[STUMP] hangs and crashes?

2 views
Skip to first unread message

Shawn Betts

unread,
May 14, 2007, 10:35:37 PM5/14/07
to stumpw...@nongnu.org
Hi folks,

There are 2 bugs that I'm trying to figure out. If you are
experiencing one of these bugs, can you post the version of sbcl
you're using (so far i've only seen it on sbcl), the version of clx,
and uname -a, and anything else that you feel is relevant? I want to
see if these errors only occur on certain architectures or certain
versions of sbcl or...something that ties them together.

Thanks!

1. modeline crash bug

The error is: Caught 'attempt to RETURN-FROM a block or GO to a tag that no longer exists' at the top

And the meat of the backtrace looks like this:

7: ("foreign function: call_into_lisp")
8: ("foreign function: funcall2")
9: ("foreign function: interrupt_internal_error")
10: ("foreign function: handle_trap")
11: ("no debug information for frame")
12: (SIGNAL #<SB-SYS:MEMORY-FAULT-ERROR {B9F8899}>)
13: (ERROR SB-SYS:MEMORY-FAULT-ERROR)
14: (SB-SYS:MEMORY-FAULT-ERROR)
15: (SB-SYS:MEMORY-FAULT-ERROR)
16: ("foreign function: call_into_lisp")
17: ("foreign function: post_signal_tramp")
18: ((SB-PCL::FAST-METHOD REDRAW-MODE-LINE-FOR (T SCREEN))

2. stumpwm hang

There is no error, stumpwm just becomes unresponsive. The keyboard is
frozen to the point where switching to another console doesn't work
(control+alt+F1). Killing stumpwm unfreezes the keyboard. If you
manage to get a backtrace it looks like this:

1: (SB-SYS:INVOKE-INTERRUPTION #<CLOSURE (LAMBDA #) {49C8EAF5}>)
2: ("foreign function: call_into_lisp")
3: ("foreign function: funcall3")
4: ("foreign function: interrupt_handle_now")
5: ("foreign function: maybe_defer_handler")
6: ("bogus stack frame")
7: (SB-SYS:WAIT-UNTIL-FD-USABLE 5 :INPUT NIL)
8: (XLIB::BUFFER-INPUT-WAIT-DEFAULT
#<XLIB:DISPLAY :0 (The X.Org Foundation R60900000)>
NIL)
9: (XLIB::BUFFER-INPUT-WAIT
#<XLIB:DISPLAY :0 (The X.Org Foundation R60900000)>
NIL)
10: (XLIB::READ-INPUT
#<XLIB:DISPLAY :0 (The X.Org Foundation R60900000)>
NIL
NIL
#<FUNCTION (LAMBDA #) {49855D85}>)
11: (XLIB::READ-REPLY #<unavailable argument> #<unavailable argument>)
12: (XLIB:DISPLAY-FINISH-OUTPUT
#<XLIB:DISPLAY :0 (The X.Org Foundation R60900000)>)
13: (STUMPWM::STUMPWM-INTERNAL-LOOP)
14: (STUMPWM:STUMPWM ":0" NIL)

There are instructions on getting a backtrace for this bug here:

http://stumpwm.antidesktop.net/cgi-bin/wiki/Freezes

-Shawn


_______________________________________________
Stumpwm-devel mailing list
Stumpw...@nongnu.org
http://lists.nongnu.org/mailman/listinfo/stumpwm-devel

Tassilo Horn

unread,
May 15, 2007, 2:24:37 AM5/15/07
to stumpw...@nongnu.org
Shawn Betts <sab...@vcn.bc.ca> writes:

Hi Shawn,

> 1. modeline crash bug
>
> The error is: Caught 'attempt to RETURN-FROM a block or GO to a tag
> that no longer exists' at the top
>
> And the meat of the backtrace looks like this:
>
> 7: ("foreign function: call_into_lisp")
> 8: ("foreign function: funcall2")
> 9: ("foreign function: interrupt_internal_error")
> 10: ("foreign function: handle_trap")
> 11: ("no debug information for frame")
> 12: (SIGNAL #<SB-SYS:MEMORY-FAULT-ERROR {B9F8899}>)
> 13: (ERROR SB-SYS:MEMORY-FAULT-ERROR)
> 14: (SB-SYS:MEMORY-FAULT-ERROR)
> 15: (SB-SYS:MEMORY-FAULT-ERROR)
> 16: ("foreign function: call_into_lisp")
> 17: ("foreign function: post_signal_tramp")
> 18: ((SB-PCL::FAST-METHOD REDRAW-MODE-LINE-FOR (T SCREEN))

Ah, that's mine.

,----[ eix sbcl ]
| [I] dev-lisp/sbcl
| [...]
| Installed versions: 1.0.5 (doc -ldb -source threads unicode)
| [...]
`----

But the bug occured with 1.0.4, too.

,----[ eix clx ]
| [I] dev-lisp/cl-clx
| [...]
| Installed versions: 0.7.3
| [...]
`----

,----[ uname -a ]
| Linux baldur 2.6.21-gentoo #1 SMP PREEMPT Wed May 9 10:08:07 CEST 2007
| i686 Genuine Intel(R) CPU T2300 @ 1.66GHz GenuineIntel GNU/Linux
`----

Bye,
Tassilo
--
If programmers deserve to be rewarded for creating innovative programs,
by the same token they deserve to be punished if they restrict the use
of these programs. (Richard M. Stallman)

Tassilo Horn

unread,
May 15, 2007, 2:25:17 AM5/15/07
to stumpw...@nongnu.org
Shawn Betts <sab...@vcn.bc.ca> writes:

Hi Shawn,

> 1. modeline crash bug


>
> The error is: Caught 'attempt to RETURN-FROM a block or GO to a tag
> that no longer exists' at the top
>
> And the meat of the backtrace looks like this:
>
> 7: ("foreign function: call_into_lisp")
> 8: ("foreign function: funcall2")
> 9: ("foreign function: interrupt_internal_error")
> 10: ("foreign function: handle_trap")
> 11: ("no debug information for frame")
> 12: (SIGNAL #<SB-SYS:MEMORY-FAULT-ERROR {B9F8899}>)
> 13: (ERROR SB-SYS:MEMORY-FAULT-ERROR)
> 14: (SB-SYS:MEMORY-FAULT-ERROR)
> 15: (SB-SYS:MEMORY-FAULT-ERROR)
> 16: ("foreign function: call_into_lisp")
> 17: ("foreign function: post_signal_tramp")
> 18: ((SB-PCL::FAST-METHOD REDRAW-MODE-LINE-FOR (T SCREEN))

Ah, that's mine.

,----[ eix sbcl ]
| [I] dev-lisp/sbcl
| [...]
| Installed versions: 1.0.5 (doc -ldb -source threads unicode)
| [...]
`----

But the bug occured with 1.0.4, too.

,----[ eix clx ]
| [I] dev-lisp/cl-clx
| [...]
| Installed versions: 0.7.3
| [...]
`----

,----[ uname -a ]
| Linux baldur 2.6.21-gentoo #1 SMP PREEMPT Wed May 9 10:08:07 CEST 2007
| i686 Genuine Intel(R) CPU T2300 @ 1.66GHz GenuineIntel GNU/Linux
`----

Bye,
Tassilo
--
If programmers deserve to be rewarded for creating innovative programs,
by the same token they deserve to be punished if they restrict the use
of these programs. (Richard M. Stallman)

Antonis Antoniadis

unread,
May 15, 2007, 5:47:38 AM5/15/07
to stumpw...@nongnu.org
Hello,

I have the freeze too, here is my data:
sbcl: 1.0.4
clx: 0.7.3
uname -a:  FreeBSD fbsd 6.2-STABLE FreeBSD 6.2-STABLE #0: Sun Apr 15 15:57:50
       EEST 2007     plutonas@fbsd:/usr/obj/usr/src/sys/GENERIC  i386

bye,
plutonas

Tassilo Horn

unread,
May 15, 2007, 6:26:10 AM5/15/07
to stumpw...@nongnu.org
Tassilo Horn <tas...@member.fsf.org> writes:

Hi,

> ,----[ eix sbcl ]
> | [I] dev-lisp/sbcl
> | [...]
> | Installed versions: 1.0.5 (doc -ldb -source threads unicode)
> | [...]
> `----
>
> But the bug occured with 1.0.4, too.
>
> ,----[ eix clx ]
> | [I] dev-lisp/cl-clx
> | [...]
> | Installed versions: 0.7.3
> | [...]
> `----
>
> ,----[ uname -a ]
> | Linux baldur 2.6.21-gentoo #1 SMP PREEMPT Wed May 9 10:08:07 CEST 2007
> | i686 Genuine Intel(R) CPU T2300 @ 1.66GHz GenuineIntel GNU/Linux
> `----

I just wanted to add that I use X.org 7.2, if that matters.

Bye,
Tassilo
--
Chuck Norris can be unlocked on the hardest level of Tekken. But only
Chuck Norris is skilled enough to unlock himself. Then he roundhouse
kicks the Playstation back to Japan.

Lionel Flandrin

unread,
May 16, 2007, 5:08:55 AM5/16/07
to stumpw...@nongnu.org
On 5/15/07, Antonis Antoniadis <cereal.ki...@gmail.com> wrote:
> Hello,
>
> I have the freeze too, here is my data:
> sbcl: 1.0.4
> clx: 0.7.3
> uname -a: FreeBSD fbsd 6.2-STABLE FreeBSD 6.2-STABLE #0: Sun Apr 15
> 15:57:50
> EEST 2007 plutonas@fbsd:/usr/obj/usr/src/sys/GENERIC i386
>
> bye,
> plutonas

No freeze since several months here on both linux and freebsd 6.1

sbcl: 1.0.5 (rebuilt from source)
clx: 0.7.3
on linux 2.6.17 and freebsd 6.1 (STABLE)

(I hate gmail, i hope this mail will be correct...)

--
Simias

Tim hobbs

unread,
May 19, 2007, 12:21:47 AM5/19/07
to stumpw...@nongnu.org
I have continued to get the keyboard freeze though I'm not having trouble with it during idle any more.  I think I've had it since you merged that one window manager compat protocol patch.  I have tracked down the start of the freeze, it is upon varios window initializations, such as when you start xmms from cron.
--
-
Tim
tim.t...@gmail.com


--
-
Tim
tim.t...@gmail.com
Reply all
Reply to author
Forward
0 new messages