On 08/31/2012 11:29 AM, Andrzej Pietrasiewicz wrote:
Thank you for asking for my opinion. I would like to say that we need ccg in the tree. It seems that the big change with converting to configfs will not happen within the next two kernel releases, and we still don't know the new framework will look like and work - my first patch with "USB gadget - configfs" https://lkml.org/lkml/2012/6/21/154 contained only mass storage, and there are over a dozen functions or so.
And you realize that ccg remains unused right now from its biggest and probably only user called Android? So why should we keep it?
It will take time before things are fully settled.
Yes and this Nokia gadget fork makes things complicated for no good reason as far as I tell. I had to provide a copy of composite.c because some things were beyond fixable. I happy with fixing 15 gadgets in tree including nokia.c or multi.c since those have active users.
So I suggest we try making ccg in sync with your latest changes and reconsider keeping/removing it after the full process of gadget's transition to configfs is completed.
I reshuffled some code already I have to do more. If you still insist on keeping ccg I hope you have a good reason for it _and_ don't mind when I copy all included .c files into staging as well? btw: I've been going over the staging mailing list and looked for ccg patches. Besides the usual noise I just noticed akpm somewhere close to a heart attack. So this is parked there until we have something configfs based. Staging was meant for drivers which have to be improved and more people can work on it. This is not the case here because here we need a replacement with different API.
Thank you, Andrzej
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
