public inbox for linux-media@vger.kernel.org
 help / color / mirror / Atom feed
From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
To: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
Cc: linuxarm@huawei.com, mauro.chehab@huawei.com,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	linux-kernel@vger.kernel.org, linux-media@vger.kernel.org
Subject: Re: [PATCH 2/5] media: uvc: don't do DMA on stack
Date: Mon, 21 Jun 2021 15:09:10 +0300	[thread overview]
Message-ID: <YNCBZtuewVrAoaVu@pendragon.ideasonboard.com> (raw)
In-Reply-To: <d9d3b95daf726f3d4dc09477b1f61669763e301a.1624276138.git.mchehab+huawei@kernel.org>

Hi Mauro,

Thank you for the patch.

On Mon, Jun 21, 2021 at 01:56:46PM +0200, Mauro Carvalho Chehab wrote:
> As warned by smatch:
> 	drivers/media/usb/uvc/uvc_v4l2.c:911 uvc_ioctl_g_input() error: doing dma on the stack (&i)
> 	drivers/media/usb/uvc/uvc_v4l2.c:943 uvc_ioctl_s_input() error: doing dma on the stack (&i)
> 
> those two functions call uvc_query_ctrl passing a pointer to
> a data at the DMA stack. those are used to send URBs via
> usb_control_msg(). Using DMA stack is not supported and should
> not work anymore on modern Linux versions.
> 
> So, use a temporary buffer, allocated together with
> struct uvc_video_chain.

DMA in a memory location that may share a cache line with something else
isn't a great idea either. The buffer should be kmalloc()ed in
uvc_ioctl_g_input() and uvc_ioctl_s_input().

> Signed-off-by: Mauro Carvalho Chehab <mchehab+huawei@kernel.org>
> ---
>  drivers/media/usb/uvc/uvc_v4l2.c | 10 ++++------
>  drivers/media/usb/uvc/uvcvideo.h |  3 +++
>  2 files changed, 7 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/media/usb/uvc/uvc_v4l2.c b/drivers/media/usb/uvc/uvc_v4l2.c
> index 252136cc885c..e60d4675881a 100644
> --- a/drivers/media/usb/uvc/uvc_v4l2.c
> +++ b/drivers/media/usb/uvc/uvc_v4l2.c
> @@ -900,7 +900,6 @@ static int uvc_ioctl_g_input(struct file *file, void *fh, unsigned int *input)
>  	struct uvc_fh *handle = fh;
>  	struct uvc_video_chain *chain = handle->chain;
>  	int ret;
> -	u8 i;
>  
>  	if (chain->selector == NULL ||
>  	    (chain->dev->quirks & UVC_QUIRK_IGNORE_SELECTOR_UNIT)) {
> @@ -910,11 +909,11 @@ static int uvc_ioctl_g_input(struct file *file, void *fh, unsigned int *input)
>  
>  	ret = uvc_query_ctrl(chain->dev, UVC_GET_CUR, chain->selector->id,
>  			     chain->dev->intfnum,  UVC_SU_INPUT_SELECT_CONTROL,
> -			     &i, 1);
> +			     &chain->input, 1);
>  	if (ret < 0)
>  		return ret;
>  
> -	*input = i - 1;
> +	*input = chain->input - 1;
>  	return 0;
>  }
>  
> @@ -923,7 +922,6 @@ static int uvc_ioctl_s_input(struct file *file, void *fh, unsigned int input)
>  	struct uvc_fh *handle = fh;
>  	struct uvc_video_chain *chain = handle->chain;
>  	int ret;
> -	u32 i;
>  
>  	ret = uvc_acquire_privileges(handle);
>  	if (ret < 0)
> @@ -939,10 +937,10 @@ static int uvc_ioctl_s_input(struct file *file, void *fh, unsigned int input)
>  	if (input >= chain->selector->bNrInPins)
>  		return -EINVAL;
>  
> -	i = input + 1;
> +	chain->input = input + 1;
>  	return uvc_query_ctrl(chain->dev, UVC_SET_CUR, chain->selector->id,
>  			      chain->dev->intfnum, UVC_SU_INPUT_SELECT_CONTROL,
> -			      &i, 1);
> +			      &chain->input, 1);
>  }
>  
>  static int uvc_ioctl_queryctrl(struct file *file, void *fh,
> diff --git a/drivers/media/usb/uvc/uvcvideo.h b/drivers/media/usb/uvc/uvcvideo.h
> index cce5e38133cd..3c0ed90d6912 100644
> --- a/drivers/media/usb/uvc/uvcvideo.h
> +++ b/drivers/media/usb/uvc/uvcvideo.h
> @@ -475,6 +475,9 @@ struct uvc_video_chain {
>  	struct mutex ctrl_mutex;		/* Protects ctrl.info */
>  
>  	struct v4l2_prio_state prio;		/* V4L2 priority state */
> +
> +	u8 input;				/* buffer for set/get input */
> +
>  	u32 caps;				/* V4L2 chain-wide caps */
>  };
>  

-- 
Regards,

Laurent Pinchart

  reply	other threads:[~2021-06-21 12:09 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-21 11:56 [PATCH 0/5] some smatch fixes Mauro Carvalho Chehab
2021-06-21 11:56 ` [PATCH 1/5] media: dib8000: rewrite the init prbs logic Mauro Carvalho Chehab
2021-06-21 11:56 ` [PATCH 2/5] media: uvc: don't do DMA on stack Mauro Carvalho Chehab
2021-06-21 12:09   ` Laurent Pinchart [this message]
2021-06-21 11:56 ` [PATCH 3/5] media: v4l2-flash-led-class: drop an useless check Mauro Carvalho Chehab
2021-06-24  9:31   ` Sakari Ailus
2021-06-24  9:59     ` Mauro Carvalho Chehab
2021-06-24 10:14       ` Sakari Ailus
2021-06-24 11:12         ` Sakari Ailus
2021-06-24 11:32         ` Mauro Carvalho Chehab
2021-06-24 11:37           ` Sakari Ailus
2021-06-21 11:56 ` [PATCH 4/5] media: ivtv: prevent going past the hw arrays Mauro Carvalho Chehab
2021-06-21 15:24   ` Hans Verkuil
2021-06-21 11:56 ` [PATCH 5/5] media: sti: don't copy past the size Mauro Carvalho Chehab

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YNCBZtuewVrAoaVu@pendragon.ideasonboard.com \
    --to=laurent.pinchart@ideasonboard.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linuxarm@huawei.com \
    --cc=mauro.chehab@huawei.com \
    --cc=mchehab+huawei@kernel.org \
    --cc=mchehab@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox