On Thu, Jul 30, 2015 at 06:06:32PM +0800, Chen-Yu Tsai wrote:
> >> Last, I plan to add the new drivers to the defconfigs. OTG (musb)
> >> will not be added as the host/gadget/otg setting must match the dts
> >> setting, and that is per-board.  IIRC a mismatch might result in a
> >> oops.
> >
> > That seems quite bad. Doesn't the dual role option makes everything
> > works for everyone?
> 
> My setup isn't good for testing OTG/gadget. That needs modules, due
> to the way musb it hooked up.

No, it doesn't. There's an issue currently that if there's no gadget
registered when the USB gadget driver is probed, things simply don't
work.

The modules stuff is simply a workaround that race, but there's also
this patch: https://patchwork.kernel.org/patch/5999591/ that just
works, and will remove that race.

> I can test how it reacts if compiled as host with dts set to OTG. If
> it works out I guess we can set USB_MUSB_HOST=y.
> 
> So the problem is actually two-fold.
> 
>   1. There might be a oops. (or something else... don't really
>      remember specifics.)

And that one is a bit fishy.

Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

-- 
You received this message because you are subscribed to the Google Groups 
"linux-sunxi" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Attachment: signature.asc
Description: Digital signature

Reply via email to