Re: [PATCH 1/3] usb: dwc2: override PHY input signals with usb role switch support

[Amelie DELAUNAY]

Hi Felipe,
Hi Martin,

I saw the issue reported ([balbi-usb:testing/next 2/17] 
drivers/usb/dwc2/drd.c:80:38: error: no member named 'test_mode' in 
'struct dwc2_hsotg'). I prepare a V2 fixing it + adressing Martin's 
comments.

Regards,
Amelie

On 7/19/20 9:56 PM, Martin Blumenstingl wrote:
> Hello Amelie,
>
> sorry for the late reply
>
> On Wed, Jul 8, 2020 at 6:00 PM Amelie DELAUNAY <amelie.delaunay@xxxxxx> 
> wrote:
> [...]
> > Could you please test with:
> >
> > static int dwc2_drd_role_sw_set(struct device *dev, enum usb_role role)
> > {
> > ÂÂÂÂÂÂÂÂ struct dwc2_hsotg *hsotg = dev_get_drvdata(dev);
> > ÂÂÂÂÂÂÂÂ unsigned long flags;
> > ÂÂÂÂÂÂÂÂ int already = 0;
> >
> > ÂÂÂÂÂÂÂÂ /* Skip session not in line with dr_mode */
> > ÂÂÂÂÂÂÂÂ if ((role == USB_ROLE_DEVICE && hsotg->dr_mode == USB_DR_MODE_HOST) ||
> > ÂÂÂÂÂÂÂÂÂÂÂÂ (role == USB_ROLE_HOST && hsotg->dr_mode == USB_DR_MODE_PERIPHERAL))
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ return -EINVAL;
> >
> > ÂÂÂÂÂÂÂÂ /* Skip session if core is in test mode */
> > ÂÂÂÂÂÂÂÂ if (role == USB_ROLE_NONE && hsotg->test_mode) {
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ dev_dbg(hsotg->dev, "Core is in test mode\n");
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ return -EBUSY;
> > ÂÂÂÂÂÂÂÂ }
> >
> > ÂÂÂÂÂÂÂÂ spin_lock_irqsave(&hsotg->lock, flags);
> >
> > ÂÂÂÂÂÂÂÂ if (role == USB_ROLE_HOST) {
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ already = dwc2_ovr_avalid(hsotg, true);
> > ÂÂÂÂÂÂÂÂ } else if (role == USB_ROLE_DEVICE) {
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ already = dwc2_ovr_bvalid(hsotg, true);
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ /* This clear DCTL.SFTDISCON bit */
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ dwc2_hsotg_core_connect(hsotg);
> > ÂÂÂÂÂÂÂÂ } else {
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ if (dwc2_is_device_mode(hsotg)) {
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ if (!dwc2_ovr_bvalid(hsotg, false))
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ /* This set DCTL.SFTDISCON bit */
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ dwc2_hsotg_core_disconnect(hsotg);
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ } else {
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ dwc2_ovr_avalid(hsotg, false);
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ }
> > ÂÂÂÂÂÂÂÂ }
> >
> > ÂÂÂÂÂÂÂÂ spin_unlock_irqrestore(&hsotg->lock, flags);
> >
> > ÂÂÂÂÂÂÂÂ if (!already &&
> > ÂÂÂÂÂÂÂÂÂÂÂÂ role != USB_ROLE_NONE && hsotg->dr_mode == USB_DR_MODE_OTG)
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ /* This will raise a Connector ID Status Change Interrupt */
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ dwc2_force_mode(hsotg, role == USB_ROLE_HOST);
> >
> > ÂÂÂÂÂÂÂÂ dev_dbg(hsotg->dev, "%s-session valid\n",
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ role == USB_ROLE_NONE ? "No" :
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ role == USB_ROLE_HOST ? "A" : "B");
> >
> > ÂÂÂÂÂÂÂÂ return 0;
> > }
> >
> >
> > dwc2_force_mode is called outside the spin_lock_irqsave so the kernel
> > should not complain. I've tested on my setup and the behavior seems the
> > same.
> this one is looking good - the previous kernel warnings are now gone!
> thank you very much
>
>
> Best regards,
> Martin