From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 21756 invoked by alias); 31 May 2011 18:06:39 -0000 Received: (qmail 21748 invoked by uid 22791); 31 May 2011 18:06:38 -0000 X-SWARE-Spam-Status: No, hits=-1.1 required=5.0 tests=AWL,BAYES_00,RCVD_IN_DNSWL_NONE X-Spam-Check-By: sourceware.org Received: from mailrelay007.isp.belgacom.be (HELO mailrelay007.isp.belgacom.be) (195.238.6.173) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 31 May 2011 18:06:22 +0000 X-Belgacom-Dynamic: yes X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AjoFAKUr5U1tgO4S/2dsb2JhbABThEmFTpwIeIhqB60IkFIOgR2DbIEHBJ9r Received: from 18.238-128-109.adsl-dyn.isp.belgacom.be (HELO soleil) ([109.128.238.18]) by relay.skynet.be with SMTP; 31 May 2011 20:06:20 +0200 Message-ID: <1A6A06B5CD4346FEB0EAE0998527F07D@soleil> From: "Philippe Waroquiers" To: "Pedro Alves" , Cc: "Yao Qi" , "Tom Tromey" References: <201105271858.53944.pedro@codesourcery.com> <4DE317B1.1040200@codesourcery.com> <201105311830.59328.pedro@codesourcery.com> Subject: Re: ping: Re: PATCH : allow to set length of hw watchpoints (e.g. for Valgrind gdbserver) Date: Tue, 31 May 2011 18:06:00 -0000 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="utf-8"; reply-type=original Content-Transfer-Encoding: 7bit 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: 2011-05/txt/msg00701.txt.bz2 > If that was the only problem, than it'd be okay --- the user just > shouldn't use the command then. GDB will just do what the > user told it to. But, it looks like the patch changes the > behavior _even_ if the user doesn't use the command. Effectively, the patch changes the behaviour (but I believe in a more consistent way). But if that is considered as not good, I can change the patch so as to keep by default the old behaviour. Note that thanks to the pointer Joel gave me for the gdbserver testing, I have run the regression test suite (on debian 5.0 amd64) with and without the patch, and there is no regression. Now that I understand better how to test gdbserver, I will try to add a test which reproduces the gdbserver crash. > >> The ideal solution, IMO, is remote side gives GDB the value of >> hardware-watchpoint-length-limit, however, I don't know it is easy or >> hard to do such thing. > > We've also discussed completely getting rid of watchpoint > resources accounting recently. For sure, if that would appear, that would be nice. I guess that what we need is a packet such as: "here is a list of hw watchpoint, is this list ok ?" packet. Note that one other thing that I find confusing in the current behaviour is that if you have a certain set of hw watchpoints that were accepted and you add a new one, you might obtain an error back referencing an "old" accepted watchpoint. I think it would be better if the watchpoints would always be re-inserted by gdb in the same order. Philippe