platform/upstream/v4l-utils.git
11 years agov4l2-ctl: initial attempt to support M2M device streaming
Tzu-Jung Lee [Wed, 10 Apr 2013 10:57:06 +0000 (18:57 +0800)]
v4l2-ctl: initial attempt to support M2M device streaming

Signed-off-by: Tzu-Jung Lee <tjlee@ambarella.com>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: break down the streaming_set()
Tzu-Jung Lee [Wed, 10 Apr 2013 10:57:05 +0000 (18:57 +0800)]
v4l2-ctl: break down the streaming_set()

This patch breaks down the streaming_set() into smaller
ones, which can be resued for supporting m2m devices.

Further cleanup or consolidation can be applied with
separate patches, since this one tries not to modify
logics.

Signed-off-by: Tzu-Jung Lee <tjlee@ambarella.com>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: add eos and decoder-stop support.
Hans Verkuil [Tue, 9 Apr 2013 06:46:49 +0000 (08:46 +0200)]
v4l2-ctl: add eos and decoder-stop support.

When dealing with compressed streams it is useful if the capture streaming
code would check for V4L2_EVENT_EOS and if the output streaming side would
be able to use a STOP DECODER command before calling STREAMOFF to give the
decoder the time to flush any pending data.

This patch adds support for that.

Thanks to Tzu-Jung Lee for testing this.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-compliance: really fix the RDS and V4L2_CAP_READWRITE tests
Hans Verkuil [Mon, 8 Apr 2013 19:04:23 +0000 (21:04 +0200)]
v4l2-compliance: really fix the RDS and V4L2_CAP_READWRITE tests

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-compliance: fix RDS + CAP_READWRITE tests.
Hans Verkuil [Mon, 8 Apr 2013 10:40:51 +0000 (12:40 +0200)]
v4l2-compliance: fix RDS + CAP_READWRITE tests.

The READWRITE capability should only be set if RDS_BLOCK_IO is set.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: set bytesused for streaming to the output.
Hans Verkuil [Mon, 8 Apr 2013 10:31:55 +0000 (12:31 +0200)]
v4l2-ctl: set bytesused for streaming to the output.

For output stream I/O bytesused should be filled in.

Note that v4l2-ctl only supports uncompressed formats for output streaming
or compressed streams where each buffer is fully filled up (i.e. bytesused
== buffer length).

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agolibdvbv5: fix anonymous unions on older gcc
André Roth [Sat, 6 Apr 2013 16:36:22 +0000 (18:36 +0200)]
libdvbv5: fix anonymous unions on older gcc

11 years agov4l2-compliance: remove ChipIdent tests
Hans Verkuil [Wed, 3 Apr 2013 06:31:50 +0000 (08:31 +0200)]
v4l2-compliance: remove ChipIdent tests

These tests are no longer necessary: VIDIOC_DBG_G_CHIP_IDENT is deprecated
and is replaced by VIDIOC_DBG_G_CHIP_NAME, which is a core ioctl that is
always available.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l-utils: sync-with-kernel.
Hans Verkuil [Tue, 2 Apr 2013 09:19:47 +0000 (11:19 +0200)]
v4l-utils: sync-with-kernel.

Update to the latest kernel headers.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l-utils: sync to the latest kernel.
Hans Verkuil [Tue, 26 Mar 2013 08:25:17 +0000 (09:25 +0100)]
v4l-utils: sync to the latest kernel.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l-utils: sync with kernel, remove all references to the DV_PRESET API.
Hans Verkuil [Mon, 25 Mar 2013 17:09:57 +0000 (18:09 +0100)]
v4l-utils: sync with kernel, remove all references to the DV_PRESET API.

The DV_PRESET API has been removed from the kernel, so remove all references
to that API.

It has been superceded by the DV TIMINGS API.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: add the --concise option.
Hans Verkuil [Wed, 20 Mar 2013 18:24:30 +0000 (19:24 +0100)]
v4l2-ctl: add the --concise option.

With the concise option you can get a more concise overview of some of the
lists. Implemented this for list-dv-timings and list-standards.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agoqv4l2/v4l2-ctl: fix rounding error in field-per-second calculation
Hans Verkuil [Wed, 20 Mar 2013 17:59:31 +0000 (18:59 +0100)]
qv4l2/v4l2-ctl: fix rounding error in field-per-second calculation

The total height was divided by 2, since height is odd we should make sure the
height is a double, not unsigned or you get a rounding error.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agoqv4l2/v4l2-ctl: fix the fields-per-second calculation
Hans Verkuil [Wed, 20 Mar 2013 17:11:22 +0000 (18:11 +0100)]
qv4l2/v4l2-ctl: fix the fields-per-second calculation

The calculation did not take the vertical blanking of the second field
into account, which is typically one line longer. So that will slightly
affect the fields-per-second value.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvbv5-zap: Fix build warning
Jean Delvare [Wed, 20 Mar 2013 14:08:36 +0000 (11:08 -0300)]
dvbv5-zap: Fix build warning

dvbv5-zap.c: In function "do_traffic_monitor":
dvbv5-zap.c:582:12: warning: format "%d" expects argument of type "int",
  but argument 5 has type "long long unsigned int" [-Wformat]

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agov4l2-ctl: fix the field-per-second calculation
Hans Verkuil [Wed, 20 Mar 2013 08:15:57 +0000 (09:15 +0100)]
v4l2-ctl: fix the field-per-second calculation

The height in the bt_timings struct is the height of the full frame,
so for the fields-per-second calculation the height should be divided
by two.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agoqv4l2: fix the way interlaced formats are displayed
Hans Verkuil [Wed, 20 Mar 2013 08:15:07 +0000 (09:15 +0100)]
qv4l2: fix the way interlaced formats are displayed

The timings string shows frames instead of fields, which is non-standard.
Switch to fields for interlaced formats.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvbv5-zap: add traffic monitor support
Mauro Carvalho Chehab [Tue, 19 Mar 2013 21:28:43 +0000 (18:28 -0300)]
dvbv5-zap: add traffic monitor support

Add some code to do traffic monitor. The code there came from
dvb-apps dvbtraffic.c file, released under public domain.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agov4l2-ctl: don't produce verbose output for QBUF while streaming.
Hans Verkuil [Tue, 19 Mar 2013 11:50:06 +0000 (12:50 +0100)]
v4l2-ctl: don't produce verbose output for QBUF while streaming.

You are not interested in the verbose logging of QBUF while streaming.
It's not done for DQBUF, so silence QBUF as well.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-compliance: fix s_std test: ENODATA is a valid error code.
Hans Verkuil [Tue, 19 Mar 2013 11:45:56 +0000 (12:45 +0100)]
v4l2-compliance: fix s_std test: ENODATA is a valid error code.

The S_STD test failed because the driver returned ENODATA. This is a
valid error code, but the test didn't check for it.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agolibv4l2/log.c: update list of ioctls.
Hans Verkuil [Tue, 12 Mar 2013 07:42:13 +0000 (08:42 +0100)]
libv4l2/log.c: update list of ioctls.

This list was out of date, update it.

Also update a very small log printf (missing colon).

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agoqv4l2: handle control range updates.
Hans Verkuil [Tue, 12 Mar 2013 06:40:24 +0000 (07:40 +0100)]
qv4l2: handle control range updates.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agoqv4l2: do nothing if querystd does not detect a standard.
Hans Verkuil [Mon, 11 Mar 2013 21:55:11 +0000 (22:55 +0100)]
qv4l2: do nothing if querystd does not detect a standard.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-compliance: improve tests when TRY/S_FMT returns EINVAL.
Hans Verkuil [Mon, 11 Mar 2013 20:45:14 +0000 (21:45 +0100)]
v4l2-compliance: improve tests when TRY/S_FMT returns EINVAL.

If a wrong pixelformat is specified, then TRY/S_FMT can return EINVAL.
However, part of the test sequence was skipped in that case, which led
to missed errors.

Now get a valid pixelformat and put that in and try again. That way all
other tests are executed as well.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-compliance: missing check on the frameinterval
Hans Verkuil [Sun, 10 Mar 2013 10:40:00 +0000 (11:40 +0100)]
v4l2-compliance: missing check on the frameinterval

A wrong discrete frameinterval as returned by ENUM_FRAMEINTERVALS
wasn't detected.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-compliance: add support for the new 'COPY' timestamp type.
Hans Verkuil [Sun, 10 Mar 2013 09:31:50 +0000 (10:31 +0100)]
v4l2-compliance: add support for the new 'COPY' timestamp type.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: add support for 'COPY' timestamp and ctrl range events.
Hans Verkuil [Sun, 10 Mar 2013 09:29:38 +0000 (10:29 +0100)]
v4l2-ctl: add support for 'COPY' timestamp and ctrl range events.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l-utils: run sync-with-kernel.
Hans Verkuil [Sun, 10 Mar 2013 09:25:24 +0000 (10:25 +0100)]
v4l-utils: run sync-with-kernel.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-* utils: allow -d11 to refer to /dev/video11
Hans Verkuil [Sat, 9 Mar 2013 17:59:11 +0000 (18:59 +0100)]
v4l2-* utils: allow -d11 to refer to /dev/video11

Lately I have been working with devices with lots of video nodes. Various
v4l2 utilities allow -dX to refer to /dev/videoX, but 'X' was limited to
a single digit. Extend this to support up to three digits. So -d11 would
refer to /dev/video11 and -d128 to /dev/video128 (that really should be
enough!)

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: show buffers or key/P/B-frames when capturing.
Hans Verkuil [Mon, 4 Mar 2013 22:45:09 +0000 (23:45 +0100)]
v4l2-ctl: show buffers or key/P/B-frames when capturing.

Add support in v4l2-ctl to show any captured buffer data if --verbose is
given, or to show whether a key, P or B frame was captured.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: support data_offset when streaming multiplanar formats.
Hans Verkuil [Mon, 4 Mar 2013 15:01:09 +0000 (16:01 +0100)]
v4l2-ctl: support data_offset when streaming multiplanar formats.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: improve dv_timings reporting.
Hans Verkuil [Mon, 4 Mar 2013 14:58:58 +0000 (15:58 +0100)]
v4l2-ctl: improve dv_timings reporting.

Show fields per second instead of frame per second as the latter is confusing
for interlaced formats.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvbv5-scan: timeout-multiply should also affect the timeout for scan
Mauro Carvalho Chehab [Sun, 3 Mar 2013 15:12:39 +0000 (12:12 -0300)]
dvbv5-scan: timeout-multiply should also affect the timeout for scan

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agov4l2-ctl: when streaming to file use bytesused instead of the full buffer.
Hans Verkuil [Sun, 3 Mar 2013 14:21:29 +0000 (15:21 +0100)]
v4l2-ctl: when streaming to file use bytesused instead of the full buffer.

This makes it possible to use v4l2-ctl to stream compressed video using
stream I/O.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvb-fe: Supress statistics if signal lock vanishes
Mauro Carvalho Chehab [Sat, 2 Mar 2013 19:07:58 +0000 (16:07 -0300)]
dvb-fe: Supress statistics if signal lock vanishes

If for some reason the signal unlocks, the DVB driver should be
marking the stats as unavailable. Make the dvbv5 library to
honourg it.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-scan: Don't call dvb_fe_get_stats() twice
Mauro Carvalho Chehab [Sat, 2 Mar 2013 17:51:26 +0000 (14:51 -0300)]
dvb-scan: Don't call dvb_fe_get_stats() twice

This routine was called already be check_frontend. No need to
call it again.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvbv5-scan: don't be too verbose with status
Mauro Carvalho Chehab [Sat, 2 Mar 2013 15:23:45 +0000 (12:23 -0300)]
dvbv5-scan: don't be too verbose with status

As it tries to get status on every 100ms, it ends that too much
verbosity is produced. Don't do that. Instead, keep printing the
status at the same line. This is nicer.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb5-scan: Improve output to better show the status
Mauro Carvalho Chehab [Sat, 2 Mar 2013 12:38:28 +0000 (09:38 -0300)]
dvb5-scan: Improve output to better show the status

Show intermediate status, in order to allow better adjusting
the antenna.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvbv5-zap: Only go back if is there anything to clean
Mauro Carvalho Chehab [Sat, 2 Mar 2013 12:39:39 +0000 (09:39 -0300)]
dvbv5-zap: Only go back if is there anything to clean

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agov4l2-ctl: improve interval reporting.
Hans Verkuil [Sat, 2 Mar 2013 11:20:49 +0000 (12:20 +0100)]
v4l2-ctl: improve interval reporting.

Shorten the text and drop the fps step value which was wrong and not useful.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agoqv4l2: correctly update frame width and height for fixed resolutions.
Hans Verkuil [Sat, 2 Mar 2013 10:28:23 +0000 (11:28 +0100)]
qv4l2: correctly update frame width and height for fixed resolutions.

When selecting a specific resolution the width and height fields were
only disabled, but not updated to that resolution. Fixed.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvb-fe: Reserve space for the other DVBv5 properties
Mauro Carvalho Chehab [Fri, 1 Mar 2013 13:40:55 +0000 (10:40 -0300)]
dvb-fe: Reserve space for the other DVBv5 properties

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-fe: Fix check for stats property
Mauro Carvalho Chehab [Fri, 1 Mar 2013 13:39:29 +0000 (10:39 -0300)]
dvb-fe: Fix check for stats property

Instead of [i], the tests were doing [2]. That broke DTV_STATUS
retrival.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agoqv4l2: don't show "Query Standard" button if QUERYSTD isn't supported.
Hans Verkuil [Fri, 15 Feb 2013 12:44:02 +0000 (13:44 +0100)]
qv4l2: don't show "Query Standard" button if QUERYSTD isn't supported.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvbv5-scan: Fix get_signal logic
Mauro Carvalho Chehab [Wed, 6 Feb 2013 13:26:27 +0000 (11:26 -0200)]
dvbv5-scan: Fix get_signal logic

As reported by Hans Verkuil:

dvbv5-scan.c: In function ‘main’:
dvbv5-scan.c:113:23: warning: ‘i’ may be used uninitialized in this function [-Wuninitialized]
dvbv5-scan.c:97:10: note: ‘i’ was declared here

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agov4l-utils: use openat when available
Riku Voipio [Tue, 22 Jan 2013 15:37:22 +0000 (12:37 -0300)]
v4l-utils: use openat when available

New architectures such as 64-Bit arm build kernels without legacy
system calls - Such as the the no-at system calls. Thus, use
SYS_openat whenever it is available.

Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agov4l2-compliance: improve read/write error reporting.
Hans Verkuil [Wed, 6 Feb 2013 07:27:41 +0000 (08:27 +0100)]
v4l2-compliance: improve read/write error reporting.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agoioctl-test: expand it to also test DVB frontend/demux
Mauro Carvalho Chehab [Mon, 4 Feb 2013 19:59:11 +0000 (17:59 -0200)]
ioctl-test: expand it to also test DVB frontend/demux

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agoAdd DVB headers to make sync-with-kernel
Mauro Carvalho Chehab [Mon, 4 Feb 2013 17:13:37 +0000 (15:13 -0200)]
Add DVB headers to make sync-with-kernel

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agoioctl-test: Auto-generate the list of ioctls to test
Mauro Carvalho Chehab [Mon, 4 Feb 2013 19:00:16 +0000 (17:00 -0200)]
ioctl-test: Auto-generate the list of ioctls to test

Instead of hardcoding the V4L ioctl magic numbers, determine it
from the Kernel headers.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvbv5-scan: also prints signal strength on tuner fail
Mauro Carvalho Chehab [Thu, 24 Jan 2013 19:55:29 +0000 (17:55 -0200)]
dvbv5-scan: also prints signal strength on tuner fail

Even when tuner fails, it makes sense to print the signal strength.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agoioctl-test: improve comments and add checks against ioctl num changes.
Hans Verkuil [Mon, 4 Feb 2013 12:56:02 +0000 (13:56 +0100)]
ioctl-test: improve comments and add checks against ioctl num changes.

Improve the comments at the top of the source.

Also add explicit checks against the ioctl command values to verify that
they haven't changed after modifying structs used by existing ioctls.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agolibv4l: Add Founder PC T14MF to upside down table
Gregor Jasny [Sun, 27 Jan 2013 15:57:54 +0000 (16:57 +0100)]
libv4l: Add Founder PC T14MF to upside down table

Signed-off-by: Gregor Jasny <gjasny@googlemail.com>
11 years agov4l2-compliance: reorganize verbosity options
Hans Verkuil [Fri, 25 Jan 2013 08:49:14 +0000 (09:49 +0100)]
v4l2-compliance: reorganize verbosity options

The --verbose option had two levels: 1 would show warnings, 2 would show
warnings and info messages. But this was confusing since by default the
warnings weren't shown, although a warning count was printed in the summary.

Changed to show warnings by default, and --verbose now turns on just info
messages and the new -n option turns off warnings.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvbv5-scan: use library support for stats print
Mauro Carvalho Chehab [Thu, 24 Jan 2013 19:30:59 +0000 (17:30 -0200)]
dvbv5-scan: use library support for stats print

Use the libdvbv5 support to display status.

Now, an ISDB-T channel scanning will look like:

INFO     Scanning frequency #1 485142857
Lock   (0x1f) Quality= Poor Signal= 6.25% C/N= 17.05dB postBER= 517x10^-3 preBER= 505x10^-3 PER= 927x10^-3
  Layer A: Quality= Poor C/N= 12.82dB postBER= 2.10x10^-3 preBER= 35.5x10^-3 PER= 62.6x10^-3
  Layer B: Quality= Poor C/N= 17.81dB postBER= 15.6x10^-3 preBER= 83.4x10^-3 PER= 939x10^-3
Service #0 (59833) GLOBO 1SEG channel 4.0
Service #1 (59808) GLOBO HD channel 4.1
INFO     Scanning frequency #2 527142857
RF     (0x01): Tuning failed.
INFO     Scanning frequency #3 593142857
RF     (0x01): Tuning failed.
INFO     Scanning frequency #4 629142857
Lock   (0x1f) Quality= Poor Signal= 6.25% C/N= 10.98dB postBER= 872x10^-6 preBER= 21.7x10^-3 PER= 11.4x10^-3
  Layer A: Quality= Good C/N= 8.74dB postBER= 0 preBER= 2.38x10^-3 PER= 0
  Layer B: Quality= Poor C/N= 13.13dB postBER= 914x10^-6 preBER= 22.2x10^-3 PER= 11.9x10^-3
Service #0 (60696) RECORD 1SEG channel 8.0
Service #1 (60672) RECORD HD channel 8.1
INFO     Scanning frequency #5 725142857
Lock   (0x1f) Quality= Good Signal= 6.25% C/N= 17.90dB postBER= 1.58x10^-3 preBER= 26.0x10^-3 PER= 0
  Layer A: Quality= Good C/N= 19.33dB postBER= 0 preBER= 26.0x10^-3 PER= 0
  Layer B: Quality= Good C/N= 17.53dB postBER= 1.66x10^-3 preBER= 22.2x10^-3 PER= 0
Service #0 (60320) BAND HD channel 6.0
Service #1 (60345) BAND 1SEG channel 6.1

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvbv5-zap: improve status display
Mauro Carvalho Chehab [Thu, 24 Jan 2013 17:03:01 +0000 (15:03 -0200)]
dvbv5-zap: improve status display

When multiple layers are present, show each layer on a separate line.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb: if the signal vanishes, bit/block error counters should vanish
Mauro Carvalho Chehab [Thu, 24 Jan 2013 11:11:48 +0000 (09:11 -0200)]
dvb: if the signal vanishes, bit/block error counters should vanish

The library should detect if the BER measures aren't available anymore,
as the signal may be dropped.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agolibv4l: Don't try to access sysfs 256 times when there is no sysfs
Hans Petter Selasky [Thu, 24 Jan 2013 12:06:30 +0000 (13:06 +0100)]
libv4l: Don't try to access sysfs 256 times when there is no sysfs

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
11 years agov4l-utils: fix two 'dereferencing type-punned pointer' warnings
Hans Verkuil [Thu, 24 Jan 2013 11:58:15 +0000 (12:58 +0100)]
v4l-utils: fix two 'dereferencing type-punned pointer' warnings

After this change v4l-utils now builds without warnings.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvb: add support for pre-BER measurement
Mauro Carvalho Chehab [Thu, 24 Jan 2013 11:03:41 +0000 (09:03 -0200)]
dvb: add support for pre-BER measurement

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb: Better name the post-BER data
Mauro Carvalho Chehab [Thu, 24 Jan 2013 10:51:58 +0000 (08:51 -0200)]
dvb: Better name the post-BER data

As pre-BER will also be added, better name post-BER ones.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb: use a macro for the number of Kernel stats
Mauro Carvalho Chehab [Thu, 24 Jan 2013 10:45:40 +0000 (08:45 -0200)]
dvb: use a macro for the number of Kernel stats

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agov4l-utils: fix make install problems with ln -s
Hans Verkuil [Thu, 24 Jan 2013 10:54:16 +0000 (11:54 +0100)]
v4l-utils: fix make install problems with ln -s

When re-installing the linking of the some libraries fail with a 'File exists'
error:

make  install-exec-hook
make[4]: Entering directory `/home/hve/work/src/v4l/v4l-utils/lib/libv4l2'
/bin/mkdir -p //usr/local/lib
(cd //usr/local/lib && ln -s libv4l/v4l2convert.so v4l2convert.so)
ln: failed to create symbolic link 'v4l2convert.so': File exists
make[4]: *** [install-exec-hook] Error 1
make[4]: Leaving directory `/home/hve/work/src/v4l/v4l-utils/lib/libv4l2'
make[3]: *** [install-exec-am] Error 2
make[3]: Leaving directory `/home/hve/work/src/v4l/v4l-utils/lib/libv4l2'
make[2]: *** [install-am] Error 2
make[2]: Leaving directory `/home/hve/work/src/v4l/v4l-utils/lib/libv4l2'
make[1]: *** [install-recursive] Error 1
make[1]: Leaving directory `/home/hve/work/src/v4l/v4l-utils/lib'
make: *** [install-recursive] Error 1

Fix this by removing the old link first.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agodvb: Add support for retrieving pre-BER data
Mauro Carvalho Chehab [Thu, 24 Jan 2013 10:42:01 +0000 (08:42 -0200)]
dvb: Add support for retrieving pre-BER data

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agolibdvbv5: fix compile error
André Roth [Wed, 23 Jan 2013 22:47:47 +0000 (23:47 +0100)]
libdvbv5: fix compile error

Signed-off-by: André Roth <neolynx@gmail.com>
11 years agodvb: Rename a few DVB statistics
Mauro Carvalho Chehab [Wed, 23 Jan 2013 21:16:42 +0000 (19:16 -0200)]
dvb: Rename a few DVB statistics

At the last moment, a few DVB stats properties got renamed.
Their meaning didn't change, so the only thing it is needed is
to synchronize with the new nomenclature.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-fe: Add support for a quality measurement
Mauro Carvalho Chehab [Mon, 21 Jan 2013 19:14:54 +0000 (17:14 -0200)]
dvb-fe: Add support for a quality measurement

Handle the technical specifics for each delivery system and provide an
userspace-friendly quality measurement.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-fe: Fix PER statistics measurement
Mauro Carvalho Chehab [Mon, 21 Jan 2013 19:14:09 +0000 (17:14 -0200)]
dvb-fe: Fix PER statistics measurement

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb: Move the core of the stats logic to dvb-fe
Mauro Carvalho Chehab [Sat, 19 Jan 2013 15:04:49 +0000 (13:04 -0200)]
dvb: Move the core of the stats logic to dvb-fe

In order to duplicate similar work at dvbv5-utils and on other
userspace tools that uses this library, move the code that handles
the stats into the core.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-fe: Add a method to report PER
Mauro Carvalho Chehab [Sat, 19 Jan 2013 14:15:00 +0000 (12:15 -0200)]
dvb-fe: Add a method to report PER

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvbv5-zap: better display BER
Mauro Carvalho Chehab [Sat, 19 Jan 2013 11:30:18 +0000 (09:30 -0200)]
dvbv5-zap: better display BER

Instead of masking the BER scale, if the device has DVBv5 stats, it
will now display BER as it should be:

BER= 1.45x10^-6
BER= 329x10^-9

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-fe: Allow returning BER in a float value
Mauro Carvalho Chehab [Sat, 19 Jan 2013 11:28:36 +0000 (09:28 -0200)]
dvb-fe: Allow returning BER in a float value

BER is a fractional measure; allow returning it as such, in order
to let userspace to better display it.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-fe: Improve BER measurements for DVBv5
Mauro Carvalho Chehab [Wed, 23 Jan 2013 19:41:46 +0000 (17:41 -0200)]
dvb-fe: Improve BER measurements for DVBv5

Instead to display the BER rate calculated since the beginning of
the measurement, display the BER between two time intervals.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-fe: Add a flag to indicate if DVBv5 stats is in use
Mauro Carvalho Chehab [Sat, 19 Jan 2013 09:25:13 +0000 (07:25 -0200)]
dvb-fe: Add a flag to indicate if DVBv5 stats is in use

Instead of trying every time to see if DVBv5 stats are available,
if version is equal or upper than 5.10, a call to FE_GET_PROPERTY
will fail only if a more permanent error happened (EFAULT, ENOMEM,
and similar errors).

A successful call to FE_GET_PROPERTY that returns all stat props
with len=0 means that the frontend doesn't implement any DVBv5
stats. So, if it fails once, it can be assumed that there's no
DVBv5 stats implemented for that frontend.

As a bonus, the logic now knows exactly if a measure came from
DVBv3 or from DVBv5 API.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvbv5-zap: use the new DVBv5 stats API
Mauro Carvalho Chehab [Fri, 18 Jan 2013 20:04:32 +0000 (18:04 -0200)]
dvbv5-zap: use the new DVBv5 stats API

The statistics are now presented using the DVBv5 API information:
- proper scales;
- per-layer information.

The code should be backward compatible with legacy drivers.

On new drivers with multiple layers, this will now look like:

Lock    (0x1f) Signal= 6.25% C/N= 18.86 dB BER= 31550862 UCB= 17452451  Layer A: C/N= 15.09 dB BER= 4401807  Layer B: C/N= 20.50 dB BER= 31882245 UCB= 17452451

PS.: UCB and BER on the above are still not ok (UCB is due to a driver's
issue; BER requires more work inside the userspace code - and maybe at
driver's level - to work fine).

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvb-fe: add support for DVBv5 stats
Mauro Carvalho Chehab [Tue, 15 Jan 2013 01:25:41 +0000 (23:25 -0200)]
dvb-fe: add support for DVBv5 stats

In order to avoid breakage on packages that depend on this
library, backward support was preserved. Of course, the backward
call dvb_fe_retrieve_stats() won't bring anything new; it will
only return what it was provided already by the old API.

In order to get full benefit of the new API, the new DVBv5
library call should be used: dvb_fe_retrieve_stats_layer().

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agoSync with kernel, with DVB stats
Mauro Carvalho Chehab [Tue, 15 Jan 2013 00:44:00 +0000 (22:44 -0200)]
Sync with kernel, with DVB stats

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agodvbv5-zap: Allow to enable FE debug
Mauro Carvalho Chehab [Fri, 18 Jan 2013 14:25:33 +0000 (12:25 -0200)]
dvbv5-zap: Allow to enable FE debug

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agoAdd ASUS F2F with STK DC-1125 to upside down list
Gregor Jasny [Sat, 19 Jan 2013 16:22:53 +0000 (17:22 +0100)]
Add ASUS F2F with STK DC-1125 to upside down list

Signed-off-by: Gregor Jasny <gjasny@googlemail.com>
11 years agov4l2-compliance: add check whether the timestamp is monotonic.
Hans Verkuil [Thu, 10 Jan 2013 12:47:15 +0000 (13:47 +0100)]
v4l2-compliance: add check whether the timestamp is monotonic.

Note: today we only have 'Unknown' (legacy) and 'Monotonic' timestamps.
If new timestamp types are added, then v4l2-compliance needs to be modified
to add support for those.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-ctl: show used timestamp type.
Hans Verkuil [Thu, 10 Jan 2013 12:45:49 +0000 (13:45 +0100)]
v4l2-ctl: show used timestamp type.

Implement support for the new timestamp type encoded in the buffer flags.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l-utils: sync with latest kernel.
Hans Verkuil [Thu, 10 Jan 2013 12:24:13 +0000 (13:24 +0100)]
v4l-utils: sync with latest kernel.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agoqv4l2: fix setting frequency for radio modulators
Hans Verkuil [Tue, 8 Jan 2013 07:28:39 +0000 (08:28 +0100)]
qv4l2: fix setting frequency for radio modulators

Setting the frequency worked fine for TV and radio receivers, but not
for modulators.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-compliance: allow EINVAL to be returned by S/TRY_FMT.
Hans Verkuil [Mon, 7 Jan 2013 09:44:45 +0000 (10:44 +0100)]
v4l2-compliance: allow EINVAL to be returned by S/TRY_FMT.

See http://www.mail-archive.com/linux-media@vger.kernel.org/msg56550.html why
this is needed for now.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agolibv4l-mplane: Fix compile flags to allow anonymous unions with gcc 4.5.5
André Roth [Thu, 3 Jan 2013 13:44:24 +0000 (14:44 +0100)]
libv4l-mplane: Fix compile flags to allow anonymous unions with gcc 4.5.5

In c99 mode, anonymous unions are ignored with "declaration does not
declare anything", at least on gcc 4.5.5.

Signed-off-by: André Roth <neolynx@gmail.com>
11 years agocontrib/m920x/m920x_parse.pl: silence a warning
Antonio Ospite [Sat, 29 Dec 2012 21:37:08 +0000 (18:37 -0300)]
contrib/m920x/m920x_parse.pl: silence a warning

Silence a warning due to the way get_line() is supposed to be called:
  Use of uninitialized value $cmd in split at m920x_parse.pl line 118

Signed-off-by: Antonio Ospite <ospite@studenti.unina.it>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agocontrib/m920x/m920x_parse.pl: stricter check when extracting firmware
Antonio Ospite [Sat, 29 Dec 2012 21:37:07 +0000 (18:37 -0300)]
contrib/m920x/m920x_parse.pl: stricter check when extracting firmware

Extract firmware only from the right messages, skip the other messages.

Signed-off-by: Antonio Ospite <ospite@studenti.unina.it>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agov4l2-compliance: Print invalid error_idx values in control tests
Laurent Pinchart [Mon, 24 Dec 2012 12:55:42 +0000 (13:55 +0100)]
v4l2-compliance: Print invalid error_idx values in control tests

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agov4l2-compliance: Print invalid return codes in control tests
Laurent Pinchart [Mon, 24 Dec 2012 12:55:41 +0000 (13:55 +0100)]
v4l2-compliance: Print invalid return codes in control tests

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
11 years agokeytable: always use hexadecimal values for scancodes
Mauro Carvalho Chehab [Mon, 17 Dec 2012 12:45:07 +0000 (10:45 -0200)]
keytable: always use hexadecimal values for scancodes

Makes it more consistent.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: be clear when a value is in hexadecimal
Mauro Carvalho Chehab [Fri, 14 Dec 2012 18:12:34 +0000 (16:12 -0200)]
keytable: be clear when a value is in hexadecimal

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: properly print the other MSC_* event names
Mauro Carvalho Chehab [Fri, 14 Dec 2012 18:07:49 +0000 (16:07 -0200)]
keytable: properly print the other MSC_* event names

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: Make the parsing code more generic
Mauro Carvalho Chehab [Fri, 14 Dec 2012 18:02:36 +0000 (16:02 -0200)]
keytable: Make the parsing code more generic

It will now be a little more generic, as the code will be able
to print all input event messages with its proper name.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: Add the remaining other unhandled events
Mauro Carvalho Chehab [Fri, 14 Dec 2012 17:13:47 +0000 (15:13 -0200)]
keytable: Add the remaining other unhandled events

That helps to latter add parsing code for those events, if ever
needed.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: Put events in numeric order
Mauro Carvalho Chehab [Fri, 14 Dec 2012 17:11:56 +0000 (15:11 -0200)]
keytable: Put events in numeric order

That helps to identify what's missing

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: also print code for unknown/unhandled events
Mauro Carvalho Chehab [Fri, 14 Dec 2012 17:10:18 +0000 (15:10 -0200)]
keytable: also print code for unknown/unhandled events

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: Add handler for EV_REL type
Mauro Carvalho Chehab [Fri, 14 Dec 2012 17:05:17 +0000 (15:05 -0200)]
keytable: Add handler for EV_REL type

Those are typically used on mouse events.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: reorder functions to put similar code together
Mauro Carvalho Chehab [Fri, 14 Dec 2012 17:04:18 +0000 (15:04 -0200)]
keytable: reorder functions to put similar code together

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agokeytable: merge two identical structures into one
Mauro Carvalho Chehab [Fri, 14 Dec 2012 17:02:27 +0000 (15:02 -0200)]
keytable: merge two identical structures into one

While here, better name the event structures.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
11 years agoir-keytable: Add support for testing EV_ABS events
Mauro Carvalho Chehab [Fri, 14 Dec 2012 16:00:23 +0000 (14:00 -0200)]
ir-keytable: Add support for testing EV_ABS events

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