Instead of relying on the cdev release callback we should rely on the
release callback from the device struct. This requires that we use
get_device/put_device to do proper refcounting. In order to do this
safely v4l2-dev.c now sets up its own file_operations that call
out to the driver's ops.
Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

usbvideo_ClientIncModCount may return value < 0 in the case of error, not > 0.
Signed-off-by: NJiri Slaby <jirislaby@gmail.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Converted to v4l2_subdev. While doing that I also discovered a stray 'break'
that broke the S_CTRL handling. Fixed that as well.
Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

The em28xx driver can be coupled to an anciliary AC97 chip. This patch
allows read/write AC97 registers directly.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

The comment said CX2584X instead of CX2341X.
Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Add the missing video_device documentation to v4l2-framework.txt.
Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

get_index requires a valid parent device in order to discover which
indices are in use. Some drivers (e.g. pvrusb2) do not set the parent
device. In that case just return 0.
Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

The bt832 i2c driver was never used or even compiled and is no longer
maintained. It is now removed completely.
Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Use a consistent crystal value of 28.636360 MHz for computing all PLL
parameters so clocks don't have relative error due to assumed crystal
value mismatches.  Also aimed to have all PLLs run their VOCs at close to
400 MHz to minimze the error of these PLLs as frequency synthesizers. Also
set the VDCLK and AIMCLK PLLs to sane values before the APU and CPU firmware
are loaded.  Also fixed I2S Master clock dividers.

Many thanks to Mike Bradley and Jeff Campbell for reporting this problem and
suggesting the solution, researching and experimenting, and performing
extensive testing to support their suggested solution.
Reported-by: NJeff Campbell <jac1dlists@gmail.com>
Reported-by: NMike Bradley <mike.bradley@incanetworks.com>
Signed-off-by: NAndy Walls <awalls@radix.net>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Disable AV_LOCK, locking of audio PLL to video PLL in the cx18-av-core for
analog captures.  It seems to have adverse effects on captures from
SVideo and CVBS sources with current clock & crystal settings in the driver.

Many thanks to Jeff Campbell and Mike Bradley for reporting this problem,
and suggesting the solution and performing extensive testing to support their
suggestion.
Reported-by: NJeff Campbell <jac1dlists@gmail.com>
Reported-by: NMike Bradley <mike.bradley@incanetworks.com>
Signed-off-by: NAndy Walls <awalls@radix.net>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Added all missing v4l1/2 ioctls and fix several broken conversions.
Partially based on work done by Cody Pisto <cpisto@gmail.com>.
Tested-by: NBrandon Jenkins <bcjenkins@tvwhere.com>
Signed-off-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

vidioc_g_ctrl() were using an uneeded confusing logic. Instead, use the
direct approach.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

This patch converts the gp8psk module to use the S2API.
It pretends to be  DVB-S2 capable in order to allow the various
supported modulations (8PSK, QPSK-Turbo, etc), and keep software
compatibility with the S2API patches for Mythtv and VDR.

Signed-off by: Alan Nisota <alannisota@gmail.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

This patch uses the same code for enumberating video formats that are
present on cx88, bttv and saa7134 drivers.
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

The changeset 78aa52a159cf introduced a bug on em28xx: buffer setup should be
awaking xc3028. Instead, since we didn't specify the tuner mode, the device
were going to sleep, due to the lack of tuner mode when asking tuner to handle
VIDIOC_S_FREQUENCY:

xc2028 0-0061: Device is Xceive 3028 version 1.0, firmware version 2.7
xc2028 0-0061: divisor= 00 00 14 d0 (freq=83.250)
xc2028 0-0061: Putting xc2028/3028 into poweroff mode.
xc2028 0-0061: Printing sleep stack trace:
Pid: 10936, comm: mplayer Tainted: P   M      2.6.27.8 #1

Call Trace:
 [<ffffffffa0b759ea>] xc2028_sleep+0x89/0x1ab [tuner_xc2028]
 [<ffffffffa0b48fb9>] tuner_s_frequency+0xf5/0x165 [tuner]
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

The init_fw option was broken for the HVR-950q because we would call the reset
callback inside of dvb_attach() and the callback had not been setup yet.

Michael Krufky (who added the init_fw feature) says it's no longer required,
so just remove the option completely.

Thanks to user Zzeiss from #linuxtv chat for reporting the issue and
Michael Krufky <mkrufky@linuxtv.org> for proposing the fix.
Signed-off-by: NDevin Heitmueller <dheitmueller@linuxtv.org>
Acked-by: NMichael Krufky <mkrufky@linuxtv.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Don't operate under the assumption that every device that uses the em2820
default USB ID is a Prolink PlayTV USB.  We have an eeprom hash, so use that,
since otherwise we cannot support other devices with the 2820 default USB ID
(such as the ADS Tech Instant TV USB USBAV-704)
Signed-off-by: NDevin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Add the em2870 to the list of known em28xx chip ids.
Signed-off-by: NDevin Heitmueller <dheitmueller@linuxtv.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

This cleans-up the advise/acquire methods.

This has been tested on the hvr-1300/4000 and assumed to be
correct on the hvr-3000.

This update also fixes analogue tuning on the hvr-1300
when in blackbird mode.
Signed-off-by: NDarron Broad <darron@kewl.org>
Cc: Steven Toth <stoth@hauppauge.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Slow tuning, EG:

> szap-s2 -r -c 39.FTATV "1=Pgm1;Net1"
reading channels from file '39.FTATV'
zapping to 3 '1=Pgm1;Net1':
delivery DVB-S, modulation QPSK
sat 0, frequency 11140 MHz V, symbolrate 1425000, coderate 5/6, rolloff 0.35
vpid 0x0200, apid 0x0300, sid 0x0001
using '/dev/dvb/adapter0/frontend0' and '/dev/dvb/adapter0/demux0'
status 00 | signal c2c0 | snr 0000 | ber 00012d4a | unc 00000000 |
status 03 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 03 | signal c040 | snr 0000 | ber 00000000 | unc 00000000 |
status 03 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 03 | signal c040 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal c040 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 01 | signal c040 | snr 0000 | ber 00000000 | unc 00000000 |
status 03 | signal c040 | snr 0000 | ber 00000000 | unc 00000000 |
status 1f | signal c040 | snr bb33 | ber 00000000 | unc 00000000 | FE_HAS_LOCK

Appears to be resolved by dropping zig-zag tuning and relying on
hardware only:

> szap-s2 -r -c 39.FTATV "1=Pgm1;Net1"
reading channels from file '39.FTATV'
zapping to 3 '1=Pgm1;Net1':
delivery DVB-S, modulation QPSK
sat 0, frequency 11140 MHz V, symbolrate 1425000, coderate 5/6, rolloff 0.35
vpid 0x0200, apid 0x0300, sid 0x0001
using '/dev/dvb/adapter0/frontend0' and '/dev/dvb/adapter0/demux0'
status 01 | signal ff80 | snr 0000 | ber 00000000 | unc 00000000 |
status 1f | signal ff80 | snr c199 | ber 00000000 | unc 00000000 | FE_HAS_LOCK
status 1f | signal c040 | snr c199 | ber 00000000 | unc 00000000 | FE_HAS_LOCK
status 1f | signal ff80 | snr c199 | ber 00000000 | unc 00000000 | FE_HAS_LOCK

Tuning to a weaker channel shows that the hardware recovers sync:

status 1f | signal fe40 | snr 5199 | ber 00000000 | unc 00000000 | FE_HAS_LOCK
status 03 | signal fd40 | snr 5000 | ber 00000000 | unc 00000000 |
status 1f | signal fe40 | snr 5000 | ber 00000000 | unc 00000000 | FE_HAS_LOCK
status 03 | signal fd80 | snr 5000 | ber 00000000 | unc 00000000 |
status 01 | signal fd80 | snr 4e66 | ber 00000000 | unc 00000000 |
status 1f | signal fe40 | snr 5000 | ber 00000000 | unc 00000000 | FE_HAS_LOCK
status 03 | signal fe40 | snr 5000 | ber 00000000 | unc 00000000 |
status 01 | signal fd80 | snr 0000 | ber 00000000 | unc 00000000 |
status 1f | signal fe40 | snr 5199 | ber 00000000 | unc 00000000 | FE_HAS_LOCK

Also, tuning, moving the dish and then restoring the dish also retunes.
Signed-off-by: NDarron Broad <darron@kewl.org>
Cc: Steven Toth <stoth@hauppauge.com>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

It has been noticed that in HW tuning mode in cx24123.c that
a check is made to not alter the status var when in one shot
mode, a simpler solution is to not update here for all cards.

Cc: Steven Toth <stoth@hauppauge.com>
Signed-off-by: NDarron Broad <darron@kewl.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Thanks to AKPM for spotting this.

Cc: Steven Toth <stoth@hauppauge.com>
Signed-off-by: NDarron Broad <darron@kewl.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>

Delsys was missing when searching for FEC in table.

This fixes FEC lookup for DVB-S2 QPSK modulation.

Cc: Steven Toth <stoth@hauppauge.com>
Signed-off-by: NDarron Broad <darron@kewl.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab@redhat.com>