linux-kernel - Re: [PATCH v11 07/12] media: mediatek: jpeg: fix decoding resolution change operation

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: <41e00b0f4c667ff8e8a8f8b81f64fe1bbca7c001.camel@ndufresne.ca>
Date: Tue, 16 Dec 2025 16:40:44 -0500
From: Nicolas Dufresne <nicolas@...fresne.ca>
To: Kyrie Wu <kyrie.wu@...iatek.com>, Hans Verkuil
 <hverkuil-cisco@...all.nl>,  Mauro Carvalho Chehab	 <mchehab@...nel.org>,
 Rob Herring <robh@...nel.org>, Krzysztof Kozlowski	 <krzk+dt@...nel.org>,
 Conor Dooley <conor+dt@...nel.org>, Matthias Brugger	
 <matthias.bgg@...il.com>, AngeloGioacchino Del Regno	
 <angelogioacchino.delregno@...labora.com>, linux-media@...r.kernel.org, 
	devicetree@...r.kernel.org, linux-kernel@...r.kernel.org, 
	linux-arm-kernel@...ts.infradead.org, linux-mediatek@...ts.infradead.org
Subject: Re: [PATCH v11 07/12] media: mediatek: jpeg: fix decoding
 resolution change operation

Hi,

Le mardi 02 décembre 2025 à 17:47 +0800, Kyrie Wu a écrit :
> 1.add a judgement for src buffer to avoid kernel crash
> in the stop streaming function;
> 2.When a resolution changing occurs, it needs to set new
> resolution parameter immediately and then report this event.
> Otherwise, if the original software process is maintained,
> the resolution change event is reported firstly, the CPU is
> dispatched to the app to process the event, and the driver
> does not set a new resolution, which will cause parameter errors.
> 3.After a resolution change occurred, decoding should not continue,
> needs to wait until new buffers are ready and the state machine
> changed.

I mention this in other patchset, very often, 3 bullets means 3 distinct
changes. Don't use bullets. Reflow this text, rework this text, there is many
syntax error here.

> 
> Fixes: dedc21500334 ("media: mtk-jpegdec: add jpeg decode worker interface")
> 
> Signed-off-by: Kyrie Wu <kyrie.wu@...iatek.com>
> ---
>  drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c | 9 +++++++--
>  1 file changed, 7 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c
> b/drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c
> index 5ffaee4dcd19..9233bbfe2d97 100644
> --- a/drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c
> +++ b/drivers/media/platform/mediatek/jpeg/mtk_jpeg_core.c
> @@ -887,7 +887,8 @@ static void mtk_jpeg_dec_stop_streaming(struct vb2_queue
> *q)
>  
>  		vb = v4l2_m2m_next_src_buf(ctx->fh.m2m_ctx);
>  		src_buf = mtk_jpeg_vb2_to_srcbuf(&vb->vb2_buf);
> -		mtk_jpeg_set_queue_data(ctx, &src_buf->dec_param);
> +		if (!IS_ERR_OR_NULL(src_buf))
> +			mtk_jpeg_set_queue_data(ctx, &src_buf->dec_param);

The lack of vb2_wait_for_all_buffers() might explains this better, you might not
need to do random null checks like this.

Nicolas

>  		ctx->state = MTK_JPEG_RUNNING;
>  	} else if (V4L2_TYPE_IS_OUTPUT(q->type)) {
>  		ctx->state = MTK_JPEG_INIT;
> @@ -1749,11 +1750,15 @@ static void mtk_jpegdec_worker(struct work_struct
> *work)
>  
>  	if (mtk_jpeg_check_resolution_change(ctx,
>  					     &jpeg_src_buf->dec_param)) {
> -		mtk_jpeg_queue_src_chg_event(ctx);
> +		mtk_jpeg_set_queue_data(ctx, &jpeg_src_buf->dec_param);
>  		ctx->state = MTK_JPEG_SOURCE_CHANGE;
> +		mtk_jpeg_queue_src_chg_event(ctx);
>  		goto getbuf_fail;
>  	}
>  
> +	if (ctx->state == MTK_JPEG_SOURCE_CHANGE)
> +		goto getbuf_fail;
> +
>  	mtk_jpegdec_set_hw_param(ctx, hw_id, src_buf, dst_buf);
>  	ret = pm_runtime_resume_and_get(comp_jpeg[hw_id]->dev);
>  	if (ret < 0) {

Download attachment "signature.asc" of type "application/pgp-signature" (229 bytes)