Re: SIGSEGV in X export display.

[Michaël Cadilhac]

Richard Stallman <address@hidden> writes:

> I don't see anything in that backtrace that relates
> to losing an X connection,

I do agree, it just happens in such a situation. It can be, I don't
know, that the frame being deleted, some objects are no longer valid
but still are in a GC list.

> to SIGSEGV,

Okey, I didn't add the one line that says « Sigsegv ».

> Please do NOT use `bt full' in your bug reports.

Just following the rules of « M-x report-emacs-bugs ».

Well, otherwise, this problem still appears quite frequently (say,
once a day), with still the same backtrace, that starts with:

######################################################################
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 16384 (LWP 19690)]
0x0817ef1d in mark_object (arg=135891787) at alloc.c:5509
5509            MARK_INTERVAL_TREE (ptr->intervals);
(gdb) bt full
#0  0x0817ef1d in mark_object (arg=135891787) at alloc.c:5509
        ptr = (struct Lisp_String *) 0x8198b48
        obj = 135891787
        cdr_count = 0
#1  0x081cd803 in mark_byte_stack () at bytecode.c:294
        stack = (struct byte_stack *) 0xbf81c210
        obj = (Lisp_Object *) 0xbf81c288
######################################################################

etc.

This happens with a gnuclient  connection and, maybe, a frame crashing
will rendering  « illegal » X  things: it is usually  triggered by the
viewing of a gnus article.

Well, there's still gdb running on my computer if you can have any
idea.

-- 
/!\ My mail address changed, please update your files accordingly.
 |      Michaël `Micha' Cadilhac   |  Would someone please DTRT with this,  |
 |         Epita/LRDE Promo 2007   |        then ACK?                       |
 |  http://michael.cadilhac.name   |          -- Richard Stallman           |
 `--  -   JID: address@hidden --'                                   -  --'

pgptYhH36c4mr.pgp
Description: PGP signature