Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-25 Thread Valentin Schneider
On 25/02/21 09:05, Vincent Guittot wrote: >> One last thing for patch 7: mayhaps we could do a tad better to avoid >> duplicate updates going through a heapful of leaf cfs rqs, see >> >> http://lore.kernel.org/r/jhj4kiht7oh.mog...@arm.com > > rq->last_blocked_load_update_tick is there only to fil

Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-25 Thread Vincent Guittot
Hi Valentin, On Wed, 24 Feb 2021 at 19:46, Valentin Schneider wrote: > > On 24/02/21 14:30, Vincent Guittot wrote: > > Joel reported long preempt and irq off sequence in newidle_balance because > > of a large number of CPU cgroups in use and having to be updated. This > > patchset moves the updat

Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-24 Thread Valentin Schneider
On 24/02/21 14:30, Vincent Guittot wrote: > Joel reported long preempt and irq off sequence in newidle_balance because > of a large number of CPU cgroups in use and having to be updated. This > patchset moves the update outside newidle_imblance. This enables to early > abort during the updates in c

Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-24 Thread Peter Zijlstra
On Wed, Feb 24, 2021 at 06:51:01PM +0100, Vincent Guittot wrote: > > OK, shall I add something like: > > > > This reduces the IRQ latency from O(nr_cgroups * nr_nohz_cpus) to > > O(nr_cgroups). > > > > To the changelog of patch #1 ? > > Yes, good point. This will clarify the range of improvement

Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-24 Thread Vincent Guittot
On Wed, 24 Feb 2021 at 18:41, Peter Zijlstra wrote: > > On Wed, Feb 24, 2021 at 04:57:15PM +0100, Vincent Guittot wrote: > > On Wed, 24 Feb 2021 at 16:54, Peter Zijlstra wrote: > > > > > > On Wed, Feb 24, 2021 at 02:30:00PM +0100, Vincent Guittot wrote: > > > > Joel reported long preempt and irq

Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-24 Thread Peter Zijlstra
On Wed, Feb 24, 2021 at 04:57:15PM +0100, Vincent Guittot wrote: > On Wed, 24 Feb 2021 at 16:54, Peter Zijlstra wrote: > > > > On Wed, Feb 24, 2021 at 02:30:00PM +0100, Vincent Guittot wrote: > > > Joel reported long preempt and irq off sequence in newidle_balance because > > > of a large number o

Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-24 Thread Vincent Guittot
On Wed, 24 Feb 2021 at 16:54, Peter Zijlstra wrote: > > On Wed, Feb 24, 2021 at 02:30:00PM +0100, Vincent Guittot wrote: > > Joel reported long preempt and irq off sequence in newidle_balance because > > of a large number of CPU cgroups in use and having to be updated. This > > patchset moves the

Re: [PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-24 Thread Peter Zijlstra
On Wed, Feb 24, 2021 at 02:30:00PM +0100, Vincent Guittot wrote: > Joel reported long preempt and irq off sequence in newidle_balance because > of a large number of CPU cgroups in use and having to be updated. This > patchset moves the update outside newidle_imblance. This enables to early > abort

[PATCH 0/7 v4] move update blocked load outside newidle_balance

2021-02-24 Thread Vincent Guittot
Joel reported long preempt and irq off sequence in newidle_balance because of a large number of CPU cgroups in use and having to be updated. This patchset moves the update outside newidle_imblance. This enables to early abort during the updates in case of pending irq as an example. Instead of kick