Re: [PATCH]: pselect: new module

2011-07-06 Thread Pádraig Brady
On 06/07/11 08:52, Paul Eggert wrote: No serial number here? I still don't get why hand-maintained serial numbers are useful in m4/*.m4 files (and nowhere else?!). Maybe they were useful long ago, but their cost outweighs their benefit now, as far as I can see. I got this answer when

Re: [PATCH]: pselect: new module

2011-07-04 Thread Bruno Haible
Paul, I wrote: (which is probably why you didn't notice the missing 'test $HAVE_PSELECT = 0'). This comment was not fair. There was no link error. I apologize. Bruno -- In memoriam Yonatan Netanyahu http://en.wikipedia.org/wiki/Yonatan_Netanyahu

Re: [PATCH]: pselect: new module

2011-07-03 Thread Bruno Haible
Hi Paul, +/* Get definition of 'sigset_t'. + But avoid namespace pollution on glibc systems. */ +#if !(defined __GLIBC__ !defined __UCLIBC__) +# include signal.h +#endif According to doc/posix-headers/signal.texi, it is not enough to include signal.h. You also need add a module

Re: [PATCH]: pselect: new module

2011-07-03 Thread Bruno Haible
Hi Paul, More comments: +++ b/m4/pselect.m4 @@ -0,0 +1,31 @@ +# pselect.m4 No serial number here? + if test $ac_cv_func_pselect = no || test $gl_cv_sig_pselect = no; then +REPLACE_PSELECT=1 + fi +#undef pselect + +int +rpl_pselect (int nfds, fd_set *restrict rfds, These are

[PATCH]: pselect: new module

2011-06-30 Thread Paul Eggert
I pushed this; it's for Emacs, but I expect other applications can use nanosecond-resolution 'select' as well. No doubt there will be porting glitches. I've tested it on Solaris 8 (!) and Fedora 14 for starters. pselect: new module * lib/sys_select.in.h: Include signal.h, for 'sigset_t'.