Panic after update: stable/12-n233561-42cb78bdd49 - n233682-a436245ccf1

But only after starting X11.

The former (stable/12-n233561-42cb78bdd49) is from 10 August; the latter
(stable/12-n233682-a436245ccf1) is from today (28 August).  I normally
update this machine (my laptop) daily, but was out of town for a couple
of weeks with fairly spotty connectivity (with a lot of cycling), and
did not attempt updates during that period.

Copies of the crash dump and associated files, as well as the kernel
config file ("CANARY") may be found in
https://www.catwhisker.org/~david/FreeBSD/FreeBSD/stable_12/n233672-a436245ccf1/

The core.txt.0 file claims that the backtrace is:

panic: vm_fault: fault on nofault entry, addr: 0xfffffe10384e2000
cpuid = 7
time = 1630206013
KDB: stack backtrace:
#0 0xffffffff80bebfc5 at kdb_backtrace+0x65
#1 0xffffffff80ba017b at vpanic+0x17b
#2 0xffffffff80b9fff3 at panic+0x43
#3 0xffffffff80f5be62 at vm_fault+0x24f2
#4 0xffffffff80f59850 at vm_fault_trap+0x60
#5 0xffffffff8104fb4c at trap_pfault+0x19c
#6 0xffffffff8104f046 at trap+0x286
#7 0xffffffff81026eb8 at calltrap+0x8
Uptime: 2m18s

This is the most recent instantiation of the crash; following the
initial one, which occurred after the Subject update, but before I
updated any ports (excpt for x11/nvidia-driver, which gets
re-built/-installed every time I build the kernel anyway), I performed
the following series of actions:

* Noted that another laptop did not exhibit this panic after the update
  in question.
* Updated the installed ports (via "portmaster -da").
* Updated the "head" slice on the laptop (from
  main-n248581-4a5a67fe679d - main-n248984-08b9cc316a31; same date range),
  where a very similar (possibly the same) panic occurred.  Note that I
  build the ports under stable/12 and use the misc/compat12x port to be
  able to use the ports thus built under head and stable/13 (during this
  period where I still depend on using stable/12).
* Noticed that I had 'WITHOUT_DEBUG_FILES=1' in /etc/src.conf, commented
  it out, and re-built/-installed stable/12.
* Explicitly re-built/installed x11/xdm and x11/xorg.
* Explicitly re-built/installed all installed ports that matched x11/lib*.
* Since the other laptop uses vt, rather than sc for kern.vty (ref.
  /boot/loader.conf), tried commenting out the 'kern.vty="sc"' line and
  rebooting.  This at least permitted me to see the console panic message.

A very similar (possibly "the same") panic occurred after each of these
(save the first, "Noted that..." entry).

I expect to defer updating my critical machines (here at home) until
this is resolved -- I normally update them every Sunday, following a
successful update on my laptop & the build machine.  (This is not what I
would call a "successful update" for my laptop. :-} )

So: any hints?  (Yes, I did check src/UPDATING; most recent entry is
20210730, and about pf, which I do not use.)

Thanks!

Peace,
david
-- 
David H. Wolfskill                              david@catwhisker.org
Life is not intended to be a zero-sum game.

See https://www.catwhisker.org/~david/publickey.gpg for my public key.