On Mon, Mar 07, 2005 at 09:49:40PM +0000, Linux Kernel Mailing List wrote:
> ChangeSet 1.1982.132.4, 2005/03/07 13:49:40-08:00, [email protected]
>
> [PATCH] Restore PWC driver
>
> PWC has a new maintainer (Luc Saillard) and also the various contentious
> binary hooks removed and replaced with reverse engineering work.
>
> Please restore it to the kernel
Ick, Alan, couldn't you have had the decency to run this through the USB
developers, and at least pinged me on it? Especially due to all of the
hate-email I have gotten over this driver in the past.
As it is, the coding style sucks in places, and you didn't really need
to make it a new subdirectory (although due to the increased size of the
driver, it's probably better now...)
And, there's no MAINTAINERS entry for who I need to bug about this
thing.
Bleah.
So, who's going to fix up:
- the MAINTAINERS entry
- the coding style
- drop that unneeded changelog file
- fix the module help text to point to the proper file (or put
the file in the proper place.)
- get rid of the c++ crud in the header file
- drop the "magic" nonsense
- the ioctls to work on 64bit machines
?
And I found all of that in just a quick glance over the code...
Double bleah.
greg k-h
On Mon, Mar 07, 2005 at 09:26:43PM -0800, Greg KH wrote:
> So, who's going to fix up:
Add:
- the sparse warnings.
to that list.
Oh, and those sparse warnings show that this driver is now completely
broken on big-endian boxes.
So, who should I be bouncing the emails that I'm about to get from the
angry mob of ppc users/developers that always throw large, blunt objects
at me whenever I add code that breaks on their machines?
Ick, ick, ick...
greg k-h
On Mon, Mar 07, 2005 at 09:26:43PM -0800, Greg KH wrote:
> Ick, Alan, couldn't you have had the decency to run this through the USB
> developers, and at least pinged me on it? Especially due to all of the
> hate-email I have gotten over this driver in the past.
>
> As it is, the coding style sucks in places, and you didn't really need
> to make it a new subdirectory (although due to the increased size of the
> driver, it's probably better now...)
>
> And, there's no MAINTAINERS entry for who I need to bug about this
> thing.
Agree with greg on all these bits, and similar (although not as bad) for
the other bits Alan sent. Alan, could you please go through the subsystem
maintainer & -mm process like everyone else? A little public review
can't hurt your patches either.
On Mon, Mar 07, 2005 at 09:26:43PM -0800, Greg KH wrote:
> On Mon, Mar 07, 2005 at 09:49:40PM +0000, Linux Kernel Mailing List wrote:
> > ChangeSet 1.1982.132.4, 2005/03/07 13:49:40-08:00, [email protected]
> >
> > [PATCH] Restore PWC driver
> >
> > PWC has a new maintainer (Luc Saillard) and also the various contentious
> > binary hooks removed and replaced with reverse engineering work.
> >
> > Please restore it to the kernel
Hi all,
thanks for your comments, because this why i've waited for long time when i
post the first version of the driver. I didn't post another patch on the lkml
because i want to fix the v4l2 layer before another round. I want (if
possible) to remove or deprecated a lot of ioctl in favor of sysfs and v4l2
API. Some ugly kmalloc need to be remove (the last thing of the plugins
architecture) and this:
> So, who's going to fix up:
> - the MAINTAINERS entry
> - the coding style
oops (anyone have a vim syntax file for lkml indenting ?)
> - drop that unneeded changelog file
already done
> - fix the module help text to point to the proper file (or put
> the file in the proper place.)
> - get rid of the c++ crud in the header file
> - drop the "magic" nonsense
> - the ioctls to work on 64bit machines
Can you help me about this 64bits problems ? i've now a amd64 and the webcam
works fine. But perhaps i need to test with a 32 bits app?
Luc
On Mon, Mar 07, 2005 at 09:26:43PM -0800, Greg KH wrote:
> As it is, the coding style sucks in places, and you didn't really need
> to make it a new subdirectory (although due to the increased size of the
> driver, it's probably better now...)
Luc I assume
> And, there's no MAINTAINERS entry for who I need to bug about this
> thing.
Try the email address at the top of the files.. ?
> So, who's going to fix up:
> - the MAINTAINERS entry
> - the coding style
> - drop that unneeded changelog file
> - fix the module help text to point to the proper file (or put
> the file in the proper place.)
> - get rid of the c++ crud in the header file
> - drop the "magic" nonsense
> - the ioctls to work on 64bit machines
> ?
Luc and I'm happy to help doing further work on it. However it's been like that
in kernel for years so it might also be a good one for the janitors to join in
on ?
On Tue, Mar 08, 2005 at 05:34:06AM +0000, Christoph Hellwig wrote:
> Agree with greg on all these bits, and similar (although not as bad) for
> the other bits Alan sent. Alan, could you please go through the subsystem
> maintainer & -mm process like everyone else? A little public review
> can't hurt your patches either.
I sent it to Andrew the day Linus removed it, nothing happened. I got fed up
of people asking why the most used webcam driver is -ac only.
On Mon, Mar 07, 2005 at 09:32:19PM -0800, Greg KH wrote:
> So, who should I be bouncing the emails that I'm about to get from the
> angry mob of ppc users/developers that always throw large, blunt objects
> at me whenever I add code that breaks on their machines?
Relax dude. Lets have a plan and an order to things. Nobody is going to dump
it all on your lap, me included.
On Tue, Mar 08, 2005 at 09:15:08AM +0100, Luc Saillard wrote:
> > - the coding style
> oops (anyone have a vim syntax file for lkml indenting ?)
indent -kr -i8 -bri0 -l255
comes very close as a starter. If you do that don't mix it with any actual
code changes.
> > - drop that unneeded changelog file
Alan Cox <[email protected]> wrote:
>
> On Tue, Mar 08, 2005 at 05:34:06AM +0000, Christoph Hellwig wrote:
> > Agree with greg on all these bits, and similar (although not as bad) for
> > the other bits Alan sent. Alan, could you please go through the subsystem
> > maintainer & -mm process like everyone else? A little public review
> > can't hurt your patches either.
>
> I sent it to Andrew the day Linus removed it, nothing happened.
You did? I must have missed it. That's the sort of thing which I'll hang
onto for long periods of time so people know where to go to get it.
On Tue, Mar 08, 2005 at 04:59:38AM -0500, Alan Cox wrote:
> On Mon, Mar 07, 2005 at 09:26:43PM -0800, Greg KH wrote:
>
> > So, who's going to fix up:
> > - the MAINTAINERS entry
> > - the coding style
> > - drop that unneeded changelog file
> > - fix the module help text to point to the proper file (or put
> > the file in the proper place.)
> > - get rid of the c++ crud in the header file
> > - drop the "magic" nonsense
> > - the ioctls to work on 64bit machines
> > ?
>
> Luc and I'm happy to help doing further work on it. However it's been like that
> in kernel for years so it might also be a good one for the janitors to join in
> on ?
I'd like to see this driver back in mainline too. Luc, please contact
me and we'll get this working correctly on a bunch of machines.
mh
--
Martin Hicks Wild Open Source Inc.
[email protected] 613-266-2296