From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 796 invoked by alias); 6 Jun 2012 11:55:16 -0000 Received: (qmail 786 invoked by uid 22791); 6 Jun 2012 11:55:15 -0000 X-SWARE-Spam-Status: No, hits=-3.0 required=5.0 tests=AWL,BAYES_00,KHOP_RCVD_UNTRUST,KHOP_THREADED,RCVD_IN_DNSWL_LOW,RCVD_IN_HOSTKARMA_YE,TW_WF,TW_XF X-Spam-Check-By: sourceware.org Received: from mail-we0-f175.google.com (HELO mail-we0-f175.google.com) (74.125.82.175) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 06 Jun 2012 11:55:03 +0000 Received: by werg55 with SMTP id g55so4774064wer.20 for ; Wed, 06 Jun 2012 04:55:01 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding:x-gm-message-state; bh=+a9OsQjbz5Gci5Vn/+4UgS7JdyNUjp/HQ8MajdyS92g=; b=YHRDe4YMQO+iu4twj1E8w0SWalA6zHyxGIkV6a+aD3iXVW7uAKqHBLxeWLpGcqHBV9 qsNSqdb0lt3W+IcwyWCV8ntreM8a/NA6bvmhDTz/CI53iyY0lrRwviLs9L8yafAVLeB3 ZvffEgbAMwwKc6iDzeRoXtYsChsIySY8CdvyHXsL+RSdpMWnLHcvS7joyh3drM/+9YE3 Cw4/YJ2Lc6P1hi2lEJn52EcjBQzRzfM7LTPyzlkBB0jkpsuC7m7KzJODpsVqm2PH6fSc J5jiJ2K+CXpedhHlSsvch3ZYZOaRPRcSUUCdo/6xGQiEx7U26c9FhnGZObieQuUzfQjG /RdQ== MIME-Version: 1.0 Received: by 10.216.194.196 with SMTP id m46mr3506437wen.197.1338983701573; Wed, 06 Jun 2012 04:55:01 -0700 (PDT) Received: by 10.216.132.97 with HTTP; Wed, 6 Jun 2012 04:55:01 -0700 (PDT) In-Reply-To: <1338264799-12374-1-git-send-email-mattst88@gmail.com> References: <1338264799-12374-1-git-send-email-mattst88@gmail.com> Date: Wed, 06 Jun 2012 11:59:00 -0000 Message-ID: Subject: Re: [PATCH ARM iWMMXt 0/5] Improve iWMMXt support From: Ramana Radhakrishnan To: Matt Turner Cc: gcc-patches@gcc.gnu.org, Richard Earnshaw , Nick Clifton , Paul Brook , Xinyu Qi Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable X-Gm-Message-State: ALoCoQkYammgbegv1ztjZjdFXa5vJW6CjWK1gg2iFb9jb1TwoKxp2sRJRopgi2Gd7hLFijGn9Ylb X-IsSubscribed: yes Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org X-SW-Source: 2012-06/txt/msg00364.txt.bz2 On 29 May 2012 05:13, Matt Turner wrote: > > This series was written by Marvell and sent by Xinyu Qi > a number of times in the last year. > > We (One Laptop per Child) need these patches for reasonable iWMMXt support > and performance. Without them, logical and shift intrinsics cause ICEs, > see PR 35294 and its duplicates 36798 and 36966. > > The software compositing library pixman uses MMX intrinsics to optimize > various compositing routines. The following are the minimum execution tim= es > of cairo-perf-trace graphics work loads without and with iWMMXt-optimized > pixman for the image and image16 backends (32-bpp and 16-bpp respectively= ). > > =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 image =A0 =A0 =A0= =A0 =A0 =A0 =A0 image16 > =A0 =A0 =A0 =A0 =A0 evolution =A0 33.492 -> =A029.590 =A0 =A030.334 -> = =A024.751 > firefox-planet-gnome =A0191.465 -> 173.835 =A0 211.297 -> 187.570 > gnome-system-monitor =A0 51.956 -> =A044.549 =A0 =A052.272 -> =A040.525 > =A0gnome-terminal-vim =A0 53.625 -> =A054.554 =A0 =A047.593 -> =A047.341 > =A0 =A0 =A0grads-heat-map =A0 =A04.439 -> =A0 4.165 =A0 =A0 4.548 -> =A0 = 4.624 > =A0 =A0 =A0 midori-zoomed =A0 38.033 -> =A028.500 =A0 =A038.576 -> =A026.= 937 > =A0 =A0 =A0 =A0 =A0 =A0 poppler =A0 41.096 -> =A031.949 =A0 =A041.230 -> = =A031.749 > =A0swfdec-giant-steps =A0 20.062 -> =A016.912 =A0 =A028.294 -> =A017.286 > =A0 =A0 =A0swfdec-youtube =A0 42.281 -> =A037.335 =A0 =A052.848 -> =A047.= 053 > =A0 xfce4-terminal-a1 =A0 64.311 -> =A051.011 =A0 =A062.592 -> =A051.191 > > We have cleaned up some white-space issues with the patches and fixed a > small bug in patch 4/5 since the last time they were posted in December > (added tandc,textrc,torc,torvsc to the "wtype" attribute) > > Please commit them for 4.8. You do not mention how these patches have been tested with trunk after you've rebased them - I understand that you are using them in your port but can you specify how these were tested and what the results looked like ? Ramana