- 24 Mar, 2013 40 commits
-
-
Hans Verkuil authored
Convert this driver to videobuf2 and core locking. Signed-off-by:
Hans Verkuil <hans.verkuil@cisco.com> Signed-off-by:
Mauro Carvalho Chehab <mchehab@redhat.com>
-
Hans Verkuil authored
Remove struct go7007_file: all fields contained in that struct are moved to the go7007 struct since they are really global values. The lock has just been deleted (what's the point of a per-fh lock??). Signed-off-by:
-
Hans Verkuil authored
The s_crystal_freq operation has to be called for the saa7115 to set up the audio clock correctly. Signed-off-by:
-
Hans Verkuil authored
There really is no need to split out the board IDs for each tuner. That's what the tuner_type is for, after all. Signed-off-by:
-
Hans Verkuil authored
The go7007 produces elementary streams, so we shouldn't use the STREAM_TYPE control, since that is for multiplexed streams. Instead use V4L2_PIX_FMT_MPEG1/2/4. Initially set up all the values for MPEG-2 dvd-mode, and select the dvd_mode field if those values match that setup precisely. Clean up lots of obsolete code relating to the custom ioctls for the MPEG support. Signed-off-by:
-
Hans Verkuil authored
- The Adlink is initially detected as a sensor board, so the driver config is set up as if it was a sensor. Later the driver discovers that it really is a video capture card and that means that the driver configuration has to be updated for a PAL/NTSC capture card. - Setup the correct initial std based on the TV tuner type. - Remember the exact std the user selected, don't map it to STD_PAL or NTSC. - Use v4l2_disable_ioctl to disable ioctls based on the board config and drop a lot of checking code in each of those ioctls since that is no longer needed. - enum_input should use tvnorms to fill its std field. - configure the input, std and initial tuner frequency at driver initialization. - fix std handling in the s2250 driver. - the code handling scaling for devices without a scaler was broken. This is now fixed. - correctly set readbuffers and capability in g/s_parm. - remove the bogus test for capturemode in s_parm. - correctly implement enum_framesizes/intervals. Signed-off-by:
-
Hans Verkuil authored
Signed-off-by:
-
Hans Verkuil authored
Signed-off-by:
-
Hans Verkuil authored
Signed-off-by:
-
Hans Verkuil authored
Just add a read-only V4L2_CID_JPEG_ACTIVE_MARKER control to replace the JPEGCOMP support. Signed-off-by:
-
Hans Verkuil authored
- use the v4l2_device's release() callback - remove the unnecessary ref_count - don't free usb data structures on disconnect, only do that in the final release callback. This is the correct way in order to safely handle disconnect and removal of modules. Signed-off-by:
-
Hans Verkuil authored
Don't load it everytime you stop encoding. Instead remember it. Another reason for not loading it every time is that this could be called from within the release() file operation, which turns out to be deadly. Signed-off-by:
-
Hans Verkuil authored
- Don't pass data allocated on the stack to usb_control_msg. - Use dma_mapping_error after calling dma_map_page(). Signed-off-by:
-
Hans Verkuil authored
It's deprecated and replaced by g_std. Since this driver already implements g_std the use of current_norm can just be removed. Signed-off-by:
-
Hans Verkuil authored
And don't set the version field, the core does that for you. Signed-off-by:
-
Hans Verkuil authored
The v2 of this function doesn't do DMA to objects on the stack like its predecessor does. Signed-off-by:
-
Hans Verkuil authored
Since we now know what audio inputs there are, we can also get/set and enumerate them. Signed-off-by:
-
Hans Verkuil authored
These are now no longer used and so they can be removed. Signed-off-by:
-
Hans Verkuil authored
Instead of using the wis-* drivers we now use the standard 'proper' subdev drivers. The board configuration tables now also list the possible audio inputs, this will be used later to implement audio inputs. Special mention deserves a little change in set_capture_size() where the height passed to s_mbus_fmt is doubled: that is because the saa7115 driver expects to see the frame height, not the field height as the wis_saa7115 driver did. Another change is that the tuner input is moved from last to the first input, which is consistent with the common practice in other video drivers. Signed-off-by:
-
Volokh Konstantin authored
Do i2c initialization via struct item as tw2804 has a 0x00 i2c address, so we need to use the I2C_CLIENT_TEN flag for validity. Signed-off-by:
Volokh Konstantin <volokh84@gmail.com> Signed-off-by:
-
Volokh Konstantin authored
Switch off all ADC (max 4) with first init, we control it when starting/stopping streaming. Signed-off-by:
-
Hans Verkuil authored
This is based on the wis-tw2804.c driver that's part of the go7007 driver. It has been converted to a v4l subdev driver by Volokh Konstantin, and I made additional cleanups. Based on work by: Volokh Konstantin <volokh84@gmail.com> Signed-off-by:
-
Hans Verkuil authored
This based on the wis-tw9903.c driver that's part of the go7007 driver. It has been converted to a v4l subdev driver by Pete Eberlein, and I made additional cleanups. Based on work by: Pete Eberlein <pete@sensoray.com> Signed-off-by:
-
Hans Verkuil authored
This based on the wis-uda1342.c driver that's part of the go7007 driver. It has been converted to a v4l subdev driver by Pete Eberlein, and I made additional cleanups. Based on work by: Pete Eberlein <pete@sensoray.com> Signed-off-by:
-
Hans Verkuil authored
This based on the wis-ov7640.c driver that's part of the go7007 driver. It has been converted to a v4l subdev driver by Pete Eberlein, and I made additional cleanups. Based on work by: Pete Eberlein <pete@sensoray.com> Signed-off-by:
-
Hans Verkuil authored
The Sony BTF PG472Z has an internal MPX to deal with mono/stereo/bilingual audio. This is split off from the wis-sony-tuner driver that is part of the go7007 driver as it should be a separate i2c sub-device driver. The wis-sony-tuner is really three i2c devices: a standard tuner, a tda9887 compatible demodulator and this mpx. After this patch the wis-sony-tuner can be replaced by this driver and the standard tuner driver. Signed-off-by:
-
Hans Verkuil authored
This adds support for three Sony BTF tuners: TUNER_SONY_BTF_PG472Z: PAL+SECAM TUNER_SONY_BTF_PK467Z: NTSC-M-JP TUNER_SONY_BTF_PB463Z: NTSC-M These come from the go7007 staging driver where they were implemented in the wis-sony-tuner i2c driver. Adding support for these tuners to tuner-types.c is the first step towards removing the wis-sony-tuner driver. Signed-off-by:
-
Hans Verkuil authored
The i2c_xfer functions didn't return the proper error codes and (especially important) on success they returned 0 instead of the number of transferred messages. Signed-off-by:
-
Hans Verkuil authored
Some devices expect a double rate ASCLK. Add a flag to let the driver know through the s_crystal_freq call. Signed-off-by:
-
Hans Verkuil authored
The saa7115 has better PAL/NTSC detection, so it can detect PAL even though the chip is currently set up for NTSC. Signed-off-by:
-
Hans Verkuil authored
Needed by the go7007 driver: it assumes a different polarity of the IDQ signal, so we need to be able to tell the saa7115 about this. Signed-off-by:
-
Hans Verkuil authored
Control whether video sequence headers should be repeated. Signed-off-by:
-
Hans Verkuil authored
Just checking if the op exists isn't correct, you should check if the ioctl is valid (which implies that the op exists as well). One exception is g_std: if current_norm is non-zero, then the g_std op may be absent. This sort of weird behavior is one of the reasons why I am trying to get rid of current_norm. This patch fixes the case where the g/s_std op is set, but these ioctls are disabled. This can happen in drivers supporting multiple models, some that have TV input (and support the STD API) and some that have a sensor (and do not support the STD API). In the latter case qv4l2 would still show the Standards combobox. Signed-off-by:
-
Hans Verkuil authored
This patch adds a v4l2_subdev_has_op() macro and a v4l2_device_has_op macro to quickly check if a specific subdev or any subdev supports a particular subdev operation. This makes it easy for drivers to disable certain ioctls if none of the subdevs supports the necessary functionality. Signed-off-by:
-
Andy Walls authored
When calling v4l2_ctrl_add_handler(), lockdep would detect a potential recursive locking problem on a situation that is by design intended and not a recursive lock. This happened because all struct v4l2_ctrl_handler.lock mutexes were created as members of the same lock class in v4l2_ctrl_handler_init(), and v4l2_ctrl_add_handler() takes the hdl->lock on two different v4l2_ctrl_handler instances. This change breaks the large lockdep lock class for struct v4l2_ctrl_handler.lock and breaks it into v4l2_ctrl_handler instantiation specific lock classes with meaningful class names. This will validly eliminate lockdep alarms for v4l2_ctrl_handler locking validation, as long as the relationships between drivers adding v4l2 controls to their own handler from other v4l2 drivers' control handlers remains straightforward. struct v4l2_ctrl_handler.lock lock classes are created with names such that the output of cat /proc/lockdep indicates where in the v4l2 driver code v4l2_ctrl_handle_init() is being called on instantiations: ffffffffa045f490 FD: 10 BD: 8 +.+...: cx2341x:1534:(hdl)->lock ffffffffa0497d20 FD: 12 BD: 2 +.+.+.: saa7115:1581:(hdl)->lock ffffffffa04ac660 FD: 14 BD: 2 +.+.+.: msp3400_driver:756:(hdl)->lock ffffffffa0484b90 FD: 12 BD: 1 +.+.+.: ivtv_gpio:366:(&itv->hdl_gpio)->lock ffffffffa04eb530 FD: 11 BD: 2 +.+.+.: cx25840_core:1982:(&state->hdl)->lock ffffffffa04fbc80 FD: 11 BD: 3 +.+.+.: wm8775:246:(&state->hdl)->lock Some lock chains, that were previously causing the recursion alarms, are now visible in the output of cat /proc/lockdep_chains: irq_context: 0 [ffffffffa0497d20] saa7115:1581:(hdl)->lock [ffffffffa045f490] cx2341x:1534:(hdl)->lock irq_context: 0 [ffffffffa04ac660] msp3400_driver:756:(hdl)->lock [ffffffffa045f490] cx2341x:1534:(hdl)->lock irq_context: 0 [ffffffffa0484b90] ivtv_gpio:366:(&itv->hdl_gpio)->lock [ffffffffa045f490] cx2341x:1534:(hdl)->lock irq_context: 0 [ffffffffa04eb530] cx25840_core:1982:(&state->hdl)->lock [ffffffffa045f490] cx2341x:1534:(hdl)->lock irq_context: 0 [ffffffffa04fbc80] wm8775:246:(&state->hdl)->lock [ffffffffa045f490] cx2341x:1534:(hdl)->lock Signed-off-by:
Andy Walls <awalls@md.metrocast.net> [hans.verkuil@cisco.com: keep mutex_init in v4l2_ctrl_handler_init_class] Signed-off-by:
-
Masanari Iida authored
Correct spelling typos. Signed-off-by:
Masanari Iida <standby24x7@gmail.com> Signed-off-by:
-
Hans Verkuil authored
The core debug code can now be simplified since all the write-only ioctls are now const and will not modify the data they pass to the drivers. So instead of logging write-only ioctls before the driver is called this can now be done afterwards, which is cleaner when it comes to error reporting as well. This also fixes a logic error in the debugging code where there was one 'else' too many. Signed-off-by:
-
Hans Verkuil authored
This ioctl is defined as IOW, so pass the argument as const. Signed-off-by:
Guennadi Liakhovetski <g.liakhovetski@gmx.de> Acked-by:
Lad, Prabhakar <prabhakar.csengg@gmail.com> Signed-off-by:
-
Hans Verkuil authored
The ivtv_itvc function receives a pointer to v4l2_dbg_register. When we add const to that pointer in the s_register case we will run into a problem here since this common function would discard const in that case. So change this function so it receives the address and a pointer to a value. In addition we now set the size field in the g_register function which is where it belongs. This will simplify the next patch where const will be added to s_register. Signed-off-by:
-
Hans Verkuil authored
Never print more than the size of the buffer containing the name. Signed-off-by:
-
