Re: [linux-usb-devel] [patch 2.6.0-test6] usbcore and driver model(0/3)

2003-10-11 Thread Oliver Neukum
Am Freitag, 10. Oktober 2003 17:00 schrieb David Brownell: Oliver Neukum wrote: So this is a feedback please. Essentially all of the changes are consequences of the first one, to remove usb-internal state duplicating driver model state. Looking very cool. You are replacing

Re: [linux-usb-devel] [patch 2.6.0-test6] usbcore and driver model (0/3)

2003-10-10 Thread Oliver Neukum
So this is a feedback please. Essentially all of the changes are consequences of the first one, to remove usb-internal state duplicating driver model state. Looking very cool. You are replacing driver-serialize with driver core's bus rwsem. It seems to me that this is a per bus lock. Is that

Re: [linux-usb-devel] [patch 2.6.0-test6] usbcore and driver model (0/3)

2003-10-10 Thread David Brownell
Oliver Neukum wrote: So this is a feedback please. Essentially all of the changes are consequences of the first one, to remove usb-internal state duplicating driver model state. Looking very cool. You are replacing driver-serialize with driver core's bus rwsem. It seems to me that this is a per

[linux-usb-devel] [patch 2.6.0-test6] usbcore and driver model (0/3)

2003-10-07 Thread David Brownell
Following this message are three related patches: - Use interface.dev.driver instead of interface.driver, updating usbcore to use device_{bind,release}_driver. - Percolate that change through usbfs. - Updates usbnet accordingly ... some other drivers may need corresponding