From: Peter Zijlstra <peterz@infradead.org> To: Andrea Parri <andrea.parri@amarulasolutions.com> Cc: linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org, Ingo Molnar <mingo@redhat.com>, Will Deacon <will.deacon@arm.com>, Alan Stern <stern@rowland.harvard.edu>, Boqun Feng <boqun.feng@gmail.com>, Nicholas Piggin <npiggin@gmail.com>, David Howells <dhowells@redhat.com>, Jade Alglave <j.alglave@ucl.ac.uk>, Luc Maranget <luc.maranget@inria.fr>, "Paul E . McKenney" <paulmck@linux.vnet.ibm.com>, Akira Yokosawa <akiyks@gmail.com>, Daniel Lustig <dlustig@nvidia.com>, Jonathan Corbet <corbet@lwn.net>, Randy Dunlap <rdunlap@infradead.org>, Matthew Wilcox <willy@infradead.org> Subject: Re: [PATCH v2 2/3] locking: Clarify requirements for smp_mb__after_spinlock() Date: Mon, 2 Jul 2018 17:37:35 +0200 [thread overview] Message-ID: <20180702153735.GQ2494@hirez.programming.kicks-ass.net> (raw) In-Reply-To: <1530544315-14614-1-git-send-email-andrea.parri@amarulasolutions.com> On Mon, Jul 02, 2018 at 05:11:55PM +0200, Andrea Parri wrote: > /* > + * smp_mb__after_spinlock() provides the equivalent of a full memory barrier > + * between program-order earlier lock acquisitions and program-order later > + * memory accesses. > * > + * This guarantees that the following two properties hold: > * > + * 1) Given the snippet: > * > + * { X = 0; Y = 0; } > * > + * CPU0 CPU1 > * > + * WRITE_ONCE(X, 1); WRITE_ONCE(Y, 1); > + * spin_lock(S); smp_mb(); > + * smp_mb__after_spinlock(); r1 = READ_ONCE(X); > + * r0 = READ_ONCE(Y); > + * spin_unlock(S); > * > + * it is forbidden that CPU0 does not observe CPU1's store to Y (r0 = 0) > + * and CPU1 does not observe CPU0's store to X (r1 = 0); see the comments > + * preceding the call to smp_mb__after_spinlock() in __schedule() and in > + * try_to_wake_up(). > + * > + * 2) Given the snippet: > + * > + * { X = 0; Y = 0; } > + * > + * CPU0 CPU1 CPU2 > + * > + * spin_lock(S); spin_lock(S); r1 = READ_ONCE(Y); > + * WRITE_ONCE(X, 1); smp_mb__after_spinlock(); smp_rmb(); > + * spin_unlock(S); r0 = READ_ONCE(X); r2 = READ_ONCE(X); > + * WRITE_ONCE(Y, 1); > + * spin_unlock(S); > + * > + * it is forbidden that CPU0's critical section executes before CPU1's > + * critical section (r0 = 1), CPU2 observes CPU1's store to Y (r1 = 1) > + * and CPU2 does not observe CPU0's store to X (r2 = 0); see the comments > + * preceding the calls to smp_rmb() in try_to_wake_up() for similar > + * snippets but "projected" onto two CPUs. Maybe explicitly note that 2) is the RCsc lock upgrade. > * Since most load-store architectures implement ACQUIRE with an smp_mb() after > * the LL/SC loop, they need no further barriers. Similarly all our TSO > diff --git a/kernel/sched/core.c b/kernel/sched/core.c > index da8f12119a127..ec9ef0aec71ac 100644 > +++ b/kernel/sched/core.c > @@ -1999,21 +1999,20 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) > * 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() > + * STORE p->on_rq = 1 LOAD p->state > + * UNLOCK rq->lock > + * > + * __schedule() (switch to task 'p') > + * LOCK rq->lock smp_rmb(); > + * smp_mb__after_spinlock(); > + * UNLOCK rq->lock > * > * [task p] > + * STORE p->state = UNINTERRUPTIBLE LOAD p->on_rq > * > + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in > + * __schedule(). See the comment for smp_mb__after_spinlock(). > */ > smp_rmb(); > if (p->on_rq && ttwu_remote(p, wake_flags)) > @@ -2027,15 +2026,17 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) > * One must be running (->on_cpu == 1) in order to remove oneself > * from the runqueue. > * > + * __schedule() (switch to task 'p') try_to_wake_up() > + * STORE p->on_cpu = 1 LOAD p->on_rq > + * UNLOCK rq->lock > + * > + * __schedule() (put 'p' to sleep) > + * LOCK rq->lock smp_rmb(); > + * smp_mb__after_spinlock(); > + * STORE p->on_rq = 0 LOAD p->on_cpu > * > + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in > + * __schedule(). See the comment for smp_mb__after_spinlock(). > */ > smp_rmb(); Ah yes, good. Ack!
WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <peterz@infradead.org> To: Andrea Parri <andrea.parri@amarulasolutions.com> Cc: linux-kernel@vger.kernel.org, linux-doc@vger.kernel.org, Ingo Molnar <mingo@redhat.com>, Will Deacon <will.deacon@arm.com>, Alan Stern <stern@rowland.harvard.edu>, Boqun Feng <boqun.feng@gmail.com>, Nicholas Piggin <npiggin@gmail.com>, David Howells <dhowells@redhat.com>, Jade Alglave <j.alglave@ucl.ac.uk>, Luc Maranget <luc.maranget@inria.fr>, "Paul E . McKenney" <paulmck@linux.vnet.ibm.com>, Akira Yokosawa <akiyks@gmail.com>, Daniel Lustig <dlustig@nvidia.com>, Jonathan Corbet <corbet@lwn.net>, Randy Dunlap <rdunlap@infradead.org>, Matthew Wilcox <willy@infradead.org> Subject: Re: [PATCH v2 2/3] locking: Clarify requirements for smp_mb__after_spinlock() Date: Mon, 2 Jul 2018 17:37:35 +0200 [thread overview] Message-ID: <20180702153735.GQ2494@hirez.programming.kicks-ass.net> (raw) In-Reply-To: <1530544315-14614-1-git-send-email-andrea.parri@amarulasolutions.com> On Mon, Jul 02, 2018 at 05:11:55PM +0200, Andrea Parri wrote: > /* > + * smp_mb__after_spinlock() provides the equivalent of a full memory barrier > + * between program-order earlier lock acquisitions and program-order later > + * memory accesses. > * > + * This guarantees that the following two properties hold: > * > + * 1) Given the snippet: > * > + * { X = 0; Y = 0; } > * > + * CPU0 CPU1 > * > + * WRITE_ONCE(X, 1); WRITE_ONCE(Y, 1); > + * spin_lock(S); smp_mb(); > + * smp_mb__after_spinlock(); r1 = READ_ONCE(X); > + * r0 = READ_ONCE(Y); > + * spin_unlock(S); > * > + * it is forbidden that CPU0 does not observe CPU1's store to Y (r0 = 0) > + * and CPU1 does not observe CPU0's store to X (r1 = 0); see the comments > + * preceding the call to smp_mb__after_spinlock() in __schedule() and in > + * try_to_wake_up(). > + * > + * 2) Given the snippet: > + * > + * { X = 0; Y = 0; } > + * > + * CPU0 CPU1 CPU2 > + * > + * spin_lock(S); spin_lock(S); r1 = READ_ONCE(Y); > + * WRITE_ONCE(X, 1); smp_mb__after_spinlock(); smp_rmb(); > + * spin_unlock(S); r0 = READ_ONCE(X); r2 = READ_ONCE(X); > + * WRITE_ONCE(Y, 1); > + * spin_unlock(S); > + * > + * it is forbidden that CPU0's critical section executes before CPU1's > + * critical section (r0 = 1), CPU2 observes CPU1's store to Y (r1 = 1) > + * and CPU2 does not observe CPU0's store to X (r2 = 0); see the comments > + * preceding the calls to smp_rmb() in try_to_wake_up() for similar > + * snippets but "projected" onto two CPUs. Maybe explicitly note that 2) is the RCsc lock upgrade. > * Since most load-store architectures implement ACQUIRE with an smp_mb() after > * the LL/SC loop, they need no further barriers. Similarly all our TSO > diff --git a/kernel/sched/core.c b/kernel/sched/core.c > index da8f12119a127..ec9ef0aec71ac 100644 > +++ b/kernel/sched/core.c > @@ -1999,21 +1999,20 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) > * 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() > + * STORE p->on_rq = 1 LOAD p->state > + * UNLOCK rq->lock > + * > + * __schedule() (switch to task 'p') > + * LOCK rq->lock smp_rmb(); > + * smp_mb__after_spinlock(); > + * UNLOCK rq->lock > * > * [task p] > + * STORE p->state = UNINTERRUPTIBLE LOAD p->on_rq > * > + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in > + * __schedule(). See the comment for smp_mb__after_spinlock(). > */ > smp_rmb(); > if (p->on_rq && ttwu_remote(p, wake_flags)) > @@ -2027,15 +2026,17 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) > * One must be running (->on_cpu == 1) in order to remove oneself > * from the runqueue. > * > + * __schedule() (switch to task 'p') try_to_wake_up() > + * STORE p->on_cpu = 1 LOAD p->on_rq > + * UNLOCK rq->lock > + * > + * __schedule() (put 'p' to sleep) > + * LOCK rq->lock smp_rmb(); > + * smp_mb__after_spinlock(); > + * STORE p->on_rq = 0 LOAD p->on_cpu > * > + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in > + * __schedule(). See the comment for smp_mb__after_spinlock(). > */ > smp_rmb(); Ah yes, good. Ack! -- To unsubscribe from this list: send the line "unsubscribe linux-doc" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2018-07-02 15:38 UTC|newest] Thread overview: 40+ messages / expand[flat|nested] mbox.gz Atom feed top 2018-06-28 10:41 [PATCH 0/3] sched/locking/doc: Miscellaneous fixes Andrea Parri 2018-06-28 10:41 ` Andrea Parri 2018-06-28 10:41 ` [PATCH 1/3] sched: Use smp_mb() in wake_woken_function() Andrea Parri 2018-06-28 10:41 ` Andrea Parri 2018-06-28 10:41 ` [PATCH 2/3] locking: Clarify requirements for smp_mb__after_spinlock() Andrea Parri 2018-06-28 10:41 ` Andrea Parri 2018-06-28 13:02 ` Matthew Wilcox 2018-06-28 13:02 ` Matthew Wilcox 2018-06-28 13:10 ` Andrea Parri 2018-06-28 13:10 ` Andrea Parri 2018-06-28 13:49 ` Alan Stern 2018-06-28 13:49 ` Alan Stern 2018-06-28 13:52 ` Andrea Parri 2018-06-28 13:52 ` Andrea Parri 2018-06-28 15:05 ` Peter Zijlstra 2018-06-28 15:05 ` Peter Zijlstra 2018-06-28 17:30 ` Andrea Parri 2018-06-28 17:30 ` Andrea Parri 2018-07-02 12:50 ` Peter Zijlstra 2018-07-02 12:50 ` Peter Zijlstra 2018-07-02 15:11 ` [PATCH v2 " Andrea Parri 2018-07-02 15:11 ` Andrea Parri 2018-07-02 15:37 ` Peter Zijlstra [this message] 2018-07-02 15:37 ` Peter Zijlstra 2018-07-03 8:49 ` Andrea Parri 2018-07-03 8:49 ` Andrea Parri 2018-07-03 14:53 ` [PATCH v3 " Andrea Parri 2018-07-03 14:53 ` Andrea Parri 2018-07-03 15:39 ` Paul E. McKenney 2018-07-03 15:39 ` Paul E. McKenney 2018-07-03 17:07 ` Andrea Parri 2018-07-03 17:07 ` Andrea Parri 2018-06-28 10:41 ` [PATCH 3/3] doc: Update wake_up() & co. memory-barrier guarantees Andrea Parri 2018-06-28 10:41 ` Andrea Parri 2018-07-05 22:28 ` [PATCH 0/3] sched/locking/doc: Miscellaneous fixes Andrea Parri 2018-07-05 22:28 ` Andrea Parri 2018-07-06 10:36 ` Peter Zijlstra 2018-07-06 10:36 ` Peter Zijlstra 2018-07-06 14:43 ` Paul E. McKenney 2018-07-06 14:43 ` Paul E. McKenney
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=20180702153735.GQ2494@hirez.programming.kicks-ass.net \ --to=peterz@infradead.org \ --cc=akiyks@gmail.com \ --cc=andrea.parri@amarulasolutions.com \ --cc=boqun.feng@gmail.com \ --cc=corbet@lwn.net \ --cc=dhowells@redhat.com \ --cc=dlustig@nvidia.com \ --cc=j.alglave@ucl.ac.uk \ --cc=linux-doc@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=luc.maranget@inria.fr \ --cc=mingo@redhat.com \ --cc=npiggin@gmail.com \ --cc=paulmck@linux.vnet.ibm.com \ --cc=rdunlap@infradead.org \ --cc=stern@rowland.harvard.edu \ --cc=will.deacon@arm.com \ --cc=willy@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: 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.