> AFAIU, this is nowhere near the C stack limit.  The Lisp nesting is
> near its limit, but that's a far cry from exhausting the C stack,
> AFAICT.

Thanks; I'm still learning the internals so I might confuse some terms.

With #71243 fixed, I could reproduce this one!
And there's indeed a breach of the Lisp stack limit.

The variables you requested:


process server <1328> no longer connected to pipe; closed it
process server <1329> no longer connected to pipe; closed it
process server <1330> no longer connected to pipe; closed it
Lisp nesting exceeds ‘max-lisp-eval-depth’: 1601
Lisp nesting exceeds ‘max-lisp-eval-depth’: 1601
Lisp nesting exceeds ‘max-lisp-eval-depth’: 1601
Lisp nesting exceeds ‘max-lisp-eval-depth’: 1601
Lisp nesting exceeds ‘max-lisp-eval-depth’: 1601
Lisp nesting exceeds ‘max-lisp-eval-depth’
Lisp nesting exceeds ‘max-lisp-eval-depth’
Lisp nesting exceeds ‘max-lisp-eval-depth’: 1601
error in process sentinel: Lisp nesting exceeds `max-lisp-eval-depth': 1601
error in process filter: Lisp nesting exceeds `max-lisp-eval-depth': 1601
Lisp nesting exceeds ‘max-lisp-eval-depth’
process server <1349> no longer connected to pipe; closed it
process server <1350> no longer connected to pipe; closed it
process server <1359> no longer connected to pipe; closed it
process server <1360> no longer connected to pipe; closed it
process server <1369> no longer connected to pipe; closed it
process server <1370> no longer connected to pipe; closed it
process server <1378> no longer connected to pipe; closed it
process server <1379> no longer connected to pipe; closed it
process server <1380> no longer connected to pipe; closed it

Program received signal SIGSEGV, Segmentation fault.
start_display (it=0x7fffffb96bd0, w=0x5555621f25a0, pos=...) at xdisp.c:3734
3734  row = w->desired_matrix->rows + first_vpos;
(gdb) p w->desired_matrix
$1 = (struct glyph_matrix *) 0x0
(gdb) p w->desired_matrix->nrows
Cannot access memory at address 0x18
(gdb) p window_wants_tab_line (w)
$2 = false
(gdb) p window_wants_header_line (w)
$3 = false
(gdb)

And others (but I don't know what else to check):

(gdb) p w->current_matrix
$18 = (struct glyph_matrix *) 0x0
(gdb) p w->redisplay
$19 = true

(gdb) p w->window_parameters
$12 = XIL(0)
(gdb) p w->must_be_updated_p
$13 = true

(gdb) xframe w->frame
$8 = (struct frame *) 0x5555621f2380
"F861"
(gdb) p w->header
$9 = {
  size = 4611686018612097049
}
(gdb) p w->parent
$10 = XIL(0)
(gdb) p w->normal_lines
$11 = XIL(0x7fffd663a24f)


On Tue, 28 May 2024 at 15:55, Eli Zaretskii <eliz@gnu.org> wrote:
> From: Daniel Clemente <n142857@gmail.com>
> Date: Tue, 28 May 2024 15:36:05 +0000
> Cc: 71224@debbugs.gnu.org
>
> I forgot whether in this case (#71224) the stack was also near the stack limit. Maybe when it's so close to a
> stack overflow, these SIGSEGV can happen first.

AFAIU, this is nowhere near the C stack limit.  The Lisp nesting is
near its limit, but that's a far cry from exhausting the C stack,
AFAICT.