Re: [PATCH] sched: support dynamiQ cluster

2018-04-13 Thread Joel Fernandes (Google)
On Fri, Apr 6, 2018 at 5:58 AM, Morten Rasmussen wrote: > On Thu, Apr 05, 2018 at 06:22:48PM +0200, Vincent Guittot wrote: >> Hi Morten, >> >> On 5 April 2018 at 17:46, Morten Rasmussen wrote: >> > On Wed, Apr 04, 2018 at 03:43:17PM +0200, Vincent Guittot wrote: >> >> On 4 April 2018 at 12:44, Va

Re: [PATCH] sched: support dynamiQ cluster

2018-04-13 Thread Vincent Guittot
On 12 April 2018 at 20:22, Peter Zijlstra wrote: > On Tue, Apr 10, 2018 at 02:19:50PM +0100, Morten Rasmussen wrote: >> As said above, I see your point about completion time might suffer in >> some cases for low utilization tasks, but I don't see how you can fix >> that automagically. ASYM_PACKING

Re: [PATCH] sched: support dynamiQ cluster

2018-04-13 Thread Morten Rasmussen
On Thu, Apr 12, 2018 at 08:22:11PM +0200, Peter Zijlstra wrote: > On Tue, Apr 10, 2018 at 02:19:50PM +0100, Morten Rasmussen wrote: > > As said above, I see your point about completion time might suffer in > > some cases for low utilization tasks, but I don't see how you can fix > > that automagica

Re: [PATCH] sched: support dynamiQ cluster

2018-04-12 Thread Peter Zijlstra
On Tue, Apr 10, 2018 at 02:19:50PM +0100, Morten Rasmussen wrote: > As said above, I see your point about completion time might suffer in > some cases for low utilization tasks, but I don't see how you can fix > that automagically. ASYM_PACKING has a lot of problematic side-effects. > If use-space

Re: [PATCH] sched: support dynamiQ cluster

2018-04-10 Thread Morten Rasmussen
On Mon, Apr 09, 2018 at 09:34:00AM +0200, Vincent Guittot wrote: > Hi Morten, > > On 6 April 2018 at 14:58, Morten Rasmussen wrote: > > On Thu, Apr 05, 2018 at 06:22:48PM +0200, Vincent Guittot wrote: > >> Hi Morten, > >> > >> On 5 April 2018 at 17:46, Morten Rasmussen > >> wrote: > >> > On Wed

Re: [PATCH] sched: support dynamiQ cluster

2018-04-09 Thread Vincent Guittot
Hi Morten, On 6 April 2018 at 14:58, Morten Rasmussen wrote: > On Thu, Apr 05, 2018 at 06:22:48PM +0200, Vincent Guittot wrote: >> Hi Morten, >> >> On 5 April 2018 at 17:46, Morten Rasmussen wrote: >> > On Wed, Apr 04, 2018 at 03:43:17PM +0200, Vincent Guittot wrote: >> >> On 4 April 2018 at 12:

Re: [PATCH] sched: support dynamiQ cluster

2018-04-06 Thread Morten Rasmussen
On Thu, Apr 05, 2018 at 06:22:48PM +0200, Vincent Guittot wrote: > Hi Morten, > > On 5 April 2018 at 17:46, Morten Rasmussen wrote: > > On Wed, Apr 04, 2018 at 03:43:17PM +0200, Vincent Guittot wrote: > >> On 4 April 2018 at 12:44, Valentin Schneider > >> wrote: > >> > Hi, > >> > > >> > On 03/0

Re: [PATCH] sched: support dynamiQ cluster

2018-04-05 Thread Vincent Guittot
Hi Morten, On 5 April 2018 at 17:46, Morten Rasmussen wrote: > On Wed, Apr 04, 2018 at 03:43:17PM +0200, Vincent Guittot wrote: >> On 4 April 2018 at 12:44, Valentin Schneider >> wrote: >> > Hi, >> > >> > On 03/04/18 13:17, Vincent Guittot wrote: >> >> Hi Valentin, >> >> >> > [...] >> >>> >> >>

Re: [PATCH] sched: support dynamiQ cluster

2018-04-05 Thread Morten Rasmussen
On Wed, Apr 04, 2018 at 03:43:17PM +0200, Vincent Guittot wrote: > On 4 April 2018 at 12:44, Valentin Schneider > wrote: > > Hi, > > > > On 03/04/18 13:17, Vincent Guittot wrote: > >> Hi Valentin, > >> > > [...] > >>> > >>> I believe ASYM_PACKING behaves better here because the workload is only >

Re: [PATCH] sched: support dynamiQ cluster

2018-04-04 Thread Vincent Guittot
On 4 April 2018 at 12:44, Valentin Schneider wrote: > Hi, > > On 03/04/18 13:17, Vincent Guittot wrote: >> Hi Valentin, >> > [...] >>> >>> I believe ASYM_PACKING behaves better here because the workload is only >>> sysbench threads. As stated above, since task utilization is disregarded, I >> >> I

Re: [PATCH] sched: support dynamiQ cluster

2018-04-04 Thread Valentin Schneider
Hi, On 03/04/18 13:17, Vincent Guittot wrote: > Hi Valentin, > [...] >> >> I believe ASYM_PACKING behaves better here because the workload is only >> sysbench threads. As stated above, since task utilization is disregarded, I > > It behaves better because it doesn't wait for the task's utilizati

Re: [PATCH] sched: support dynamiQ cluster

2018-04-03 Thread Vincent Guittot
Hi Valentin, On 3 April 2018 at 00:27, Valentin Schneider wrote: > Hi, > > On 30/03/18 13:34, Vincent Guittot wrote: >> Hi Morten, >> > [..] >>> >>> As I see it, the main differences is that ASYM_PACKING attempts to pack >>> all tasks regardless of task utilization on the higher capacity cpus >>>

Re: [PATCH] sched: support dynamiQ cluster

2018-04-02 Thread Valentin Schneider
Hi, On 30/03/18 13:34, Vincent Guittot wrote: > Hi Morten, > [..] >> >> As I see it, the main differences is that ASYM_PACKING attempts to pack >> all tasks regardless of task utilization on the higher capacity cpus >> whereas the "misfit task" series carefully picks cpus with tasks they >> can't

Re: [PATCH] sched: support dynamiQ cluster

2018-03-30 Thread Vincent Guittot
Hi Morten, On 29 March 2018 at 14:53, Morten Rasmussen wrote: > On Wed, Mar 28, 2018 at 09:46:55AM +0200, Vincent Guittot wrote: >> Arm DynamiQ system can integrate cores with different micro architecture >> or max OPP under the same DSU so we can have cores with different compute >> capacity at

Re: [PATCH] sched: support dynamiQ cluster

2018-03-29 Thread Morten Rasmussen
On Wed, Mar 28, 2018 at 09:46:55AM +0200, Vincent Guittot wrote: > Arm DynamiQ system can integrate cores with different micro architecture > or max OPP under the same DSU so we can have cores with different compute > capacity at the LLC (which was not the case with legacy big/LITTLE > architecture

Re: [PATCH] sched: support dynamiQ cluster

2018-03-28 Thread Vincent Guittot
On 28 March 2018 at 11:12, Will Deacon wrote: > On Wed, Mar 28, 2018 at 09:46:55AM +0200, Vincent Guittot wrote: >> >> The SD_ASYM_PACKING flag is disabled by default and I'm preparing another >> patch >> to enable this dynamically at boot time by detecting the system topology. >> >> arch/arm64

Re: [PATCH] sched: support dynamiQ cluster

2018-03-28 Thread Will Deacon
On Wed, Mar 28, 2018 at 09:46:55AM +0200, Vincent Guittot wrote: > Arm DynamiQ system can integrate cores with different micro architecture > or max OPP under the same DSU so we can have cores with different compute > capacity at the LLC (which was not the case with legacy big/LITTLE > architecture

[PATCH] sched: support dynamiQ cluster

2018-03-28 Thread Vincent Guittot
Arm DynamiQ system can integrate cores with different micro architecture or max OPP under the same DSU so we can have cores with different compute capacity at the LLC (which was not the case with legacy big/LITTLE architecture). Such configuration is similar in some way to ITMT on intel platform wh