On Mon, Feb 04, 2008 at 06:08:51PM +0100, Michael Buesch wrote: > On Monday 04 February 2008 18:01:31 Greg KH wrote: > > > Firmware upload -> Not sure. Do we already have some generic device > > > firmware > > > update API/subsystem? > > > > Yes, we already have a firmware interface within the kernel, that's been > > standard for a number of years now. See include/linux/firmware.h > > Yeah, well. But this is different. > This interface is for devices that need to load a firmware on initialization. > This mouse has the fw flashed onto the device. So the driver doesn't need > it on init. We rather need an interface here, over which the user can reflash > this firmware. So no request_firmware() interface is needed, which the driver > has to call. So this is the other way around. The user will write the fw to > the > hardware and it will permanently be stored.
No, I think it's the same interface. Systems use that interface to update the BIOS in the machine, which is the same type of need here. When a user wants to update the firmware, they just write it to the file, no need to do it all the time. There is no requirement that the device only ask for the firmware in order to initialize itself :) Just use the request_firmware_nowait(), and if you get called back, then update the firmware. Actually everyone should be using the _nowait() call, but that's another different topic... thanks, greg k-h - To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html