From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 1039) id B29003858022; Fri, 12 Nov 2021 18:32:55 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org B29003858022 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit From: H.J. Lu To: glibc-cvs@sourceware.org Subject: [glibc] Add LLL_MUTEX_READ_LOCK [BZ #28537] X-Act-Checkin: glibc X-Git-Author: H.J. Lu X-Git-Refname: refs/heads/master X-Git-Oldrev: 49302b8fdf9103b6fc0a398678668a22fa19574c X-Git-Newrev: d672a98a1af106bd68deb15576710cd61363f7a6 Message-Id: <20211112183255.B29003858022@sourceware.org> Date: Fri, 12 Nov 2021 18:32:55 +0000 (GMT) X-BeenThere: glibc-cvs@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Glibc-cvs mailing list List-Unsubscribe: , List-Archive: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 12 Nov 2021 18:32:55 -0000 https://sourceware.org/git/gitweb.cgi?p=glibc.git;h=d672a98a1af106bd68deb15576710cd61363f7a6 commit d672a98a1af106bd68deb15576710cd61363f7a6 Author: H.J. Lu Date: Tue Nov 2 18:33:07 2021 -0700 Add LLL_MUTEX_READ_LOCK [BZ #28537] CAS instruction is expensive. From the x86 CPU's point of view, getting a cache line for writing is more expensive than reading. See Appendix A.2 Spinlock in: https://www.intel.com/content/dam/www/public/us/en/documents/white-papers/xeon-lock-scaling-analysis-paper.pdf The full compare and swap will grab the cache line exclusive and cause excessive cache line bouncing. Add LLL_MUTEX_READ_LOCK to do an atomic load and skip CAS in spinlock loop if compare may fail to reduce cache line bouncing on contended locks. Reviewed-by: Szabolcs Nagy Diff: --- nptl/pthread_mutex_lock.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/nptl/pthread_mutex_lock.c b/nptl/pthread_mutex_lock.c index 59a28cff1a..762059b230 100644 --- a/nptl/pthread_mutex_lock.c +++ b/nptl/pthread_mutex_lock.c @@ -64,6 +64,11 @@ lll_mutex_lock_optimized (pthread_mutex_t *mutex) # define PTHREAD_MUTEX_VERSIONS 1 #endif +#ifndef LLL_MUTEX_READ_LOCK +# define LLL_MUTEX_READ_LOCK(mutex) \ + atomic_load_relaxed (&(mutex)->__data.__lock) +#endif + static int __pthread_mutex_lock_full (pthread_mutex_t *mutex) __attribute_noinline__; @@ -141,6 +146,8 @@ PTHREAD_MUTEX_LOCK (pthread_mutex_t *mutex) break; } atomic_spin_nop (); + if (LLL_MUTEX_READ_LOCK (mutex) != 0) + continue; } while (LLL_MUTEX_TRYLOCK (mutex) != 0);