* Alan Stern | 2011-05-13 13:52:36 [-0400]:
>> struct usb_gadget;
>> +struct usb_gadget_driver;
>> +typedef int (*usb_gad_bind_t)(struct usb_gadget *gadget);
>
>Ugh! Please don't do this. Write out the full function type at the
>place where it is used, just like all the other function pointer
>declarations do.
I tried to save here key strokes and since it is not a complex (like a
struct) it is not a big deal. And we do it in other places in kernel.
But if you prefer it written out, I will change it.
>> /* the rest of the api to the controller hardware: device operations,
>> * which don't involve endpoints (or i/o).
>> @@ -430,6 +432,8 @@ struct usb_gadget_ops {
>> int (*pullup) (struct usb_gadget *, int is_on);
>> int (*ioctl)(struct usb_gadget *,
>> unsigned code, unsigned long param);
>> + int (*start)(struct usb_gadget *, struct usb_gadget_driver *);
>> + int (*stop)(struct usb_gadget *, struct usb_gadget_driver *);
>> };
>>
>> /**
>> @@ -769,6 +773,7 @@ static inline int usb_gadget_disconnect(struct usb_gadget *gadget)
>> struct usb_gadget_driver {
>> char *function;
>> enum usb_device_speed speed;
>> + usb_gad_bind_t bind;
>> void (*unbind)(struct usb_gadget *);
>> int (*setup)(struct usb_gadget *,
>> const struct usb_ctrlrequest *);
>
>After all, why should bind be different from unbind or setup?
Yeah that is messy. I did not want to change the prototype for setup
which has nothing to do with this patch.
>Also, there should be kerneldoc for the new bind method in the comments
>preceding the definition of struct usb_gadget_driver.
Okay, I will add it.
>
>Alan Stern
>
Sebastian
--
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
