bug#70790: 30.0.50; xfaces crash

[Eli Zaretskii]

> Date: Sun, 05 May 2024 23:10:19 +0530 (IST)
> From: Madhu <enometh@meer.net>
> 
> Hello, FTR I'm attaching the decoded addr2line crash backtrace, no gdb
> backtrace yet.  In this case the crash happened when I evaluated the
> following form:
> 
> ```
> (set-fontset-font "fontset-default" (cons #x0591 #xFB4F) "Taamey D")
> ```
> 
> I think I've seen this a few times but I don't have a reproducible
> recipe yet and way too many fonts on my system.  I'll follow up on any
> suggestions which might produce a better test case.

Does this happen only with that font?

Is it reproducible (i.e., happens each time you evaluate that form)?

> deliver_thread_signal at 
> /14/build/emacs/build-xt-native/src/../../src/sysdep.c:1782
> stack_overflow at /14/build/emacs/build-xt-native/src/../../src/sysdep.c:1892
>  (inlined by) handle_sigsegv at 
> /14/build/emacs/build-xt-native/src/../../src/sysdep.c:1937
> ?? ??:0

The above confuses me: it says "stack overflow", so does it mean there
was some kind of infinite recursion?

Please try triggering this in Emacs run from GDB, and show the full
backtrace when it happens.