On 07/10/2013 04:35 PM, Felipe Balbi wrote: > Hi, > > On Wed, Jul 10, 2013 at 04:11:52PM +0300, Roger Quadros wrote: >> On 07/10/2013 02:42 PM, Felipe Balbi wrote: >>> On Tue, Jul 09, 2013 at 02:16:32PM +0300, Roger Quadros wrote: >>>> Hi, >>>> >>>> On 07/06/2013 03:52 PM, Ruchika Kharwar wrote: >>>>> This patch adapts the dwc3 to use the device tree helper >>>>> "of_usb_get_dr_mode" for the mode of operation of the dwc3 instance >>>>> being probed. >>>>> >>>>> Signed-off-by: Ruchika Kharwar <ruchika@xxxxxx> >>>>> --- >>>>> drivers/usb/dwc3/core.c | 51 +++++++++++++++++++++++++---------------------- >>>>> drivers/usb/dwc3/core.h | 5 ----- >>>>> 2 files changed, 27 insertions(+), 29 deletions(-) >>>>> >>>>> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c >>>>> index c35d49d..7b98e4f 100644 >>>>> --- a/drivers/usb/dwc3/core.c >>>>> +++ b/drivers/usb/dwc3/core.c >>>>> @@ -517,14 +517,17 @@ static int dwc3_probe(struct platform_device *pdev) >>>>> } >>>>> >>>>> if (IS_ENABLED(CONFIG_USB_DWC3_HOST)) >>>>> - mode = DWC3_MODE_HOST; >>>>> + mode = USB_DR_MODE_HOST; >>>>> else if (IS_ENABLED(CONFIG_USB_DWC3_GADGET)) >>>>> - mode = DWC3_MODE_DEVICE; >>>>> + mode = USB_DR_MODE_PERIPHERAL; >>>>> else >>>>> - mode = DWC3_MODE_DRD; >>>>> + mode = of_usb_get_dr_mode(node); >>>> >>>> You need to check if "node" is not NULL before using >>>> of_usb_get_dr_mode. >>>> >>>> Also what would happen if DT passes a mode which can't be supported >>>> due to missing device driver? e.g. DT passes mode = "host" whereas >>>> HOST is not enabled. >>> >>> hmm... look closely, she's already handling that, right ? If DWC3 >>> Host-only, then mode is hardcoded to host, if DWC3 is Gadget-only, then >>> mode is hardcoded to peripheral and if DWC3 is DRD, then she checks >>> DeviceTree. >>> >> >> In the above example if DT passes mode = "host" but >> CONFIG_USB_DWC3_GADGET is enabled in config, the dwc3 driver will work >> in gadget only mode. >> There is no warning/information to the user that it can't be enabled >> in host mode. >> >> All I'm saying is that there must be some kind of indication about the >> requested mode not being available because of a configuration issue. > > oh, alright. But that's not part of $subject ;-) Agreed. Warning the user can be a different patch. > > So there are just two fixes which needs to be made to this patch: > > 1) call of_usb_get_dr_mode() conditionally on dev->of_node > 2) make sure platform_data can be used to pass the mode perfect :) cheers, -roger -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
