- 24 Feb, 2020 40 commits
-
-
Andrzej Pietrasiewicz authored
Luma and chroma qtables need to be written into two 16-register blocks, each table consisting of 64 bytes total. The blocks are contiguous and start at offset 0 for luma and at offset 0x40 for chroma. The seemingly innocent optimization of writing the two blocks using one loop causes side effects which result in improper values of quantization tables being used by the hardware during encoding. Visually this results in macroblocking artifacts around contrasting edges in encoded images. The artifacts look like horizontally flipped shadows of the said edges. Changing the write operations to non-relaxed variant doesn't help. This patch removes this premature optimization and after this change the macroblocking artifacts around contrasting edges are gone. Signed-off-by:
Andrzej Pietrasiewicz <andrzej.p@collabora.com> Tested-by:
Ezequiel Garcia <ezequiel@collabora.com> Signed-off-by:
Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by:
Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
-
Andrzej Pietrasiewicz authored
The quantization tables as defined in the file (luma_q_table, chroma_q_table) are in fact in linear order. The JPEG file header, which is not generated by the hardware, but must be programatically created with the CPU, expects the table in zigzag order. On the other hand, the hardware doesn't expect neither linear, nor zigzag order. Instead it expects the quantization tables in vertical groups of four quantization parameters, and the groups are organized in blocks of two vertically adjacent groups. On top of that the blocks must be provided to the hardware in this order: leftmost top block, leftmost bottom block, second leftmost top block, second leftmost bottom block and so on. So, if this is the quantization table in linear order: 0x10, 0x0b, 0x0a, 0x10, 0x18, 0x28, 0x33, 0x3d, 0x0c, 0x0c, 0x0e, 0x13, 0x1a, 0x3a, 0x3c, 0x37, 0x0e, 0x0d, 0x10, 0x18, 0x28, 0x39, 0x45, 0x38, 0x0e, 0x11, 0x16, 0x1d, 0x33, 0x57, 0x50, 0x3e, 0x12, 0x16, 0x25, 0x38, 0x44, 0x6d, 0x67, 0x4d, 0x18, 0x23, 0x37, 0x40, 0x51, 0x68, 0x71, 0x5c, 0x31, 0x40, 0x4e, 0x57, 0x67, 0x79, 0x78, 0x65, 0x48, 0x5c, 0x5f, 0x62, 0x70, 0x64, 0x67, 0x63 then the hardware expects this in its consecutive registers: 0x100c0e0e, 0x0b0c0d11, 0x12183148, 0x1623405c, 0x0a0e1016, 0x1013181d, 0x25374e5f, 0x38405762, and so on. Consequently, the same area of memory cannot be used both for dumping it into the JPEG file header and writing its contents to the hardware registers. Instead, a separate pair of arrays is added for properly reordered quantization tables, to be read with get_unaligned_be32() and linearly written to the registers. The "ctx" parameter is not needed any more for hantro_jpeg_get_qtable(). Signed-off-by:
-
Andrzej Pietrasiewicz authored
The table is actually different in the document than in this file, so align this file with the document. Signed-off-by:
-
Andrzej Pietrasiewicz authored
Since (luma/chroma)_qtable is an array of unsigned char, indexing it returns consecutive byte locations, but we are supposed to read the arrays in four-byte words. Consequently, we should be pointing get_unaligned_be32() at consecutive word locations instead. Signed-off-by:
-
Colin Ian King authored
There is a spelling mistake in a v4l2_err message. Fix it. Signed-off-by:
Colin Ian King <colin.king@canonical.com> Reviewed-by:
Michael Tretter <m.tretter@pengutronix.de> Signed-off-by:
-
Dafna Hirschfeld authored
Ignore errors returned from kthread_stop since the vimc subdevices should still be notified that streaming stopped so they can release the memory for the streaming, and also kthread should be set to NULL. kthread_stop can return -EINTR in case the thread did not yet run. This can happen if userspace calls streamon and streamoff right after. Signed-off-by:
Dafna Hirschfeld <dafna.hirschfeld@collabora.com> Acked-by:
Helen Koike <helen.koike@collabora.com> Signed-off-by:
-
Helen Koike authored
The only places which make sese to allow users to enable or disable links are: * between sensors and isp: So users can select which sensor should be used while streaming * between isp and the resizers: | v here rkisp1_isp:2 -> rkisp1_resizer_mainpath -> rkisp1_mainpath (capture) \-> rkisp1_resizer_selfpath -> rkisp1_selfpath (capture) ^ here | So users can disable one of the capture paths when unused, to avoid worring about matching formats. Make the following links immutable to simplify userspace: rkisp1_resizer_mainpath -> rkisp1_mainpath rkisp1_resizer_selfpath -> rkisp1_selfpath rkisp1_params -> rkisp1_isp rkisp1_isp -> rkisp1_stats Signed-off-by:Niklas Söderlund <niklas.soderlund@ragnatech.se> Signed-off-by:
-
Cengiz Can authored
There's a mistakenly written self assignment in `static int vpfe_enum_fmt_vid_cap(..)`. Fixed it according to Prabhakar Lad's feedback. Signed-off-by:
Cengiz Can <cengiz@kernel.wtf> Signed-off-by:
-
YueHaibing authored
platform_get_irq() will call dev_err() itself on failure, so there is no need for the driver to also do this. This is detected by coccinelle. Signed-off-by:
YueHaibing <yuehaibing@huawei.com> Signed-off-by:
-
Hans Verkuil authored
The last driver was converted to use VFL_TYPE_VIDEO, so the old _GRABBER enum can now be removed. Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
-
Hans Verkuil authored
'GRABBER' is a weird name, all other types map to the /dev device names. Rename to 'VIDEO' to be consistent with the other types. Signed-off-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by:
-
Hans Verkuil authored
We currently have the following devnode types: enum vfl_devnode_type { VFL_TYPE_GRABBER = 0, VFL_TYPE_VBI, VFL_TYPE_RADIO, VFL_TYPE_SUBDEV, VFL_TYPE_SDR, VFL_TYPE_TOUCH, VFL_TYPE_MAX /* Shall be the last one */ }; They all make sense, except for the first: GRABBER really refers to /dev/videoX devices, which can be capture, output or m2m, so 'grabber' doesn't even refer to their function anymore. Let's call a spade a spade and rename this to VFL_TYPE_VIDEO. Signed-off-by:Sakari Ailus <sakari.ailus@linux.intel.com> Reviewed-by:
-
Chuhong Yuan authored
allegro_open() misses checks for v4l2_m2m_ctx_init() and results of v4l2_ctrl_new* calls. Add checks and error handlers to fix the problems. Signed-off-by:
Chuhong Yuan <hslester96@gmail.com> Signed-off-by:
-
Chuhong Yuan authored
This driver forgets to disable and unprepare clock when remove. Add a call to clk_disable_unprepare() to fix it. Signed-off-by:
Krzysztof Kozlowski <krzk@kernel.org> Signed-off-by:
-
Mohammad Rasim authored
This is an NEC remote control device shipped with the Videostrong KII Pro tv box as well as other devices from videostrong. Signed-off-by:
Mohammad Rasim <mohammad.rasim96@gmail.com> Signed-off-by:
Sean Young <sean@mess.org> Signed-off-by:
-
Colin Ian King authored
The variable rc is being initialized with a value that is never read and it is being updated later with a new value. The initialization is redundant and can be removed. Addresses-Coverity: ("Unused value") Signed-off-by: -
Johan Hovold authored
Make sure to use the current alternate setting when verifying the interface descriptors to avoid submitting an URB to an invalid endpoint. Failing to do so could cause the driver to misbehave or trigger a WARN() in usb_submit_urb() that kernels with panic_on_warn set would choke on. Fixes: c4018fa2 ("[media] dib0700: fix RC support on Hauppauge Nova-TD") Cc: stable <stable@vger.kernel.org> # 3.16 Signed-off-by:
Johan Hovold <johan@kernel.org> Signed-off-by:
-
Johan Hovold authored
commit 1b976fc6 ("media: b2c2-flexcop-usb: add sanity checking") added an endpoint sanity check to address a NULL-pointer dereference on probe. Unfortunately the check was done on the current altsetting which was later changed. Fix this by moving the sanity check to after the altsetting is changed. Fixes: 1b976fc6 ("media: b2c2-flexcop-usb: add sanity checking") Cc: Oliver Neukum <oneukum@suse.com> Cc: stable <stable@vger.kernel.org> Signed-off-by:
-
Niklas Söderlund authored
The hardware does not support capturing the field types V4L2_FIELD_SEQ_TB and V4L2_FIELD_SEQ_BT. To capture in these formats the driver needs to adjust the offset of the capture buffer and capture twice to each vb2 buffer. Signed-off-by:
Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se> Reviewed-by:
Jacopo Mondi <jacopo@jmondi.org> Signed-off-by:
-
Niklas Söderlund authored
To support SEQ_TB/BT not all buffers given to the hardware will be equal, the driver needs to keep track of different buffer types. Move the tracking of buffers given to hardware into a struct so additional tracking fields can be associated with each buffer. [hverkuil: fix small checkpatch warning] Signed-off-by:
-
Peter Ujfalusi authored
dma_request_slave_channel() is a wrapper on top of dma_request_chan() eating up the error code. By using dma_request_chan() directly the driver can support deferred probing against DMA. Signed-off-by:
Peter Ujfalusi <peter.ujfalusi@ti.com> Reviewed-by:
-
Dafna Hirschfeld authored
In case kthread_run fails, the vimc subdevices should be notified that streaming stopped so they can release the memory for the streaming. Also, kthread should be set to NULL. Signed-off-by:
-
Niklas Söderlund authored
Add a init_cfg pad operation so that configurations allocated with v4l2_subdev_alloc_pad_config() are initialized. Signed-off-by:
-
Hans Verkuil authored
An attempt to get a BUTTON control returns EACCES, not EINVAL. BUTTON controls have V4L2_CTRL_FLAG_WRITE_ONLY set, and the documentation for that flag correctly says that getting a write-only control returns EACCES. It's the description of the BUTTON type that's wrong, so fix that so they are consistent. Signed-off-by:
-
Johan Hovold authored
The driver was issuing synchronous uninterruptible control requests without using a timeout. This could lead to the driver hanging on various user requests due to a malfunctioning (or malicious) device until the device is physically disconnected. The USB upper limit of five seconds per request should be more than enough. Fixes: f3d27f34 ("[media] usbtv: Add driver for Fushicai USBTV007 video frame grabber") Fixes: c53a846c ("[media] usbtv: add video controls") Cc: stable <stable@vger.kernel.org> # 3.11 Signed-off-by:
Lubomir Rintel <lkundrak@v3.sk> Reviewed-by:
-
Eugen Hristev authored
Add new file include/linux/atmel-isc-media.h to ATMEL ISC entry. Signed-off-by:
Eugen Hristev <eugen.hristev@microchip.com> Signed-off-by:
-
Eugen Hristev authored
This exposes the white balance configuration of the ISC as v4l2 controls into userspace. There are 8 controls available: 4 gain controls, sliders, for each of the BAYER components: R, B, GR, GB. These gains are multipliers for each component, in format unsigned 0:4:9 with a default value of 512 (1.0 multiplier). 4 offset controls, sliders, for each of the BAYER components: R, B, GR, GB. These offsets are added/substracted from each component, in format signed 1:12:0 with a default value of 0 (+/- 0) To expose this to userspace, added 8 custom controls, in an auto cluster. To summarize the functionality: The auto cluster switch is the auto white balance control, and it works like this: AWB == 1: autowhitebalance is on, the do_white_balance button is inactive, the gains/offsets are inactive, but volatile and readable. Thus, the results of the whitebalance algorithm are available to userspace to read at any time. AWB == 0: autowhitebalance is off, cluster is in manual mode, user can configure the gain/offsets directly. More than that, if the do_white_balance button is pressed, the driver will perform one-time-adjustment, (preferably with color checker card) and the userspace can read again the new values. With this feature, the userspace can save the coefficients and reinstall them for example after reboot or reprobing the driver. [hverkuil: fix checkpatch warning] [hverkuil: minor spacing adjustments in the functionality description] Signed-off-by:
-
Joel Stanley authored
Use the preferred form for GPL2+, and add IBM and Intel as the authors. The driver was initially upstreamed by Eddie from IBM, and has had major contributions from Jae at Intel. Signed-off-by:
Joel Stanley <joel@jms.id.au> Acked-by:
Jae Hyun Yoo <jae.hyun.yoo@linux.intel.com> Reviewed-by:
Eddie James <eajames@linux.ibm.com> Signed-off-by:
-
Joel Stanley authored
The aspeed video IP has some differences between SoC families. Currently the driver decides which registers to use at compile time, which means a single kernel can not be used between platforms. Switch to using runtime configuration of the registers that vary between SoC families. This is in preparation for upcoming ast2600 support. [hverkuil: replace ;; by ;] Signed-off-by:
-
Joel Stanley authored
Use the recently introduced function devm_platform_ioremap_resource to save a few lines of code. This makes the driver match common platform device probe patterns. Signed-off-by:
-
Jae Hyun Yoo authored
The AST2600 has Video Engine so add the compatible string into the document. Signed-off-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Peter Ujfalusi authored
dma_request_slave_channel() is a wrapper on top of dma_request_chan() eating up the error code. By using dma_request_chan() directly the driver can support deferred probing against DMA. Signed-off-by:
-
