nanosleep() patch

Jason Tishler
Tue Jan 21 18:01:00 GMT 2003


Thanks for your feedback.

On Tue, Jan 21, 2003 at 05:17:06PM +0100, Corinna Vinschen wrote:
> On Tue, Jan 21, 2003 at 11:02:01AM -0500, Christopher Faylor wrote:
> > On Tue, Jan 21, 2003 at 04:58:42PM +0100, Corinna Vinschen wrote:
> > >I'm wondering if we could do without an extra function
> > >sleep_worker() and let nanosleep() be the basic implementation.  So
> > >sleep() as well as usleep() could call nanosleep().  Isn't that
> > >done that way in the Linux kernel, too?
> > 
> > In that case, nanosleep needs to be rewritten to deal with the same
> > issues as sleep().
> Sure.  nanosleep would be sleep_worker with timespec arguments.

OK, I will rework the patch as specified above.

Regarding usleep(), I was afraid to change it to use nanosleep() (aka
sleep_worker()) because its implementation was different than sleep().
Additionally, its current behavior does not seem to agree with what is
documented in "man 3 usleep" under Red Hat Linux 8.0.  Should I include
a reworked usleep() in the next version of this patch?


PGP/GPG Key: or key servers
Fingerprint: 7A73 1405 7F2B E669 C19D  8784 1AFD E4CC ECF4 8EF6

More information about the Cygwin-patches mailing list