[U-Boot] [linux-sunxi] [PATCH] phy: sun4i-usb: Fix PHY0 routing and passby configuration for MUSB

[Paul Kocialkowski]

Hi,

On Mon, 2019-05-27 at 00:50 +0100, André Przywara wrote:
> On 17/04/2019 12:28, Jagan Teki wrote:
> > On Mon, Apr 15, 2019 at 1:52 PM Paul Kocialkowski
> > <paul.kocialkowski at bootlin.com> wrote:
> 
> Hi,
> 
> > > Le vendredi 12 avril 2019 à 14:49 +0530, Jagan Teki a écrit :
> > > > On Thu, Mar 14, 2019 at 4:08 PM Paul Kocialkowski
> > > > <paul.kocialkowski at bootlin.com> wrote:
> > > > > Recent Allwinner platforms (starting with the H3) only use the MUSB
> > > > > controller for peripheral mode and use HCI for host mode. As a result,
> > > > > extra steps need to be taken to properly route USB signals to one or
> > > > > the other. More precisely, the following is required:
> > > > > * Routing the pins to either HCI/MUSB (controlled by PHY);
> > > > > * Enabling USB PHY passby in HCI mode (controlled by PMU).
> > > > > 
> > > > > The current code will enable passby for each PHY and reroute PHY0 to
> > > > > MUSB, which is inconsistent and results in broken USB peripheral support.
> > > > > 
> > > > > Passby on PHY0 must only be enabled when we want to use HCI. Since
> > > > > host/device mode detection is not available from the PHY code and
> > > > > because U-Boot does not support changing the mode dynamically anyway,
> > > > > we can just mux the controller to MUSB if it is enabled and mux it to
> > > > > HCI otherwise.
> > > > > 
> > > > > This fixes USB peripheral support for platforms with PHY0 dual-route,
> > > > > especially H3/H5 and V3s.
> > > > > 
> > > > > Signed-off-by: Paul Kocialkowski <paul.kocialkowski at bootlin.com>
> > > > > ---
> > > > >  drivers/phy/allwinner/phy-sun4i-usb.c | 14 +++++++++++++-
> > > > >  1 file changed, 13 insertions(+), 1 deletion(-)
> > > > > 
> > > > > diff --git a/drivers/phy/allwinner/phy-sun4i-usb.c b/drivers/phy/allwinner/phy-sun4i-usb.c
> > > > > index f206fa3f5d48..4f1c7e519d71 100644
> > > > > --- a/drivers/phy/allwinner/phy-sun4i-usb.c
> > > > > +++ b/drivers/phy/allwinner/phy-sun4i-usb.c
> > > > > @@ -302,9 +302,21 @@ static int sun4i_usb_phy_init(struct phy *phy)
> > > > >                                     data->cfg->disc_thresh, PHY_DISCON_TH_LEN);
> > > > >         }
> > > > > 
> > > > > +#ifdef CONFIG_USB_MUSB_SUNXI
> > > > > +       /* Needed for HCI and conflicts with MUSB, keep PHY0 on MUSB */
> > > > > +       if (usb_phy->id != 0)
> > > > > +               sun4i_usb_phy_passby(phy, true);
> > > > > +
> > > > > +       /* Route PHY0 to MUSB to allow USB gadget */
> > > > > +       if (data->cfg->phy0_dual_route)
> > > > > +               sun4i_usb_phy0_reroute(data, true);
> > > > > +#else
> > > > >         sun4i_usb_phy_passby(phy, true);
> > > > > 
> > > > > -       sun4i_usb_phy0_reroute(data, true);
> > > > > +       /* Route PHY0 to HCI to allow USB host */
> > > > > +       if (data->cfg->phy0_dual_route)
> > > > > +               sun4i_usb_phy0_reroute(data, false);
> > > > > +#endif
> > > > 
> > > > I think we can manage this route and passby dynamically by detecting
> > > > id since we have dr_mode verify the MUSB host or peripheral via
> > > > usb_get_dr_mode, any chance to try that way?
> > > 
> > > Oh, I didn't know that U-Boot has support for usb_get_dr_mode these
> > > days. Thanks!
> > > 
> > > So far, the sunxi port has been using Kconfig to distinguish between
> > > host/device (unless I'm mistaken?) so I feel like this should be a
> > > separate follow-up patch to convert the sunxi MUSB glue + PHY to
> > > detecting dr_mode using usb_get_dr_mode. This feels like a significant
> > > rework, too.
> > 
> > Yes.
> > 
> > > Also, how should we handle the OTG case? I'm not sure we can support
> > > having both musb host and gadget built-in at this point. But that would
> > > certainly be welcome as part of the rework, too.
> > > 
> > > What do you think?
> > 
> > You mean handling dr_mode at runtime.
> > 
> > If yes, It is bit new where we can register the musb as UCLASS_MISC
> > wrapper and decide to bind driver for host and peripheral by decoding
> > dr_mode. and indeed host should register with UCLASS_USB and
> > peripheral with UCLASS_USB_GADGET_GENERIC.
> > 
> > I tried this wrapper before, not placed in-between because of other
> > work but TI musb has similar code to manage this
> > drivers/usb/musb-new/ti-musb.c
> 
> Before we go wild with any fancy rework, can we possibly take this patch
> as it? As I realised, this is basically a better version of the patch I
> sent two weeks ago [1]. I tried Paul's patch back then, but was missing
> the phys property in the DT, which I addressed in patch 2/2.

I fully agree, and believe the rework should be a separate subsequent
series.

> So I would appreciate if we can take this patch, as it solves a real
> problem (upper USB port not working) on many Pine64 boards (given the
> small DT change is in place). And on those boards the OTG functionality
> is not really feasible anyway, as VBUS is either permanently enabled or
> at least tied to the other host port's supply, so we can't turn it off
> for peripheral mode.

It makes USB OTG work on my V3/V3s boards as well and there's a chance
it fixes A83t as well.

Cheers,

Paul

> Cheers,
> Andre.
> 
> [1] https://lists.denx.de/pipermail/u-boot/2019-May/369521.html
-- 
Paul Kocialkowski, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com