From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 13556 invoked by alias); 28 May 2009 08:08:55 -0000 Received: (qmail 13537 invoked by uid 22791); 28 May 2009 08:08:52 -0000 X-SWARE-Spam-Status: No, hits=-1.7 required=5.0 tests=AWL,BAYES_00,SPF_SOFTFAIL X-Spam-Check-By: sourceware.org Received: from mtaout4.012.net.il (HELO mtaout3.012.net.il) (84.95.2.10) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Thu, 28 May 2009 08:08:47 +0000 Received: from conversion-daemon.i_mtaout3.012.net.il by i_mtaout3.012.net.il (HyperSendmail v2004.12) id <0KKC00500GW1HD00@i_mtaout3.012.net.il> for gdb-patches@sourceware.org; Thu, 28 May 2009 11:08:44 +0300 (IDT) Received: from HOME-C4E4A596F7 ([84.228.115.215]) by i_mtaout3.012.net.il (HyperSendmail v2004.12) with ESMTPA id <0KKC009EYHAJ7FB0@i_mtaout3.012.net.il>; Thu, 28 May 2009 11:08:44 +0300 (IDT) Date: Thu, 28 May 2009 08:08:00 -0000 From: Eli Zaretskii Subject: Re: [RFA] Fix troubles with watchpoints in DJGPP In-reply-to: <200905280030.28382.pedro@codesourcery.com> To: Pedro Alves Cc: gdb-patches@sourceware.org, muller@ics.u-strasbg.fr Reply-to: Eli Zaretskii Message-id: <8363flsl80.fsf@gnu.org> References: <001401c9de9f$928b9f80$b7a2de80$@u-strasbg.fr> <200905280030.28382.pedro@codesourcery.com> 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: 2009-05/txt/msg00600.txt.bz2 > From: Pedro Alves > Date: Thu, 28 May 2009 00:30:28 +0100 > Cc: "Pierre Muller \(IMAP\)" , > "'Eli Zaretskii'" > > On Wednesday 27 May 2009 08:48:49, Pierre Muller (IMAP) wrote: > > > PS-1) Are there not other native targets, without > > dynamic libraries, that will suffer the same troubles? > > I think so. I've just tried on x86_64-linux, with a statically > linked binary (I used gdb.threads/staticthreads, set a watchpoint > on semaphore), and although there are no shared libraries loaded, > the problem is masked by adding the symbols of the vsyscall page > (sysfile-mem.c:add_vsyscall_page). If I hack that function to > do nothing, I see that same thing you're seeing on djgpp. > > Maybe there's a place for a generic fix? Somewhere after > having opened a connection to the target interface. I was > thinking of post_create_inferior, but sounds like opening > a connection to a remote target with "target remote" that > happens to not pull in any more symbols (like most embedded > targets) is having the same problem? Maybe there should be > a target_post_open ... I'd indeed prefer a generic fix. Pierre, could you please try making a patch along the lines Pedro suggests above?