Re: [bug-gawk] Behavior of fflush with SIGPIPE on stdout [PATCH]

[Eli Zaretskii]

> From: address@hidden
> Date: Mon, 27 Mar 2017 12:50:07 -0600
> Cc: address@hidden
> 
> I have set things up (I hope) so that gawk behaves like other awks and
> dies via SIGPIPE when that should happen. This then automatically
> has the exit status set to what it should be in POSIX systems.

What does this mean for systems that don't have SIGPIPE?  You will
likely see a long series of EINVAL from the write routines there.