Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1440365 > unrolled thread
| Started by | Morten Rasmussen <morten.rasmussen@arm.com> |
|---|---|
| First post | 2016-07-11 10:40 +0200 |
| Last post | 2016-07-12 16:40 +0200 |
| Articles | 4 — 3 participants |
Back to article view | Back to linux.kernel
This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by
below is the oldest one visible, not the original post.
Re: [PATCH v2 00/13] sched: Clean-ups and asymmetric cpu capacity support Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-11 10:40 +0200
Re: [PATCH v2 00/13] sched: Clean-ups and asymmetric cpu capacity support Vincent Guittot <vincent.guittot@linaro.org> - 2016-07-11 14:50 +0200
Re: [PATCH v2 00/13] sched: Clean-ups and asymmetric cpu capacity support Peter Zijlstra <peterz@infradead.org> - 2016-07-12 15:30 +0200
Re: [PATCH v2 00/13] sched: Clean-ups and asymmetric cpu capacity support Morten Rasmussen <morten.rasmussen@arm.com> - 2016-07-12 16:40 +0200
| From | Morten Rasmussen <morten.rasmussen@arm.com> |
|---|---|
| Date | 2016-07-11 10:40 +0200 |
| Subject | Re: [PATCH v2 00/13] sched: Clean-ups and asymmetric cpu capacity support |
| Message-ID | <rTEUW-4zH-43@gated-at.bofh.it> |
On Wed, Jun 22, 2016 at 06:03:11PM +0100, Morten Rasmussen wrote: > Hi, > > The scheduler is currently not doing much to help performance on systems with > asymmetric compute capacities (read ARM big.LITTLE). This series improves the > situation with a few tweaks mainly to the task wake-up path that considers > compute capacity at wake-up and not just whether a cpu is idle for these > systems. This gives us consistent, and potentially higher, throughput in > partially utilized scenarios. SMP behaviour and performance should be > unaffected. Peter, Vincent: Any chance you have time for another look? Thanks, Morten
[toc] | [next] | [standalone]
| From | Vincent Guittot <vincent.guittot@linaro.org> |
|---|---|
| Date | 2016-07-11 14:50 +0200 |
| Subject | Re: [PATCH v2 00/13] sched: Clean-ups and asymmetric cpu capacity support |
| Message-ID | <rTIOT-762-41@gated-at.bofh.it> |
| In reply to | #1440365 |
Hi Morten, On 11 July 2016 at 10:33, Morten Rasmussen <morten.rasmussen@arm.com> wrote: > On Wed, Jun 22, 2016 at 06:03:11PM +0100, Morten Rasmussen wrote: >> Hi, >> >> The scheduler is currently not doing much to help performance on systems with >> asymmetric compute capacities (read ARM big.LITTLE). This series improves the >> situation with a few tweaks mainly to the task wake-up path that considers >> compute capacity at wake-up and not just whether a cpu is idle for these >> systems. This gives us consistent, and potentially higher, throughput in >> partially utilized scenarios. SMP behaviour and performance should be >> unaffected. > > Peter, Vincent: Any chance you have time for another look? I will have a look but have only partial network access for now Vincent > > Thanks, > Morten
[toc] | [prev] | [next] | [standalone]
| From | Peter Zijlstra <peterz@infradead.org> |
|---|---|
| Date | 2016-07-12 15:30 +0200 |
| Message-ID | <rU5V7-5rJ-13@gated-at.bofh.it> |
| In reply to | #1440365 |
On Mon, Jul 11, 2016 at 09:33:54AM +0100, Morten Rasmussen wrote: > On Wed, Jun 22, 2016 at 06:03:11PM +0100, Morten Rasmussen wrote: > > Hi, > > > > The scheduler is currently not doing much to help performance on systems with > > asymmetric compute capacities (read ARM big.LITTLE). This series improves the > > situation with a few tweaks mainly to the task wake-up path that considers > > compute capacity at wake-up and not just whether a cpu is idle for these > > systems. This gives us consistent, and potentially higher, throughput in > > partially utilized scenarios. SMP behaviour and performance should be > > unaffected. > > Peter, Vincent: Any chance you have time for another look? So no fundamental objections, just these few issues that need sorting.
[toc] | [prev] | [next] | [standalone]
| From | Morten Rasmussen <morten.rasmussen@arm.com> |
|---|---|
| Date | 2016-07-12 16:40 +0200 |
| Message-ID | <rU70S-6aC-47@gated-at.bofh.it> |
| In reply to | #1441345 |
On Tue, Jul 12, 2016 at 03:25:48PM +0200, Peter Zijlstra wrote: > On Mon, Jul 11, 2016 at 09:33:54AM +0100, Morten Rasmussen wrote: > > On Wed, Jun 22, 2016 at 06:03:11PM +0100, Morten Rasmussen wrote: > > > Hi, > > > > > > The scheduler is currently not doing much to help performance on systems with > > > asymmetric compute capacities (read ARM big.LITTLE). This series improves the > > > situation with a few tweaks mainly to the task wake-up path that considers > > > compute capacity at wake-up and not just whether a cpu is idle for these > > > systems. This gives us consistent, and potentially higher, throughput in > > > partially utilized scenarios. SMP behaviour and performance should be > > > unaffected. > > > > Peter, Vincent: Any chance you have time for another look? > > So no fundamental objections, just these few issues that need sorting. Thanks for the review. I will put together a v3.
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web