 18:34         ` Peter Zijlstra
2016-08-30 21:28           ` Benjamin Herrenschmidt
2016-08-31  7:18             ` Peter Zijlstra
2016-08-31 10:56               ` Benjamin Herrenschmidt
2016-08-31 13:31             ` Peter Zijlstra
2016-08-31 21:47               ` Benjamin Herrenschmidt
2016-09-01  6:49                 ` Balbir Singh
2016-09-01  6:57                 ` Peter Zijlstra
2016-09-01 14:17                   ` Boqun Feng
2016-09-01 15:33                     ` Peter Zijlstra
2016-08-30 21:25     ` Benjamin Herrenschmidt
2016-08-31  7:20       ` Peter Zijlstra
2016-08-31 10:55         ` Benjamin Herrenschmidt
2016-08-31  3:41   ` Balbir Singh
2016-08-31  7:28     ` Peter Zijlstra
2016-08-31 10:17       ` Balbir Singh
2016-08-31 10:57       ` Benjamin Herrenschmidt
2016-09-01  1:48       ` Alexey Kardashevskiy
2016-09-01 12:16         ` Alexey Kardashevskiy
2016-08-30 12:58 ` Oleg Nesterov
2016-08-31  3:25   ` Balbir Singh

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=20160830130426.GA17795@redhat.com \
    --to=oleg@redhat.com \
    --cc=benh@kernel.crashing.org \
    --cc=bsingharora@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nicholas.piggin@gmail.com \
    --cc=peterz@infradead.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).