Re: [PATCH v2 0/6] Update last busy timestamp in Runtime PM autosuspend callbacks

[Sakari Ailus <sakari.ailus@linux.intel.com>]

Hi Oliver,

On Mon, Jun 23, 2025 at 03:28:45PM +0200, Oliver Neukum wrote:
> On 16.06.25 08:12, Sakari Ailus wrote:
> > Folks,
> 
> Hi,
> > This set extends the inclusion of the pm_runtime_mark_last_busy() call to
> > the _autosuspend() variants of the Runtime PM functions dealing with
> > suspending devices, i.e. pm_runtime_put_autosuspend(),
> > pm_runtime_put_sync_autosuspend(), pm_runtime_autosuspend() and
> > pm_request_autosuspend(). This will introduce, for a brief amount of time,
> > unnecessary calls to pm_runtime_mark_last_busy() but this wasn't seen as
> > an issue. Also, all users of these functions, including those that did not
> > call pm_runtime_mark_last_busy(), will now include that call. Presumably
> > in the vast majority of the cases a missing call would have been a bug.
> 
> Now that I think about this, I am not sure of the logic behind this.
> It seems to me that you are making marking a device busy and dropping
> the reference synonymous.
> 
> Is the time you use to derive the likelihood determined by the start
> of IO or the end of IO?

The purpose of autosuspend is to delay device suspend in case it might be
used again soon after the point of dropping Runtime PM usage_count. Are
there cases you're aware of where pm_runtime_mark_last_busy() isn't
directly followed pm_runtime_put_autosuspend() or one of its variants and
it is desirable? Just stamping the last busy timestamp is a trivial
operation so one or two extra in a few places isn't really a problem.

-- 
Kind regards,

Sakari Ailus