GNU bug report logs - #11867
24.1.50; Windows bootstrap crash (converting tit files?)

Previous Next

Full log

View this message in rfc822 format

From: Juanma Barranquero <lekktu <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: rcopley <at> gmail.com, 11867 <at> debbugs.gnu.org
Subject: bug#11867: 24.1.50; Windows bootstrap crash (converting tit files?)
Date: Mon, 9 Jul 2012 22:54:22 +0200

On Mon, Jul 9, 2012 at 10:42 PM, Eli Zaretskii <eliz <at> gnu.org> wrote:

> Sorry, I don't understand: how did you know to attach the debugger
> exactly when Emacs crashed or was about to crash?  Just attaching a
> debugger stops Emacs wherever it was, which could be anywhere (judging
> by the backtrace, it was in some system call, waiting for some other
> syscall to finish).  That place is random, normally having nothing to
> do with the crash.
[...]
> But you said that Emacs crashes while byte-compiling japanese.el.
> When it did crash, did you see the abort dialog, or did you see
> something else?

During the bootstrap, while Emacs is byte-compiling files, it crashes.
I know that because:

1) I get a message from Windows (not the usual w32_abort dialog)
telling me that Emacs has crashed and asking whether I want to send a
report to Microsoft.
2) Bootstrapping stops until I dismiss that message.

At that point, until I tell Windows to dismiss the report sending, I
have all the time in the world to get the id of the only Emacs running
and attach gdb to it.

Additionally,
3) I know that it is in japanese.el, because it happens while
bytecompiling lisp/language/*, and japanese.el is the only missing
.elc afterwards.

I get the crash in all bootstraps, so if you want me to try something, just ask.

    Juanma

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.