On Mon, 25 Jul 2016, Oliver Neukum wrote: > On Fri, 2016-07-22 at 09:59 -0400, Alan Stern wrote: > > On Fri, 22 Jul 2016, Oliver Neukum wrote: > > > > > It is always possible to go back to the usb_device if you know > > > the interface. It is unnecessary to store the device and thus should > > > not be done. > > > > That is not a hard-and-fast law -- the kernel isn't based on Codd's > > rules for relational databases. This patch eliminates one pointer from > > one data structure, while adding 8 extra nontrivial computations to the > > code. The end result is not obviously an improvement. > > OK, let's drop it. But it seems to me that something is wrong with > our APIs if you always need the parent device although the driver > is intended for a leaf interface. It isn't _needed_ -- it's an optimization. Alan Stern -- 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
