Right, let's give a fresh look at the todo list. We need:
a) a set of working libs
b) hooks (sorry if the term is not appropriate) in the compiler
In principle, I have no problem with my present knowledge in tackling A,
but B would be another thing... I could do something there as well, but I
need to be guided as a baby who hasn't learnt how to walk :-) I see many
of the potential contributors there to be up to something else (mainly,
tree-ssa), so probably noone will put big commitment on this stage before
that project has reached production-stage.
Will it be sensible to do some counting? Who is still there and still
wants to give a hand, even if not in a near future?
Biagio
On Thu, 15 May 2003, Lars Segerlund wrote:
As far as I am concerned, there is tons of stuff to do, and nobody is
doing zit !
If you make a sugestion of a way to progress, it's all well do that
later and so on.
We have everything to start off with the libs, ( if anybody have
looked at makeing a build from the stubs I submitted ).
In the style of exa ( an example language for gcc & friens ), we could
emit code for some 'static' cases ( namely the example from the spec )
and verify correct behaviour in the backend.
This has NO parsing, NO fiddling with trees or anything such but would
enable us to do work on a set of working libs.
I really got ticked of when nobody was looking at the cvs or the code
in it :-( .... but if there is interest I am all ready for some coding :-) .
_______________________________________________
Gomp-discuss mailing list
address@hidden
http://mail.nongnu.org/mailman/listinfo/gomp-discuss