2005-02-10 23:56:11

by Adrian Bunk

[permalink] [raw]
Subject: DVB at76c651.c driver seems to be dead code

I didn't find any way how the drivers/media/dvb/frontends/at76c651.c
driver would do anything inside kernel 2.6.11-rc3-mm2. All it does is to
EXPORT_SYMBOL a function at76c651_attach that isn't used anywhere.

Is a patch to remove this driver OK or did I miss anything?

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed


2005-02-11 01:11:22

by Mws

[permalink] [raw]
Subject: Re: DVB at76c651.c driver seems to be dead code

On Friday 11 February 2005 00:56, Adrian Bunk wrote:
> I didn't find any way how the drivers/media/dvb/frontends/at76c651.c
> driver would do anything inside kernel 2.6.11-rc3-mm2. All it does is to
> EXPORT_SYMBOL a function at76c651_attach that isn't used anywhere.
>
> Is a patch to remove this driver OK or did I miss anything?
>
> cu
> Adrian
>
FYI

The atmel at76c651 frontend driver is used for the
Sagem DBox2 Digital Cable Receiver.

As all other parts of the dbox2 drivers are atm not hosted at kernel cvs but at
cvs.tuxbox.org you won't find any components in mainline kernel tree using this.

thus we are a hobby project - but even well known - there are not so many developer
available to make every kernel driver and other parts of it "kernel-style-alike".
maybe there is more progress and kernel driver patching into mainline in the future.
we are having 2.6.9 running on dbox2 - higher versions are atm broken for support of
the mpc 823 architecture :/

removing this driver is not wanted.

regards
Marcel Siegert
[email protected]




Attachments:
(No filename) (1.03 kB)
(No filename) (189.00 B)
Download all attachments

2005-02-11 09:36:06

by Holger Waechtler

[permalink] [raw]
Subject: Re: [linux-dvb-maintainer] DVB at76c651.c driver seems to be dead code

Adrian Bunk wrote:

>I didn't find any way how the drivers/media/dvb/frontends/at76c651.c
>driver would do anything inside kernel 2.6.11-rc3-mm2. All it does is to
>EXPORT_SYMBOL a function at76c651_attach that isn't used anywhere.
>
>Is a patch to remove this driver OK or did I miss anything?
>
>

no, please let it there. This driver is the GPL'd part of the dbox2
driver which is not part of the official kernel tree.

Since frontend and demod drivers are reusable elsewhere and mainstream
hardware that makes use of this demodulator may show up every week it's
just stupid to remove this code as long we know it is working and
continously tested by the dbox2 folks.

Instead it may make sense to move the dbox2 sources into the mainstream
source tree. Andreas, what do you think?

Holger

2005-02-11 09:49:50

by Adrian Bunk

[permalink] [raw]
Subject: Re: DVB at76c651.c driver seems to be dead code

On Fri, Feb 11, 2005 at 02:11:17AM +0100, Mws wrote:

> FYI
>
> The atmel at76c651 frontend driver is used for the
> Sagem DBox2 Digital Cable Receiver.
>
> As all other parts of the dbox2 drivers are atm not hosted at kernel cvs but at
> cvs.tuxbox.org you won't find any components in mainline kernel tree using this.
>
> thus we are a hobby project - but even well known - there are not so many developer
> available to make every kernel driver and other parts of it "kernel-style-alike".
> maybe there is more progress and kernel driver patching into mainline in the future.
> we are having 2.6.9 running on dbox2 - higher versions are atm broken for support of
> the mpc 823 architecture :/
>
> removing this driver is not wanted.

If I understand it correctly, there are several drivers that only make
sense if they are used together. at76c651.c alone makes zero sense?
This means it would be highly appreciated to have all parts inside the
kernel at some time in the future.

Something different:
The atmel at76c651 frontend driver is specific to the MPC823
architecture?
In this case this dependency should be expressed in the Kconfig file
since it then makes no sense to offer this driver on other
architectures.

> regards
> Marcel Siegert

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed

2005-02-11 12:09:52

by Andreas Oberritter

[permalink] [raw]
Subject: Re: [linux-dvb-maintainer] DVB at76c651.c driver seems to be dead code

On Fri, 2005-02-11 at 10:36 +0100, Holger Waechtler wrote:
> Adrian Bunk wrote:
>
> >I didn't find any way how the drivers/media/dvb/frontends/at76c651.c
> >driver would do anything inside kernel 2.6.11-rc3-mm2. All it does is to
> >EXPORT_SYMBOL a function at76c651_attach that isn't used anywhere.
> >
> >Is a patch to remove this driver OK or did I miss anything?
> >
> >
>
> no, please let it there. This driver is the GPL'd part of the dbox2
> driver which is not part of the official kernel tree.

(Actually all dbox2 drivers are GPL-licensed, you can get them at
cvs.tuxbox.org)

> Since frontend and demod drivers are reusable elsewhere and mainstream
> hardware that makes use of this demodulator may show up every week it's
> just stupid to remove this code as long we know it is working and
> continously tested by the dbox2 folks.
>
> Instead it may make sense to move the dbox2 sources into the mainstream
> source tree. Andreas, what do you think?

It has been a long term goal since months, but I don't have the time for
it now. We are still waiting for mpc8xx to become stable in 2.6.

Regards,
Andreas

2005-02-11 12:22:10

by Andreas Oberritter

[permalink] [raw]
Subject: Re: DVB at76c651.c driver seems to be dead code

On Fri, 2005-02-11 at 10:49 +0100, Adrian Bunk wrote:
> If I understand it correctly, there are several drivers that only make
> sense if they are used together. at76c651.c alone makes zero sense?
> This means it would be highly appreciated to have all parts inside the
> kernel at some time in the future.

It makes sense if
- the dbox2 core code gets merged into mainline, which is our goal, but
can take a huge amount of time.
- someone rips off the frontend module of a dbox2 and puts it on his
PCI DVB card because this Atmel chip rocks so much :-)
It will require only very few changes to the PCI driver...
- or a company decides to use this chip on their brand new DVB-C device
and john doe decides to write a Linux driver for it. He will then
notice that there is already a driver for the frontend module and can
therefore save a lot of work.

> Something different:
> The atmel at76c651 frontend driver is specific to the MPC823
> architecture?

no.

Regards,
Andreas

2006-01-06 00:35:45

by Adrian Bunk

[permalink] [raw]
Subject: Re: [linux-dvb-maintainer] DVB at76c651.c driver seems to be dead code

On Fri, Feb 11, 2005 at 01:11:08PM +0100, Andreas Oberritter wrote:
> On Fri, 2005-02-11 at 10:36 +0100, Holger Waechtler wrote:
> > Adrian Bunk wrote:
> >
> > >I didn't find any way how the drivers/media/dvb/frontends/at76c651.c
> > >driver would do anything inside kernel 2.6.11-rc3-mm2. All it does is to
> > >EXPORT_SYMBOL a function at76c651_attach that isn't used anywhere.
> > >
> > >Is a patch to remove this driver OK or did I miss anything?
> > >
> > >
> >
> > no, please let it there. This driver is the GPL'd part of the dbox2
> > driver which is not part of the official kernel tree.
>
> (Actually all dbox2 drivers are GPL-licensed, you can get them at
> cvs.tuxbox.org)
>
> > Since frontend and demod drivers are reusable elsewhere and mainstream
> > hardware that makes use of this demodulator may show up every week it's
> > just stupid to remove this code as long we know it is working and
> > continously tested by the dbox2 folks.
> >
> > Instead it may make sense to move the dbox2 sources into the mainstream
> > source tree. Andreas, what do you think?
>
> It has been a long term goal since months, but I don't have the time for
> it now. We are still waiting for mpc8xx to become stable in 2.6.

Are thre any news regarding merging code actually using at76c651.c?

> Regards,
> Andreas

cu
Adrian

--

"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon Seed