From: Linus Torvalds <torvalds@linux-foundation.org> To: Will Deacon <will.deacon@arm.com> Cc: Luc Maranget <luc.maranget@inria.fr>, Daniel Lustig <dlustig@nvidia.com>, Peter Zijlstra <peterz@infradead.org>, "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>, Andrea Parri <parri.andrea@gmail.com>, Linux Kernel Mailing List <linux-kernel@vger.kernel.org>, Palmer Dabbelt <palmer@sifive.com>, Albert Ou <albert@sifive.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>, Akira Yokosawa <akiyks@gmail.com>, Ingo Molnar <mingo@kernel.org>, linux-riscv@lists.infradead.org Subject: Re: [RFC PATCH] riscv/locking: Strengthen spin_lock() and spin_unlock() Date: Mon, 26 Feb 2018 09:00:43 -0800 [thread overview] Message-ID: <CA+55aFyT=+8m0mmpeegP8LOVtkCYv3O2vyO-ZcQdLAR=s309jg@mail.gmail.com> (raw) In-Reply-To: <20180226162426.GB17158@arm.com> On Mon, Feb 26, 2018 at 8:24 AM, Will Deacon <will.deacon@arm.com> wrote: > > Strictly speaking, that's not what we've got implemented on arm64: only > the read part of the RmW has Acquire semantics, but there is a total > order on the lock/unlock operations for the lock. Hmm. I thought we had exactly that bug on some architecture with the queued spinlocks, and people decided it was wrong. But it's possible that I mis-remember, and that we decided it was ok after all. > spin_lock(&lock); > WRITE_ONCE(foo, 42); > > then another CPU could do: > > if (smp_load_acquire(&foo) == 42) > BUG_ON(!spin_is_locked(&lock)); > > and that could fire. Is that relied on somewhere? I have a distinct memory that we said the spinlock write is seen in order, wrt the writes inside the spinlock, and the reason was something very similar to the above, except that "spin_is_locked()" was about our spin_unlock_wait(). Because we had something very much like the above in the exit path, where we would look at some state and do "spin_unlock_wait()" and expect to be guaranteed to be the last user after that. But a few months ago we obviously got rid of spin_unlock_wait exactly because people were worried about the semantics. So maybe I just remember an older issue that simply became a non-issue with that. Linus
WARNING: multiple messages have this Message-ID (diff)
From: torvalds@linux-foundation.org (Linus Torvalds) To: linux-riscv@lists.infradead.org Subject: [RFC PATCH] riscv/locking: Strengthen spin_lock() and spin_unlock() Date: Mon, 26 Feb 2018 09:00:43 -0800 [thread overview] Message-ID: <CA+55aFyT=+8m0mmpeegP8LOVtkCYv3O2vyO-ZcQdLAR=s309jg@mail.gmail.com> (raw) In-Reply-To: <20180226162426.GB17158@arm.com> On Mon, Feb 26, 2018 at 8:24 AM, Will Deacon <will.deacon@arm.com> wrote: > > Strictly speaking, that's not what we've got implemented on arm64: only > the read part of the RmW has Acquire semantics, but there is a total > order on the lock/unlock operations for the lock. Hmm. I thought we had exactly that bug on some architecture with the queued spinlocks, and people decided it was wrong. But it's possible that I mis-remember, and that we decided it was ok after all. > spin_lock(&lock); > WRITE_ONCE(foo, 42); > > then another CPU could do: > > if (smp_load_acquire(&foo) == 42) > BUG_ON(!spin_is_locked(&lock)); > > and that could fire. Is that relied on somewhere? I have a distinct memory that we said the spinlock write is seen in order, wrt the writes inside the spinlock, and the reason was something very similar to the above, except that "spin_is_locked()" was about our spin_unlock_wait(). Because we had something very much like the above in the exit path, where we would look at some state and do "spin_unlock_wait()" and expect to be guaranteed to be the last user after that. But a few months ago we obviously got rid of spin_unlock_wait exactly because people were worried about the semantics. So maybe I just remember an older issue that simply became a non-issue with that. Linus
next prev parent reply other threads:[~2018-02-26 17:00 UTC|newest] Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top 2018-02-22 12:19 [RFC PATCH] riscv/locking: Strengthen spin_lock() and spin_unlock() Andrea Parri 2018-02-22 12:19 ` Andrea Parri 2018-02-22 12:44 ` Andrea Parri 2018-02-22 12:44 ` Andrea Parri 2018-02-22 13:40 ` Peter Zijlstra 2018-02-22 13:40 ` Peter Zijlstra 2018-02-22 14:12 ` Andrea Parri 2018-02-22 14:12 ` Andrea Parri 2018-02-22 17:27 ` Daniel Lustig 2018-02-22 17:27 ` Daniel Lustig 2018-02-22 18:13 ` Paul E. McKenney 2018-02-22 18:13 ` Paul E. McKenney 2018-02-22 18:27 ` Peter Zijlstra 2018-02-22 18:27 ` Peter Zijlstra 2018-02-22 19:47 ` Daniel Lustig 2018-02-22 19:47 ` Daniel Lustig 2018-02-23 11:16 ` Andrea Parri 2018-02-23 11:16 ` Andrea Parri 2018-02-26 10:39 ` Will Deacon 2018-02-26 10:39 ` Will Deacon 2018-02-26 14:21 ` Luc Maranget 2018-02-26 14:21 ` Luc Maranget 2018-02-26 16:06 ` Linus Torvalds 2018-02-26 16:06 ` Linus Torvalds 2018-02-26 16:24 ` Will Deacon 2018-02-26 16:24 ` Will Deacon 2018-02-26 17:00 ` Linus Torvalds [this message] 2018-02-26 17:00 ` Linus Torvalds 2018-02-26 17:10 ` Will Deacon 2018-02-26 17:10 ` Will Deacon 2018-03-06 13:00 ` Peter Zijlstra 2018-03-06 13:00 ` Peter Zijlstra 2018-02-27 5:06 ` Boqun Feng 2018-02-27 5:06 ` Boqun Feng 2018-02-27 10:16 ` Boqun Feng 2018-02-27 10:16 ` Boqun Feng 2018-03-01 15:11 ` Andrea Parri 2018-03-01 15:11 ` Andrea Parri 2018-03-01 21:54 ` Palmer Dabbelt 2018-03-01 21:54 ` Palmer Dabbelt 2018-03-01 22:21 ` Daniel Lustig 2018-03-01 22:21 ` Daniel Lustig 2018-02-22 20:02 ` Paul E. McKenney 2018-02-22 20:02 ` Paul E. McKenney 2018-02-22 18:21 ` Peter Zijlstra 2018-02-22 18:21 ` Peter Zijlstra
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='CA+55aFyT=+8m0mmpeegP8LOVtkCYv3O2vyO-ZcQdLAR=s309jg@mail.gmail.com' \ --to=torvalds@linux-foundation.org \ --cc=akiyks@gmail.com \ --cc=albert@sifive.com \ --cc=boqun.feng@gmail.com \ --cc=dhowells@redhat.com \ --cc=dlustig@nvidia.com \ --cc=j.alglave@ucl.ac.uk \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-riscv@lists.infradead.org \ --cc=luc.maranget@inria.fr \ --cc=mingo@kernel.org \ --cc=npiggin@gmail.com \ --cc=palmer@sifive.com \ --cc=parri.andrea@gmail.com \ --cc=paulmck@linux.vnet.ibm.com \ --cc=peterz@infradead.org \ --cc=stern@rowland.harvard.edu \ --cc=will.deacon@arm.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: 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.