From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9082 invoked by alias); 7 Mar 2010 19:10:05 -0000 Received: (qmail 9074 invoked by uid 22791); 7 Mar 2010 19:10:05 -0000 X-SWARE-Spam-Status: No, hits=-2.3 required=5.0 tests=AWL,BAYES_00 X-Spam-Check-By: sourceware.org Received: from mail.codesourcery.com (HELO mail.codesourcery.com) (38.113.113.100) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Sun, 07 Mar 2010 19:10:01 +0000 Received: (qmail 10666 invoked from network); 7 Mar 2010 19:09:59 -0000 Received: from unknown (HELO localhost) (froydnj@127.0.0.2) by mail.codesourcery.com with ESMTPA; 7 Mar 2010 19:09:59 -0000 Date: Sun, 07 Mar 2010 19:10:00 -0000 From: Nathan Froyd To: "H.J. Lu" Cc: Mark Kettenis , gdb-patches@sourceware.org Subject: Re: PATCH: 0/6 [2nd try]: Add AVX support Message-ID: <20100307190959.GY16726@codesourcery.com> References: <20100304180219.GA10826@intel.com> <20100306221634.GA21133@intel.com> <201003071416.o27EGPfu007140@glazunov.sibelius.xs4all.nl> <6dc9ffc81003070637q3e554ba4n2ee6053c2fd24511@mail.gmail.com> <6dc9ffc81003070831yf923b9bj83da56cb9c83a751@mail.gmail.com> <6dc9ffc81003070840x649dcc23qa088fddd69df7e54@mail.gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <6dc9ffc81003070840x649dcc23qa088fddd69df7e54@mail.gmail.com> User-Agent: Mutt/1.5.17+20080114 (2008-01-14) X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2010-03/txt/msg00278.txt.bz2 On Sun, Mar 07, 2010 at 08:40:10AM -0800, H.J. Lu wrote: > My first approach works here since XMM and YMM register have the same > register numbers. We can solve it with 2 alternatives: > > 1. Give a different DWARF register number for YMM register, > which is an incompatible ABI change. > 2. Implement YMM registers as a super set of XMM registers, which > is my first approach. The third alternative--again, what's adopted for the PPC SPE 64-bit registers--is to give %ymmNh their own DWARF register numbers. I suppose it's also ABI-incompatible, but it seems like it fits with your approach much better than either of the above alternatives. -Nathan