* [GIT PULL FOR v3.5] Various fixes
@ 2012-04-19 15:48 Hans Verkuil
2012-04-20 0:07 ` Andy Walls
2012-04-20 11:57 ` Sakari Ailus
0 siblings, 2 replies; 5+ messages in thread
From: Hans Verkuil @ 2012-04-19 15:48 UTC (permalink / raw)
To: linux-media
Cc: Sakari Ailus, Andy Walls, Mauro Carvalho Chehab, Manjunath Hadli
While I was cleaning up some older drivers I came across a few bugs that are
fixed here. The fixes are all trivial one-liners.
Regards,
Hans
The following changes since commit f4d4e7656b26a6013bc5072c946920d2e2c44e8e:
[media] em28xx: Make em28xx-input.c a separate module (2012-04-10 20:45:41 -0300)
are available in the git repository at:
git://linuxtv.org/hverkuil/media_tree.git fixes
for you to fetch changes up to f85e735051e71410bfd695536a25c1013bceeabc:
vivi: fix duplicate line. (2012-04-19 17:38:52 +0200)
----------------------------------------------------------------
Hans Verkuil (4):
V4L: fix incorrect refcounting.
V4L2: drivers implementing vidioc_default should also return -ENOTTY
v4l2-ctrls.c: zero min/max/step/def values for 64 bit integers.
vivi: fix duplicate line.
Documentation/video4linux/v4l2-framework.txt | 14 +++++++++-----
drivers/media/radio/dsbr100.c | 1 -
drivers/media/radio/radio-keene.c | 1 -
drivers/media/video/cx18/cx18-ioctl.c | 2 +-
drivers/media/video/davinci/vpfe_capture.c | 2 +-
drivers/media/video/ivtv/ivtv-ioctl.c | 2 +-
drivers/media/video/meye.c | 2 +-
drivers/media/video/mxb.c | 2 +-
drivers/media/video/v4l2-ctrls.c | 1 +
drivers/media/video/vivi.c | 2 +-
10 files changed, 16 insertions(+), 13 deletions(-)
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [GIT PULL FOR v3.5] Various fixes
2012-04-19 15:48 [GIT PULL FOR v3.5] Various fixes Hans Verkuil
@ 2012-04-20 0:07 ` Andy Walls
2012-04-20 11:57 ` Sakari Ailus
1 sibling, 0 replies; 5+ messages in thread
From: Andy Walls @ 2012-04-20 0:07 UTC (permalink / raw)
To: Hans Verkuil
Cc: linux-media, Sakari Ailus, Mauro Carvalho Chehab, Manjunath Hadli
On Thu, 2012-04-19 at 17:48 +0200, Hans Verkuil wrote:
> While I was cleaning up some older drivers I came across a few bugs that are
> fixed here. The fixes are all trivial one-liners.
>
> Regards,
>
> Hans
The fixes for ivtv and cx18 look good to me. Thanks Hans.
Reviewed-by: Andy Walls <awalls@md.metrocast.net>
Regards,
Andy
> The following changes since commit f4d4e7656b26a6013bc5072c946920d2e2c44e8e:
>
> [media] em28xx: Make em28xx-input.c a separate module (2012-04-10 20:45:41 -0300)
>
> are available in the git repository at:
>
> git://linuxtv.org/hverkuil/media_tree.git fixes
>
> for you to fetch changes up to f85e735051e71410bfd695536a25c1013bceeabc:
>
> vivi: fix duplicate line. (2012-04-19 17:38:52 +0200)
>
> ----------------------------------------------------------------
> Hans Verkuil (4):
> V4L: fix incorrect refcounting.
> V4L2: drivers implementing vidioc_default should also return -ENOTTY
> v4l2-ctrls.c: zero min/max/step/def values for 64 bit integers.
> vivi: fix duplicate line.
>
> Documentation/video4linux/v4l2-framework.txt | 14 +++++++++-----
> drivers/media/radio/dsbr100.c | 1 -
> drivers/media/radio/radio-keene.c | 1 -
> drivers/media/video/cx18/cx18-ioctl.c | 2 +-
> drivers/media/video/davinci/vpfe_capture.c | 2 +-
> drivers/media/video/ivtv/ivtv-ioctl.c | 2 +-
> drivers/media/video/meye.c | 2 +-
> drivers/media/video/mxb.c | 2 +-
> drivers/media/video/v4l2-ctrls.c | 1 +
> drivers/media/video/vivi.c | 2 +-
> 10 files changed, 16 insertions(+), 13 deletions(-)
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [GIT PULL FOR v3.5] Various fixes
2012-04-19 15:48 [GIT PULL FOR v3.5] Various fixes Hans Verkuil
2012-04-20 0:07 ` Andy Walls
@ 2012-04-20 11:57 ` Sakari Ailus
2012-04-20 12:15 ` Hans Verkuil
1 sibling, 1 reply; 5+ messages in thread
From: Sakari Ailus @ 2012-04-20 11:57 UTC (permalink / raw)
To: Hans Verkuil
Cc: linux-media, Andy Walls, Mauro Carvalho Chehab, Manjunath Hadli
Hi Hans,
On Thu, Apr 19, 2012 at 05:48:51PM +0200, Hans Verkuil wrote:
> While I was cleaning up some older drivers I came across a few bugs that are
> fixed here. The fixes are all trivial one-liners.
>
> Regards,
>
> Hans
>
> The following changes since commit f4d4e7656b26a6013bc5072c946920d2e2c44e8e:
>
> [media] em28xx: Make em28xx-input.c a separate module (2012-04-10 20:45:41 -0300)
>
> are available in the git repository at:
>
> git://linuxtv.org/hverkuil/media_tree.git fixes
>
> for you to fetch changes up to f85e735051e71410bfd695536a25c1013bceeabc:
>
> vivi: fix duplicate line. (2012-04-19 17:38:52 +0200)
>
> ----------------------------------------------------------------
> Hans Verkuil (4):
> V4L: fix incorrect refcounting.
> V4L2: drivers implementing vidioc_default should also return -ENOTTY
> v4l2-ctrls.c: zero min/max/step/def values for 64 bit integers.
Thanks for the patches.
I'd put the setting of the min/max/step values for 64-bit controls to
v4l2_ctrl_new() instead for two reasons:
- All 64-bit controls require this for the time being, independent of the
purpose, at least until min/max/step are implemented for them and
- that's where part of the string control initialisation is done as well.
What do you think?
I think that in the long run we'll need to add capability to get
min/max/step for 64-bit controls also to user space, but the information
could be added to the controls in the kernel already earlier.
Kind regards,
--
Sakari Ailus
e-mail: sakari.ailus@iki.fi jabber/XMPP/Gmail: sailus@retiisi.org.uk
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [GIT PULL FOR v3.5] Various fixes
2012-04-20 11:57 ` Sakari Ailus
@ 2012-04-20 12:15 ` Hans Verkuil
2012-04-20 12:32 ` Sakari Ailus
0 siblings, 1 reply; 5+ messages in thread
From: Hans Verkuil @ 2012-04-20 12:15 UTC (permalink / raw)
To: Sakari Ailus
Cc: linux-media, Andy Walls, Mauro Carvalho Chehab, Manjunath Hadli
On Friday, April 20, 2012 13:57:00 Sakari Ailus wrote:
> Hi Hans,
>
> On Thu, Apr 19, 2012 at 05:48:51PM +0200, Hans Verkuil wrote:
> > While I was cleaning up some older drivers I came across a few bugs that are
> > fixed here. The fixes are all trivial one-liners.
> >
> > Regards,
> >
> > Hans
> >
> > The following changes since commit f4d4e7656b26a6013bc5072c946920d2e2c44e8e:
> >
> > [media] em28xx: Make em28xx-input.c a separate module (2012-04-10 20:45:41 -0300)
> >
> > are available in the git repository at:
> >
> > git://linuxtv.org/hverkuil/media_tree.git fixes
> >
> > for you to fetch changes up to f85e735051e71410bfd695536a25c1013bceeabc:
> >
> > vivi: fix duplicate line. (2012-04-19 17:38:52 +0200)
> >
> > ----------------------------------------------------------------
> > Hans Verkuil (4):
> > V4L: fix incorrect refcounting.
> > V4L2: drivers implementing vidioc_default should also return -ENOTTY
> > v4l2-ctrls.c: zero min/max/step/def values for 64 bit integers.
>
> Thanks for the patches.
>
> I'd put the setting of the min/max/step values for 64-bit controls to
> v4l2_ctrl_new() instead for two reasons:
>
> - All 64-bit controls require this for the time being, independent of the
> purpose, at least until min/max/step are implemented for them and
>
> - that's where part of the string control initialisation is done as well.
>
> What do you think?
I disagree :-)
v4l2_ctrl_fill is the only function at the moment that fills in these fields.
Doing it elsewhere just for that type is inconsistent.
Also, v4l2_ctrl_fill is also called by some older drivers that do not use the
control framework yet, so it really has to be done there.
Another matter is that ivtv-controls.c uses non-zero max and step values for
these 64 bit controls, that should be fixed as well.
>
> I think that in the long run we'll need to add capability to get
> min/max/step for 64-bit controls also to user space, but the information
> could be added to the controls in the kernel already earlier.
It wouldn't be difficult to make min/max/def of type s64. At this moment I do
not see a compelling reason, but if we get more 64-bit integer controls that
do need this, then we will have to add it.
Regards,
Hans
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [GIT PULL FOR v3.5] Various fixes
2012-04-20 12:15 ` Hans Verkuil
@ 2012-04-20 12:32 ` Sakari Ailus
0 siblings, 0 replies; 5+ messages in thread
From: Sakari Ailus @ 2012-04-20 12:32 UTC (permalink / raw)
To: Hans Verkuil
Cc: linux-media, Andy Walls, Mauro Carvalho Chehab, Manjunath Hadli
Hi,
Hans Verkuil wrote:
> On Friday, April 20, 2012 13:57:00 Sakari Ailus wrote:
>> Hi Hans,
>>
>> On Thu, Apr 19, 2012 at 05:48:51PM +0200, Hans Verkuil wrote:
>>> While I was cleaning up some older drivers I came across a few bugs that are
>>> fixed here. The fixes are all trivial one-liners.
>>>
>>> Regards,
>>>
>>> Hans
>>>
>>> The following changes since commit f4d4e7656b26a6013bc5072c946920d2e2c44e8e:
>>>
>>> [media] em28xx: Make em28xx-input.c a separate module (2012-04-10 20:45:41 -0300)
>>>
>>> are available in the git repository at:
>>>
>>> git://linuxtv.org/hverkuil/media_tree.git fixes
>>>
>>> for you to fetch changes up to f85e735051e71410bfd695536a25c1013bceeabc:
>>>
>>> vivi: fix duplicate line. (2012-04-19 17:38:52 +0200)
>>>
>>> ----------------------------------------------------------------
>>> Hans Verkuil (4):
>>> V4L: fix incorrect refcounting.
>>> V4L2: drivers implementing vidioc_default should also return -ENOTTY
>>> v4l2-ctrls.c: zero min/max/step/def values for 64 bit integers.
>>
>> Thanks for the patches.
>>
>> I'd put the setting of the min/max/step values for 64-bit controls to
>> v4l2_ctrl_new() instead for two reasons:
>>
>> - All 64-bit controls require this for the time being, independent of the
>> purpose, at least until min/max/step are implemented for them and
>>
>> - that's where part of the string control initialisation is done as well.
>>
>> What do you think?
>
> I disagree :-)
>
> v4l2_ctrl_fill is the only function at the moment that fills in these fields.
> Doing it elsewhere just for that type is inconsistent.
>
> Also, v4l2_ctrl_fill is also called by some older drivers that do not use the
> control framework yet, so it really has to be done there.
Ok; after taking another look there I agree with you.
> Another matter is that ivtv-controls.c uses non-zero max and step values for
> these 64 bit controls, that should be fixed as well.
Right. Feel free to add to that patch:
Acked-by: Sakari Ailus <sakari.ailus@iki.fi>
>> I think that in the long run we'll need to add capability to get
>> min/max/step for 64-bit controls also to user space, but the information
>> could be added to the controls in the kernel already earlier.
>
> It wouldn't be difficult to make min/max/def of type s64. At this moment I do
> not see a compelling reason, but if we get more 64-bit integer controls that
> do need this, then we will have to add it.
The controls the SMIA++ driver uses these for are also read-only and so
set by the driver. Once we get controls that would benefit from
min/max/step they should be added at the same time.
Regards,
--
Sakari Ailus
sakari.ailus@iki.fi
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2012-04-20 12:32 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-04-19 15:48 [GIT PULL FOR v3.5] Various fixes Hans Verkuil
2012-04-20 0:07 ` Andy Walls
2012-04-20 11:57 ` Sakari Ailus
2012-04-20 12:15 ` Hans Verkuil
2012-04-20 12:32 ` Sakari Ailus
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.