From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7560 invoked by alias); 12 Jan 2011 19:33:20 -0000 Received: (qmail 7552 invoked by uid 22791); 12 Jan 2011 19:33:19 -0000 X-SWARE-Spam-Status: No, hits=1.3 required=5.0 tests=AWL,BAYES_05,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,RCVD_IN_DNSWL_LOW,RFC_ABUSE_POST X-Spam-Check-By: sourceware.org Received: from mail-bw0-f41.google.com (HELO mail-bw0-f41.google.com) (209.85.214.41) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 12 Jan 2011 19:33:14 +0000 Received: by bwz16 with SMTP id 16so918981bwz.0 for ; Wed, 12 Jan 2011 11:33:12 -0800 (PST) MIME-Version: 1.0 Received: by 10.204.57.197 with SMTP id d5mr1077886bkh.63.1294860791879; Wed, 12 Jan 2011 11:33:11 -0800 (PST) Received: by 10.204.79.12 with HTTP; Wed, 12 Jan 2011 11:33:10 -0800 (PST) In-Reply-To: <1292271927.3622.6.camel@localhost.localdomain> References: <1283448864.3854.56.camel@gargoyle> <1283526020.3854.77.camel@gargoyle> <1286218436.3679.1.camel@gargoyle> <1292271927.3622.6.camel@localhost.localdomain> Date: Tue, 18 Jan 2011 13:16:00 -0000 Message-ID: Subject: Re: [PATCH] powerpc: 405/440/464/476 support and optimizations From: Ryan Arnold To: libc-ports Cc: luisgpm@linux.vnet.ibm.com, Todd Iglehart , Josh Boyer , rsa@us.ibm.com Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable 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: 2011-01/txt/msg00004.txt.bz2 On Mon, Dec 13, 2010 at 2:25 PM, Ryan Arnold wrote: > Sorry for the delinquent response. =C2=A0This looks good to me and I thin= k it > should be checked in. > > I'd like for someone with a 405, 440, or 464 to test it further. =C2=A0As= far > as we know the code only uses instructions available on all of these > platforms. > > I'd like to stress that it was authored by Todd Iglehart > and contributed by IBM. =C2=A0Luis did the fixup and > authored the implies structure. > > Ryan S. Arnold I've checked this patch into glibc-ports under: commit # a72cc2b29d00207fd8e2ee4612502339a14816b6 Just a general note on configuration; some of these processors have a floating point unit but I believe all of them default to soft-fp. GLIBC configure won't recognize --with-cpu=3D476fp even though the compiler might recognize -mcpu=3D476fp. If you want to configure a hard-fp build just pass --with-cpu=3D476 --with-fp instead and a new Makefile fragment will make sure that -mhard-float is added to CFLAGS and ASFLAGS. Ryan S. Arnold