Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-08-10 Thread Vincent Guittot
On Mon, 6 Aug 2018 at 12:53, Valentin Schneider wrote: > > Hi, > > On 06/08/18 11:20, Vincent Guittot wrote: > > Hi Valentin, > > > > On Tue, 31 Jul 2018 at 14:33, Valentin Schneider > > wrote: > >> > >> Hi, > >> > >> On 31/07/18 13:17, Vincent Guittot wrote: > >> [...] > > This can

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-08-10 Thread Vincent Guittot
On Mon, 6 Aug 2018 at 12:53, Valentin Schneider wrote: > > Hi, > > On 06/08/18 11:20, Vincent Guittot wrote: > > Hi Valentin, > > > > On Tue, 31 Jul 2018 at 14:33, Valentin Schneider > > wrote: > >> > >> Hi, > >> > >> On 31/07/18 13:17, Vincent Guittot wrote: > >> [...] > > This can

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-08-06 Thread Valentin Schneider
Hi, On 06/08/18 11:20, Vincent Guittot wrote: > Hi Valentin, > > On Tue, 31 Jul 2018 at 14:33, Valentin Schneider > wrote: >> >> Hi, >> >> On 31/07/18 13:17, Vincent Guittot wrote: >> [...] This can easily happen with SD_PREFER_SIBLING enabled too so I wouldn't say that this

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-08-06 Thread Valentin Schneider
Hi, On 06/08/18 11:20, Vincent Guittot wrote: > Hi Valentin, > > On Tue, 31 Jul 2018 at 14:33, Valentin Schneider > wrote: >> >> Hi, >> >> On 31/07/18 13:17, Vincent Guittot wrote: >> [...] This can easily happen with SD_PREFER_SIBLING enabled too so I wouldn't say that this

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-08-06 Thread Vincent Guittot
Hi Valentin, On Tue, 31 Jul 2018 at 14:33, Valentin Schneider wrote: > > Hi, > > On 31/07/18 13:17, Vincent Guittot wrote: > > On Fri, 6 Jul 2018 at 16:31, Morten Rasmussen > > wrote: > >> > >> On Fri, Jul 06, 2018 at 12:18:17PM +0200, Vincent Guittot wrote: > >>> [...] > >> > >> Scheduling

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-08-06 Thread Vincent Guittot
Hi Valentin, On Tue, 31 Jul 2018 at 14:33, Valentin Schneider wrote: > > Hi, > > On 31/07/18 13:17, Vincent Guittot wrote: > > On Fri, 6 Jul 2018 at 16:31, Morten Rasmussen > > wrote: > >> > >> On Fri, Jul 06, 2018 at 12:18:17PM +0200, Vincent Guittot wrote: > >>> [...] > >> > >> Scheduling

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-07-31 Thread Valentin Schneider
Hi, On 31/07/18 13:17, Vincent Guittot wrote: > On Fri, 6 Jul 2018 at 16:31, Morten Rasmussen > wrote: >> >> On Fri, Jul 06, 2018 at 12:18:17PM +0200, Vincent Guittot wrote: >>> [...] >> >> Scheduling one task per cpu when n_task == n_cpus on asymmetric >> topologies is generally broken already

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-07-31 Thread Valentin Schneider
Hi, On 31/07/18 13:17, Vincent Guittot wrote: > On Fri, 6 Jul 2018 at 16:31, Morten Rasmussen > wrote: >> >> On Fri, Jul 06, 2018 at 12:18:17PM +0200, Vincent Guittot wrote: >>> [...] >> >> Scheduling one task per cpu when n_task == n_cpus on asymmetric >> topologies is generally broken already

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-07-31 Thread Vincent Guittot
On Fri, 6 Jul 2018 at 16:31, Morten Rasmussen wrote: > > On Fri, Jul 06, 2018 at 12:18:17PM +0200, Vincent Guittot wrote: > > On Wed, 4 Jul 2018 at 12:18, Morten Rasmussen > > wrote: > > > > > > The 'prefer sibling' sched_domain flag is intended to encourage > > > spreading tasks to sibling

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-07-31 Thread Vincent Guittot
On Fri, 6 Jul 2018 at 16:31, Morten Rasmussen wrote: > > On Fri, Jul 06, 2018 at 12:18:17PM +0200, Vincent Guittot wrote: > > On Wed, 4 Jul 2018 at 12:18, Morten Rasmussen > > wrote: > > > > > > The 'prefer sibling' sched_domain flag is intended to encourage > > > spreading tasks to sibling

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-07-06 Thread Morten Rasmussen
On Fri, Jul 06, 2018 at 12:18:17PM +0200, Vincent Guittot wrote: > On Wed, 4 Jul 2018 at 12:18, Morten Rasmussen > wrote: > > > > The 'prefer sibling' sched_domain flag is intended to encourage > > spreading tasks to sibling sched_domain to take advantage of more caches > > and core for SMT

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-07-06 Thread Morten Rasmussen
On Fri, Jul 06, 2018 at 12:18:17PM +0200, Vincent Guittot wrote: > On Wed, 4 Jul 2018 at 12:18, Morten Rasmussen > wrote: > > > > The 'prefer sibling' sched_domain flag is intended to encourage > > spreading tasks to sibling sched_domain to take advantage of more caches > > and core for SMT

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-07-06 Thread Vincent Guittot
On Wed, 4 Jul 2018 at 12:18, Morten Rasmussen wrote: > > The 'prefer sibling' sched_domain flag is intended to encourage > spreading tasks to sibling sched_domain to take advantage of more caches > and core for SMT systems. It has recently been changed to be on all > non-NUMA topology level.

Re: [PATCHv4 12/12] sched/core: Disable SD_PREFER_SIBLING on asymmetric cpu capacity domains

2018-07-06 Thread Vincent Guittot
On Wed, 4 Jul 2018 at 12:18, Morten Rasmussen wrote: > > The 'prefer sibling' sched_domain flag is intended to encourage > spreading tasks to sibling sched_domain to take advantage of more caches > and core for SMT systems. It has recently been changed to be on all > non-NUMA topology level.