You've already forked FFmpeg
mirror of
https://github.com/FFmpeg/FFmpeg.git
synced 2025-08-04 22:03:09 +02:00
6bad55eb17
This patch adds a fully-featured level 3 and 4 decoder for FFv1, supporting Golomb and all Range coding variants, all pixel formats, and all features, except for the newly added floating-point formats. On a 6000 Ada, for 3840x2160 bgr0 content at 50Mbps (standard desktop recording), it is able to do 400fps. An Alder Lake with 24 threads can barely do 100fps.
91 lines
2.7 KiB
Plaintext
91 lines
2.7 KiB
Plaintext
/*
|
|
* FFv1 codec
|
|
*
|
|
* Copyright (c) 2024 Lynne <dev@lynne.ee>
|
|
*
|
|
* This file is part of FFmpeg.
|
|
*
|
|
* FFmpeg is free software; you can redistribute it and/or
|
|
* modify it under the terms of the GNU Lesser General Public
|
|
* License as published by the Free Software Foundation; either
|
|
* version 2.1 of the License, or (at your option) any later version.
|
|
*
|
|
* FFmpeg is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
* Lesser General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU Lesser General Public
|
|
* License along with FFmpeg; if not, write to the Free Software
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
|
*/
|
|
|
|
ivec4 load_components(ivec2 pos)
|
|
{
|
|
ivec4 pix = ivec4(imageLoad(src[0], pos));
|
|
if (planar_rgb != 0) {
|
|
for (int i = 1; i < (3 + transparency); i++)
|
|
pix[i] = int(imageLoad(src[i], pos)[0]);
|
|
}
|
|
|
|
return ivec4(pix[fmt_lut[0]], pix[fmt_lut[1]],
|
|
pix[fmt_lut[2]], pix[fmt_lut[3]]);
|
|
}
|
|
|
|
void bypass_sample(ivec2 pos)
|
|
{
|
|
imageStore(dst[0], pos, load_components(pos));
|
|
}
|
|
|
|
void bypass_block(in SliceContext sc)
|
|
{
|
|
ivec2 start = ivec2(gl_LocalInvocationID) + sc.slice_pos;
|
|
ivec2 end = sc.slice_pos + sc.slice_dim;
|
|
for (uint y = start.y; y < end.y; y += gl_WorkGroupSize.y)
|
|
for (uint x = start.x; x < end.x; x += gl_WorkGroupSize.x)
|
|
bypass_sample(ivec2(x, y));
|
|
}
|
|
|
|
void transform_sample(ivec2 pos, ivec2 rct_coef)
|
|
{
|
|
ivec4 pix = load_components(pos);
|
|
pix.b -= offset;
|
|
pix.r -= offset;
|
|
pix.g -= (pix.r*rct_coef.x + pix.b*rct_coef.y) >> 2;
|
|
pix.b += pix.g;
|
|
pix.r += pix.g;
|
|
imageStore(dst[0], pos, pix);
|
|
}
|
|
|
|
void transform_sample(ivec2 pos, ivec2 rct_coef)
|
|
{
|
|
ivec4 pix = load_components(pos);
|
|
pix.b -= pix.g;
|
|
pix.r -= pix.g;
|
|
pix.g += (pix.r*rct_coef.x + pix.b*rct_coef.y) >> 2;
|
|
pix.b += offset;
|
|
pix.r += offset;
|
|
imageStore(dst[0], pos, pix);
|
|
}
|
|
|
|
void transform_block(in SliceContext sc)
|
|
{
|
|
const ivec2 rct_coef = sc.slice_rct_coef;
|
|
const ivec2 start = ivec2(gl_LocalInvocationID) + sc.slice_pos;
|
|
const ivec2 end = sc.slice_pos + sc.slice_dim;
|
|
|
|
for (uint y = start.y; y < end.y; y += gl_WorkGroupSize.y)
|
|
for (uint x = start.x; x < end.x; x += gl_WorkGroupSize.x)
|
|
transform_sample(ivec2(x, y), rct_coef);
|
|
}
|
|
|
|
void main()
|
|
{
|
|
const uint slice_idx = gl_WorkGroupID.y*gl_NumWorkGroups.x + gl_WorkGroupID.x;
|
|
|
|
if (slice_ctx[slice_idx].slice_coding_mode == 1)
|
|
bypass_block(slice_ctx[slice_idx]);
|
|
else
|
|
transform_block(slice_ctx[slice_idx]);
|
|
}
|