Re: AIX status

[Ulrich Drepper <drepper at redhat dot com>]

David Edelsohn <dje@watson.ibm.com> writes:

> 	Please note that I tried discussing AIX pthread support back in
> the beginning of May and sent a patch, but received no response.  I then
> inquired again during June, but received no response. 

And your problem is?  Just as I said, it's nobody obligation to talk
to you about this.  You have the interest that means you have to
provide the code.

> 	AIX is fully supported in libstdc++-v2; v3 should not be a step
> backwards.

That's absolutely no reason.  The requirements to support v3 are much
higher.  It's a rewrite, not a little change.

> In that sense, it is not a completely new, non-existent port at all,
> so that is misleading.

Crap.  Everything is written from scratch excluding libio.  I.e.,
everything *is* completely new on every new architecture.

> 	From the beginning, all that I have been asking for is a flexible
> framework for portability and a discussion of a design for acceptable
> patches given the AIX requirements.  Without that, the port developer is
> being asked to waste his or her time stumbling around in the dark creating
> patches which are sure to be rejected.

Well, yeah, to some extend this is true.  But it is what others did
and they didn't complain, only you did.

Also, you cannot criticize the framework at all.  It is very flexible.
And adding a solution which does not require changes in the rest of
the system would definitely not be rejected.


> The maintainer needs to take the lead on the overall design
> framework given his or her broader view and discuss an approach with
> the developer.

No and I don't understand why you don't get it.  You're no newbie, you
should know it meanwhile.  The maintainer has no obligations.  Get
this in your head.  Everything which involves work has to be of some
interest on both sides.

The problem is that you want to get out of this with an investment of
time as small as possible.


> One cannot do this work alone without the guidance of the
> maintainers.  One should not have to pay the maintainers for them to
> be engaged in the discussion.

"One should not": nice words.  When have you stepped up and wasted
your time on something you have not the slightest interest in?  I
certainly am not doing this since my time is too precious.

I'm only continuing this thread because I don't want you to be seen as
the poor victim of those vicious people who don't want to work with
you on *your* problems.

> 	There are AIX patches for libtool actively being developed in the
> community, so that complaint is a red herring as well.

It's not.  libtool is not ready.

> 	If libstdc++-v3 is intended to be a portable C++ library for GCC,
> then it needs to be designed that way from the beginning.

This only shows how little you understand of the design.  Nathan and I
spend quite some time designing it.  The model is similar to glibc's
and glibc definitely allows everything you want.  If you don't see it
you simply don't understand the concept.

By continuing to claim that the design is bad and does not allow full
replacement of v2 on all previously supported platforms you simply
disqualify your comments.  Fact is that with a complete rewrite (and
v3 is just that) every platform has to start from scratch and
complaining about leaving behind some of the platforms is completely
unjustified.  If nobody has interest for the platform, it's dead.  If
the platform is still used but nobody wants to spend time porting v3,
there is still the possibility to use v2.  One cannot stop progress
just because some environments which could use the old version are not
supported (in the moment).

> 	Placing all of the blame on me and AIX and Mark and the GCC
> Steering Committee is incorrect.

I don't blame you for having interests.  We all have.  But I cannot
stand the way you try to push people.  The comments in your last mail
should make obvious to everybody how you want people to behave,
namely, to serve your goals.  You cannot do this, the gcc SC cannot do
this.  Contributors here are volunteers and you must not demand
*anything*.  People are working on a) what they need and possibly b)
what is fun.  Support for AIX obviously does not fall in any of these
categories (otherwise the port would have happened without your
intervention).  You want to put the blame for this somewhere and this
is completely unacceptable if you put it anywhere but on yourself,
other AIX users, and to some extend AIX itself.

I criticized Mark only for bowing too easily to the pressure put on
him to add AIX to the first tier platforms.  A thorough investigation
would have shown him how hard this is and how little willingness to
solve the problem there is.  Next time around he will be better
prepared, this is his first release and therefore I expect mistakes
like this.

-- 
---------------.                          ,-.   1325 Chesapeake Terrace
Ulrich Drepper  \    ,-------------------'   \  Sunnyvale, CA 94089 USA
Red Hat          `--' drepper at redhat.com   `------------------------