firefly-linux-kernel-4.4.55.git
13 years ago[media] V4L: ov9740: Fixed some settings
Andrew Chew [Thu, 23 Jun 2011 23:19:41 +0000 (20:19 -0300)]
[media] V4L: ov9740: Fixed some settings

Based on vendor feedback, should issue a software reset at start of day.

Also, OV9740_ANALOG_CTRL15 needs to be changed so the sensor does not begin
streaming until it is ready (otherwise, results in a nonsense frame for the
initial frame).

Added a comment on using discontinuous clock.

Finally, OV9740_ISP_CTRL19 needs to be changed to really use YUYV ordering
(the previous value was for VYUY).

Signed-off-by: Andrew Chew <achew@nvidia.com>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: ov9740: Correct print in ov9740_reg_rmw()
Andrew Chew [Thu, 23 Jun 2011 23:19:40 +0000 (20:19 -0300)]
[media] V4L: ov9740: Correct print in ov9740_reg_rmw()

The register width of the ov9740 is 16-bits, so correct the debug print
to reflect this.

Signed-off-by: Andrew Chew <achew@nvidia.com>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: ov9740: Cleanup hex casing inconsistencies
Andrew Chew [Thu, 23 Jun 2011 23:19:39 +0000 (20:19 -0300)]
[media] V4L: ov9740: Cleanup hex casing inconsistencies

Made all hex number casing use lower-case throughout the entire driver
for consistency.

Signed-off-by: Andrew Chew <achew@nvidia.com>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: at91: add Atmel Image Sensor Interface (ISI) support
Josh Wu [Wed, 8 Jun 2011 01:40:19 +0000 (22:40 -0300)]
[media] V4L: at91: add Atmel Image Sensor Interface (ISI) support

This patch is to enable Atmel Image Sensor Interface (ISI) driver support.

Signed-off-by: Josh Wu <josh.wu@atmel.com>
Acked-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: soc-camera: remove several now unused soc-camera client operations
Guennadi Liakhovetski [Tue, 7 Jun 2011 09:24:32 +0000 (06:24 -0300)]
[media] V4L: soc-camera: remove several now unused soc-camera client operations

This patch removes .enum_input(), .suspend() and .resume() soc-camera
client operations.

Functionality, provided by .enum_input(), if needed, can be implemented
using the v4l2-subdev API.

As for .suspend() and .resume(), the only client driver, implementing
these methods has been mt9m111, and the only host driver, using them
has been pxa-camera. Now that both those drivers have been converted
to the standard subdev .s_power() operation, .suspend() and .resume()
can be removed.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: mt9m111: switch to v4l2-subdev .s_power() method
Guennadi Liakhovetski [Sun, 5 Jun 2011 11:27:39 +0000 (08:27 -0300)]
[media] V4L: mt9m111: switch to v4l2-subdev .s_power() method

Eliminate soc-camera specific .suspend() and .restore() methods in favour
of the standard v4l2-subdev .s_power() method

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: mt9m111: propagate higher level abstraction down in functions
Guennadi Liakhovetski [Tue, 7 Jun 2011 09:47:30 +0000 (06:47 -0300)]
[media] V4L: mt9m111: propagate higher level abstraction down in functions

It is more convenient to propagate the higher level abstraction - the
struct mt9m111 object into functions and then retrieve a pointer to
the i2c client, if needed, than to do the reverse.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Acked-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: soc-camera: MIPI flags are not sensor flags
Guennadi Liakhovetski [Sat, 4 Jun 2011 20:24:22 +0000 (17:24 -0300)]
[media] V4L: soc-camera: MIPI flags are not sensor flags

SOCAM_MIPI_[1234]LANE flags are not board-specific sensor flags, they
are bus configuration flags.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: tw9910: remove bogus ENUMINPUT implementation
Guennadi Liakhovetski [Sat, 4 Jun 2011 19:48:10 +0000 (16:48 -0300)]
[media] V4L: tw9910: remove bogus ENUMINPUT implementation

tw9910 is a TV decoder, it doesn't have a tuner. Besides, the
.enum_input soc-camera operation is optional and normally not needed.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: sh_mobile_ceu_camera: remove redundant calculations
Guennadi Liakhovetski [Sat, 4 Jun 2011 18:09:02 +0000 (15:09 -0300)]
[media] V4L: sh_mobile_ceu_camera: remove redundant calculations

soc_camera core now performs the standard .bytesperline and .sizeimage
calculations internally, no need to duplicate in drivers.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: mx2_camera: .try_fmt shouldn't fail
Guennadi Liakhovetski [Sat, 4 Jun 2011 18:06:47 +0000 (15:06 -0300)]
[media] V4L: mx2_camera: .try_fmt shouldn't fail

If the user is requesting too large a frame, instead of failing
select an acceptable geometry, preserving the requested aspect ratio.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: pxa-camera: switch to using subdev .s_power() core operation
Guennadi Liakhovetski [Tue, 7 Jun 2011 08:50:15 +0000 (05:50 -0300)]
[media] V4L: pxa-camera: switch to using subdev .s_power() core operation

soc-camera specific .suspend() and .resume() methods are deprecated
and should be replaced by the subdev standard .s_power() operation.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Acked-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: pxa-camera: try to force progressive video format
Guennadi Liakhovetski [Tue, 7 Jun 2011 08:44:19 +0000 (05:44 -0300)]
[media] V4L: pxa-camera: try to force progressive video format

The pxa-camera driver only supports progressive video so far. Passing
down to the client the same format, as what the user has requested can
result in interlaced video, even if the client supports both. This
patch avoids such cases.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Acked-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: pxa_camera: remove redundant calculations
Guennadi Liakhovetski [Tue, 7 Jun 2011 08:42:57 +0000 (05:42 -0300)]
[media] V4L: pxa_camera: remove redundant calculations

soc_camera core now performs the standard .bytesperline and .sizeimage
calculations internally, no need to duplicate in drivers.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Acked-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L: mx3_camera: remove redundant calculations
Guennadi Liakhovetski [Sat, 4 Jun 2011 17:11:10 +0000 (14:11 -0300)]
[media] V4L: mx3_camera: remove redundant calculations

soc_camera core now performs the standard .bytesperline and .sizeimage
calculations internally, no need to duplicate in drivers.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: make manual_mode_value 8 bits and check against control range
Hans Verkuil [Tue, 28 Jun 2011 13:43:13 +0000 (10:43 -0300)]
[media] v4l2-ctrls: make manual_mode_value 8 bits and check against control range

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-compat-ioctl32: add VIDIOC_DQEVENT support
Hans Verkuil [Tue, 7 Jun 2011 14:43:57 +0000 (11:43 -0300)]
[media] v4l2-compat-ioctl32: add VIDIOC_DQEVENT support

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] ivtv: add control event support
Hans Verkuil [Tue, 28 Jun 2011 13:40:42 +0000 (10:40 -0300)]
[media] ivtv: add control event support

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] vivi: support control events
Hans Verkuil [Tue, 7 Jun 2011 13:20:23 +0000 (10:20 -0300)]
[media] vivi: support control events

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L2 spec: document control events
Hans Verkuil [Tue, 7 Jun 2011 12:53:37 +0000 (09:53 -0300)]
[media] V4L2 spec: document control events

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: simplify event subscription
Hans Verkuil [Fri, 10 Jun 2011 08:56:39 +0000 (05:56 -0300)]
[media] v4l2-ctrls: simplify event subscription

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: add control events
Hans Verkuil [Tue, 7 Jun 2011 14:13:44 +0000 (11:13 -0300)]
[media] v4l2-ctrls: add control events

Whenever a control changes value or state an event is sent to anyone
that subscribed to it.

This functionality is useful for control panels but also for applications
that need to wait for (usually status) controls to change value.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: add v4l2_fh pointer to the set control functions
Hans Verkuil [Tue, 7 Jun 2011 09:47:18 +0000 (06:47 -0300)]
[media] v4l2-ctrls: add v4l2_fh pointer to the set control functions

When an application changes a control you want to generate an event.
However, you want to avoid sending such an event back to the application
(file handle) that caused the change.

Add the filehandle to the various set control functions.

The filehandle isn't used yet, but the control event patches will need
this.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] vivi: add autogain/gain support to test the autocluster functionality
Hans Verkuil [Tue, 7 Jun 2011 09:34:41 +0000 (06:34 -0300)]
[media] vivi: add autogain/gain support to test the autocluster functionality

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] Documentation: Improve cluster documentation and document the new autoclusters
Hans Verkuil [Tue, 7 Jun 2011 08:46:53 +0000 (05:46 -0300)]
[media] Documentation: Improve cluster documentation and document the new autoclusters

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: add v4l2_ctrl_auto_cluster to simplify autogain/gain scenarios
Hans Verkuil [Fri, 10 Jun 2011 08:44:36 +0000 (05:44 -0300)]
[media] v4l2-ctrls: add v4l2_ctrl_auto_cluster to simplify autogain/gain scenarios

It is a bit tricky to handle autogain/gain type scenerios correctly. Such
controls need to be clustered and the V4L2_CTRL_FLAG_UPDATE should be set on
the autofoo controls. In addition, the manual controls should be marked
inactive when the automatic mode is on, and active when the manual mode is on.
This also requires specialized volatile handling.

The chances of drivers doing all these things correctly are pretty remote.
So a new v4l2_ctrl_auto_cluster function was added that takes care of these
issues.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-controls.txt: update to latest v4l2-ctrl.c changes
Hans Verkuil [Fri, 27 May 2011 11:53:37 +0000 (08:53 -0300)]
[media] v4l2-controls.txt: update to latest v4l2-ctrl.c changes

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: fix and improve volatile control handling
Hans Verkuil [Fri, 10 Jun 2011 08:43:34 +0000 (05:43 -0300)]
[media] v4l2-ctrls: fix and improve volatile control handling

If you have a cluster of controls that is a mix of volatile and non-volatile
controls, then requesting the value of the volatile control would fail if the
master control of that cluster was non-volatile. The code assumed that the
volatile state of the master control was the same for all other controls in
the cluster.

This is now fixed.

In addition, it was clear from bugs in some drivers that it was confusing that
the ctrl->cur union had to be used in g_volatile_ctrl. Several drivers used the
'new' values instead. The framework was changed so that drivers now set the new
value instead of the current value.

This has an additional benefit as well: the volatile values are now only stored
in the 'new' value, leaving the current value alone. This is useful for
autofoo/foo control clusters where you want to have a 'foo' control act like a
volatile control if 'autofoo' is on, but as a normal control when it is off.

Since with this change the cur value is no longer overwritten when g_volatile_ctrl
is called, you can use it to remember the original 'foo' value. For example:

autofoo = 0, foo = 10 and foo is non-volatile.

Now autofoo is set to 1 and foo is marked volatile. Retrieving the foo value
will get the volatile value. Set autofoo back to 0, which marks foo as non-
volatile again, and retrieving foo will get the old current value of 10.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-subdev: implement per-filehandle control handlers
Hans Verkuil [Wed, 25 May 2011 11:52:13 +0000 (08:52 -0300)]
[media] v4l2-subdev: implement per-filehandle control handlers

To be consistent with v4l2-ioctl.c.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ioctl: add ctrl_handler to v4l2_fh
Hans Verkuil [Sat, 12 Mar 2011 11:54:43 +0000 (08:54 -0300)]
[media] v4l2-ioctl: add ctrl_handler to v4l2_fh

This is required to implement control events and is also needed to allow
for per-filehandle control handlers.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: drivers should be able to ignore the READ_ONLY flag
Hans Verkuil [Tue, 7 Jun 2011 07:50:31 +0000 (04:50 -0300)]
[media] v4l2-ctrls: drivers should be able to ignore the READ_ONLY flag

When applications try to set READ_ONLY controls an error should
be returned. However, when drivers do that it should be accepted.

Those controls could reflect some driver status which the application
can't change but the driver obviously has to be able to change it.

This is needed among others for future HDMI status controls.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: simplify error_idx handling
Hans Verkuil [Tue, 7 Jun 2011 07:40:04 +0000 (04:40 -0300)]
[media] v4l2-ctrls: simplify error_idx handling

The lower-level prepare functions just set error_idx for each control that
might have an error. The high-level functions will override this with
cs->count in the get and set cases. Only try will keep the error_idx.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ctrls: introduce call_op define
Hans Verkuil [Wed, 25 May 2011 09:04:58 +0000 (06:04 -0300)]
[media] v4l2-ctrls: introduce call_op define

Add the call_op define to safely call the control ops. This also allows
for controls without any ops such as the 'control class' controls.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] DocBook/v4l: Document the new system-wide version behavior
Mauro Carvalho Chehab [Sat, 25 Jun 2011 17:11:52 +0000 (14:11 -0300)]
[media] DocBook/v4l: Document the new system-wide version behavior

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] radio: Use the subsystem version control for VIDIOC_QUERYCAP
Mauro Carvalho Chehab [Sat, 25 Jun 2011 13:15:42 +0000 (10:15 -0300)]
[media] radio: Use the subsystem version control for VIDIOC_QUERYCAP

Just like the video drivers, the right thing to do is to use
the per-subsystem version control.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] Stop using linux/version.h on the remaining video drivers
Mauro Carvalho Chehab [Sat, 25 Jun 2011 14:28:37 +0000 (11:28 -0300)]
[media] Stop using linux/version.h on the remaining video drivers

Standardize the remaining video drivers to return the API version
for the VIDIOC_QUERYCAP version, instead of a per-driver version.

Those drivers had the version updated more recently or are SoC
drivers. Even so, it doesn't sound a good idea to keep a per-driver
version control, so, let's use the per-subsystem version control
instead.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] gspca: don't include linux/version.h
Mauro Carvalho Chehab [Tue, 28 Jun 2011 01:28:38 +0000 (22:28 -0300)]
[media] gspca: don't include linux/version.h

Instead of handling a per-driver driver version, use the
per-subsystem one.

As reviewed by Jean-Francois Moine <moinejf@free.fr>:
- the 'info' may be simplified:

Reviewed-by: Jean-Francois Moine <moinejf@free.fr>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] uvcvideo: Use LINUX_VERSION_CODE for VIDIOC_QUERYCAP
Mauro Carvalho Chehab [Sat, 25 Jun 2011 16:50:37 +0000 (13:50 -0300)]
[media] uvcvideo: Use LINUX_VERSION_CODE for VIDIOC_QUERYCAP

uvcvideo doesn't use vidioc_ioctl2. As the API is changing to use
a common version for all drivers, we need to expliticly fix this
driver.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] sn9c102: Use LINUX_VERSION_CODE for VIDIOC_QUERYCAP
Mauro Carvalho Chehab [Sat, 25 Jun 2011 16:43:22 +0000 (13:43 -0300)]
[media] sn9c102: Use LINUX_VERSION_CODE for VIDIOC_QUERYCAP

sn9c102 doesn't use vidioc_ioctl2. As the API is changing to use
a common version for all drivers, we need to expliticly fix this
driver.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] pvrusb2: Use LINUX_VERSION_CODE for VIDIOC_QUERYCAP
Mauro Carvalho Chehab [Sat, 25 Jun 2011 16:34:24 +0000 (13:34 -0300)]
[media] pvrusb2: Use LINUX_VERSION_CODE for VIDIOC_QUERYCAP

pvrusb2 doesn't use vidioc_ioctl2. As the API is changing to use
a common version for all drivers, we need to expliticly fix this
driver.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] et61x251: Use LINUX_VERSION_CODE for VIDIOC_QUERYCAP
Mauro Carvalho Chehab [Sat, 25 Jun 2011 15:57:04 +0000 (12:57 -0300)]
[media] et61x251: Use LINUX_VERSION_CODE for VIDIOC_QUERYCAP

et61x251 doesn't use vidioc_ioctl2. As the API is changing to use
a common version for all drivers, we need to expliticly fix this
driver.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] ivtv,cx18: Use default version control for VIDIOC_QUERYCAP
Mauro Carvalho Chehab [Sat, 25 Jun 2011 14:23:01 +0000 (11:23 -0300)]
[media] ivtv,cx18: Use default version control for VIDIOC_QUERYCAP

After discussing with Andy Walls on irc, we've agreed that this
is the best thing to do. No regressions will be introduced, as 3.x.y
is greater then the current versions for cx18 and ivtv.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] pwc: Use the default version for VIDIOC_QUERYCAP
Mauro Carvalho Chehab [Sat, 25 Jun 2011 13:07:39 +0000 (10:07 -0300)]
[media] pwc: Use the default version for VIDIOC_QUERYCAP

After discussing with Hans, change pwc to use the default version
control.

The only version ever used for pwc driver is 10.0.12, due to
commit 2b455db6d456ef2d44808a8377fd3bc832e08317.

Changing it to 3.x.y won't conflict with the old version.
There's no namespace conflicts in any predictable future.

Even on the remote far-away case where we might have a conflict,
it will be on just one specific stable Kernel release (Kernel 10.0.12),
if we ever have such stable release.

So, it is safe and consistent on using 3.x.y numering schema for
it.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] Stop using linux/version.h on most video drivers
Mauro Carvalho Chehab [Fri, 24 Jun 2011 17:45:49 +0000 (14:45 -0300)]
[media] Stop using linux/version.h on most video drivers

All the modified drivers didn't have any version increment since
Jan, 1 2011. Several of them didn't have any version increment
for a long time, even having new features and important bug fixes
happening.

As we're now filling the QUERYCAP version with the current Kernel
Release, we don't need to maintain a per-driver version control
anymore. So, let's just use the default.

In order to preserve the Kernel module version history, a
KERNEL_VERSION() macro were added to all modified drivers, and
the extraver number were incremented.

I opted to preserve the per-driver version control to a few
pwc, pvrusb2, s2255, s5p-fimc and sh_vou.

A few drivers are still using the legacy way to handle ioctl's.
So, we can't do such change on them, otherwise, they'll break.
Those are: uvc, et61x251 and sn9c102.

The rationale is that the per-driver version control seems to be
actively maintained on those.

Yet, I think that the better for them would be to just use the
default version numbering, instead of doing that by themselves.

While here, removed a few uneeded include linux/version.h

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] drxd, siano: Remove unused include linux/version.h
Mauro Carvalho Chehab [Sat, 25 Jun 2011 13:39:36 +0000 (10:39 -0300)]
[media] drxd, siano: Remove unused include linux/version.h

Both drxd and siano drivers were including linux/version.h without
any reason. Probably, this is due to some compatibility code that
used to exist before having their support added into the Linux
Kernel.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l2-ioctl: Add a default value for kernel version
Mauro Carvalho Chehab [Fri, 24 Jun 2011 16:14:14 +0000 (13:14 -0300)]
[media] v4l2-ioctl: Add a default value for kernel version

Most drivers don't increase kernel versions as newer features are added or
bug fixes are solved. So, vidioc_querycap returned value for cap->version is
meaningless. Instead of keeping this situation forever, let's add a default
value matching the current Linux version.

Drivers that want to keep their own version control can still do it, as they
can override the default value for cap->version.

Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] DocBook/v4l: Remove references to the old V4L1 compat layer
Mauro Carvalho Chehab [Sat, 25 Jun 2011 17:28:08 +0000 (14:28 -0300)]
[media] DocBook/v4l: Remove references to the old V4L1 compat layer

The old V4L1 Kernel copatibility layer was removed, but the API
spec still says that it is there.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: implement contiguous DMA operation
Jonathan Corbet [Mon, 20 Jun 2011 19:14:40 +0000 (16:14 -0300)]
[media] marvell-cam: implement contiguous DMA operation

The core driver can now operate in either vmalloc or dma-contig modes;
obviously the latter is preferable when it is supported.  Default is
currently vmalloc on all platforms; load the module with buffer_mode=1 for
contiguous DMA mode.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Don't spam the logs on frame loss
Jonathan Corbet [Mon, 20 Jun 2011 19:14:39 +0000 (16:14 -0300)]
[media] marvell-cam: Don't spam the logs on frame loss

The sequence numbers already give that information if user space cares;
this is a frequent occurrence on slower machines, alas.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: no need to initialize the DMA buffers
Jonathan Corbet [Mon, 20 Jun 2011 19:14:38 +0000 (16:14 -0300)]
[media] marvell-cam: no need to initialize the DMA buffers

This was an old debugging thing from years ago.  It's only done at
initialization time, but it's still unnecessary; take it out.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: include file cleanup
Jonathan Corbet [Mon, 20 Jun 2011 19:14:37 +0000 (16:14 -0300)]
[media] marvell-cam: include file cleanup

Put the includes into a slightly more readable ordering and get rid of a
few unneeded ones.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: convert to videobuf2
Jonathan Corbet [Mon, 20 Jun 2011 19:14:36 +0000 (16:14 -0300)]
[media] marvell-cam: convert to videobuf2

This is a basic, naive conversion to the videobuf2 infrastructure, removing
a lot of code in the process.  For now, we're using vmalloc, which is
suboptimal, but it does match what the cafe driver did before.  In the cafe
case, it may have to stay that way just because memory is too tight to do
direct streaming; mmp-camera will be able to do better.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] v4l: Fix minor typos in the documentation
Sylwester Nawrocki [Sat, 18 Jun 2011 08:52:21 +0000 (05:52 -0300)]
[media] v4l: Fix minor typos in the documentation

Signed-off-by: Sylwester Nawrocki <snjw23@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] davinci vpbe: Readme text for Dm6446 vpbe
Manjunath Hadli [Fri, 17 Jun 2011 07:01:36 +0000 (04:01 -0300)]
[media] davinci vpbe: Readme text for Dm6446 vpbe

Please refer to this file for detailed documentation of
davinci vpbe v4l2 driver.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Acked-by: Muralidharan Karicheri <m-karicheri2@ti.com>
Acked-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] davinci vpbe: Build infrastructure for VPBE driver
Manjunath Hadli [Fri, 17 Jun 2011 07:01:35 +0000 (04:01 -0300)]
[media] davinci vpbe: Build infrastructure for VPBE driver

This patch adds the build infra-structure for Davinci
VPBE dislay driver.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Acked-by: Muralidharan Karicheri <m-karicheri2@ti.com>
Acked-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] davinci vpbe: VENC( Video Encoder) implementation
Manjunath Hadli [Fri, 17 Jun 2011 07:01:34 +0000 (04:01 -0300)]
[media] davinci vpbe: VENC( Video Encoder) implementation

This patch adds the VENC or the Video encoder, which is responsible
for the blending of all source planes and timing generation for Video
modes like NTSC, PAL and other digital outputs. the VENC implementation
currently supports COMPOSITE and COMPONENT outputs and NTSC and PAL
resolutions through the analog DACs. The venc block is implemented
as a subdevice, allowing for additional external and internal encoders
of other kind to plug-in.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Acked-by: Muralidharan Karicheri <m-karicheri2@ti.com>
Acked-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] davinci vpbe: OSD(On Screen Display) block
Manjunath Hadli [Fri, 17 Jun 2011 07:01:33 +0000 (04:01 -0300)]
[media] davinci vpbe: OSD(On Screen Display) block

This patch implements the functionality of the OSD block
of the VPBE. The OSD in total supports 4 planes or Video
sources - 2 mainly RGB and 2 Video. The patch implements general
handling of all the planes, with specific emphasis on the Video
plane capabilities as the Video planes are supported through the
V4L2 driver.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Acked-by: Muralidharan Karicheri <m-karicheri2@ti.com>
Acked-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] davinci vpbe: VPBE display driver
Manjunath Hadli [Fri, 17 Jun 2011 07:01:32 +0000 (04:01 -0300)]
[media] davinci vpbe: VPBE display driver

This patch implements the core functionality of the display driver,
mainly controlling the VENC and other encoders, and acting as
the one point interface for the main V4L2 driver. This implements
the core of each of the V4L2 IOCTLs.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Acked-by: Muralidharan Karicheri <m-karicheri2@ti.com>
Acked-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] davinci vpbe: V4L2 display driver for DM644X SoC
Manjunath Hadli [Fri, 17 Jun 2011 07:01:31 +0000 (04:01 -0300)]
[media] davinci vpbe: V4L2 display driver for DM644X SoC

This is the display driver for Texas Instruments's DM644X family
SoC. This patch contains the main implementation of the driver with the
V4L2 interface. The driver implements the streaming model with
support for both kernel allocated buffers and user pointers. It also
implements all of the necessary IOCTLs necessary and supported by the
video display device.

Signed-off-by: Manjunath Hadli <manjunath.hadli@ti.com>
Acked-by: Muralidharan Karicheri <m-karicheri2@ti.com>
Acked-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] DVB: dvb-net, make the kconfig text helpful
Jiri Slaby [Thu, 16 Jun 2011 19:06:56 +0000 (16:06 -0300)]
[media] DVB: dvb-net, make the kconfig text helpful

Telling the user they can disable an option if they want is not the
much useful. Describe what it is good for instead.

The text was derived from Mauro's email.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Mauro Carvalho Chehab <mchehab@redhat.com>
Cc: Hans Petter Selasky <hselasky@c2i.net>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Acked-by: Hans Petter Selasky <hselasky@c2i.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] dvb: remove unnecessary code
Greg Dietsche [Thu, 16 Jun 2011 16:33:35 +0000 (13:33 -0300)]
[media] dvb: remove unnecessary code

remove unnecessary code that matches this coccinelle pattern
if (...)
return ret;
return ret;

Signed-off-by: Greg Dietsche <Gregory.Dietsche@cuw.edu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] dvb-usb/technisat-usb2: don't use flush_scheduled_work()
Tejun Heo [Wed, 15 Jun 2011 13:32:06 +0000 (10:32 -0300)]
[media] dvb-usb/technisat-usb2: don't use flush_scheduled_work()

flush_scheduled_work() is deprecated and scheduled to be removed.
technisat-usb2 already sync-cancels the only work item it uses and
there's no reason for it to call flush_scheduled_work().  Don't use
it.

Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Patrick Boettcher <pboettcher@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] DVB API: Mention the dvbM device
Bjørn Mork [Tue, 14 Jun 2011 12:59:14 +0000 (09:59 -0300)]
[media] DVB API: Mention the dvbM device

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] V4L1 API has been moved into "legacy" on the linuxtv.org site
Bjørn Mork [Tue, 14 Jun 2011 12:42:47 +0000 (09:42 -0300)]
[media] V4L1 API has been moved into "legacy" on the linuxtv.org site

Signed-off-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] tea575x: remove useless input ioctls
Ondrej Zary [Sat, 11 Jun 2011 13:37:29 +0000 (10:37 -0300)]
[media] tea575x: remove useless input ioctls

Remove empty and useless g_input and s_input ioctls.
This fixes one fail of v4l2-compliance test.

Signed-off-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] tea575x: allow multiple opens
Ondrej Zary [Sat, 11 Jun 2011 13:28:59 +0000 (10:28 -0300)]
[media] tea575x: allow multiple opens

Change locking to allow tea575x-radio device to be opened multiple times.

Signed-off-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] radio-timb: Simplified platform data
Richard Röjfors [Fri, 10 Jun 2011 14:48:48 +0000 (11:48 -0300)]
[media] radio-timb: Simplified platform data

This patch simplifies the platform data slightly, by removing
unused elements.

Signed-off-by: Richard Röjfors <richard.rojfors@pelagicore.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] media: DVB_NET must depend on DVB_CORE
Guennadi Liakhovetski [Fri, 10 Jun 2011 10:43:26 +0000 (07:43 -0300)]
[media] media: DVB_NET must depend on DVB_CORE

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] cx23885: Add IR Rx support for HVR-1270 boards
Andy Walls [Thu, 9 Jun 2011 00:24:25 +0000 (21:24 -0300)]
[media] cx23885: Add IR Rx support for HVR-1270 boards

Signed-off-by: Andy Walls <awalls@md.metrocast.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] cx231xx: Add support for Hauppauge WinTV USB2-FM
Peter Moon [Wed, 8 Jun 2011 18:54:19 +0000 (15:54 -0300)]
[media] cx231xx: Add support for Hauppauge WinTV USB2-FM

This patch adds support for the "Hauppauge WinTV USB2-FM" Analog TV Stick.
It includes support for both the PAL and NTSC variants of the device.

Signed-off-by: Peter Moon <pomoon@gmail.com>
Reviewed-by: Devin Heitmueller <dheitmueller@hauppauge.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] frontends/s5h1420: Change: clock_settting to clock_setting
Justin P. Mattock [Mon, 6 Jun 2011 16:11:52 +0000 (13:11 -0300)]
[media] frontends/s5h1420: Change: clock_settting to clock_setting

Changes clock_settting to clock_setting.

Note: This could be intentionally set this way from the beginning and/or
is a typo.

Signed-off-by: Justin P. Mattock <justinmattock@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] cx88: replaced duplicated code with function call
istvan_v@mailbox.hu [Tue, 7 Jun 2011 16:21:02 +0000 (13:21 -0300)]
[media] cx88: replaced duplicated code with function call

The following patch replaces code to reset the XC3028 tuner with a call
to the tuner reset callback.

Signed-off-by: Istvan Varga <istvan_v@mailbox.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] cx88: added support for Leadtek WinFast DTV1800 H with XC4000 tuner
istvan_v@mailbox.hu [Tue, 7 Jun 2011 16:16:56 +0000 (13:16 -0300)]
[media] cx88: added support for Leadtek WinFast DTV1800 H with XC4000 tuner

This patch implements support for the Leadtek WinFast DTV1800 H card with
XC4000 tuner (107d:6f38).

Signed-off-by: Istvan Varga <istvan_v@mailbox.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] cx88: added support for Leadtek WinFast DTV2000 H Plus
istvan_v@mailbox.hu [Tue, 7 Jun 2011 16:14:53 +0000 (13:14 -0300)]
[media] cx88: added support for Leadtek WinFast DTV2000 H Plus

This patch implements support for the Leadtek WinFast DTV2000 H Plus card
with XC4000 tuner (107d:6f42).

Signed-off-by: Istvan Varga <istvan_v@mailbox.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] cx88: added XC4000 tuner callback and DVB attach functions
istvan_v@mailbox.hu [Tue, 7 Jun 2011 16:12:29 +0000 (13:12 -0300)]
[media] cx88: added XC4000 tuner callback and DVB attach functions

Added functions for XC4000 tuner reset and attaching DVB frontend.

Signed-off-by: Istvan Varga <istvan_v@mailbox.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx: Mark Kworld 305 as validated
Mauro Carvalho Chehab [Sat, 18 Jun 2011 02:18:21 +0000 (23:18 -0300)]
[media] em28xx: Mark Kworld 305 as validated

This board were used for testing the em28xx-alsa using a separate interface.
So, it is obviously validated ;)

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx: Add support for devices with a separate audio interface
Mauro Carvalho Chehab [Fri, 17 Jun 2011 18:15:12 +0000 (15:15 -0300)]
[media] em28xx: Add support for devices with a separate audio interface

Some devices use a separate interface for the vendor audio class.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx-audio: Some Alsa API fixes
Mauro Carvalho Chehab [Sun, 19 Jun 2011 16:39:31 +0000 (13:39 -0300)]
[media] em28xx-audio: Some Alsa API fixes

Mark the alsa stream with SNDRV_PCM_INFO_BATCH,
as the timing to get audio streams can vary.

Also, add SNDRV_PCM_TRIGGER for pause/release.

while here, fix the stop indicator, to be sure that audio
will be properly released at the stop events.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx-audio: Properly report failures to start stream
Mauro Carvalho Chehab [Sun, 19 Jun 2011 13:15:35 +0000 (10:15 -0300)]
[media] em28xx-audio: Properly report failures to start stream

If the audio stream fails for any reason, it should:
1) Report an error via dmesg;
2) Mark internally that the stream didn't started.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx-audio: add debug info for the volume control
Mauro Carvalho Chehab [Sat, 18 Jun 2011 14:00:20 +0000 (11:00 -0300)]
[media] em28xx-audio: add debug info for the volume control

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx-audio: volumes are inverted
Mauro Carvalho Chehab [Sun, 19 Jun 2011 16:14:07 +0000 (13:14 -0300)]
[media] em28xx-audio: volumes are inverted

While here, fix volume mask to 5 bits

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx-audio: add support for mute controls
Mauro Carvalho Chehab [Sun, 19 Jun 2011 16:08:46 +0000 (13:08 -0300)]
[media] em28xx-audio: add support for mute controls

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx-alsa: add mixer support for AC97 volume controls
Mauro Carvalho Chehab [Sun, 19 Jun 2011 16:06:40 +0000 (13:06 -0300)]
[media] em28xx-alsa: add mixer support for AC97 volume controls

Export ac97 volume controls via mixer.

Pulseaudio will probably handle it very badly, as it has
no idea about how volumes are wired, and how are they
associated with each TV input. Those wirings are
card model dependent, and we don't have the wiring mappings
for each supported device.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx: Allow to compile it without RC/input support
Mauro Carvalho Chehab [Sat, 18 Jun 2011 13:19:11 +0000 (10:19 -0300)]
[media] em28xx: Allow to compile it without RC/input support

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx: Fix a wrong enum at the ac97 control tables
Mauro Carvalho Chehab [Sat, 18 Jun 2011 10:02:49 +0000 (07:02 -0300)]
[media] em28xx: Fix a wrong enum at the ac97 control tables

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] em28xx: Don't initialize a var if won't be using it
Mauro Carvalho Chehab [Fri, 17 Jun 2011 23:28:51 +0000 (20:28 -0300)]
[media] em28xx: Don't initialize a var if won't be using it

Fixes most cases of initializing a var but not using it.

There are still 3 cases at em28xx-alsa, were those vars should
probably be used.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Basic working MMP camera driver
Jonathan Corbet [Sat, 11 Jun 2011 17:46:49 +0000 (14:46 -0300)]
[media] marvell-cam: Basic working MMP camera driver

Now we have a camera working over the marvell cam controller core.  It
works like the cafe driver and has all the same limitations, contiguous DMA
only being one of them.  But it's a start.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Allocate the i2c adapter in the platform driver
Jonathan Corbet [Sat, 11 Jun 2011 17:46:48 +0000 (14:46 -0300)]
[media] marvell-cam: Allocate the i2c adapter in the platform driver

The upcoming mmp-camera driver will need an i2c_adapter structure allocated
externally, so change the core adapter to a pointer and require the
platform code to fill it in.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Right-shift i2c slave ID's in the cafe driver
Jonathan Corbet [Sat, 11 Jun 2011 17:46:47 +0000 (14:46 -0300)]
[media] marvell-cam: Right-shift i2c slave ID's in the cafe driver

This makes the cafe i2c implement consistent with the rest of Linux so that
the core can use the same slave ID everywhere.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Move Cafe-specific register definitions to cafe-driver.c
Jonathan Corbet [Sat, 11 Jun 2011 17:46:46 +0000 (14:46 -0300)]
[media] marvell-cam: Move Cafe-specific register definitions to cafe-driver.c

Nobody else ever needs to see these, so let's hide them.

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Remove the "untested" comment
Jonathan Corbet [Sat, 11 Jun 2011 17:46:45 +0000 (14:46 -0300)]
[media] marvell-cam: Remove the "untested" comment

This code is, indeed, tested :)

Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Pass sensor parameters from the platform
Jonathan Corbet [Sat, 11 Jun 2011 17:46:44 +0000 (14:46 -0300)]
[media] marvell-cam: Pass sensor parameters from the platform

Depending on the controller, the ov7670 sensor may be told to work with a
different clock speed or to use the SMBUS protocol.  Remove the wired-in
code and pass that information from the platform layer.  The Cafe driver
now just assumes it's running on an OLPC XO 1; I do not believe it has ever
run anywhere else.

Cc: Daniel Drake <dsd@laptop.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Separate out the Marvell camera core
Jonathan Corbet [Sat, 11 Jun 2011 17:46:43 +0000 (14:46 -0300)]
[media] marvell-cam: Separate out the Marvell camera core

There will eventually be multiple users of the core camera controller, so
separate it from the bus/platform/i2c stuff.  I've tried to do the minimal
set of changes to get the driver functioning in this configuration; I did
clean up a bunch of old checkpatch gripes in the process.  This driver
works like the old one did on OLPC XO 1 systems.

Cc: Daniel Drake <dsd@laptop.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] marvell-cam: Move cafe-ccic into its own directory
Jonathan Corbet [Sat, 11 Jun 2011 17:46:42 +0000 (17:46 +0000)]
[media] marvell-cam: Move cafe-ccic into its own directory

This driver will soon become a family of drivers, so let's give it its own
place to live.  This move requires putting ov7670.h into include/media, but
there are no code changes.

Cc: Daniel Drake <dsd@laptop.org>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] DocBook: Use base64 for gif/png files
Mauro Carvalho Chehab [Thu, 16 Jun 2011 02:13:19 +0000 (23:13 -0300)]
[media] DocBook: Use base64 for gif/png files

The patch utility doesn't work with non-binary files. This causes some
tools to break, like generating tarball targets and the scripts that
generate diff patches at http://www.kernel.org/pub/linux/kernel/v2.6/.

So, let's convert all binaries to ascii using base64, and add a
logic at Makefile to convert them back into binaries at runtime.

Reported-by: Randy Dunlap <randy.dunlap@oracle.com>
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Michal Marek <mmarek@suse.cz>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] drivers/media/video: fix memory leak of snd_cx18_init()
Andre Bartke [Fri, 3 Jun 2011 18:06:58 +0000 (15:06 -0300)]
[media] drivers/media/video: fix memory leak of snd_cx18_init()

cxsc is not freed in the error case.

Signed-off-by: Andre Bartke <andre.bartke@gmail.com>
Cc: Andy Walls <awalls@md.metrocast.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] Correct and add some parameter descriptions
Hans Petter Selasky [Thu, 26 May 2011 08:06:09 +0000 (05:06 -0300)]
[media] Correct and add some parameter descriptions

>From 70d02ce19f64fae4ceee563501e8462a76e17b91 Mon Sep 17 00:00:00 2001

Signed-off-by: Hans Petter Selasky <hselasky@c2i.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] Remove unused definitions which can cause conflict with definitions in usb...
Hans Petter Selasky [Thu, 26 May 2011 07:56:55 +0000 (04:56 -0300)]
[media] Remove unused definitions which can cause conflict with definitions in usb/ch9.h

>From 0dd2949dfeae431ed4ffbd00fd14a10dc3747ad0 Mon Sep 17 00:00:00 2001

Signed-off-by: Hans Petter Selasky <hselasky@c2i.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] Correct error code from -ENOMEM to -EINVAL.
Hans Petter Selasky [Thu, 26 May 2011 07:49:52 +0000 (04:49 -0300)]
[media] Correct error code from -ENOMEM to -EINVAL.

Make sure the return value is set in all cases

>From 9b38a5c9878b5e4be2899ae291c4524f5f5fc218 Mon Sep 17 00:00:00 2001

Signed-off-by: Hans Petter Selasky <hselasky@c2i.net>
Cc: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
13 years ago[media] rc-core: lirc use unsigned int
David Härdeman [Thu, 28 Apr 2011 15:13:58 +0000 (12:13 -0300)]
[media] rc-core: lirc use unsigned int

Durations can never be negative, so it makes sense to consistently use
unsigned int for LIRC transmission. Contrary to the initial impression,
this shouldn't actually change the userspace API.

Signed-off-by: David Härdeman <david@hardeman.nu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>