[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CA+55aFyLp7DMp4jgJjV_Dj_KvMD--umZkqCDKBPbzH+sOZu24g@mail.gmail.com>
Date: Tue, 17 Jul 2018 13:13:51 -0700
From: Linus Torvalds <torvalds@...ux-foundation.org>
To: Alan Stern <stern@...land.harvard.edu>
Cc: Paul McKenney <paulmck@...ux.vnet.ibm.com>,
Michael Ellerman <mpe@...erman.id.au>,
Peter Zijlstra <peterz@...radead.org>,
andrea.parri@...rulasolutions.com,
Will Deacon <will.deacon@....com>,
Akira Yokosawa <akiyks@...il.com>,
Boqun Feng <boqun.feng@...il.com>,
Daniel Lustig <dlustig@...dia.com>,
David Howells <dhowells@...hat.com>,
Jade Alglave <j.alglave@....ac.uk>,
Luc Maranget <luc.maranget@...ia.fr>,
Nick Piggin <npiggin@...il.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] tools/memory-model: Add extra ordering for locks and
remove it for ordinary release/acquire
On Tue, Jul 17, 2018 at 12:37 PM Alan Stern <stern@...land.harvard.edu> wrote:
>
> Why not? Instructions are allowed to migrate _into_ critical sections,
> just not _out_ of them. So a store preceding the start of a spinlocked
> region can migrate in and be executed after a load that is inside the
> region.
Hmm, yes of course. But the isync instruction description I found also
talks about the previous instructions being "completed".
But yeah, that obviously can mean just "in the store buffer", not
actually ordered.
Linus
Powered by blists - more mailing lists