kernel/linux.git/drivers/media/usb, branch v7.2-rc1

Merge tag 'media/v7.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media

2026-06-19T00:21:54+00:00

Pull media updates from Mauro Carvalho Chehab: - v4l2: - core: fix subdev sensor ownership - subdev: Allow accessing routes with STREAMS client capability - ctrls: Add validation for HEVC active reference counts and background detection control - common: Add YUV24 format info and has_alpha helper - vb2: Change vb2_read() and vb2_write() return types to ssize_t - i2c: cvs: Add driver of Intel Computer Vision Sensing Controller(CVS) - atmel-isc: remove deprecated driver - cec: Add CEC Latency Indication Protocol (LIP) support - imon: Add iMON VFD HID OEM v1.2 key mappings - AVMatrix: new HWS capture driver - isp4: new AMD capture driver - qcom: - iris: Add hierarchical coding, B-frame, and Long-Term Reference support for encoder - camss: Add SM6350 platform support - venus: Add SM6115 platform support - chips-media: wave5: Add support for Packed YUV422, CBP profile, and background detection - csi2rx: Add multistream support and 32 dma chans - Several cleanups and fixes * tag 'media/v7.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media: (394 commits) media: v4l2-fwnode: Fix subdev owner overwritten in v4l2_async_register_subdev_sensor() media: qcom: iris: vdec: allow GEN2 decoding into 10bit format media: qcom: iris: vdec: update find_format to handle 8bit and 10bit formats media: qcom: iris: vdec: update size and stride calculations for 10bit formats media: qcom: iris: gen2: add support for 10bit decoding media: qcom: iris: add QC10C & P010 buffer size calculations media: qcom: iris: add helpers for 8bit and 10bit formats media: qcom: iris: Fix FPS calculation and VPP FW overhead media: qcom: camss: vfe-340: Support for PIX client media: qcom: camss: vfe-340: Proper client handling media: qcom: camss: csid-340: Enable PIX interface routing media: qcom: camss: csid-340: Add port-to-interface mapping media: qcom: camss: csid-340: Switch to generic CSID_CFG/CTRL registers media: iris: Initialize HFI ops after firmware load in core init media: iris: drop struct iris_fmt media: iris: Add platform data for X1P42100 media: iris: Add hardware power on/off ops for X1P42100 media: iris: optimize COMV buffer allocation for VPU3x and VPU4x media: iris: add FPS calculation and VPP FW overhead in frequency formula media: qcom: iris: Simplify COMV size calculation ...

media: uvcvideo: Only do uvc_video_get_time() if needed

2026-05-21T19:14:08+00:00

There is no need to calculate the current time if the sample is going to be filtered. Move the assignment close to uvc_video_clock_add_sample(). Suggested-by: Hans de Goede Signed-off-by: Ricardo Ribalda Tested-by: Yunke Cao Reviewed-by: Hans de Goede Link: https://patch.msgid.link/20260513-uvc-hwtimestamp-v3-6-7a64838b0b02@chromium.org Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil

media: uvcvideo: Do not add samples if dev_sof has not changed

2026-05-21T19:14:08+00:00

We only save relevant samples into the circular buffer. If the data is very similar to the previous one, exit early, this allows us to avoid some expensive operations such as usb_get_current_frame_number(). Suggested-by: Laurent Pinchart Signed-off-by: Ricardo Ribalda Tested-by: Yunke Cao Reviewed-by: Hans de Goede Link: https://patch.msgid.link/20260513-uvc-hwtimestamp-v3-5-7a64838b0b02@chromium.org Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil

media: uvcvideo: Do not add clock samples with small sof delta

2026-05-21T19:14:08+00:00

Some UVC 1.1 cameras running in fast isochronous mode tend to spam the USB host with a lot of empty packets. These packets contain clock information and are added to the clock buffer but do not add any accuracy to the calculation. In fact, it is quite the opposite, in our calculations, only the first and the last timestamp is used, and we only have 32 slots. Ignore the samples that will produce less than MIN_HW_TIMESTAMP_DIFF data. Fixes: 141270bd95d4 ("media: uvcvideo: Refactor clock circular buffer") Cc: stable@vger.kernel.org Tested-by: Yunke Cao Reviewed-by: Hans de Goede Signed-off-by: Ricardo Ribalda Link: https://patch.msgid.link/20260513-uvc-hwtimestamp-v3-4-7a64838b0b02@chromium.org Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil

media: uvcvideo: Relax the constrains for interpolating the hw clock

2026-05-21T19:14:07+00:00

In the initial version we set the min value to 250msec. Looks like 100msec can also provide a good value. Now that we are at it, add a macro to make it cleaner. Fixes: 6243c83be6ee8 ("media: uvcvideo: Allow hw clock updates with buffers not full") Cc: stable@vger.kernel.org Reviewed-by: Hans de Goede Tested-by: Yunke Cao Signed-off-by: Ricardo Ribalda Link: https://patch.msgid.link/20260513-uvc-hwtimestamp-v3-3-7a64838b0b02@chromium.org Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil

media: uvcvideo: Use hw timestaming if the clock buffer is full

2026-05-21T19:14:07+00:00

In some situations, even with a full clock buffer, it does not contain 250msec of data. This results in the driver jumping back from software to hardware timestapsing creating a nasty artifact in the video. If the clock buffer is full, use it to calculate the timestamp instead of defaulting to software stamps, the reduced accuracy is less visible than jumping from one timestamping mechanism to the other. Fixes: 6243c83be6ee8 ("media: uvcvideo: Allow hw clock updates with buffers not full") Cc: stable@vger.kernel.org Reviewed-by: Hans de Goede Tested-by: Yunke Cao Signed-off-by: Ricardo Ribalda Link: https://patch.msgid.link/20260513-uvc-hwtimestamp-v3-2-7a64838b0b02@chromium.org Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil

media: uvcvideo: Fix dev_sof filtering in hw timestamp

2026-05-21T19:14:07+00:00

To avoid filling the clock circular buffer with duplicated data we only add it if the new value sof is different than the last added sof. The issue is that we compare the unprocess sof with the processed sof. If there is a sof_offset, or UVC_QUIRK_INVALID_DEVICE_SOF is enabled, the comparison will not work as expected. This patch moves the comparison to the right place. Fixes: 141270bd95d4 ("media: uvcvideo: Refactor clock circular buffer") Cc: stable@vger.kernel.org Reviewed-by: Hans de Goede Tested-by: Yunke Cao Signed-off-by: Ricardo Ribalda Link: https://patch.msgid.link/20260513-uvc-hwtimestamp-v3-1-7a64838b0b02@chromium.org Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil

media: uvcvideo: Avoid partial metadata buffers

2026-05-21T19:14:07+00:00

If the metadata queue that is empty receives a new buffer while we are in the middle of processing a frame, the first metadata buffer will contain partial information. Avoid this by tracking the state of the metadata buffer and making sure that it is in sync with the data buffer. Now that we are at it, make sure that we skip buffers of size 1 or 0. They are not allowed by the spec... but it is a simple check to add and better be safe than sorry. Fixes: 088ead255245 ("media: uvcvideo: Add a metadata device node") Cc: stable@vger.kernel.org Signed-off-by: Ricardo Ribalda Link: https://patch.msgid.link/20260417-uvc-meta-partial-v2-2-31d274af7d2d@chromium.org Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil

media: uvcvideo: Do not open code uvc_queue_get_current_buffer

2026-05-21T19:14:07+00:00

Do not re-implement uvc_queue_get_current_buffer() logic inside uvc_video_complete(), just call the function. Reviewed-by: Laurent Pinchart Signed-off-by: Ricardo Ribalda Link: https://patch.msgid.link/20260417-uvc-meta-partial-v2-1-31d274af7d2d@chromium.org Reviewed-by: Hans de Goede Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil

media: uvcvideo: Fix deadlock if uvc_status_stop is called from async_ctrl.work

2026-05-21T19:14:07+00:00

If a UVC camera has an asynchronous control, uvc_status_stop may be called from async_ctrl.work: uvc_ctrl_status_event_work() uvc_ctrl_status_event() uvc_ctrl_clear_handle() uvc_pm_put() uvc_status_put() uvc_status_stop() cancel_work_sync() This will cause a deadlock, since cancel_work_sync will wait for uvc_ctrl_status_event_work to complete before returning. Fix this by returning early from uvc_status_stop if we are currently in the work function. flush_status now remains false until uvc_status_start is called again, ensuring that uvc_ctrl_status_event_work won't resubmit the URB. Fixes: a32d9c41bdb8 ("media: uvcvideo: Make power management granular") Cc: stable@vger.kernel.org Closes: https://lore.kernel.org/all/6733bdfb-3e88-479f-8956-ab09c04c433e@linux.dev/ Signed-off-by: Sean Anderson Link: https://patch.msgid.link/20260316155823.1855434-1-sean.anderson@linux.dev Reviewed-by: Ricardo Ribalda Tested-by: Ricardo Ribalda Reviewed-by: Laurent Pinchart Signed-off-by: Hans de Goede Signed-off-by: Hans Verkuil