MUSB dual-role on AM335x behaving weirdly

[balbi@ti.com (Felipe Balbi)]

On Thu, May 14, 2015 at 12:07:00PM -0500, Felipe Balbi wrote:
> Hi,
> 
> On Wed, Feb 25, 2015 at 01:11:22PM +0100, Yegor Yefremov wrote:
> > On 25.02.2015 12:11, Maxime Ripard wrote:
> > > On Tue, Feb 24, 2015 at 11:33:57AM -0600, Felipe Balbi wrote:
> > >> Hi,
> > >>
> > >> On Tue, Feb 24, 2015 at 05:50:50PM +0100, Maxime Ripard wrote:
> > >>> Hi Felipe,
> > >>>
> > >>> On Tue, Feb 24, 2015 at 08:54:01AM -0600, Felipe Balbi wrote:
> > >>>> Hi,
> > >>>>
> > >>>> On Tue, Feb 24, 2015 at 11:39:11AM +0100, Maxime Ripard wrote:
> > >>>>> On Thu, Feb 05, 2015 at 02:21:42PM +0100, Maxime Ripard wrote:
> > >>>>>> Hi,
> > >>>>>>
> > >>>>>> On Thu, Jan 22, 2015 at 08:37:45AM +0100, Yegor Yefremov wrote:
> > >>>>>>> I have the same experience with 3.15. The switching is working when
> > >>>>>>> CONFIG_USB_MUSB_DUAL_ROLE is set and dr_mode = "otg". But since 3.16
> > >>
> > >> since 3.16 ?
> > > 
> > > That's what Yegor said. I never saw it working with 3.15 either.
> > 
> > I've used 3.15.1 and 3.15.2 with this set of patches:
> > https://github.com/visionsystemsgmbh/onrisc_br_bsp/tree/master/board/vscom/kernel-patches/linux-3.15
> > 
> > And it worked so far. The system:
> > http://www.visionsystems.de/produkte/baltos-ir-5221.html
> 
> I've had more time to look into this (thanks Yegor for sponsoring a
> test/dev platform) what I noticed is that Connect IRQ takes seconds to
> fire up.  Below a tiny log snippet after pluging USB OTG adapter cable
> that came with IR5521:
> 
> | [ 1227.200514] musb-hdrc musb-hdrc.1.auto: usbintr (100) epintr(0)
> 
> Cable connected. ID is grounded. 0x100 == DRVVBUS IRQ
> 
> | [ 1227.206788] musb-hdrc musb-hdrc.1.auto: VBUS on (a_wait_vrise), devctl 19
> 
> MUSB starts to wait for VBUS to reach Session valid threshold
> 
> | [ 1230.281159] musb-hdrc musb-hdrc.1.auto: usbintr (10) epintr(0)
> 
> 3 seconds later connect interrupt happens. Looking at VBUS charge time
> with a scope, it's quite ok. VBUS charges in about 1.77ms. I'll dig
> further into this.

even more weird. If I disconnect device from OTG adapter, rather than
OTG adapter from IR5521, this leave ID pin grounded, which means DRVVBUS
is still asserted and VBUS remains above session valid threshold.

Even in this case, when I connect the device on the other end of the
cable, I still see some 3 seconds delay from the time device is
connected, to the time connect IRQ fires up.

-- 
balbi
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20150514/3ef6f526/attachment.sig>