Re: [PATCH 02/10] usb: musb: Fix idling after host mode by increasing autosuspend delay

[Tony Lindgren <tony-4v6yS6AI5VpBDgjK7y7TUQ@public.gmane.org>]

* Bin Liu <b-liu-l0cyMroinI0@public.gmane.org> [160505 08:04]:
> Hi,
> 
> On Thu, Apr 28, 2016 at 10:33:12AM -0700, Tony Lindgren wrote:
> > Looks like at least 2430 glue won't idle reliably with the 200 ms
> > autosuspend delay. This causes deeper idle states being blocked for
> > the whole SoC when disconnecting OTG A cable.
> > 
> > Increasing the delay to 500 ms seems to idle both MUSB and the PHY
> > reliably. This is probably because of time needed by the hardware
> > based negotiation between MUSB and the PHY.
> 
> Unless we know other glues also have the same issue, this change
> affects other glues, right? Can we just increase it for 2430 only?

Well it's hard to say with no PM working with other glue layers right
now. I would not be suprised if similar issue exists with other phys
too talking with musb.

Does the current 200 ms autosuspend timeout relate to anything real
other than what I found out with my measurements?

If there's no other data available for the 200 ms timeout, I suggest
we just set it to 500 ms and add comments there that at least 2430
glue needs that. If things are not idling properly for the whole
SoC because musb, the power consumption will be way worse than any
power savings between 200 and 500 ms :)

BTW, I noticed 300 ms did not make a difference there, and 500 ms
seems to work so far.

Regards,

Tony
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html