[FFmpeg-devel] [PATCH 1/3] lavu/pixfmt: Add Y216, Y410, and Y416 formats
Xiang, Haihao
haihao.xiang at intel.com
Mon Aug 15 09:12:20 EEST 2022
On Sun, 2022-08-14 at 14:33 -0700, Philip Langdale wrote:
> These are the formats returned by the Intel VAAPI decoder for 12bit
> 4:2:2, 10bit 4:4:4, and 12bit 4:4:4 respectively. As with the already
> supported Y210 and YUVA (AVUY) formats, they are the formats Microsoft
> picked as their preferred 4:2:2 and 4:4:4 video formats, and Intel ran
> with it.
>
> Y216 is simply an extension of Y210 to say all 16bits will be used, and
> Y416 is a normal looking packed 4 channel format. Y410 is an annoying
> format that packs three 10bit channels into 32bits with 2bits of alpha.
>
> As a result, I had to define Y410 as a bitstream format, even though
> each pixel is byte-aligned. If it is in-fact possible to define as a
> normal byte-aligned format, please let me know how.
>
> As with VUYA, I have kept the formal definition of Y410 and Y416 as
> formats with alpha channels to maintain fidelity. The Intel folks say
> they prefer this and they would rather explicitly use a format defined
> as not having alpha than to silently drop it. The hardware decoder
> does at least ensure the alpha channel is set to full opacity.
>
> Signed-off-by: Philip Langdale <philipl at overt.org>
> ---
> libavutil/pixdesc.c | 77 +++++++++++++++++++++++++++++++-
> libavutil/pixfmt.h | 12 +++++
> tests/ref/fate/imgutils | 6 +++
> tests/ref/fate/sws-pixdesc-query | 25 +++++++++++
> 4 files changed, 119 insertions(+), 1 deletion(-)
>
> diff --git a/libavutil/pixdesc.c b/libavutil/pixdesc.c
> index f7558ff8b9..5dee3a95d3 100644
> --- a/libavutil/pixdesc.c
> +++ b/libavutil/pixdesc.c
> @@ -2532,6 +2532,81 @@ static const AVPixFmtDescriptor
> av_pix_fmt_descriptors[AV_PIX_FMT_NB] = {
> .flags = AV_PIX_FMT_FLAG_RGB | AV_PIX_FMT_FLAG_ALPHA |
> AV_PIX_FMT_FLAG_FLOAT,
> },
> + [AV_PIX_FMT_Y216LE] = {
> + .name = "y216le",
> + .nb_components = 3,
> + .log2_chroma_w = 1,
> + .log2_chroma_h = 0,
> + .comp = {
> + { 0, 4, 0, 0, 16 }, /* Y */
> + { 0, 8, 2, 0, 16 }, /* U */
> + { 0, 8, 6, 0, 16 }, /* V */
> + },
> + },
> + [AV_PIX_FMT_Y216BE] = {
> + .name = "y216be",
> + .nb_components = 3,
> + .log2_chroma_w = 1,
> + .log2_chroma_h = 0,
> + .comp = {
> + { 0, 4, 0, 0, 16 }, /* Y */
> + { 0, 8, 2, 0, 16 }, /* U */
> + { 0, 8, 6, 0, 16 }, /* V */
> + },
> + .flags = AV_PIX_FMT_FLAG_BE,
> + },
> + [AV_PIX_FMT_Y410LE] = {
> + .name = "y410le",
> + .nb_components= 4,
> + .log2_chroma_w= 0,
> + .log2_chroma_h= 0,
> + .comp = {
> + { 0, 32, 10, 0, 10 }, /* Y */
> + { 0, 32, 0, 0, 10 }, /* U */
> + { 0, 32, 20, 0, 10 }, /* V */
> + { 0, 32, 30, 0, 2 }, /* A */
> + },
> + .flags = AV_PIX_FMT_FLAG_ALPHA | AV_PIX_FMT_FLAG_BITSTREAM,
> + },
> + [AV_PIX_FMT_Y410BE] = {
> + .name = "y410be",
> + .nb_components= 4,
> + .log2_chroma_w= 0,
> + .log2_chroma_h= 0,
> + .comp = {
> + { 0, 32, 10, 0, 10 }, /* Y */
> + { 0, 32, 0, 0, 10 }, /* U */
> + { 0, 32, 20, 0, 10 }, /* V */
> + { 0, 32, 30, 0, 2 }, /* A */
> + },
> + .flags = AV_PIX_FMT_FLAG_BE | AV_PIX_FMT_FLAG_ALPHA |
> AV_PIX_FMT_FLAG_BITSTREAM,
> + },
> + [AV_PIX_FMT_Y416LE] = {
> + .name = "y416le",
> + .nb_components= 4,
> + .log2_chroma_w= 0,
> + .log2_chroma_h= 0,
> + .comp = {
> + { 0, 8, 2, 0, 16 }, /* Y */
> + { 0, 8, 0, 0, 16 }, /* U */
> + { 0, 8, 4, 0, 16 }, /* V */
> + { 0, 8, 6, 0, 16 }, /* A */
> + },
> + .flags = AV_PIX_FMT_FLAG_ALPHA,
> + },
> + [AV_PIX_FMT_Y416BE] = {
> + .name = "y416be",
> + .nb_components= 4,
> + .log2_chroma_w= 0,
> + .log2_chroma_h= 0,
> + .comp = {
> + { 0, 8, 2, 0, 16 }, /* Y */
> + { 0, 8, 0, 0, 16 }, /* U */
> + { 0, 8, 4, 0, 16 }, /* V */
> + { 0, 8, 6, 0, 16 }, /* A */
> + },
> + .flags = AV_PIX_FMT_FLAG_BE | AV_PIX_FMT_FLAG_ALPHA,
> + },
> };
>
> static const char * const color_range_names[] = {
> @@ -2767,7 +2842,7 @@ void ff_check_pixfmt_descriptors(void){
>
> if (!d->name && !d->nb_components && !d->log2_chroma_w && !d-
> >log2_chroma_h && !d->flags)
> continue;
> -// av_log(NULL, AV_LOG_DEBUG, "Checking: %s\n", d->name);
> + av_log(NULL, AV_LOG_INFO, "Checking: %s\n", d->name);
> av_assert0(d->log2_chroma_w <= 3);
> av_assert0(d->log2_chroma_h <= 3);
> av_assert0(d->nb_components <= 4);
> diff --git a/libavutil/pixfmt.h b/libavutil/pixfmt.h
> index 86c9bdefeb..485655f0c0 100644
> --- a/libavutil/pixfmt.h
> +++ b/libavutil/pixfmt.h
> @@ -372,6 +372,15 @@ enum AVPixelFormat {
> AV_PIX_FMT_RGBAF16BE, ///< IEEE-754 half precision packed RGBA
> 16:16:16:16, 64bpp, RGBARGBA..., big-endian
> AV_PIX_FMT_RGBAF16LE, ///< IEEE-754 half precision packed RGBA
> 16:16:16:16, 64bpp, RGBARGBA..., little-endian
>
> + AV_PIX_FMT_Y216BE, ///< packed YUV 4:2:2 like YUYV422, 32bpp, big-
> endian
> + AV_PIX_FMT_Y216LE, ///< packed YUV 4:2:2 like YUYV422, 32bpp, big-
> endian
> +
> + AV_PIX_FMT_Y410BE, ///< packed AVYU 2:10:10:10, 32bpp, (msb)2A 10V
> 10Y 10U(lsb), big-endian
> + AV_PIX_FMT_Y410LE, ///< packed AVYU 2:10:10:10, 32bpp, (msb)2A 10V
> 10Y 10U(lsb), little-endian
> +
> + AV_PIX_FMT_Y416BE, ///< packed AVYU 16:16:16:16, 64bpp, big-endian
> + AV_PIX_FMT_Y416LE, ///< packed AVYU 16:16:16:16, 64bpp, little-
> endian
> +
> AV_PIX_FMT_NB ///< number of pixel formats, DO NOT USE THIS if
> you want to link with shared libav* because the number of formats might differ
> between versions
> };
>
> @@ -461,6 +470,9 @@ enum AVPixelFormat {
> #define AV_PIX_FMT_P016 AV_PIX_FMT_NE(P016BE, P016LE)
>
> #define AV_PIX_FMT_Y210 AV_PIX_FMT_NE(Y210BE, Y210LE)
> +#define AV_PIX_FMT_Y216 AV_PIX_FMT_NE(Y216BE, Y216LE)
> +#define AV_PIX_FMT_Y410 AV_PIX_FMT_NE(Y410BE, Y410LE)
> +#define AV_PIX_FMT_Y416 AV_PIX_FMT_NE(Y416BE, Y416LE)
> #define AV_PIX_FMT_X2RGB10 AV_PIX_FMT_NE(X2RGB10BE, X2RGB10LE)
> #define AV_PIX_FMT_X2BGR10 AV_PIX_FMT_NE(X2BGR10BE, X2BGR10LE)
>
> diff --git a/tests/ref/fate/imgutils b/tests/ref/fate/imgutils
> index 01c9877de5..ea959c26b1 100644
> --- a/tests/ref/fate/imgutils
> +++ b/tests/ref/fate/imgutils
> @@ -249,3 +249,9 @@ p416le planes: 2, linesizes: 128 256 0 0,
> plane_sizes: 6144 12288
> vuya planes: 1, linesizes: 256 0 0 0, plane_sizes:
> 12288 0 0 0, plane_offsets: 0 0 0, total_size: 12288
> rgbaf16be planes: 1, linesizes: 512 0 0 0, plane_sizes:
> 24576 0 0 0, plane_offsets: 0 0 0, total_size: 24576
> rgbaf16le planes: 1, linesizes: 512 0 0 0, plane_sizes:
> 24576 0 0 0, plane_offsets: 0 0 0, total_size: 24576
> +y216be planes: 1, linesizes: 256 0 0 0, plane_sizes:
> 12288 0 0 0, plane_offsets: 0 0 0, total_size: 12288
> +y216le planes: 1, linesizes: 256 0 0 0, plane_sizes:
> 12288 0 0 0, plane_offsets: 0 0 0, total_size: 12288
> +y410be planes: 1, linesizes: 256 0 0 0, plane_sizes:
> 12288 0 0 0, plane_offsets: 0 0 0, total_size: 12288
> +y410le planes: 1, linesizes: 256 0 0 0, plane_sizes:
> 12288 0 0 0, plane_offsets: 0 0 0, total_size: 12288
> +y416be planes: 1, linesizes: 512 0 0 0, plane_sizes:
> 24576 0 0 0, plane_offsets: 0 0 0, total_size: 24576
> +y416le planes: 1, linesizes: 512 0 0 0, plane_sizes:
> 24576 0 0 0, plane_offsets: 0 0 0, total_size: 24576
> diff --git a/tests/ref/fate/sws-pixdesc-query b/tests/ref/fate/sws-pixdesc-
> query
> index f79d99e513..2dea9c5a3c 100644
> --- a/tests/ref/fate/sws-pixdesc-query
> +++ b/tests/ref/fate/sws-pixdesc-query
> @@ -23,6 +23,10 @@ is16BPS:
> rgba64le
> rgbaf16be
> rgbaf16le
> + y216be
> + y216le
> + y416be
> + y416le
> ya16be
> ya16le
> yuv420p16be
> @@ -75,6 +79,8 @@ isNBPS:
> xyz12le
> y210be
> y210le
> + y410be
> + y410le
> yuv420p10be
> yuv420p10le
> yuv420p12be
> @@ -164,6 +170,9 @@ isBE:
> x2rgb10be
> xyz12be
> y210be
> + y216be
> + y410be
> + y416be
> ya16be
> yuv420p10be
> yuv420p12be
> @@ -223,6 +232,12 @@ isYUV:
> xyz12le
> y210be
> y210le
> + y216be
> + y216le
> + y410be
> + y410le
> + y416be
> + y416le
> ya16be
> ya16le
> ya8
> @@ -665,6 +680,10 @@ ALPHA:
> rgbaf16be
> rgbaf16le
> vuya
> + y410be
> + y410le
> + y416be
> + y416le
> ya16be
> ya16le
> ya8
> @@ -761,6 +780,12 @@ Packed:
> xyz12le
> y210be
> y210le
> + y216be
> + y216le
> + y410be
> + y410le
> + y416be
> + y416le
> ya16be
> ya16le
> ya8
Hi Philip,
May we add new formats P012, Y212 and Y412 for 12bit contents ? I agree with
Mark's comment in
https://patchwork.ffmpeg.org/project/ffmpeg/patch/20200619015248.21873-1-fei.w.wang@intel.com/
"
Tracking it separately does not seem fun - it looks to me like it would require
adding a new bit depth field to AVFrame.
FFmpeg has always used pixfmt as defining both the memory layout and which bits
are used in that (so, for example, ARGB and 0RGB are not the same thing), unlike
most of the graphics APIs which tend to
define those two separately.
"
The bit depth is known if using p012 for 12bit contents in the command below:
$ ffmpeg -init_hw_device vaapi=vaapi:/dev/dri/renderD128 -f lavfi -i yuvtestsrc
-vf "format=p012,hwupload" -c:v hevc_vaapi -f null -
If using p016 for 12bit contents, how do we know the bit depth is 12 when
converting yuv444 to p016 in the command below ?
$ ffmpeg -init_hw_device vaapi=vaapi:/dev/dri/renderD128 -f lavfi -i yuvtestsrc
-vf "format=p016,hwupload" -c:v hevc_vaapi -f null -
Thanks
Haihao
More information about the ffmpeg-devel
mailing list