diff mbox series

[v3] usb: host: xhci-plat: Set XHCI_SKIP_PHY_INIT quirk for DWC3 controller

Message ID 1638339338-6731-2-git-send-email-quic_c_sanm@quicinc.com
State Superseded
Headers show
Series [v3] usb: host: xhci-plat: Set XHCI_SKIP_PHY_INIT quirk for DWC3 controller | expand

Commit Message

Sandeep Maheswaram Dec. 1, 2021, 6:15 a.m. UTC
Set XHCI_SKIP_PHY_INIT quirk to avoid phy initialization twice.
Runtime suspend of phy drivers was failing from DWC3 driver as
runtime usage value is 2 because the phy is initialized from
DWC3 core and HCD core.
DWC3 manages phy in their core drivers.
Set this quirk to avoid phy initialization in HCD core.

Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
---
 drivers/usb/host/xhci-plat.c | 3 +++
 1 file changed, 3 insertions(+)

Comments

Greg KH Dec. 13, 2021, 1:57 p.m. UTC | #1
On Wed, Dec 01, 2021 at 11:45:38AM +0530, Sandeep Maheswaram wrote:
> Set XHCI_SKIP_PHY_INIT quirk to avoid phy initialization twice.
> Runtime suspend of phy drivers was failing from DWC3 driver as
> runtime usage value is 2 because the phy is initialized from
> DWC3 core and HCD core.
> DWC3 manages phy in their core drivers.
> Set this quirk to avoid phy initialization in HCD core.
> 
> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> ---
>  drivers/usb/host/xhci-plat.c | 3 +++
>  1 file changed, 3 insertions(+)

What changed from previous versions of this patch?  That goes below the
--- line always, right?

Please fix up and send a v4.

thanks,

greg k-h
Sandeep Maheswaram Dec. 14, 2021, 4:47 a.m. UTC | #2
On 12/2/2021 8:47 AM, Stephen Boyd wrote:
> Quoting Sandeep Maheswaram (2021-11-30 22:15:38)
>> Set XHCI_SKIP_PHY_INIT quirk to avoid phy initialization twice.
>> Runtime suspend of phy drivers was failing from DWC3 driver as
>> runtime usage value is 2 because the phy is initialized from
>> DWC3 core and HCD core.
>> DWC3 manages phy in their core drivers.
> This looks wrapped weirdly.
>
>> Set this quirk to avoid phy initialization in HCD core.
>>
>> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> Any Fixes tag?
Not sure about the commit id to be used here.
>
>> ---
>>   drivers/usb/host/xhci-plat.c | 3 +++
>>   1 file changed, 3 insertions(+)
>>
>> diff --git a/drivers/usb/host/xhci-plat.c b/drivers/usb/host/xhci-plat.c
>> index c1edcc9..9bbd939 100644
>> --- a/drivers/usb/host/xhci-plat.c
>> +++ b/drivers/usb/host/xhci-plat.c
>> @@ -327,6 +327,9 @@ static int xhci_plat_probe(struct platform_device *pdev)
>>                                           &xhci->imod_interval);
>>          }
>>
>> +       if (of_device_is_compatible(pdev->dev.parent->of_node, "snps,dwc3"))
> Are we sure that pdev->dev.parent isn't NULL here?

+    if (of_device_is_compatible(to_of_node(sysdev->fwnode), "snps,dwc3"))

Will this be ok ?

>
>> +               xhci->quirks |= XHCI_SKIP_PHY_INIT;
>> +
>>          hcd->usb_phy = devm_usb_get_phy_by_phandle(sysdev, "usb-phy", 0);
>>          if (IS_ERR(hcd->usb_phy)) {
Stephen Boyd Dec. 15, 2021, 1:02 a.m. UTC | #3
Quoting Sandeep Maheswaram (2021-12-13 20:47:48)
>
> On 12/2/2021 8:47 AM, Stephen Boyd wrote:
> > Quoting Sandeep Maheswaram (2021-11-30 22:15:38)
> >> Set XHCI_SKIP_PHY_INIT quirk to avoid phy initialization twice.
> >> Runtime suspend of phy drivers was failing from DWC3 driver as
> >> runtime usage value is 2 because the phy is initialized from
> >> DWC3 core and HCD core.
> >> DWC3 manages phy in their core drivers.
> > This looks wrapped weirdly.
> >
> >> Set this quirk to avoid phy initialization in HCD core.
> >>
> >> Signed-off-by: Sandeep Maheswaram <quic_c_sanm@quicinc.com>
> > Any Fixes tag?
> Not sure about the commit id to be used here.

Maybe whenever the phy started being initialized twice?

> >
> >> ---
> >>   drivers/usb/host/xhci-plat.c | 3 +++
> >>   1 file changed, 3 insertions(+)
> >>
> >> diff --git a/drivers/usb/host/xhci-plat.c b/drivers/usb/host/xhci-plat.c
> >> index c1edcc9..9bbd939 100644
> >> --- a/drivers/usb/host/xhci-plat.c
> >> +++ b/drivers/usb/host/xhci-plat.c
> >> @@ -327,6 +327,9 @@ static int xhci_plat_probe(struct platform_device *pdev)
> >>                                           &xhci->imod_interval);
> >>          }
> >>
> >> +       if (of_device_is_compatible(pdev->dev.parent->of_node, "snps,dwc3"))
> > Are we sure that pdev->dev.parent isn't NULL here?
>
> +    if (of_device_is_compatible(to_of_node(sysdev->fwnode), "snps,dwc3"))
>
> Will this be ok ?

Probably? But I see this function already uses 'sysdev->of_node' so
that's probably all that's necessary assuming sysdev == pdev->dev.parent
in this case.

	if (of_device_is_compatible(sysdev->of_node, "snps,dwc3"))

>
> >
> >> +               xhci->quirks |= XHCI_SKIP_PHY_INIT;
> >> +
> >>          hcd->usb_phy = devm_usb_get_phy_by_phandle(sysdev, "usb-phy", 0);
> >>          if (IS_ERR(hcd->usb_phy)) {
diff mbox series

Patch

diff --git a/drivers/usb/host/xhci-plat.c b/drivers/usb/host/xhci-plat.c
index c1edcc9..9bbd939 100644
--- a/drivers/usb/host/xhci-plat.c
+++ b/drivers/usb/host/xhci-plat.c
@@ -327,6 +327,9 @@  static int xhci_plat_probe(struct platform_device *pdev)
 					 &xhci->imod_interval);
 	}
 
+	if (of_device_is_compatible(pdev->dev.parent->of_node, "snps,dwc3"))
+		xhci->quirks |= XHCI_SKIP_PHY_INIT;
+
 	hcd->usb_phy = devm_usb_get_phy_by_phandle(sysdev, "usb-phy", 0);
 	if (IS_ERR(hcd->usb_phy)) {
 		ret = PTR_ERR(hcd->usb_phy);