bug#52141: 29.0.50; Crash during redisplay

[Eli Zaretskii <eliz@gnu.org>]

> From: Po Lu <luangruo@yahoo.com>
> Cc: 52141@debbugs.gnu.org
> Date: Sat, 27 Nov 2021 19:32:29 +0800
> 
> >> #9  set_next_vectorPython Exception <class 'gdb.MemoryError'>: Cannot access memory at address 0xfe94cc8
> >>  (p=#10 setup_on_free_list (nbytes=<optimized out>, v=0x2ce6c10) at alloc.c:2977
> >> #11 sweep_vectors () at alloc.c:3242
> >> #12 0x0000000000561328 in gc_sweep () at alloc.c:7255
> >> #13 garbage_collect () at alloc.c:6190
> >> #14 0x0000000000561811 in maybe_garbage_collect () at alloc.c:6053
> >> #15 0x00000000005804ad in maybe_gc () at /home/oldosfan/emacs-dev/emacs-gc/src/lisp.h:5161
> >> #16 eval_sub (form=0x1e753d3) at eval.c:2449
> >> #17 0x0000000000582149 in Feval (form=0x1e753d3, lexical=<optimized out>) at eval.c:2372
> >                                    ^^^^
> > What is this form we are evaluating here?  Can you show its contents?
> > (Not that I think it is related to the crash.)
> 
> Its type is Lisp_Cons.  However, its address is invalid.  Running
> `xcons' in gdb complains:
> 
> (gdb) xcons
> $9 = (struct Lisp_Cons *) 0x1e753d0
> Cannot access memory at address 0x1e753d0

Hmm... and the address is very different from the one which caused the
segfault.  Strange.

Any ideas, anyone?