Re: [PATCH v5 09/10] media: Avoid parsing quantization and huffman tables

[Philipp Zabel]

Hi Mirela,

On Thu, 2020-11-12 at 05:05 +0200, Mirela Rabulea (OSS) wrote:
> From: Mirela Rabulea <mirela.rabulea@xxxxxxx>
> 
> These are optional in struct v4l2_jpeg_header, so do not parse if
> not requested, save some time.
> 
> Signed-off-by: Mirela Rabulea <mirela.rabulea@xxxxxxx>
> ---
>  drivers/media/v4l2-core/v4l2-jpeg.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/media/v4l2-core/v4l2-jpeg.c b/drivers/media/v4l2-core/v4l2-jpeg.c
> index d77e04083d57..7576cd0ce6b9 100644
> --- a/drivers/media/v4l2-core/v4l2-jpeg.c
> +++ b/drivers/media/v4l2-core/v4l2-jpeg.c
> @@ -307,6 +307,9 @@ static int jpeg_parse_quantization_tables(struct jpeg_stream *stream,
>  {
>  	int len = jpeg_get_word_be(stream);
>  
> +	if (!tables)
> +		return 0;
> +
>  	if (len < 0)
>  		return len;
>  	/* Lq = 2 + n * 65 (for baseline DCT), n >= 1 */
> @@ -361,6 +364,9 @@ static int jpeg_parse_huffman_tables(struct jpeg_stream *stream,
>  	int mt;
>  	int len = jpeg_get_word_be(stream);
>  
> +	if (!tables)
> +		return 0;
> +
>  	if (len < 0)
>  		return len;
>  	/* Table B.5 - Huffman table specification parameter sizes and values */

I don't understand this. jpeg_parse_quantization_tables() is only called
if v4l2_jpeg_parse_header() finds a DQT marker. The entire quantization
table-specification parameter block is optional, but when a DQT marker
is present, IIUC the block must contain at least one table (see B.2.4.1,
specifically figure B.6).

regards
Philipp