From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:41342) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1R5cSZ-0007rr-0I for qemu-devel@nongnu.org; Mon, 19 Sep 2011 07:56:30 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1R5cSX-0007aY-R5 for qemu-devel@nongnu.org; Mon, 19 Sep 2011 07:56:26 -0400 Received: from mx1.redhat.com ([209.132.183.28]:15723) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1R5cSX-0007aU-I9 for qemu-devel@nongnu.org; Mon, 19 Sep 2011 07:56:25 -0400 Message-ID: <4E772DE1.2090407@redhat.com> Date: Mon, 19 Sep 2011 14:56:17 +0300 From: Avi Kivity MIME-Version: 1.0 References: <4E74FC29.1050003@mail.berlios.de> <1316289634-18786-5-git-send-email-weil@mail.berlios.de> <4E75865B.5040106@mail.berlios.de> <4E759C43.5070708@redhat.com> <4E76305B.80108@redhat.com> <51221ff5190e1823f0d6b602c4ca0f4d.squirrel@www.firstfloor.org> In-Reply-To: <51221ff5190e1823f0d6b602c4ca0f4d.squirrel@www.firstfloor.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 5/8] tcg: Add interpreter for bytecode List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Andi Kleen Cc: Paolo Bonzini , QEMU Developers On 09/19/2011 09:52 AM, Andi Kleen wrote: > > I think it also improves branch target prediction - if you have a tight > > loop of a few opcodes the predictor can guess where you're headed (since > > there is a separate lookup key for each opcode), whereas with the > > original code, there's a single key which cannot be used to predict the > > branch target. > > At least usually. I caught at least one version of gcc to CSE the jump > instruction into a common location in one of my interpreters > :-( But not all do it at least, and I hope gcc gets fixed. You generally want CSE, yes? So you can't blame gcc for getting it wrong sometimes. -- error compiling committee.c: too many arguments to function