pxa_camera: remove init() callback

pxa_camera init() callback is sometimes abused to setup MFP for PXA CIF, or
even to request GPIOs to be used by the camera *sensor*. These initializations
can be performed statically in machine init functions.

The current semantics for this init() callback is ambiguous anyways, it is
invoked in pxa_camera_activate(), hence at device node open, but its users use
it like a generic initialization to be done at module init time (configure
MFP, request GPIOs for *sensor* control).

Signed-off-by: Antonio Ospite <ospite@studenti.unina.it>
Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

soc-camera: adjust coding style to match V4L preferences

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

soc-camera: ov772x: Modify buswidth control

This patch removes "buswidth" struct member, and sets the default buswidth
to the natively supported 10 bit. You can select 8 bit buswidth by new flag.
This patch also modify ap325rxa/migor setup.c

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>

tw9910: use TABs for indentation

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

soc-camera: update mt9v022 to take into account board signal routing

Use soc_camera_apply_sensor_flags() in mt9v022 to account for any inverters in
video signal paths.

Signed-off-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: az6027: az6027_read_mac_addr is currently unused

drivers/media/dvb/dvb-usb/az6027.c:759: warning: ‘az6027_read_mac_addr’ defined but not used

While there's some code that uses it, it is currently commented. So, comment also
the function itself.

CC: Manu Abraham <abraham.manu@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: az6027: IR RC keys are using the old struct with 3 parameters, instead of 2

drivers/media/dvb/dvb-usb/az6027.c:390: warning: excess elements in struct initializer
drivers/media/dvb/dvb-usb/az6027.c:390: warning: (near initialization for ‘az6027_rc_keys[0]’)
drivers/media/dvb/dvb-usb/az6027.c:391: warning: excess elements in struct initializer
drivers/media/dvb/dvb-usb/az6027.c:391: warning: (near initialization for ‘az6027_rc_keys[1]’)

CC: Manu Abraham <abraham.manu@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: AZ6027: Fix checkpatch violations

Signed-off-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: AZ6027: Add driver supported ID's

Signed-off-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: AZ6027: Update Build

Signed-off-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: AZ6027: Add driver supported ID's

Signed-off-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: AZ6027: Initial import of the driver

Signed-off-by: Manu Abaraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: tlg2300: remove the country code for analog tv and radio

video :
use the V4L2_STD macros to select the proper audio setting.

radio :
add preemphasis ctr.
test it by the command:
v4l2-ctl -d /dev/radio0 --set-ctrl=pre_emphasis_settings=1

[mchehab@redhat.com: folded documentation patch]
Signed-off-by: Huang Shijie <shijie8@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: Add driver for Telegent tlg2300

pd-common.h contains the common data structures, while
vendorcmds.h contains the vendor commands for firmware.

[mchehab@redhat.com: Folded the 10 patches with the driver]
Signed-off-by: Huang Shijie <shijie8@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: Kworld 315U remote support

Adds remote support for the Kworld 315U device

I have added the change for the IR_TYPE_NEC that Mauro suggested.

Note: I believe I got most of the mappings correct.  Though the
source and shutdown button probably could be mapped to something
better.

[mchehab@redhat.com: Fix CodingStyle]
Signed-off-by: Franklin Meng <fmeng2002@yahoo.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: drivers/media/radio/si470x/radio-si470x-usb.c fix use after free

In si470x_usb_driver_disconnect() radio->disconnect_lock is accessed
after it is freed. This fixes the problem.

Coverity CID: 2530

Signed-off-by: Darren Jenkins <darrenrjenkins@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: af901x: inform NXP TDA18218 tuner as know but not supported

There is no driver for new NXP TDA18218 tuner which is used by few recent
af9015 designs. af9015 uses number 179 as ID for that tuner. Inform this
tuner is not supported when device using that tuner is meet.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: af9015: MYGICTV U718 remote autodetection

Autodetect MYGICTV U718 from eeprom hash and select remote automatically.
Remove old iManufacturer string compare based remote selection logic.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: af9015: A-Link DTU(m) remote autodetection

Autodetect A-Link DTU(m) based eeprom hash and select correct remote.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: af9015: support for Leadtek WinFast DTV2000DS

Add USB ID 0413:6a04 for Leadtek WinFast DTV2000DS.

Thanks to Gavin Ramm <gavin_ramm@hotmail.com> for reporting this.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: af9015: support for DigitalNow TinyTwin v2

Add USB ID 1b80:e402 for DigitalNow TinyTwin version 2.

Thanks to Trevor Phillips <trevor.phillips@gmail.com> for reporting this.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: DVB: ngene, fix memset parameters

Switch second and third memset parameter to stamp the length buffer bytes
by 0xff's, not 255 bytes by low 8 bits of Length.

Cc: linux-media@vger.kernel.org, linux-kernel@vger.kernel.org, jirislaby@gmail.com, Matthias Benesch <twoof7@freenet.de>, Ralph Metzler <rjkm@metzlerbros.de>, Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Matthias Benesch <twoof7@freenet.de>
Cc: Ralph Metzler <rjkm@metzlerbros.de>
Cc: Oliver Endriss <o.endriss@gmx.de>
Cc: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: em28xx: add Dikom DK300 hybrid USB tuner

Adds digital and analogue tv support for Dikom DK300 hybrid usb card.

Not working: remote controller

To be done: it seems that with the proposed patch the digital demodulator
remains activated if the tuner is switched from digital to analogue mode.

Workaround is to unplug and replug the device when switching from digital to
analogue. If someone can explain how to verify the gpio settings using the
usbsnoop, the above issue perhaps can be resolved.

Signed-off-by: Andrea Amorosi <Andrea.Amorosi76@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: drivers/media: Correct NULL test

In each case, the NULL test has been performed already.

A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)

// <smpl>
@r@
expression *x;
expression e;
identifier l;
@@

if (x == NULL || ...) {
    ... when forall
    return ...; }
... when != goto l;
    when != x = e
    when != &x
*x == NULL
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: dst: fixes for DVB-C Twinhan VP2031

Remove check  "state->dst_type == DST_DTYPE_IS_CABLE"  in function
dst_get_tuna (around line 1352) to select the correct checksum
computation

Fill in the .caps field in struct dst_dvbc_ops (around line 1824) with
all the supported QAM modulation methods to match the capabilities of
the card as implemented in function dst_set_modulation (around line
502). Note that beginning with linux kernel version 2.6.32 the
modulation method is checked (by function
dvb_frontend_check_parameters in file
drivers/media/dvb/dvb-core/dvb_frontend.c) and thus tuning fails if
you use a modulation method that is not present in the .caps field.

This patch has been tested on a Twinhan VP2031A DVB-C card with the
2.6.32.2 kernel.

Signed-off-by: Klaas de Waal <klaas.de.waal@gmail.com>
Signed-off-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: saa7134 can capture 720x480 when capturing NTSC

When capturing NTSC, the saa7134 can capture 720x480.

While doing Laserdisc captures using this card, I noticed that right side
was truncated/cropped. The highest geometry the driver allows is 704x480,
even though in Windows XP it is 720x480. This results in no cropping
and the same results as in Windows.

Tested on an AverMedia GO 007 FM Plus.

Signed-off-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: STV0900 Cut 3.0 AGC2 fix for NetUP Dual DVB-S2-CI card

if the AGC2 > 1400 then no signal on this frequency for STV0900 Cut 3.0. For
Cut 2.0 this value is 700.

Signed-off-by: Abylay Ospan <aospan@netup.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: Fix logic for Leadtek winfast tv usbii deluxe

As pointed by Magnus Alm <magnus.alm@gmail.com>, commit 99dbd128bb applied
a hunk at the wrong place. This patch moves the code to the right place.

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

V4L/DVB: get_dvb_firmware: Fix typo, sort list of components

Commiter: Oliver Endriss <o.endriss@gmx.de>

Fix typo. Sort list of components for better readability.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: get_dvb_firmware: Add function to retrieve nGene firmwares

Commiter: Oliver Endriss <o.endriss@gmx.de>

Use 'get_dvb_firmware ngene' to download 'ngene_15.fw' and 'ngene_17.fw'.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: using vmalloc requires include linux/vmalloc.h

After merging the scsi-post-merge tree, today's linux-next build (powerpc
allyesconfig) failed like this:

drivers/media/dvb/ngene/ngene-core.c: In function 'ngene_release_buffers':
drivers/media/dvb/ngene/ngene-core.c:1443: error: implicit declaration of function 'vfree'
drivers/media/dvb/ngene/ngene-core.c: In function 'ngene_get_buffers':
drivers/media/dvb/ngene/ngene-core.c:1454: error: implicit declaration of function 'vmalloc'
drivers/media/dvb/ngene/ngene-core.c:1454: warning: assignment makes pointer from integer without a cast
drivers/media/dvb/ngene/ngene-core.c:1461: warning: assignment makes pointer from integer without a cast
drivers/media/dvb/ngene/ngene-core.c:1467: warning: assignment makes pointer from integer without a cast
drivers/media/dvb/ngene/ngene-core.c: In function 'ngene_probe':
drivers/media/dvb/ngene/ngene-core.c:1836: warning: assignment makes pointer from integer without a cast

Caused by commit b54c755d6263efe376f3dfc8c72c8e7b4fb4f4c1 ("V4L/DVB: ngene: Initial check-in").

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: mfd: Add support for the timberdale FPGA

The timberdale FPGA is found on the Intel in-Vehicle Infotainment reference board
russelville.

The driver is a PCI driver which chunks up the I/O memory and distributes interrupts
to a number of platform devices for each IP inside the FPGA.

Signed-off-by: Richard Röjfors <richard.rojfors@pelagicore.com>
Signed-off-by: Samuel Ortiz <sameo@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: radio: Add radio-timb

This patch add supports for the radio system on the Intel Russellville board.

It's a In-Vehicle Infotainment board with a radio tuner and DSP.

This umbrella driver has the DSP and tuner as V4L2 subdevs and calls them
when needed.

Signed-off-by: Richard Röjfors <richard.rojfors@pelagicore.com>
Reviewed-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Fix I2C initialisation

Use I2C_CLASS_TV_DIGITAL, not I2C_ADAP_CLASS_TV_DIGITAL.
Remove initialisation of bus_lock mutex.
Thanks to Mauro for pointing out.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Thanks-to: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Make checkpatch.pl happy

Make checkpatch.pl happy.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Use correct name for cineS2 cards

Cards 18c3:abc3/18c3:abc4 should be displayed as 'Linux4Media cineS2'.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Take care of late interrupts

Request might already have been processed when the interrupt arrives.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Workaround for command timeout

Implement workaround for comamnd timeout.
Use FW15 and avoid switching DMA on and off.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Set ADC ranges to 1Vpp

Set ADC1 and ADC2 range of the stv0900 to 1Vpp.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Adapt to current frontend drivers

Adapt to current stv090x + stv6110 drivers.

[mchehab@redhat.com: removed compilation hack added by me on a prev patch]
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Mystique SaTiX-S2 Dual

Identify card 18c3:db01 as 'Mystique SaTiX-S2 Dual'.

Signed-off-by: Matthias Benesch <twoof7@freenet.de>
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: [nGene] Trivial I2C related fixes/cleanups

Trivial I2C related fixes/cleanups
 - Remove Unused I2C ID
 - point the I2C device to the physical device

Signed-off-by: Manu Abraham <manu@linuxtv.org>
Acked-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Code cleanup

Remove/comment-out unused code, make some functions/declarations static.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Added module parameter "one_adapter"

If parameter "one_adapter" is set, only one adapter per device will be attached.
Otherwise an adapter for every frontend will be attached.

Signed-off-by: Matthias Benesch <twoof7@freenet.de>
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: DVB: Export dvb_ringbuffer_flush() again

Add EXPORT_SYMBOL(dvb_ringbuffer_flush) again,
replace my_dvb_ringbuffer_flush() in ngene driver.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Remove kernel version check for dvb_[un]register_adapter

Remove kernel version check for dvb_[un]register_adapter.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Change severity of some diagnostic messages

Changed severity of some diagnostic messages.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Speed-up tuning

Remove stray mleep() which prevented fast tuning.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Fix unloading of module

Fixed unloading of module ngene.

Signed-off-by: Roland Praml <pram@gmx.de>
Signed-off-by: Matthias Benesch <twoof7@freenet.de>
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Remove firmware debugging

Remove firmware debugging, because setting ngene_fw_debug and requesting
firmware from file causes a kernel error.

Signed-off-by: Matthias Benesch <twoof7@freenet.de>
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Added Media-Pointer MP-S2/CineS2 DVB-S2 Twin Tuner

Added support for nGene based Media-Pointer DVB-S2 twin tuner cards.

[mchehab@redhat.com: Fix compilation error: unknown field ‘ref_clk’]

Signed-off-by: Matthias Benesch <twoof7@freenet.de>
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Additional clean-up

Remove and disable unnecessary code. Disable define ONE_ADAPTER.

Signed-off-by: Matthias Benesch <twoof7@freenet.de>
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Always use firmware loader

Always use firmware loader.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Driver compiles now

Remove LNBH21 routines, disable code which broke compilation.

Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: ngene: Initial check-in

Add Micronas nGene PCIe bridge driver.

The source code was provided by Micronas / Ralph Metzler,
and has been reformatted to comply with Linux Codingstyle.

Signed-off-by: Matthias Benesch <twoof7@freenet.de>
Signed-off-by: Ralph Metzler <rjkm@metzlerbros.de>
Signed-off-by: Oliver Endriss <o.endriss@gmx.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: hdpvr: add new USB product id

Signed-off-by: Janne Grunau <j@jannau.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: hdpvr-video: cleanup signedness

The fifth parameter of usb_bulk_msg() is a pointer to signed
(see <linux/usb.h>) so also call this function with pointer to signed.

This will remove the following sparse warning (see "make C=1"):
 * warning: incorrect type in argument 5 (different signedness)
       expected int *actual_length
       got unsigned int *<noident>

Signed-off-by: Márton Németh <nm127@freemail.hu>
Signed-off-by: Janne Grunau <j@jannau.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: media: dvb/af9015, add hashes support

So as a final patch, add support for hash and one hash entry
for MSI digi vox mini II:
iManufacturer 1 Afatech
iProduct      2 DVB-T 2
iSerial       3 010101010600001

It is now handled with proper IR and key map tables.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Acked-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: media: dvb/af9015, refactor remote setting

Add af9015_setup structure to hold (right now only remote) setup
of distinct receivers.

Add af9015_setup_match for matching ids against tables.

This is for easier matching different kind of ids against tables
to obtain setups. Currently module parameters and usb vendor ids
are switched into and matched against tables. Hashes will follow.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Acked-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: media: dvb/af9015, factor out remote setting

This is just a code shuffle without functional changes. For easier
review of later changes, i.e. preparation.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Acked-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: media: dvb/af9015, implement eeprom hashing

This will be useful for matching of IR tables later.

We read the eeprom anyway for dumping. Switch the dumping to
print_hex_dump_bytes and compute hash above that by
hash = 0;
for (u32 VAL) in (eeprom):
  hash *= GOLDEN_RATIO_PRIME_32
  hash += VAL; // while preserving endinaness

The computation is moved earlier to the flow, namely from
af9015_af9013_frontend_attach to af9015_read_config, so that
we can access the sum in af9015_read_config already.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Acked-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: cx23885: Add support for LEADTEK WinFast PxTV1200

I tested only tv and composite. Video works fine but no audio.

Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: hdpvr-core: make module parameters local

The default_video_input and default_audio_input module parameters are
only used inside the hdpvr-core.c file so make them static.

This will remove the following sparse warnings (see "make C=1"):
 * warning: symbol 'default_video_input' was not declared. Should it be static?
 * warning: symbol 'default_audio_input' was not declared. Should it be static?

Signed-off-by: Márton Németh <nm127@freemail.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: radio: add support for SAA7706H Car Radio DSP

Initial support for the SAA7706H Car Radio DSP.

It is a I2C device and currently the mute control is supported.

When the device is unmuted it is brought out of reset and initiated using
the proposed intialisation sequence.

When muted the DSP is brought into reset state.

[akpm@linux-foundation.org: include delay.h]

Signed-off-by: Richard Röjfors <richard.rojfors@pelagicore.com>
Cc: Douglas Schilling Landgraf <dougsland@gmail.com>
Cc: Hans Verkuil <hverkuil@xs4all.nl>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: drivers/media/dvb/frontends/stv090x.c: fix use-uninitlalised

Mad guess.

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Manu Abraham <manu@linuxtv.org>
Acked-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: drivers/media/dvb/frontends/stv090x.c: fix use-uninitialised

drivers/media/dvb/frontends/stv090x.c: In function 'stv090x_blind_search':
drivers/media/dvb/frontends/stv090x.c:1967: warning: 'coarse_fail' may be used uninitialized in this function

Cc: Manu Abraham <manu@linuxtv.org>
Cc: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Reviewed-by: Manu Abraham <manu@linuxtv.org>
Acked-by: Manu Abraham <manu@linuxtv.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: drivers/media/video/cx18/cx18-alsa-pcm.c: fix printk warning

drivers/media/video/cx18/cx18-alsa-pcm.c: In function 'cx18_alsa_announce_pcm_data':
drivers/media/video/cx18/cx18-alsa-pcm.c:82: warning: format '%d' expects type 'int', but argument 5 has type 'size_t'

Cc: Andy Walls <awalls@radix.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: dvb: return -ENOMEM if kzalloc failed in dvb_usb_device_init()

If in a cold state and the download succeeded ret is zero, but we
should return -ENOMEM.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: em28xx: fix for "Leadtek winfast tv usbii deluxe"

fix Video/Sound support "Leadtek winfast tv usbii deluxe".

Now, it is working Stereo, IR, Radio, TV, Svideo and Composite.

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

V4L/DVB: smsir: make local variables static

Make the file local parameters static.

This will remove the following sparse warnings (see "make C=1"):
 * warning: symbol 'ir_pos' was not declared. Should it be static?
 * warning: symbol 'ir_word' was not declared. Should it be static?
 * warning: symbol 'ir_toggle' was not declared. Should it be static?

Signed-off-by: Márton Németh <nm127@freemail.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: m920x: Add support for Pinnacle PCTV310e card

Adds support to Pinnacle PCTV310e hybrid tuner card, for DVB-T and
remote control, still no analog video.

[mchehab@redhat.com: fix CodingStyle issues]

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

V4L/DVB: cafe_ccic: Fix lock imbalances

Fix lock imbalances in function device_authorization.

Signed-off-by: Alexander Strakh <strakh@ispras.ru>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: stv0900: dereference of state->internal in fe_stv0900_error stv0900_init_internal()

state->internal allocation may fail as well as the allocation of
stv0900_first_inode or new_node->next_inode in append_internal().

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: cx18: Add missing serialization locks to cx18-dvb

Operations that manipulates the state of the encoder or global instance flags
and variables should be serialized.  This adds some serialization locks that
have been missing from cx18-dvb.  Since the DVB part of the CX23418 doesn't
affect the encoder, no one would likely have ever noticed a race in changing
card instance variables.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: cx18: Increment driver version for the addition of cx18-alsa and fixes

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: cx18: Fix memory leak in cx18-alsa starting of PCM captures

The cx18_open_id is normally dynamically allocated and stored in the
filp->private_data for v4l2 file operations.  The cx18-alsa routines should
not dynamically allocate a cx18_open_id because they never store it anywhere
and never free it.  This change fixes that and plugs a memory leak.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: cx18: Add missing serialization locking to cx18-alsa

File operations that manipulate the state of the encoder, or video or audio
subdevices or some of the stream flags need to be serialized.  Add the
serilization locks to cx18-alsa in a manner consistent with the locking for
the v4l2 file operations.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: cx18: Rename snd_cx18_mixer_lock to snd_cx18_lock and increase visibility

Rename snd_cx18_mixer_lock() to snd_cx18_lock() in anticpation of using it
in the cx18-alsa-pcm.c file routines.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: cx25840: Fix composite detection

If CX25840_VIN1_CH1 and the like is used, input is not detected as composite.
Their value is 0x800000XX and CX25840_COMPONENT_ON is 0x80000200. So

   739 else if ((vid_input & CX25840_COMPONENT_ON) == 0)

this condition never be true.

Signed-off-by: Kusanagi Kouichi <slash@ac.auone-net.jp>
Reviewed-by: Andy Walls <awalls@radix.net>
Acked-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - main: Change version to 2.9.0.

Signed-off-by: Jean-François Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - t613: Support for LT168G sensor.

Signed-off-by: Nicolau Werneck <nwerneck@gmail.com>
Signed-off-by: Jean-François Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - pac7302: Change latin1 text to utf-8.

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - pac7302: Add support for camera button.

Signed-off-by: Márton Németh <nm127@freemail.hu>
Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - main: Add input support for interrupt endpoints.

Signed-off-by: Márton Németh <nm127@freemail.hu>
Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: dvb/bt8xx: Clean-up init and exit functions

The init and exit functions are needlessly complex. Remove the bloat:
* Drop irrelevant/outdated comments.
* Remove useless bt878_pci_driver_registered global variable.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Mauro Carvalho chehab <mchehab@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: Fix VIDIOC_QBUF compat ioctl32

When using VIDIOC_QBUF with memory type set to V4L2_MEMORY_MMAP, the
v4l2_buffer buffer gets unmodified on drivers like uvc (well, only
bytesused field is modified). Then some apps like gstreamer are reusing
the same buffer later to call munmap (eg passing the buffer "length"
field as 2nd parameter of munmap).

It's working fine on full 32bits but on 32bits systems with 64bit
kernel, the get_v4l2_buffer32() doesn't copy length/m.offset values and
then copy garbage to userspace in put_v4l2_buffer32().

This has for consequence things like that in the libv4l2 logs:

libv4l2: v4l2 unknown munmap 0x2e2b0000, -2145144908
libv4l2: v4l2 unknown munmap 0x2e530000, -2145144908

The buffer are not unmap'ed and then if the application close and open
again the device, it won't work and logs will show something like:

libv4l2: error setting pixformat: Device or resource busy

The easy solution is to read length and m.offset in get_v4l2_buffer32().

Signed-off-by: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: zoran: remove variable shadowing

The loop counter j is declared twice in function error_handler().
Remove the redundant declaration.

This will remove the following sparse warning (see "make C=1"):
 * symbol 'j' shadows an earlier one

Signed-off-by: Márton Németh <nm127@freemail.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: zoran: match parameter signedness of g_input_status

The second parameter of g_input_status operation in <media/v4l2-subdev.h>
is unsigned so also call it with unsigned paramter.

This will remove the following sparse warning (see "make C=1"):
 * incorrect type in argument 2 (different signedness)
       expected unsigned int [usertype] *status
       got int *<noident>

Signed-off-by: Márton Németh <nm127@freemail.hu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: cx23885: Wrong command printed in cmd_to_str()

The wrong command was printed for case CX2341X_ENC_SET_DNR_FILTER_MODE,
and a typo in case CX2341X_ENC_SET_PCR_ID.

Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Acked-By: Steven Toth <stoth@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: firedtv: reduce memset()s

Before each FCP transdaction, the entire 512 bytes of the FCP frame were
cleared, then values filled in.

Clear only the bytes between filled-in bytes and end of the
  - request frame, or
  - response frame if data from a larger response will be needed, or
  - whole frame if data from a variable length response will be taken.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: firedtv: remove check for interrupting signal

FCP transactions as well as CMP transactions were serialized with
mutex_lock_interruptible.  It is extremely unlikly though that a signal
will arrive while a concurrent process holds the mutex.  And even if one
does, the duration of a transaction is reasonably short (1.2 seconds if
all retries time out, usually much shorter).

Hence simplify the code to plain mutex_lock.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: firedtv: do not DMA-map stack addresses

This is a portability fix and reduces stack usage.

The DMA mapping API cannot map on-stack addresses, as explained in
Documentation/DMA-mapping.txt.  Convert the two cases of on-stack packet
payload buffers in firedtv (payload of write requests in avc_write and
of lock requests in cmp_lock) to slab-allocated memory.

We use the 512 bytes sized FCP frame buffer in struct firedtv for this
purpose.  Previously it held only incoming FCP responses, now it holds
pending FCP requests and is then overwriten by an FCP response from the
tuner subunit.  Ditto for CMP lock requests and responses.  Accesses to
the payload buffer are serialized by fdtv->avc_mutex.

As a welcome side effect, stack usage of the AV/C transaction functions
is reduced by 512 bytes.

Alas, avc_register_remote_control() is a special case:  It previously
did not wait for a response.  To fit better in with the other FCP
transactions, let it wait for an interim response.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

Fix the __init/__exit annotations

WARNING: drivers/media/video/cx18/cx18-alsa.o(.text+0x4de): Section
mismatch in reference from the function cx18_alsa_load() to the function
.init.text:snd_cx18_init()
The function cx18_alsa_load() references
the function __init snd_cx18_init().
This is often because cx18_alsa_load lacks a __init
annotation or the annotation of snd_cx18_init is wrong.

WARNING: drivers/media/video/cx18/built-in.o(.text+0x1c022): Section
mismatch in reference from the function cx18_alsa_load() to the function
.init.text:snd_cx18_init()
The function cx18_alsa_load() references
the function __init snd_cx18_init().
This is often because cx18_alsa_load lacks a __init
annotation or the annotation of snd_cx18_init is wrong.

WARNING: drivers/media/video/built-in.o(.text+0x28cc56): Section
mismatch in reference from the function cx18_alsa_load() to the function
.init.text:snd_cx18_init()
The function cx18_alsa_load() references
the function __init snd_cx18_init().
This is often because cx18_alsa_load lacks a __init
annotation or the annotation of snd_cx18_init is wrong.

WARNING: drivers/media/built-in.o(.text+0x2d2432): Section mismatch in
reference from the function cx18_alsa_load() to the function
.init.text:snd_cx18_init()
The function cx18_alsa_load() references
the function __init snd_cx18_init().
This is often because cx18_alsa_load lacks a __init
annotation or the annotation of snd_cx18_init is wrong.

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

cx18: Fix compilation when CONFIG_MODULE is disabled

drivers/media/video/cx18/cx18-driver.c:252: warning: ‘request_modules’ used but never defined

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

V4L/DVB: gspca - sq905c: Fix a compilation warning

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - ov534: Fix end of frame handling

Fix a regression, introduced in the driver split, which made
the ov534 driver unusable.
Plus, the debug message should reflect that we discard also packets
beyond the expected frame size.

Signed-off-by: Antonio Ospite <ospite@studenti.unina.it>
Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - ov534: Allow enumerating supported framerates

Signed-off-by: Antonio Ospite <ospite@studenti.unina.it>
Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - ov534_9: Propagate USB errors to higher level

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - main: Clear any previous USB error when starting the transfer

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

V4L/DVB: gspca - pac7311: Use usb_err to propagate USB errors

Signed-off-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>