linux-kernel - Re: [PATCH] media: imx-jpeg: Drop the first error frames

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <d3668dc7-b8c8-497c-ae3a-b476a8658c40@oss.nxp.com>
Date: Mon, 21 Apr 2025 14:37:39 +0800
From: "Ming Qian(OSS)" <ming.qian@....nxp.com>
To: Nicolas Dufresne <nicolas@...fresne.ca>, mchehab@...nel.org,
 hverkuil-cisco@...all.nl, mirela.rabulea@....nxp.com
Cc: shawnguo@...nel.org, s.hauer@...gutronix.de, kernel@...gutronix.de,
 festevam@...il.com, xiahong.bao@....com, eagle.zhou@....com,
 linux-imx@....com, imx@...ts.linux.dev, linux-media@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH] media: imx-jpeg: Drop the first error frames

Hi Nicolas,

On 2025/4/18 21:12, Nicolas Dufresne wrote:
> Le mercredi 05 mars 2025 à 10:25 +0800, ming.qian@....nxp.com a écrit :
>> From: Ming Qian <ming.qian@....nxp.com>
>>
>> When an output buffer contains error frame header,
>> v4l2_jpeg_parse_header() will return error, then driver will mark this
>> buffer and a capture buffer done with error flag in device_run().
>>
>> But if the error occurs in the first frames, before setup the capture
>> queue, there is no chance to schedule device_run(), and there may be no
>> capture to mark error.
>>
>> So we need to drop this buffer with error flag, and make the decoding
>> can continue.
>>
>> Signed-off-by: Ming Qian <ming.qian@....nxp.com>
> 
> This seems like a relatively important bug fix to be, perhaps you can
> offer "Fixes" tag to this commit ?
> 

thanks for the reminder

>> ---
>>   drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c | 9 +++++++++
>>   1 file changed, 9 insertions(+)
>>
>> diff --git a/drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c b/drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c
>> index 1221b309a916..0e6ee997284b 100644
>> --- a/drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c
>> +++ b/drivers/media/platform/nxp/imx-jpeg/mxc-jpeg.c
>> @@ -1921,6 +1921,15 @@ static void mxc_jpeg_buf_queue(struct vb2_buffer *vb)
>>   	if (ret)
>>   		jpeg_src_buf->jpeg_parse_error = true;
>>   
>> +	/*
>> +	 * if the capture queue is not setup, the device_run() won't be scheduled,
>> +	 * need to drop the error buffer, so that the decoding can continue
>> +	 */
>> +	if (jpeg_src_buf->jpeg_parse_error &&
>> +	    !vb2_is_streaming(v4l2_m2m_get_dst_vq(ctx->fh.m2m_ctx))) {
>> +		v4l2_m2m_buf_done(vbuf, VB2_BUF_STATE_ERROR);
>> +		return;
>> +	}
> 
> Looks good to me, an alternative implementation could have been to
> merged into the error branch above.
> 
>   	if (ret) {
>   		jpeg_src_buf->jpeg_parse_error = true;
>   
> 		/*
> 	 	 * If the capture queue is not setup, the device_run() won't
> 		 * be scheduled, drop the error buffer so that the decoding
> 		 * can continue.
> 	 	 */
> 		if (!vb2_is_streaming(v4l2_m2m_get_dst_vq(ctx->fh.m2m_ctx))) {
> 			v4l2_m2m_buf_done(vbuf, VB2_BUF_STATE_ERROR);
> 			return;
> 		}
> 	}
> 
> With or without this suggestion, but with the Fixes tag, you can include in your v2:
> 
> Reviewed-by: Nicolas Dufresne <nicolas.dufresne@...labora.com>
> 

I'll apply your suggestion and with the Fixes tag in V2

Regards,
Ming

>>   end:
>>   	v4l2_m2m_buf_queue(ctx->fh.m2m_ctx, vbuf);
>>   }