X-Authentication-Warning: delorie.com: mail set sender to djgpp-workers-bounces using -f
X-Recipient: djgpp-workers AT delorie DOT com
Date: Tue, 08 Sep 2009 20:45:41 +0300
From: Eli Zaretskii <eliz AT gnu DOT org>
Subject: Re: Building latest GDB snapshots
In-reply-to: <000301ca305d$bff70850$3fe518f0$@u-strasbg.fr>
X-012-Sender: halo1 AT inter DOT net DOT il
To: Pierre Muller <muller AT ics DOT u-strasbg DOT fr>
Cc: djgpp-workers AT delorie DOT com
Message-id: <83ljkp2uuy.fsf@gnu.org>
References: <83hbvj80ht DOT fsf AT gnu DOT org> <93c172b50909040715n791c007ema4715db95e946e2e AT mail DOT gmail DOT com> <83y6ou4sls DOT fsf AT gnu DOT org> <83tyzi4nn6 DOT fsf AT gnu DOT org> <83fxb14c4i DOT fsf AT gnu DOT org> <001801ca2f96$43b892d0$cb29b870$@u-strasbg.fr> <83vdju3ami DOT fsf AT gnu DOT org> <000301ca305d$bff70850$3fe518f0$@u-strasbg.fr>
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> From: "Pierre Muller" <muller AT ics DOT u-strasbg DOT fr>
> Cc: "'Eli Zaretskii'" <eliz AT gnu DOT org>
> Date: Tue, 8 Sep 2009 10:24:14 +0200
> 
> > What are the problems that you need to hack the script?  It works for
> > me unaltered (assuming you are using the version from after I posted
> > the message to which you respond).  What stage(s) crash, and what are
> > the error messages?
> 
>   I was trying to use the script with CVS HEAD directly,
> but that apparently doesn't work :(

Again, I'd like to see the error messages and/or any other evidence of
the problem, and also to know where it crashed in the process.

I no longer have time to try building GDB in several possible ways, so
I only build the snapshots (which have the same structure as a release
tarball), and only in the source tree.  However, if I can understand
the problems by looking at the error messages, I still can try to keep
GDB capable of building in other ways.  So please do show the
information about the problems.

> 
>   Nevertheless, I first tried to compile in a separate 
> directory (I wanted to check both 2.03 and 2.04...)
> and it crashed when I tried this :(
>   But directly in source directory your way works !

See above: I'd like to know where it crashed and how.

>   I already discovered one strange thing about the 
> multiple inferiors support:
>   Trying to debug gdb with itself,
> 
>   I restarted gdb several times (as debuggee)
> and finally, when I try to quit, I get this:
> (top-gdb) q
> A debugging session is active.
> 
>         Inferior 2 [process 42] will be killed.
>         Inferior 1 [process 42] will be killed.
> 
> Quit anyway? (y or n)

If you can debug this, fine.  If not, can you provide a recipe for
reproducing this?  (I won't expect the DJGPP build to even support
multiple inferiors, because I'm quite sure dbgcom.c is not designed
for that.)