Bug#1017711: emacs-gtk: terminated with signal SIGABRT, 137 MB coredump

[Sean Whitton]

Hello Vincent,

Upstream says there isn't enough information in the backtrace to say
anything helpful about this.  Could you take a look at
 and consider supplying more information
over there, please?

Also, are you able to reproduce this with 'emacs -q' (not -Q)?

Currently we have no way to reproduce this, and you're the only person
who's seen anything like this, so we'll probably have to move the
severity back down to 'important'.

-- 
Sean Whitton


signature.asc
Description: PGP signature

Bug#1017711: emacs-gtk: terminated with signal SIGABRT, 137 MB coredump

[Sean Whitton]

Hello,

On Tue 01 Nov 2022 at 11:17PM +01, Vincent Lefevre wrote:

> On 2022-10-31 15:59:17 +0100, Vincent Lefevre wrote:
>> Hi,
>>
>> On 2022-10-30 06:31:27 -0700, Sean Whitton wrote:
>> > I've backported a couple of patches from upstream related to native
>> > compilation, now, so would you be able to test again with the latest
>> > upload?  Thanks.
>>
>> If you mean 1:28.2+1-3, I still got the same issue. But it seems
>> that after a reboot, the problem no longer occurs (after removing
>> the .emacs.d directory). Was a reboot needed?
>
> Unfortunately, it has just occurred again (with emacs run by
> Subversion to enter a commit message).
>
> Core was generated by `/usr/bin/emacs -no-comp-spawn --batch -l 
> /tmp/emacs-async-comp-url.el-FGov4z.el'.
> Program terminated with signal SIGABRT, Aborted.
> #0  __pthread_kill_implementation (threadid=, 
> signo=signo@entry=6, no_tid=no_tid@entry=0) at ./nptl/pthread_kill.c:44
> 44  ./nptl/pthread_kill.c: No such file or directory.
>
> I've attached the backtrace. Seems too many recursive calls to
> mark_object and mark_objects.
>
> Worse, not just a background emacs process, the UI also crashes
> (IIRC, this didn't happen before).

Thanks for following up.

-- 
Sean Whitton


signature.asc
Description: PGP signature

Bug#1017711: emacs-gtk: terminated with signal SIGABRT, 137 MB coredump

[Vincent Lefevre]

On 2022-10-31 15:59:17 +0100, Vincent Lefevre wrote:
> Hi,
> 
> On 2022-10-30 06:31:27 -0700, Sean Whitton wrote:
> > I've backported a couple of patches from upstream related to native
> > compilation, now, so would you be able to test again with the latest
> > upload?  Thanks.
> 
> If you mean 1:28.2+1-3, I still got the same issue. But it seems
> that after a reboot, the problem no longer occurs (after removing
> the .emacs.d directory). Was a reboot needed?

Unfortunately, it has just occurred again (with emacs run by
Subversion to enter a commit message).

Core was generated by `/usr/bin/emacs -no-comp-spawn --batch -l 
/tmp/emacs-async-comp-url.el-FGov4z.el'.
Program terminated with signal SIGABRT, Aborted.
#0  __pthread_kill_implementation (threadid=, 
signo=signo@entry=6, no_tid=no_tid@entry=0) at ./nptl/pthread_kill.c:44
44  ./nptl/pthread_kill.c: No such file or directory.

I've attached the backtrace. Seems too many recursive calls to
mark_object and mark_objects.

Worse, not just a background emacs process, the UI also crashes
(IIRC, this didn't happen before).

-- 
Vincent Lefèvre  - Web: 
100% accessible validated (X)HTML - Blog: 
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

Thread 1 (Thread 0x7f5b914cb380 (LWP 35005)):
#0  __pthread_kill_implementation (threadid=, 
signo=signo@entry=6, no_tid=no_tid@entry=0) at ./nptl/pthread_kill.c:44
tid = 
ret = 0
pd = 
old_mask = {__val = {0 }}
ret = 
#1  0x7f5b922895df in __pthread_kill_internal (signo=6, threadid=) at ./nptl/pthread_kill.c:78
#2  0x7f5b9223da02 in __GI_raise (sig=sig@entry=6) at 
../sysdeps/posix/raise.c:26
ret = 
#3  0x55fdfd114864 in terminate_due_to_signal (sig=sig@entry=6, 
backtrace_limit=backtrace_limit@entry=40) at ./debian/build-src/src/emacs.c:437
#4  0x55fdfd114d27 in emacs_abort () at ./debian/build-src/src/sysdep.c:2282
#5  0x55fdfd111e99 in check_message_stack () at 
./debian/build-src/src/xdisp.c:12157
#6  0x55fdfd20492a in shut_down_emacs (sig=0, stuff=0x0) at 
./debian/build-src/src/emacs.c:2789
tpgrp = 
#7  0x55fdfd114765 in Fkill_emacs (arg=arg@entry=0x6) at 
./debian/build-src/src/emacs.c:2692
exit_code = 
#8  0x55fdfd114827 in terminate_due_to_signal (sig=sig@entry=1, 
backtrace_limit=backtrace_limit@entry=40) at ./debian/build-src/src/emacs.c:417
#9  0x55fdfd114d00 in handle_fatal_signal (sig=sig@entry=1) at 
./debian/build-src/src/sysdep.c:1762
#10 0x55fdfd114d17 in deliver_process_signal (handler=0x55fdfd114cf5 
, sig=1) at ./debian/build-src/src/sysdep.c:1720
old_errno = 2
on_main_thread = true
#11 deliver_fatal_signal (sig=1) at ./debian/build-src/src/sysdep.c:1768
#12 0x7f5b9223daa0 in  () at 
/lib/x86_64-linux-gnu/libc.so.6
#13 0x55fdfd2641c3 in mark_object (arg=0x295d90329a68) at 
./debian/build-src/src/alloc.c:6628
obj = 0x295d90329a68
po = 
cdr_count = 0
#14 0x55fdfd26528e in mark_objects (n=3, obj=0x7f5b8e029930) at 
./debian/build-src/src/alloc.c:6607
i = 0
ptr = 0x7f5b8e029928
size = 3
#15 mark_vectorlike (header=0x7f5b8e029928) at 
./debian/build-src/src/alloc.c:6382
ptr = 0x7f5b8e029928
size = 3
#16 0x55fdfd26528e in mark_objects (n=3, obj=0x7f5b8e0298c0) at 
./debian/build-src/src/alloc.c:6607
i = 2
ptr = 0x7f5b8e0298b8
size = 3
#17 mark_vectorlike (header=0x7f5b8e0298b8) at 
./debian/build-src/src/alloc.c:6382
ptr = 0x7f5b8e0298b8
size = 3
#18 0x55fdfd26528e in mark_objects (n=711, obj=0x7f5b8e027fc8) at 
./debian/build-src/src/alloc.c:6607
i = 614
ptr = 0x7f5b8e027fc0
size = 711
#19 mark_vectorlike (header=0x7f5b8e027fc0) at 
./debian/build-src/src/alloc.c:6382
ptr = 0x7f5b8e027fc0
size = 711
#20 0x55fdfd26528e in mark_objects (n=7, obj=0x7f5b8da7eab8) at 
./debian/build-src/src/alloc.c:6607
i = 5
ptr = 0x7f5b8da7eab0
size = 7
#21 mark_vectorlike (header=0x7f5b8da7eab0) at 
./debian/build-src/src/alloc.c:6382
ptr = 0x7f5b8da7eab0
size = 7
#22 0x55fdfd2646ed in mark_object (arg=) at 
./debian/build-src/src/alloc.c:6787
subr = 0x7f5b8df6d260
ptr = 0x7f5b8df6d260
pvectype = 
obj = 0x7f5b8df6d265
po = 0x7f5b8df6d260
cdr_count = 0
#23 0x55fdfd2642d3 in mark_object (arg=) at 
./debian/build-src/src/alloc.c:6815
ptr = 0x7f5b8df6d230
obj = 
po = 0x7f5b8df6d230
cdr_count = 
#24 0x55fdfd26528e in mark_objects (n=5, obj=0x55fdff0253d8) at 
./debian/build-src/src/alloc.c:6607
i = 3
ptr = 0x55fdff0253d0
size = 5
#25 mark_vectorlike (header=0x55fdff0253d0) at 
./debian/build-src/src/alloc.c:6382
ptr = 0x55fdff0253d0
size = 5
#26 0x55fdfd2642d3 in mark_object (arg=) at 

Bug#1017711: emacs-gtk: terminated with signal SIGABRT, 137 MB coredump

[Vincent Lefevre]

Hi,

On 2022-10-30 06:31:27 -0700, Sean Whitton wrote:
> I've backported a couple of patches from upstream related to native
> compilation, now, so would you be able to test again with the latest
> upload?  Thanks.

If you mean 1:28.2+1-3, I still got the same issue. But it seems
that after a reboot, the problem no longer occurs (after removing
the .emacs.d directory). Was a reboot needed?

-- 
Vincent Lefèvre  - Web: 
100% accessible validated (X)HTML - Blog: 
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

Bug#1017711: emacs-gtk: terminated with signal SIGABRT, 137 MB coredump

[Sean Whitton]

Hello,

On Fri 14 Oct 2022 at 12:53PM +02, Vincent Lefevre wrote:

> Control: found -1 1:28.2+1-1
>
> Hi,
>
> On 2022-10-13 22:28:24 -0700, Sean Whitton wrote:
>> Can you reproduce this with what's in sid now?
>
> Yes, this still occurs:
>
> zira% ls -lh
> total 41M
> -rw--- 1 vinc17 vinc17 43M 2022-10-14 12:49:40 core
>
> Core was generated by `/usr/bin/emacs --batch -l 
> /tmp/emacs-async-comp-url-cookie.el-A86dRV.el'.
> Program terminated with signal SIGABRT, Aborted.
> #0  0x7f59efe8957c in ?? () from /lib/x86_64-linux-gnu/libc.so.6

I've backported a couple of patches from upstream related to native
compilation, now, so would you be able to test again with the latest
upload?  Thanks.

-- 
Sean Whitton


signature.asc
Description: PGP signature

Processed: Re: Bug#1017711: emacs-gtk: terminated with signal SIGABRT, 137 MB coredump

[Debian Bug Tracking System]

Processing control commands:

> found -1 1:28.2+1-1
Bug #1017711 [emacs-gtk] emacs-gtk: terminated with signal SIGABRT, 137 MB 
coredump
Marked as found in versions emacs/1:28.2+1-1.

-- 
1017711: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1017711
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems

Bug#1017711: emacs-gtk: terminated with signal SIGABRT, 137 MB coredump

[Vincent Lefevre]

Control: found -1 1:28.2+1-1

Hi,

On 2022-10-13 22:28:24 -0700, Sean Whitton wrote:
> Can you reproduce this with what's in sid now?

Yes, this still occurs:

zira% ls -lh
total 41M
-rw--- 1 vinc17 vinc17 43M 2022-10-14 12:49:40 core

Core was generated by `/usr/bin/emacs --batch -l 
/tmp/emacs-async-comp-url-cookie.el-A86dRV.el'.
Program terminated with signal SIGABRT, Aborted.
#0  0x7f59efe8957c in ?? () from /lib/x86_64-linux-gnu/libc.so.6

-- 
Vincent Lefèvre  - Web: 
100% accessible validated (X)HTML - Blog: 
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)

Bug#1017711: emacs-gtk: terminated with signal SIGABRT, 137 MB coredump

[Sean Whitton]

Hello,

On Mon 22 Aug 2022 at 02:55AM +02, Vincent Lefevre wrote:

> Control: found -1 1:28.1+1-2
> Control: severity -1 serious
>
> On 2022-08-19 12:10:31 +0200, Vincent Lefevre wrote:
>> I have noticed that Emacs left a 137 MB coredump.
>
> This has occurred again (this time a 54 MB coredump).
> This is going to use much disk space...
>
>> gdb says:
>>
>> Reading symbols from /usr/bin/emacs...
>> Reading symbols from 
>> /usr/lib/debug/.build-id/d0/b7c40dc33110b0623ea2ca797a6d3f3eb166b5.debug...
>> [New LWP 1483812]
>> [New LWP 1483816]
>> [Thread debugging using libthread_db enabled]
>> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
>> Core was generated by `/usr/bin/emacs --batch -l 
>> /tmp/emacs-async-comp-cc-engine.el-2UV1nf.el'.
>
> This time,
>
> Core was generated by `/usr/bin/emacs --batch -l 
> /tmp/emacs-async-comp-auth-source.el-84YT0a.el'.
> Program terminated with signal SIGABRT, Aborted.
>
> I'm wondering whether this is related to bug 1017845, as a
> "/usr/bin/emacs --batch -l /tmp/emacs-async-..." is also involved.

Can you reproduce this with what's in sid now?

It may have been fixed by adding the emacs-el dependency.

Thanks.

-- 
Sean Whitton


signature.asc
Description: PGP signature