From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9067 invoked by alias); 19 Feb 2015 07:25:12 -0000 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 Received: (qmail 9056 invoked by uid 89); 19 Feb 2015 07:25:11 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=0.3 required=5.0 tests=AWL,BAYES_00,FREEMAIL_ENVFROM_END_DIGIT,FREEMAIL_FROM,KAM_FROM_URIBL_PCCC,RCVD_IN_DNSWL_LOW,SPF_PASS autolearn=no version=3.3.2 X-HELO: mail-qg0-f46.google.com Received: from mail-qg0-f46.google.com (HELO mail-qg0-f46.google.com) (209.85.192.46) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES128-GCM-SHA256 encrypted) ESMTPS; Thu, 19 Feb 2015 07:25:10 +0000 Received: by mail-qg0-f46.google.com with SMTP id z107so4987931qgd.5 for ; Wed, 18 Feb 2015 23:25:08 -0800 (PST) X-Received: by 10.229.80.72 with SMTP id s8mr10424441qck.24.1424330708410; Wed, 18 Feb 2015 23:25:08 -0800 (PST) MIME-Version: 1.0 Received: by 10.229.168.4 with HTTP; Wed, 18 Feb 2015 23:24:48 -0800 (PST) In-Reply-To: <1384464221.2484.86.camel@ubuntu-sellcey> References: <1384464221.2484.86.camel@ubuntu-sellcey> From: Matt Turner Date: Thu, 19 Feb 2015 07:25:00 -0000 Message-ID: Subject: Re: [Patch, mips] Faster strcmp for mips To: Steve Ellcey Cc: libc-ports@sourceware.org Content-Type: text/plain; charset=UTF-8 X-IsSubscribed: yes X-SW-Source: 2015-02/txt/msg00000.txt.bz2 On Thu, Nov 14, 2013 at 1:23 PM, Steve Ellcey wrote: > > I have created an optimized strcmp routine for MIPS and would like to check > it in. I verified it by running 'make check' and I also ran 'make bench' > to check the performance. This function is slightly slower then the current > generic strcmp for strings that are not aligned, but much faster when strings > are aligned because it will then load and compare either 2, 4, or 8 bytes at a > time. > > This means it could be loading bytes beyond the end of the strings being > compared but it looks like other architecture specific strcmp functions > are also doing this optimization and the newlib version of strcmp also does > this. > > OK to checkin? > > I have attached the original and new bench-strcmp.out files in case anyone > wants to compare the old and new strcmp performance results. > > Steve Ellcey > sellcey@mips.com > > > 2013-11-14 Steve Ellcey > > * sysdeps/mips/strcmp.c: New. > > Doesn't seem that this was ever committed?