Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752565AbaFDVrQ (ORCPT ); Wed, 4 Jun 2014 17:47:16 -0400 Received: from g2t2354.austin.hp.com ([15.217.128.53]:45101 "EHLO g2t2354.austin.hp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751405AbaFDVrO (ORCPT ); Wed, 4 Jun 2014 17:47:14 -0400 Message-ID: <1401918431.2232.32.camel@j-VirtualBox> Subject: Re: [RFC PATCH 3/3] locking/mutex: Optimize mutex trylock slowpath From: Jason Low To: Davidlohr Bueso Cc: mingo@kernel.org, peterz@infradead.org, tglx@linutronix.de, akpm@linux-foundation.org, linux-kernel@vger.kernel.org, paulmck@linux.vnet.ibm.com, tim.c.chen@linux.intel.com, peter@hurleysoftware.com, riel@redhat.com, hpa@zytor.com, walken@google.com, Waiman.Long@hp.com, aswin@hp.com, scott.norton@hp.com, chegu_vinod@hp.com Date: Wed, 04 Jun 2014 14:47:11 -0700 In-Reply-To: <1401913714.13877.11.camel@buesod1.americas.hpqcorp.net> References: <1401908911-8947-1-git-send-email-jason.low2@hp.com> <1401908911-8947-4-git-send-email-jason.low2@hp.com> <1401913714.13877.11.camel@buesod1.americas.hpqcorp.net> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.2.3-0ubuntu6 Content-Transfer-Encoding: 7bit Mime-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 2014-06-04 at 13:28 -0700, Davidlohr Bueso wrote: > On Wed, 2014-06-04 at 12:08 -0700, Jason Low wrote: > > In __mutex_trylock_slowpath(), we acquire the wait_lock spinlock, > > xchg() lock->count with -1, then set lock->count back to 0 if there > > are no waiters, and return true if the prev lock count was 1. > > > > However, if we the mutex is already locked, then there may not be > ^^ leave that out. > > > much point in attempting the above operations. > > Isn't this redundant? I mean, if we enter the slowpath its because > __mutex_fastpath_trylock() already failed so we already know that the > lock is taken. This function is really just used as an alternative method of trylock for !__HAVE_ARCH_CMPXCHG. In that case, the fastpath can call directly into the slowpath function, without checking for if the lock is taken. > What kind of testing has this change been put through? Any advantages? > (ie: how many cycles are we saving here?), the trylock mechanism is > already pretty darn fast. While I did run tests with this patch, this particular patch shouldn't show benefits on my machine as it should be using the more efficient atomic_cmpxchg. The advantage is in !__HAVE_ARCH_CMPXCHG, where we would avoid taking a spinlock and 2 atomic operations when the mutex is already taken. Thanks. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/