legacy drivers in staging?

legacy drivers in staging?

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 438 bytes --]

Hi Jean,

when having another look at the status page of legacy drivers in the i2c-wiki,
I stumbled over the drivers from 'staging'. Do they really need to appear
there? I hope the removal of the legacy binding will not depend on those
drivers ;)

Regards,

   Wolfram

-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 197 bytes --]

Re: legacy drivers in staging?

[Jean Delvare]

Hi Wolfram,

On Sun, 1 Mar 2009 11:13:06 +0100, Wolfram Sang wrote:
> when having another look at the status page of legacy drivers in the i2c-wiki,
> I stumbled over the drivers from 'staging'. Do they really need to appear
> there? I hope the removal of the legacy binding will not depend on those
> drivers ;)

I am not sure myself. I agree we don't want to delay the removal of the
legacy binding just because of these drivers. But OTOH I see little
point in having a staging directory [1] if drivers being placed there
break as soon as there is a core kernel change. I guess we are supposed
to update these drivers as we do for all other drivers.

So, ideally, someone would convert these drivers so that they do not
break when legacy drivers are removed. My hope was that the author of
these drivers would do it when seeing the warnings which i2c.h spits
out. But if this doesn't happen, someone else will have to take care.
That being said I think that there is much more work needed on these
drivers, on the V4L side.

Bottom line is that, is these drivers are the last ones blocking the
removal of the legacy i2c driver, I will either attempt to convert them
the quick-and-dirty way, or I will mark them as broken and leave it to
somebody else to fix them. They will not delay us.

[1] To be honest I was not really fond of the idea of a staging drivers
directory in the first place anyway.

-- 
Jean Delvare

Re: legacy drivers in staging?

[Wolfram Sang]

[-- Attachment #1: Type: text/plain, Size: 1633 bytes --]

Hi Jean,

> I am not sure myself. I agree we don't want to delay the removal of the
> legacy binding just because of these drivers. But OTOH I see little
> point in having a staging directory [1] if drivers being placed there
> break as soon as there is a core kernel change. I guess we are supposed
> to update these drivers as we do for all other drivers.

My view is a bit like the comment found in staging/rt28[67]0/TODO:

===

Please send any patches or complaints about this driver to Greg
Kroah-Hartman <greg-U8xfFu+wG4EAvxtiuMwx3w@public.gmane.org> and don't bother the upstream wireless
kernel developers about it, they want nothing to do with it.

===

> That being said I think that there is much more work needed on these
> drivers, on the V4L side.

As some of these drivers are not just ugly but also conceptually wrong
(epl for example), I would vote for not accepting any duties which come
from them. Especially when it is a long awaited task like the removal of
the legacy binding.

> Bottom line is that, is these drivers are the last ones blocking the
> removal of the legacy i2c driver, I will either attempt to convert them
> the quick-and-dirty way, or I will mark them as broken and leave it to
> somebody else to fix them. They will not delay us.

That is nice to hear. I don't think you need to mark them as broken,
staging is per se broken :) And please don't spend time on fixing them ;)

Regards,

   Wolfram

-- 
Pengutronix e.K.                           | Wolfram Sang                |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 197 bytes --]