Re: [PATCH] usb: dwc3: Addition of "dr_mode" dt property.

[Ruchika Kharwar <ruchika@ti.com>]

On 05/30/2013 03:35 PM, Dan Murphy wrote:
> Fix spelling in my own comments
>
> On 05/30/2013 03:31 PM, Dan Murphy wrote:
>> On 05/30/2013 03:14 PM, Ruchika Kharwar wrote:
>>> This patch adds an optional parameter "dr_mode" to the dwc3 core device node.
>>> In the case the compile flag for the DWC3 controller is set to
>>> "USB_DWC3_DUAL_ROLE" a device tree could restrain to either functionality of
>>> host or gadget. In the case the device tree does not use this optional flag or
>>> specifies it superfluously to "drd" the functionality will be that
>>> of a dual role device.
>>>
>>> Signed-off-by: Ruchika Kharwar <ruchika@ti.com>
>>> ---
>> Can you add patch history if this is truly a new patch to a previous submission
         Will do in the next pach submitted.
>>>   Documentation/devicetree/bindings/usb/dwc3.txt |    3 ++-
>>>   drivers/usb/dwc3/core.c                        |   21 +++++++++++++++++----
>>>   2 files changed, 19 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/Documentation/devicetree/bindings/usb/dwc3.txt b/Documentation/devicetree/bindings/usb/dwc3.txt
>>> index 7a95c65..2f5d584 100644
>>> --- a/Documentation/devicetree/bindings/usb/dwc3.txt
>>> +++ b/Documentation/devicetree/bindings/usb/dwc3.txt
>>> @@ -10,7 +10,8 @@ Required properties:
>>>   
>>>   Optional properties:
>>>    - tx-fifo-resize: determines if the FIFO *has* to be reallocated.
>>> -
>>> + - dr_mode: determines the mode of core. Supported modes are "gadget", "host"
>>> +   and "drd".
>> My previous comments still stand for this section.  Nothing was addressed or commented back
>>
>> https://patchwork.kernel.org/patch/2638511/
>>>   This is usually a subnode to DWC3 glue to which it is connected.
>>>   
>>>   dwc3@4a030000 {
>>> diff --git a/drivers/usb/dwc3/core.c b/drivers/usb/dwc3/core.c
>>> index c35d49d..e11660a 100644
>>> --- a/drivers/usb/dwc3/core.c
>>> +++ b/drivers/usb/dwc3/core.c
>>> @@ -378,7 +378,7 @@ static int dwc3_probe(struct platform_device *pdev)
>>>   	void			*mem;
>>>   
>>>   	u8			mode;
>>> -
>>> +	char			*dr_mode;
>>>   	mem = devm_kzalloc(dev, sizeof(*dwc) + DWC3_ALIGN_MASK, GFP_KERNEL);
>>>   	if (!mem) {
>>>   		dev_err(dev, "not enough memory\n");
>>> @@ -520,9 +520,22 @@ static int dwc3_probe(struct platform_device *pdev)
>>>   		mode = DWC3_MODE_HOST;
>>>   	else if (IS_ENABLED(CONFIG_USB_DWC3_GADGET))
>>>   		mode = DWC3_MODE_DEVICE;
>>> -	else
>>> -		mode = DWC3_MODE_DRD;
>>> -
>>> +	else {
>>> +		if (of_property_read_string(node, "dr_mode", &dr_mode))
>>> +			mode = DWC3_MODE_DRD;
>>> +		else {
>>> +			if (strcmp(dr_mode, "host") == 0)
>>> +				mode = DWC3_MODE_HOST;
>>> +			else if (strcmp(dr_mode, "gadget") == 0)
>>> +				mode = DWC3_MODE_DEVICE;
>>> +			else if (strcmp(dr_mode, "drd") == 0)
>>> +				mode = DWC3_MODE_DRD;
>> My previous comments still stand for this section.  Nothing was addressed or commented back
>>
>> https://patchwork.kernel.org/patch/2638511/
         DRD mode is the super set mode for Host and Gadget. The compile 
time flag is primary. If DRD is set, the host and gadget code is 
compiled in.
         The dr_mode property can be used *if* specified in the device 
tree. A specific example may be a board could have constraints where the 
port could only be a host
         or gadget. In that case, the dr_mode property is specified 
overrides the compile flag.
>>> +			else {
>>> +				dev_err(dev, "invalid dr_mode property value\n");
>>> +				goto err2;
>> This is still wrong.  You have not initialized and of the device's
> s/and/any
        This is correct as per 3.10-rc3. If you're looking at an older 
tree you'd be right.
>>> +			}
>>> +		}
>>> +	}
>>>   	switch (mode) {
>>>   	case DWC3_MODE_DEVICE:
>>>   		dwc3_set_mode(dwc, DWC3_GCTL_PRTCAP_DEVICE);
>> Is this a new patch?
>> Subject should say v2.
Yes, my bad.. learning.. next patch will have it.
>>
>