Re: kernel/timer: avoid spurious ksoftirqd wakeups (v2)

2015-04-13 Thread Luiz Capitulino
On Sat, 11 Apr 2015 11:25:49 +0200 (CEST) Thomas Gleixner wrote: > On Fri, 10 Apr 2015, Luiz Capitulino wrote: > > On Fri, 10 Apr 2015 15:09:07 -0300 > > > It seems a valid solution for this interrupt is to program > > > sched_timer to the nearest future possible. > > > > What about calling th

Re: kernel/timer: avoid spurious ksoftirqd wakeups (v2)

2015-04-11 Thread Thomas Gleixner
On Fri, 10 Apr 2015, Luiz Capitulino wrote: > On Fri, 10 Apr 2015 15:09:07 -0300 > > It seems a valid solution for this interrupt is to program > > sched_timer to the nearest future possible. > > What about calling the timer function right there, like > hrtimer_interrupt() does? No, you cannot

Re: kernel/timer: avoid spurious ksoftirqd wakeups (v2)

2015-04-10 Thread Luiz Capitulino
On Fri, 10 Apr 2015 15:09:07 -0300 Marcelo Tosatti wrote: > On Wed, Apr 08, 2015 at 12:12:45AM +0200, Frederic Weisbecker wrote: > > On Tue, Apr 07, 2015 at 11:10:49PM +0200, Thomas Gleixner wrote: > > > On Mon, 6 Apr 2015, Marcelo Tosatti wrote: > > > > It is only necessary to raise timer softir

Re: kernel/timer: avoid spurious ksoftirqd wakeups (v2)

2015-04-07 Thread Frederic Weisbecker
On Tue, Apr 07, 2015 at 11:10:49PM +0200, Thomas Gleixner wrote: > On Mon, 6 Apr 2015, Marcelo Tosatti wrote: > > It is only necessary to raise timer softirq > > in case there are active timers. > > Depends. See below. > > > Limit the ksoftirqd wakeup to that case. > > > > Fixes a latency spike

Re: kernel/timer: avoid spurious ksoftirqd wakeups (v2)

2015-04-07 Thread Thomas Gleixner
On Mon, 6 Apr 2015, Marcelo Tosatti wrote: > It is only necessary to raise timer softirq > in case there are active timers. Depends. See below. > Limit the ksoftirqd wakeup to that case. > > Fixes a latency spike with isolated CPUs and > nohz full mode. This lacks a proper explanation of the o

Re: kernel/timer: avoid spurious ksoftirqd wakeups (v2)

2015-04-06 Thread Rik van Riel
On 04/06/2015 07:15 PM, Marcelo Tosatti wrote: > > > It is only necessary to raise timer softirq > in case there are active timers. > > Limit the ksoftirqd wakeup to that case. > > Fixes a latency spike with isolated CPUs and > nohz full mode. > > v2: fix variable initialization > do not r