From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 5611 invoked by alias); 30 Aug 2011 12:42:12 -0000 Received: (qmail 5602 invoked by uid 22791); 30 Aug 2011 12:42:11 -0000 X-SWARE-Spam-Status: No, hits=-1.5 required=5.0 tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,RCVD_IN_DNSWL_NONE X-Spam-Check-By: sourceware.org Received: from nm4-vm1.bt.bullet.mail.ukl.yahoo.com (HELO nm4-vm1.bt.bullet.mail.ukl.yahoo.com) (217.146.182.252) by sourceware.org (qpsmtpd/0.43rc1) with SMTP; Tue, 30 Aug 2011 12:41:51 +0000 Received: from [217.146.183.196] by nm4.bt.bullet.mail.ukl.yahoo.com with NNFMP; 30 Aug 2011 12:41:50 -0000 Received: from [217.146.183.206] by tm2.bt.bullet.mail.ukl.yahoo.com with NNFMP; 30 Aug 2011 12:41:50 -0000 Received: from [127.0.0.1] by omp1004.bt.mail.ukl.yahoo.com with NNFMP; 30 Aug 2011 12:41:50 -0000 Received: (qmail 43738 invoked by uid 60001); 30 Aug 2011 12:41:50 -0000 Received: from [94.72.254.2] by web86707.mail.ird.yahoo.com via HTTP; Tue, 30 Aug 2011 13:41:49 BST References: <20110830014851.78030246131@ruffy.mtv.corp.google.com> <20110830091623.GA5329@host1.jankratochvil.net> <1314698487.57217.YahooMailRC@web86705.mail.ird.yahoo.com> <20110830115627.GA21003@host1.jankratochvil.net> Message-ID: <1314708109.43672.YahooMailRC@web86707.mail.ird.yahoo.com> Date: Tue, 30 Aug 2011 12:42:00 -0000 From: pfee@talk21.com Subject: Re: [RFC] stept, nextt, finisht, untilt, continuet To: Jan Kratochvil Cc: Doug Evans , gdb-patches@sourceware.org In-Reply-To: <20110830115627.GA21003@host1.jankratochvil.net> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-IsSubscribed: yes Mailing-List: contact gdb-patches-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-patches-owner@sourceware.org X-SW-Source: 2011-08/txt/msg00594.txt.bz2 > > I like the 'step' mode of scheduler-locking, but often wish it applied= to=20 >the=20 > > > "next" command, not just step. >=20 > I agree, it does not apply to the continue-over-call part of `next'. Bu= t=20 that > is a bug which should be fixed. >=20 >=20 > > My suggestion would be to create a "set scheduler-locking next" mode i= n=20 >which=20 > > > both "step" and "next" operate with other threads locked out. >=20 > Do you think the "step" mode would be still useful if "next" exists? To be honest, I would always use "set scheduler-locking next" if it existed= and=20 would never use "step" locking mode. My suggestion was influenced by a desire to minimise the code changes. It = would=20 be better for me if "set scheduler-locking step" was replaced with "set=20 scheduler-locking next" though I haven't investigated how easy that would = be to=20 implement. Another point, regarding "continue". In non-stop mode, there's a "-a" flag= to=20 continue all threads. Would it be worth using the same flag when in=20 scheduler-locking mode to cause the entire process to resume? Thanks, Paul