linux-kernel - Re: [PATCH v3 3/5] media: imx-jpeg: Propagate the output frame size to the capture side

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: <76cc3643-db74-607a-a6b9-c5e0d10081e3@oss.nxp.com>
Date:   Mon, 11 Apr 2022 14:50:13 +0300
From:   "mirela.rabulea@....nxp.com" <mirela.rabulea@....nxp.com>
To:     Ming Qian <ming.qian@....com>, mchehab@...nel.org,
        shawnguo@...nel.org, s.hauer@...gutronix.de
Cc:     hverkuil-cisco@...all.nl, kernel@...gutronix.de,
        festevam@...il.com, linux-imx@....com, linux-media@...r.kernel.org,
        linux-kernel@...r.kernel.org, devicetree@...r.kernel.org,
        linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH v3 3/5] media: imx-jpeg: Propagate the output frame size
 to the capture side

Hi Ming,

On 23.03.2022 11:05, Ming Qian wrote:
> The GStreamer v4l2videodec only ever calls S_FMT on the output side
> and then expects G_FMT on the capture side to return a valid format.
> 
> Signed-off-by: Ming Qian <ming.qian@....com>
> ---
>   .../media/platform/nxp/imx-jpeg/mxc-jpeg.c    | 30 ++++++++++++++++++-
>   1 file changed, 29 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c b/drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c
> index a95305639dd9..5dfa6f87871e 100644
> --- a/drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c
> +++ b/drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c
> @@ -1831,12 +1831,40 @@ static int mxc_jpeg_s_fmt_vid_out(struct file *file, void *priv,
>   				  struct v4l2_format *f)
>   {
>   	int ret;
> +	struct mxc_jpeg_ctx *ctx = mxc_jpeg_fh_to_ctx(priv);
> +	struct vb2_queue *dst_vq;
> +	struct mxc_jpeg_q_data *q_data_cap;
> +	enum v4l2_buf_type cap_type = V4L2_BUF_TYPE_VIDEO_CAPTURE_MPLANE;
> +	struct v4l2_format fc;
>   
>   	ret = mxc_jpeg_try_fmt_vid_out(file, priv, f);
>   	if (ret)
>   		return ret;
>   
> -	return mxc_jpeg_s_fmt(mxc_jpeg_fh_to_ctx(priv), f);
> +	ret = mxc_jpeg_s_fmt(mxc_jpeg_fh_to_ctx(priv), f);
> +	if (ret)
> +		return ret;
> +
> +	if (ctx->mxc_jpeg->mode != MXC_JPEG_DECODE)
> +		return 0;
> +
> +	dst_vq = v4l2_m2m_get_vq(ctx->fh.m2m_ctx, cap_type);
> +	if (!dst_vq)
> +		return -EINVAL;
> +
> +	if (vb2_is_busy(dst_vq))
> +		return 0;
> +
> +	q_data_cap = mxc_jpeg_get_q_data(ctx, cap_type);
> +	if (q_data_cap->w == f->fmt.pix_mp.width && q_data_cap->h == f->fmt.pix_mp.height)
> +		return 0;

Is this an optimization, to avoid propagating the format to the capture 
queue, if it is already set? If so, shouldn't fourcc also be part of the 
comparison?

Thanks,
Mirela

> +	memset(&fc, 0, sizeof(fc));
> +	fc.type = cap_type;
> +	fc.fmt.pix_mp.pixelformat = q_data_cap->fmt->fourcc;
> +	fc.fmt.pix_mp.width = f->fmt.pix_mp.width;
> +	fc.fmt.pix_mp.height = f->fmt.pix_mp.height;
> +
> +	return mxc_jpeg_s_fmt_vid_cap(file, priv, &fc);
>   }
>   
>   static int mxc_jpeg_g_fmt_vid(struct file *file, void *priv,