Re: [PATCH] sched/fair: Revert boost in cpu_util()

[Hongyan Xia]

On 5/27/2026 1:16 AM, Dietmar Eggemann wrote:
> On 22.05.26 11:36, hongyan.xia wrote:
>> On 5/22/2026 3:49 PM, Dietmar Eggemann wrote:
>>> On 18.05.26 04:40, hongyan.xia(夏弘彦) wrote:
>>>> From: Hongyan Xia <hongyan.xia@xxxxxxxxxxxxx>
> 
> [...]
> 
>>> Are you using Android on your devices? I remember there was some
>>> functionality added to avoid janks in display pipeline.
>>
>> Yes, this is Android but with scheduler vendor hooks stripped, so kernel
>> 6.6 schedutil with some GKI patches on top. There is "something to avoid
>> janks", but to make sure the results can be shared upstream, such
>> Android things are disabled, leaving basically vanilla schedutil.
> 
> OK.
> 
> Would be interesting to know how this 'avoid janks' feature can be
> disabled. So all ADPF ('Performance Hints' and/or 'Performance boost for
> games') hints are off? And this can be achieved by simply stripping all
> Vendor Hooks?
> I haven't looked at this for quite some time so I don't know how this is
> all wired up and how to disable this.

For our devices, the ADPF/PowerHAL is entirely written or controlled by 
ourselves so it's fairly easy to disable everything. After disabling 
everything it's pretty much vanilla schedutil.

Your mileage may vary because I assume you use Pixels, which might not 
be easy to know how everything can be enabled/disabled. Maybe the Pixel 
mainline kernels can help?
> [...]