[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: crash in cvs head
From: |
Richard Stallman |
Subject: |
Re: crash in cvs head |
Date: |
Fri, 30 Jun 2006 23:21:45 -0400 |
(gdb) p dpyinfo
$1 = (struct x_display_info *) 0x0
(gdb) up
#1 0x080cedf4 in unwind_create_frame (frame=193660300) at xfns.c:2968
2968 x_free_frame_resources (f);
(gdb) p f
No symbol "f" in current context.
The variable f was apparently optimized away.
It was surely created from FRAME. If you had examined the data inside
FRAME, you could have diagnosed more.
Nonetheless, I think I see the problem. Does this fix it?
*** xfns.c 25 Jun 2006 12:16:32 -0400 1.668
--- xfns.c 30 Jun 2006 15:46:37 -0400
***************
*** 3075,3081 ****
f->output_data.x->scroll_bar_top_shadow_pixel = -1;
f->output_data.x->scroll_bar_bottom_shadow_pixel = -1;
#endif /* USE_TOOLKIT_SCROLL_BARS */
- record_unwind_protect (unwind_create_frame, frame);
f->icon_name
= x_get_arg (dpyinfo, parms, Qicon_name, "iconName", "Title",
--- 3075,3080 ----
***************
*** 3084,3089 ****
--- 3083,3091 ----
f->icon_name = Qnil;
FRAME_X_DISPLAY_INFO (f) = dpyinfo;
+
+ /* With FRAME_X_DISPLAY_INFO set up, this unwind-protect is safe. */
+ record_unwind_protect (unwind_create_frame, frame);
#if GLYPH_DEBUG
image_cache_refcount = FRAME_X_IMAGE_CACHE (f)->refcount;
dpyinfo_refcount = dpyinfo->reference_count;
- crash in cvs head, Sam Steingold, 2006/06/29
- Re: crash in cvs head,
Richard Stallman <=