From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1KcPE9-0002fc-8F
	for qemu-devel@nongnu.org; Sun, 07 Sep 2008 14:43:13 -0400
Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43)
	id 1KcPE7-0002f8-MD
	for qemu-devel@nongnu.org; Sun, 07 Sep 2008 14:43:12 -0400
Received: from [199.232.76.173] (port=49026 helo=monty-python.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1KcPE7-0002eh-5x
	for qemu-devel@nongnu.org; Sun, 07 Sep 2008 14:43:11 -0400
Received: from mail.codesourcery.com ([65.74.133.4]:40108)
	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 1KcPE4-0005M5-Pa
	for qemu-devel@nongnu.org; Sun, 07 Sep 2008 14:43:10 -0400
From: Paul Brook <paul@codesourcery.com>
Subject: Re: [Qemu-devel] TCG native 32->64 concatenation
Date: Sun, 7 Sep 2008 19:43:03 +0100
References: <200809071753.27384.paul@codesourcery.com>
	<f43fc5580809071115s1013a287tb31297f1620119f3@mail.gmail.com>
In-Reply-To: <f43fc5580809071115s1013a287tb31297f1620119f3@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain;
  charset="utf-8"
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Message-Id: <200809071943.04030.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: Blue Swirl <blauwirbel@gmail.com>

On Sunday 07 September 2008, Blue Swirl wrote:
> On 9/7/08, Paul Brook <paul@codesourcery.com> wrote:
> > The patch below adds a new concat_i32_i64 TCG op.  This allows a pair of
> >  32-bit values to be efficiently combined to form a 64-bit value.  I've
> >  converted all the cases I could find to use this, and tested the arm
> > code on both 32 and 64-bit hosts.
>
> Sparc's helper_pack64 does the same thing, I used it to reduce
> register pressure on i386 for 64 bit stores.

They're slightly different because helper_pack64 takes target_ulong arguments.
Replacing helper_pack64 with concat_i32_i64 would need explicit truncation on 
64-bit targets. The current TCG implementation will work if you use the wrong 
register type, however I don't want targets relying on that.

Paul