Re: Another paralell make patch for libf2c

[Jeffrey A Law <law at cygnus dot com>]

  > > >Be realistic, since when was I appointed as an egcs change reviewer?
  > > 
  > > I don't get this -- what prevents you from reading the patches
  > > submitted to egcs-patches, and commenting on them, when you,
  > > an apparent "GNU make expert", see changes that don't take into
  > > account potential parallel-make problems?
  > 
  > I don't have the time to check out all the patches submitted
  > to egcs. I don't want to spend time on something which may
  > not be accepted in the first place. However, I will spend
  > time on it if I am asked to do so.
It would be appreciated if you could keep an eye on Makefile patches and how
they interact with parallel builds.

I think it would also be in your best interest to spend some time explaining
your patches.  Else they're likely to just get dropped on the floor.

  > > So, you think refusing to send patches to egcs to egcs-patches, instead
  > > sending them to others, forcing more people to keep more sets of
  > > random patches floating around, is better for everyone than simply
  > > taking the time to explain what your patches do, at least upon request,
  > > if not initially?
  > > 
  > 
  > Please go back to see if I was asked to explain my patch before
  > it was installed.
I've been asking you to do this for years.  It took me quite a while to figure
out what your patch was doing.  That's the absolutely wrong approach.  You
already know what the problem is, all you have to do is explain it to us and
how your patch fixes the problem.

Why do you find that so difficult to understand?  I've been asking you to do
this for 14 months.


  > My patch is very simple and basic. I thought it should not take
  > anyone who knows something about the GNU make, by that I mean
  > he/she knows how to modify the makefiles in glibc, more than 5
  > minutes to review it. I am sorry it took more than that for Jeff
  > to figure out what was going on. I just thought his time might
  > be better spent somewhere else and he would let people who know
  > review makefile changes. It will save everyone's time.
You can not, must not, assume that people have that kind of knowledge unless
you know that someone with that knowledge will review the patch.  We've been
through this before too.  Making such assumptions about the knowledge base of
the people reviewing patches, is wrong.  Plain and simple.

  > BTW, comparing with building glibc 2, libf2c is just a piece of
  > cake. Roland, Ulrich, Zack, Andreas and many other people have
  > done a very good job on makefiles. Thanks, folks.
That doesn't have anything to do with the core issue at hand -- you're ongoing
refusal to explain your problems & patches.

jeff