Re: non-free firmware in the linux kernel
On Tue, Jan 10, 2006 at 05:19:47PM +0100, Andreas Barth wrote:
> * Kyle McMartin (firstname.lastname@example.org) [060110 16:20]:
> > I would argue it's the former. I can see the argument when it's a part of
> > the source code, but not when it's a completely seperate entity.
> Sorry, but there is no difference regarding DFSG: If the binary blob is
> actually seperated from the kernel implementation, it can be treated as
> binary blob (which I personally consider as non-program - but see
how can you consider it as non-program. It is indeed composed of machine code
destined to run on the controller of the device the driver is written for.
Micro-code being another synonym for the concerned firmware in some cases. So
how could anyone honestly claim it is a non-program is beyond me :) I know we
all wish it where so, but ...
> Steve's mail for details). Whether this binary blob is embedded in the
> kernel source code (like we have IIRC for the pinguin boot logo) or not
> (or is linked to the kernel during link time) doesn't make any
> difference to this. It might however make an difference to
> GPL-compatibility, unless the license is GPL-compatible anyways.
Nope, please read my posts on debian-legal about this topic 6 month or so ago.
The firmware is destined to run on another processor, and thus there is no way
you can claim it is a derived work from the actual linux driver, since the
communication channel is clearly delimited. The counter-example would make the
linux kernel a derivative of any expansion card or motherboard with embedded
bios, or even make it a derivative of the actual hardware.
So, no, there is no GPL issue there, and things are clearly separate.