Re: mktime() problem when building print/a2ps

[Robert Elz <kre%munnari.OZ.AU@localhost>]

    Date:        Fri, 14 Oct 2022 11:22:45 +0200
    From:        Martin Husemann <martin%duskware.de@localhost>
    Message-ID:  <20221014092244.GC12480%mail.duskware.de@localhost>

  | For reasons beyound me gdb thinks the program has two threads running,
  | and both idle looping - not sure which thread gets the sigalarm.

I don't know enough gdb magic to be able to tell that, but at least
as far as ps is concerned, I seem to have just a single thread.

However, given that there's no SIGALRM handler, nothing should be
catching it, or processing it at all, the signal should simply kill
the process (however many threads it managed to create).

Since the SIGALRM works for me, it does terminate the process, perhaps
someone for whom that isn't true can work out why, if that seems important,
since I cannot.

But IMO this isn't worth chasing any further, the configure script is
obviously broken (and with a modern gcc would be broken everywhere,
there's nothing NetBSD specific about it that I can see).   Someone
should figure out how to file an upstream bug report, and in the
meantime, patch the configure script when used by pkgsrc not to use -O
(at least not for that test).

kre