On Fri, May 04, 2001 at 05:01:25PM -0400, Benjamin Sugars wrote:
> On Fri, 4 May 2001, Jarkko Hietaniemi wrote:
>
> > Let us agree to disagree and move on.
>
> I'm more than happy to move on, but I do have to ask if this whole
> wonderful experience has had any influence on your plans for 5.8?
I'm not too enamored by the idea of adding Configure probes for few
dozen system/library calls, all slightly different, but I guess we
(I and/or Andy) can be convinced.
The "Fun" part is then to add the copying of the thread-specific data
back and forth, given all the _r variants. I suggest building some
sort of intermediate abstractions, say, Perl_getpwuid_r(), that will
hide the ugliness from pp_getpwblah(). Mark my words: "I suggest".
As in: "Somebody *else* will do". (Since I have no experience with
the ithreads guts, and no burning desire to become an expert on that.)
> Cheers,
> -Ben
>
> --
> signer: can't create ~/.sig: No such device
--
$jhi++; # http://www.iki.fi/jhi/
# There is this special biologist word we use for 'stable'.
# It is 'dead'. -- Jack Cohen
