[RFC/PATCH 01/32] usb: gadget: udc-core: allow udc class register gadget device

Tomasz Figa tomasz.figa at gmail.com
Thu Jan 24 13:57:46 EST 2013


Hi Felipe,

On Thursday 24 of January 2013 17:45:39 Felipe Balbi wrote:
> Currently all UDC drivers are calling
> device_register() before calling
> usb_add_gadget_udc(). In order to avoid
> code duplication, we can allow udc-core.c
> register that device.
> 
> However that would become a really large patch,
> so to cope with the meanwhile and allow us
> to write bite-sized patches, we're adding
> a flag which will be set by UDC driver once
> it removes the code for registering the
> gadget device.
> 
> Once all are converted, the new flag will
> be removed.
> 
> Signed-off-by: Felipe Balbi <balbi at ti.com>
> ---
>  drivers/usb/gadget/udc-core.c | 23 +++++++++++++++++++----
>  include/linux/usb/gadget.h    |  4 ++++
>  2 files changed, 23 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/usb/gadget/udc-core.c
> b/drivers/usb/gadget/udc-core.c index 2a9cd36..9195054 100644
> --- a/drivers/usb/gadget/udc-core.c
> +++ b/drivers/usb/gadget/udc-core.c
> @@ -173,6 +173,14 @@ int usb_add_gadget_udc(struct device *parent,
> struct usb_gadget *gadget) if (!udc)
>  		goto err1;
> 
> +	if (gadget->register_my_device) {
> +		dev_set_name(&gadget->dev, "gadget");
> +
> +		ret = device_register(&gadget->dev);
> +		if (ret)
> +			goto err2;
> +	}
> +
>  	device_initialize(&udc->dev);
>  	udc->dev.release = usb_udc_release;
>  	udc->dev.class = udc_class;
> @@ -180,7 +188,7 @@ int usb_add_gadget_udc(struct device *parent, struct
> usb_gadget *gadget) udc->dev.parent = parent;
>  	ret = dev_set_name(&udc->dev, "%s", kobject_name(&parent->kobj));
>  	if (ret)
> -		goto err2;
> +		goto err3;

Just a nitpick: If you are at changing labels of error path, wouldn't it 
be better to give them some meaningful names? Like err_del_unlock, 
err_put_udc, err_put_gadget, err_ret.

Otherwise looks good. Nice idea.

Reviewed-by: Tomasz Figa <tomasz.figa at gmail.com>

Best regards,
Tomasz Figa

>  	udc->gadget = gadget;
> 
> @@ -189,18 +197,22 @@ int usb_add_gadget_udc(struct device *parent,
> struct usb_gadget *gadget)
> 
>  	ret = device_add(&udc->dev);
>  	if (ret)
> -		goto err3;
> +		goto err4;
> 
>  	mutex_unlock(&udc_lock);
> 
>  	return 0;
> -err3:
> +
> +err4:
>  	list_del(&udc->list);
>  	mutex_unlock(&udc_lock);
> 
> -err2:
> +err3:
>  	put_device(&udc->dev);
> 
> +err2:
> +	if (gadget->register_my_device)
> +		put_device(&gadget->dev);
>  err1:
>  	return ret;
>  }
> @@ -254,6 +266,9 @@ found:
> 
>  	kobject_uevent(&udc->dev.kobj, KOBJ_REMOVE);
>  	device_unregister(&udc->dev);
> +
> +	if (gadget->register_my_device)
> +		device_unregister(&gadget->dev);
>  }
>  EXPORT_SYMBOL_GPL(usb_del_gadget_udc);
> 
> diff --git a/include/linux/usb/gadget.h b/include/linux/usb/gadget.h
> index 2e297e8..fcd9ef8 100644
> --- a/include/linux/usb/gadget.h
> +++ b/include/linux/usb/gadget.h
> @@ -494,6 +494,9 @@ struct usb_gadget_ops {
>   *	only supports HNP on a different root port.
>   * @b_hnp_enable: OTG device feature flag, indicating that the A-Host
>   *	enabled HNP support.
> + * @register_my_device: Flag telling udc-core that UDC driver didn't
> + *	register the gadget device to the driver model. Temporary until
> + *	all UDC drivers are fixed up properly.
>   * @name: Identifies the controller hardware type.  Used in diagnostics
> *	and sometimes configuration.
>   * @dev: Driver model state for this abstract device.
> @@ -531,6 +534,7 @@ struct usb_gadget {
>  	unsigned			b_hnp_enable:1;
>  	unsigned			a_hnp_support:1;
>  	unsigned			a_alt_hnp_support:1;
> +	unsigned			register_my_device:1;
>  	const char			*name;
>  	struct device			dev;
>  	unsigned			out_epnum;



More information about the linux-arm-kernel mailing list