Re: [PATCH 1/2] sched/fair: move cpufreq hook to update_cfs_rq_load_avg()

[Steve Muckle]

On 03/31/2016 12:37 AM, Peter Zijlstra wrote:
> On Wed, Mar 30, 2016 at 06:42:20PM -0700, Steve Muckle wrote:
>> On 03/30/2016 12:35 PM, Peter Zijlstra wrote:
>>> On Mon, Mar 28, 2016 at 12:38:26PM -0700, Steve Muckle wrote:
>>>> Without covering all the paths where CFS utilization changes it's
>>>> possible to have to wait up to a tick to act on some changes, since the
>>>> tick is the only guaranteed regularly-occurring instance of the hook.
>>>> That's an unacceptable amount of latency IMO...
>>>
>>> Note that even with your patches that might still be the case. Remote
>>> wakeups might not happen on the destination CPU at all, so it might not
>>> be until the next tick (which always happens locally) that we'll
>>> 'observe' the utilization change brought with the wakeups.
>>>
>>> We could force all the remote wakeups to IPI the destination CPU, but
>>> that comes at a significant performance cost.
>>
>> What about only IPI'ing the destination when the utilization change is
>> known to require a higher CPU frequency?
> 
> Can't, the way the wakeup path is constructed we would be sending the
> IPI way before we know about utilization.

Sorry I thought we were referring to the possibility of sending an IPI
to just run the cpufreq driver rather than to conduct the whole wakeup
operation.

My thinking was in CFS we get rid of the (cpu == smp_processor_id())
condition for calling the cpufreq hook.

The sched governor can then calculate utilization and frequency required
for cpu. If (cpu == smp_processor_id()), the update is processed
normally. If (cpu != smp_processor_id()) and the new frequency is higher
than cpu's Fcur, the sched gov IPIs cpu to continue running the update
operation. Otherwise, the update is dropped.

Does that sound plausible?