[PATCH v5 0/2] media: uvcvideo: Support partial control reads and minor changes

[PATCH v5 0/2] media: uvcvideo: Support partial control reads and minor changes

[Ricardo Ribalda]

Some cameras do not return all the bytes requested from a control
if it can fit in less bytes. Eg: returning 0xab instead of 0x00ab.
Support these devices.

Also, now that we are at it, improve uvc_query_ctrl() logging.

Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>
---
Changes in v5:
- Improve comment.
- Link to v4: https://lore.kernel.org/r/20241120-uvc-readless-v4-0-4672dbef3d46@chromium.org

Changes in v4:
- Improve comment.
- Keep old likely(ret == size)
- Link to v3: https://lore.kernel.org/r/20241118-uvc-readless-v3-0-d97c1a3084d0@chromium.org

Changes in v3:
- Improve documentation.
- Do not change return sequence.
- Use dev_ratelimit and dev_warn_once
- Link to v2: https://lore.kernel.org/r/20241008-uvc-readless-v2-0-04d9d51aee56@chromium.org

Changes in v2:
- Rewrite error handling (Thanks Sakari)
- Discard 2/3. It is not needed after rewriting the error handling.
- Link to v1: https://lore.kernel.org/r/20241008-uvc-readless-v1-0-042ac4581f44@chromium.org

---
Ricardo Ribalda (2):
      media: uvcvideo: Support partial control reads
      media: uvcvideo: Add more logging to uvc_query_ctrl()

 drivers/media/usb/uvc/uvc_video.c | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)
---
base-commit: 9852d85ec9d492ebef56dc5f229416c925758edc
change-id: 20241008-uvc-readless-23f9b8cad0b3

Best regards,
-- 
Ricardo Ribalda <ribalda@chromium.org>

[PATCH v5 1/2] media: uvcvideo: Support partial control reads

[Ricardo Ribalda]

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: a763b9fb58be ("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>
---
 drivers/media/usb/uvc/uvc_video.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/drivers/media/usb/uvc/uvc_video.c b/drivers/media/usb/uvc/uvc_video.c
index cd9c29532fb0..67f714bca417 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -79,6 +79,27 @@ int uvc_query_ctrl(struct uvc_device *dev, u8 query, u8 unit,
 	if (likely(ret == size))
 		return 0;
 
+	/*
+	 * Some devices return shorter USB control packets than expected if the
+	 * returned value can fit in less bytes. Zero all the bytes that the
+	 * device has not written.
+	 *
+	 * This quirk is applied to all controls, regardless of their data type.
+	 * Most controls are little-endian integers, in which case the missing
+	 * bytes become 0 MSBs. For other data types, a different heuristic
+	 * could be implemented if a device is found needing it.
+	 *
+	 * We exclude UVC_GET_INFO from the quirk. UVC_GET_LEN does not need
+	 * to be excluded because its size is always 1.
+	 */
+	if (ret > 0 && query != UVC_GET_INFO) {
+		memset(data + ret, 0, size - ret);
+		dev_warn_once(&dev->udev->dev,
+			      "UVC non compliance: %s control %u on unit %u returned %d bytes when we expected %u.\n",
+			      uvc_query_name(query), cs, unit, ret, size);
+		return 0;
+	}
+
 	if (ret != -EPIPE) {
 		dev_err(&dev->udev->dev,
 			"Failed to query (%s) UVC control %u on unit %u: %d (exp. %u).\n",

-- 
2.47.0.338.g60cca15819-goog

[PATCH v5 2/2] media: uvcvideo: Add more logging to uvc_query_ctrl()

[Ricardo Ribalda]

If we fail to query the ctrl 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>
---
 drivers/media/usb/uvc/uvc_video.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/media/usb/uvc/uvc_video.c b/drivers/media/usb/uvc/uvc_video.c
index 67f714bca417..a650d886e922 100644
--- a/drivers/media/usb/uvc/uvc_video.c
+++ b/drivers/media/usb/uvc/uvc_video.c
@@ -117,8 +117,12 @@ int uvc_query_ctrl(struct uvc_device *dev, u8 query, u8 unit,
 	error = *(u8 *)data;
 	*(u8 *)data = tmp;
 
-	if (ret != 1)
+	if (ret != 1) {
+		dev_err_ratelimited(&dev->udev->dev,
+				    "Failed to query (%s) UVC error code control %u on unit %u: %d (exp. 1).\n",
+				    uvc_query_name(query), cs, unit, ret);
 		return ret < 0 ? ret : -EPIPE;
+	}
 
 	uvc_dbg(dev, CONTROL, "Control error %u\n", error);
 

-- 
2.47.0.338.g60cca15819-goog

Re: [PATCH v5 0/2] media: uvcvideo: Support partial control reads and minor changes

[Hans de Goede]

Hi,

On 28-Nov-24 9:53 PM, Ricardo Ribalda wrote:
> Some cameras do not return all the bytes requested from a control
> if it can fit in less bytes. Eg: returning 0xab instead of 0x00ab.
> Support these devices.
> 
> Also, now that we are at it, improve uvc_query_ctrl() logging.
> 
> Signed-off-by: Ricardo Ribalda <ribalda@chromium.org>

Thank you for the new version, I've replaced the 2 patches in:
https://gitlab.freedesktop.org/linux-media/users/uvc/-/commits/next/

with this new version.

Regards,

Hans

 
> ---
> Changes in v5:
> - Improve comment.
> - Link to v4: https://lore.kernel.org/r/20241120-uvc-readless-v4-0-4672dbef3d46@chromium.org
> 
> Changes in v4:
> - Improve comment.
> - Keep old likely(ret == size)
> - Link to v3: https://lore.kernel.org/r/20241118-uvc-readless-v3-0-d97c1a3084d0@chromium.org
> 
> Changes in v3:
> - Improve documentation.
> - Do not change return sequence.
> - Use dev_ratelimit and dev_warn_once
> - Link to v2: https://lore.kernel.org/r/20241008-uvc-readless-v2-0-04d9d51aee56@chromium.org
> 
> Changes in v2:
> - Rewrite error handling (Thanks Sakari)
> - Discard 2/3. It is not needed after rewriting the error handling.
> - Link to v1: https://lore.kernel.org/r/20241008-uvc-readless-v1-0-042ac4581f44@chromium.org
> 
> ---
> Ricardo Ribalda (2):
>       media: uvcvideo: Support partial control reads
>       media: uvcvideo: Add more logging to uvc_query_ctrl()
> 
>  drivers/media/usb/uvc/uvc_video.c | 27 ++++++++++++++++++++++++++-
>  1 file changed, 26 insertions(+), 1 deletion(-)
> ---
> base-commit: 9852d85ec9d492ebef56dc5f229416c925758edc
> change-id: 20241008-uvc-readless-23f9b8cad0b3
> 
> Best regards,

Re: [PATCH v5 1/2] media: uvcvideo: Support partial control reads

[Laurent Pinchart]

On Thu, Nov 28, 2024 at 08:53:41PM +0000, Ricardo Ribalda wrote:
> 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: a763b9fb58be ("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>

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

> ---
>  drivers/media/usb/uvc/uvc_video.c | 21 +++++++++++++++++++++
>  1 file changed, 21 insertions(+)
> 
> diff --git a/drivers/media/usb/uvc/uvc_video.c b/drivers/media/usb/uvc/uvc_video.c
> index cd9c29532fb0..67f714bca417 100644
> --- a/drivers/media/usb/uvc/uvc_video.c
> +++ b/drivers/media/usb/uvc/uvc_video.c
> @@ -79,6 +79,27 @@ int uvc_query_ctrl(struct uvc_device *dev, u8 query, u8 unit,
>  	if (likely(ret == size))
>  		return 0;
>  
> +	/*
> +	 * Some devices return shorter USB control packets than expected if the
> +	 * returned value can fit in less bytes. Zero all the bytes that the
> +	 * device has not written.
> +	 *
> +	 * This quirk is applied to all controls, regardless of their data type.
> +	 * Most controls are little-endian integers, in which case the missing
> +	 * bytes become 0 MSBs. For other data types, a different heuristic
> +	 * could be implemented if a device is found needing it.
> +	 *
> +	 * We exclude UVC_GET_INFO from the quirk. UVC_GET_LEN does not need
> +	 * to be excluded because its size is always 1.
> +	 */
> +	if (ret > 0 && query != UVC_GET_INFO) {
> +		memset(data + ret, 0, size - ret);
> +		dev_warn_once(&dev->udev->dev,
> +			      "UVC non compliance: %s control %u on unit %u returned %d bytes when we expected %u.\n",
> +			      uvc_query_name(query), cs, unit, ret, size);
> +		return 0;
> +	}
> +
>  	if (ret != -EPIPE) {
>  		dev_err(&dev->udev->dev,
>  			"Failed to query (%s) UVC control %u on unit %u: %d (exp. %u).\n",

-- 
Regards,

Laurent Pinchart

Re: [PATCH v5 2/2] media: uvcvideo: Add more logging to uvc_query_ctrl()

[Laurent Pinchart]

On Thu, Nov 28, 2024 at 08:53:42PM +0000, Ricardo Ribalda wrote:
> If we fail to query the ctrl error code there is no information on dmesg

s/ctrl/control/

> 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>

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

I'll update the commit message, no need to resend.

> ---
>  drivers/media/usb/uvc/uvc_video.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/media/usb/uvc/uvc_video.c b/drivers/media/usb/uvc/uvc_video.c
> index 67f714bca417..a650d886e922 100644
> --- a/drivers/media/usb/uvc/uvc_video.c
> +++ b/drivers/media/usb/uvc/uvc_video.c
> @@ -117,8 +117,12 @@ int uvc_query_ctrl(struct uvc_device *dev, u8 query, u8 unit,
>  	error = *(u8 *)data;
>  	*(u8 *)data = tmp;
>  
> -	if (ret != 1)
> +	if (ret != 1) {
> +		dev_err_ratelimited(&dev->udev->dev,
> +				    "Failed to query (%s) UVC error code control %u on unit %u: %d (exp. 1).\n",
> +				    uvc_query_name(query), cs, unit, ret);
>  		return ret < 0 ? ret : -EPIPE;
> +	}
>  
>  	uvc_dbg(dev, CONTROL, "Control error %u\n", error);
>  

-- 
Regards,

Laurent Pinchart