Re: [PATCH 0/8] nohz: Fix nohz kick irq work on tick v2

2014-09-12 Thread Peter Zijlstra
On Fri, Sep 12, 2014 at 08:05:49PM +0200, Frederic Weisbecker wrote: > > Acked-by: Peter Zijlstra (Intel) > > Intel? :-) Wow. Yeah, moved about a month ago :-) > Note, I fear I'll have to turn the static_cpu_has into cpu_has. The > alternatives code which initialize all static_cpu_has() things

Re: [PATCH 0/8] nohz: Fix nohz kick irq work on tick v2

2014-09-12 Thread Frederic Weisbecker
On Thu, Sep 11, 2014 at 10:28:12AM +0200, Peter Zijlstra wrote: > On Wed, Sep 10, 2014 at 11:33:20PM +0200, Frederic Weisbecker wrote: > > > Frederic Weisbecker (7): > > nohz: Move nohz full init call to tick init > > irq_work: Force raised irq work to run on irq work interrupt > >

Re: [PATCH 0/8] nohz: Fix nohz kick irq work on tick v2

2014-09-12 Thread Frederic Weisbecker
On Thu, Sep 11, 2014 at 10:28:12AM +0200, Peter Zijlstra wrote: On Wed, Sep 10, 2014 at 11:33:20PM +0200, Frederic Weisbecker wrote: Frederic Weisbecker (7): nohz: Move nohz full init call to tick init irq_work: Force raised irq work to run on irq work interrupt x86:

Re: [PATCH 0/8] nohz: Fix nohz kick irq work on tick v2

2014-09-12 Thread Peter Zijlstra
On Fri, Sep 12, 2014 at 08:05:49PM +0200, Frederic Weisbecker wrote: Acked-by: Peter Zijlstra (Intel) pet...@infradead.org Intel? :-) Wow. Yeah, moved about a month ago :-) Note, I fear I'll have to turn the static_cpu_has into cpu_has. The alternatives code which initialize all

Re: [PATCH 0/8] nohz: Fix nohz kick irq work on tick v2

2014-09-11 Thread Peter Zijlstra
On Wed, Sep 10, 2014 at 11:33:20PM +0200, Frederic Weisbecker wrote: > Frederic Weisbecker (7): > nohz: Move nohz full init call to tick init > irq_work: Force raised irq work to run on irq work interrupt > x86: Tell irq work about self IPI support > arm: Tell irq work

Re: [PATCH 0/8] nohz: Fix nohz kick irq work on tick v2

2014-09-11 Thread Peter Zijlstra
On Wed, Sep 10, 2014 at 11:33:20PM +0200, Frederic Weisbecker wrote: Frederic Weisbecker (7): nohz: Move nohz full init call to tick init irq_work: Force raised irq work to run on irq work interrupt x86: Tell irq work about self IPI support arm: Tell irq work about

[PATCH 0/8] nohz: Fix nohz kick irq work on tick v2

2014-09-10 Thread Frederic Weisbecker
Hi, The nohz kick uses irq work to restart the tick but it relies on irq work to run on its own IPI. Otherwise the kick runs on the tick and bad things happen when it tries to cancel the tick hrtimer from itself. That 2nd version addresses Peterz review. Note I only tested x86. Just waiting for

[PATCH 0/8] nohz: Fix nohz kick irq work on tick v2

2014-09-10 Thread Frederic Weisbecker
Hi, The nohz kick uses irq work to restart the tick but it relies on irq work to run on its own IPI. Otherwise the kick runs on the tick and bad things happen when it tries to cancel the tick hrtimer from itself. That 2nd version addresses Peterz review. Note I only tested x86. Just waiting for