From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7846 invoked by alias); 7 Feb 2007 23:57:05 -0000 Received: (qmail 7836 invoked by uid 22791); 7 Feb 2007 23:57:05 -0000 X-Spam-Check-By: sourceware.org Received: from tmpnat1.honeywell.com (HELO AZ18CN849.global.ds.honeywell.com) (199.64.0.252) by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 07 Feb 2007 23:56:59 +0000 Received: from AZ18EV808.global.ds.honeywell.com ([131.127.167.102]) by AZ18CN849.global.ds.honeywell.com with Microsoft SMTPSVC(6.0.3790.1830); Wed, 7 Feb 2007 16:56:57 -0700 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Subject: RE: GDB solib interface Date: Wed, 07 Feb 2007 23:57:00 -0000 Message-ID: In-Reply-To: From: "Smith, Stephen \(SWCOE\)" To: "Smith, Stephen \(SWCOE\)" , "Daniel Jacobowitz" Cc: X-IsSubscribed: yes Mailing-List: contact gdb-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: gdb-owner@sourceware.org X-SW-Source: 2007-02/txt/msg00046.txt.bz2 We figured out a fix.=20=20 -----Original Message----- From: Smith, Stephen (SWCOE)=20 Sent: Wednesday, February 07, 2007 12:41 PM To: Smith, Stephen (SWCOE); Daniel Jacobowitz Cc: gdb@sourceware.org Subject: RE: GDB solib interface Weird issue, we have implemented the patch for the remote protocol using the patch that you posted with some minor tweeks. We are getting ready to internally release the updated gdbserver and found a problem. On our platform, it is not possible to hook the function that dynamically loads the libraries, so I was checking to see if new libraries were loaded after the processor hit a breakpoint (i.e. immediatetly following the prepare_resume_reply() for the C, S, c, and s packets). Everything works well unless the following two things are true: * there are new libraries which are then reported * the program counter stopped at the location=20 which has a GDB applied breakpoint. In this case, the instruction which is where the breakpoint was inserted is not executed and in our case a fault occurred. Since on my platform I won't halt because of a library load, where should I place this check?