bug#31233: 26.1; Emacs terminated unexpectedly

[nljlistbox2@gmail.com (N. Jackson)]

At 19:51 +0300 on Sunday 2018-04-22, Eli Zaretskii wrote:
>
> Hmm... signal 7 is SIGBUS on your system, right?

Yes, that's right.

> So this sounds like one of the pointers in the red-black tree is not
> aligned correctly.

Actually, it looks like this is probably nothing to do with Emacs but is
rather a serious problem with my system. [Looking back in the logs it
seems that for the past week or two, at the moment my system resumes
from suspend, sometimes a program will terminate with SIGBUS following
an ANOM_ABEND error (whatever that is) -- which is exactly what happened
to Emacs yesterday. I hadn't noticed the previous events, but it
happened again this afternoon and this time the program affected was
Xorg, and I certainly noticed that. Awful.]

For what it's worth the logged message when Emacs terminated yesterday
was:

  Apr 21 20:04:54 moondust.localdomain audit[3617]: ANOM_ABEND auid=1000 uid=1000 gid=1000 ses=2 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 pid=3617 comm="emacs" exe="/data/projects/vc/emacs/git/emacs/src/emacs" sig=7 res=1

> I think we will need to see which pointer is that and what's its
> relation to the root of the tree and to mem_z sentinel node concocted
> by mem_find. You should be able to show these values when/if the crash
> happens again.

I will continue to run Emacs under gdb and report if it crashes again.
Meanwhile, do you want to close this bug, or should we keep it open for
a little while?

> Thanks.

Thank you.

Regards,
N.