From: Ingo Molnar <mingo@kernel.org> To: Tim Chen <tim.c.chen@linux.intel.com> Cc: Ingo Molnar <mingo@elte.hu>, Andrew Morton <akpm@linux-foundation.org>, Linus Torvalds <torvalds@linux-foundation.org>, Andrea Arcangeli <aarcange@redhat.com>, Alex Shi <alex.shi@linaro.org>, Andi Kleen <andi@firstfloor.org>, Michel Lespinasse <walken@google.com>, Davidlohr Bueso <davidlohr.bueso@hp.com>, Matthew R Wilcox <matthew.r.wilcox@intel.com>, Dave Hansen <dave.hansen@intel.com>, Peter Zijlstra <a.p.zijlstra@chello.nl>, Rik van Riel <riel@redhat.com>, Peter Hurley <peter@hurleysoftware.com>, "Paul E.McKenney" <paulmck@linux.vnet.ibm.com>, Jason Low <jason.low2@hp.com>, Waiman Long <Waiman.Long@hp.com>, linux-kernel@vger.kernel.org, linux-mm <linux-mm@kvack.org> Subject: Re: [PATCH v8 0/9] rwsem performance optimizations Date: Thu, 10 Oct 2013 09:54:44 +0200 [thread overview] Message-ID: <20131010075444.GD17990@gmail.com> (raw) In-Reply-To: <1381336441.11046.128.camel@schen9-DESK> * Tim Chen <tim.c.chen@linux.intel.com> wrote: > The throughput of pure mmap with mutex is below vs pure mmap is below: > > % change in performance of the mmap with pthread-mutex vs pure mmap > #threads vanilla all rwsem without optspin > patches > 1 3.0% -1.0% -1.7% > 5 7.2% -26.8% 5.5% > 10 5.2% -10.6% 22.1% > 20 6.8% 16.4% 12.5% > 40 -0.2% 32.7% 0.0% > > So with mutex, the vanilla kernel and the one without optspin both run > faster. This is consistent with what Peter reported. With optspin, the > picture is more mixed, with lower throughput at low to moderate number > of threads and higher throughput with high number of threads. So, going back to your orignal table: > % change in performance of the mmap with pthread-mutex vs pure mmap > #threads vanilla all without optspin > 1 3.0% -1.0% -1.7% > 5 7.2% -26.8% 5.5% > 10 5.2% -10.6% 22.1% > 20 6.8% 16.4% 12.5% > 40 -0.2% 32.7% 0.0% > > In general, vanilla and no-optspin case perform better with > pthread-mutex. For the case with optspin, mmap with pthread-mutex is > worse at low to moderate contention and better at high contention. it appears that 'without optspin' appears to be a pretty good choice - if it wasn't for that '1 thread' number, which, if I correctly assume is the uncontended case, is one of the most common usecases ... How can the single-threaded case get slower? None of the patches should really cause noticeable overhead in the non-contended case. That looks weird. It would also be nice to see the 2, 3, 4 thread numbers - those are the most common contention scenarios in practice - where do we see the first improvement in performance? Also, it would be nice to include a noise/sttdev figure, it's really hard to tell whether -1.7% is statistically significant. Thanks, Ingo
WARNING: multiple messages have this Message-ID (diff)
From: Ingo Molnar <mingo@kernel.org> To: Tim Chen <tim.c.chen@linux.intel.com> Cc: Ingo Molnar <mingo@elte.hu>, Andrew Morton <akpm@linux-foundation.org>, Linus Torvalds <torvalds@linux-foundation.org>, Andrea Arcangeli <aarcange@redhat.com>, Alex Shi <alex.shi@linaro.org>, Andi Kleen <andi@firstfloor.org>, Michel Lespinasse <walken@google.com>, Davidlohr Bueso <davidlohr.bueso@hp.com>, Matthew R Wilcox <matthew.r.wilcox@intel.com>, Dave Hansen <dave.hansen@intel.com>, Peter Zijlstra <a.p.zijlstra@chello.nl>, Rik van Riel <riel@redhat.com>, Peter Hurley <peter@hurleysoftware.com>, "Paul E.McKenney" <paulmck@linux.vnet.ibm.com>, Jason Low <jason.low2@hp.com>, Waiman Long <Waiman.Long@hp.com>, linux-kernel@vger.kernel.org, linux-mm <linux-mm@kvack.org> Subject: Re: [PATCH v8 0/9] rwsem performance optimizations Date: Thu, 10 Oct 2013 09:54:44 +0200 [thread overview] Message-ID: <20131010075444.GD17990@gmail.com> (raw) In-Reply-To: <1381336441.11046.128.camel@schen9-DESK> * Tim Chen <tim.c.chen@linux.intel.com> wrote: > The throughput of pure mmap with mutex is below vs pure mmap is below: > > % change in performance of the mmap with pthread-mutex vs pure mmap > #threads vanilla all rwsem without optspin > patches > 1 3.0% -1.0% -1.7% > 5 7.2% -26.8% 5.5% > 10 5.2% -10.6% 22.1% > 20 6.8% 16.4% 12.5% > 40 -0.2% 32.7% 0.0% > > So with mutex, the vanilla kernel and the one without optspin both run > faster. This is consistent with what Peter reported. With optspin, the > picture is more mixed, with lower throughput at low to moderate number > of threads and higher throughput with high number of threads. So, going back to your orignal table: > % change in performance of the mmap with pthread-mutex vs pure mmap > #threads vanilla all without optspin > 1 3.0% -1.0% -1.7% > 5 7.2% -26.8% 5.5% > 10 5.2% -10.6% 22.1% > 20 6.8% 16.4% 12.5% > 40 -0.2% 32.7% 0.0% > > In general, vanilla and no-optspin case perform better with > pthread-mutex. For the case with optspin, mmap with pthread-mutex is > worse at low to moderate contention and better at high contention. it appears that 'without optspin' appears to be a pretty good choice - if it wasn't for that '1 thread' number, which, if I correctly assume is the uncontended case, is one of the most common usecases ... How can the single-threaded case get slower? None of the patches should really cause noticeable overhead in the non-contended case. That looks weird. It would also be nice to see the 2, 3, 4 thread numbers - those are the most common contention scenarios in practice - where do we see the first improvement in performance? Also, it would be nice to include a noise/sttdev figure, it's really hard to tell whether -1.7% is statistically significant. Thanks, Ingo -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2013-10-10 7:54 UTC|newest] Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top [not found] <cover.1380748401.git.tim.c.chen@linux.intel.com> 2013-10-02 22:38 ` [PATCH v8 0/9] rwsem performance optimizations Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-03 7:32 ` Ingo Molnar 2013-10-03 7:32 ` Ingo Molnar 2013-10-07 22:57 ` Tim Chen 2013-10-07 22:57 ` Tim Chen 2013-10-09 6:15 ` Ingo Molnar 2013-10-09 6:15 ` Ingo Molnar 2013-10-09 7:28 ` Peter Zijlstra 2013-10-09 7:28 ` Peter Zijlstra 2013-10-10 3:14 ` Linus Torvalds 2013-10-10 3:14 ` Linus Torvalds 2013-10-10 5:03 ` Davidlohr Bueso 2013-10-10 5:03 ` Davidlohr Bueso 2013-10-09 16:34 ` Tim Chen 2013-10-09 16:34 ` Tim Chen 2013-10-10 7:54 ` Ingo Molnar [this message] 2013-10-10 7:54 ` Ingo Molnar 2013-10-16 0:09 ` Tim Chen 2013-10-16 0:09 ` Tim Chen 2013-10-16 6:55 ` Ingo Molnar 2013-10-16 6:55 ` Ingo Molnar 2013-10-16 18:28 ` Tim Chen 2013-10-16 18:28 ` Tim Chen 2013-11-04 22:36 ` Tim Chen 2013-11-04 22:36 ` Tim Chen 2013-10-16 21:55 ` Tim Chen 2013-10-16 21:55 ` Tim Chen 2013-10-18 6:52 ` Ingo Molnar 2013-10-18 6:52 ` Ingo Molnar 2013-10-02 22:38 ` [PATCH v8 1/9] rwsem: check the lock before cpmxchg in down_write_trylock Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-02 22:38 ` [PATCH v8 2/9] rwsem: remove 'out' label in do_wake Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-02 22:38 ` [PATCH v8 3/9] rwsem: remove try_reader_grant label do_wake Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-02 22:38 ` [PATCH v8 4/9] rwsem/wake: check lock before do atomic update Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-02 22:38 ` [PATCH v8 5/9] MCS Lock: Restructure the MCS lock defines and locking code into its own file Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-08 19:51 ` Rafael Aquini 2013-10-08 19:51 ` Rafael Aquini 2013-10-08 20:34 ` Tim Chen 2013-10-08 20:34 ` Tim Chen 2013-10-08 21:31 ` Rafael Aquini 2013-10-08 21:31 ` Rafael Aquini 2013-10-02 22:38 ` [PATCH v8 6/9] MCS Lock: optimizations and extra comments Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-02 22:38 ` [PATCH v8 7/9] MCS Lock: Barrier corrections Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-02 22:38 ` [PATCH v8 8/9] rwsem: do optimistic spinning for writer lock acquisition Tim Chen 2013-10-02 22:38 ` Tim Chen 2013-10-02 22:38 ` [PATCH v8 9/9] rwsem: reduce spinlock contention in wakeup code path Tim Chen 2013-10-02 22:38 ` Tim Chen
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=20131010075444.GD17990@gmail.com \ --to=mingo@kernel.org \ --cc=Waiman.Long@hp.com \ --cc=a.p.zijlstra@chello.nl \ --cc=aarcange@redhat.com \ --cc=akpm@linux-foundation.org \ --cc=alex.shi@linaro.org \ --cc=andi@firstfloor.org \ --cc=dave.hansen@intel.com \ --cc=davidlohr.bueso@hp.com \ --cc=jason.low2@hp.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=matthew.r.wilcox@intel.com \ --cc=mingo@elte.hu \ --cc=paulmck@linux.vnet.ibm.com \ --cc=peter@hurleysoftware.com \ --cc=riel@redhat.com \ --cc=tim.c.chen@linux.intel.com \ --cc=torvalds@linux-foundation.org \ --cc=walken@google.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.