torvalds/linux
3
0
Fork 0
mirror of git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git synced 2026-04-07 06:59:25 +08:00
Code Issues Packages Projects Releases Wiki Activity
1,369,264 Commits 2 Branches 926 Tags
f9db1fc56281b96fe8748632b3894de970a8a850
Commit Graph

431 Commits

Author SHA1 Message Date
Ricardo Ribalda
3328eb4dfe media: uvcvideo: Use dev_err_probe for devm_gpiod_get_optional 
Use the dev_err_probe() helper for devm_gpiod_get_optional(), like we do
with gpiod_to_irq()

That eventually calls device_set_deferred_probe_reason() which can be
helpful for tracking down problems.

Now that all the error paths in uvc_gpio_parse have dev_err_probe, we
can remove the error message in uvc_probe.

Suggested-by: Doug Anderson <dianders@chromium.org>
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250313-uvc-eprobedefer-v3-2-a1d312708eef@chromium.org>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:13 +02:00
Ricardo Ribalda
387e893930 media: uvcvideo: Fix deferred probing error 
uvc_gpio_parse() can return -EPROBE_DEFER when the GPIOs it depends on
have not yet been probed. This return code should be propagated to the
caller of uvc_probe() to ensure that probing is retried when the required
GPIOs become available.

Currently, this error code is incorrectly converted to -ENODEV,
causing some internal cameras to be ignored.

This commit fixes this issue by propagating the -EPROBE_DEFER error.

Cc: stable@vger.kernel.org
Fixes: 2886477ff9 ("media: uvcvideo: Implement UVC_EXT_GPIO_UNIT")
Reviewed-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250313-uvc-eprobedefer-v3-1-a1d312708eef@chromium.org>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
a70705d3c0 media: uvcvideo: Rollback non processed entities on error 
If we fail to commit an entity, we need to restore the
UVC_CTRL_DATA_BACKUP for the other uncommitted entities. Otherwise the
control cache and the device would be out of sync.

Cc: stable@kernel.org
Fixes: b4012002f3 ("[media] uvcvideo: Add support for control events")
Reported-by: Hans de Goede <hdegoede@redhat.com>
Closes: https://lore.kernel.org/linux-media/fe845e04-9fde-46ee-9763-a6f00867929a@redhat.com/
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250224-uvc-data-backup-v2-3-de993ed9823b@chromium.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
5c791467ae media: uvcvideo: Send control events for partial succeeds 
Today, when we are applying a change to entities A, B. If A succeeds and B
fails the events for A are not sent.

This change changes the code so the events for A are send right after
they happen.

Cc: stable@kernel.org
Fixes: b4012002f3 ("[media] uvcvideo: Add support for control events")
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250224-uvc-data-backup-v2-2-de993ed9823b@chromium.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
ba4fafb02a media: uvcvideo: Return the number of processed controls 
If we let know our callers that we have not done anything, they will be
able to optimize their decisions.

Cc: stable@kernel.org
Fixes: b4012002f3 ("[media] uvcvideo: Add support for control events")
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250224-uvc-data-backup-v2-1-de993ed9823b@chromium.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
d1b618e795 media: uvcvideo: Do not turn on the camera for some ioctls 
There are some ioctls that do not need to turn on the camera. Do not
call uvc_pm_get in those cases.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250327-uvc-granpower-ng-v6-5-35a2357ff348@chromium.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
a32d9c41bd media: uvcvideo: Make power management granular 
Now that every ioctl takes care of their power management we can remove
the "global" power management.

Despite its size, this is a relatively big change. We hope that there
are no size effects of it. If there are some specific devices that
miss-behave, we can add a small quirk for them.

This patch introduces a behavioral change for the uvc "trigger" button.
Before the "trigger" button would work as long as userspace has opened
/dev/videoX. Now it only works when the camera is actually streaming. We
consider that this the most common (if not the only) usecase and
therefore we do not think of this as a regression.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250327-uvc-granpower-ng-v6-4-35a2357ff348@chromium.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
10acb91013 media: uvcvideo: Increase/decrease the PM counter per IOCTL 
Now we call uvc_pm_get/put from the device open/close. This low
level of granularity might leave the camera powered on in situations
where it is not needed.

Increase the granularity by increasing and decreasing the Power
Management counter per ioctl. There are two special cases where the
power management outlives the ioctl: async controls and streamon. Handle
those cases as well.

In a future patch, we will remove the uvc_pm_get/put from open/close.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250327-uvc-granpower-ng-v6-3-35a2357ff348@chromium.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
2f101572c0 media: uvcvideo: Create uvc_pm_(get|put) functions 
Most of the times that we have to call uvc_status_(get|put) we need to
call the usb_autopm_ functions.

Create a new pair of functions that automate this for us. This
simplifies the current code and future PM changes in the driver.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250327-uvc-granpower-ng-v6-2-35a2357ff348@chromium.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
14f6e205e5 media: uvcvideo: Keep streaming state in the file handle 
Add a variable in the file handle state to figure out if a camera is in
the streaming state or not. This variable will be used in the future for
power management policies.

Now that we are at it, make use of guards to simplify the code.

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Message-ID: <20250327-uvc-granpower-ng-v6-1-35a2357ff348@chromium.org>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-05-09 15:29:12 +02:00
Ricardo Ribalda
b810697e1c media: uvcvideo: Remove vidioc_queryctrl 
It can be implemented by the v4l2 ioctl framework using
vidioc_query_ext_ctrl.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
[hverkuil: rebased]
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2025-03-05 10:43:25 +01:00
Laurent Pinchart
fd4f68174d media: uvcvideo: Drop the uvc_driver structure 
The uvc_driver structure used to contain more fields, but those got
removed in commit ba2fa99668 ("[media] uvcvideo: Hardcode the
index/selector relationship for XU controls"). The structure is now just
a wrapper around usb_driver. Drop it.

Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:36 +01:00
Ricardo Ribalda
8c54e58f94 media: uvcvideo: Add quirk for Actions UVC05 
Actions UVC05 is a HDMI to USB dongle that implements the UVC protocol.

When the device suspends, its firmware seems to enter a weird mode when it
does not produce more frames.

Add the device to the quirk list to disable autosuspend.

Bus 001 Device 007: ID 1de1:f105 Actions Microelectronics Co. Display
 capture-UVC05
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               2.00
  bDeviceClass          239 Miscellaneous Device
  bDeviceSubClass         2 [unknown]
  bDeviceProtocol         1 Interface Association
  bMaxPacketSize0        64
  idVendor           0x1de1 Actions Microelectronics Co.
  idProduct          0xf105 Display capture-UVC05
  bcdDevice            4.09
  iManufacturer           1 Actions Micro
  iProduct                2 Display capture-UVC05
  iSerial                 3 -1005308387
  bNumConfigurations      1

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241210-uvc-hdmi-suspend-v1-1-01f5dec023ea@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:36 +01:00
Yunke Cao
2dc768d71b media: uvcvideo: implement UVC v1.5 ROI 
Implement support for ROI as described in UVC 1.5:
4.2.2.1.20 Digital Region of Interest (ROI) Control

ROI control is implemented using V4L2 control API as
two UVC-specific controls:
V4L2_CID_UVC_REGION_OF_INTEREST_RECT and
V4L2_CID_UVC_REGION_OF_INTEREST_AUTO.

Reviewed-by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Yunke Cao <yunkec@google.com>
Tested-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-16-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
[hverkuil: fix control names: "Of" -> "of", "Controls" -> "Ctrls"]
 2025-03-03 18:23:36 +01:00
Ricardo Ribalda
990262fdfc media: uvcvideo: Add sanity check to uvc_ioctl_xu_ctrl_map 
Do not process unknown data types.

Tested-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-15-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:36 +01:00
Ricardo Ribalda
dc6d7e050c media: uvcvideo: Introduce uvc_mapping_v4l2_size 
Centralize the calculation for the v4l2_size of a mapping.

Reviewed-by: Yunke Cao <yunkec@google.com>
Tested-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-14-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:36 +01:00
Ricardo Ribalda
1fde21ea0b media: uvcvideo: let v4l2_query_v4l2_ctrl() work with v4l2_query_ext_ctrl 
v4l2_query_ext_ctrl contains information that is missing in
v4l2_queryctrl, like elem_size and elems.

With this change we can handle all the element_size information inside
uvc_ctrl.c.

Now that we are at it, remove the memset of the reserved fields, the
v4l2 ioctl handler should do that for us.

There is no functional change expected from this change.

Reviewed-by: Yunke Cao <yunkec@google.com>
Tested-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-13-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:36 +01:00
Yunke Cao
39d2c891c9 media: uvcvideo: support V4L2_CTRL_WHICH_MIN/MAX_VAL 
Add support for V4L2_CTRL_WHICH_MIN/MAX_VAL in uvc driver.
It is needed for the V4L2_CID_UVC_REGION_OF_INTEREST_RECT control.

Signed-off-by: Yunke Cao <yunkec@google.com>
Tested-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-12-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:36 +01:00
Ricardo Ribalda
cc9a7cb3f7 media: uvcvideo: Factor out query_boundaries from query_ctrl 
Split the function in two parts. queryctrl_boundaries will be used in
future patches.

No functional change expected from this patch.

Reviewed-by: Yunke Cao <yunkec@google.com>
Tested-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-11-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:36 +01:00
Yunke Cao
8c3d9d4dc6 media: uvcvideo: add support for compound controls 
This patch adds support for compound controls. This is required to
support controls that cannot be represented with a s64 data, such as the
Region of Interest.

Signed-off-by: Yunke Cao <yunkec@google.com>
Tested-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-10-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:36 +01:00
Ricardo Ribalda
813b015678 media: uvcvideo: Factor out clamping from uvc_ctrl_set 
Move the logic to a separated function. Do not expect any change.
This is a preparation for supporting compound controls.

Reviewed-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Yunke Cao <yunkec@google.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-9-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:35 +01:00
Ricardo Ribalda
4e15c53565 media: uvcvideo: Support any size for mapping get/set 
Right now, we only support mappings for v4l2 controls with a max size of
s32. This patch modifies the prototype of get/set so it can support any
size.

This is done to prepare for compound controls.

Suggested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Yunke Cao <yunkec@google.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Yunke Cao <yunkec@google.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-8-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:35 +01:00
Ricardo Ribalda
66dcae2c3b media: uvcvideo: uvc_ioctl_(g|s)_ext_ctrls: handle NoP case 
If nothing needs to be done. Exit early.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Yunke Cao <yunkec@google.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-7-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:35 +01:00
Ricardo Ribalda
2002ce4443 media: uvcvideo: refactor uvc_ioctl_g_ext_ctrls 
We want to support fetching the min and max values with g_ext_ctrls,
this patch is a preparation for that.

Instead of abusing uvc_query_v4l2_ctrl(), add an extra parameter to
uvc_ctrl_get, so it can support fetching the default value.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Yunke Cao <yunkec@google.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-6-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:35 +01:00
Ricardo Ribalda
7f1556a5d8 media: uvcvideo: Handle uvc menu translation inside uvc_set_le_value 
Be consistent with uvc_get_le_value() and do the menu translation there.

Note that in this case, the refactor does not provide much... but
consistency is a nice feature.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Yunke Cao <yunkec@google.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-5-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:35 +01:00
Ricardo Ribalda
9109a0b4cb media: uvcvideo: Handle uvc menu translation inside uvc_get_le_value 
map->get() gets a value from an uvc_control in "UVC format" and converts
it to a value that can be consumed by v4l2.

Instead of using a special get function for V4L2_CTRL_TYPE_MENU, we
were converting from uvc_get_le_value in two different places.

Move the conversion to uvc_get_le_value().

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Yunke Cao <yunkec@google.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250203-uvc-roi-v17-4-5900a9fed613@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
 2025-03-03 18:23:35 +01:00
Thadeu Lima de Souza Cascardo
8004d635f2 Revert "media: uvcvideo: Require entities to have a non-zero unique ID" 
This reverts commit 3dd075fe8e.

Tomasz has reported that his device, Generalplus Technology Inc. 808 Camera,
with ID 1b3f:2002, stopped being detected:

$ ls -l /dev/video*
zsh: no matches found: /dev/video*
[    7.230599] usb 3-2: Found multiple Units with ID 5

This particular device is non-compliant, having both the Output Terminal
and Processing Unit with ID 5. uvc_scan_fallback, though, is able to build
a chain. However, when media elements are added and uvc_mc_create_links
call uvc_entity_by_id, it will get the incorrect entity,
media_create_pad_link will WARN, and it will fail to register the entities.

In order to reinstate support for such devices in a timely fashion,
reverting the fix for these warnings is appropriate. A proper fix that
considers the existence of such non-compliant devices will be submitted in
a later development cycle.

Reported-by: Tomasz Sikora <sikora.tomus@gmail.com>
Fixes: 3dd075fe8e ("media: uvcvideo: Require entities to have a non-zero unique ID")
Cc: stable@vger.kernel.org
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@igalia.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20250114200045.1401644-1-cascardo@igalia.com
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2025-01-31 22:07:43 +01:00
Isaac Scott
2762eab6d4 media: uvcvideo: Add Kurokesu C1 PRO camera 
Add support for the Kurokesu C1 PRO camera. This camera experiences the
same issues faced by the Sonix Technology Co. 292A IPC AR0330. As such,
enable the UVC_QUIRK_MJPEG_NO_EOF quirk for this device to prevent
frames from being erroneously dropped.

Signed-off-by: Isaac Scott <isaac.scott@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2025-01-02 09:58:30 +01:00
Isaac Scott
81f8c0e138 media: uvcvideo: Add new quirk definition for the Sonix Technology Co. 292a camera 
The Sonix Technology Co. 292A camera (which uses an AR0330 sensor), can
produce MJPEG and H.264 streams concurrently. When doing so, it drops
the last packets of MJPEG frames every time the H.264 stream generates a
key frame. Set the UVC_QUIRK_MJPEG_NO_EOF quirk to work around the
issue.

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Isaac Scott <isaac.scott@ideasonboard.com>
Link: https://lore.kernel.org/r/20241128145144.61475-3-isaac.scott@ideasonboard.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2025-01-02 09:58:30 +01:00
Isaac Scott
c2eda35e67 media: uvcvideo: Implement dual stream quirk to fix loss of usb packets 
Some cameras, such as the Sonix Technology Co. 292A, exhibit issues when
running two parallel streams, causing USB packets to be dropped when an
H.264 stream posts a keyframe while an MJPEG stream is running
simultaneously. This occasionally causes the driver to erroneously
output two consecutive JPEG images as a single frame.

To fix this, we inspect the buffer, and trigger a new frame when we
find an SOI.

Signed-off-by: Isaac Scott <isaac.scott@ideasonboard.com>
Reviewed-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20241128145144.61475-2-isaac.scott@ideasonboard.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2025-01-02 09:58:30 +01:00
Ricardo Ribalda
40ed9e9b28 media: uvcvideo: Announce the user our deprecation intentions 
If the user sets the nodrop parameter, print a deprecation warning once.
Hopefully they will come to the mailing list if it is an ABI change.

Now that we have a callback, take this chance to parse the parameter as
a boolean. We still say to userspace that it is a uint to avoid ABI
changes.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20241218-uvc-deprecate-v2-4-ab814139e983@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
8869eb654f media: uvcvideo: Allow changing noparam on the fly 
Right now the parameter value is read during video_registration and
cannot be changed afterwards, despite its permissions 0644, that makes
the user believe that the value can be written.

The parameter only affects the behaviour of uvc_queue_buffer_complete(),
with only one check per buffer.

We can read the value directly from uvc_queue_buffer_complete() and
therefore allowing changing it with sysfs on the fly.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20241218-uvc-deprecate-v2-3-ab814139e983@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
52fbe173ba media: uvcvideo: Invert default value for nodrop module param 
The module param `nodrop` defines what to do with frames that contain an
error: drop them or sending them to userspace.

The default in the rest of the media subsystem is to return buffers with
an error to userspace with V4L2_BUF_FLAG_ERROR set in v4l2_buffer.flags.
In UVC we drop buffers with errors by default.

Change the default behaviour of uvcvideo to match the rest of the
drivers and maybe get rid of the module parameter in the future.

Suggested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20241218-uvc-deprecate-v2-2-ab814139e983@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
87ce177654 media: uvcvideo: Propagate buf->error to userspace 
Now we return VB2_BUF_STATE_DONE for valid and invalid frames. Propagate
the correct value, so the user can know if the frame is valid or not via
struct v4l2_buffer->flags.

Reported-by: Hans de Goede <hdegoede@redhat.com>
Closes: https://lore.kernel.org/linux-media/84b0f212-cd88-46bb-8e6f-b94ec3eccba6@redhat.com
Fixes: 6998b6fb4b ("[media] uvcvideo: Use videobuf2-vmalloc")
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20241218-uvc-deprecate-v2-1-ab814139e983@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
d6b874ff9c media: uvcvideo: Flush the control cache when we get an event 
Asynchronous controls trigger an event when they have completed their
operation.

This can make that the control cached value does not match the value in
the device.

Let's flush the cache to be on the safe side.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241203-uvc-fix-async-v6-5-26c867231118@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
02baaa09d1 media: uvcvideo: Annotate lock requirements for uvc_ctrl_set 
Make it explicit that the function is always called with ctrl_mutex
being held.

Suggested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20241203-uvc-fix-async-v6-4-26c867231118@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
221cd51efe media: uvcvideo: Remove dangling pointers 
When an async control is written, we copy a pointer to the file handle
that started the operation. That pointer will be used when the device is
done. Which could be anytime in the future.

If the user closes that file descriptor, its structure will be freed,
and there will be one dangling pointer per pending async control, that
the driver will try to use.

Clean all the dangling pointers during release().

To avoid adding a performance penalty in the most common case (no async
operation), a counter has been introduced with some logic to make sure
that it is properly handled.

Cc: stable@vger.kernel.org
Fixes: e5225c820c ("media: uvcvideo: Send a control event when a Control Change interrupt arrives")
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241203-uvc-fix-async-v6-3-26c867231118@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
04d3398f66 media: uvcvideo: Remove redundant NULL assignment 
ctrl->handle will only be different than NULL for controls that have
mappings. This is because that assignment is only done inside
uvc_ctrl_set() for mapped controls.

Cc: stable@vger.kernel.org
Fixes: e5225c820c ("media: uvcvideo: Send a control event when a Control Change interrupt arrives")
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20241203-uvc-fix-async-v6-2-26c867231118@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
d9fecd096f media: uvcvideo: Only save async fh if success 
Now we keep a reference to the active fh for any call to uvc_ctrl_set,
regardless if it is an actual set or if it is a just a try or if the
device refused the operation.

We should only keep the file handle if the device actually accepted
applying the operation.

Cc: stable@vger.kernel.org
Fixes: e5225c820c ("media: uvcvideo: Send a control event when a Control Change interrupt arrives")
Suggested-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20241203-uvc-fix-async-v6-1-26c867231118@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Benoit Sevens
082dd785e2 media: uvcvideo: Refactor frame parsing code into a uvc_parse_frame function 
The ftype value does not change in the while loop so we can check it
before entering the while loop. Refactoring the frame parsing code into
a dedicated uvc_parse_frame function makes this more readable.

Signed-off-by: Benoit Sevens <bsevens@google.com>
Link: https://lore.kernel.org/r/20241107142204.1182969-3-bsevens@google.com
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
840fb2c339 media: uvcvideo: Remove duplicated cap/out code 
The *_vid_cap and *_vid_out helpers seem to be identical:
- Remove all the cap/out duplicated code.
- Remove s/g_parm helpers
- Reorder uvc_ioctl_ops

And now that we are at it, fix a comment for uvc_acquire_privileges()

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20241202-uvc-dup-cap-out-v3-1-d40b11bb74b7@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Xiong Nandi
7309f3e2b7 media: uvcvideo: Use uvc_query_name in uvc_get_video_ctrl 
uvc_query_name was introduced to print query name in uvc_query_ctrl.
So we can also use it in uvc_get_video_ctrl.

Signed-off-by: Xiong Nandi <xndchn@gmail.com>
Reviewed-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241129151723.48275-1-xndchn@gmail.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
060950f7f7 media: uvcvideo: Add more logging to uvc_query_ctrl() 
If we fail to query the control error code there is no information on
dmesg or in uvc_dbg. This makes difficult to debug the issue.

Print a proper error message when we cannot retrieve the error code from
the device.

Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20241128-uvc-readless-v5-2-cf16ed282af8@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
f00ee2ca8d media: uvcvideo: Support partial control reads 
Some cameras, like the ELMO MX-P3, do not return all the bytes
requested from a control if it can fit in less bytes.
Eg: Returning 0xab instead of 0x00ab.
usb 3-9: Failed to query (GET_DEF) UVC control 3 on unit 2: 1 (exp. 2).

Extend the returned value from the camera and return it.

Cc: stable@vger.kernel.org
Fixes: a763b9fb58 ("media: uvcvideo: Do not return positive errors in uvc_query_ctrl()")
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Link: https://lore.kernel.org/r/20241128-uvc-readless-v5-1-cf16ed282af8@chromium.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
c31cffd5ae media: uvcvideo: Fix event flags in uvc_ctrl_send_events 
If there is an event that needs the V4L2_EVENT_CTRL_CH_FLAGS flag, all
the following events will have that flag, regardless if they need it or
not.

This is because we keep using the same variable all the time and we do
not reset its original value.

Cc: stable@vger.kernel.org
Fixes: 805e9b4a06 ("[media] uvcvideo: Send control change events for slave ctrls when the master changes")
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241114-uvc-roi-v15-1-64cfeb56b6f8@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
a67f75c2b5 media: uvcvideo: Fix deadlock during uvc_probe 
If uvc_probe() fails, it can end up calling uvc_status_unregister() before
uvc_status_init() is called.

Fix this by checking if dev->status is NULL or not in
uvc_status_unregister().

Reported-by: syzbot+9446d5e0d25571e6a212@syzkaller.appspotmail.com
Closes: https://lore.kernel.org/linux-media/20241020160249.GD7770@pendragon.ideasonboard.com/T/#m506744621d72a2ace5dd2ab64055be9898112dbd
Fixes: c5fe3ed618 ("media: uvcvideo: Avoid race condition during unregister")
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241022-race-unreg-v1-1-2212f364d9de@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Laurent Pinchart
c6ef3a7fa9 media: uvcvideo: Fix double free in error path 
If the uvc_status_init() function fails to allocate the int_urb, it will
free the dev->status pointer but doesn't reset the pointer to NULL. This
results in the kfree() call in uvc_status_cleanup() trying to
double-free the memory. Fix it by resetting the dev->status pointer to
NULL after freeing it.

Fixes: a31a405547 ("V4L/DVB:usbvideo:don't use part of buffer for USB transfer #4")
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20241107235130.31372-1-laurent.pinchart@ideasonboard.com
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Reviewed by: Ricardo Ribalda <ribalda@chromium.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
a9ea1a3d88 media: uvcvideo: Fix crash during unbind if gpio unit is in use 
We used the wrong device for the device managed functions. We used the
usb device, when we should be using the interface device.

If we unbind the driver from the usb interface, the cleanup functions
are never called. In our case, the IRQ is never disabled.

If an IRQ is triggered, it will try to access memory sections that are
already free, causing an OOPS.

We cannot use the function devm_request_threaded_irq here. The devm_*
clean functions may be called after the main structure is released by
uvc_delete.

Luckily this bug has small impact, as it is only affected by devices
with gpio units and the user has to unbind the device, a disconnect will
not trigger this error.

Cc: stable@vger.kernel.org
Fixes: 2886477ff9 ("media: uvcvideo: Implement UVC_EXT_GPIO_UNIT")
Reviewed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241106-uvc-crashrmmod-v6-1-fbf9781c6e83@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Ricardo Ribalda
33656034de media: uvcvideo: Reorder uvc_status_init() 
Only initialize the input device if the interrupt endpoint has been
properly initialized.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Link: https://lore.kernel.org/r/20241022-order_status-v1-1-3904fafca340@chromium.org
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
 2024-12-19 14:44:43 +00:00
Linus Torvalds
70e8ef2d67 Merge tag 'media/v6.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media 
Pull media fix from Mauro Carvalho Chehab:

 - uvcvideo: Skip parsing frames of type UVC_VS_UNDEFINED in
   uvc_parse_format

* tag 'media/v6.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
  media: uvcvideo: Skip parsing frames of type UVC_VS_UNDEFINED in uvc_parse_format
 2024-11-20 14:05:34 -08:00