bug#71744: 29.4; SIGSEGV during completion-at-point in lsp-mode with corfu and cape

[Eli Zaretskii <eliz@gnu.org>]

> Date: Wed, 14 Aug 2024 13:22:18 +0000
> From:  Sigve Indregard via "Bug reports for GNU Emacs,
>  the Swiss army knife of text editors" <bug-gnu-emacs@gnu.org>
> 
> I experience the exact same bug, in that the backtrace looks identical.
> 
> Similar to the original reporter, it occurs while completing symbols for Rust, and my setup also involves Corfu. It does not happen on every complete, but usually within 10 minutes of development.
> 
> But some differences are probably important: my setup is with eglot, not lsp-mode, and I do not use Cape.
> 
> Maybe it is relevant that the trace ending with the failing Fget starts with jsonrpc. Is there a way to monitor the exact communication emacs has with the language server?
> 
> * Backtrace (snippet):
> 
> Thread 1 (Thread 0x7ffff2837080 (LWP 150113) "emacs"):
> #0  Fget (symbol=0x5555564e3dc0, propname=0x66f0) at /usr/src/debug/emacs/emacs-29.4-wayland/src/lisp.h:1152
>         propval = <optimized out>
>         propval = <optimized out>
> #1  Fget (symbol=0x5555564e3dc0, propname=0x66f0) at /usr/src/debug/emacs/emacs-29.4-wayland/src/fns.c:2516
>         propval = <optimized out>
> #2  0x0000555555681201 in parse_modifiers (symbol=0x5555564e3dc0) at /usr/src/debug/emacs/emacs-29.4-wayland/src/keyboard.c:6897
>         elements = <optimized out>
>         elements = <optimized out>
>         end = <optimized out>
>         modifiers = <optimized out>
>         unmodified = <optimized out>
>         mask = <optimized out>
> #3  parse_modifiers (symbol=0x5555564e3dc0) at /usr/src/debug/emacs/emacs-29.4-wayland/src/keyboard.c:6888
                       ^^^^^^^^^^^^^^^^^^^^^
Please show what is this symbol.  Something like this:

  (gdb) source /path/to/emacs/src/.gdbinit
  (gdb) p symbol
  (gdb) xsymbol

Also, what is the value of globals.f_Voverriding_plist_environment?