Re: [RFC PATCH 0/2] dwc2/s3c-hsotg: Initial steps to combine the 2 driver

[Dinh Nguyen <dinguyen@altera.com>]

On Tue, 2014-01-14 at 06:21 -0800, Greg KH wrote:
> On Tue, Jan 14, 2014 at 05:01:00AM -0600, dinguyen@altera.com wrote:
> > From: Dinh Nguyen <dinguyen@altera.com>
> > 
> > Hi,
> > 
> > I'm starting work downstream on combining the DWC2 host driver and the s3c-hsotg
> > gadget driver into a dual-role OTG driver. Before I go further, I was hoping to
> > solicit comments on whether or not my initial approach is correct? I know there
> > are plans to combine the 2, so would like to solicit comments/suggestions so
> > that I can also upstream it as well.
> > 
> > These 2 patches:
> > 
> > * Moves the DWC2 driver out of drivers/staging into drivers/usb/dwc2/
> 
> This already happened yesterday in my tree, so you should see this in
> linux-next by now, no need to do it again :)
> 

I see it now. Thanks for the pointer.

> > * Moves the s3c-hsotg driver into drivers/usb/dwc2/
> > * Delete s3c-hsotg.h 
> > * Make the s3c-hsotg.c file use the defines in hw.h from the DWC2 driver.
> > 
> > This initial patch has been tested on the SOCFPGA platform only in Host-only
> > and Gadget-only mode.
> > 
> > The next step would be to do the combining of the driver into a dual-role OTG
> > driver.
> 
> I was told that merging the two of these isn't going to work as the
> silicon is just too different, which is why I allowed the code to move
> out of staging.  If you feel differently, and think you can combine the
> two drivers, that's wonderful, I'll gladly take patches to do so, but be
> sure to test on the proper platforms to make sure nothing breaks.
> 

I wasn't aware of the silicon differences. I just took the s3c-hsotg
driver as is and it worked fine on my version 2.93a of the USB IP. I'll
search the ML for information, or perhaps Paul can comment?

Thanks,
Dinh
> thanks,
> 
> greg k-h
>