Messages in this thread | | | Date | Mon, 13 Mar 2023 10:06:41 +0100 | Subject | Re: [PATCH v2] sched/fair: sanitize vruntime of entity being migrated | From | Dietmar Eggemann <> |
| |
On 10/03/2023 15:29, Vincent Guittot wrote: > Le jeudi 09 mars 2023 � 16:14:38 (+0100), Vincent Guittot a �crit : >> On Thu, 9 Mar 2023 at 15:37, Peter Zijlstra <peterz@infradead.org> wrote: >>> >>> On Thu, Mar 09, 2023 at 03:28:25PM +0100, Peter Zijlstra wrote: >>>> On Thu, Mar 09, 2023 at 02:34:05PM +0100, Vincent Guittot wrote:
[...]
>> you mean taking the patch 1/2 that you mentioned here to add a >> migrated field: >> https://lore.kernel.org/all/68832dfbb60fda030540b5f4e39c5801942689b1.1648228023.git.tim.c.chen@linux.intel.com/T/#ma5637eb8010f3f4a4abff778af8db705429d003b >> >> And assume that the divergence between the rq_clock_task() can be ignored ? >> >> That could probably work but we need to replace the (60LL * >> NSEC_PER_SEC) by ((1ULL << 63) / NICE_0_LOAD) because 60sec divergence >> would not be unrealistic. >> and a comment to explain why it's acceptable > > Zhang, > > Could you try the patch below ? > This is a rebase/merge/update of: > -patch 1/2 above and > -https://lore.kernel.org/lkml/20230209193107.1432770-1-rkagan@amazon.de/ > > The proposal accepts a divergence of up to 52 days between the 2 rqs. > > If this work, we will prepare a proper patch
Looks to me that this patch brings back the old numbers:
model name : Intel(R) Xeon(R) Silver 4314 CPU @ 2.40GHz
perf stat --null --repeat 10 -- perf bench sched messaging -g 50 -l 5000
tip sched/core
a2e90611b9f4 - sched/fair: Remove capacity inversion detection (2023-02-11 Vincent Guittot)
5.7295 +- 0.0219 seconds time elapsed ( +- 0.38% )
829c1651e9c4 - sched/fair: sanitize vruntime of entity being placed (2023-02-11 Zhang Qiao)
6.0961 +- 0.0297 seconds time elapsed ( +- 0.49% )
this patch on top 829c1651e9c4
5.7165 +- 0.0231 seconds time elapsed ( +- 0.40% )
[...]
| |