On Thu, Feb 19, 2009 at 01:57:24PM +0100, maximilian attems wrote:
> first of all thanks for your work on the relicensing front,
> know that it is no fun..
>
> On Thu, Feb 19, 2009 at 02:19:36AM +, Ben Hutchings wrote:
> > I wrote:
> > > No word from Sun re Cassini.
> >
> > Still no word, I've tried contacting Simon Phipps now.
> >
> > > There is a FreeBSD driver for the Kawasaki USB network chips (kaweth
> > > driver) under 4-clause BSD but the stated copyright holder for the
> > > firmware is the driver author, which is not correct. I will try
> > > contacting him.
> >
> > Bill Paul wrote:
> > > First, look here:
> > >
> > > http://www.freebsd.org/~wpaul/KLSI
> > >
> > > These are all of the files that I originally received from Kawasaki LSI
> > > when I first contacted them about writing a driver, as I originally
> > > received them. This includes the file containing the firmwre image. I
> > > was never given any firmware source. You will note that there are no
> > > copyright headers at all, and when I spoke to the Kawasaki representative
> > > on the phone the subject of copyright on the firmware image never came up.
> >
> > I wrote:
> > > No news from Tehuti, but I found firmware in OpenBSD under 4-clause BSD.
> >
> > Pinchas Ziv (CEO of Tehuti) wrote:
> > > We are satisfied with the last statement:
> > >
> > >" Permission is hereby granted for the distribution of this firmware
> > > data
> > > in hexadecimal or equivalent format, provided this copyright notice is
> > > accompanying it."
> >
> > I wrote:
> > > WhiteHEAT hardware is still avalable so there may be some mileage in
> > > contacting the manufacturer.
> >
> > I talked back and forth with them and they seemed to be happy to fix the
> > license text but I never got a final statement of what exactly it would
> > be.
> >
> > I think we should move ahead with removal of sourceless firmware, but
> > include everything we have clear permission to distribute in the
> > non-free section. This would exclude the 3 blobs mentioned above and
> > several others that are already stripped from Debian kernels.
>
> ack.
>
> > I propose that future versions of firmware-nonfree, or a second firmware
> > "source" package, should be based on the firmware directory of Linux
> > releases, using a script to exclude the files with unclear license
> > status. The contents of the firmware directory should be excluded from
> > the linux-2.6 package except where source is available.
>
> wouldn't vote for yet another package,
> can we integrate them in firmware-nonfree?
>
> > Of the 13 sourceless firmware blobs I was aware of in the lenny kernel,
> > several have subsequently been moved to the firmware directory:
> >
> > - 2.6.27: dabusb, dsp56k (with source), kaweth, whiteheat
> > - 2.6.28: cassini
> > - 2.6.29: e100, starfire
> > - linux-next: qla1280
>
> right.
>
> > The remaining firmware - found in mga, r128, radeon, tehuti, typhoon -
> > is at least clearly redistributable.
>
> hmm bnx2 and bnx2x seem still not yet using request_firmware(),
> waldi should fire off that patch upstream.
John Wright has a bnx2x patch in #509646 & #509647.
--
dann frazier
--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org