Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-19 Thread Steven Rostedt
On Thu, 2013-04-18 at 18:34 +0200, Vincent Guittot wrote: > The current update of the rq's load can be erroneous when RT tasks are > involved > > The update of the load of a rq that becomes idle, is done only if the avg_idle > is less than sysctl_sched_migration_cost. If RT tasks and short idle du

Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-19 Thread Peter Zijlstra
On Thu, 2013-04-18 at 18:34 +0200, Vincent Guittot wrote: > The current update of the rq's load can be erroneous when RT tasks are > involved > > The update of the load of a rq that becomes idle, is done only if the avg_idle > is less than sysctl_sched_migration_cost. If RT tasks and short idle du

Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-19 Thread Vincent Guittot
On 19 April 2013 11:21, Mike Galbraith wrote: > On Fri, 2013-04-19 at 10:50 +0200, Vincent Guittot wrote: >> On 19 April 2013 10:14, Mike Galbraith wrote: >> > On Fri, 2013-04-19 at 09:49 +0200, Vincent Guittot wrote: >> >> On 19 April 2013 06:30, Mike Galbraith wrote: >> >> > On Thu, 2013-04-18

Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-19 Thread Mike Galbraith
On Fri, 2013-04-19 at 11:21 +0200, Mike Galbraith wrote: > On Fri, 2013-04-19 at 10:50 +0200, Vincent Guittot wrote: > > On 19 April 2013 10:14, Mike Galbraith wrote: > > > On Fri, 2013-04-19 at 09:49 +0200, Vincent Guittot wrote: > > >> On 19 April 2013 06:30, Mike Galbraith wrote: > > >> > On

Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-19 Thread Mike Galbraith
On Fri, 2013-04-19 at 10:50 +0200, Vincent Guittot wrote: > On 19 April 2013 10:14, Mike Galbraith wrote: > > On Fri, 2013-04-19 at 09:49 +0200, Vincent Guittot wrote: > >> On 19 April 2013 06:30, Mike Galbraith wrote: > >> > On Thu, 2013-04-18 at 18:34 +0200, Vincent Guittot wrote: > >> >> The

Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-19 Thread Vincent Guittot
On 19 April 2013 10:14, Mike Galbraith wrote: > On Fri, 2013-04-19 at 09:49 +0200, Vincent Guittot wrote: >> On 19 April 2013 06:30, Mike Galbraith wrote: >> > On Thu, 2013-04-18 at 18:34 +0200, Vincent Guittot wrote: >> >> The current update of the rq's load can be erroneous when RT tasks are >>

Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-19 Thread Mike Galbraith
On Fri, 2013-04-19 at 09:49 +0200, Vincent Guittot wrote: > On 19 April 2013 06:30, Mike Galbraith wrote: > > On Thu, 2013-04-18 at 18:34 +0200, Vincent Guittot wrote: > >> The current update of the rq's load can be erroneous when RT tasks are > >> involved > >> > >> The update of the load of a r

Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-19 Thread Vincent Guittot
On 19 April 2013 06:30, Mike Galbraith wrote: > On Thu, 2013-04-18 at 18:34 +0200, Vincent Guittot wrote: >> The current update of the rq's load can be erroneous when RT tasks are >> involved >> >> The update of the load of a rq that becomes idle, is done only if the >> avg_idle >> is less than s

Re: [PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-18 Thread Mike Galbraith
On Thu, 2013-04-18 at 18:34 +0200, Vincent Guittot wrote: > The current update of the rq's load can be erroneous when RT tasks are > involved > > The update of the load of a rq that becomes idle, is done only if the avg_idle > is less than sysctl_sched_migration_cost. If RT tasks and short idle d

[PATCH Resend v6] sched: fix wrong rq's runnable_avg update with rt tasks

2013-04-18 Thread Vincent Guittot
The current update of the rq's load can be erroneous when RT tasks are involved The update of the load of a rq that becomes idle, is done only if the avg_idle is less than sysctl_sched_migration_cost. If RT tasks and short idle duration alternate, the runnable_avg will not be updated correctly and