On Wed, Feb 15, 2012 at 5:11 PM, Peter Zijlstra <a.p.zijlstra@xxxxxxxxx> wrote: > > On Fri, 2012-02-10 at 22:24 +0200, Gilad Ben-Yossef wrote: > > I think the concept of giving the task some way to know if the tick is > > disabled or not is nice. > > Not sure the exact feature and surely not the interface are what we > > should adopt - maybe > > allow registering to receive a signal at the end of the tick when it > > is disabled an re-enabled? > > Fair enough, I indeed missed that property. And yes that makes sense. > > It might be a tad tricky to implement as things currently stand, because > AFAICR Frederic's stuff re-enables the tick on kernel entry (syscall) > things like signal delivery or a blocking wait for it might be 'fun'. > > But I'll have to defer to Frederic, its been too long since I've seen > his patches to remember most details. Yes, what I had in mind is that since Frederic's patch set always disables the tick from inside the (last) timer tick, we can have the tick return to user code from the timer with a signal whenever it is disabled or re-enabled. Basically, have the timer code make the signal pending from inside the timer, so that the return to user space on the special timer ticks (the last before disable or the first after re-enable) will be to a signal handler. I don't know if what I wrote above actually makes sense or not :-) I'll try to hack something up and see. Thanks, Gilad -- Gilad Ben-Yossef Chief Coffee Drinker gilad@xxxxxxxxxxxxx Israel Cell: +972-52-8260388 US Cell: +1-973-8260388 http://benyossef.com "If you take a class in large-scale robotics, can you end up in a situation where the homework eats your dog?" -- Jean-Baptiste Queru -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href