X-Authentication-Warning: delorie.com: mail set sender to djgpp-workers-bounces using -f Date: Thu, 15 Jan 2004 13:46:41 +0200 From: "Eli Zaretskii" Sender: halo1 AT zahav DOT net DOT il To: Andris Pavenis Message-Id: <7704-Thu15Jan2004134640+0200-eliz@elta.co.il> X-Mailer: emacs 21.3.50 (via feedmail 8 I) and Blat ver 1.8.9 CC: djgpp-workers AT delorie DOT com In-reply-to: <200401151324.12246.pavenis@latnet.lv> (message from Andris Pavenis on Thu, 15 Jan 2004 13:24:12 +0200) Subject: Re: rhide and gdb-6.0 [WAS: Re: Rhide 1.5 crashes when using Help menu] References: <200401151122 DOT 00249 DOT pavenis AT latnet DOT lv> <9787-Thu15Jan2004123110+0200-eliz AT elta DOT co DOT il> <200401151324 DOT 12246 DOT pavenis AT latnet DOT lv> 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: Andris Pavenis > Date: Thu, 15 Jan 2004 13:24:12 +0200 > > > > Probably due to changes in the annotation functions, sigh. > > For other commands one can successfully get output with > ui_file_xstrdup() from gdb_stdout, but for backtrace it doesn't work. Simply > there is nothing in buffer when anotate_*() functions are called. Maybe that > is wrong way how to retrieve output from GDB. I don't know GDB internals so > well. It might be a good idea to present the problem on the GDB mailing list (gdb AT sources DOT redhat DOT com), they could give some ideas to solve this or at least recognize that it's a bug. > Function list is a different thing. There is attempt to mess directly with > debug info and my first attempt to modify it for GDB-6.0 failed. If you could tell more about the problem, I will try to look into this when I have time (which isn't going to be soon, but it's better than never ;-). > PS. Looked, what is annotation level 2 (set annotate 2). As I remember there > were discussions about deprecating it.It seems that it could be rather > usefull if we would use it here and analyze full output instead of trying to > get it from anotate_*() callbacks. Actually, The Future is in using the MI interface (see the node "GDB/MI" in the GDB manual for more details). This interface is actively developed and the GDB developers would like all front-ends to switch to it.