From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 12182 invoked by alias); 30 May 2011 05:34:20 -0000 Received: (qmail 11922 invoked by uid 22791); 30 May 2011 05:34:18 -0000 X-SWARE-Spam-Status: No, hits=-1.0 required=5.0 tests=AWL,BAYES_00,RCVD_IN_DNSWL_NONE X-Spam-Check-By: sourceware.org Received: from mailrelay004.isp.belgacom.be (HELO mailrelay004.isp.belgacom.be) (195.238.6.170) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 30 May 2011 05:33:56 +0000 X-Belgacom-Dynamic: yes X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AgUFAFAq401tgHac/2dsb2JhbABVhEmFTpwSeIhqB61cj1sOgR2DbIEHBJ9r Received: from 156.118-128-109.adsl-dyn.isp.belgacom.be (HELO soleil) ([109.128.118.156]) by relay.skynet.be with SMTP; 30 May 2011 07:33:54 +0200 Message-ID: From: "Philippe Waroquiers" To: "Yao Qi" , "Pedro Alves" Cc: , "Tom Tromey" References: <4DDF196F.5010202@codesourcery.com> <201105271858.53944.pedro@codesourcery.com> <4DE317B1.1040200@codesourcery.com> Subject: Re: ping: Re: PATCH : allow to set length of hw watchpoints (e.g. for Valgrind gdbserver) Date: Mon, 30 May 2011 05:34:00 -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/msg00671.txt.bz2 > One new command "set remote hardware-watchpoint-length-limit" > is added in the patch, which is only useful to gdbserver+valgrind. > When gdb is talking with normal gdbserver, it may be wrong to set > hardware-watchpoint-length-limit in gdb side. Users should be careful > when using this command. I think the command is also useful in other cases (cfr my other post of yesterday http://sourceware.org/ml/gdb-patches/2011-05/msg00664.html). But if you believe this should be aimed only (or mainly) at Valgrind gdbserver, I can update the patch (e.g. the documentation + change length default value) to make it more Valgrind gdbserver specific. > 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. An approach where the remote side would tell which hw watchpoints are ok is the best. It was discussed in another thread, but that is not easy to do. E.g. for the moment, which hw watchpoints are accepted depends on the order in which gdb inserts the watchpoints. Currently, depending of the order of watchpoint insertion, you will even trigger a bug in x86/amd64 gdbserver (cfr http://sourceware.org/ml/gdb-patches/2011-05/msg00664.html). Philippe