The VIDIOC_G_STD ioctl may not be present in the case of radio receivers.
In that case G_STD will return an error. The v4l1-compat layer should not
attempt to propagate that error to the caller, instead it should be
ignored.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Steven Toth [Thu, 27 Aug 2009 21:08:21 +0000 (18:08 -0300)]
V4L/DVB (12942): SAA7164: HVR2250 changes related to attach time tuner configuration
Ensure that by default all tuners are set correctly to master/slave mode.
For all HVR2250's, ensure slave based tuners are caliberated during attach
to avoid locking problems on tuner# above channel 91.
HVR2200 tuner attach time to be reviewed in a future patch.
Signed-off-by: Steven Toth <stoth@kernellabs.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Steven Toth [Wed, 12 Aug 2009 15:14:37 +0000 (12:14 -0300)]
V4L/DVB (12940): SAA7164: IRQ / message timeout related change
In some cases we're seeing large timeouts on commands. I'm changing the
implementation so that the deferred worker checks the PCI bus for
any messages and signals the waiting caller accordingly. The previous
mechanism was too unreliable.
Signed-off-by: Steven Toth <stoth@kernellabs.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Steven Toth [Sat, 8 Aug 2009 13:22:02 +0000 (10:22 -0300)]
V4L/DVB (12938): SAA7164: Increase the firmware command timeout to avoid firmware errors.
The firmware typically responds in < 50ms and, via the interrupts and
deferred work queue the caller (blocked in the driver) is signalled very
efficiently. In a highly stressed system this can take many multiples
of seconds. So, we need a larger maximum timeout for busy systems.
Signed-off-by: Steven Toth <stoth@kernellabs.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Steven Toth [Sun, 10 May 2009 00:17:28 +0000 (21:17 -0300)]
V4L/DVB (12923): SAA7164: Add support for the NXP SAA7164 silicon
This patch adds support for all of the known shipping Hauppauge HVR-2200
and HVR-2250 boards. Digital TV ATSC/QAM and DVB-T is enabled at this
time. Both tuners are supported.
Volatiles and typedefs need rework, the rest is coding style compliant.
Signed-off-by: Steven Toth <stoth@kernellabs.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Uwe Bugla [Mon, 17 Aug 2009 11:09:06 +0000 (08:09 -0300)]
V4L/DVB (12902): Documentation: synchronize documentation for Technisat cards
The patch does the following:
a. it synchronizes /Documentation/dvb/technisat.txt to the current kernel
design
b. it syncs Trent Piepho' s latest changes to the flexcop driver concept:
implementation of the dvb pll library for some cards that need it
V4L/DVB (12892): DVB-API: add support for ISDB-T and ISDB-Tsb (version 5.1)
This patch increments the DVB-API to version 5.1 in order to reflect the addition of ISDB-T and ISDB-Tsb on Linux' DVB-API.
Changes in detail:
- added a small document to describe how to use the API to tune to an ISDB-T or ISDB-Tsb channel
- added necessary fields to dtv_frontend_cache
- added a smarter clear-cache function which resets all fields of the dtv_frontend_cache
- added a TRANSMISSION_MODE_4K to fe_transmit_mode_t
By default, when a callout is used, DocBook will try to use
a graphics image for callouts. This requires that the graphics
to be copied to the documentation directory. As this is not
done, use the text callouts: (1), (2), ...
DocBook makefile expects that the documents to be in tmpl extension,
since it has some preprocessing that it is done on it. This preprocessing
is not needed currently, but, as it removes the xml versions, we're
forced to rename anyway.
Andy Walls [Mon, 7 Sep 2009 17:32:45 +0000 (14:32 -0300)]
V4L/DVB (12884): cx18: Eliminate warning about discarding 'const' is assignment for IR init
i2c-kbd-i2c allows a bridge driver to pass information about IR configuration,
but uses a "void *" to pass along what is essentially constant data. This
change casts a const * to a void * to remove the warning.
Signed-off-by: Andy Walls <awalls@radix.net> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
V4L/DVB (12881): em28xx: fix codingstyle issues in em28xx-video.c
Fix some codingstyle issues introduced during the addition of em28xx VBI
support. The patch makes no functional changes other than converting a few
debug printk() statements to em28xx_isocdbg.
V4L/DVB (12878): gspca - m5602-s5k4aa: Add vflip quirk for the GX700
The MSI GX700 is a tricky machine to support. Some revisions do need the sensor flipped, but not all.
Add another quirk, distinguished by its BIOS date.
V4L/DVB (12870): tda18271: update temperature compensation calculatation formula
Update the tda18271c2_rf_tracking_filters_correction function to include
the modified temperature compensation calculatation formula as described
in Rev.04 of the TDA18271HD datasheet.
Michael Krufky [Sun, 30 Aug 2009 06:07:10 +0000 (03:07 -0300)]
V4L/DVB (12867): tda18271: ensure that configuration options are set for multiple instances
For the case of multiple tuner instances, ensure that non-default
configuration options are saved into the driver's state.
This resolves an issue where a configuration option may not be
carried into the driver if the analog side of a hybrid driver
initializes before the digital side.
Signed-off-by: Michael Krufky <mkrufky@kernellabs.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Michael Krufky [Sun, 11 May 2008 15:46:52 +0000 (12:46 -0300)]
V4L/DVB (12861): tda18271: add support for additional low-power standby modes
By default, the driver enters standby mode with slave tuner output
loop thru enabled and xtal oscillator on.
Not all designs require that slave tuner output loop thru and
xtal oscillator remain active while in standby mode, so two
additional standby modes have been added:
- standby mode with xtal oscillator on (loop thru off)
- total power off
Signed-off-by: Michael Krufky <mkrufky@kernellabs.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Antti Palosaari [Sat, 12 Sep 2009 23:46:30 +0000 (20:46 -0300)]
V4L/DVB (12755): af9015: improve usb control message function slightly
* define names for few values
* decrease buffer len by one byte which was not used
* add check for buffer overflow for sure
* indentation fixes
* remove useless 0 len check from memcpy
It should not happen and if it happens memcpy should not do anything.
Signed-off-by: Antti Palosaari <crope@iki.fi> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
V4L/DVB (12753): af9015: [1/2] fix USB TS configuration
Fix wrongly configured USB2.0 TS EP.
Signed-off-by: Jose Alberto Reguero <jareguero@telefonica.net> Signed-off-by: Antti Palosaari <crope@iki.fi> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
V4L/DVB (12749): em28xx: remove unneeded code that set VINCTRL register
Remove redundant call to set the vinctrl register. This eliminates any
ambiguity as to how the register is configured (since it is now always set in
em28xx_set_outfmt).
We need to implement the g_std call, or else the default norm always gets
returned, which breaks VBI capturing if you had changed the standard to NTSC
using s_std.
I had temporarily changed the default norm to NTSC so that zvbi-ntsc-cc
wouldn't choke, so now that we are returning the correct value, switch it back
to PAL as the default.
V4L/DVB (12744): em28xx: restructure fh/dev locking to handle both video and vbi
The current locking infrastructure didn't support having multiple fds accessing
the device (such as video and vbi). Rework the locking infrastructure,
borrowing the design from cx88.
V4L/DVB (12742): em28xx: add raw VBI support for NTSC
Add support for raw VBI capture for the em28xx bridge, currently only for
NTSC. Support for PAL capture to follow shortly (including the removal of
numerous hard-coded NTSC-specific sizes for capture buffers, etc).
Note that the code currently changes the default current norm from PAL to
NTSC (so that zvbi-ntsc-cc works properly). The default norm really should
be moved into a board-level parameter.
V4L/DVB (12741): em28xx: make video isoc stream work when VBI is enabled
Add code enabling the VBI registers for variants of the em28xx chip that
support VBI, and make sure the isoc streaming code continues to work for
the video component of the stream (note the video and vbi data arrive
intermixed on the same isoc pipe).
Note that this version just drops the actual VBI data onto the floor as
opposed to processing it. The "#ifdef 0" tags are for the videobuf code that
appears in the next patch in this series.
We created a separate version of the isoc_copy version for parsing the version
of the stream that includes VBI data. In theory, they might be able to be
merged at some point in the future, but the initial goal is to ensure that we
do not cause any regressions with devices that do not have VBI support.
V4L/DVB (12734): cx25821: Fix some compilation troubles
Lindent caused some compilation breakages. There were also others
related to some other changes at kernel KABI.
There's still one missing warning fix against 2.6.30:
/home/v4l/cx25821/v4l/cx25821-alsa.c: In function 'cx25821_audio_initdev':
/home/v4l/cx25821/v4l/cx25821-alsa.c:706: warning: 'snd_card_new' is deprecated (declared at include/sound/core.h:306)
/home/v4l/master/v4l/cx25821-audups11.c:99: error: implicit declaration of function 'lock_kernel'
/home/v4l/master/v4l/cx25821-audups11.c:112: error: implicit declaration of function 'unlock_kernel'