From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 4175 invoked by alias); 24 Oct 2007 22:14:36 -0000 Received: (qmail 4163 invoked by uid 22791); 24 Oct 2007 22:14:36 -0000 X-Spam-Check-By: sourceware.org Received: from ug-out-1314.google.com (HELO ug-out-1314.google.com) (66.249.92.174) by sourceware.org (qpsmtpd/0.31) with ESMTP; Wed, 24 Oct 2007 22:14:34 +0000 Received: by ug-out-1314.google.com with SMTP id m2so585541ugc for ; Wed, 24 Oct 2007 15:14:31 -0700 (PDT) Received: by 10.78.162.4 with SMTP id k4mr848155hue.1193264071398; Wed, 24 Oct 2007 15:14:31 -0700 (PDT) Received: by 10.78.188.17 with HTTP; Wed, 24 Oct 2007 15:14:31 -0700 (PDT) Message-ID: <4053daab0710241514s3a94f1fds6ef17c226546ccc4@mail.gmail.com> Date: Wed, 24 Oct 2007 22:23:00 -0000 From: "Pedro Alves" To: "Joel Brobecker" , "Luis Machado" , gdb-patches@sources.redhat.com Subject: Re: [patch] Backtrace prints wrong argument value In-Reply-To: <20071024204654.GP10943@caradoc.them.org> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <1190035779.4651.16.camel@localhost> <1190037624.4651.24.camel@localhost> <20070917140416.GA15635@caradoc.them.org> <1190049306.4651.42.camel@localhost> <20070917171821.GA2107@caradoc.them.org> <1192997096.5584.2.camel@localhost> <20071021232837.GD6180@adacore.com> <20071022010550.GA12211@caradoc.them.org> <20071022033114.GB764@adacore.com> <20071024204654.GP10943@caradoc.them.org> X-Google-Sender-Auth: 546cec9144c3cb67 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: 2007-10/txt/msg00640.txt.bz2 Daniel Jacobowitz wrote: > With Pedro's help, two are now fixed. Only build/2339 remains. But > that's with SunPro and I don't think it's a blocker - so we are > probably good to go now. > Irk. I've seen that happen too a few months ago, but totally forgot about it. That happens due to the your's trully and Nick's new STRING_COMMA_LEN in bfd. bfd-in.h:#define STRING_COMMA_LEN(STR) (STR), ((STR) ? sizeof (STR) - 1 : 0) SunPro doesn't like the '?' operator for constant initilizers. elf.c", line 856: non-constant initializer: op "?" The fix is easy - inline the STRING_COMMA_LEN(NULL) cases. Wait, a quick 'grep STRING_COMMA_LEN * -rn | grep NULL' shows there aren't any? This should fix it. I'll see if I can give it a spin with SunPro this week, but no promises. -#define STRING_COMMA_LEN(STR) (STR), ((STR) ? sizeof (STR) - 1 : 0) +#define STRING_COMMA_LEN(STR) (STR), (sizeof (STR) - 1) Cheers, Pedro Alves