Re: Perl 6 Summary for 2004-10-23 through 2004-11-01

[Brent 'Dax' Royal-Gordon]

Matt Fowles <uber...@gmail.com> wrote:
> Parrot on AMD64
> Brent 'Dax' Royal-Gordon managed to find an AMD64 for himself. Not only
> am I jealous, but Parrot passes all expected tests on it when one adds
> --ccflags=':add{ -fPIC }'.

I then patched Parrot so that Configure would automagically add that
flag on AMD64--although that message doesn't seem to have reached the
list, likely because of my own idiocy. Ah well.

--
Brent 'Dax' Royal-Gordon <br...@brentdax.com>
Perl and Parrot hacker

There is no cabal.

[Leopold Toetsch]

Matt Fowles <uber...@gmail.com> wrote:
> Perl 6 Summary for 2004-10-23 through 2004-11-01

> newsub opcodes
> Leo wanted to add some newsub opcodes to save allocations. Dan countered
> that it would be a better thing to get constant pmcs working properly.

Not quite. The C<newsub> opcodes were already there and got used in PIR
code for function calls:

x = foo(y) # $P0 = newsub .Sub, foo ; ... ; $P0()

> So Leo did.

Yep. Now the PMC constant of the Sub object "foo" is inserted at compile
time, saving the creation of a new PMC for every call.

Thanks for the detailed summary,
leo

[Stéphane Payrard]

Rereading this excellent summary, I note that I did not keep
people posted on my problem.

> Stephane Payrard asked about some problems he had encountered creating
> the magical all in one Siva PMC. Leo provided a helpful prod towards
> LVALUE_CAST. The other problems may or may not related to the mandrake
> cooker. Hopefully someone will tell us.
>

So I do.

Mandrake cooker was not in cause. My sloppiness was. As a general
rule, when the parrot compilation fails, one should do a 'make realclean'
to restart with a clean slate.
I did not and pick remnants of a obolete configuration.

--
stef