2:02 p.m.
I managed to reproduce this with 21.4.6 and 21.5.4 under cygwin and 21.4.5 under Linux.
quimby$ cat foo.el
(let* ((buf (compile "make"))
(proc (get-buffer-process buf)))
(while (= 0 (process-exit-status proc))
(sleep-for 1))
(message "%s" (buffer-string buf)))
quimby$
quimby$
quimby$ xemacs -vanilla -batch -load foo.el
(No files need saving)
Fatal error (11).
Your files have been auto-saved.
Use `M-x recover-session' to recover them.
If you have access to the PROBLEMS file that came with your
version of XEmacs, please check to see if your crash is described
there, as there may be a workaround available.
Otherwise, please report this bug by running the send-pr
script included with XEmacs, or selecting `Send Bug Report'
from the help menu.
As a last resort send ordinary email to `crashes(a)xemacs.org'.
*MAKE SURE* to include the information in the command
M-x describe-installation.
If at all possible, *please* try to obtain a C stack backtrace;
it will help us immensely in determining what went wrong.
To do this, locate the core file that was produced as a result
of this crash (it's usually called `core' and is located in the
directory in which you started the editor, or maybe in your home
directory), and type
gdb /usr/local/bin/xemacs core
then type `where' when the debugger prompt comes up.
(If you don't have GDB on your system, you might have DBX,
or XDB, or SDB. A similar procedure should work for all of
these. Ask your system administrator if you need more help.)
Lisp backtrace follows:
pos-visible-in-window-p(152 #<window on "*compilation*" 0x2436>)
# bind (buffer-read-only status)
# (unwind-protect ...)
# bind (obuf omax opoint estatus window buffer msg proc)
compilation-sentinel(#<process "compilation" pid 3612 state:exit>
"exited abnormally with code 2\n")
# (condition-case ... . error)
# bind (inhibit-quit)
# (unwind-protect ...)
# (unwind-protect ...)
sleep-for(1)
(while (= 0 (process-exit-status proc)) (sleep-for 1))
# bind (proc buf)
(let* ((buf ...) (proc ...)) (while (= 0 ...) (sleep-for 1)) (message "%s"
(buffer-string buf)))
# bind (current-load-list)
# (unwind-protect ...)
# bind (load-file-name)
# (unwind-protect ...)
# (unwind-protect ...)
# (unwind-protect ...)
# (unwind-protect ...)
# (unwind-protect ...)
load-internal("/home/stodghil/foo.el" nil t nil undecided)
# bind (path handler filename nosuffix nomessage noerror file)
load("/home/stodghil/foo.el" nil t)
# bind (file arg)
command-line-do-load("-load")
# bind (dir file-count line end-of-options file-p arg tem)
command-line-1()
# bind (command-line-args-left)
command-line()
# (unwind-protect ...)
normal-top-level()
# (condition-case ... . error)
# (catch top-level ...)
Segmentation fault
quimby$
4:27 p.m.
"Paul Stodghill" <stodghil(a)cs.cornell.edu> writes:
I managed to reproduce this with 21.4.6 and 21.5.4 under cygwin and
21.4.5 under Linux.
Jarl Friis <jarl(a)diku.dk> writes:
This one even crashes my 21.1.14 !
Please, please _always_ provide backtraces when crashing on Unix
boxes..
However in this I can see it myself...
Program received signal SIGSEGV, Segmentation fault.
0x818705e in generate_modeline (w=0x85bd8f0, dl=0x85c10e8, type=2)
at /local/xemacs/trunk/src/redisplay.c:3654
3654 font_inst = WINDOW_FACE_CACHEL_FONT (w, MODELINE_INDEX, Vcharset_ascii);
(gdb) where
#0 0x818705e in generate_modeline (w=0x85bd8f0, dl=0x85c10e8,
type=2) at /local/xemacs/trunk/src/redisplay.c:3654
#1 0x8188093 in ensure_modeline_generated (w=0x85bd8f0, type=2)
at /local/xemacs/trunk/src/redisplay.c:4217
#2 0x8189204 in regenerate_window (w=0x85bd8f0, start_pos=1,
point=1, type=2) at /local/xemacs/trunk/src/redisplay.c:5122
#3 0x818da8b in update_line_start_cache (w=0x85bd8f0, from=1,
to=168, point=1, no_regen=0)
at /local/xemacs/trunk/src/redisplay.c:7967
#4 0x818c7c3 in point_in_line_start_cache (w=0x85bd8f0, point=1,
min_past=0) at /local/xemacs/trunk/src/redisplay.c:7277
#5 0x818cda4 in point_would_be_visible (w=0x85bd8f0, startp=1,
point=397) at /local/xemacs/trunk/src/redisplay.c:7454
#6 0x81e76cc in Fpos_visible_in_window_p (pos=795, window=140237040)
at /local/xemacs/trunk/src/window.c:1324
#7 0x80bd4da in Ffuncall (nargs=3, args=0xbfffe520)
at /local/xemacs/trunk/src/eval.c:3477
#8 0x8091e0e in execute_optimized_program (
Note in this context that
/* #### This can be referenced by various functions,
but face_cachels isn't initialized for the stream device.
Since it doesn't need the value we just return nil here to avoid
blowing up in multiple places. */
#define WINDOW_FACE_CACHEL_FONT(window, index, charset) \
((window)->face_cachels \
? FACE_CACHEL_FONT (WINDOW_FACE_CACHEL (window, index), charset) \
And indeed (window)->face_cachels is not 0, but an empty array.
Jan
5:48 p.m.
Jan Vroonhof <jan(a)xemacs.org> writes:
/* #### This can be referenced by various functions,
but face_cachels isn't initialized for the stream device.
Since it doesn't need the value we just return nil here to avoid
blowing up in multiple places. */
#define WINDOW_FACE_CACHEL_FONT(window, index, charset) \
((window)->face_cachels \
? FACE_CACHEL_FONT (WINDOW_FACE_CACHEL (window, index), charset) \
And indeed (window)->face_cachels is not 0, but an empty array.
This whole area needs a audit. The line-start cache suffers from
similar problems on stream devices
Starting program: /local/xemacs/trunk/normal/src/./xemacs -batch -vanilla -f recenter
Program received signal SIGSEGV, Segmentation fault.
0x818d2b4 in start_with_line_at_pixpos (w=0x830e1b8, point=1,
pixpos=4) at /local/xemacs/trunk/src/redisplay.c:7203
7203 Dynarr_reset (w->line_start_cache);
(gdb) where
#0 0x818d2b4 in start_with_line_at_pixpos (w=0x830e1b8, point=1,
pixpos=4) at /local/xemacs/trunk/src/redisplay.c:7203
#1 0x81ef316 in Fcenter_to_window_line (n=137117028,
window=137117028) at /local/xemacs/trunk/src/window.c:4733
#2 0x80bd4da in Ffuncall (nargs=3, args=0xbfffefd8)
at /local/xemacs/trunk/src/eval.c:3477
7:50 a.m.
>>>> "Paul" == Paul Stodghill
<stodghil(a)cs.cornell.edu> writes:
Paul> I managed to reproduce this with 21.4.6 and 21.5.4 under cygwin and 21.4.5
under Linux.
Paul> quimby$ cat foo.el
Paul> (let* ((buf (compile "make"))
Paul> (proc (get-buffer-process buf)))
Paul> (while (= 0 (process-exit-status proc))
Paul> (sleep-for 1))
Paul> (message "%s" (buffer-string buf)))
Paul> quimby$
This does not happen in native Windows 2000 XEmacs.
Evaluating the above in the scratch buffer of my running emacs-version
"21.4 (patch 6) \"Common Lisp (Windows)\" XEmacs Lucid" does not
crash
it.
Instead it hangs in the while loop.
I can get out with C-g, which gives me (after setting debug-on-quit):
Signaling: (quit)
sleep-for(1)
(while (= 0 (process-exit-status proc)) (sleep-for 1))
(let* ((buf ...) (proc ...)) (while (= 0 ...) (sleep-for 1)) (message "%s"
(buffer-string buf)))
eval((let* ((buf ...) (proc ...)) (while (= 0 ...) (sleep-for 1)) (message
"%s" (buffer-string buf))))
eval-interactive((let* ((buf ...) (proc ...)) (while (= 0 ...) (sleep-for 1)) (message
"%s" (buffer-string buf))))
eval-last-sexp(nil)
call-interactively(eval-last-sexp)
Paul> quimby$
Paul> quimby$ xemacs -vanilla -batch -load foo.el
Paul> (No files need saving)
Paul> Fatal error (11).
Paul> Your files have been auto-saved.
Paul> Use `M-x recover-session' to recover them.
Paul> If you have access to the PROBLEMS file that came with your
Paul> version of XEmacs, please check to see if your crash is described
Paul> there, as there may be a workaround available.
Paul> Otherwise, please report this bug by running the send-pr
Paul> script included with XEmacs, or selecting `Send Bug Report'
> from the help menu.
Paul> As a last resort send
ordinary email to `crashes(a)xemacs.org'.
Paul> *MAKE SURE* to include the information in the command
Paul> M-x describe-installation.
Paul> If at all possible, *please* try to obtain a C stack backtrace;
Paul> it will help us immensely in determining what went wrong.
Paul> To do this, locate the core file that was produced as a result
Paul> of this crash (it's usually called `core' and is located in the
Paul> directory in which you started the editor, or maybe in your home
Paul> directory), and type
Paul> gdb /usr/local/bin/xemacs core
Paul> then type `where' when the debugger prompt comes up.
Paul> (If you don't have GDB on your system, you might have DBX,
Paul> or XDB, or SDB. A similar procedure should work for all of
Paul> these. Ask your system administrator if you need more help.)
Paul> Lisp backtrace follows:
Paul> pos-visible-in-window-p(152 #<window on "*compilation*"
0x2436>)
Paul> # bind (buffer-read-only status)
Paul> # (unwind-protect ...)
Paul> # bind (obuf omax opoint estatus window buffer msg proc)
Paul> compilation-sentinel(#<process "compilation" pid 3612
state:exit> "exited abnormally with code 2\n")
Paul> # (condition-case ... . error)
Paul> # bind (inhibit-quit)
Paul> # (unwind-protect ...)
Paul> # (unwind-protect ...)
Paul> sleep-for(1)
Paul> (while (= 0 (process-exit-status proc)) (sleep-for 1))
Paul> # bind (proc buf)
Paul> (let* ((buf ...) (proc ...)) (while (= 0 ...) (sleep-for 1)) (message
"%s" (buffer-string buf)))
Paul> # bind (current-load-list)
Paul> # (unwind-protect ...)
Paul> # bind (load-file-name)
Paul> # (unwind-protect ...)
Paul> # (unwind-protect ...)
Paul> # (unwind-protect ...)
Paul> # (unwind-protect ...)
Paul> # (unwind-protect ...)
Paul> load-internal("/home/stodghil/foo.el" nil t nil undecided)
Paul> # bind (path handler filename nosuffix nomessage noerror file)
Paul> load("/home/stodghil/foo.el" nil t)
Paul> # bind (file arg)
Paul> command-line-do-load("-load")
Paul> # bind (dir file-count line end-of-options file-p arg tem)
Paul> command-line-1()
Paul> # bind (command-line-args-left)
Paul> command-line()
Paul> # (unwind-protect ...)
Paul> normal-top-level()
Paul> # (condition-case ... . error)
Paul> # (catch top-level ...)
Paul> Segmentation fault
Paul> quimby$
--
Adrian Aichner
mailto:adrianï¼ xemacs.org
http://www.xemacs.org/
8342
days inactive
8346
days old
3 comments
3 participants
participants (3)
-
Adrian Aichner -
Jan Vroonhof -
Paul Stodghill