lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <YSLjQqzmR2ZqN6+p@pendragon.ideasonboard.com>
Date:   Mon, 23 Aug 2021 02:52:34 +0300
From:   Laurent Pinchart <laurent.pinchart@...asonboard.com>
To:     Ricardo Ribalda <ribalda@...omium.org>
Cc:     Hans Verkuil <hverkuil-cisco@...all.nl>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Sergey Senozhatsky <sergey.senozhatsky@...il.com>,
        linux-media@...r.kernel.org, linux-kernel@...r.kernel.org,
        tfiga@...omium.org, Hans Verkuil <hans.verkuil@...co.com>
Subject: Re: [PATCH v10 18/21] uvcvideo: improve error handling in
 uvc_query_ctrl()

Hi Ricardo and Hans,

Thank you for the patch.

Please add a "media: " prefix to the subject line. Same for the other
patches in the series that are missing this.

On Fri, Jun 18, 2021 at 02:29:20PM +0200, Ricardo Ribalda wrote:
> From: Hans Verkuil <hverkuil-cisco@...all.nl>
> 
> - If __uvc_query_ctrl() failed with a non-EPIPE error, then
>   report that with dev_err. If an error code is obtained, then
>   report that with dev_dbg.
> 
> - For error 2 (Wrong state) return -EACCES instead of -EILSEQ.
>   EACCES is a much more appropriate error code. EILSEQ will return
>   "Invalid or incomplete multibyte or wide character." in strerror(),
>   which is a *very* confusing message.

I would still have split the patch in two :-)

> Reviewed-by: Ricardo Ribalda <ribalda@...omium.org>
> Signed-off-by: Hans Verkuil <hans.verkuil@...co.com>
> ---
>  drivers/media/usb/uvc/uvc_video.c | 38 ++++++++++++++++++-------------
>  1 file changed, 22 insertions(+), 16 deletions(-)
> 
> diff --git a/drivers/media/usb/uvc/uvc_video.c b/drivers/media/usb/uvc/uvc_video.c
> index daba5fe352ea..00488f15cdbf 100644
> --- a/drivers/media/usb/uvc/uvc_video.c
> +++ b/drivers/media/usb/uvc/uvc_video.c
> @@ -79,15 +79,11 @@ int uvc_query_ctrl(struct uvc_device *dev, u8 query, u8 unit,
>  	if (likely(ret == size))
>  		return 0;
>  
> -	dev_err(&dev->udev->dev,
> -		"Failed to query (%s) UVC control %u on unit %u: %d (exp. %u).\n",
> -		uvc_query_name(query), cs, unit, ret, size);
> -
> -	if (ret != -EPIPE)
> -		return ret;
> +	if (ret < 0 && ret != -EPIPE)

What if ret >= 0 ? There's a change of behaviour here that isn't
documented in the commit message. It shouldn't happen, but it would be
nice to handle it correctly.

> +		goto err;
>  
> +	/* reuse data[0] to request the error code. */
>  	tmp = *(u8 *)data;
> -
>  	ret = __uvc_query_ctrl(dev, UVC_GET_CUR, 0, intfnum,
>  			       UVC_VC_REQUEST_ERROR_CODE_CONTROL, data, 1,
>  			       UVC_CTRL_CONTROL_TIMEOUT);
> @@ -95,19 +91,21 @@ int uvc_query_ctrl(struct uvc_device *dev, u8 query, u8 unit,
>  	error = *(u8 *)data;
>  	*(u8 *)data = tmp;
>  
> -	if (ret != 1)
> -		return ret < 0 ? ret : -EPIPE;
> +	if (ret != 1) {
> +		ret = ret < 0 ? ret : -EPIPE;
> +		goto err;

This will print an error message that doesn't match the error.

> +	}
>  
> -	uvc_dbg(dev, CONTROL, "Control error %u\n", error);
> +	if (error >=1 && error <=8)

Missing space before 1 and 8.

> +		uvc_dbg(dev, CONTROL,
> +			"Failed to query (%s) UVC control %u on unit %u: got error %u.\n",
> +			uvc_query_name(query), cs, unit, error);
>  
>  	switch (error) {
> -	case 0:
> -		/* Cannot happen - we received a STALL */
> -		return -EPIPE;
>  	case 1: /* Not ready */
>  		return -EBUSY;
>  	case 2: /* Wrong state */
> -		return -EILSEQ;
> +		return -EACCES;
>  	case 3: /* Power */
>  		return -EREMOTE;
>  	case 4: /* Out of range */
> @@ -123,10 +121,18 @@ int uvc_query_ctrl(struct uvc_device *dev, u8 query, u8 unit,
>  	case 8: /* Invalid value within range */
>  		return -EINVAL;
>  	default: /* reserved or unknown */
> -		break;
> +		dev_err(&dev->udev->dev,
> +			"Failed to query (%s) UVC control %u on unit %u: got error %u.\n",
> +			uvc_query_name(query), cs, unit, error);
> +		return -EPIPE;
>  	}
>  
> -	return -EPIPE;
> +err:
> +	dev_err(&dev->udev->dev,
> +		"Failed to query (%s) UVC control %u on unit %u: %d (exp. %u).\n",
> +		uvc_query_name(query), cs, unit, ret, size);
> +
> +	return ret;
>  }
>  
>  static void uvc_fixup_video_ctrl(struct uvc_streaming *stream,

-- 
Regards,

Laurent Pinchart

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ