From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 31717 invoked by alias); 4 Jun 2002 16:19:00 -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 31560 invoked from network); 4 Jun 2002 16:18:53 -0000 Received: from unknown (HELO kiruna.synopsys.com) (204.176.20.18) by sources.redhat.com with SMTP; 4 Jun 2002 16:18:53 -0000 Received: from mother.synopsys.com (mother.synopsys.com [146.225.100.171]) by kiruna.synopsys.com (Postfix) with ESMTP id 92A75F5BF; Tue, 4 Jun 2002 09:18:52 -0700 (PDT) Received: from atrus.synopsys.com (localhost [127.0.0.1]) by mother.synopsys.com (8.9.1/8.9.1) with ESMTP id JAA00333; Tue, 4 Jun 2002 09:18:16 -0700 (PDT) From: Joe Buck Received: (from jbuck@localhost) by atrus.synopsys.com (8.9.3+Sun/8.9.1) id JAA01096; Tue, 4 Jun 2002 09:18:50 -0700 (PDT) Message-Id: <200206041618.JAA01096@atrus.synopsys.com> Subject: Re: Speeding up GC To: ak@suse.de (Andi Kleen) Date: Tue, 04 Jun 2002 09:21:00 -0000 Cc: davem@redhat.com (David S. Miller), dberlin@dberlin.org, pkoning@equallogic.com, levon@movementarian.org, ak@suse.de, rth@redhat.com, gcc@gcc.gnu.org In-Reply-To: <20020604165254.A697@wotan.suse.de> from "Andi Kleen" at Jun 04, 2002 04:52:54 PM MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-SW-Source: 2002-06/txt/msg00227.txt.bz2 > > This is kind of the "next level" of what I was proposing. Once you > > have this "enter short-term mode" thing, you can have the "exit > > short-term" call just reset the state of all the temporary GC pools as > > if GC has been run on them and no references had been found. > > Just doesn't this bring all the bugs back that got fixed by the GC introduction? You make the short-term area just a pool of the GC, and you have two modes, settable by a flag. In the production mode, when "exit short-term" is called you just nuke the short-term pool. In the GC-debug mode, when exit short-term is called you do a GC and abort if there's anything live in the short-term pool. You then run the full regression suite in the GC-debug mode. GC-debug is shipped, so that if any user gets a mysterious ICE you can see if this is the cause.