From: Vineet Gupta <Vineet.Gupta1@synopsys.com> To: Peter Zijlstra <peterz@infradead.org>, Will Deacon <will.deacon@arm.com>, Paul McKenney <paulmck@linux.vnet.ibm.com>, Boqun Feng <boqun.feng@gmail.com> Cc: <linux-kernel@vger.kernel.org>, Ingo Molnar <mingo@kernel.org>, "Thomas Gleixner" <tglx@linutronix.de>, <rkuo@codeaurora.org>, <james.hogan@imgtec.com>, <jejb@parisc-linux.org>, <davem@davemloft.net>, <cmetcalf@mellanox.com>, arcml <linux-snps-arc@lists.infradead.org> Subject: Re: [RFC][PATCH] atomic: Fix atomic_set_release() for 'funny' architectures Date: Fri, 9 Jun 2017 10:28:50 -0700 [thread overview] Message-ID: <ecffd3a6-167b-c0ed-0121-1a3a4141f799@synopsys.com> (raw) In-Reply-To: <20170609111305.bn4ca4uscbp6pgxn@hirez.programming.kicks-ass.net> On 06/09/2017 04:13 AM, Peter Zijlstra wrote: > On Fri, Jun 09, 2017 at 01:05:06PM +0200, Peter Zijlstra wrote: > >> The spinlock based atomics should be SC, that is, none of them appear to >> place extra barriers in atomic_cmpxchg() or any of the other SC atomic >> primitives and therefore seem to rely on their spinlock implementation >> being SC (I did not fully validate all that). > > So I did see that ARC and PARISC have 'superfluous' smp_mb() calls > around their spinlock implementation. > > That is, for spinlock semantics you only need one _after_ lock and one > _before_ unlock. But the atomic stuff relies on being SC and thus would > need one before and after both lock and unlock. Right we discussed this a while back: https://lkml.org/lkml/2015/6/11/276 At the time when I tried removing these extra barriers, hackbench regressed. I'm about to get a new quad core 1GHz chip (vs. the FPGA before) and will re-experiment. Likely we don't need it otherwise I will add a comment of this "feature" > But ARC could probably optimize (if they still care about that hardware) > by pulling out those barriers and putting it in the atomic > implementation. A bit confused here. Reading the lkml posting for this thread, you posted 2 patches, and they had to do with atomic_set() for EZChip platform which is really special (no ll/sc). The extra smp_mb() is related to ll/sc variants. Just tryign to make sure that we are talking 2 different things here :-) -Vineet
WARNING: multiple messages have this Message-ID (diff)
From: Vineet.Gupta1@synopsys.com (Vineet Gupta) To: linux-snps-arc@lists.infradead.org Subject: [RFC][PATCH] atomic: Fix atomic_set_release() for 'funny' architectures Date: Fri, 9 Jun 2017 10:28:50 -0700 [thread overview] Message-ID: <ecffd3a6-167b-c0ed-0121-1a3a4141f799@synopsys.com> (raw) In-Reply-To: <20170609111305.bn4ca4uscbp6pgxn@hirez.programming.kicks-ass.net> On 06/09/2017 04:13 AM, Peter Zijlstra wrote: > On Fri, Jun 09, 2017@01:05:06PM +0200, Peter Zijlstra wrote: > >> The spinlock based atomics should be SC, that is, none of them appear to >> place extra barriers in atomic_cmpxchg() or any of the other SC atomic >> primitives and therefore seem to rely on their spinlock implementation >> being SC (I did not fully validate all that). > > So I did see that ARC and PARISC have 'superfluous' smp_mb() calls > around their spinlock implementation. > > That is, for spinlock semantics you only need one _after_ lock and one > _before_ unlock. But the atomic stuff relies on being SC and thus would > need one before and after both lock and unlock. Right we discussed this a while back: https://lkml.org/lkml/2015/6/11/276 At the time when I tried removing these extra barriers, hackbench regressed. I'm about to get a new quad core 1GHz chip (vs. the FPGA before) and will re-experiment. Likely we don't need it otherwise I will add a comment of this "feature" > But ARC could probably optimize (if they still care about that hardware) > by pulling out those barriers and putting it in the atomic > implementation. A bit confused here. Reading the lkml posting for this thread, you posted 2 patches, and they had to do with atomic_set() for EZChip platform which is really special (no ll/sc). The extra smp_mb() is related to ll/sc variants. Just tryign to make sure that we are talking 2 different things here :-) -Vineet
next prev parent reply other threads:[~2017-06-09 17:29 UTC|newest] Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top 2017-06-09 9:24 [RFC][PATCH]: documentation,atomic: Add a new atomic_t document Peter Zijlstra 2017-06-09 11:05 ` [RFC][PATCH] atomic: Fix atomic_set_release() for 'funny' architectures Peter Zijlstra 2017-06-09 11:13 ` Peter Zijlstra 2017-06-09 17:28 ` Vineet Gupta [this message] 2017-06-09 17:28 ` Vineet Gupta 2017-06-09 18:49 ` Peter Zijlstra 2017-06-09 18:49 ` Peter Zijlstra 2017-06-09 18:58 ` James Bottomley 2017-06-09 14:03 ` Chris Metcalf 2017-08-10 12:10 ` [tip:locking/core] locking/atomic: " tip-bot for Peter Zijlstra 2017-06-09 15:44 ` [RFC][PATCH]: documentation,atomic: Add a new atomic_t document Will Deacon 2017-06-09 19:36 ` Peter Zijlstra 2017-06-11 13:56 ` Boqun Feng 2017-06-12 14:49 ` Peter Zijlstra 2017-06-13 6:39 ` Boqun Feng 2017-06-14 12:33 ` Will Deacon 2017-07-12 12:53 ` Boqun Feng 2017-07-12 13:08 ` Peter Zijlstra 2017-07-12 19:13 ` Paul E. McKenney 2017-07-26 11:53 ` [RFC][PATCH v3]: documentation,atomic: Add new documents Peter Zijlstra 2017-07-26 12:47 ` Boqun Feng 2017-07-31 9:05 ` Peter Zijlstra 2017-07-31 11:04 ` Boqun Feng 2017-07-31 17:43 ` Paul E. McKenney 2017-08-01 2:14 ` Boqun Feng 2017-08-01 9:01 ` Peter Zijlstra 2017-08-01 10:19 ` Will Deacon 2017-08-01 11:47 ` Peter Zijlstra 2017-08-01 12:17 ` Will Deacon 2017-08-01 12:52 ` Peter Zijlstra 2017-08-01 16:14 ` Paul E. McKenney 2017-08-01 16:42 ` Peter Zijlstra 2017-08-01 16:53 ` Will Deacon 2017-08-01 22:18 ` Paul E. McKenney 2017-08-02 8:46 ` Will Deacon 2017-08-01 18:37 ` Paul E. McKenney 2017-08-02 9:45 ` Will Deacon 2017-08-02 16:17 ` Paul E. McKenney 2017-08-03 14:05 ` Boqun Feng 2017-08-03 14:55 ` Paul E. McKenney 2017-08-03 16:12 ` Will Deacon 2017-08-03 16:58 ` Paul E. McKenney 2017-08-01 13:35 ` Paul E. McKenney 2017-07-26 16:28 ` Randy Dunlap 2017-06-09 18:15 ` [RFC][PATCH]: documentation,atomic: Add a new atomic_t document Randy Dunlap
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=ecffd3a6-167b-c0ed-0121-1a3a4141f799@synopsys.com \ --to=vineet.gupta1@synopsys.com \ --cc=boqun.feng@gmail.com \ --cc=cmetcalf@mellanox.com \ --cc=davem@davemloft.net \ --cc=james.hogan@imgtec.com \ --cc=jejb@parisc-linux.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-snps-arc@lists.infradead.org \ --cc=mingo@kernel.org \ --cc=paulmck@linux.vnet.ibm.com \ --cc=peterz@infradead.org \ --cc=rkuo@codeaurora.org \ --cc=tglx@linutronix.de \ --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.