From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1IQP7e-0005Of-ML
	for qemu-devel@nongnu.org; Wed, 29 Aug 2007 11:06:22 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1IQP7d-0005Nl-2G
	for qemu-devel@nongnu.org; Wed, 29 Aug 2007 11:06:22 -0400
Received: from [199.232.76.173] (helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1IQP7c-0005Na-S8
	for qemu-devel@nongnu.org; Wed, 29 Aug 2007 11:06:20 -0400
Received: from mail.codesourcery.com ([65.74.133.4])
	by monty-python.gnu.org with esmtps
	(TLS-1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.60)
	(envelope-from <paul@codesourcery.com>) id 1IQP7c-0005dy-2O
	for qemu-devel@nongnu.org; Wed, 29 Aug 2007 11:06:20 -0400
From: Paul Brook <paul@codesourcery.com>
Subject: Re: [Qemu-devel] [patch] make qemu work with GCC 4
Date: Wed, 29 Aug 2007 16:06:10 +0100
References: <Pine.LNX.4.64.0708282124590.23011@wotan.suse.de>
In-Reply-To: <Pine.LNX.4.64.0708282124590.23011@wotan.suse.de>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
Message-Id: <200708291606.14173.paul@codesourcery.com>
Reply-To: qemu-devel@nongnu.org
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: qemu-devel@nongnu.org
Cc: Michael Matz <matz@suse.de>, Alexander Graf <agraf@suse.de>

>        I solved that by placing one of the T[012] operands into memory
>        for HOST_I386, thereby freeing one reg.  Here's some justification
>        of why that doesn't really cost performance: with three free regs
>        GCC is already spilling like mad in the snippets, we just trade one
>        of those memory accesses (to stack) with one other mem access to
>        the cpu_state structure, which will be in cache.

Do you have any evidence to support this claim? Last time I did this it cau=
sed=20
a significant performance hit. I'd guess that most common ops are simple=20
enough that we don't need more than 3 registers.

> --- qemu-0.9.0.cvs.orig/softmmu_header.h
> - =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0: "%eax", "%ecx", "%edx", "memory", =
"cc");
> + =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0: "%eax", "%edx", "memory", "cc");

This change is wrong. The inline asm calls C code which clobbers %ecx.

Paul