From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 16515 invoked by alias); 21 Sep 2004 21:13:09 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 16494 invoked from network); 21 Sep 2004 21:13:06 -0000 Received: from unknown (HELO lana.hrz.tu-chemnitz.de) (134.109.132.3) by sourceware.org with SMTP; 21 Sep 2004 21:13:06 -0000 Received: from [62.159.236.194] (helo=[192.168.0.2]) by lana.hrz.tu-chemnitz.de with asmtp (TLSv1:AES256-SHA:256) (Exim 4.41) id 1C9rwg-0006Sh-2S for gcc@gcc.gnu.org; Tue, 21 Sep 2004 23:13:06 +0200 Message-ID: <41509991.2000909@cs.tu-chemnitz.de> Date: Wed, 22 Sep 2004 01:21:00 -0000 From: =?ISO-8859-1?Q?Adrian_Str=E4tling?= User-Agent: Thunderbird 0.7.3 (Windows/20040803) MIME-Version: 1.0 To: GCC Subject: Register Allocation Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Score: 0.0 (/) X-Spam-Report: --- Start der SpamAssassin 2.64 Textanalyse (0.0 Punkte) Fragen an/questions to: Postmaster TU Chemnitz --- Ende der SpamAssassin Textanalyse X-Scan-Signature: d7bb2b01975254b8f41cda1475e3c4ea X-SW-Source: 2004-09/txt/msg01257.txt.bz2 Hi, I'm working on a target that does support up to 8 parallel instructions, but they have to be grouped at compile time. The scheduler could do much better than it currently does if there were less (output) depencies. Is it possible to tell the register allocator not to reduce the register count to the minimum? example: ( || means that this insn is executed in parallel to the last one ) 1) mvkl L4, b4 \ 2) mvkh L4, b4 + push ret_label 3) stw b4, *--b15 / || mvkl times2, b4 \ 4) mvkh times2, b4 + branch 5) b b4 / L4: If the second temporary were not assigned to b4 but to b5, the schedule would be shorter: 1) mvkl L4, b4 || mvkl times2, b5 2) mvkh L4, b4 || mvkh times2, b5 3) stw b4, *--b15 || b b5 L4: Thanks, Adrian