From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 22394 invoked by alias); 26 Aug 2002 20:51:51 -0000 Mailing-List: contact gdb-help@sources.redhat.com; run by ezmlm Precedence: bulk List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sources.redhat.com Received: (qmail 22387 invoked from network); 26 Aug 2002 20:51:51 -0000 Received: from unknown (HELO crack.them.org) (65.125.64.184) by sources.redhat.com with SMTP; 26 Aug 2002 20:51:51 -0000 Received: from nevyn.them.org ([66.93.61.169] ident=mail) by crack.them.org with asmtp (Exim 3.12 #1 (Debian)) id 17jRmD-0001KX-00; Mon, 26 Aug 2002 16:52:01 -0500 Received: from drow by nevyn.them.org with local (Exim 3.35 #1 (Debian)) id 17jQqw-0003bt-00; Mon, 26 Aug 2002 16:52:50 -0400 Date: Mon, 26 Aug 2002 13:51:00 -0000 From: Daniel Jacobowitz To: Mark Kettenis Cc: Andrew Cagney , gdb@sources.redhat.com Subject: Re: i386 register groups? Message-ID: <20020826205250.GA12224@nevyn.them.org> Mail-Followup-To: Mark Kettenis , Andrew Cagney , gdb@sources.redhat.com References: <3D6A75EC.2090807@ges.redhat.com> <20020826184705.GA6735@nevyn.them.org> <3D6A81CD.10207@ges.redhat.com> <86lm6tfiv7.fsf@elgar.kettenis.dyndns.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <86lm6tfiv7.fsf@elgar.kettenis.dyndns.org> User-Agent: Mutt/1.5.1i X-SW-Source: 2002-08/txt/msg00335.txt.bz2 On Mon, Aug 26, 2002 at 10:46:52PM +0200, Mark Kettenis wrote: > Andrew Cagney writes: > > > > On Mon, Aug 26, 2002 at 02:39:40PM -0400, Andrew Cagney wrote: > > > > > >> mxcsr 40 40 304 4 int general,all > > > > > > > > > With the other MMX registers, perhaps? > > > > Done (I wouldn't know an MMX register if I tripped over it :-). > > And Daniel doesn't either ;-) (sorry Daniel). mxcsr is the SSE > control/status register, so the sse group seems more appropriate to > me ;-). Can't win 'em all. > And while I'm at it, I don't think orig_eax belongs in the "generic" > group. It's some sort of OS-specific pseudo-register. We might > create a special register group for it, but putting it only in the > "all" group is fine by me. Fiddling with it is quite useless (and GDB > does it for you if it's needed). Agreed, only "all". > And what do people think about moving the segment registers out of the > "general" group into their own group? If you don't program at the OS > level, you're not very likely to need them (although on Linux you > might be interested in %gs when debugging a multi-threaded program). > We could name the group "segment". GCC can now generate code that uses these, with the TLS extensions; I think it'd be worthwhile to keep them. How useful having them is, though, I don't know... -- Daniel Jacobowitz MontaVista Software Debian GNU/Linux Developer