From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 21970 invoked by alias); 28 Jun 2012 17:30:42 -0000 Received: (qmail 21948 invoked by uid 22791); 28 Jun 2012 17:30:39 -0000 X-SWARE-Spam-Status: No, hits=-3.1 required=5.0 tests=AWL,BAYES_00,KHOP_THREADED,T_RP_MATCHES_RCVD X-Spam-Check-By: sourceware.org Received: from usmamail.tilera.com (HELO USMAMAIL.TILERA.COM) (12.216.194.151) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 28 Jun 2012 17:30:26 +0000 Received: from [10.7.0.95] (10.9.0.23) by USMAExch2.tad.internal.tilera.com (10.3.0.33) with Microsoft SMTP Server id 14.0.694.0; Thu, 28 Jun 2012 13:30:24 -0400 Message-ID: <4FEC94AF.40301@tilera.com> Date: Thu, 28 Jun 2012 17:30:00 -0000 From: Chris Metcalf User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:13.0) Gecko/20120614 Thunderbird/13.0.1 MIME-Version: 1.0 To: Maxim Kuvyrkov CC: "Joseph S. Myers" , GLIBC Devel , , Tom de Vries Subject: Re: [PATCH] Optimize libc_lock_lock for MIPS XLP. References: <4FD9DB74.8080905@tilera.com> <40CBC472-71CC-4FF3-A452-073B76701215@codesourcery.com> <4FDAA190.3050706@tilera.com> <15EB7E17-5692-4221-A1B1-FC16EA236BFF@codesourcery.com> In-Reply-To: <15EB7E17-5692-4221-A1B1-FC16EA236BFF@codesourcery.com> Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit X-IsSubscribed: yes Mailing-List: contact libc-ports-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Post: List-Help: , Sender: libc-ports-owner@sourceware.org X-SW-Source: 2012-06/txt/msg00094.txt.bz2 On 6/27/2012 5:45 PM, Maxim Kuvyrkov wrote: > On 15/06/2012, at 2:49 PM, Maxim Kuvyrkov wrote: > >> > On 15/06/2012, at 2:44 PM, Chris Metcalf wrote: >> > >>> >> On 6/14/2012 9:20 PM, Maxim Kuvyrkov wrote: >> > ... >>>> >>> As I read it, in case of a contended lock __lll_lock_wait will reset the value of the lock to "2" before calling lll_futex_wait(). I agree that there is a timing window in which the other threads will see a value of the lock greater than "2", but the value will not get as high as hundreds or billions as it will be constantly reset to "2" in atomic_exchange in lll_lock_wait(). >>>> >>> >>>> >>> I do not see how threads will get into a busywait state, though. Would you please elaborate on that? >>> >> >>> >> You are correct. I was thinking the that the while loop had a cmpxchg, but >>> >> since it's just a straight-up exchange, the flow will be something like: >>> >> >>> >> - Fail to initially call lll_futex_wait() if the lock is contended >>> >> - Fall through to while loop >>> >> - Spin as long as the lock is contended enough that *futex > 2 >>> >> - Enter futex_wait >>> >> >>> >> So a little busy under high contention, but probably settles out reasonably >>> >> well. >> > > Attached is an improved patch that also optimizes __libc_lock_trylock using XLP's atomic instructions. > > The patch also removes unnecessary indirection step represented by new macros lll_add_lock, which is then used to define __libc_lock_lock, and defines __libc_lock_lock and __libc_lock_trylock directly in lowlevellock.h . This makes changes outside of ports/ trivial. > > Tested on MIPS XLP with no regressions. OK to apply for 2.17? It looks OK to me. I would want someone else to sign off on it before applying to 2.17. -- Chris Metcalf, Tilera Corp. http://www.tilera.com