GNU bug report logs - #76970
31.0.50; master emacs crash with stack overflow

Previous Next

Full log

Message #11 received at 76970 <at> debbugs.gnu.org (full text, mbox):

From: Eval Exec <execvy <at> gmail.com>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 76970 <at> debbugs.gnu.org
Subject: Re: bug#76970: 31.0.50; master emacs crash with stack overflow
Date: Wed, 12 Mar 2025 22:00:06 +0800

> Would it be possible for you to run Emacs under GDB at all times?

it's possible.

> Any idea what Emacs was doing when this happened?

I forget, maybe I'm doing `(profiler-start)` that times? I'm not sure.

On Wed, Mar 12, 2025 at 9:56 PM Eli Zaretskii <eliz <at> gnu.org> wrote:
>
> > From: Eval Exec <execvy <at> gmail.com>
> > Date: Wed, 12 Mar 2025 10:43:59 +0800
> >
> >
> > Hello, I got a master branch emacs crash:
> >
> > (gdb) bt full
> > #0  0x000000000055979e in stack_overflow (siginfo=0x85d1b0 <sigsegv_stack+62640>) at /home/exec/Projects/git.savannah.gnu.org/git/emacs/src/sysdep.c:1895
> >         addr = 0x70 <error: Cannot access memory at address 0x70>
> >         bot = <optimized out>
> >         top = <optimized out>
> >         LG_STACK_HEURISTIC = LG_STACK_HEURISTIC
> > #1  handle_sigsegv (sig=11, siginfo=0x85d1b0 <sigsegv_stack+62640>, arg=<optimized out>) at /home/exec/Projects/git.savannah.gnu.org/git/emacs/src/sysdep.c:1930
> >         fatal = false
> > #2  <signal handler called>
> > No symbol table info available.
> > #3  backtrace_top () at /home/exec/Projects/git.savannah.gnu.org/git/emacs/src/eval.c:174
> >         pdl = <optimized out>
> >         pdl = <optimized out>
> > #4  backtrace_top_function () at /home/exec/Projects/git.savannah.gnu.org/git/emacs/src/eval.c:4270
> >         pdl = <optimized out>
> >         pdl = <optimized out>
> > #5  add_sample (count=1, plog=<optimized out>) at /home/exec/Projects/git.savannah.gnu.org/git/emacs/src/profiler.c:328
> > No locals.
> > #6  handle_profiler_signal (signal=27) at /home/exec/Projects/git.savannah.gnu.org/git/emacs/src/profiler.c:380
> >         count = 1
> >         count = <optimized out>
> >         overruns = <optimized out>
> > #7  deliver_process_signal (handler=<optimized out>, sig=27) at /home/exec/Projects/git.savannah.gnu.org/git/emacs/src/sysdep.c:1751
> >         old_errno = 1
> >         on_main_thread = true
> >         old_errno = <optimized out>
> >         on_main_thread = <optimized out>
> >         blocked = <optimized out>
> > #8  deliver_profiler_signal (signal=27) at /home/exec/Projects/git.savannah.gnu.org/git/emacs/src/profiler.c:386
> > No locals.
> > Backtrace stopped: Cannot access memory at address 0x7ffe5c5f2678
> > You can't do that without a process to debug.
> > (gdb)
>
> Is this the main thread ("thread 1"), or some other thread?  What does
> "info threads" say?
>
> Would it be possible for you to run Emacs under GDB at all times?
> That would allow to do more when Emacs crashes, because some things
> GDB cannot do when all it has is the core file, as opposed to a live
> process.  E.g., the above backtrace doesn't give a clue what caused
> stack overflow, because it shows too few stack frames.
>
> Any idea what Emacs was doing when this happened?

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.