platform/kernel/linux-rpi.git
15 years agoV4L/DVB (11152): hdpvr: Fix build with Config_I2C not set
Janne Grunau [Mon, 23 Mar 2009 21:18:54 +0000 (18:18 -0300)]
V4L/DVB (11152): hdpvr: Fix build with Config_I2C not set

Signed-off-by: Janne Grunau <j@jannau.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11146): gspca - vc032x: Change the probe sequence.
Jean-Francois Moine [Sun, 22 Mar 2009 19:33:47 +0000 (16:33 -0300)]
V4L/DVB (11146): gspca - vc032x: Change the probe sequence.

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11145): gspca - t613: Greater delay after om6802 reset.
Jean-Francois Moine [Sun, 22 Mar 2009 19:31:32 +0000 (16:31 -0300)]
V4L/DVB (11145): gspca - t613: Greater delay after om6802 reset.

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11144): gspca - t613: Don't re-read the ID registers at probe time.
Jean-Francois Moine [Sun, 22 Mar 2009 19:30:42 +0000 (16:30 -0300)]
V4L/DVB (11144): gspca - t613: Don't re-read the ID registers at probe time.

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11143): gspca - t613: Bad sensor detection.
Jean-Francois Moine [Sun, 22 Mar 2009 19:29:36 +0000 (16:29 -0300)]
V4L/DVB (11143): gspca - t613: Bad sensor detection.

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11142): au0828: fix oops on ARM platform when allocating transfer buffers
Devin Heitmueller [Mon, 23 Mar 2009 02:42:26 +0000 (23:42 -0300)]
V4L/DVB (11142): au0828: fix oops on ARM platform when allocating transfer buffers

Add missing URB_NO_TRANSFER_DMA_MAP flag, since the use of consistent memory
is not permitted for DMA on the ARM platform.

Thanks to Paul Thomas <pthomas8589@gmail.com> for providing sample ARM
hardware that was experiencing the oops (tested on the at91rm9200 based
LinuxStamp).

Thanks to David Brownell <david-b@pacbell.net> for providing insight into the
ARM memory architecture.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11141): em28xx: fix oops on ARM platform when allocating transfer buffers
Devin Heitmueller [Mon, 23 Mar 2009 02:41:28 +0000 (23:41 -0300)]
V4L/DVB (11141): em28xx: fix oops on ARM platform when allocating transfer buffers

Add missing URB_NO_TRANSFER_DMA_MAP flag, since the use of consistent memory
is not permitted for DMA on the ARM platform.

Thanks to Paul Thomas <pthomas8589@gmail.com> for providing sample ARM
hardware that was experiencing the oops (tested on the at91rm9200 based
LinuxStamp).

Thanks to David Brownell <david-b@pacbell.net> for providing insight into the
ARM memory architecture.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11140): usbvision: fix oops on ARM platform when allocating transfer buffers
Devin Heitmueller [Mon, 23 Mar 2009 02:39:12 +0000 (23:39 -0300)]
V4L/DVB (11140): usbvision: fix oops on ARM platform when allocating transfer buffers

Add missing URB_NO_TRANSFER_DMA_MAP flag, since the use of consistent memory
is not permitted for DMA on the ARM platform.

Thanks to Paul Thomas <pthomas8589@gmail.com> for providing sample ARM hardware
that was experiencing the oops (tested on the at91rm9200 based LinuxStamp).

Thanks to David Brownell <david-b@pacbell.net> for providing insight into the
ARM memory architecture.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11139): em28xx: add remote control definition for HVR-900 (both versions)
Devin Heitmueller [Mon, 23 Mar 2009 01:58:46 +0000 (22:58 -0300)]
V4L/DVB (11139): em28xx: add remote control definition for HVR-900 (both versions)

The HVR-900 did not have a remote control defined, so it would not work.  Add
the line for both versions of the product.

Thanks to Jens-Michael Hoffmann (#linuxtv user "jmho") for pointing out the
issue and testing the patch.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11138): get_dvb_firmware: add support for downloading the cx2584x firmware...
Mauro Carvalho Chehab [Mon, 23 Mar 2009 15:35:03 +0000 (12:35 -0300)]
V4L/DVB (11138): get_dvb_firmware: add support for downloading the cx2584x firmware for pvrusb2

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11137): get_dvb_firmware: add cx23885 firmwares
Mauro Carvalho Chehab [Mon, 23 Mar 2009 14:46:19 +0000 (11:46 -0300)]
V4L/DVB (11137): get_dvb_firmware: add cx23885 firmwares

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11136): get_dvb_firmware: Add download code for cx18 firmwares
Mauro Carvalho Chehab [Mon, 23 Mar 2009 13:57:15 +0000 (10:57 -0300)]
V4L/DVB (11136): get_dvb_firmware: Add download code for cx18 firmwares

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11127): Kconfig: replace all occurrences of CUSTOMIZE to CUSTOMISE
Mauro Carvalho Chehab [Fri, 20 Mar 2009 22:33:59 +0000 (19:33 -0300)]
V4L/DVB (11127): Kconfig: replace all occurrences of CUSTOMIZE to CUSTOMISE

There are several Kconfig items using CUSTOMIZE. Yet, most use the English
writing CUSTOMISE. This generates lots of trouble, because people sometimes type
the Kbuild item different.

Let's standardise every occurrence using the same syntax.

The changes were generated by this small shell script:

for i in `find linux -type f`; do sed s,CUSTOMIZE,CUSTOMISE,g $i >/tmp/a && mv /tmp/a $i; done

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11125): fix mispelled Hauppauge in HD PVR and PVR USB2 driver comments
Janne Grunau [Thu, 19 Mar 2009 22:00:35 +0000 (19:00 -0300)]
V4L/DVB (11125): fix mispelled Hauppauge in HD PVR and PVR USB2 driver comments

Signed-off-by: Janne Grunau <j@jannau.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11124): Add support for ProVideo PV-183 to bttv
Alan McIvor [Fri, 13 Mar 2009 00:43:34 +0000 (21:43 -0300)]
V4L/DVB (11124): Add support for ProVideo PV-183 to bttv

Add support for ProVideo PV-183 to bttv

This patch adds support for the ProVideo PV-183 card to the bttv
device driver. The PV-183 is a PCI card with 8 BT878 devices plus a Hint
Corp HiNT HB4 PCI-PCI Bridge. Each BT878 has two composite input channels
available. There are no tuners on this card.

Signed-off-by: Alan McIvor <alan.mcivor@reveal.co.nz>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11120): cafe_ccic: stick in a comment with a request for test results
Hans Verkuil [Wed, 18 Mar 2009 16:31:56 +0000 (13:31 -0300)]
V4L/DVB (11120): cafe_ccic: stick in a comment with a request for test results

Due to lack of hardware this conversion to v4l2_device/v4l2_subdev is
untested. If all goes well I should be able to test it in about a month,
but just in case I can't manage that it should be made clear in the code
that this isn't tested.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11118): cafe_ccic: replace debugfs with g/s_register ioctls.
Hans Verkuil [Wed, 18 Mar 2009 16:25:54 +0000 (13:25 -0300)]
V4L/DVB (11118): cafe_ccic: replace debugfs with g/s_register ioctls.

Using VIDIOC_DBG_S/G_REGISTER is the standard way of reading/writing register
for advanced debugging under v4l2. In addition, using this means that the
cafe_ccic driver doesn't need to have knowledge about the used sensor: the
debug ioctl can be passed on to the sensor if it isn't for the host.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Acked-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11117): ov7670: add support to get/set registers
Hans Verkuil [Wed, 18 Mar 2009 16:24:05 +0000 (13:24 -0300)]
V4L/DVB (11117): ov7670: add support to get/set registers

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Acked-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11116): ov7670: cleanup and remove legacy code.
Hans Verkuil [Wed, 18 Mar 2009 16:23:13 +0000 (13:23 -0300)]
V4L/DVB (11116): ov7670: cleanup and remove legacy code.

v4l2-i2c-drv-legacy.h and ov7670_command are no longer needed after the
cafe driver is converted to use v4l2_subdev.

Rather than using a large ov7670_control array, just call the handlers
via a simple switch and use v4l2_ctrl_query_fill() to handle queryctrl.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Acked-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11115): cafe_ccic: use v4l2_subdev to talk to the ov7670 sensor.
Hans Verkuil [Wed, 18 Mar 2009 16:16:44 +0000 (13:16 -0300)]
V4L/DVB (11115): cafe_ccic: use v4l2_subdev to talk to the ov7670 sensor.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Acked-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11114): cafe_ccic: convert to v4l2_device.
Hans Verkuil [Wed, 18 Mar 2009 16:13:09 +0000 (13:13 -0300)]
V4L/DVB (11114): cafe_ccic: convert to v4l2_device.

Convert this driver to v4l2_device and removed the unnecessary cafe_dev_list.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Acked-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11113): ov7670: convert to v4l2_subdev
Hans Verkuil [Wed, 18 Mar 2009 16:01:06 +0000 (13:01 -0300)]
V4L/DVB (11113): ov7670: convert to v4l2_subdev

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Acked-by: Jonathan Corbet <corbet@lwn.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11112): v4l2-subdev: add support for TRY_FMT, ENUM_FMT and G/S_PARM.
Hans Verkuil [Fri, 6 Feb 2009 21:59:35 +0000 (18:59 -0300)]
V4L/DVB (11112): v4l2-subdev: add support for TRY_FMT, ENUM_FMT and G/S_PARM.

These ops are used by the ov7670 driver, so these need to be added.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11111a): MAINTAINERS: Drop references to deprecated video4linux list
Jean Delvare [Sat, 14 Mar 2009 18:21:24 +0000 (15:21 -0300)]
V4L/DVB (11111a): MAINTAINERS: Drop references to deprecated video4linux list

Mailing list video4linux-list@redhat.com is deprecated, so drop
references to it in MAINTAINERS.

MAINTAINERS |    2 --

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11111): dvb_dummy_fe: Fix compilation breakage
Mauro Carvalho Chehab [Fri, 20 Mar 2009 00:41:08 +0000 (21:41 -0300)]
V4L/DVB (11111): dvb_dummy_fe: Fix compilation breakage

As reported by Randy Dunlap <randy.dunlap@oracle.com>:
ERROR: "dvb_dummy_fe_ofdm_attach" [drivers/media/video/cx231xx/cx231xx-dvb.ko] undefined!

This happens since cx231xx DVB part still misses the frontend modules.
So, the dummy frontend were used for development. The proper fix is to
implement the DVB modules there, as they will be required.

While this won't happen, lets allow the compilation with or without the
dummy FE testing module.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11110): au8522/au0828: Fix Kconfig dependencies
Mauro Carvalho Chehab [Fri, 20 Mar 2009 00:03:09 +0000 (21:03 -0300)]
V4L/DVB (11110): au8522/au0828: Fix Kconfig dependencies

au8522 is now dependent of V4L2, as reported by Randy Dunlap
<randy.dunlap@oracle.com>:

au8522_decoder.c:(.text+0x199898): undefined reference to `v4l2_ctrl_query_fill'
au8522_decoder.c:(.text+0x1998b3): undefined reference to `v4l2_ctrl_query_fill'
au8522_decoder.c:(.text+0x199944): undefined reference to `v4l2_device_unregister_subdev'
au8522_decoder.c:(.text+0x19997c): undefined reference to `v4l2_chip_ident_i2c_client'
au8522_decoder.c:(.text+0x199f1e): undefined reference to `v4l2_i2c_subdev_init'

Cc: Devin Heitmueller <devin.heitmueller@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11109): au0828: Fix compilation when VIDEO_ADV_DEBUG = n
Mauro Carvalho Chehab [Thu, 19 Mar 2009 22:26:23 +0000 (19:26 -0300)]
V4L/DVB (11109): au0828: Fix compilation when VIDEO_ADV_DEBUG = n

As reported by Kyle McMartin and Randy Dunlap:

drivers/media/video/au0828/au0828-video.c:1438: error: 'const struct v4l2_subdev_core_ops' has no member named 'g_register'
drivers/media/video/au0828/au0828-video.c:1453: error: 'const struct v4l2_subdev_core_ops' has no member named 's_register'

This patch properly implements those two API ioctls only when debug is
enabled.

Cc: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11108): get_dvb_firmware: Add option to download firmware for cx231xx
Mauro Carvalho Chehab [Thu, 19 Mar 2009 15:10:00 +0000 (12:10 -0300)]
V4L/DVB (11108): get_dvb_firmware: Add option to download firmware for cx231xx

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11106): gspca - ov534: New sensor ov965x and re-enable the webcam 06f8:3003
Jean-Francois Moine [Thu, 19 Mar 2009 09:15:21 +0000 (06:15 -0300)]
V4L/DVB (11106): gspca - ov534: New sensor ov965x and re-enable the webcam 06f8:3003

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11105): gspca - ov534: Adjust the packet scan function
Jean-Francois Moine [Thu, 19 Mar 2009 09:12:59 +0000 (06:12 -0300)]
V4L/DVB (11105): gspca - ov534: Adjust the packet scan function

- change max payload size to 2040 bytes (was 2048)
- optimize

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11104): gspca - ov534: Bad frame pointer after adding the last packet
Jean-Francois Moine [Thu, 19 Mar 2009 09:05:06 +0000 (06:05 -0300)]
V4L/DVB (11104): gspca - ov534: Bad frame pointer after adding the last packet

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11103): gspca - main: May have isochronous transfers on altsetting 0
Jean-Francois Moine [Thu, 19 Mar 2009 08:55:22 +0000 (05:55 -0300)]
V4L/DVB (11103): gspca - main: May have isochronous transfers on altsetting 0

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11100): au8522: fix compilation warning.
Hans Verkuil [Wed, 18 Mar 2009 19:39:07 +0000 (16:39 -0300)]
V4L/DVB (11100): au8522: fix compilation warning.

normal_i2c and I2C_CLIENT_INSMOD are only necessary for kernels < 2.6.22.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11098): v4l2-common: remove incorrect MODULE test
Hans Verkuil [Wed, 18 Mar 2009 18:48:01 +0000 (15:48 -0300)]
V4L/DVB (11098): v4l2-common: remove incorrect MODULE test

v4l2-common doesn't have to be a module for it to call request_module().
Just remove that test.

Thanks-to: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11097): use video_ioctl2 as ioctl handler directly
Janne Grunau [Wed, 18 Mar 2009 23:57:04 +0000 (20:57 -0300)]
V4L/DVB (11097): use video_ioctl2 as ioctl handler directly

The encoder commands ioctls are available in v4l2_ioctl_ops. Use them
and get rid of the custom ioctl handler and use video_ioctl2.

Signed-off-by: Janne Grunau <j@jannau.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11096): V4L2 Driver for the Hauppauge HD PVR usb capture device
Janne Grunau [Wed, 18 Mar 2009 21:10:04 +0000 (18:10 -0300)]
V4L/DVB (11096): V4L2 Driver for the Hauppauge HD PVR usb capture device

The device encodes component video up to 1080i to a MPEG-TS stream with
H.264 video and stereo AAC audio. Newer firmwares accept also AC3
(up to 5.1) audio over optical SPDIF without reencoding.
Firmware upgrade is unimplemeted but rather unimportant since
the firmware sits on a flash chip.

The I2C adapter to drive the integrated infrared receiver/sender is
currently disabled due to a conflict with cx18-based devices.

Tested-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Janne Grunau <j@jannau.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11095): adds V4L2_CID_SHARPNESS to v4l2_ctrl_query_fill()
Janne Grunau [Wed, 18 Mar 2009 20:00:39 +0000 (17:00 -0300)]
V4L/DVB (11095): adds V4L2_CID_SHARPNESS to v4l2_ctrl_query_fill()

Signed-off-by: Janne Grunau <j@jannau.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11092): cx18: Optimize processing of VBI buffers from the capture unit
Andy Walls [Sun, 15 Mar 2009 02:20:49 +0000 (23:20 -0300)]
V4L/DVB (11092): cx18: Optimize processing of VBI buffers from the capture unit

Removed some unnecessary memcpy()'s by reworking the compress_*_vbi_buf()
functions.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11091): cx18, ivtv: Ensure endianess for linemasks in VBI embedded in MPEG...
Andy Walls [Sat, 14 Mar 2009 23:03:26 +0000 (20:03 -0300)]
V4L/DVB (11091): cx18, ivtv: Ensure endianess for linemasks in VBI embedded in MPEG stream

The sliced VBI payloads that cx18 and ivtv would insert in the MPEG stream
did not have consistent endianess for the linemasks in the payload (a big
endian platform would write them out big endian).  This change ensures the
linemasks are always stored as little-endian in the MPEG stream to ensure
cross platform consistency in parsing the generated MPEG stream.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11089): au8522: finish conversion to v4l2_device/subdev
Devin Heitmueller [Sun, 15 Mar 2009 23:05:51 +0000 (20:05 -0300)]
V4L/DVB (11089): au8522: finish conversion to v4l2_device/subdev

Per Hans Verkuil <hverkuil@xs4all.nl> instruction, remove the au8522_command
and replace v4l2-i2c-drv-legacy.h with v4l2-i2c-drv.h

Thanks to Hans Verkuil <hverkuil@xs4all.nl> for reviewing the au8522 analog
support.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11088): au0828: finish videodev/subdev conversion
Devin Heitmueller [Sun, 15 Mar 2009 23:01:53 +0000 (20:01 -0300)]
V4L/DVB (11088): au0828: finish videodev/subdev conversion

Per Hans Verkuil <hverkuil@xs4all.nl> instruction, remove the deprecated
attach_inform/detach_inform routines, and convert over the i2c calls to
subdev calls.

Thanks to Hans Verkuil <hverkuil@xs4all.nl> for providing feedback on the
au0828 analog support.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11086): au0828: rename macro for currently non-function VBI support
Devin Heitmueller [Sun, 15 Mar 2009 21:52:10 +0000 (18:52 -0300)]
V4L/DVB (11086): au0828: rename macro for currently non-function VBI support

The VBI support or the au0828 has the framework written but it does not yet
work.  Rename the macro per Mauro's request.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11085): au0828/au8522: Codingstyle fixes
Devin Heitmueller [Sun, 15 Mar 2009 21:48:52 +0000 (18:48 -0300)]
V4L/DVB (11085): au0828/au8522: Codingstyle fixes

Take a pass over all of the au0828/au8522 files and cleanup all the codingstyle
issues.  This patch does not make *any* functional change to the code.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11084): au0828: add entry for undefined input type
Devin Heitmueller [Sun, 15 Mar 2009 20:48:26 +0000 (17:48 -0300)]
V4L/DVB (11084): au0828: add entry for undefined input type

For the sake of completeness, include the "undefined" input type enumeration,
even though there is no path that can actually call it.

Thanks to Mauro Carvalho Chehab <mchehab@infradead.org> for pointing this out.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11083): au0828: remove some unneeded braces
Devin Heitmueller [Sun, 15 Mar 2009 20:43:13 +0000 (17:43 -0300)]
V4L/DVB (11083): au0828: remove some unneeded braces

There were some braces left behind from when there was more code in the block.
Remove it.

Thanks to Mauro Carvalho Chehab <mchehab@infradead.org> for pointing this out.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11082): au0828: remove memset calls in v4l2 routines.
Devin Heitmueller [Sun, 15 Mar 2009 20:38:47 +0000 (17:38 -0300)]
V4L/DVB (11082): au0828: remove memset calls in v4l2 routines.

The userland callers are responsible for clearing the output buffers, so
remove the unneeded memset calls.

Thanks to Mauro Carvalho Chehab <mchehab@infradead.org> for pointing this out.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11081): au0828: make sure v4l2_device name is unique
Devin Heitmueller [Thu, 12 Mar 2009 00:58:04 +0000 (21:58 -0300)]
V4L/DVB (11081): au0828: make sure v4l2_device name is unique

Make sure newly created v4l2 devices have a unique name, modeling the logic
after the cx18 driver.

Thanks to Andy Walls <awalls@radix.net> for pointing out the issue.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11080): au0828: Convert to use v4l2_device/subdev framework
Devin Heitmueller [Wed, 11 Mar 2009 06:01:04 +0000 (03:01 -0300)]
V4L/DVB (11080): au0828: Convert to use v4l2_device/subdev framework

Convert over to using the new subdev framework for the au0828 bridge.  This
includes using the new i2c probing mechanism.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11079): au0828: fix panic on disconnect if analog initialization failed
Devin Heitmueller [Wed, 11 Mar 2009 06:01:02 +0000 (03:01 -0300)]
V4L/DVB (11079): au0828: fix panic on disconnect if analog initialization failed

If the analog initialization failed to create the video device, we never
actually add the entry to the au0828_devlist.  Therefore a panic occurs when
unregistering the analog subsystem.

Make it so we only remove the entry from the list if we added it to the list
in the first place.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11078): au0828: properly handle non-existent analog inputs
Devin Heitmueller [Wed, 11 Mar 2009 06:01:01 +0000 (03:01 -0300)]
V4L/DVB (11078): au0828: properly handle non-existent analog inputs

It is not valid to look for dev->board.input == NULL to detect an undefined
analog configuration section, since it is a member of the struct and not a
pointer (hence it will *always* be non-NULL).  Do the check based on whether
the first input is actually a valid input type instead.

Thanks to Michael Krufky <mkrufky@linuxtv.org> for providing sample hardware
of various configurations to test with.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11077): au0828: properly handle missing analog USB endpoint
Devin Heitmueller [Wed, 11 Mar 2009 06:01:00 +0000 (03:01 -0300)]
V4L/DVB (11077): au0828: properly handle missing analog USB endpoint

Move the setup of the analog isoc handler into au0828-video.c, so it does not
occur if there is not an .input section defined for the board.  Also fixes a
case where if there is an input section but the board does not actually have
analog support, the digital support will continue to work as expected.

Thanks to Michael Krufky <mkrufky@linuxtv.org> for providing sample hardware
of various configurations to test with.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11076): au0828: make g_chip_ident call work properly
Devin Heitmueller [Wed, 11 Mar 2009 06:00:58 +0000 (03:00 -0300)]
V4L/DVB (11076): au0828: make g_chip_ident call work properly

Make the g_chip_ident call work for the au0828/au8522.  Discovered when testing
with the v4l2_compliance tool

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
[mchehab@redhat.com: fix merge conflict, due to a path change for analog demod]
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11075): au0828: make register debug lines easier to read
Devin Heitmueller [Wed, 11 Mar 2009 06:00:57 +0000 (03:00 -0300)]
V4L/DVB (11075): au0828: make register debug lines easier to read

Make it a little easier to read the debug messages for register read/write
operations

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11074): au0828: fix i2c enumeration bug
Devin Heitmueller [Wed, 11 Mar 2009 06:00:56 +0000 (03:00 -0300)]
V4L/DVB (11074): au0828: fix i2c enumeration bug

There was a bug where enumerating the i2c for devices would result in false
positives.  The root of the issue was the scanning was using SMBUS_QUICK
messages, which are zero length write requests (which our i2c adapter
implementation didn't handle).  Because we never strobed any bytes onto the
bus, the status register would still contain the value from the previous
request.

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11073): au0828: disable VBI code since it doesn't yet work
Devin Heitmueller [Wed, 11 Mar 2009 06:00:55 +0000 (03:00 -0300)]
V4L/DVB (11073): au0828: disable VBI code since it doesn't yet work

Since the VBI support is not yet working for the au0828, don't advertise the
capability or create the /dev/vbi device.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11072): au0828: advertise only NTSC-M (as opposed to all NTSC standards)
Devin Heitmueller [Wed, 11 Mar 2009 06:00:53 +0000 (03:00 -0300)]
V4L/DVB (11072): au0828: advertise only NTSC-M (as opposed to all NTSC standards)

We don't now how to make any variant of NTSC work other than NTSC-M, so don't
advertise that we support the other variants.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11071): tveeprom: add the xc5000 tuner to the tveeprom definition
Devin Heitmueller [Wed, 11 Mar 2009 06:00:49 +0000 (03:00 -0300)]
V4L/DVB (11071): tveeprom: add the xc5000 tuner to the tveeprom definition

Prior to now, we never relied on the Hauppauge tveeprom to identify which
tuner is associated with the board.  Now that the HVR-950q determines what
the tuner is based on the eeprom, set the tuner ID appropriately.

Thanks to Michael Krufky <mkrufky@linuxtv.org> for providing information on
the tveeprom organization.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11070): au0828: Rework the way the analog video binding occurs
Devin Heitmueller [Wed, 11 Mar 2009 06:00:47 +0000 (03:00 -0300)]
V4L/DVB (11070): au0828: Rework the way the analog video binding occurs

Rework the way boards are managed so that we can change the board description
based on the Hauppauge eeprom (modeled after cx88-cards.c).

Also, make sure that we don't load the analog stack if there are no analog
inputs defined in the board profile.

Thanks to Michael Krufky <mkrufky@linuxtv.org> for providing information on
the various ways different Hauppauge boards can be configured.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11069): au8522: add mutex protecting use of hybrid state
Devin Heitmueller [Wed, 11 Mar 2009 06:00:45 +0000 (03:00 -0300)]
V4L/DVB (11069): au8522: add mutex protecting use of hybrid state

Access using the hybrid state framework requires the list to be protected by
a mutex.

Thanks to Michael Krufky <mkrufky@linuxtv.org> for reporting this during a
code review.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11068): au0828: add analog profile for the HVR-850
Devin Heitmueller [Wed, 11 Mar 2009 06:00:43 +0000 (03:00 -0300)]
V4L/DVB (11068): au0828: add analog profile for the HVR-850

Add the analog parameters to the device profile for the HVR-850

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11067): au0828: workaround a bug in the au0828 i2c handling
Devin Heitmueller [Wed, 11 Mar 2009 06:00:41 +0000 (03:00 -0300)]
V4L/DVB (11067): au0828: workaround a bug in the au0828 i2c handling

There is an issue related to the i2c clock for addressing the xc5000.  The
au0828 chip does not support clock stretching, which the xc5000 makes use of.
This results in cases where we silently get back garbage in i2c read
operations.  To work around this issue until we slow down the i2c clock when
talking with that specific device.

This was not an issue before we had analog support because we never needed to
enumerate the i2c bus, and digital tuning never actually needed to perform
read operations against the xc5000.

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11066): au0828: add support for analog functionality in bridge
Devin Heitmueller [Wed, 11 Mar 2009 06:00:40 +0000 (03:00 -0300)]
V4L/DVB (11066): au0828: add support for analog functionality in bridge

Add support for the analog functionality found in the au0828 bridge

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
[mchehab@redhat.com: fix compilation by adding linux/version.h]
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11065): au8522: add support for analog side of demodulator
Devin Heitmueller [Wed, 11 Mar 2009 06:00:38 +0000 (03:00 -0300)]
V4L/DVB (11065): au8522: add support for analog side of demodulator

Add support for the analog functionality in the au8522 analog/digital
demodulator

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
[mchehab: renamed drivers/media/video/au8522_decoder.c as drivers/media/dvb/frontends/au8522_decoder.c to avoid breaking bisect]
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11064): au8522: make use of hybrid framework so analog/digital demod can...
Devin Heitmueller [Wed, 11 Mar 2009 06:00:36 +0000 (03:00 -0300)]
V4L/DVB (11064): au8522: make use of hybrid framework so analog/digital demod can share state

Make use of the hybrid tuner framework so the analog and digital parts of the
au8522 demodulator can make use of the same shared state.

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11063): au8522: power down the digital demod when not in use
Devin Heitmueller [Wed, 11 Mar 2009 06:00:34 +0000 (03:00 -0300)]
V4L/DVB (11063): au8522: power down the digital demod when not in use

When the au8522 is idle, put the chip into a low power mode (reduces power
consumption from 450ma to 346ma)

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11062): au8522: fix register read/write high bits
Devin Heitmueller [Wed, 11 Mar 2009 06:00:19 +0000 (03:00 -0300)]
V4L/DVB (11062): au8522: fix register read/write high bits

For the i2c messages to read and write registers, the two high order bits
of the first byte dictates whether it is a read or a write operation.

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11061): au8522: move shared state and common functions into a separate heade...
Devin Heitmueller [Wed, 11 Mar 2009 05:59:56 +0000 (02:59 -0300)]
V4L/DVB (11061): au8522: move shared state and common functions into a separate header files

Move the au8522 state structure, as well as exposing functions needed by the
analog side of the demodulator into a common header file.

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11060): au8522: rename the au8522.c source file
Devin Heitmueller [Wed, 11 Mar 2009 05:59:30 +0000 (02:59 -0300)]
V4L/DVB (11060): au8522: rename the au8522.c source file

Rename the au8522.c file to au8522_dig.c so that more source files can be
added to the driver while preserving the original name of au8522.ko

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11059): xc5000: fix bug for hybrid xc5000 devices with IF other than 5380
Devin Heitmueller [Wed, 11 Mar 2009 05:58:53 +0000 (02:58 -0300)]
V4L/DVB (11059): xc5000: fix bug for hybrid xc5000 devices with IF other than 5380

The xc5000 driver has a bug where the IF is always set to whatever the first
caller to dvb_attach() provides.  This fails when the device requires an IF
other than 5380 and the analog driver is loaded first through tuner-core
(which always supplies the hard-coded value of 5380).

Thanks to Michael Krufky <mkrufky@linuxtv.org> and Steven Toth
<stoth@linuxtv.org> for providing sample hardware, engineering level support,
and testing.

Signed-off-by: Devin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11057): Fix CiMax stability in Netup Dual DVB-S2 CI
Abylay Ospan [Tue, 17 Mar 2009 21:21:18 +0000 (18:21 -0300)]
V4L/DVB (11057): Fix CiMax stability in Netup Dual DVB-S2 CI

It appears TS discontinuity about one per 10 hrs if CAM used.
Patch to fix it.

Signed-off-by: Abylay Ospan <aospan@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11056): Bug fix in NetUP: restore high address lines in CI
Abylay Ospan [Tue, 17 Mar 2009 21:13:52 +0000 (18:13 -0300)]
V4L/DVB (11056): Bug fix in NetUP: restore high address lines in CI

CI high address lines disappears due to wrong data type used.
Patch to fix it.

Signed-off-by: Abylay Ospan <aospan@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11055): Fix typo in stv0900
Igor M. Liplianin [Sun, 15 Mar 2009 10:31:45 +0000 (07:31 -0300)]
V4L/DVB (11055): Fix typo in stv0900

Signed-off-by: Igor M. Liplianin <liplianin@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11054): Shorten some lines in stv0900 to less then 81 characters
Igor M. Liplianin [Sun, 15 Mar 2009 10:28:45 +0000 (07:28 -0300)]
V4L/DVB (11054): Shorten some lines in stv0900 to less then 81 characters

Signed-off-by: Igor M. Liplianin <liplianin@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11053): saa7134: set v4l2_dev field of video_device
Hans Verkuil [Sun, 15 Mar 2009 09:53:32 +0000 (06:53 -0300)]
V4L/DVB (11053): saa7134: set v4l2_dev field of video_device

The v4l2_dev field of video_device wasn't initialized. The parent field
is derived from v4l2_dev, so that doesn't need to be set anymore.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11052): bt819: remove an unused header
Hans Verkuil [Sat, 14 Mar 2009 19:34:07 +0000 (16:34 -0300)]
V4L/DVB (11052): bt819: remove an unused header

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11051): v4l-dvb: replace remaining references to the old mailinglist.
Hans Verkuil [Sat, 14 Mar 2009 18:06:08 +0000 (15:06 -0300)]
V4L/DVB (11051): v4l-dvb: replace remaining references to the old mailinglist.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11048): zoran: fix incorrect return type of notify function.
Hans Verkuil [Sat, 14 Mar 2009 15:53:37 +0000 (12:53 -0300)]
V4L/DVB (11048): zoran: fix incorrect return type of notify function.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11047): cx88: convert to v4l2_device.
Hans Verkuil [Sat, 14 Mar 2009 15:40:51 +0000 (12:40 -0300)]
V4L/DVB (11047): cx88: convert to v4l2_device.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11046): bttv: convert to v4l2_device.
Hans Verkuil [Sat, 14 Mar 2009 15:36:54 +0000 (12:36 -0300)]
V4L/DVB (11046): bttv: convert to v4l2_device.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11045): v4l2: call v4l2_device_disconnect in USB drivers.
Hans Verkuil [Sat, 14 Mar 2009 15:27:01 +0000 (12:27 -0300)]
V4L/DVB (11045): v4l2: call v4l2_device_disconnect in USB drivers.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11044): v4l2-device: add v4l2_device_disconnect
Hans Verkuil [Sat, 14 Mar 2009 11:28:45 +0000 (08:28 -0300)]
V4L/DVB (11044): v4l2-device: add v4l2_device_disconnect

Call v4l2_device_disconnect when the parent of a hotpluggable device
disconnects. This ensures that you do not have a pointer to a device that
is no longer present.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11042): v4l2-api: Add definitions for V4L2_MPEG_STREAM_VBI_FMT_IVTV payloads
Andy Walls [Sat, 14 Mar 2009 20:06:07 +0000 (17:06 -0300)]
V4L/DVB (11042): v4l2-api: Add definitions for V4L2_MPEG_STREAM_VBI_FMT_IVTV payloads

This addition to the v4l2-api add definitions for the constants and
data structures used for sliced VBI data insertion into MPEG streams triggered
by V4L2_MPEG_STREAM_VBI_FMT_IVTV.  This simply declares what the ivtv and
cx18 drivers and MythTV have already been doing and provides a proper data
structure definition to user space.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11040): gspca - most jpeg subdrivers: Have the JPEG quality settable.
Jean-Francois Moine [Mon, 2 Mar 2009 09:40:52 +0000 (06:40 -0300)]
V4L/DVB (11040): gspca - most jpeg subdrivers: Have the JPEG quality settable.

The JPEG quality of the images (quantization tables) is now settable by the
VIDIOC_S_JPEGCOMP ioctl.

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11039): gspca - most jpeg subdrivers: Change the JPEG header creation.
Jean-Francois Moine [Tue, 3 Mar 2009 08:33:41 +0000 (05:33 -0300)]
V4L/DVB (11039): gspca - most jpeg subdrivers: Change the JPEG header creation.

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11035): mt9t031 bugfix
Philippe Rétornaz [Fri, 13 Mar 2009 12:42:32 +0000 (09:42 -0300)]
V4L/DVB (11035): mt9t031 bugfix

- The video device is not allocated when mt9t031_init() is called, don't use
it in debug printk.

- The clock polarity is inverted in mt9t031_set_bus_param(), use the correct
one.

Signed-off-by: Philippe Rétornaz <philippe.retornaz@epfl.ch>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11034): soc-camera: remove now unused gpio member of struct soc_camera_link
Sascha Hauer [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11034): soc-camera: remove now unused gpio member of struct soc_camera_link

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11033): mt9v022: allow setting of bus width from board code
Sascha Hauer [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11033): mt9v022: allow setting of bus width from board code

This patch removes the phytec specific setting of the bus width
and switches to the more generic query_bus_param/set_bus_param
hooks

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11032): mt9m001: allow setting of bus width from board code
Sascha Hauer [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11032): mt9m001: allow setting of bus width from board code

This patch removes the phytec specific setting of the bus width
and switches to the more generic query_bus_param/set_bus_param
hooks

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11031): pcm990 baseboard: add camera bus width switch setting
Sascha Hauer [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11031): pcm990 baseboard: add camera bus width switch setting

Some Phytec cameras have a I2C GPIO expander which allows it to
switch between different sensor bus widths. This was previously
handled in the camera driver. Since handling of this switch
varies on several boards the cameras are used on, the board
support seems a better place to handle the switch

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11030): soc-camera: add board hook to specify the buswidth for camera sensors
Sascha Hauer [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11030): soc-camera: add board hook to specify the buswidth for camera sensors

Camera sensors have a native bus width say support, but on some
boards not all sensor data lines are connected to the image
interface and thus support a different bus width than the sensors
native one. Some boards even have a bus driver which dynamically
switches between different bus widths with a GPIO.

This patch adds a hook which board code can use to support different
bus widths.

Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11029): video: use videobuf_waiton() in sh_mobile_ceu free_buffer()
Magnus Damm [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11029): video: use videobuf_waiton() in sh_mobile_ceu free_buffer()

Make sure videobuf_waiton() is used before freeing a buffer.
Without this fix we may return the buffer to the allocator
before the bus mastering operation is finished.

Reported-by: Matthieu CASTET <matthieu.castet@parrot.com>
Tested-by: Kuninori Morimoto <morimoto.kuninori@renesas.com>
Signed-off-by: Magnus Damm <damm@igel.co.jp>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11028): ov772x: use soft sleep mode in stop_capture
Kuninori Morimoto [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11028): ov772x: use soft sleep mode in stop_capture

Signed-off-by: Kuninori Morimoto <morimoto.kuninori@renesas.com>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11027): soc-camera: configure drivers with a default format at probe time
Guennadi Liakhovetski [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11027): soc-camera: configure drivers with a default format at probe time

Currently soc-camera doesn't set up any image format without an explicit
S_FMT. According to the API this should be supported, for example,
capture-example.c from v4l2-apps by default doesn't issue an S_FMT. This
patch moves negotiating of available host-camera format translations to
probe() time, and restores the state from the last close() on the next
open(). This is needed for some drivers, which power down or reset
hardware after the last user closes the interface. This patch also has a
nice side-effect of avoiding multiple allocation anf freeing of format
translation tables.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11026): sh-mobile-ceu-camera: set field to the value, configured at open()
Guennadi Liakhovetski [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11026): sh-mobile-ceu-camera: set field to the value, configured at open()

For the case, that we have to capture with a default format, i.e., when the
user doesn't call S_FMT, we have to use the field value according to the
default, configured at open() time.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11025): soc-camera: configure drivers with a default format on open
Guennadi Liakhovetski [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11025): soc-camera: configure drivers with a default format on open

Currently soc-camera doesn't set up any image format without an explicit S_FMT.
It seems this should be supported, since, for example, capture-example.c from
v4l2-apps by default doesn't issue an S_FMT. This patch configures a default
image format on open().

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11024): soc-camera: separate S_FMT and S_CROP operations
Guennadi Liakhovetski [Fri, 13 Mar 2009 09:08:20 +0000 (06:08 -0300)]
V4L/DVB (11024): soc-camera: separate S_FMT and S_CROP operations

As host and camera drivers become more complex, differences between S_FMT and
S_CROP functionality grow, this patch separates them.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11022): zoran/bt819: use new notify functionality.
Hans Verkuil [Sun, 8 Mar 2009 20:04:38 +0000 (17:04 -0300)]
V4L/DVB (11022): zoran/bt819: use new notify functionality.

Bt819 needs the parent driver to drive a GPIO pin low and high in order to
reset its fifo. Use the new notify callback for this.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (11021): v4l2-device: add a notify callback.
Hans Verkuil [Sun, 8 Mar 2009 20:02:10 +0000 (17:02 -0300)]
V4L/DVB (11021): v4l2-device: add a notify callback.

Add a notify callback to v4l2_device to let sub-devices notify their
parent of special events.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (10988): v4l2-dev: use parent field if the v4l2_device has no parent set.
Hans Verkuil [Fri, 13 Mar 2009 13:03:04 +0000 (10:03 -0300)]
V4L/DVB (10988): v4l2-dev: use parent field if the v4l2_device has no parent set.

Normally the parent device of v4l2_device is used as the video device
node's parent. But if it was not set, then use the parent field in the
video_device struct.

This is needed in the cx88 driver, which has one core v4l2_device but
creates multiple pci devices (one each for raw and mpeg video).

So you cannot associate the core v4l2_device with a particular PCI device,
but you can do that for each video_device.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
15 years agoV4L/DVB (10987): cx23885: fix crash on non-netup cards
Hans Verkuil [Fri, 13 Mar 2009 16:24:19 +0000 (13:24 -0300)]
V4L/DVB (10987): cx23885: fix crash on non-netup cards

The new support for the CX23885_BOARD_NETUP_DUAL_DVBS2_CI board broke the
existing boards. Interrupts for the netup part were enabled and handled
without testing whether the current board actually had a netup -> instant
and fatal crash.

I've added tests to do this only for the CX23885_BOARD_NETUP_DUAL_DVBS2_CI
board.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Reviewed-by: Michael Krufky <mkrufky@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>