From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-return-89695-listarch-gcc=gcc.gnu.org@gcc.gnu.org>
Received: (qmail 10774 invoked by alias); 21 Jan 2004 15:29:53 -0000
Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm
Precedence: bulk
List-Archive: <http://gcc.gnu.org/ml/gcc/>
List-Post: <mailto:gcc@gcc.gnu.org>
List-Help: <http://gcc.gnu.org/ml/>
Sender: gcc-owner@gcc.gnu.org
Received: (qmail 10749 invoked from network); 21 Jan 2004 15:29:52 -0000
Received: from unknown (HELO localhost.localdomain) (212.143.109.38)
  by sources.redhat.com with SMTP; 21 Jan 2004 15:29:52 -0000
Received: from localhost (localhost.localdomain [127.0.0.1])
	by localhost.localdomain (8.12.8/8.12.8) with ESMTP id i0LFTnJM027110;
	Wed, 21 Jan 2004 17:29:50 +0200
From: xyzzy@hotpop.com
To: Jim Wilson <wilson@specifixinc.com>, GCC list <gcc@gcc.gnu.org>
Subject: Re: RTL/SYMREF/VOLATILE question
Date: Wed, 21 Jan 2004 15:33:00 -0000
User-Agent: KMail/1.5.4
References: <200401201428.54425.xyzzy@hotpop.com> <400DC917.9070405@specifixinc.com>
In-Reply-To: <400DC917.9070405@specifixinc.com>
MIME-Version: 1.0
Content-Type: Text/Plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Description: clearsigned data
Content-Disposition: inline
Message-Id: <200401211729.49662.xyzzy@hotpop.com>
X-SW-Source: 2004-01/txt/msg01650.txt.bz2

-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160

On Wednesday 21 January 2004 2:34 am, Jim Wilson wrote:
> xyzzy@hotpop.com wrote:
> > Is there a simple way to do what I want (that is, get the least amount =
of
> > indirection possible for a SYMREF and be optimization safe?)
>
> You haven't provided enough information.  None of your code samples will
> actually produce code for instance.  Just generating a symbol_ref does
> not do anything interesting.  You have to emit insns that use it.  How
> it gets optimized depends on what insns you emit.  Since you did not
> give us this info, there isn't much we can do.


Sorry...

In essence I am passing this rtx up the chain to the expand_va_arg code, so=
 I=20
don't KNOW what insns are emitted unless I do a lot of painful digging.

Again, the code sequence is:
rtx bar;
and either
bar =3D gen_rtx_SYMBOL_REF(Pmode,"*foo");

or

bar =3D gen_rtx_MEM(Pmode,gen_rtx_SYMBOL_REF(Pmode,"*foo"));
MEM_VOLATILE_P(bar) =3D 1;

... whereupon I return force_operand(copy_to_reg(bar),NULL_RTX) as the resu=
lt=20
of EXPAND_BUILTIN_VA_ARG.
-----BEGIN PGP SIGNATURE-----

iD8DBQFADprlrT0x/nxLkjwRA4G8AJ4026dmw84/b4189tv/jrnypkHUkwCeKQv5
MlEcCADWFoYP5oX2kpJM7Zc=3D
=3DGTQd
-----END PGP SIGNATURE-----