[media] siano: split debugfs code into a separate file

To avoid mixing two different things at the same place, move the
debugfs code into a separate file.

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

[media] siano: allow showing the complete statistics via debugfs

Outputs the result of the statistics responses via debugfs.
That can help to track bugs at the stats filling.

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

[media] siano: fix start of statistics

It seems that the first u32 after the header for some stats are used by
something not documented.
The stats struct starts after it.

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

[media] siano: Convert it to report DVBv5 stats

While this frontend provides a nice set of statistics, the
way it is currently reported to userspace is poor. Worse than
that, instead of using quality indicators that range from 0 to 65535,
as expected by userspace, most indicators range from 0 to 100.
Improve it by using DVBv5 statistics API. The legacy indicators
are still reported using the very same old way, but they're now
using a proper range (0 to 65535 for quality indicadors; 0.1 dB
for SNR).

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

[media] siano: split debug logic from the status update routine

It is confusing to merge both status updates with debug stuff.
Also, it is a better idea to move those status updates to
debugfs, instead of doing a large amount of printk's like that.
So, break them into a separate block of routines.

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

[media] siano: split get_frontend into per-std functions

Instead of handling both DVB-T and ISDB-T at the same get_frontend
function, break it intow one function per-delivery system.
That makes the code clearer as we start to add support for DVBv5
statistics, and for ISDB-T get frontend stuff.

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

[media] siano: simplify message endianness logic

Currently, every time a message is sent or received, the endiannes
need to be fixed on big endian machines. This is currently done
on every call to the send API, and on every msg reception logic.
Instead of doing that, move it to the send/receive functions.
That simplifies the logic and avoids the risk of forgetting to
fix it somewhere.

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

[media] siano: call MSG_SMS_INIT_DEVICE_REQ

Newer firmwares seem to require an init device message. Apply
such change from Doron Cohen's patch:
http://patchwork.linuxtv.org/patch/7889/

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

[media] siano: Configure board's mtu and xtal

Backported from Doron Cohen's patch:
http://patchwork.linuxtv.org/patch/7889/

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

[media] siano: add new devices to the Siano Driver

This patch is based on Doron Cohen's patches:
http://patchwork.linuxtv.org/patch/7881/
http://patchwork.linuxtv.org/patch/7888/
http://patchwork.linuxtv.org/patch/7883/
It basically merges the above patches, rebasing them to
the macro definitions used upstream, with are different
 than the ones used by them internally.

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

[media] siano: use the newer stats message for recent firmwares

The old statistics request don't work with newer firmwares.
Add a logic to use the newer stats if firmware major is 8.
Note that I have only 2 devices here, one with firmware 2.1
(Hauppauge model 55009 Rev B1F7) and another one with
firmware 8.1. We may need to adjust the firmware minimal
version for the *_EX message variants, as we start finding
firmware versions between 2.x and 8.x.
This patch was based on Doron Cohen patch:
http://patchwork.linuxtv.org/patch/7886/

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

[media] siano: add support for LNA on ISDB-T

The very same code also exists for DVB-T. Add it for ISDB-T.

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

[media] siano: add support for ISDB-T full-seg

Fix the DVBv5 API handling for ISDB-T and add support
for 13 segments.

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

[media] siano: use a separate completion for stats

Instead of re-use tune_done also for stats, the better is to use
a different completion.
Also, it was noticed that sometimes, the driver answers with
MSG_SMS_SIGNAL_DETECTED_IND for status request. Fix the code to
also handle those other signal indicators.

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

[media] siano: add some new messages to the smscoreapi

Based on Doron Cohen's patch:
http://patchwork.linuxtv.org/patch/7887/

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

[media] siano: cleanups at smscoreapi.c

Some cleanups at smscoreapi. Most are just CodingStyle.
Also, use kzalloc when allocating a new buffer, as it initializes
the allocated space with zero.

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

[media] siano: always load smsdvb

Without smsdvb, the driver actually does nothing, as it
lacks the userspace API.
While I wrote it independently, in order to make a sms2270 board
I have here to work, this patch is functionally identical to this
patch from Doron Cohen:
http://patchwork.linuxtv.org/patch/7894/

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

[media] siano: fix the debug message

Instead of displaying this:
[   61.869415] smscore_load_firmware_family2: rc=0, postload=0x          (null)
Display, instead:
[ 1348.441160] smscore_load_firmware_family2: rc=0

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

[media] siano: report the choosed firmware in debug

Don't keep in the dark: report the firmware file name after
lookup. That helps to debug what's happening when a firmware is not
found.

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

[media] siano: make load firmware logic to work with newer firmwares

There are new firmwares for sms2xxx devices. Change the firmware
load logic to handle those newer firmwares and devices.

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

[media] siano: store firmware version

As there are some changes that seem to be firmware-dependent,
we need to store the firmware version, as we don't want to break
support for existing cards that use a legacy (and sometimes
custom) firmware.

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

[media] siano: use USB endpoint descriptors for in/out endp

Instead of using hardcoded descriptors, detect them from the
USB descriptors.
This patch is rebased form Doron Cohen's patch:
http://patchwork.linuxtv.org/patch/7883/

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

[media] siano: add additional attributes to cards entries

Those attributes will be used by the newer sms2xxx cards.

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

[media] siano: Properly initialize board information

Board #0 is an existing one. Instead of initializing the driver
with it, use a different value to detect if board is unknown.

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

[media] siano: add the remaining new defines from new driver

Add the remaining new defines/enums from Doron Cohen's patch:
        http://patchwork.linuxtv.org/patch/7882/

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

[media] siano: better debug send/receive messages

Instead of printing a message for some random messages, print
it for all sent/received ones. That helps a lot to debug
what's going on.

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

[media] siano: update message macros

Convert from #define into an enum and add the newer message
macros as found on this patch from Doron Cohen:
http://patchwork.linuxtv.org/patch/7882/
No messages got supressed.

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

[media] siano: remove a duplicated structure definition

The same GPIO config struct was declared twice at the
driver, with different names and different macros:
struct smscore_config_gpio
struct smscore_config_gpio
Remove the one that uses CamelCase and fix the references to
its attributes/macros.
No functional changes.

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

[media] siano: Add the new voltage definitions for GPIO

Those new definitions came from this patch, from Doron Cohen:
http://patchwork.linuxtv.org/patch/7882/

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

[media] siano: Change GPIO voltage setting names

Siano changed the namespace on more recent API, and re-used some
of the old names. In order to be able to update the API to support
newer chips, the better is to follow this change.

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

[media] davinci: vpbe: fix module build

add a null entry in platform_device_id {}.
This patch fixes following error:
drivers/media/platform/davinci/vpbe_venc: struct platform_device_id is 24 bytes.  The last of 3 is:
0x64 0x6d 0x33 0x35 0x35 0x2c 0x76 0x70 0x62 0x65 0x2d 0x76 0x65 0x6e 0x63 0x00 0x00 0x00 0x00 0x00 0x03 0x00 0x00 0x00
FATAL: drivers/media/platform/davinci/vpbe_venc: struct platform_device_id is not terminated with a NULL entry!
make[1]: *** [__modpost] Error 1

Reported-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Lad, Prabhakar <prabhakar.csengg@gmail.com>
Tested-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] blackfin: replace V4L2_IN/OUT_CAP_CUSTOM_TIMINGS by DV_TIMINGS

The use of V4L2_IN/OUT_CAP_CUSTOM_TIMINGS is obsolete, use DV_TIMINGS instead.
Note that V4L2_IN/OUT_CAP_CUSTOM_TIMINGS is just a #define for
V4L2_IN/OUT_CAP_DV_TIMINGS.
At some point in the future these CUSTOM_TIMINGS defines might be removed.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Scott Jiang <scott.jiang.linux@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] davinci: more gama -> gamma typo fixes

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] davinci/dm644x_ccdc: fix compiler warning

drivers/media/platform/davinci/dm644x_ccdc.c: In function ‘validate_ccdc_param’:
drivers/media/platform/davinci/dm644x_ccdc.c:233:32: warning: comparison between ‘enum ccdc_gama_width’ and ‘enum ccdc_data_size’ [-Wenum-compare]
It took a bit of work, see this thread of an earlier attempt to fix this:
https://patchwork.kernel.org/patch/1923091/
I've chosen not to follow the suggestions in that thread since gamma_width is
really a different property from data_size. What you really want is to know if
gamma_width fits inside data_size and for that you need to translate each
enum into a maximum bit number so you can safely compare the two.
So I put in two static inline translation functions instead, keeping the rest
of the code the same (except for fixing the 'gama' typo).

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] davinci/vpfe_capture: remove current_norm

Since vpfe_capture already provided a g_std op setting current_norm
does not actually do anything. Remove it.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] davinci/vpbe_display: remove deprecated current_norm

Since vpbe_display already provides a g_std op setting current_norm
didn't do anything. Remove that code.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] davinci/vpfe_capture: convert to the control framework

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] davinci: replace V4L2_OUT_CAP_CUSTOM_TIMINGS by V4L2_OUT_CAP_DV_TIMINGS

The use of V4L2_OUT_CAP_CUSTOM_TIMINGS is deprecated, use DV_TIMINGS instead.
Note that V4L2_OUT_CAP_CUSTOM_TIMINGS is just a #define for
V4L2_OUT_CAP_DV_TIMINGS.
At some point in the future these CUSTOM_TIMINGS defines might be removed.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Acked-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] davinci: remove VPBE_ENC_DV_PRESET and rename VPBE_ENC_CUSTOM_TIMINGS

Remove VPBE_ENC_DV_PRESET (the DV_PRESET API is no longer supported) and
VPBE_ENC_CUSTOM_TIMINGS is renamed to VPBE_ENC_DV_TIMINGS since the old
"CUSTOM_TIMINGS" name is deprecated in favor of "DV_TIMINGS".

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Acked-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] davinci_vpfe: fix copy-paste errors in several comments

This removes some incorrect dv_preset references left over from copy-and-paste
errors.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] tvp7002: remove dv_preset support

Finally remove the dv_preset support from this driver. Note that dv_preset
support was already removed from any bridge drivers that use this i2c
driver, so the dv_preset ops were no longer called and can be removed
safely.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] tvp7002: use dv_timings structs instead of presets

In the functions tvp7002_mbus_fmt(), tvp7002_log_status and tvp7002_probe()
we should use the dv_timings data structures instead of dv_preset data
structures and functions.
This is the second step towards removing the deprecated preset support of this
driver.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] tvp7002: replace 'preset' by 'timings' in various structs/variables

This is the first step towards removing the deprecated preset support of this
driver.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Acked-by: Lad, Prabhakar <prabhakar.lad@ti.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] [1/1,dvb-usb] GOTVIEW SatelliteHD card support

Added support for the GOTVIEW SatelliteHD card which is based on
Montage M88DS3000 and works very well with this driver.

Signed-off-by: Andrey Pavlenko <andrey.a.pavlenko@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] rc-core: rename ir_input_class to rc_class

The name is already misleading and will be more so in the future as the
connection to the input subsystem is obscured away further.

Signed-off-by: David Härdeman <david@hardeman.nu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] rc-core: initialize rc-core earlier if built-in

rc-core is a subsystem so it should be registered earlier if built into the
kernel.

Signed-off-by: David Härdeman <david@hardeman.nu>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] lmedm04: Remove redundant NULL check before kfree

kfree on NULL pointer is a no-op.

Signed-off-by: Syam Sidhardhan <s.syam@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] em28xx: update cardlist

There's one missing USB ID at the card list. Add it.

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

[media] dvb_demux: Transport stream continuity check fix

This patch avoids incrementing continuity counter
demux->cnt_storage[pid] for TS packets without payload in accordance
with ISO /IEC 13818-1.

[mchehab@redhat.com: unmangle whitespacing and fix CodingStyle.
 Also checked ISO/IEC spec: patch is according with it]
Reviewed-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Signed-off-by: John Smith <johns90812@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] redrat3: missing endian conversions and warnings

Spotted by sparse.

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

[media] redrat3: remove memcpys and fix unaligned memory access

In stead of doing a memcpy from #defined offset, declare structs which
describe the incoming and outgoing data accurately.
Tested on first generation RedRat.

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

[media] redrat3: limit periods to hardware limits

The redrat hardware cannot handle periods of larger than 32767us,
limit appropriately. Also fix memory leak in redrat3_get_timeout.

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

[media] em28xx: Add ISDB support for c3tech Digital duo

This is an hybrid board. However, for analog, it requires
a new driver for saa7136. So, for now, let's just add
support for Digital TV.

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

[media] v4l2-mem2mem: drop rdy_queue on STREAMOFF

When a v4l2-mem2mem context gets a STREAMOFF call on either its CAPTURE
or OUTPUT queues, we should:
* Drop the corresponding rdy_queue, since a subsequent STREAMON expects
  an empty queue.
* Deschedule the context, as it now has at least one empty queue and
  cannot run.

Signed-off-by: John Sheu <sheu@google.com>
Acked-by: Pawel Osciak <pawel@osciak.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] v4l2-mem2mem: use CAPTURE queue lock

In v4l2_m2m_try_schedule(), use the CAPTURE queue lock when accessing
the CAPTURE queue, instead of relying on just holding the OUTPUT queue
lock.

Signed-off-by: John Sheu <sheu@google.com>
Acked-by: Pawel Osciak <pawel@osciak.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] media: implement 32-on-64 bit compat IOCTL handling

Use the same handlers where the structs are the same. Implement a new
handler for link enumeration since struct media_links_enum is different on
32-bit and 64-bit systems.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] media: Add 64--32 bit compat ioctl handler

Provide an ioctl handler for 32-bit binaries on 64-bit systems.

Signed-off-by: Sakari Ailus <sakari.ailus@iki.fi>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] cx25821: Cleanup filename assignment code

I'm pasting the original code and my proposal on the commit message for
make it easy to compare the two versions.
Line 62 of cx25821-audio-upstream.h contains:
char *_defaultAudioName = "/root/audioGOOD.wav";
Original code after replace kmemdup for kstrdup, and after fix return error
code:
if (dev->input_audiofilename) {
dev->_audiofilename = kstrdup(dev->input_audiofilename,
      GFP_KERNEL);
if (!dev->_audiofilename) {
err = -ENOMEM;
goto error;
}
/* Default if filename is empty string */
if (strcmp(dev->input_audiofilename, "") == 0)
dev->_audiofilename = "/root/audioGOOD.wav";
} else {
dev->_audiofilename = kstrdup(_defaultAudioName,
      GFP_KERNEL);
if (!dev->_audiofilename) {
err = -ENOMEM;
goto error;
}
}
Code proposed in this patch:
if ((dev->input_audiofilename) &&
    (strcmp(dev->input_audiofilename, "") != 0))
dev->_audiofilename = kstrdup(dev->input_audiofilename,
      GFP_KERNEL);
else
dev->_audiofilename = kstrdup(_defaultAudioName,
      GFP_KERNEL);
if (!dev->_audiofilename) {
err = -ENOMEM;
goto error;
}

Signed-off-by: Peter Senna Tschudin <peter.senna@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] dvb_frontend: Simplify the emulation logic

The current logic was broken and too complex; while it works
fine for DVB-S2/DVB-S, it is broken for ISDB-T.
Make the logic simpler, fixes it for ISDB-T and make it clearer.

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

[media] dvb-frontend: split set_delivery_system()

This function is complex, and has different workflows, one for
DVBv3 calls, and another one for DVBv5 calls. Break it into 3
functions, in order to make easier to understand what each
block does.
No functional changes so far. A few comments got improved.

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

[media] em28xx-i2c: relax error check in em28xx_i2c_recv_bytes()

It turned out that some devices return less bytes then requested via i2c when
ALL of the following 3 conditions are met:
- i2c bus B is used
- there was no attempt to write to the specified slave address before
- no device present at the specified slave address
With the current code, this triggers an -EIO error and prints a message to the
system log.
Because it can happen very often during device probing, it is better to ignore
this error and bail out silently after the follwing i2c transaction success
check with -ENODEV.

[mchehab@redhat.com: a small CodingStyle fix]
Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] em28xx: set the timestamp type for video and vbi vb2_queues

The em28xx driver obtains the timestamps using function v4l2_get_timestamp(),
which produces a montonic timestamp.
Fixes the warnings appearing in the system log since commit 6aa69f99
"[media] vb2: Add support for non monotonic timestamps"

Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] em28xx: add support for registering multiple i2c buses

Register both buses 0 and 1 via I2C API. For now, bus 0 is used
only by eeprom on all known devices. Later patches will be needed
if this changes in the future.

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

[media] em28xx: Add a separate config dir for secondary bus

Prepare to register a separate bus for the second bus.
For now, just add a new field. A latter patch will add the
bits to make it work.
This patch was generated by this script:
perl -e 'while (<>) { if (s/EM2874_I2C_SECONDARY_BUS_SELECT.*\n//) {
printf "\t\t.def_i2c_bus  = 1,\n"; $found = 1; print $_ } else { if ($found) { s/^\s+// }; $found = 0; print $_; } }' \
drivers/media/usb/em28xx/em28xx-cards.c >a && mv a drivers/media/usb/em28xx/em28xx-cards.c

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

[media] em28xx: Prepare to support 2 different I2C buses

Newer em28xx devices have 2 buses. Change the logic to allow
using both buses.
This patch was generated by this small script:
for i in drivers/media/usb/em28xx/*.c; do
sed 's,->i2c_adap,->i2c_adap[dev->def_i2c_bus],g;s,->i2c_client,->i2c_client[dev->def_i2c_bus],'
done
Of course, em28xx.h needed manual edit.

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

[media] media/v4l2: VIDEOBUF2_DMA_CONTIG should depend on HAS_DMA

m68k/sun3:
drivers/media/v4l2-core/videobuf2-dma-contig.c: In function ‘vb2_dc_mmap’:
drivers/media/v4l2-core/videobuf2-dma-contig.c:204: error: implicit declaration of function ‘dma_mmap_coherent’
drivers/media/v4l2-core/videobuf2-dma-contig.c: In function ‘vb2_dc_get_base_sgt’:
drivers/media/v4l2-core/videobuf2-dma-contig.c:387: error: implicit declaration of function ‘dma_get_sgtable’
Make VIDEOBUF2_DMA_CONTIG and VIDEO_SH_VEU (which selects the former and
doesn't have a platform dependency) depend on HAS_DMA to fix this.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] m920x: let GCC see 'ret' is used initialized

Since commit 7543f344e9b06afe86b55a2620f5c11b38bd5642 ("[media] m920x:
factor out a m920x_write_seq() function") building m920x.o triggers this
GCC warning:
    drivers/media/usb/dvb-usb/m920x.c: In function ‘m920x_probe’:
    drivers/media/usb/dvb-usb/m920x.c:91:6: warning: ‘ret’ may be used uninitialized in this function [-Wuninitialized]
This warning is caused by m920x_write_seq(), which is apparently inlined
into m920x_probe(). It is clear why GCC thinks 'ret' may be used
uninitialized. But in practice the first seq->address will always be
non-zero when this function is called. That means we can change the
while()-do{} loop into a do{}-while() loop. And that suffices to make
GCC see that 'ret' will not be used uninitialized.

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] MAINTAINERS: remove include/media/sh_veu.h

Apparently a copy-paste mistake; the similar sh_vou.h exists, and both
were added to MAINTAINERS by commit b618b69 ([media] MAINTAINERS: add
entries for sh_veu and sh_vou V4L2 drivers).

Signed-off-by: Cesar Eduardo Barros <cesarb@cesarb.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] MAINTAINERS: fix Documentation/video4linux/saa7134/

That directory never existed. The intention was probably to match
CARDLIST.saa7134 and README.saa7134.

Signed-off-by: Cesar Eduardo Barros <cesarb@cesarb.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] MAINTAINERS: fix drivers/media/i2c/cx2341x.c

This file was moved to drivers/media/common/ by commit 6259582 ([media]
cx2341x: move from media/i2c to media/common).

Signed-off-by: Cesar Eduardo Barros <cesarb@cesarb.net>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] s5p-mfc: Staticize symbols in s5p_mfc_opr_v5.c

Some symbols are used only in this file. Make them static.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] s5p-mfc: Staticize symbols in s5p_mfc_opr_v6.c

Symbols used only in this file should be made static.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] s5p-mfc: Staticize some symbols in s5p_mfc_cmd_v5.c

These symbols are used only in this file and can be made static.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] s5p-mfc: Staticize some symbols in s5p_mfc_cmd_v6.c

Since these symbols are used only in this file, they can be made static.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] cx231xx : Add support for OTG102 aka EZGrabber2

Thanks for the response, I have done as you suggested.
Below is an updated patch for the OTG102 device against http://git.linuxtv.org/hverkuil/media_tree.git/shortlog/refs/heads/cx231xx, kernel version 3.8.
With further testing it appears the extra clauses in cx231xx-cards.c were not necessary (in static in cx231xx_init_dev and static int cx231xx_usb_probe), so those have been also been removed.

Signed-off-by: Matt Gomboc <gomboc0@gmail.com>
Acked-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] drivers/staging/media/as102: using ccflags-y instead of EXTRA_FLAGS in Makefile

need using ccflags-y instead of EXTRA_CFLAGS
    can reference scripts/checkpatch.pl (1755..1766)
  when make EXTRA_CFLAGS=-W, the compiling issue will be occured.

Signed-off-by: Chen Gang <gang.chen@asianux.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] MAINTAINERS: Add maintainer entry for si4713 FM transmitter driver

Add maintainer entry for the files composing si4713 FM transmitter driver.

Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] dvb-usb: Remove redundant NULL check before kfree

kfree on NULL pointer is a no-op.

Signed-off-by: Syam Sidhardhan <s.syam@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] media: tuners: Remove redundant NULL check before kfree

kfree on NULL pointer is a no-op.

Signed-off-by: Syam Sidhardhan <s.syam@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] media: ivtv: Remove redundant NULL check before kfree

kfree on NULL pointer is a no-op.

Signed-off-by: Syam Sidhardhan <s.syam@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] siano: Remove redundant NULL check before kfree

kfree on NULL pointer is a no-op.

Signed-off-by: Syam Sidhardhan <s.syam@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] Media: remove incorrect __init/__exit markups

Even if bus is not hot-pluggable, the devices can be unbound from the
driver via sysfs, so we should not be using __exit annotations on
remove() methods. The only exception is drivers registered with
platform_driver_probe() which specifically disables sysfs bind/unbind
attributes.
Similarly probe() methods should not be marked __init unless
platform_driver_probe() is used.

Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Acked-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Acked-by: Sakari Ailus <sakari.ailus@iki.fi>
Acked-by: Timo Kokkonen <timo.t.kokkonen@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] anysee: coding style changes

I did what I liked to do. Also corrected two long log writings
as checkpatch.pl was complaining about those.

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

[media] anysee: do not use buffers from stack for usb_bulk_msg()

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

[media] af9035: do not use buffers from stack for usb_bulk_msg()

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

[media] af9015: do not use buffers from stack for usb_bulk_msg()

WARNING: at lib/dma-debug.c:947 check_for_stack+0xa7/0xf0()
ehci-pci 0000:00:04.1: DMA-API: device driver maps memory fromstack

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

[media] dvb_usb_v2: locked versions of USB bulk IO functions

Implement:
dvb_usbv2_generic_rw_locked()
dvb_usbv2_generic_write_locked()
Caller must hold device lock when locked versions are called.

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

[media] hdpvr: Fix memory leak

This patch fixes the print_buf leaking.

Signed-off-by: Syam Sidhardhan <s.syam@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] lmedm04: Fix possible NULL pointer dereference

Check for (adap == NULL) has to done before accessing adap.

Signed-off-by: Syam Sidhardhan <s.syam@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] solo6x10: Maintainer change

Signed-off-by: Ismael Luceno <ismael.luceno@corp.bluecherry.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] media/usb: cx231xx-pcb-cfg.h: Remove unused enum _true_false

Signed-off-by: Thiago Farina <tfarina@chromium.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] media: Terratec Cinergy S2 USB HD Rev.2

Terratec Cinergy S2 USB HD Rev.2 support.
This commit is a corrected cherry-pick of 03228792 which got reverted in
b7e38636 because it was rebased incorrectly and introduced compilation
errors.

Signed-off-by: Stephan Hilb <stephan@ecshi.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] stv090x: do not unlock unheld mutex in stv090x_sleep()

goto err and goto err_gateoff before mutex_lock(&state->internal->demod_lock)
lead to unlock of unheld mutex in stv090x_sleep().
Found by Linux Driver Verification project (linuxtesting.org).

Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Cc: Manu Abraham <abraham.manu@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] timblogiw: Fix sparse warning

Fixes the below warning:
drivers/media/platform/timblogiw.c:81:31: warning:
symbol 'timblogiw_tvnorms' was not declared. Should it be static?

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] af9035: add ID [0ccd:00aa] TerraTec Cinergy T Stick (rev. 2)

This patch adds USB ID for alternative "Terratec Cinergy T Stick".
Tested by a friend: works similarly to 0ccd:0093 version (af9035+tua9001)

Signed-off-by: Fabrizio Gazzato <fabrizio.gazzato@gmail.com>
Acked-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] cx231xx: fix undefined function cx231xx_g_chip_ident()

This patch:
http://git.linuxtv.org/media_tree.git/commit/b86d15440b683f8634c0cb26fc0861a5bc4913ac
is missing a chunk when compared to an older version:
https://patchwork.kernel.org/patch/2063281/
probably because of an unresolved merging conflict.
This causes the following error:
WARNING: "cx231xx_g_chip_ident" [/home/jena/media_build/v4l/cx231xx.ko] undefined!

Signed-off-by: Gianluca Gennari <gennarone@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] bttv: move fini_bttv_i2c() from bttv-input.c to bttv-i2c.c

Like init_bttv_i2c(), fini_bttv_i2c() belongs to bttv-i2c.c.

Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] bttv: make remote controls of devices with i2c ir decoder working

Request module ir-kbd-i2c if an i2c ir decoder is detected.
Tested with device "Hauppauge WinTV Theatre" (model 37284 rev B421).

Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] rtl28xxu: Add USB ID for MaxMedia HU394-T

Add USB ID for MaxMedia HU394-T USB DVB-T Multi (FM, DAB, DAB+)
dongle (RTL2832U+FC0012)

In Italy, is branded as "DIKOM USB-DVBT HD"
lsusb: ID 1b80:d394 Afatech

Signed-off-by: Fabrizio Gazzato <fabrizio.gazzato@gmail.com>
Acked-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>

[media] s2255: fix big-endian support

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>