[PATCH v4 3/4] USB: Gadget: Add Samsung S3C24XX USB High-Speed controller driver

[Greg KH]

On Sat, Apr 16, 2011 at 12:00:31PM +0200, Heiko Stübner wrote:
> Am Donnerstag 14 April 2011, 19:15:23 schrieb Alan Stern:
> > On Thu, 14 Apr 2011, Greg KH wrote:
> > > On Thu, Apr 14, 2011 at 11:35:43AM -0400, Alan Stern wrote:
> > > > On Thu, 14 Apr 2011, Heiko [iso-8859-1] St?bner wrote:
> > > > > From: Thomas Abraham <thomas.ab at samsung.com>
> > > > > 
> > > > > The Samsung's S3C2416, S3C2443 and S3C2450 includes a USB High-Speed
> > > > > device controller module. This driver enables support for USB
> > > > > high-speed gadget functionality for the Samsung S3C24xx SoC's that
> > > > > include this controller.
> > > > > 
> > > > > Signed-off-by: Thomas Abraham <thomas.ab at samsung.com>
> > > > > Signed-off-by: Sangbeom Kim <sbkim73 at samsung.com>
> > > > > Signed-off-by: Kukjin Kim <kgene.kim at samsung.com>
> > > > > Signed-off-by: Alexander Neumann <alexander at bumpern.de>
> > > > > Signed-off-by: Heiko Stuebner <heiko at sntech.de>
> > > > 
> > > > ...
> > > > 
> > > > > +static struct usb_ep_ops s3c_hsudc_ep_ops = {
> > > > > +	.enable = s3c_hsudc_ep_enable,
> > > > > +	.disable = s3c_hsudc_ep_disable,
> > > > > +	.alloc_request = s3c_hsudc_alloc_request,
> > > > > +	.free_request = s3c_hsudc_free_request,
> > > > > +	.queue = s3c_hsudc_queue,
> > > > > +	.dequeue = s3c_hsudc_dequeue,
> > > > > +	.set_halt = s3c_hsudc_set_halt,
> > > > > +};
> > > > 
> > > > There's no .set_wedge method.  Why do people always leave this out?
> > > 
> > > Does the code spit out a nasty warning if this isn't set?  If not, I
> > > would suggest adding it so that this doesn't keep happening.
> > > 
> > > Or just refuse to be able to register the structure, that would stop it
> > > right away :)
> > 
> > In fact, set_wedge is optional.  But it's so easy to implement, there's
> > no good reason for leaving it out.
> 
> It seems Thomas [original author of the driver] will be able to implement said 
> set_wedge function for it.
> As he will need a bit of time for this, two possible ways for going forward 
> come to mind:
> (1) use current driver [as set_wedge is optional] and add it later via patch
> (2) resubmit whole driver again when set_wedge is added to it
> 
> Obviously I would prefer option 1 :-), but in the end it's your decision.

It shouldn't take that much time to do this, what is the delay?

I'd prefer to get the correct version implemented and would not like to
accept a patch that everyone knows is wrong.

thanks,

greg k-h