This is the mail archive of the
gdb@sourceware.org
mailing list for the GDB project.
Re: Debugging through exec() (Linux MAY_FOLLOW_EXEC)
- From: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- To: Eli Zaretskii <eliz at gnu dot org>
- Cc: gdb at sourceware dot org
- Date: Mon, 4 Sep 2006 19:58:27 +0200
- Subject: Re: Debugging through exec() (Linux MAY_FOLLOW_EXEC)
- References: <20060729185317.GA16200@host0.dyn.jankratochvil.net> <200607312038.k6VKchKj018729@elgar.sibelius.xs4all.nl> <20060805164144.GA23819@host0.dyn.jankratochvil.net> <20060808160113.GC21032@nevyn.them.org> <20060814150628.GA24544@host0.dyn.jankratochvil.net> <u7j1b581f.fsf@gnu.org> <20060814213849.GA1433@nevyn.them.org> <u3bby65ua.fsf@gnu.org> <20060815142819.GA26405@host0.dyn.jankratochvil.net> <usljui5z3.fsf@gnu.org>
On Fri, 18 Aug 2006 14:27:44 +0200, Eli Zaretskii wrote:
...
> Thanks for the explanations, but I seem to be dense today, because I
> don't get to the cheese yet.
>
> Perhaps if you could describe a scenario where the current behavior is
> inconvenient, and what does your patch change in this respect, I will
> understand the issues.
I have a Perl script spawning scp(1) in PTY. This scp(1) will not print the
progress bar if spawned by this Perl script, it shows it otherwise (it uses
isatty() to detect it). I would like to debug scp(1) there as the bug is
(apparently) not reproducible when scp(1) is executed standalone.
Sure there are many ways how to debug it but I feel this is the right sample
problem for "catch exec".
Unfortunately that my patch will not provide correct enough debugging state
after the "exec" event for the new inferior. Patch needs to be improven.
Regards,
Jan