RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-27 09:27:35 +00:00
Code Issues Activity Actions

LibVideo/VP9: Implement simple FIXMEs that use now supported data

Browse source
This commit is contained in:
FalseHonesty 2021-06-26 12:14:23 -04:00 • committed by Andreas Kling
parent e4f015ce3d
commit 559f5a087d
2 changed files with 31 additions and 18 deletions
Download patch file Download diff file Expand all files Collapse all files

47
Userland/Libraries/LibVideo/VP9/Decoder.cpp
View file

@ -31,9 +31,6 @@ bool Decoder::parse_frame(ByteBuffer const& frame_data)
dbgln("Finished reading uncompressed header");
SAFE_CALL(trailing_bits());
if (m_header_size_in_bytes == 0) {
// FIXME: Do we really need to read all of these bits?
// while (m_bit_stream->get_position() < m_start_bit_pos + (8 * frame_data.size()))
// RESERVED_ZERO;
dbgln("No header");
return true;
}
@ -218,9 +215,8 @@ bool Decoder::frame_size_with_refs()
for (auto i = 0; i < 3; i++) {
found_ref = m_bit_stream->read_bit();
if (found_ref) {
// TODO:
// - FrameWidth = RefFrameWidth[ref_frame_idx[ i] ];
// - FrameHeight = RefFrameHeight[ref_frame_idx[ i] ];
m_frame_width = m_ref_frame_width[m_ref_frame_idx[i]];
m_frame_height = m_ref_frame_height[m_ref_frame_idx[i]];
break;
}
}
@ -262,14 +258,12 @@ bool Decoder::loop_filter_params()
if (m_loop_filter_delta_enabled) {
if (m_bit_stream->read_bit()) {
for (auto i = 0; i < 4; i++) {
if (m_bit_stream->read_bit()) {
// TODO: loop_filter_ref_deltas[i] = s(6);
}
if (m_bit_stream->read_bit())
m_loop_filter_ref_deltas[i] = m_bit_stream->read_s(6);
}
for (auto i = 0; i < 2; i++) {
if (m_bit_stream->read_bit()) {
// TODO: loop_filter_mode_deltas[i] = s(6);
}
if (m_bit_stream->read_bit())
m_loop_filter_mode_deltas[i] = m_bit_stream->read_s(6);
}
}
}
@ -380,10 +374,15 @@ bool Decoder::setup_past_independence()
}
}
m_segmentation_abs_or_delta_update = false;
m_prev_segment_ids.clear();
m_prev_segment_ids.ensure_capacity(m_mi_rows);
for (auto row = 0u; row < m_mi_rows; row++) {
Vector<u8> sub_vector = {};
sub_vector.ensure_capacity(m_mi_cols);
for (auto col = 0u; col < m_mi_cols; col++) {
// TODO: m_prev_segment_ids[row][col] = 0;
sub_vector.append(0);
}
m_prev_segment_ids.append(sub_vector);
}
m_loop_filter_delta_enabled = true;
m_loop_filter_ref_deltas[IntraFrame] = 1;
@ -712,7 +711,6 @@ bool Decoder::decode_tiles()
for (auto tile_row = 0; tile_row < tile_rows; tile_row++) {
for (auto tile_col = 0; tile_col < tile_cols; tile_col++) {
auto last_tile = (tile_row == tile_rows - 1) && (tile_col == tile_cols - 1);
// FIXME: Spec has `sz -= tile_size + 4`, but I think we don't need this because our bit stream manages how much data we have left?
auto tile_size = last_tile ? m_bit_stream->bytes_remaining() : m_bit_stream->read_f(32);
m_mi_row_start = get_tile_offset(tile_row, m_mi_rows, m_tile_rows_log2);
m_mi_row_end = get_tile_offset(tile_row + 1, m_mi_rows, m_tile_rows_log2);
@ -790,11 +788,24 @@ bool Decoder::decode_partition(u32 row, u32 col, u8 block_subsize)
SAFE_CALL(decode_block(row, col, subsize));
} else if (partition == PartitionHorizontal) {
SAFE_CALL(decode_block(row, col, subsize));
// FIXME: if (hasRows)
// decode_block(r + halfBlock8x8, c, subsize)
if (m_has_rows)
SAFE_CALL(decode_block(row + half_block_8x8, col, subsize));
} else if (partition == PartitionVertical) {
SAFE_CALL(decode_block(row, col, subsize));
if (m_has_cols)
SAFE_CALL(decode_block(row, col + half_block_8x8, subsize));
} else {
SAFE_CALL(decode_partition(row, col, subsize));
SAFE_CALL(decode_partition(row, col + half_block_8x8, subsize));
SAFE_CALL(decode_partition(row + half_block_8x8, col, subsize));
SAFE_CALL(decode_partition(row + half_block_8x8, col + half_block_8x8, subsize));
}
if (block_subsize == Block_8x8 || partition != PartitionSplit) {
for (size_t i = 0; i < m_num_8x8; i++) {
m_above_partition_context[col + i] = 15 >> b_width_log2_lookup[subsize];
m_left_partition_context[row + i] = 15 >> b_width_log2_lookup[subsize];
}
}
// FIXME: Finish implementing partition decoding
return true;
}