bug#36609: 27.0.50; Possible race-condition in threading implementation

[Eli Zaretskii]

> From: dick.r.chiang@gmail.com
> Cc: larsi@gnus.org,  politza@hochschule-trier.de,  pipcet@gmail.com,
>   36609@debbugs.gnu.org
> Date: Wed, 09 Jun 2021 17:40:28 -0400
> 
> There is no guarantee the static variable `threads_holding_glib_lock`
> introduced in xgselect.c in commit 9c62ffb is synchronized across threads.

How can that happen?  It's a single word, and its change should
therefore be atomic.