XDarwin 1.0a3 running on 10.1 on DP 533
machine:
Both rooted and rootless servers consistently
cause kernel panics when running xemacs. The
beginning text of the kernel panic refers to a
problem in lockmgr.
This is probably a very good thing to report to Apple. If
xemacs is running without root privileges, it is clearly a
kernel problem. Non-root user land apps shouldn't be able to
panic the kernel (period). For a bug report to be useful, you
should try to copy the full message you get from the kernel
panic. Unfortunately this means copying down by hand. If you
don't have any good way to report bugs to Apple, you can post
it here and perhaps we can reproduce it/send it on.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm the original poster of the bug. Update: I get similar kernel panics running xemacs under Xtools. OS X is otherwise rock-solid for me, so I suspect that there is some obscure multithreading bug in the kernel that is exposed only by xemacs on DP machines.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I've send on the complete kernel panic message to the Darwin Developers mailing list. This is apparently a common problem with xemacs and dual processor machines. A kernel patch was suggested, which may address this problem, so I would guess that a fix will be forthcoming. Thanks for the full kernel panic message as earlier reports only included the first few lines which isn't enough to go on.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have had EXACTLY the same problem (G4 DP 450). It seems that xemacs + the XAqua usually cause kernel
panics within minutes of use. Nothing else I run seems to trigger the problem.
However, since I'm a long time xemacs user (and programmer), this problem is extremely irritating. I just
searched the Darwin archives, and looks like there hasn't been any news of a fix or work-around.
Anyone know of any?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We've seen the problem with Xemacs and regular
emacs running in a terminal window. The patched
kernel won't even boot on a DP 800. I'm inclined to think
it's a problem with lisp, but who knows, the kernel
should not be vulnerable to panics caused by a user
program. As painful as it is, the only work around we
have found is to use some other editor, nedit, bbedit,
gvim or whatever. This is a problem that Apple will have
to fix.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I am now running fink 0.3.1 and XDarwin 1.0.4. Last
night I was able to run Xemacs for a few minutes
without a problem. This morning, I got the same kernel
panic immediately upon starting Xemacs. So I guess I
can't agree with jcbertin
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
From what I can see it looks worth adding my own
experiences here. I have also gotten kernel panics on
a dual monitor, DP800 with 10.1.1 while using emacs
(NOT xemacs) in an Xwindow in rootless mode. The
panic always seems to happen when I am using the
mouse to select text in the emacs window. It happens
WHILE I am dragging to select the text. I did not see
the problem with 10.1, but I also was not using emacs
very much during that time.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Logged In: YES
user_id=133579
This is probably a very good thing to report to Apple. If
xemacs is running without root privileges, it is clearly a
kernel problem. Non-root user land apps shouldn't be able to
panic the kernel (period). For a bug report to be useful, you
should try to copy the full message you get from the kernel
panic. Unfortunately this means copying down by hand. If you
don't have any good way to report bugs to Apple, you can post
it here and perhaps we can reproduce it/send it on.
Logged In: NO
I have had a similar problem with XDarwin 1.0a3 on
10.1 on a DP 800 and DP 533 while running emacs
and xemacs:
I got the following panic just now with emacs:
panic (cpu0): lockmgr: pid -2, thread 0x 26e32f0, not
exclusive lock holder pid -2 thread 0x 1d5b7c8
unlocking, exclusive count 1
Latest stack backtrace for cpu 0:
Backtrace:
0x0008122c 0x000815cc 0x00027d38 0x001831b0
0x001886c8 0x00188bf8 0x00026794 0x0008ec0c
0x00191168 0x0018949c 0x0018931c
Proceeding back via exception chain:
Exception state (sv=0x117f8000)
PC=0x70025CA8, MSR=0x0200d030; DAR=
0x007c81de; DSISR=0x40000000; LR=0x001a3ebc;
R1=0xbfffd580; XCP=0x00000030 (0xC00-System call)
Kernel version:
Darwin Kernel version 1.4:
Sun Sep 9 15:39:59 PDT 2001; root:xnu/xnu-201.obj~1/
RELEASE_PPC
Logged In: YES
user_id=341864
I'm the original poster of the bug. Update: I get similar kernel panics running xemacs under Xtools. OS X is otherwise rock-solid for me, so I suspect that there is some obscure multithreading bug in the kernel that is exposed only by xemacs on DP machines.
Logged In: YES
user_id=133579
I've send on the complete kernel panic message to the Darwin Developers mailing list. This is apparently a common problem with xemacs and dual processor machines. A kernel patch was suggested, which may address this problem, so I would guess that a fix will be forthcoming. Thanks for the full kernel panic message as earlier reports only included the first few lines which isn't enough to go on.
Logged In: YES
user_id=155842
I have had EXACTLY the same problem (G4 DP 450). It seems that xemacs + the XAqua usually cause kernel
panics within minutes of use. Nothing else I run seems to trigger the problem.
However, since I'm a long time xemacs user (and programmer), this problem is extremely irritating. I just
searched the Darwin archives, and looks like there hasn't been any news of a fix or work-around.
Anyone know of any?
Logged In: YES
user_id=226250
We've seen the problem with Xemacs and regular
emacs running in a terminal window. The patched
kernel won't even boot on a DP 800. I'm inclined to think
it's a problem with lisp, but who knows, the kernel
should not be vulnerable to panics caused by a user
program. As painful as it is, the only work around we
have found is to use some other editor, nedit, bbedit,
gvim or whatever. This is a problem that Apple will have
to fix.
Logged In: YES
user_id=123278
im wondering if someone already tried with 10.1.1 ?
Logged In: YES
user_id=123278
I update to fink 0.3.1 and XDarwin 1.0.4 and the bug looks
vanished...
Is somebody agree with me ???
Logged In: YES
user_id=226250
I am now running fink 0.3.1 and XDarwin 1.0.4. Last
night I was able to run Xemacs for a few minutes
without a problem. This morning, I got the same kernel
panic immediately upon starting Xemacs. So I guess I
can't agree with jcbertin
Logged In: YES
user_id=123278
Works with remote X and was able to run xemacs for days now.
I have to try it locally...
Logged In: YES
user_id=123278
The bug looks closed on Darwin's Radar. Maybe next update...
Logged In: YES
user_id=226250
IThis problem is "believed" to be fixed in 10.1.2 which
according to the rumor sites is due out "any day now".
Phil
Logged In: NO
Hi,
From what I can see it looks worth adding my own
experiences here. I have also gotten kernel panics on
a dual monitor, DP800 with 10.1.1 while using emacs
(NOT xemacs) in an Xwindow in rootless mode. The
panic always seems to happen when I am using the
mouse to select text in the emacs window. It happens
WHILE I am dragging to select the text. I did not see
the problem with 10.1, but I also was not using emacs
very much during that time.
Logged In: YES
user_id=133579
Now that Mac OS X 10.1.2 is out, are there any further cases
of this kernel panic with 10.1.2?
Logged In: YES
user_id=12935
I had a report that this is fixed in 10.1.2, by one user on
the fink-users mailing list.