From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-patches-return-320956-listarch-gcc-patches=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 19871 invoked by alias); 14 Jun 2012 20:13:21 -0000
Received: (qmail 19855 invoked by uid 22791); 14 Jun 2012 20:13:18 -0000
X-SWARE-Spam-Status: No, hits=-3.5 required=5.0	tests=BAYES_00,FREEMAIL_FROM,KHOP_THREADED,RCVD_IN_DNSWL_NONE,RCVD_IN_HOSTKARMA_NO,T_RP_MATCHES_RCVD
X-Spam-Check-By: sourceware.org
Received: from qmta01.westchester.pa.mail.comcast.net (HELO qmta01.westchester.pa.mail.comcast.net) (76.96.62.16)    by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 14 Jun 2012 20:13:03 +0000
Received: from omta02.westchester.pa.mail.comcast.net ([76.96.62.19])	by qmta01.westchester.pa.mail.comcast.net with comcast	id NL7E1j0020QuhwU51LD2ew; Thu, 14 Jun 2012 20:13:02 +0000
Received: from bag6-1-pt.tunnel.tserv3.fmt2.ipv6.he.net ([IPv6:2001:470:1f04:ae1::2])	by omta02.westchester.pa.mail.comcast.net with comcast	id NLD11j0010P3DwE3NLD2Pi; Thu, 14 Jun 2012 20:13:02 +0000
Subject: Re: [PATCH] ARM: exclude fixed_regs for stack-alignment save/restore
Mime-Version: 1.0 (Apple Message framework v1084)
Content-Type: text/plain; charset=us-ascii
From: Mike Stump <mikestump@comcast.net>
In-Reply-To: <x57jboklg6f9.fsf@frobland.mtv.corp.google.com>
Date: Thu, 14 Jun 2012 20:22:00 -0000
Cc: gcc-patches@gcc.gnu.org
Content-Transfer-Encoding: quoted-printable
Message-Id: <AC3806B1-20EB-4444-A7BC-D74EBF470D57@comcast.net>
References: <x57jboklg6f9.fsf@frobland.mtv.corp.google.com>
To: Roland McGrath <mcgrathr@google.com>
X-IsSubscribed: yes
Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Id: <gcc-patches.gcc.gnu.org>
List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
List-Post: <mailto:gcc-patches@gcc.gnu.org>
List-Help: <mailto:gcc-patches-help@gcc.gnu.org>
Sender: gcc-patches-owner@gcc.gnu.org
X-SW-Source: 2012-06/txt/msg00961.txt.bz2

On Jun 14, 2012, at 10:16 AM, Roland McGrath wrote:
> But if e.g. I use -ffixed-r9 then I think it's a reasonable expectation
> that no code is generated that touches r9 in any way, shape, or form.

Also,  I can't help but wonder if global_regs is respected.  In theory, peo=
ple are allowed to declare global registers, and nothing should be stopping=
 them, though, this is abi breaking, and one does need to recompile the wor=
ld as I recall to use it, so, most people don't use it and can't use it, bu=
t the bare metal people can.

Your change looks good to me.

I'll note in passing that cse.c does:

/* Determine whether register number N is considered a fixed register for t=
he=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20=20
   purpose of approximating register costs.=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20
   It is desirable to replace other regs with fixed regs, to reduce need fo=
r=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20=20=20
   non-fixed hard regs.=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=
=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20=20
   A reg wins if it is either the frame pointer or designated as fixed.  */
#define FIXED_REGNO_P(N)  \
  ((N) =3D=3D FRAME_POINTER_REGNUM || (N) =3D=3D HARD_FRAME_POINTER_REGNUM \
   || fixed_regs[N] || global_regs[N])