Re: Fwd: [bug #33138] .PARLLELSYNC enhancement with patch

[Eli Zaretskii]

> Date: Fri, 26 Apr 2013 09:05:18 +0100
> From: Tim Murphy <address@hidden>
> Cc: "Paul D. Smith" <address@hidden>, "address@hidden" <address@hidden>
> 
> > How much would you use for the timeout, though?  A sub-Make could
> > legitimately run for a very long time, depending on what's in the
> > Makefile.
> >
> > FWIW, I currently let Make wait forever for the mutex.
> >
> 
> It's a horrible question and no answer makes people happy.

Yes, I know.  Which is why I think we would be better off not asking
it in the first place.

> So timeouts have to be configurable by users who can set whatever they
> think is the least bad compromise in their particular case.

If humans need to be in the loop, they can impose timeouts even if
Make does not offer any such customizations.  E.g., run the build
under the 'timeout' command (from Coreutils), or just plain Ctrl-C the
instance of Make that hogs the mutex.