Re: [PATCH] PM: runtime: Update runtime_pm.rst for non-fatal resume errors

[Stanley Jhu]

On Tue, Nov 03, 2026 at 11:47:58AM -0800, Brian Norris wrote:
> Those need a close eye, as the code involved is a bit subtle. I tried to
> review, but I'm not really an expert in those subsystems. (I'm also not
> quite sure the second one was 100% a regression, vs further exposing an
> existing underlying problem.)

Agreed on the UFS clock-gating hold release. Keeping the hold until
resume succeeds was cleaner regardless of retry behavior. That fix has
been applied to 7.4/scsi-fixes. I have also sent v2 of the block-layer
q->pm_only fix.

> I think it would be better to flip the comparison to the other
> direction, and deemphasize power.runtime_error. For example:
>
>   "Unlike with the suspend callback, failures in the resume callback are
>   never considered fatal, [...]"

I have sent v2 using that comparison, without mentioning -EBUSY,
-EAGAIN, or power.runtime_error, and added Fixes for commit
aedc51762a0d covering the ->runtime_idle() paragraph.