Re: Timestamp race avoidance in do_update()

[Derek R. Price]

Larry Jones wrote:

> Laine Stump writes [quoting me]:
> > >
> > > Unfortunately, usleep() isn't portable.
> >
> > Okay, then how about nanosleep()?
>
> nanosleep() is even less portable.  (It's part of the POSIX real-time
> extensions.  I think usleep() is part of the base POSIX standard;
> sleep() definitely is and it's been in Unix "forever" so it's much more
> widely implemented.)

Well, it didn't seem very hard to do, so I had configure search for
nanosleep, usleep, and select in that order, then use them in the same
order of preference, falling back on sleep as a last resort.  It speeds up
'make check' noticeably.

Checked in.

Derek

--
Derek Price                      CVS Solutions Architect ( http://CVSHome.org )
mailto:address@hidden     OpenAvenue ( http://OpenAvenue.com )
--
I will not squeak chalk.
I will not squeak chalk.
I will not squeak chalk...

          - Bart Simpson on chalkboard, _The Simpsons_