Looking at the perlxs page I don't see any problem with the use of NO_INIT although I did notice that a bunch of the cases seemed to have a function parameter 'position' which was not used anywhere within the code. It was also unclear where/how some of these routines like at_c are called....
--Chris On Wed, Aug 17, 2011 at 8:16 AM, Sisyphus <[email protected]> wrote: > > ----- Original Message ----- From: "chm" > >>> I think Core.xs.PL might be setting up an incorrect Core.xs. And it may >>> well be that it has been doing that for quite some time - but without >>> any undesirable consequences until recent changes to ExtUtils::ParseXS. >> >> According to git, the code in question has been there >> since 1998. > > That's quite amazing. After 13 years, code that has worked fine suddenly > starts to fail. > Yet (partly as the result of a link to the NO_INIT documentation, posted on > the perlmonks thread) I'm presently inclined to think that Core.xs.PL has > been misusing the NO_INIT keyword for all that time. > I repeat, however, that I'm not an XSpert. > > Usually, when I have some problem with XS, I'm able to come up with a fairly > simple Inline::C demo but this particular case doesn't lend itself well to > Inline::C, afaict. > >>> >>> For anyone interested, that thread can be found at: >>> http://www.perlmonks.org/index.pl?node_id=920621 >> >> Thanks for the ref. > > Still haven't got the authoritative response from there that I was hoping > for ... have to wait and see what turns up. > I think it's quite possible that someone on perlmonks will come up with the > answers ... if not I guess there's still the XS mailing list to try. And by > the time the perlmonks thread has run its course I should at least have a > good idea of how to effectively ask about this on that list. > > Cheers, > Rob > _______________________________________________ Perldl mailing list [email protected] http://mailman.jach.hawaii.edu/mailman/listinfo/perldl
