[GIT PATCHES] V4L/DVB fixes

[GIT PATCHES] V4L/DVB fixes

[Mauro Carvalho Chehab]

Linus,

Please pull from:
        ssh://master.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb.git master

For the following:

   - several driver fixes at ivtv, em28xx, tvp5150, tda10086, saa7146, zl10353, 
     saa7134, saa7134-alsa, saa7134-dvb, cxusb, saa5246a and saa5249;

   - section fixes for bttv and ivtv;

   - Kconfig fixes for dib3000mc;

   - videobuf-core bad lock fixes;

   - memory leak fix at videobuf-vmalloc.

Cheers,
Mauro.

---

 drivers/media/common/saa7146_video.c          |   11 +-
 drivers/media/dvb/dvb-usb/Makefile            |    3 +
 drivers/media/dvb/dvb-usb/dibusb-common.c     |    4 +
 drivers/media/dvb/frontends/tda10086.c        |   14 +-
 drivers/media/dvb/frontends/zl10353.c         |    4 +-
 drivers/media/dvb/frontends/zl10353.h         |    2 +-
 drivers/media/video/bt8xx/bttv-cards.c        |    2 +-
 drivers/media/video/bt8xx/bttv-driver.c       |    9 +-
 drivers/media/video/cx88/cx88-blackbird.c     |    5 +-
 drivers/media/video/cx88/cx88-video.c         |    5 +-
 drivers/media/video/em28xx/em28xx-video.c     |    7 +-
 drivers/media/video/ivtv/ivtv-i2c.c           |    2 +-
 drivers/media/video/ivtv/ivtv-i2c.h           |    2 +-
 drivers/media/video/ivtv/ivtv-streams.c       |    2 +-
 drivers/media/video/saa5246a.c                |   10 +-
 drivers/media/video/saa5249.c                 |    8 +-
 drivers/media/video/saa7134/saa7134-alsa.c    |   16 ++-
 drivers/media/video/saa7134/saa7134-cards.c   |    2 +-
 drivers/media/video/saa7134/saa7134-core.c    |   21 +-
 drivers/media/video/saa7134/saa7134-dvb.c     |    1 +
 drivers/media/video/saa7134/saa7134-empress.c |    5 +-
 drivers/media/video/saa7134/saa7134-video.c   |    5 +-
 drivers/media/video/tvp5150.c                 |   11 +
 drivers/media/video/videobuf-core.c           |  247 ++++++++++++++++---------
 drivers/media/video/videobuf-vmalloc.c        |    6 +-
 drivers/media/video/vivi.c                    |    1 +
 include/media/videobuf-core.h                 |    2 +
 27 files changed, 246 insertions(+), 161 deletions(-)

Adrian Bunk (2):
      V4L/DVB (6796): ivtv/ section fix
      V4L/DVB (6797): bt8xx/ section fixes

Benoit Istin (1):
      V4L/DVB (6746): saa7134-dvb: fix tuning for WinTV HVR-1110

Brandon Philips (4):
      V4L/DVB (6601): V4L: videobuf-core locking fixes and comments
      V4L/DVB (6602): V4L: Convert videobuf drivers to videobuf_stop
      V4L/DVB (6615): V4L: Fix VIDIOCGMBUF  locking in saa7146
      V4L/DVB (6751): V4L: Memory leak!  Fix count in videobuf-vmalloc mmap

Chris Pascoe (1):
      V4L/DVB (6629): zl10353: fix default adc_clock and TRL nominal rate calculation

Hans Verkuil (1):
      V4L/DVB (6485): ivtv: fix compile warning

Hartmut Hackmann (1):
      V4L/DVB (6579): Fix bug #8824: Correct support for Diseqc on tda10086

Heikki Lindholm (2):
      V4L/DVB (6666): saa7134-alsa: fix period handling
      V4L/DVB (6690): saa7134: fix ignored interrupts

Hermann Pitton (1):
      V4L/DVB (6686): saa7134: fix composite over s-video input on the Tevion MD 9717

Jiri Kosina (1):
      V4L/DVB (6733): DVB: Compile 3000MC-specific DIB code only for CONFIG_DVB_DIB3000MC

Mauro Carvalho Chehab (4):
      V4L/DVB (6542): Fix S-video mode on tvp5150
      V4L/DVB (6581): Fix: avoids negative vma usage count
      V4L/DVB (6750): Fix in-kernel compilation for cxusb
      V4L/DVB (6794): Fix compilation when dib3000mc is compiled as a module

Richard Knutsson (1):
      V4L/DVB (6684): Complement va_start() with va_end() + style fixes

Sascha Sommer (1):
      V4L/DVB (6540): em28xx: fix failing autodetection after the reboot

---------------------------------------------------
V4L/DVB development is hosted at http://linuxtv.org

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Mauro Carvalho Chehab]

Em Ter, 2007-12-11 às 21:14 -0500, Michael Krufky escreveu:
> Mauro Carvalho Chehab wrote:
> > Please pull from:
> >         ssh://master.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb.git master
> >
> > [snip]
> >
> >   
> Mauro,
> > Mauro Carvalho Chehab (4):
> >       V4L/DVB (6750): Fix in-kernel compilation for cxusb
> >   
> ^^ This patch fixes a build issue that only applies to new devices that
> depend on the tuner-xc2028 module, which is planned to be added to
> mainline for 2.6.25 ....  We don't need to send this patch upstream yet,
> because the dependency doesn't exist in 2.6.24.

Hmm... ok. Anyway, It won't hurt to apply it. The patch just adds a
missing include directory to Makefile:
+EXTRA_CFLAGS += -Idrivers/media/video

Anyway, I'll remove it from the tree and send to linus an updated pull
request.

> > Benoit Istin (1):
> >       V4L/DVB (6746): saa7134-dvb: fix tuning for WinTV HVR-1110
> >
> >   
> ^^ This patch is good for 2.6.24, but it is only the first half of the
> fix.  This fixes digital tuning, but the analog fix is the following
> changeset:
> 
> saa7134: enable LNA in analog mode for Hauppauge WinTV HVR-1110
> http://linuxtv.org/hg/~mkrufky/oxford2/rev/b227949c41ad
> 
> Please send this one in your next pull request to Linus.
Ok.

I'll pull from your oxford2 tree and keep it for a while at development tree for testing.

> Regards,
> 
> Mike
-- 
Cheers,
Mauro

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Mauro Carvalho Chehab]

> > ^^ This patch fixes a build issue that only applies to new devices
> that
> > depend on the tuner-xc2028 module, which is planned to be added to
> > mainline for 2.6.25 ....  We don't need to send this patch upstream yet,
> > because the dependency doesn't exist in 2.6.24.
> 
> Hmm... ok. Anyway, It won't hurt to apply it. The patch just adds a
> missing include directory to Makefile:
> +EXTRA_CFLAGS += -Idrivers/media/video
> 
> Anyway, I'll remove it from the tree and send to linus an updated pull
> request.

Linus already pulled from my tree. There's no need to revert this
changeset. It will not hurt and will be needed soon, anyway.

-- 
Cheers,
Mauro

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Michael Krufky]

Mauro Carvalho Chehab wrote:
>>> ^^ This patch fixes a build issue that only applies to new devices
>>>       
>> that
>>     
>>> depend on the tuner-xc2028 module, which is planned to be added to
>>> mainline for 2.6.25 ....  We don't need to send this patch upstream yet,
>>> because the dependency doesn't exist in 2.6.24.
>>>       
>> Hmm... ok. Anyway, It won't hurt to apply it. The patch just adds a
>> missing include directory to Makefile:
>> +EXTRA_CFLAGS += -Idrivers/media/video
>>
>> Anyway, I'll remove it from the tree and send to linus an updated pull
>> request.
>>     
>
> Linus already pulled from my tree. There's no need to revert this
> changeset. It will not hurt and will be needed soon, anyway.
>   
That's fine -- it doesn't hurt anything.

Cheers,

Mike

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Trent Piepho]

On Tue, 12 Feb 2008, Mauro Carvalho Chehab wrote:
>    - cx88-mpeg: Allow concurrent access to cx88-mpeg devices;

So you decided to just commit this one with the race condition anyway?

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Mauro Carvalho Chehab]

On Tue, 12 Feb 2008 21:21:43 -0800 (PST)
Trent Piepho <xyzzy@speakeasy.org> wrote:

> On Tue, 12 Feb 2008, Mauro Carvalho Chehab wrote:
> >    - cx88-mpeg: Allow concurrent access to cx88-mpeg devices;
> 
> So you decided to just commit this one with the race condition anyway?

The version with problems is the one dated by Jan, 16:
http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-January/006119.html

As Ricardo stated on Feb, 5, he fixed the lock issues, that were present on
your first revision:

http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-February/006292.html

I've pushed the reviewed version at the same day, for testing, at the
development environment:

http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-February/006326.html

Nobody pointed any newer issues on the reviewed version, since then.

If you still see any issues, please send us a patch fixing it.

Mauro

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Trent Piepho]

On Wed, 13 Feb 2008, Mauro Carvalho Chehab wrote:
> On Tue, 12 Feb 2008 21:21:43 -0800 (PST)
> Trent Piepho <xyzzy@speakeasy.org> wrote:
>
> > On Tue, 12 Feb 2008, Mauro Carvalho Chehab wrote:
> > >    - cx88-mpeg: Allow concurrent access to cx88-mpeg devices;
> >
> > So you decided to just commit this one with the race condition anyway?
>
> The version with problems is the one dated by Jan, 16:
> http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-January/006119.html
>
> As Ricardo stated on Feb, 5, he fixed the lock issues, that were present on
> your first revision:
>
> http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-February/006292.html

I don't see what you are saying in that message.  Ricardo agreed it had a
race condition and said he would, at some future point, redo it, but I
never saw a new patch.

> I've pushed the reviewed version at the same day, for testing, at the

review-by == I saw your patch in my inbox and read the title

> development environment:
>
> http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-February/006326.html
>
> Nobody pointed any newer issues on the reviewed version, since then.

Nobody points to any issues, ever, when this happens.  Look at how broken
the v4l2 only bttv driver was.  You still can't unload and load cx88-dvb
since Markus's patch for hotplug, no one's fixed that regression.  Now
there's yet another race in the cx88 subdriver code.  Maybe the original
author who added that code should feel some motiviation to fix the
regressions it caused....

The only complaints that are ever generated on commits to the development
enviroment are political, "so-and-so shouldn't be able to patch my
fiefdom."

> If you still see any issues, please send us a patch fixing it.

I don't have the time or the desire to be the janitor who cleans up after
sloppy coders' bugs.

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Ricardo Cerqueira]

On Wed, 2008-02-13 at 10:45 -0800, Trent Piepho wrote:
> On Wed, 13 Feb 2008, Mauro Carvalho Chehab wrote:
> > On Tue, 12 Feb 2008 21:21:43 -0800 (PST)
> > Trent Piepho <xyzzy@speakeasy.org> wrote:
> >
> > The version with problems is the one dated by Jan, 16:
> > http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-January/006119.html
> >
> > As Ricardo stated on Feb, 5, he fixed the lock issues, that were present on
> > your first revision:
> >
> > http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-February/006292.html
> 
> I don't see what you are saying in that message.  Ricardo agreed it had a
> race condition and said he would, at some future point, redo it, but I
> never saw a new patch.

You missed it, then:

http://linuxtv.org/pipermail/v4l-dvb-maintainer/2008-February/006293.html

was sent in 25 minutes after the message Mauro mentioned.


> Nobody points to any issues, ever, when this happens.  Look at how broken
> the v4l2 only bttv driver was.  You still can't unload and load cx88-dvb
> since Markus's patch for hotplug, no one's fixed that regression.  Now
> there's yet another race in the cx88 subdriver code.  Maybe the original
> author who added that code should feel some motiviation to fix the
> regressions it caused....

Read the code that was actually committed: here's a helpful link:

http://git.kernel.org/?p=linux/kernel/git/mchehab/v4l-dvb.git;a=commitdiff;h=41a93616082af630e7242cba766a161d7847560b


> 
> I don't have the time or the desire to be the janitor who cleans up after
> sloppy coders' bugs.

You properly raised a red flag, and got a fix in return.
Don't expect fixes to be accompanied by a "are you happy now?"
follow-up; if you took the time to check the first commit, take the time
to read the rest.

Now... it looks as if you simply missed the second patch; it happens.
But next time you feel like sending this kind of message(s) to 3
separate lists, at least check your mail history (or read the code
you're commenting).

--
RC

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Trent Piepho]

On Wed, 13 Feb 2008, Ricardo Cerqueira wrote:
> On Wed, 2008-02-13 at 10:45 -0800, Trent Piepho wrote:
> > On Wed, 13 Feb 2008, Mauro Carvalho Chehab wrote:
> > > On Tue, 12 Feb 2008 21:21:43 -0800 (PST)
> > > Trent Piepho <xyzzy@speakeasy.org> wrote:
> > >
> > Nobody points to any issues, ever, when this happens.  Look at how broken
> > the v4l2 only bttv driver was.  You still can't unload and load cx88-dvb
> > since Markus's patch for hotplug, no one's fixed that regression.  Now
> > there's yet another race in the cx88 subdriver code.  Maybe the original
> > author who added that code should feel some motiviation to fix the
> > regressions it caused....
>
> Read the code that was actually committed: here's a helpful link:
>
> http://git.kernel.org/?p=linux/kernel/git/mchehab/v4l-dvb.git;a=commitdiff;h=41a93616082af630e7242cba766a161d7847560b

I did.  Looks like there is a race on access to core->active_ref.

> > I don't have the time or the desire to be the janitor who cleans up after
> > sloppy coders' bugs.
>
> You properly raised a red flag, and got a fix in return.
> Don't expect fixes to be accompanied by a "are you happy now?"
> follow-up; if you took the time to check the first commit, take the time
> to read the rest.

You could also CC the relevant parties.  Unless you're hoping they won't
notice something.

Re: [v4l-dvb-maintainer] [GIT PATCHES] V4L/DVB fixes

[Ricardo Cerqueira]

On Wed, 2008-02-13 at 13:53 -0800, Trent Piepho wrote:
> On Wed, 13 Feb 2008, Ricardo Cerqueira wrote:
> > On Wed, 2008-02-13 at 10:45 -0800, Trent Piepho wrote:
> > > On Wed, 13 Feb 2008, Mauro Carvalho Chehab wrote:
> > > > On Tue, 12 Feb 2008 21:21:43 -0800 (PST)
> > > > Trent Piepho <xyzzy@speakeasy.org> wrote:
> > > >
> > Read the code that was actually committed: here's a helpful link:
> >
> > http://git.kernel.org/?p=linux/kernel/git/mchehab/v4l-dvb.git;a=commitdiff;h=41a93616082af630e7242cba766a161d7847560b
> 
> I did.  Looks like there is a race on access to core->active_ref.

You're right, I missed that. A patch has already been sent to the
V4L/DVB list.

> You could also CC the relevant parties.  Unless you're hoping they won't
> notice something.

Sorry, but no. I actively try *not* to send duplicate e-mails to people
unless they specifically request it.

--
RC