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

[Eli Zaretskii]

> From: Andreas Schwab <schwab@linux-m68k.org>
> Cc: dick.r.chiang@gmail.com,  36609@debbugs.gnu.org
> Date: Thu, 10 Jun 2021 17:35:54 +0200
> 
> On Jun 10 2021, Eli Zaretskii wrote:
> 
> > The idea is that a variable that is modified in a single machine
> > instruction will always have the value assigned by the last thread
> > which set that.
> 
> You cannot have that without explicit atomic operations.

We can do that if it turns out to be an issue.