- 08 May, 2022 8 commits
-
-
James_Lin authored
This patch aims to add UVC_GUID_FORMAT_H265 High Efficiency Video Coding (HEVC), also known as H.265 and MPEG-H Part 2. They describe the same video encoding method. So for handling their behavior is the same. However, when external camera device describes this encoding method, some use hevc, some use h265. There is no uniform specification to describe this encoding method. So if an external camera device use h265 to describe this encoding method, driver will not recognize it. Therefore, this patch is to enable driver to read HEVC/H265 and convert it to V4L2_PIX_FMT_HEVC. Signed-off-by:
James_Lin <Ping-lei.Lin@mediatek.com> Reviewed-by:
Ricardo Ribalda <ribalda@chromium.org> Signed-off-by:
Laurent Pinchart <laurent.pinchart@ideasonboard.com> Signed-off-by:
Mauro Carvalho Chehab <mchehab@kernel.org>
-
Ricardo Ribalda authored
Replace manual decoding of psize in uvc_parse_streaming(), with the code from uvc_endpoint_max_bpi(). It also handles usb3 devices. Signed-off-by:
-
Ricardo Ribalda authored
The case USB_SPEED_WIRELESS and the default one were doing the same. Also, make always use of usb_endpoint_maxp_mult, as it should have a sane value, even for LOW speed and WIRELESS. Signed-off-by:
-
Ricardo Ribalda authored
Move all the life cycle of the name to add_mapping. This simplifies the error handling inside uvc_ioctl_ctrl_map and solves a memory leak when kemmdup fails. Also make sure that for custom controls, the user provides a valid name. Fixes: 07adedb5c606 ("media: uvcvideo: Use control names from framework") Signed-off-by: -
Hangyu Hua authored
probe->dwMaxPayloadTransferSize is a 32bit value, but bandwidth is 16bit. This may lead to a bit overflow. Signed-off-by:
Hangyu Hua <hbh25y@gmail.com> Signed-off-by:
-
Xiaomeng Tong authored
The list iterator will point to a bogus position containing HEAD if the list is empty or the element is not found in list. This case should be checked before any use of the iterator, otherwise it will lead to a invalid memory access. The missing check here is before "pin = iterm->id;", just add check here to fix the security bug. In addition, the list iterator value will *always* be set and non-NULL by list_for_each_entry(), so it is incorrect to assume that the iterator value will be NULL if the element is not found in list, considering the (mis)use here: "if (iterm == NULL". Use a new value 'it' as the list iterator, while use the old value 'iterm' as a dedicated pointer to point to the found element, which 1. can fix this bug, due to 'iterm' is NULL only if it's not found. 2. do not need to change all the uses of 'iterm' after the loop. 3. can also limit the scope of the list iterator 'it' *only inside* the traversal loop by simply declaring 'it' inside the loop in the future, as usage of the iterator outside of the list_for_each_entry is considered harmful. https://lkml.org/lkml/2022/2/17/1032 Fixes: d5e90b7a ("[media] uvcvideo: Move to video_ioctl2") Signed-off-by:
Xiaomeng Tong <xiam0nd.tong@gmail.com> Signed-off-by:
-
Jonathan Neuschäfer authored
As discussed with other developers, the linux-uvc-devel mailing list is not very useful anymore, and it's better to send people to the general linux-media mailing list. Replace/remove the old mailing list address in uvcvideo.rst and MAINTAINERS. Signed-off-by:
Jonathan Neuschäfer <j.neuschaefer@gmx.net> Signed-off-by:
-
Sean Young authored
Do not handroll mdelay(). Suggested-by:
Geert Uytterhoeven <geert@linux-m68k.org> Signed-off-by:
Sean Young <sean@mess.org> Signed-off-by:
-
- 25 Apr, 2022 1 commit
-
-
Hans Verkuil authored
The indentation was wrong, causing the documentation build to fail with: Sphinx parallel build error: docutils.utils.SystemMessagePropagation: <system_message level="3" line="1202" source="/home/hans/work/src/v4l/media-git/Documentation/media/userspace-api/v4l/ext-ctrls-codec.rst" type="ERROR"><paragraph>Error parsing content block for the "flat-table" directive: two-level bullet list expected, but row 1 does not contain a second-level bullet list.</paragraph><literal_block xml:space="preserve">.. flat-table:: :header-rows: 0 :stub-columns: 0 * - ``V4L2_MPEG_VIDEO_INTRA_REFRESH_PERIOD_TYPE_RANDOM`` - The whole frame is completely refreshed randomly after the specified period. * - ``V4L2_MPEG_VIDEO_INTRA_REFRESH_PERIOD_TYPE_CYCLIC`` - The whole frame MBs are completely refreshed in cyclic order after the specified period. </literal_block></system_message> make[1]: *** [Documentation/Makefile:96: htmldocs] Error 2 make: *** [Makefile:1678: htmldocs] Error 2 Link: https://lore.kernel.org/linux-media/62b53cbb-f3e2-ed8a-bba2-3f4145d9b8db@xs4all.nl Fixes: fcbc4acf ("media: v4l2-ctrls: Add intra-refresh type control") Signed-off-by:Hans Verkuil <hverkuil-cisco@xs4all.nl> Signed-off-by:
-
- 24 Apr, 2022 31 commits
-
-
Fabio Estevam authored
There is no need to pass 'default n' in Kconfig, as by default the CONFIG_VIDEO_IMX_MIPI_CSIS option is not selected. Signed-off-by:
Fabio Estevam <festevam@denx.de> Signed-off-by:
-
Laurent Pinchart authored
Commit 5c0701a0 ("media: imx: csis: Store pads format separately") broke initialization of the active format on the source pad, as it forgot to update the .init_cfg() handler. Fix it. Fixes: 5c0701a0 ("media: imx: csis: Store pads format separately") Signed-off-by:
Rui Miguel Silva <rmfrfs@gmail.com> Signed-off-by:
-
Jacopo Mondi authored
The mipi_csis_dump_regs() function accesses the interface registers in order to printout their values for debug purposes. As the function access the registers, it requires the interface to be powered up. Currently this is only enforced in one of the function's callers (mipi_csis_log_status)() but not when the function is called by the debugfs attribute handler. Make sure to access registers only if the interface is powered up and remove the same check from the caller. Signed-off-by:
Jacopo Mondi <jacopo@jmondi.org> Acked-by:
-
Jacopo Mondi authored
The mipi_csis_device.powered flag only serves for the purpose of not accessing registers in mipi_csis_log_status() when the interface is not powered up. Instead of manually tracking the power state, rely on pm_runtime_get_if_in_use() to remove the 'powered' flag. Also remove the locking in the function as runtime_pm() is refcounted and there's no risk of the interface being powered down behind our backs. Signed-off-by:
-
Jacopo Mondi authored
Simplify the mipi_csis_s_stream() function. This actually fixes a bug, as if calling the subdev's s_stream(1) fails, mipi_csis_stop_stream() was not called. Signed-off-by:
-
Laurent Pinchart authored
The runtime PM resume handler is guaranteed to be called on a suspended device, and the suspend handler on a resumed device. The implementation can thus be simplified. While at it, rename the mipi_csis_device state field to powered, as the now state contains a single flag only. Signed-off-by:
-
Laurent Pinchart authored
Streaming is guaranteed to have been stopped by the time the device gets runtime suspended, as pm_runtime_put() is called from .s_stream(0) only. Drop the manual stop. Signed-off-by:
-
Laurent Pinchart authored
There's no need to implement system suspend/resume manually, as video pipelines are supposed to be suspended in a controlled and ordered manner by the data sink driver at system suspend time (and similarly at resume time). Drop the system suspend/resume handlers. Signed-off-by:
-
Laurent Pinchart authored
The subdev .s_power() operation is deprecated. Drop it, requiring sensor drivers to correctly use runtime PM instead of relying on .s_power(). As this driver has just been moved out of staging, and necessary drivers to implement a full camera pipeline are still in staging, no platform depends yet on this API being called. There is thus no risk of regression. Signed-off-by:
-
Laurent Pinchart authored
Usage of "state" for the device-specific data structure is confusing, as it can also refer to the subdev state. Rename the structure to mipi_csis_device, and the corresponding state variables to csis. Signed-off-by:
-
Chen-Yu Tsai authored
The payload size for encoder capture buffers is set by the driver upon finishing encoding each frame, based on the encoded length returned from hardware, and whatever header and padding length used. Setting a non-zero default serves no real purpose, and also causes issues if the capture buffer is returned to userspace unused, confusing the application. Instead, always set the payload size to 0 for encoder capture buffers when preparing them. Fixes: 775fec69 ("media: add Rockchip VPU JPEG encoder driver") Fixes: 082aaecf ("media: hantro: Fix .buf_prepare") Signed-off-by:
Chen-Yu Tsai <wenst@chromium.org> Reviewed-by:
Ezequiel Garcia <ezequiel@vanguardiasur.com.ar> Signed-off-by:
-
Chen-Yu Tsai authored
The V4L2 stateful encoder uAPI specification requires that drivers support the ENCODER_CMD ioctl to allow draining of buffers. This however was not implemented, and causes issues for some userspace applications. Implement support for the ENCODER_CMD ioctl using v4l2-mem2mem helpers. This is entirely based on existing code found in the vicodec test driver. Fixes: 775fec69 ("media: add Rockchip VPU JPEG encoder driver") Signed-off-by:
-
Piotr Oniszczuk authored
RK356x has Hantro G1 video decoder capable to decode MPEG2/H.264/VP8 video formats. This patch adds RK3568 compatible in rockchip-vpu dt-bindings. Tested on [1] with FFmpeg v4l2_request code taken from [2] with MPEG2, H.642 and VP8 samples with results [3]. [1] https://github.com/warpme/minimyth2 [2] https://github.com/LibreELEC/LibreELEC.tv/blob/master/packages/multimedia/ffmpeg/patches/v4l2-request/ffmpeg-001-v4l2-request.patch [3] https://github.com/warpme/minimyth2/blob/master/video-test-summary.txtSigned-off-by:
Piotr Oniszczuk <piotr.oniszczuk@gmail.com> Reviewed-by:
-
Piotr Oniszczuk authored
RK356x has Hantro G1 video decoder capable to decode MPEG2/H.264/VP8 video formats. This patch adds support for RK356x family in existing Hantro video decoder kernel driver. Tested on [1] with FFmpeg v4l2_request code taken from [2] with MPEG2, H.642 and VP8 samples with results [3]. [1] https://github.com/warpme/minimyth2 [2] https://github.com/LibreELEC/LibreELEC.tv/blob/master/packages/multimedia/ffmpeg/patches/v4l2-request/ffmpeg-001-v4l2-request.patch [3] https://github.com/warpme/minimyth2/blob/master/video-test-summary.txtSigned-off-by:
-
Ming Qian authored
For interlaced frame, the amphion vpu will store the two fields sequential into one buffer, top-bottom order so the field should be set to V4L2_FIELD_SEQ_TB. fix the previous bug that set it to V4L2_FIELD_SEQ_BT wrongly. Fixes: 6de8d628 ("media: amphion: add v4l2 m2m vpu decoder stateful driver") Signed-off-by:
Ming Qian <ming.qian@nxp.com> Reviewed-by:
Nicolas Dufresne <nicolas.dufresne@collabora.com> Signed-off-by:
-
Wan Jiabing authored
use if and else instead of if(A) and if (!A). Signed-off-by:
Wan Jiabing <wanjiabing@vivo.com> Signed-off-by:
-
Ming Qian authored
Don't change the output buffer's bytesused set by user. Drop allow_zero_bytesused, since it's deprecated. This should also guarantee it's the application's responsibility to set bytesused for the output buffer. Signed-off-by:
Mirela Rabulea <mirela.rabulea@nxp.com> Signed-off-by:
-
Zheyu Ma authored
When removing the module, we will get the following warning: [ 14.746697] remove_proc_entry: removing non-empty directory 'irq/21', leaking at least 'cx25821[1]' [ 14.747449] WARNING: CPU: 4 PID: 368 at fs/proc/generic.c:717 remove_proc_entry+0x389/0x3f0 [ 14.751611] RIP: 0010:remove_proc_entry+0x389/0x3f0 [ 14.759589] Call Trace: [ 14.759792] <TASK> [ 14.759975] unregister_irq_proc+0x14c/0x170 [ 14.760340] irq_free_descs+0x94/0xe0 [ 14.760640] mp_unmap_irq+0xb6/0x100 [ 14.760937] acpi_unregister_gsi_ioapic+0x27/0x40 [ 14.761334] acpi_pci_irq_disable+0x1d3/0x320 [ 14.761688] pci_disable_device+0x1ad/0x380 [ 14.762027] ? _raw_spin_unlock_irqrestore+0x2d/0x60 [ 14.762442] ? cx25821_shutdown+0x20/0x9f0 [cx25821] [ 14.762848] cx25821_finidev+0x48/0xc0 [cx25821] [ 14.763242] pci_device_remove+0x92/0x240 Fix this by freeing the irq before call pci_disable_device(). Signed-off-by:
Zheyu Ma <zheyuma97@gmail.com> Signed-off-by:
-
Zheyu Ma authored
When the driver fails to call the dma_set_mask(), the driver will get the following splat: [ 55.853884] BUG: KASAN: use-after-free in __process_removed_driver+0x3c/0x240 [ 55.854486] Read of size 8 at addr ffff88810de60408 by task modprobe/590 [ 55.856822] Call Trace: [ 55.860327] __process_removed_driver+0x3c/0x240 [ 55.861347] bus_for_each_dev+0x102/0x160 [ 55.861681] i2c_del_driver+0x2f/0x50 This is because the driver has initialized the i2c related resources in cx23885_dev_setup() but not released them in error handling, fix this bug by modifying the error path that jumps after failing to call the dma_set_mask(). Signed-off-by:
-
Yan Lei authored
Declaring variable "tuner_status" without initializer. Using uninitialized value "tuner_status" when calling "*fe_tuner_ops->get_status". (The function pointer resolves to "cx24113_get_status".) Signed-off-by:
Yan Lei <yan_lei@dahuatech.com> Signed-off-by:
-
Kwang Son authored
RGB/YUV Input is sensor type and it should be sub-dev node. To generate this dot graph sudo modprobe vimc media-ctl --print-dot Signed-off-by:
Kwang Son <dev.kwang.son@gmail.com> Signed-off-by:
-
Nicolas Dufresne authored
Add H264 level 1.0, 4.1, 4.2 to the list of supported formats. While the hardware does not fully support these levels, it does support most of them. The constraints on frame size and pixel formats already cover the limitation. This fixes negotiation of level on GStreamer 1.17.1. Cc: stable@vger.kernel.org Fixes: 42a68012 ("media: coda: add read-only h.264 decoder profile/level controls") Suggested-by:
Philipp Zabel <p.zabel@pengutronix.de> Signed-off-by:
Ezequiel Garcia <ezequiel@collabora.com> Signed-off-by:
-
Nicolas Dufresne authored
The CODA960 manual states that ASO/FMO features of baseline are not supported, so for this reason this driver should only report constrained baseline support. This fixes negotiation issue with constrained baseline content on GStreamer 1.17.1. ASO/FMO features are unsupported for the encoder and untested for the decoder because there is currently no userspace support. Neither GStreamer parsers nor FFMPEG parsers support ASO/FMO. Cc: stable@vger.kernel.org Fixes: 42a68012 ("media: coda: add read-only h.264 decoder profile/level controls") Signed-off-by:
Pascal Speck <kernel@iktek.de> Signed-off-by:
-
Ming Qian authored
The hardware is capable of encoding/decoding RGB and ARGB formats in whatever order the color components are, but the resulting jpegs look good if we start with raw data in BGR/ABGR order, so we will further only support V4L2_PIX_FMT_BGR24 and V4L2_PIX_FMT_ABGR32. Signed-off-by:
-
Ming Qian authored
Wait finishing jpeg job before system sleep, otherwise the encoding/decoding can't be resumed after suspend. Signed-off-by:
-
Philipp Zabel authored
Stop bailing out on JPEG header parsing errors during streamon. This allows userspace to provide valid output buffers later and fixes a v4l2-compliance streaming test failure: fail: v4l2-test-buffers.cpp(1429): node->streamon(q.g_type()) test MMAP (no poll): FAIL Signed-off-by:
-
Philipp Zabel authored
If JPEG header parsing fails, output a single message instead of two messages that say the same thing. Signed-off-by:
-
Philipp Zabel authored
If decoding fails because the output buffer does not contain a valid header, set the error flag on the returned capture buffer. Signed-off-by:
-
Philipp Zabel authored
The JPEG decoder in the CODA960 VPU can downscale images while decoding, with a factor of 1/2, 1/4, or 1/8. Signed-off-by:
-
Colin Ian King authored
The variable m is being assigned a value that is never read, it is being re-assigned in both paths of the following if-statement. The assignment is redundant and can be removed. Signed-off-by:
Colin Ian King <colin.i.king@gmail.com> Signed-off-by:
-
Philipp Zabel authored
If the decoder prepare_run callback returns an error, the hardware is never started and v4l2_m2m_job_finish() is called immediately, currently in every error path. Remove the duplicated job_finish calls from the error paths and reuse the v4l2_m2m_job_finish() at the end of coda_pic_run_work instead. Signed-off-by:
-
