From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 9512 invoked by alias); 7 Nov 2007 08:31:12 -0000 Received: (qmail 9496 invoked by uid 22791); 7 Nov 2007 08:31:10 -0000 X-Spam-Check-By: sourceware.org Received: from mail.tut.by (HELO speedy.tutby.com) (195.137.160.40) by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 07 Nov 2007 08:31:06 +0000 X-VBA32: Checked Received: from [194.105.120.70] (account spiridenok@tut.by) by speedy.tutby.com (CommuniGate Pro WEBUSER 5.1.12) with HTTP id 426909986; Wed, 07 Nov 2007 10:30:54 +0200 From: "Bobo" Subject: Re: Breakpoint in C++ class constructor is never reached To: Thiago Jung Bauermann Cc: gdb X-Mailer: CommuniGate Pro WebUser v5.1.12 Date: Wed, 07 Nov 2007 08:31:00 -0000 Message-ID: In-Reply-To: <1194379922.6746.19.camel@localhost.localdomain> References: <20071026115446.GA26327@caradoc.them.org> <20071026122115.GA28419@caradoc.them.org> <20071026143128.GA26407@caradoc.them.org> <200710261450.l9QEouxs026180@brahms.sibelius.xs4all.nl> <1193410939.5787.157.camel@localhost.localdomain> <1193495532.5787.187.camel@localhost.localdomain> <1193753619.5787.229.camel@localhost.localdomain> <1193857525.5787.258.camel@localhost.localdomain> <1194379922.6746.19.camel@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain;charset=windows-1251;format="flowed" Content-Transfer-Encoding: 8bit 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: 2007-11/txt/msg00047.txt.bz2 Thiago, Please keep me up to date about your progress with this issue. I'm ready to test your changes on Sun Blade 1500 + SUN Solaris 5.8 i'm currently using. I can also use gpatch to apply the patches so it's not needed to wait until the changes are commited to verify them. Tue, 06 Nov 2007 18:12:02 -0200, Thiago Jung Bauermann > Mmm... > > The problem here is that the automatic variable '$<' ("The name of the > first prerequisite") is not working for the decimal32.o target. It works > for the other targets in the output you pasted, though (decNumber.o and > decContext.o). > > Looks like Solaris' make didn't like the slightly fancy > "$(enable_decimal_float)/decimal32.c" dependency, as opposed to the > simpler "decNumber.c" and "decContext.c" in the other targets. > > I just got access to a Solaris machine and started looking into it. > -- > []'s > Thiago Jung Bauermann > Software Engineer > IBM Linux Technology Center > > > On Mon, 2007-11-05 at 11:01 +0200, Bobo wrote: >> Hi, >> >> Downloaded one of the latest snapshots of the HEAD release, still have >> problems with building it: >> >> source='decNumber.c' object='decNumber.o' libtool=no gcc >> -I.././libdecnumber -I. -W -Wall -Wwrite-strings -Wstrict-prototypes >> -Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute >> -Wcast-qual -pedantic -Wno-long-long -I.././libdecnumber -I. -c >>decNumber.c >> source='decContext.c' object='decContext.o' libtool=no gcc >> -I.././libdecnumber -I. -W -Wall -Wwrite-strings -Wstrict-prototypes >> -Wmissing-prototypes -Wold-style-definition -Wmissing-format-attribute >> -Wcast-qual -pedantic -Wno-long-long -I.././libdecnumber -I. -c >> decContext.c >> source='' object='decimal32.o' libtool=no gcc -I.././libdecnumber -I. >> -W >> -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes >> -Wold-style-definition -Wmissing-format-attribute -Wcast-qual -pedantic >> -Wno-long-long -I.././libdecnumber -I. -c >> gcc: no input files >> *** Error code 1 >> make: Fatal error: Command failed for target `decimal32.o' >> Current working directory /usr/local/asm/gdb-6.7.50.20071104/libdecnumber >> *** Error code 1 >> make: Fatal error: Command failed for target `all-libdecnumber' >> Current working directory /usr/local/asm/gdb-6.7.50.20071104 >> *** Error code 1 >> make: Fatal error: Command failed for target `all' >> >> >> wsasd179:/scratch/gdb-6.7.50.20071104 % nedit libdecnumber/Makefile >> >> Wed, 31 Oct 2007 17:05:25 -0200, Thiago Jung Bauermann >> >> > Hi, >> > >> > It seems that Solaris' patch tool doesn't support the unified diff >> > format. I didn't know that, sorry about this problem. >> > >> > I don't have write access to the GCC repo (which is where libdecnumber >> > is hosted), so I couldn't commit it by myself. But Janis just did. It >> > will appear in tomorrow's GDB snapshot. >> > >> > If you don't want to wait until then, you can check out now from the CVS >> > repository by following the instructions here: >> > >> > http://sourceware.org/gdb/current/ >> > >> > -- >> > []'s >> > Thiago Jung Bauermann >> > Software Engineer >> > IBM Linux Technology Center >> > >> > >> > On Wed, 2007-10-31 at 10:10 +0200, Bobo wrote: >> >> Unfortunately i can not patch the attached file. I get the following: >> >> shell:patch -p1 < /tmp/remove-gnuism.diff >> >> Looks like a unified context diff. >> >> File to patch: >> >> >> >> Is it because my version of the patch tool does not match with the diff >> >> version used to create the patch? >> >> >> >> Anyway... Thiago, when are you planning to commit the patch so that i >>get >> >>it >> >> with the latest snapshot? >> >> >> >> Tue, 30 Oct 2007 12:13:39 -0200, Thiago Jung Bauermann >> >> >> >> >> >> > Hi, >> >> > >> >> > The patch hasn't been commited yet, so it won't appear in the >>snapshot. >> >> > >> >> > In the meantime, you can use the attached patch in a recent GDB >>snapshot >> >> > with these commands: >> >> > >> >> > $ cd gdb-6.7.50.20071030 >> >> > $ patch -p1 < /path/to/attached/remove-gnuism.diff >> >> > >> >> > This should get you a GDB source code tree which will build in your >> >> > environment. >> >> > -- >> >> > []'s >> >> > Thiago Jung Bauermann >> >> > Software Engineer >> >> > IBM Linux Technology Center >> >> > >> >> > >> >> > On Tue, 2007-10-30 at 11:53 +0200, Bobo wrote: >> >> >> Sorry i don't really know what "off-list a regenerated configure >> >>script" >> >> >> means and what i need to do with it to verify if the build problem >>is >> >> >> gone... >> >> >> >> >> >> Is there maybe a snapshot available with the latest changes of >>Thiago? >> >> >> (the latest snapshot version i can find on pub/gdb/snapshots/ is >>from >> >>26 >> >> >>Oct >> >> >> 2007) >> >> >> >> >> >> Anyway, i would like to check if the build problem is solved. please >> >>let >> >> >>me >> >> >> know how i can do that. >> >> >> >> >> >> Sat, 27 Oct 2007 12:32:12 -0200, Thiago Jung Bauermann >> >> >> >> >> >> >> >> >> > On Fri, 2007-10-26 at 13:02 -0200, Thiago Jung Bauermann wrote: >> >> >> >> On Fri, 2007-10-26 at 16:50 +0200, Mark Kettenis wrote: >> >> >> >> > > Date: Fri, 26 Oct 2007 10:31:28 -0400 >> >> >> >> > > From: Daniel Jacobowitz >> >> >> >> > > Oh dear. GCC requires GNU make, but GDB is not supposed to, >>I >> >> >>don't >> >> >> >> > > think. This is a new problem. >> >> >> >> > >> >> >> >> > I regularly build GDB with OpenBSD make, so yes this is new >>(and >> >>I'd >> >> >> >> > very much appreciate it if it got fixed). >> >> >> >> >> >> >> >> I will work on this ASAP. >> >> >> > >> >> >> > I just posted a patch to fix this. Sorry for the delay, it took me >> >>some >> >> >> > time to determine if I had copyright assignment to contribute to >>GCC. >> >> >> > >> >> >> > I tested the change with NetBSD's make, so I think the GNUisms are >> >>gone. >> >> >> > If this is not enough to fix your build problems, please let me >>know. >> >> >> > >> >> >> > I can send you off-list a regenerated configure script if you want >>to >> >> >> > test the patch and don't have autoconf 2.59 handy. >> >> >> > -- >> >> >> > []'s >> >> >> > Thiago Jung Bauermann >> >> >> > Software Engineer >> >> >> > IBM Linux Technology Center ------------------ Ôîðåêñ Êëóáó 10 ëåò! http://www.forexclub.by