Re: [PATCH 5/5] nohz: Use IPI implicit full barrier against rq->nr_running r/w

2014-06-03 Thread Peter Zijlstra
On Tue, Jun 03, 2014 at 04:40:20PM +0200, Frederic Weisbecker wrote: > A full dynticks CPU is allowed to stop its tick when a single task runs. > Meanwhile when a new task gets enqueued, the CPU must be notified so that > it can restart its tick to maintain local fairness and other accounting >

[PATCH 5/5] nohz: Use IPI implicit full barrier against rq->nr_running r/w

2014-06-03 Thread Frederic Weisbecker
A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it can restart its tick to maintain local fairness and other accounting details. This notification is performed by way of an IPI. Then when the target

[PATCH 5/5] nohz: Use IPI implicit full barrier against rq-nr_running r/w

2014-06-03 Thread Frederic Weisbecker
A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it can restart its tick to maintain local fairness and other accounting details. This notification is performed by way of an IPI. Then when the target

Re: [PATCH 5/5] nohz: Use IPI implicit full barrier against rq-nr_running r/w

2014-06-03 Thread Peter Zijlstra
On Tue, Jun 03, 2014 at 04:40:20PM +0200, Frederic Weisbecker wrote: A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it can restart its tick to maintain local fairness and other accounting

[PATCH 5/5] nohz: Use IPI implicit full barrier against rq->nr_running r/w

2014-05-25 Thread Frederic Weisbecker
A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it can restart its tick to maintain local fairness and other accounting details. This notification is performed by way of an IPI. Then when the target

[PATCH 5/5] nohz: Use IPI implicit full barrier against rq-nr_running r/w

2014-05-25 Thread Frederic Weisbecker
A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it can restart its tick to maintain local fairness and other accounting details. This notification is performed by way of an IPI. Then when the target

[PATCH 5/5] nohz: Use IPI implicit full barrier against rq->nr_running r/w

2014-05-13 Thread Frederic Weisbecker
A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it restart its tick to maintain local fairness and other accounting details. This notification is performed by way of an IPI. Then when the target

[PATCH 5/5] nohz: Use IPI implicit full barrier against rq-nr_running r/w

2014-05-13 Thread Frederic Weisbecker
A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it restart its tick to maintain local fairness and other accounting details. This notification is performed by way of an IPI. Then when the target

[PATCH 5/5] nohz: Use IPI implicit full barrier against rq->nr_running r/w

2014-05-11 Thread Frederic Weisbecker
A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it restart its tick to maintain local fairness and other accounting details. This notification is performed by way of an IPI. Then when the target

[PATCH 5/5] nohz: Use IPI implicit full barrier against rq-nr_running r/w

2014-05-11 Thread Frederic Weisbecker
A full dynticks CPU is allowed to stop its tick when a single task runs. Meanwhile when a new task gets enqueued, the CPU must be notified so that it restart its tick to maintain local fairness and other accounting details. This notification is performed by way of an IPI. Then when the target