Alan Cox (alan@xxxxxxxxxx): > > I guess my annoyance is more that only one user can use /dev/rtc at > > once, and there are now alot of applications which use it. The > > Most of those apps should be ok in 2.5 with normal timers - the select > resolution is 1Khz in 2.5.x So, just to make _sure_ I've got this straight, does this mean that usleep() will suddenly be 1ms-accurate, or do I need to do something special? Also, is there some easy way to detect this accuracy? The reason I'm being picky is that currently if I usleep() at all, my frames are late, since I'm effectively always rounding up to the next 10ms time spot. If I don't usleep, I'm too soon, but at least I don't drop frames. I want to make sure that if I do usleep(), then I know I'm actually going to get better accuracy. -- Billy Biggs vektor@xxxxxxxxxxxx
