bug#12774: RE: bug#12774: RE: bug#12774: 24.2.50; EMACS 24.2.50.1 crashing‏

[Eli Zaretskii]

> From: Vincent Belaïche <vincent.b.1@hotmail.fr>
> CC: <12774@debbugs.gnu.org>
> Date: Sat, 3 Nov 2012 18:41:31 +0100
> 
> I did now exactly as you said in the first place: I launched only one 
> instance of emacs, got its PID from the task manager, got it crashed, and 
> then connected the gdb to it, and then got the backtrace.

Now we are cooking.  I need a recipe to reproduce this crash.

> I was not doing this way before, but rather launched emacs via `gdb 
> --exec=...emacs'. This is because there is no systematic way to get the crash 
> happen. Sometimes it happens, and sometimes not. It seems to have to do with 
> some real time condition, like some race between two processes. So it seemed 
> to me that it was easier to have it crashed when emacs is run from the 
> debugger, rather run it, and then connect the debugger. I must say that I 
> have no certainty about that. Indeed, the crash happens rather unpredictably. 
> So, anyway, the only truth is the trace below.

It's not related to any race condition.  It's related to memory
allocation; specifically, memory allocation for loading a large
charset.  Can you try to remember what were you doing when Emacs
crashed?