Re: [PATCH v2 tip/core/timers] Crude timer-wheel latency hacks

2014-01-15 Thread Paul E. McKenney
On Wed, Jan 15, 2014 at 11:22:41PM +0100, Thomas Gleixner wrote: > Paul, > > On Tue, 14 Jan 2014, Paul E. McKenney wrote: > > The following three patches provide some crude timer-wheel latency > > patches. I understand that a more comprehensive solution is in progress, > > but in the meantime,

Re: [PATCH v2 tip/core/timers] Crude timer-wheel latency hacks

2014-01-15 Thread Thomas Gleixner
Paul, On Tue, 14 Jan 2014, Paul E. McKenney wrote: > The following three patches provide some crude timer-wheel latency > patches. I understand that a more comprehensive solution is in progress, > but in the meantime, these patches work well in cases where a given > CPU has either zero or one

Re: [PATCH v2 tip/core/timers] Crude timer-wheel latency hacks

2014-01-15 Thread Thomas Gleixner
Paul, On Tue, 14 Jan 2014, Paul E. McKenney wrote: The following three patches provide some crude timer-wheel latency patches. I understand that a more comprehensive solution is in progress, but in the meantime, these patches work well in cases where a given CPU has either zero or one timers

Re: [PATCH v2 tip/core/timers] Crude timer-wheel latency hacks

2014-01-15 Thread Paul E. McKenney
On Wed, Jan 15, 2014 at 11:22:41PM +0100, Thomas Gleixner wrote: Paul, On Tue, 14 Jan 2014, Paul E. McKenney wrote: The following three patches provide some crude timer-wheel latency patches. I understand that a more comprehensive solution is in progress, but in the meantime, these

[PATCH v2 tip/core/timers] Crude timer-wheel latency hacks

2014-01-14 Thread Paul E. McKenney
Hello! The following three patches provide some crude timer-wheel latency patches. I understand that a more comprehensive solution is in progress, but in the meantime, these patches work well in cases where a given CPU has either zero or one timers pending, which is a common case for NO_HZ_FULL

[PATCH v2 tip/core/timers] Crude timer-wheel latency hacks

2014-01-14 Thread Paul E. McKenney
Hello! The following three patches provide some crude timer-wheel latency patches. I understand that a more comprehensive solution is in progress, but in the meantime, these patches work well in cases where a given CPU has either zero or one timers pending, which is a common case for NO_HZ_FULL