Re: [PATCH v14 04/14] task_isolation: add initial support

2016-09-09 Thread Chris Metcalf
On 9/3/2016 11:31 AM, Frederic Weisbecker wrote: On Tue, Aug 30, 2016 at 02:17:27PM -0400, Chris Metcalf wrote: On 8/30/2016 1:36 PM, Chris Metcalf wrote: See the other thread with Peter Z for the longer discussion of this. At this point I'm leaning towards replacing the set_tsk_need_resched()

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-09-03 Thread Frederic Weisbecker
On Tue, Aug 30, 2016 at 02:17:27PM -0400, Chris Metcalf wrote: > On 8/30/2016 1:36 PM, Chris Metcalf wrote: > >>>See the other thread with Peter Z for the longer discussion of this. > >>>At this point I'm leaning towards replacing the set_tsk_need_resched() with > >>> > >>> set_current_state(TA

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-08-30 Thread Chris Metcalf
On 8/30/2016 1:10 PM, Frederic Weisbecker wrote: On Tue, Aug 30, 2016 at 11:41:36AM -0400, Chris Metcalf wrote: On 8/29/2016 8:55 PM, Frederic Weisbecker wrote: On Mon, Aug 15, 2016 at 10:59:55AM -0400, Chris Metcalf wrote: On 8/11/2016 2:50 PM, Christoph Lameter wrote: On Thu, 11 Aug 2016, F

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-08-30 Thread Chris Metcalf
On 8/30/2016 1:36 PM, Chris Metcalf wrote: See the other thread with Peter Z for the longer discussion of this. At this point I'm leaning towards replacing the set_tsk_need_resched() with set_current_state(TASK_INTERRUPTIBLE); schedule(); __set_current_state(TASK_RUNNING); I don'

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-08-30 Thread Chris Metcalf
On 8/29/2016 8:55 PM, Frederic Weisbecker wrote: On Mon, Aug 15, 2016 at 10:59:55AM -0400, Chris Metcalf wrote: On 8/11/2016 2:50 PM, Christoph Lameter wrote: On Thu, 11 Aug 2016, Frederic Weisbecker wrote: Do we need to quiesce vmstat everytime before entering userspace? I thought that vmsta

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-08-30 Thread Frederic Weisbecker
On Tue, Aug 30, 2016 at 11:41:36AM -0400, Chris Metcalf wrote: > On 8/29/2016 8:55 PM, Frederic Weisbecker wrote: > >On Mon, Aug 15, 2016 at 10:59:55AM -0400, Chris Metcalf wrote: > >>On 8/11/2016 2:50 PM, Christoph Lameter wrote: > >>>On Thu, 11 Aug 2016, Frederic Weisbecker wrote: > >>> > Do

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-08-29 Thread Frederic Weisbecker
On Mon, Aug 15, 2016 at 10:59:55AM -0400, Chris Metcalf wrote: > On 8/11/2016 2:50 PM, Christoph Lameter wrote: > >On Thu, 11 Aug 2016, Frederic Weisbecker wrote: > > > >>Do we need to quiesce vmstat everytime before entering userspace? > >>I thought that vmstat only need to be offlined once and fo

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-08-15 Thread Chris Metcalf
On 8/11/2016 2:50 PM, Christoph Lameter wrote: On Thu, 11 Aug 2016, Frederic Weisbecker wrote: Do we need to quiesce vmstat everytime before entering userspace? I thought that vmstat only need to be offlined once and for all? Once is sufficient after disabling the tick. It's true that task_i

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-08-11 Thread Christoph Lameter
On Thu, 11 Aug 2016, Frederic Weisbecker wrote: > Do we need to quiesce vmstat everytime before entering userspace? > I thought that vmstat only need to be offlined once and for all? Once is sufficient after disabling the tick. -- To unsubscribe from this list: send the line "unsubscribe linux-d

Re: [PATCH v14 04/14] task_isolation: add initial support

2016-08-11 Thread Frederic Weisbecker
On Tue, Aug 09, 2016 at 04:29:46PM -0400, Chris Metcalf wrote: > +/* > + * Each time we try to prepare for return to userspace in a process > + * with task isolation enabled, we run this code to quiesce whatever > + * subsystems we can readily quiesce to avoid later interrupts. > + */ > +void task_

[PATCH v14 04/14] task_isolation: add initial support

2016-08-09 Thread Chris Metcalf
The existing nohz_full mode is designed as a "soft" isolation mode that makes tradeoffs to minimize userspace interruptions while still attempting to avoid overheads in the kernel entry/exit path, to provide 100% kernel semantics, etc. However, some applications require a "hard" commitment from th