[EMAIL PROTECTED] wrote:
>
> On Fri, 11 May 2001, Greg KH wrote:
<snip>
> > Finally! A device that uses DFU in the wild! About time... :)
Don't get carried away just yet. It might be some vendor variation on the
spec. I have to check the vendor docs and the DFU spec yet :)
> > Yes, we should probably support this, but probably through libusb, not a
> > kernel driver.
Yeah, probably. I was think along the kernel lines, but the functionality
won't be very general, and we should keep the kernel clean.
> Actually, it needs to be in a kernel driver. This device needs the driver
> to download firmware, then get it talking to the 802.11b network.
> Basically, it's download firmware, remap, download more firmware, and
> startup the device itself.
This can be done in user space, since we can tell configuration from a vendor
specific control message. If the configuration is wrong, we let the user space
tool load the firmware (and then we force a disconnection, also from user
space), and then the kernel driver will claim it.
> I think this is basically what the ezusb drivers do. Except I have to do
> it twice.
We have redistribution rights for most of the ezusb code (although the rights
vary). However after the last flamewar (lets not do that again, huh), it was
decided that the ezusb code would be moved to a user space tool.
Think about how you manage this problem if Atmel won't allow free
redistribution of the firmware image. How do you get it in the kernel?
Brad
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
http://lists.sourceforge.net/lists/listinfo/linux-usb-devel
