linux-kernel - Re: [PATCH] media: mediatek: vcodec: fix h264 cavlc bitstream fail

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: <CAGXv+5G5Lr6rCB0D+q9egRFmhAzpGL49dNrQeCT8JpeUT+OiAA@mail.gmail.com>
Date:   Tue, 18 Oct 2022 17:55:36 +0800
From:   Chen-Yu Tsai <wenst@...omium.org>
To:     Yunfei Dong <yunfei.dong@...iatek.com>
Cc:     Nicolas Dufresne <nicolas@...fresne.ca>,
        Hans Verkuil <hverkuil-cisco@...all.nl>,
        AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@...labora.com>,
        Benjamin Gaignard <benjamin.gaignard@...labora.com>,
        Tiffany Lin <tiffany.lin@...iatek.com>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Matthias Brugger <matthias.bgg@...il.com>,
        George Sun <george.sun@...iatek.com>,
        Xiaoyong Lu <xiaoyong.lu@...iatek.com>,
        Hsin-Yi Wang <hsinyi@...omium.org>,
        Fritz Koenig <frkoenig@...omium.org>,
        Daniel Vetter <daniel@...ll.ch>,
        Steve Cho <stevecho@...omium.org>, 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,
        Project_Global_Chrome_Upstream_Group@...iatek.com
Subject: Re: [PATCH] media: mediatek: vcodec: fix h264 cavlc bitstream fail

On Mon, Oct 17, 2022 at 5:02 PM Yunfei Dong <yunfei.dong@...iatek.com> wrote:
>
> Some cavlc bistream will decode fail when the frame size is small than
> 20 bytes. Need to add pending data at the end of the bitstream.

"magic terminating pattern" instead of "pending data"?

> For the size of mapped memory is at least one page, adding four bytes data
> won't lead to access unknown virtual memory.

Actually we can narrow this down a bit. The minimum dimension (16x16)
sets the minimum size of the buffer at 256 bytes.

> Fixes: 59fba9eed5a7 ("media: mediatek: vcodec: support stateless H.264 decoding for mt8192")
> Signed-off-by: Yunfei Dong <yunfei.dong@...iatek.com>
> ---
>  .../vcodec/vdec/vdec_h264_req_multi_if.c      | 27 ++++++++++++++++---
>  1 file changed, 24 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/media/platform/mediatek/vcodec/vdec/vdec_h264_req_multi_if.c b/drivers/media/platform/mediatek/vcodec/vdec/vdec_h264_req_multi_if.c
> index 4cc92700692b..c1583dddcb04 100644
> --- a/drivers/media/platform/mediatek/vcodec/vdec/vdec_h264_req_multi_if.c
> +++ b/drivers/media/platform/mediatek/vcodec/vdec/vdec_h264_req_multi_if.c
> @@ -539,6 +539,24 @@ static int vdec_h264_slice_core_decode(struct vdec_lat_buf *lat_buf)
>         return 0;
>  }
>
> +static void vdec_h264_insert_startcode(struct mtk_vcodec_dev *vcodec_dev, unsigned char *buf,
> +                                      size_t *bs_size, struct mtk_h264_pps_param *pps)
> +{
> +       struct device *dev = &vcodec_dev->plat_dev->dev;
> +
> +       /* cavlc bitstream when entropy_coding_mode_flag is false. */
> +       if (pps->entropy_coding_mode_flag || *bs_size > 20 ||
> +           !(of_device_is_compatible(dev->of_node, "mediatek,mt8192-vcodec-dec") ||
> +           of_device_is_compatible(dev->of_node, "mediatek,mt8195-vcodec-dec")))
> +               return;
> +

There should be a comment here describing what is added.


ChenYu

> +       buf[*bs_size] = 0;
> +       buf[*bs_size + 1] = 0;
> +       buf[*bs_size + 2] = 1;
> +       buf[*bs_size + 3] = 0xff;
> +       (*bs_size) += 4;
> +}
> +
>  static int vdec_h264_slice_lat_decode(void *h_vdec, struct mtk_vcodec_mem *bs,
>                                       struct vdec_fb *fb, bool *res_chg)
>  {
> @@ -582,9 +600,6 @@ static int vdec_h264_slice_lat_decode(void *h_vdec, struct mtk_vcodec_mem *bs,
>         }
>
>         inst->vsi->dec.nal_info = buf[nal_start_idx];
> -       inst->vsi->dec.bs_buf_addr = (u64)bs->dma_addr;
> -       inst->vsi->dec.bs_buf_size = bs->size;
> -
>         lat_buf->src_buf_req = src_buf_info->m2m_buf.vb.vb2_buf.req_obj.req;
>         v4l2_m2m_buf_copy_metadata(&src_buf_info->m2m_buf.vb, &lat_buf->ts_info, true);
>
> @@ -592,6 +607,12 @@ static int vdec_h264_slice_lat_decode(void *h_vdec, struct mtk_vcodec_mem *bs,
>         if (err)
>                 goto err_free_fb_out;
>
> +       vdec_h264_insert_startcode(inst->ctx->dev, buf, &bs->size,
> +                                  &share_info->h264_slice_params.pps);
> +
> +       inst->vsi->dec.bs_buf_addr = (uint64_t)bs->dma_addr;
> +       inst->vsi->dec.bs_buf_size = bs->size;
> +
>         *res_chg = inst->resolution_changed;
>         if (inst->resolution_changed) {
>                 mtk_vcodec_debug(inst, "- resolution changed -");
> --
> 2.25.1
>