Re: scratch/sigchld-fd 8f0ce42 1/2: Fix deadlock when receiving SIGCHLD during 'pselect'.

[Eli Zaretskii]

On January 20, 2021 10:37:31 AM GMT+02:00, Andreas Schwab 
<schwab@linux-m68k.org> wrote:
> On Jan 20 2021, Eli Zaretskii wrote:
> 
> > AFAIU, this is an issue for programs that install a SIGCHLD handler
> > immediately before calling pselect.  But Emacs has a SIGCHLD handler
> > installed at all times, so if the signal hits outside of the pselect
> > call, we should be able to detect that reliably.
> 
> No, the race between checking for the signal and entering pselect
> still
> exists.

Please tell more, especially what is meant by "checking for the signal".  We 
have a handler installed, so who is or should be checking for it?