openwrt/staging/blogic.git
12 years ago[media] radio-si470x: Don't unnecesarily read registers on G_TUNER
Hans de Goede [Thu, 14 Jun 2012 12:43:11 +0000 (09:43 -0300)]
[media] radio-si470x: Don't unnecesarily read registers on G_TUNER

Reading registers from the pcear USB dongles with the si470x causes a
loud pop (and an alsa buffer overrun). Since most radio apps periodically
call G_TUNER to update mono/stereo, signal and afc status this leads
to the music . pop . music . pop . music -> not good.

On the internet there is an howto for flashing the pcear with a newer
firmware from the silabs reference boardto fix this, but:
1) This howto relies on a special version of the driver which allows
   firmware flashing
2) We should try to avoid the answer to a bug report being upgrade your
   firmware, if at all possible
3) Windows does not suffer from the pop sounds

After a quick look at the driver I found at that the register reads are
not necessary at all, as the device gives us the necessary status through
usb interrupt packets, and the driver already uses these!

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] radio-shark: New driver for the Griffin radioSHARK USB radio receiver
Hans de Goede [Mon, 21 May 2012 18:24:50 +0000 (15:24 -0300)]
[media] radio-shark: New driver for the Griffin radioSHARK USB radio receiver

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] snd_tea575x: Add a cannot_mute flag
Hans de Goede [Fri, 18 May 2012 15:21:57 +0000 (12:21 -0300)]
[media] snd_tea575x: Add a cannot_mute flag

Some devices which use the tea575x tuner chip don't allow direct control
over the IO pins, and thus cannot mute the audio output.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
CC: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] snd_tea575x: Add write_/read_val operations
Hans de Goede [Fri, 18 May 2012 12:36:17 +0000 (09:36 -0300)]
[media] snd_tea575x: Add write_/read_val operations

Some devices which use the tea575x tuner chip don't allow bit banging the
lines, instead they offer a method to directly set / get the contents of the
25 bit shift-register in the chip. Notably the Griffin radioSHARK USB radio
receiver does this.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
CC: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] Fix VIDIOC_TRY_EXT_CTRLS regression
Hans Verkuil [Wed, 18 Jul 2012 12:34:59 +0000 (09:34 -0300)]
[media] Fix VIDIOC_TRY_EXT_CTRLS regression

Fixes an omission in the new v4l2_ioctls table: VIDIOC_TRY_EXT_CTRLS
must get the INFO_FL_CTRL flag, just like all the other control
related ioctls, otherwise the ioctl core won't know it also has
to check whether v4l2_fh->ctrl_handler is non-zero before it can
decide that this ioctl is not implemented.

Caught by v4l2-compliance while I was testing the mem2mem_testdev driver.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years agoMerge branch 'v4l_for_linus' into staging/for_v3.6
Mauro Carvalho Chehab [Sat, 14 Jul 2012 18:45:44 +0000 (15:45 -0300)]
Merge branch 'v4l_for_linus' into staging/for_v3.6

* v4l_for_linus: (31 commits)
  [media] Revert "[media] V4L: JPEG class documentation corrections"
  [media] s5p-fimc: Add missing FIMC-LITE file operations locking
  [media] omap3isp: preview: Fix contrast and brightness handling
  [media] omap3isp: preview: Fix output size computation depending on input format
  [media] winbond-cir: Initialise timeout, driver_type and allowed_protos
  [media] winbond-cir: Fix txandrx module info
  [media] cx23885: Silence unknown command warnings
  [media] cx23885: add support for HVR-1255 analog (cx23888 variant)
  [media] cx23885: make analog support work for HVR_1250 (cx23885 variant)
  [media] cx25840: fix vsrc/hsrc usage on cx23888 designs
  [media] cx25840: fix regression in HVR-1800 analog audio
  [media] cx25840: fix regression in analog support hue/saturation controls
  [media] cx25840: fix regression in HVR-1800 analog support
  [media] s5p-mfc: Fixed setup of custom controls in decoder and encoder
  [media] cx231xx: don't DMA to random addresses
  [media] em28xx: fix em28xx-rc load
  [media] dvb-core: Release semaphore on error path dvb_register_device()
  [media] s5p-fimc: Stop media entity pipeline if fimc_pipeline_validate fails
  [media] s5p-fimc: Fix compiler warning in fimc-lite.c
  [media] s5p-fimc: media_entity_pipeline_start() may fail
  ...

12 years agov4l: Export v4l2-common.h in include/linux/Kbuild
Sakari Ailus [Mon, 9 Jul 2012 09:10:26 +0000 (12:10 +0300)]
v4l: Export v4l2-common.h in include/linux/Kbuild

v4l2-common.h is a header file that's used in user space, thus it must be
exported using header-y.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Tested-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years agomedia: Revert "[media] Terratec Cinergy S2 USB HD Rev.2"
Mauro Carvalho Chehab [Sat, 7 Jul 2012 03:19:20 +0000 (00:19 -0300)]
media: Revert "[media] Terratec Cinergy S2 USB HD Rev.2"

This reverts commit 03228792df67a6ae231960151b8a5c87e17a73e2.

drivers/media/dvb/dvb-usb/dw2102.c:1478:18: error: ‘USB_PID_TEVII_S421’ undeclared here (not in a function)
drivers/media/dvb/dvb-usb/dw2102.c:1479:18: error: ‘USB_PID_TEVII_S632’ undeclared here (not in a function)

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] Revert "[media] V4L: JPEG class documentation corrections"
Sylwester Nawrocki [Wed, 27 Jun 2012 13:12:31 +0000 (10:12 -0300)]
[media] Revert "[media] V4L: JPEG class documentation corrections"

This reverts commit feed0258e11e04b7e0, as the same issues
are already covered in another version of that patch that
was also applied (579e92ffac65c717c9c8a50feb755a).

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] s5p-fimc: Add missing FIMC-LITE file operations locking
Sylwester Nawrocki [Mon, 4 Jun 2012 16:15:56 +0000 (13:15 -0300)]
[media] s5p-fimc: Add missing FIMC-LITE file operations locking

commit 5126f2590bee412e3053de851cb07f531e4be36a
"v4l2-dev: add flag to have the core lock all file operations"
introduced an additional bit flag (V4L2_FL_LOCK_ALL_FOPS) that
should be set by drivers that use the v4l2 core lock for all file
operations. Since this driver has been merged at the same time as
the core changes it doesn't set this flags and thus its all file
operations except IOCTL are not properly serialized. Fix this by
adding file ops locking in the driver.

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] media: Use pr_info not homegrown pr_reg macro
Joe Perches [Sun, 20 May 2012 21:45:15 +0000 (18:45 -0300)]
[media] media: Use pr_info not homegrown pr_reg macro

No need to duplicate normal kernel logging capabilities.

Add pr_fmt and convert pr_reg to pr_info.
Remove pr_reg macros.

Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] Terratec Cinergy S2 USB HD Rev.2
Igor M. Liplianin [Tue, 8 May 2012 15:28:47 +0000 (12:28 -0300)]
[media] Terratec Cinergy S2 USB HD Rev.2

Terratec Cinergy S2 USB HD Rev.2 support.

Signed-off-by: Igor M. Liplianin <liplianin@me.by>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l: Correct conflicting V4L2 subdev selection API documentation
Sakari Ailus [Fri, 15 Jun 2012 13:22:25 +0000 (10:22 -0300)]
[media] v4l: Correct conflicting V4L2 subdev selection API documentation

The API reference documents that the KEEP_CONFIG flag tells the
configuration should not be propagated by the driver whereas the interface
documentation (dev-subdev.xml) prohibited any changes to the rest of the
pipeline. Resolve the conflict by changing the API reference to disallow
changes.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] Feature removal: V4L2 selections API target and flag definitions
Sylwester Nawrocki [Mon, 25 Jun 2012 19:14:41 +0000 (16:14 -0300)]
[media] Feature removal: V4L2 selections API target and flag definitions

After unification of the V4L2 and V4L2 subdev selection targets and flags
the old flags are no longer required. Thus they can be removed. However, as
the API is present in a kernel release, this must go through the feature
removal process.

Signed-off-by: Sylwester Nawrocki <sylvester.nawrocki@gmail.com>
Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l: Unify selection flags documentation
Sakari Ailus [Wed, 13 Jun 2012 19:01:10 +0000 (16:01 -0300)]
[media] v4l: Unify selection flags documentation

As for the selection targets, the selection flags are now the same on V4L2
and V4L2 subdev interfaces. Also document them so.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l: Unify selection flags
Sakari Ailus [Wed, 13 Jun 2012 19:01:10 +0000 (16:01 -0300)]
[media] v4l: Unify selection flags

Unify flags on the selection interfaces on V4L2 and V4L2 subdev. Flags are
very similar to targets in this case: there are more similarities than
differences between the two interfaces.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l: Common documentation for selection targets
Sakari Ailus [Fri, 1 Jun 2012 16:56:53 +0000 (13:56 -0300)]
[media] v4l: Common documentation for selection targets

Both V4L2 and V4L2 subdev interface have very similar selection APIs with
differences foremost related to in-memory and media bus formats. However,
the selection targets are the same for both. Most targets are and in the
future will likely continue to be more the same than with any differences.
Thus it makes sense to unify the documentation of the targets.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l: Unify selection targets across V4L2 and V4L2 subdev interfaces
Sakari Ailus [Fri, 18 May 2012 12:31:18 +0000 (09:31 -0300)]
[media] v4l: Unify selection targets across V4L2 and V4L2 subdev interfaces

Change the users of V4L2_SUBDEV_SEL_TGT_* targets to use V4L2_SEL_TGT_*
instead. The common definitions are moved to a new header file,
include/linux/v4l2-common.h.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l: Remove "_ACTUAL" from subdev selection API target definition names
Sakari Ailus [Thu, 17 May 2012 20:50:45 +0000 (17:50 -0300)]
[media] v4l: Remove "_ACTUAL" from subdev selection API target definition names

The string "_ACTUAL" does not say anything more about the target names. Drop
it. V4L2 selection API was changed by "V4L: Remove "_ACTIVE" from the
selection target name definitions" by Sylwester Nawrocki. This patch does
the same for the V4L2 subdev API.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] V4L: Remove "_ACTIVE" from the selection target name definitions
Sylwester Nawrocki [Sun, 20 May 2012 14:17:12 +0000 (11:17 -0300)]
[media] V4L: Remove "_ACTIVE" from the selection target name definitions

This patch drops the _ACTIVE part from the selection target names
as a prerequisite to unify the selection target names across the subdev
and regular video node API.

The meaning of V4L2_SEL_TGT_*_ACTIVE and V4L2_SUBDEV_SEL_TGT_*_ACTUAL
selection targets is logically the same. Different names add to confusion
where both APIs are used in a single driver or an application. For some
system configurations different names may lead to interoperability issues.

For backwards compatibility V4L2_SEL_TGT_CROP_ACTIVE and
V4L2_SEL_TGT_COMPOSE_ACTIVE are defined as aliases to V4L2_SEL_TGT_CROP
and V4L2_SEL_TGT_COMPOSE. These aliases will be removed after deprecation
period, according to Documentation/feature-removal-schedule.txt.

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] omap3isp: preview: Fix contrast and brightness handling
Laurent Pinchart [Fri, 6 Jul 2012 11:41:25 +0000 (08:41 -0300)]
[media] omap3isp: preview: Fix contrast and brightness handling

Commit bac387efbb88cf0e8df6f46a38387897cea464ee ("omap3isp: preview:
Simplify configuration parameters access") added three fields to the
preview_update structure, but failed to properly update the related
initializers. Fix this.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] omap3isp: preview: Fix output size computation depending on input format
Laurent Pinchart [Thu, 5 Jul 2012 17:31:33 +0000 (14:31 -0300)]
[media] omap3isp: preview: Fix output size computation depending on input format

The preview engine crops 4 columns and 4 lines when CFA is enabled.
Commit b2da46e52fe7871cba36e1a435844502c0eccf39 ("omap3isp: preview: Add
support for greyscale input") inverted the condition by mistake, fix
this.

Reported-by: Florian Neuhaus <florian.neuhaus@reberinformatik.ch>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Florian Neuhaus <florian.neuhaus@reberinformatik.ch>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] winbond-cir: Initialise timeout, driver_type and allowed_protos
Anton Blanchard [Mon, 2 Jul 2012 00:58:52 +0000 (21:58 -0300)]
[media] winbond-cir: Initialise timeout, driver_type and allowed_protos

We need to set a timeout so we can go idle on no activity.

Signed-off-by: Anton Blanchard <anton@samba.org>
Acked-by: David Härdeman <david@hardeman.nu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] winbond-cir: Fix txandrx module info
Anton Blanchard [Mon, 2 Jul 2012 00:58:00 +0000 (21:58 -0300)]
[media] winbond-cir: Fix txandrx module info

We aren't getting any module info for the txandx option because
of a typo:

parm:           txandrx:bool

Signed-off-by: Anton Blanchard <anton@samba.org>
Acked-by: David Härdeman <david@hardeman.nu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx23885: Silence unknown command warnings
Anton Blanchard [Mon, 2 Jul 2012 00:38:03 +0000 (21:38 -0300)]
[media] cx23885: Silence unknown command warnings

I am seeing a constant stream of warnings on my cx23885 based card:
cx23885_tuner_callback(): Unknown command 0x2.

Add a check in cx23885_tuner_callback to silence it.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx23885: add support for HVR-1255 analog (cx23888 variant)
Devin Heitmueller [Sun, 1 Jul 2012 19:15:14 +0000 (16:15 -0300)]
[media] cx23885: add support for HVR-1255 analog (cx23888 variant)

Get the HVR-1255 analog support working for all supported inputs.  This
includes introduction of a new board profile for an OEM variant which
doesn't have all the same inputs as the retail version of the board.

Validated with the following boards:

HVR-1255 (0070:2259)

Thanks to Steven Toth and Hauppauge for loaning me various boards to
regression test with.

Thanks-to: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Devin Heitmueler <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx23885: make analog support work for HVR_1250 (cx23885 variant)
Devin Heitmueller [Sun, 1 Jul 2012 19:15:13 +0000 (16:15 -0300)]
[media] cx23885: make analog support work for HVR_1250 (cx23885 variant)

The analog support in the cx23885 driver was completely broken for the
HVR-1250.  Add the necessary code.

Note that this only implements analog for the composite and s-video
inputs. The tuner input continues to be non-functional due to a lack of
analog support in the mt2131 driver.

Validated with the following boards:

HVR-1250 (0070:7911)

Thanks to Steven Toth and Hauppauge for loaning me various boards to
regression test with.

Thanks-to: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Devin Heitmueler <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx25840: fix vsrc/hsrc usage on cx23888 designs
Devin Heitmueller [Sun, 1 Jul 2012 19:15:12 +0000 (16:15 -0300)]
[media] cx25840: fix vsrc/hsrc usage on cx23888 designs

The location of the vsrc/hsrc registers moved in the cx23888, causing
the s_mbus call to fail prematurely indicating that "720x480 is not a
valid size". The function bailed out before many pertinent registers
were set related to the scaler (causing unexpected results in video
rendering when doing raw video capture).

Use the correct registers for the cx23888.

Validated with the following boards:

HVR-1800 retail (0070:7801)
HVR-1800 OEM (0070:7809)
HVR-1850 retail (0070:8541)

Thanks to Steven Toth and Hauppauge for loaning me various boards to
regression test with.

Reported-by: Jonathan <sitten74490@mypacks.net>
Thanks-to: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Devin Heitmueler <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx25840: fix regression in HVR-1800 analog audio
Devin Heitmueller [Sun, 1 Jul 2012 19:15:11 +0000 (16:15 -0300)]
[media] cx25840: fix regression in HVR-1800 analog audio

The refactoring of the cx25840 driver to support the cx23888 caused breakage
with the existing support for cx23885/cx23887 analog audio support.  Tweak
the code so that it only uses the code if it really is a cx23888 instead of
applying it to all cx2388x based devices.

Validated with the following boards:

HVR-1800 retail (0070:7801)
HVR-1800 OEM (0070:7809)
HVR_1850 retail (0070:8541)

Thanks to Steven Toth and Hauppauge for loaning me various boards to
regression test with.

Reported-by: Jonathan <sitten74490@mypacks.net>
Thanks-to: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Devin Heitmueler <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx25840: fix regression in analog support hue/saturation controls
Devin Heitmueller [Sun, 1 Jul 2012 19:15:10 +0000 (16:15 -0300)]
[media] cx25840: fix regression in analog support hue/saturation controls

Fix regression in HVR-1800 analog support hue/saturation controls.

The changes made for the cx23888 caused regressions in the analog
support for cx23885/cx23887 based boards (partly due to changes in the
locations of the hue/saturation controls).  As a result the wrong
registers were being overwritten.

Add code to use the correct registers if it's a cx23888

Validated with the following boards:

HVR-1800 retail (0070:7801)
HVR-1800 OEM (0070:7809)
HVR-1850 retail (0070:8541)

Thanks to Steven Toth and Hauppauge for loaning me various boards to
regression test  with.

Reported-by: Jonathan <sitten74490@mypacks.net>
Thanks-to: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Devin Heitmueler <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx25840: fix regression in HVR-1800 analog support
Devin Heitmueller [Sun, 1 Jul 2012 19:15:09 +0000 (16:15 -0300)]
[media] cx25840: fix regression in HVR-1800 analog support

The refactoring of the cx25840 driver to support the cx23888 caused breakage
with the existing support for cx23885/cx23887 analog support.  Rework the
routines such that the new code is only used for the 888.

Validated with the following boards:

HVR-1800 retail (0070:7801)
HVR-1800 OEM (0070:7809)
HVR_1850 retail (0070:8541)

Thanks to Steven Toth and Hauppauge for loaning me various boards to
regression test with.

Reported-by: Jonathan <sitten74490@mypacks.net>
Thanks-to: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Devin Heitmueler <dheitmueller@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] s5p-mfc: Fixed setup of custom controls in decoder and encoder
Kamil Debski [Tue, 26 Jun 2012 07:44:40 +0000 (04:44 -0300)]
[media] s5p-mfc: Fixed setup of custom controls in decoder and encoder

Fixed bugs in functions that initialize custom controls.

Signed-off-by: Kamil Debski <k.debski@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx231xx: don't DMA to random addresses
David Dillow [Mon, 18 Jun 2012 03:15:21 +0000 (00:15 -0300)]
[media] cx231xx: don't DMA to random addresses

Commit 7a6f6c29d264cdd2fe0eb3d923217eed5f0ad134 (cx231xx: use
URB_NO_TRANSFER_DMA_MAP) was intended to avoid mapping the DMA buffer
for URB twice. This works for the URBs allocated with usb_alloc_urb(),
as those are allocated from cohernent DMA pools, but the flag was also
added for the VBI and audio URBs, which have a manually allocated area.
This leaves the random trash in the structure after allocation as the
DMA address, corrupting memory and preventing VBI and audio from
working. Letting the USB core map the buffers solves the problem.

Cc: stable@kernel.org
Signed-off-by: David Dillow <dave@thedillows.org>
Cc: Sri Deevi <srinivasa.deevi@conexant.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] em28xx: fix em28xx-rc load
Mauro Carvalho Chehab [Fri, 6 Jul 2012 01:52:34 +0000 (22:52 -0300)]
[media] em28xx: fix em28xx-rc load

The logic that checks if a device has remote control is wrong.
Due to that, the em28xx RC module is not loaded by default.

Fix the logic, in order to make it work properly.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] dvb-core: Release semaphore on error path dvb_register_device()
Santosh Nayak [Sat, 23 Jun 2012 10:59:54 +0000 (07:59 -0300)]
[media] dvb-core: Release semaphore on error path dvb_register_device()

There is a missing "up_write()" here. Semaphore should be released
before returning error value.

Cc: stable@kernel.org
Signed-off-by: Santosh Nayak <santoshprasadnayak@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] media: dvb-usb: print mac address via native %pM
Andy Shevchenko [Fri, 6 Jul 2012 14:31:51 +0000 (11:31 -0300)]
[media] media: dvb-usb: print mac address via native %pM

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] s5p-tv: Use module_i2c_driver in sii9234_drv.c file
Sachin Kamat [Wed, 4 Jul 2012 05:33:15 +0000 (02:33 -0300)]
[media] s5p-tv: Use module_i2c_driver in sii9234_drv.c file

module_i2c_driver makes the code simpler by eliminating module_init
and module_exit calls.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Acked-by: Tomasz Stanislawski <t.stanislaws@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] media: gpio-ir-recv: add allowed_protos for platform data
Du, Changbin [Tue, 3 Jul 2012 09:27:19 +0000 (06:27 -0300)]
[media] media: gpio-ir-recv: add allowed_protos for platform data

It's better to give platform code a chance to specify the allowed
protocols to use.

[mchehab@redhat.com: fix merge conflict with a patch that made
 half of this change]
Signed-off-by: Du, Changbin <changbin.du@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] s5p-jpeg: Use module_platform_driver in jpeg-core.c file
Sachin Kamat [Tue, 3 Jul 2012 08:54:33 +0000 (05:54 -0300)]
[media] s5p-jpeg: Use module_platform_driver in jpeg-core.c file

module_platform_driver makes the code simpler by eliminating module_init
and module_exit calls.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Acked-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] saa7134: fix spelling of detach in label
Peter Meerwald [Mon, 2 Jul 2012 20:27:41 +0000 (17:27 -0300)]
[media] saa7134: fix spelling of detach in label

Signed-off-by: Peter Meerwald <pmeerw@pmeerw.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88-blackbird: replace ioctl by unlocked_ioctl
Hans Verkuil [Fri, 25 May 2012 15:30:18 +0000 (12:30 -0300)]
[media] cx88-blackbird: replace ioctl by unlocked_ioctl

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: don't use current_norm
Hans Verkuil [Fri, 25 May 2012 15:15:30 +0000 (12:15 -0300)]
[media] cx88: don't use current_norm

current_norm can only be used if there is a single device node since it is
local to the device node. In this case multiple device nodes share a single
tuner.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: fix a number of v4l2-compliance violations
Hans Verkuil [Fri, 25 May 2012 15:04:10 +0000 (12:04 -0300)]
[media] cx88: fix a number of v4l2-compliance violations

- missing COMPRESSED flag for MPEG formats
- set colorspace
- set sizeimage
- add tuner index checks
- setup the frequency ranges correctly
- add missing g_chip_ident ioctl
- fix audmode handling
- don't handle vbi formats on a video node and vice versa.

cx88 now passes the v4l2-compliance tests.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: support control events
Hans Verkuil [Fri, 11 May 2012 14:25:03 +0000 (11:25 -0300)]
[media] cx88: support control events

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: add priority support
Hans Verkuil [Fri, 11 May 2012 13:57:59 +0000 (10:57 -0300)]
[media] cx88: add priority support

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: move fmt, width and height to cx8800_dev
Hans Verkuil [Fri, 11 May 2012 13:45:18 +0000 (10:45 -0300)]
[media] cx88: move fmt, width and height to cx8800_dev

These are global properties and do not belong in the filehandle struct.

Note: the overlay related fields were just removed: they weren't used at all.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: remove radio and type from cx8800_fh
Hans Verkuil [Fri, 11 May 2012 13:33:25 +0000 (10:33 -0300)]
[media] cx88: remove radio and type from cx8800_fh

This information is available elsewhere already.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: convert cx88-blackbird to the control framework
Hans Verkuil [Mon, 2 Jul 2012 11:47:11 +0000 (08:47 -0300)]
[media] cx88: convert cx88-blackbird to the control framework

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: each device node gets the right controls
Hans Verkuil [Fri, 11 May 2012 12:07:45 +0000 (09:07 -0300)]
[media] cx88: each device node gets the right controls

radio only sees audio controls, video sees video and audio and vbi
sees none.

Also disable the chroma_agc control if secam is selected.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: first phase to convert cx88 to the control framework
Hans Verkuil [Sun, 10 Jun 2012 10:39:52 +0000 (07:39 -0300)]
[media] cx88: first phase to convert cx88 to the control framework

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx88: fix querycap
Hans Verkuil [Wed, 9 May 2012 19:23:07 +0000 (16:23 -0300)]
[media] cx88: fix querycap

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-framework.txt: Update the locking documentation
Hans Verkuil [Mon, 2 Jul 2012 08:43:34 +0000 (05:43 -0300)]
[media] v4l2-framework.txt: Update the locking documentation

This documents the new queue->lock and how to use it. It also removes the
documentation of v4l2_disable_ioctl_locking: this is only used in gspca and
will be removed once gspca has been converted to vb2.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] pwc: v4l2-compliance fixes
Hans Verkuil [Sun, 10 Jun 2012 07:34:10 +0000 (04:34 -0300)]
[media] pwc: v4l2-compliance fixes

- add device_caps
- set colorspace
- s_parm should support a fps of 0 (== reset to nominal fps)

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] pwc: use the new vb2 helpers
Hans Verkuil [Mon, 2 Jul 2012 08:51:58 +0000 (05:51 -0300)]
[media] pwc: use the new vb2 helpers

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-dev.c: also add debug support for the fops
Hans Verkuil [Sat, 9 Jun 2012 15:13:29 +0000 (12:13 -0300)]
[media] v4l2-dev.c: also add debug support for the fops

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] vivi: add create_bufs/preparebuf support
Hans Verkuil [Fri, 22 Jun 2012 08:53:31 +0000 (05:53 -0300)]
[media] vivi: add create_bufs/preparebuf support

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] vivi: use vb2 helper functions
Hans Verkuil [Fri, 22 Jun 2012 08:53:02 +0000 (05:53 -0300)]
[media] vivi: use vb2 helper functions

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] vivi: embed struct video_device instead of allocating it
Hans Verkuil [Sat, 9 Jun 2012 14:27:43 +0000 (11:27 -0300)]
[media] vivi: embed struct video_device instead of allocating it

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] vivi: remove pointless g/s_std support
Hans Verkuil [Fri, 22 Jun 2012 08:49:33 +0000 (05:49 -0300)]
[media] vivi: remove pointless g/s_std support

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] videobuf2-core: add helper functions
Hans Verkuil [Mon, 2 Jul 2012 08:59:18 +0000 (05:59 -0300)]
[media] videobuf2-core: add helper functions

Add helper functions to make it easier to adapt drivers to vb2.

These helpers take care of core locking and check if the filehandle is the
owner of the queue.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-dev/ioctl.c: add vb2_queue support to video_device
Hans Verkuil [Fri, 22 Jun 2012 10:29:35 +0000 (07:29 -0300)]
[media] v4l2-dev/ioctl.c: add vb2_queue support to video_device

This prepares struct video_device for easier integration with vb2.

It also introduces a new lock that protects the vb2_queue. It is up
to the driver to use it or not. And the driver can associate an owner
filehandle with the queue to check whether queuing requests are
permitted for the calling filehandle.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] Spec: document CREATE_BUFS behavior if count == 0
Hans Verkuil [Fri, 22 Jun 2012 08:47:30 +0000 (05:47 -0300)]
[media] Spec: document CREATE_BUFS behavior if count == 0

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] vb2-core: add support for count == 0 in create_bufs
Hans Verkuil [Fri, 22 Jun 2012 08:44:14 +0000 (05:44 -0300)]
[media] vb2-core: add support for count == 0 in create_bufs

This also fixes incorrect error handling in create_bufs: the return code
of __vb2_queue_alloc is the number of allocated buffers, and not a
traditional error code.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] vb2-core: refactor reqbufs/create_bufs
Hans Verkuil [Wed, 27 Jun 2012 20:10:30 +0000 (17:10 -0300)]
[media] vb2-core: refactor reqbufs/create_bufs

Split off the memory and type validation. This is done both from reqbufs
and create_bufs, and will also be done by vb2 helpers in a later patch.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx18: don't mess with vfd->debug
Hans Verkuil [Sat, 9 Jun 2012 13:54:19 +0000 (10:54 -0300)]
[media] cx18: don't mess with vfd->debug

This is now controlled by sysfs.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] ivtv: don't mess with vfd->debug
Hans Verkuil [Sat, 9 Jun 2012 13:48:53 +0000 (10:48 -0300)]
[media] ivtv: don't mess with vfd->debug

This is now controlled by sysfs.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl: remove v4l_(i2c_)print_ioctl
Hans Verkuil [Fri, 22 Jun 2012 09:38:06 +0000 (06:38 -0300)]
[media] v4l2-ioctl: remove v4l_(i2c_)print_ioctl

v4l_i2c_print_ioctl wasn't used and v4l_print_ioctl could be replaced by
v4l_printk_ioctl.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-dev.c: add debug sysfs entry
Hans Verkuil [Fri, 22 Jun 2012 09:37:38 +0000 (06:37 -0300)]
[media] v4l2-dev.c: add debug sysfs entry

Since this could theoretically change the debug value while in the middle
of v4l2-ioctl.c, we make a copy of vfd->debug to ensure consistent debug
behavior.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: finalize table conversion
Hans Verkuil [Fri, 22 Jun 2012 09:35:01 +0000 (06:35 -0300)]
[media] v4l2-ioctl.c: finalize table conversion

Implement the default case which finalizes the table conversion and allows
us to remove the last part of the switch.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for the remaining ioctls
Hans Verkuil [Sat, 9 Jun 2012 15:55:52 +0000 (12:55 -0300)]
[media] v4l2-ioctl.c: use the new table for the remaining ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for preset/timings ioctls
Hans Verkuil [Sat, 9 Jun 2012 13:09:07 +0000 (10:09 -0300)]
[media] v4l2-ioctl.c: use the new table for preset/timings ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for debug ioctls
Hans Verkuil [Sat, 9 Jun 2012 13:06:25 +0000 (10:06 -0300)]
[media] v4l2-ioctl.c: use the new table for debug ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for compression ioctls
Hans Verkuil [Sat, 9 Jun 2012 13:02:49 +0000 (10:02 -0300)]
[media] v4l2-ioctl.c: use the new table for compression ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for selection ioctls
Hans Verkuil [Sat, 9 Jun 2012 14:57:46 +0000 (11:57 -0300)]
[media] v4l2-ioctl.c: use the new table for selection ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for control ioctls
Hans Verkuil [Sat, 9 Jun 2012 15:54:02 +0000 (12:54 -0300)]
[media] v4l2-ioctl.c: use the new table for control ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for queuing/parm ioctls
Hans Verkuil [Fri, 22 Jun 2012 09:23:59 +0000 (06:23 -0300)]
[media] v4l2-ioctl.c: use the new table for queuing/parm ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for std/tuner/modulator ioctls
Hans Verkuil [Fri, 22 Jun 2012 09:17:48 +0000 (06:17 -0300)]
[media] v4l2-ioctl.c: use the new table for std/tuner/modulator ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for overlay/streamon/off ioctls
Hans Verkuil [Sat, 9 Jun 2012 15:50:15 +0000 (12:50 -0300)]
[media] v4l2-ioctl.c: use the new table for overlay/streamon/off ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for format/framebuffer ioctls
Hans Verkuil [Fri, 22 Jun 2012 09:12:57 +0000 (06:12 -0300)]
[media] v4l2-ioctl.c: use the new table for format/framebuffer ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for priority ioctls
Hans Verkuil [Sat, 9 Jun 2012 12:27:10 +0000 (09:27 -0300)]
[media] v4l2-ioctl.c: use the new table for priority ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: use the new table for querycap and i/o ioctls
Hans Verkuil [Fri, 22 Jun 2012 09:09:54 +0000 (06:09 -0300)]
[media] v4l2-ioctl.c: use the new table for querycap and i/o ioctls

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: remove an unnecessary #ifdef
Hans Verkuil [Sat, 9 Jun 2012 13:00:51 +0000 (10:00 -0300)]
[media] v4l2-ioctl.c: remove an unnecessary #ifdef

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: v4l2-ioctl: add debug and callback/offset functionality
Hans Verkuil [Fri, 22 Jun 2012 09:04:16 +0000 (06:04 -0300)]
[media] v4l2-ioctl.c: v4l2-ioctl: add debug and callback/offset functionality

Add the necessary plumbing to make it possible to replace the switch by a
table driven implementation.

The ioctls ops can either be called directly, or by calling a small function
that does some additional work if needed.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: introduce INFO_FL_CLEAR to replace switch
Hans Verkuil [Sat, 9 Jun 2012 11:55:31 +0000 (08:55 -0300)]
[media] v4l2-ioctl.c: introduce INFO_FL_CLEAR to replace switch

The switch statement that determines how much data should be copied from
userspace is replaced by a table lookup.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] v4l2-ioctl.c: move a block of code down, no other changes
Hans Verkuil [Mon, 4 Jun 2012 08:23:40 +0000 (05:23 -0300)]
[media] v4l2-ioctl.c: move a block of code down, no other changes

A block of code is moved down in the code to make later changes easier.
Do just the move without other changes to keep the diff readable for the
upcoming patch.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] bttv-cards.c: Allow radio for CHP05x/CHP06x
Thomas Betker [Sun, 1 Jul 2012 13:22:08 +0000 (10:22 -0300)]
[media] bttv-cards.c: Allow radio for CHP05x/CHP06x

Add ".has_radio = 1" for BTTV_BOARD_MAGICTVIEW061 because there are
some CHP05x/CHP06x boards with a radio tuner.

I still have an analog Askey Magic TView card (CHP051, PCI 144f:3002)
which I use for radio only. This worked fine with kernel 2.6.37
(openSUSE 11.4), but no longer works with kernel 3.1.10 (openSUSE 12.1).

The reason apparently is that ".has_radio = 1" is missing in
bttv-cards.c for BTTV_BOARD_MAGICTVIEW061; when I add this line and
recompile the kernel, radio is working again. The line is still missing
in the latest git tree, for which I have generated the attached patch.

I can't test what will happen for CHP05x/CHP06x devices without radio;
however, it seems that other cards also have the problem that there are
some boards with radio and some without, so I hope that this can be
sorted out.

Signed-off-by: Thomas Betker <thomas.betker@freenet.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[PATCH] media: add Analog Devices ADV7393 video encoder driver
Benoît Thébaudeau [Thu, 28 Jun 2012 15:12:13 +0000 (12:12 -0300)]
[PATCH] media: add Analog Devices ADV7393 video encoder driver

Add ADV7393 I²C-based video encoder driver. This driver has been tested on
custom hardware. It has been tested for composite output. It is derived from the
ADV7343 driver.

Signed-off-by: Benoît Thébaudeau <benoit.thebaudeau@advansee.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] [V3] stv090x: variable 'no_signal' set but not used
Mauro Carvalho Chehab [Fri, 6 Jul 2012 18:11:15 +0000 (15:11 -0300)]
[media] [V3] stv090x: variable 'no_signal' set but not used

Remove variable and ignore return value of stv090x_chk_signal().

Tested by compilation only.

[mchehab@redhat.com: instead of reverting and applying V3,
 applied just the diff patch, for the sake of a cleaner history]
Signed-off-by: Peter Senna Tschudin <peter.senna@gmail.com>
Reviewed-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] uvcvideo: Fix alternate setting selection
Laurent Pinchart [Thu, 21 Jun 2012 09:35:04 +0000 (06:35 -0300)]
[media] uvcvideo: Fix alternate setting selection

The alternate setting number is not equal to the alternate setting index
in the interface alternate settings table. Use the alternate setting
number from the interface descriptor when calling usb_set_interface().

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Ming Lei <ming.lei@canonical.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] uvcvideo: Fix frame drop in bulk video stream
Jayakrishnan [Fri, 9 Mar 2012 13:10:49 +0000 (10:10 -0300)]
[media] uvcvideo: Fix frame drop in bulk video stream

When video endpoint is configured as bulk, a ZLP is sent after every
video frames with size as multiple of 512 bytes. This is done so that
host can detect end of transfer and pass data for processing.
Still, frames that are multiple of 16K in size gets dropped. The ZLP
sent by camera is ignored by uvc_video_decode_bulk(). The makes sure
that the ZLP is not part of a video frame before ignoring it. If ZLP
follows a video frame, then it triggers completion callback.

[mchehab@redhat.com: Fix a small CodingStyle issue]
Signed-off-by: Jayakrishnan Memana <jayakrishnan.memana@maxim-ic.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] uvcvideo: Document the struct uvc_xu_control_query query field
Laurent Pinchart [Mon, 14 May 2012 07:34:34 +0000 (04:34 -0300)]
[media] uvcvideo: Document the struct uvc_xu_control_query query field

Several developers have reported that the lack of macros for the struct
uvc_xu_control_query query field in uvcvideo.h was confusing and forced
them to look at the driver source code to find out what applications
were supposed to pass in that field.

Add a comment to the header to clarify the expected usage of the query
field.

Reported-by: Paulo Assis <pj.assis@gmail.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] drxk: Improve logging
Martin Blumenstingl [Wed, 4 Jul 2012 20:38:23 +0000 (17:38 -0300)]
[media] drxk: Improve logging

This patch simply fixes some logging calls:
- Use KERN_INFO when printing the chip status.
- Add a missing space when logging the drxk_gate_ctrl call.
- Use the same logging text as always if the scu_command in GetQAMLockStatus fails.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years agofixupSigned-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Mauro Carvalho Chehab [Fri, 6 Jul 2012 17:53:51 +0000 (14:53 -0300)]
fixupSigned-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

12 years ago[media] drxk: Make the QAM demodulator command parameters configurable
Martin Blumenstingl [Wed, 4 Jul 2012 20:36:55 +0000 (17:36 -0300)]
[media] drxk: Make the QAM demodulator command parameters configurable

Currently there are two different implementations (in the firmware) for
the QAM demodulator command: one takes 4 and the other takes 2
parameters. The driver shows an error in dmesg When using the
4-parameter command with firmware that implements the 2-parameter
command. Unfortunately this happens every time when chaning the
frequency (on DVB-C).

This patch simply makes configurable, how many command parameters will
be used. All existing drxk_config instances using the "drxk_a3.mc" were
updated because this firmware is the only loadable firmware where the
QAM demodulator command takes 4 parameters. Some firmwares in the ROM
might also use it. The drxk instances in the em28xx-dvb driver were also
updated to silence the warnings.

If no qam_demod_parameter_count is given in the drxk_config struct, then
the correct number of parameters will be auto-detected.

[mchehab@redhat.com: Fix a small CodingStyle issue at one comment]
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Tested-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] drxk: fix a '&' vs '|' bug
Dan Carpenter [Tue, 26 Jun 2012 07:40:38 +0000 (04:40 -0300)]
[media] drxk: fix a '&' vs '|' bug

IQM_AF_CLKNEG_CLKNEGDATA__M is 0x2 and
IQM_AF_CLKNEG_CLKNEGDATA_CLK_ADC_DATA_POS is 0.  (clkNeg | 0x2) is never
equal to zero so the condition can never be true.

I consulted with Ralph Metzler and the '|' should be changed to a '&'.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
CC: Ralph Metzler <rjkm@metzlerbros.de>
Tested-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] tuner-xc2028: tag the usual firmwares to help dracut
Mauro Carvalho Chehab [Tue, 26 Jun 2012 18:34:22 +0000 (15:34 -0300)]
[media] tuner-xc2028: tag the usual firmwares to help dracut

When tuner-xc2028 is not compiled as a module, dracut will
need to copy the firmware inside the initfs image.

So, use MODULE_FIRMWARE() to indicate such need.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] uvc/Kconfig: Fix INPUT/EVDEV dependencies
Mauro Carvalho Chehab [Fri, 6 Jul 2012 13:41:53 +0000 (10:41 -0300)]
[media] uvc/Kconfig: Fix INPUT/EVDEV dependencies

USB_VIDEO_CLASS_INPUT_EVDEV should be dependent on the UVC
selection, as otherwise, when UVC is unselected, this dependent
config still appears.

Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] rc/Kconfig: Move a LIRC sub-option to the right place
Mauro Carvalho Chehab [Fri, 6 Jul 2012 12:21:13 +0000 (09:21 -0300)]
[media] rc/Kconfig: Move a LIRC sub-option to the right place

The IR to LIRC option were at the wrong sub-menu. Move it to the right
place.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] Kconfig: Split the core support options from the driver ones
Mauro Carvalho Chehab [Fri, 6 Jul 2012 12:12:59 +0000 (09:12 -0300)]
[media] Kconfig: Split the core support options from the driver ones

Better arrange the remote controller driver items to happen after the
core support, on their proper menus, and making clerarer what is media
core options and what is media driver options.

Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
12 years ago[media] cx231xx: don't DMA to random addresses
David Dillow [Mon, 18 Jun 2012 03:15:21 +0000 (00:15 -0300)]
[media] cx231xx: don't DMA to random addresses

Commit 7a6f6c29d264cdd2fe0eb3d923217eed5f0ad134 (cx231xx: use
URB_NO_TRANSFER_DMA_MAP) was intended to avoid mapping the DMA buffer
for URB twice. This works for the URBs allocated with usb_alloc_urb(),
as those are allocated from cohernent DMA pools, but the flag was also
added for the VBI and audio URBs, which have a manually allocated area.
This leaves the random trash in the structure after allocation as the
DMA address, corrupting memory and preventing VBI and audio from
working. Letting the USB core map the buffers solves the problem.

Signed-off-by: David Dillow <dave@thedillows.org>
Cc: Sri Deevi <srinivasa.deevi@conexant.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>