A pair of new timeout functions

A pair of new timeout functions

Posted Sep 2, 2005 14:55 UTC (Fri) by forthy (guest, #1525)
Parent article: A pair of new timeout functions

Hm, again a relative delay. This is one of the primary design mistakes of
Unix: relative delays. You will always get possible race conditions with
relative delays if you want precission, but you can always easily convert
an absolute delay to a relative one (current_time+delay).

---

to post comments

A pair of new timeout functions

Posted Sep 3, 2005 11:21 UTC (Sat) by renox (guest, #23785) [Link]

Both are useful IMHO, it's true that in some usage relative delay doesn't work, but most of the time they work very well, so having to use additionnal code to compute current_time+delay nearly everywhere is realy pointless.