bug#51336: 29.0.50; Application not responding

[Kirill Korinsky]

Greetings,

On Fri, 12 Jan 2024 19:31:53 +0100,
Alan Third wrote:
>
> On Fri, Jan 12, 2024 at 08:14:18PM +0200, Eli Zaretskii wrote:
> >
> > I'm confused: 6acb3c5b05a was installed on master, not on the
> > emacs-29 branch.  AFAICT, Emacs 29 still has the code which that
> > commit removed.  And yet the bug report is for Emacs 29(?).  Am I
> > missing something?
>
> Kirill specified commit d9462e24a967e32d550ee886b5150f0cc78358f6 which
> is, I think, in the master branch.

Yes, I usually runs on something from master which I pick each months or so.

All my picks usually is distributed as `emacs-app-devel` in MacPorts.

> TBH, I'm not sure what's going on with this bug report, but I'm sure
> the current hang is the same as the hang in the other bug report.
> Unless I've misunderstood.

I've made some research in my mail archive and was able to track.

Somewhen before Emacs 29 I've encountered a crash which was reproted as 50641

After it was fixed 50534 had appeared, which also was fixed.

And after this two fixed I've start to seen on near the same setup this issue.

Before switch d9462e24a967e32d550ee886b5150f0cc78358f6 I've used
e32c57ed4d36c5c0302eeb409f96ce9155b545ea for a while and it's behave "classic"
for this bug.

Classic behaviour when I've clicked on Emacs icon inside docker by right click
and the first item on this list "Application not responding".

Sometimes Emacs freezes, but it was not that often.

I really thought that the root cause is magit with MacPorts repository, which
had used huge sqlite DB for keep all PRs.

Anyway, that file were removed and for week I haven't used it => so, it 
innocent.

Before that I've upgrade to d9462e24a967e32d550ee886b5150f0cc78358f6 and the
hell is started.

The hell that I kill emacs a few times per day because it's frozen completley.

During writing this email (~10 minutes?) it had frozen for a few seconds with
logs like this in *Message* buffer:

Error during redisplay: (jit-lock-function 626) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 627) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 633) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 674) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 675) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 682) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 686) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 687) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 688) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 687) signaled (args-out-of-range "" 
0)
Error during redisplay: (jit-lock-function 696) signaled (args-out-of-range "" 
0)

And once it was completley frozen with similar stacktrace via lldb, and I must
to restart it, and continue this draft.

My usual rutine is keep it runing, I'm using a few (like 5) virtual desktops
which and some of them has emacs frames.

Sometimes I attach and de-attach external display with much largest resolution,
and close the lid on laptop => everything moves to largest screen and resized.

--
wbr, Kirill