Commit Graph

81 Commits

Author SHA1 Message Date
Hans Verkuil
d88aab53bd [media] v4l2: make vidioc_s_jpegcomp const
Write-only ioctls should have a const argument in the ioctl op.
Do this conversion for vidioc_s_jpegcomp.
Adding const for write-only ioctls was decided during the 2012 Media Workshop.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-26 10:43:04 -03:00
Hans Verkuil
2cca7d4e4d [media] v4l2: remove experimental tag from a number of old drivers
A number of old drivers still had the experimental tag. Time to remove it.
It concerns the following drivers:
VIDEO_TLV320AIC23B
USB_STKWEBCAM
VIDEO_CX18
VIDEO_CX18_ALSA
VIDEO_ZORAN_AVS6EYES
DVB_USB_AF9005
MEDIA_TUNER_TEA5761
VIDEO_NOON010PC30
This decision was taken during the 2012 Media Workshop.

Signed-off-by: Hans Verkuil <hans.verkuil@cisco.com>
Reviewed-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-26 09:44:42 -03:00
Ezequiel Garcia
8ac456495a [media] stk1160: Stop device and unqueue buffers when start_streaming() fails
If start_streaming() fails (e.g. out of memory) the driver needs to
rewind the start procedure. This implies possibly stopping the device
and clearing the buffer queue.

Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-25 10:56:48 -03:00
Ezequiel Garcia
c6b69c6c58 [media] stk1160: Fix s_fmt and try_fmt implementation
The driver was expecting to get a valid pixelformat on s_fmt and try_fmt.
This is wrong, since the user may pass a bitmask and expect the driver
to change it, returning a valid (fourcc) pixelformat.
This problem was spotted by v4l2-compliance.

Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-25 10:53:34 -03:00
Ezequiel Garcia
065741840b [media] stk1160: Handle urb allocation failure condition properly
When an urb buffer can't be allocated, the currently allocated
buffer count must be saved so they can properly released.
Moreover, it's sufficient to call stk1160_free_isoc to have
all urb buffers released.

Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-25 10:52:58 -03:00
Ezequiel Garcia
18ad89659a [media] stk1160: Make kill/free urb debug message more verbose
This is just a cleaning patch to produce more useful
debug messages.

Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-25 10:38:43 -03:00
Antti Palosaari
b072eee034 [media] ce6230: use Kernel dev_foo() logging
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:50:42 -03:00
Antti Palosaari
3cedcab88f [media] ec168: use Kernel dev_foo() logging
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:50:28 -03:00
Antti Palosaari
8152768892 [media] gl861: use Kernel dev_foo() logging
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:50:13 -03:00
Antti Palosaari
a835ae6777 [media] au6610: use Kernel dev_foo() logging
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:50:03 -03:00
Antti Palosaari
119f7a8c5d [media] af9035: use Kernel dev_foo() logging
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:49:22 -03:00
Antti Palosaari
e1e9e510bb [media] af9015: correct few error codes
Plain '-1' is not very good error code. Use more suitable error
code definitions.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:48:58 -03:00
Antti Palosaari
2e35c66f13 [media] af9015: improve af9015_eeprom_hash()
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:48:41 -03:00
Antti Palosaari
f224749b68 [media] af9015: use Kernel dev_foo() logging
... and some minor logging changes.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:48:10 -03:00
Antti Palosaari
66b3c4deb9 [media] rtl2830: use .get_if_frequency()
Use .get_if_frequency() as all used tuner drivers
(mt2060/qt1010/mxl5005s) supports it.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:18:10 -03:00
Antti Palosaari
6a60e3f653 [media] ec168: declare MODULE_FIRMWARE
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:08:21 -03:00
Antti Palosaari
bab9b4fe08 [media] af9015: declare MODULE_FIRMWARE
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 20:07:31 -03:00
Antti Palosaari
5be65721a7 [media] rtl28xxu: add support for tua9001 tuner based devices
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 19:41:55 -03:00
Antti Palosaari
1835af1003 [media] af9035: relax frontend callback error handling
It is not good idea to return error for missing callback
handler as whole callback as optional and could be missing
by intentionally.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 19:41:17 -03:00
Antti Palosaari
832cc7cdfb [media] rtl2832: separate tuner specific init from general
It is first step closer to support multiple tuners.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 17:31:57 -03:00
Antti Palosaari
c2d246d1f0 [media] rtl28xxu: Dexatek DK DVB-T Dongle [1d19:1101]
It is RTL2832U + FC2580 reference design.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 17:13:10 -03:00
Antti Palosaari
384df49a6a [media] rtl28xxu: add support for FCI FC2580 silicon tuner driver
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 17:12:54 -03:00
Jose Alberto Reguero
aa468cc550 [media] ttusb2: add toggle to the tt3650_rc_query function
This patch add the toggle bit to the tt3650_rc_query function of the ttusb2
driver.

Signed-off-by: Jose Alberto Reguero <jareguero@telefonica.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 17:06:53 -03:00
Mauro Carvalho Chehab
fac44ee564 [media] pd-alsa: fix compilation breakage by commit da35de640
commit da35de640 broke compilation, as it reverted the name of the
usb card free function.

Cc: Alan Cox <alan@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 16:51:57 -03:00
Alan Cox
da35de640a [media] tlg2300: fix missing check for audio creation
If we fail to set up the capture device we go through negative indexes and
badness happens. Add the missing test.
Resolves-bug: https://bugzilla.kernel.org/show_bug.cgi?id=44551

Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-23 13:50:30 -03:00
Philipp Dreimann
6d60805fd2 [media] Add the usb id of the Trekstor DVB-T Stick Terres 2.0
It needs the e4000 tuner driver.

Signed-off-by: Philipp Dreimann <philipp@dreimann.net>
Acked-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-18 12:54:57 -03:00
Antti Palosaari
542f6a52b6 [media] rtl28xxu: add support for Elonics E4000 tuner
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-18 12:25:38 -03:00
Antti Palosaari
8b03663644 [media] rtl28xxu: correct usb_clear_halt() usage
It is not allowed to call usb_clear_halt() after urbs are submitted.
That causes oops sometimes. Move whole streaming_ctrl() logic to
power_ctrl() in order to avoid wrong usb_clear_halt() use. Also,
configuring streaming endpoint in streaming_ctrl() sounds like a
little bit wrong as it is aimed for control stream gate.

Reported-by: Hin-Tak Leung <htl10@users.sourceforge.net>
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-18 12:22:01 -03:00
Ezequiel García
b0c47e1ec2 [media] stk1160: Remove unused 'ifnum' variable
Since ifnum is not used anywhere it is safe to remove it.
This was spotted by Hans's media_tree daily build.

Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 11:49:14 -03:00
Ezequiel García
c11f187bd0 [media] stk1160: Remove unneeded struct vb2_queue clearing
struct vb2_queue is allocated through kzalloc as part of a larger struct,
there's no need to clear it.

Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 11:38:41 -03:00
Antti Palosaari
d89b9369e0 [media] dvb_usb_v2: use dvb_usb_dbg_usb_control_msg()
Convert drivers: au6610, ce6230, ec168, rtl28xxu for
dvb_usb_dbg_usb_control_msg() macro.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 11:35:59 -03:00
Antti Palosaari
4ab79283aa [media] dvb_usb_v2: add debug macro dvb_usb_dbg_usb_control_msg
For dumping usb_control_msg().

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 11:35:58 -03:00
Antti Palosaari
f5a8a78a1b [media] au6610: define reset_resume
After qt1010 change that device seems to survive from reset resume.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 11:34:47 -03:00
Antti Palosaari
e1f4326971 [media] rtl28xxu: fix rtl2832u module reload fails bug
This is workaround / partial fix.

rtl2832u_power_ctrl() and rtl2832u_frontend_attach() needs to
be go through carefully and fix properly. There is clearly
some logical errors when handling power-management ang GPIOs...

Signed-off-by: Antti Palosaari <crope@iki.fi>
Cc: Thomas Mair <thomas.mair86@googlemail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 09:29:49 -03:00
Antti Palosaari
62751a801d [media] rtl28xxu: stream did not start after stop on USB3.0
Stream did not start anymore after stream was stopped once.

Following error can be seen, xhci_hcd
WARN Set TR Deq Ptr cmd failed due to incorrect slot or ep state.

usb_clear_halt for streaming endpoint helps.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 09:29:27 -03:00
Antti Palosaari
0b8623cadd [media] gl861: reset_resume support
It survives now on reset_resume.

Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 09:19:43 -03:00
Laurent Pinchart
3851c34a83 [media] uvcvideo: Remove outdated comment
The uvcvideo driver now supports USERPTR, and isn't limited to YUYV and
MJPEG anymore.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-15 09:17:36 -03:00
Frank Schäfer
780d61704c [media] gspca_pac7302: extend register documentation
Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:54:49 -03:00
Frank Schäfer
0e68785db5 [media] gspca_pac7302: avoid duplicate calls of the image quality adjustment functions on capturing start
There is no need to call the image quality adjustment functions in sd_start.
The gspca main driver calls v4l2_ctrl_handler_setup in gspca_init_transfer,
which already applies all image control values.

Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:54:34 -03:00
Frank Schäfer
f58e5cdf12 [media] gspca_pac7302: increase default value for white balance temperature
The current white balance temperature default value is 4, which is much too small (possible values are 0-255).
Improve the picture quality by increasing the default value to 55, which is the default value used by the Windows driver.

Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:54:12 -03:00
Frank Schäfer
b1a19c0165 [media] gspca_pac7302: add sharpness control
The Windows driver uses page 0 register 0xb6 for sharpness adjustment.

Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:53:57 -03:00
Frank Schäfer
db43b9ca2f [media] gspca_pac7302: make red balance and blue balance controls work again
Fix a regression from kernel 3.4 which has been introduced with the conversion of the gspca driver to the v4l2 control framework.

Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Cc: stable@kernel.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:53:42 -03:00
Frank Schäfer
97d2fbf501 [media] gspca_pac7302: add support for device 1ae7:2001 Speedlink Snappy Microphone SL-6825-SBK
Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
Cc: stable@kernel.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:53:26 -03:00
Hans de Goede
36adfca94a [media] gspca: Fix input urb creation / destruction surrounding suspend resume
1) We always re-create the input-urb on resume, so we must also always
   destroy it on suspend to avoid leaking it
2) If we're going to do an init_transfer, then that will destroy the urb
   before starting the stream (nop if there is none), and (re-)create it
   once the stream is started. So there is little use in creating it, if
   we're going to do an init_transfer immediately afterward

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:53:10 -03:00
Hans de Goede
844db450e6 [media] gspca: Update / fix various comments wrt workqueue usb_lock usage
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:52:58 -03:00
Hans de Goede
ff8f25d326 [media] gspca_finepix: Remove unnecessary lock/unlock call
gspca_main: init_transfer does not do anything between calling sd_start
(which starts the workqueue) and releasing the usb_lock, so this
synchronization is a nop, remove it.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:44:34 -03:00
Hans de Goede
345321dc9c [media] gspca: Don't set gspca_dev->dev to NULL before stop0
In commit a3d6e8cc0e gspca_dev->dev is set
to NULL on disconnect, before calling stop0. The plan was to get rid of
gspca_dev->present and instead simply check for gspca_dev->dev everywhere
where we were checking for present. This should be race free since all users
of gspca_dev->dev hold the usb_lock, or so I thought.

But I was wrong, drivers which use a work-queue + synchronous bulk transfers
to get the video data don't hold the usb_lock while doing so, their stop0
callbacks stop the workqueue, so they won't be using gspca_dev->dev anymore
after the stop0 call, but they might be dereferincing it before, so we should
not set gspca_dev->dev to NULL on disconnect before calling stop0.

This also means that the workqueue functions in these drivers cannot
use gspca_dev->dev to check if they need to stop because of disconnection,
so we will need to keep gspca_dev->present around, and set that to 0 on
disconnect, before calling stop0. Unfortunately as part of the plan to remove
gspca_dev->present, these workqueues where already moved over to checking
for gspca_dev->dev instead of gspca_dev->present as part of commit
254902b01d, so this patch also reverts those
parts of that commit.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:42:17 -03:00
Peter Senna Tschudin
8c96f0a207 [media] drivers/media/usb/gspca/cpia1.c: fix error return code
Convert a nonnegative error return code to a negative one, as returned
elsewhere in the function.

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

// <smpl>
(
if@p1 (\(ret < 0\|ret != 0\))
 { ... return ret; }
|
ret@p1 = 0
)
... when != ret = e1
    when != &ret
*if(...)
{
  ... when != ret = e2
      when forall
 return ret;
}

// </smpl>

Signed-off-by: Peter Senna Tschudin <peter.senna@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:42:16 -03:00
Emil Goode
50c457a950 [media] gspca: dubious one-bit signed bitfield
This patch changes some signed integers to unsigned because
they are not intended for negative values and sparse
is making noise about it.

Sparse gives eight of these errors:
drivers/media/usb/gspca/ov519.c:144:29: error: dubious one-bit signed bitfield

Signed-off-by: Emil Goode <emilgoode@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:42:15 -03:00
Ezequiel Garcia
4a7ec2db26 [media] pwc: Remove unneeded struct vb2_queue clearing
struct vb2_queue is allocated through kzalloc as part of a larger struct,
there's no need to clear it.

Cc: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ezequiel Garcia <elezegarcia@gmail.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2012-09-13 17:42:15 -03:00