Re: [BUG] - Short freezes in gameplay due to MMC_CAP_AGGRESSIVE_PM on RTS525A card reader

[Bart Van Assche <bvanassche@acm.org>]

On 1/29/26 8:30 AM, Ulf Hansson wrote:
> Seems reasonable to me, but how do we distinguish that it's a
> battery-powered device?

RPM can be enabled or disabled from the scripts executed during boot.
At build time it should be known whether or not the device is battery-
powered.

> Are we considering UFS a technology that is used solely for
> battery-powered devices or is there something else we consider?

I think there are devices that use UFS and that are not battery-powered,
e.g. smart TVs.

> Although, a tricky part when moving it upwards into the more common
> layers, is that those latency constraint values may have a very
> different impact, as the numbers are platform specific, right?

 From the UFS driver:

cpu_latency_qos_update_request(&hba->pm_qos_req, on ?
                                0 : PM_QOS_DEFAULT_VALUE);

In other words, if no block I/O is ongoing the CPU latency is set to
PM_QOS_DEFAULT_VALUE (-1 or no constraint) and if block I/O is ongoing
the maximum CPU latency is set to 0 (no CPU power savings allowed). I
think these parameters are independent of the platform and storage
device :-)

Bart.