On Tue, Feb 17, 2015 at 09:55:21AM -0500, Alan Stern wrote:
> On Mon, 16 Feb 2015, Felipe Balbi wrote:
> 
> > Electrical tests for resume signaling. It failed with a specific hub,
> > however :-) But it's easy to note with dwc3/musb and a USB sniffer.
> > Sometimes resume signalling is cut before 20ms because we had msleep(20)
> > :-)
> 
> Maybe that should be reported to the architecture maintainers.  I'm not 
> certain, but I'm pretty sure that msleep(20) should _never_ sleep for 
> less than 20 ms.

doesn't lpj calibration play a role here ?

-- 
balbi

Attachment: signature.asc
Description: Digital signature

Reply via email to