RGBCube/serenity
1
Fork 0
mirror of https://github.com/RGBCube/serenity synced 2025-07-26 19:07:36 +00:00
Code Issues Activity Actions

LibVideo/VP9: Remove m_tokens and m_token_cache from Parser

Browse source 
Only the residual tokens array needs to be kept for the transforms to
use after all the tokens have been parsed. The token cache is able to
be kept in the stack only for the duration of the token parsing loop.
This commit is contained in:
Zaggy1024 2022-11-27 00:41:09 -06:00 • committed by Andreas Kling
parent a4f14f220d
commit 316dad7bf7
6 changed files with 32 additions and 27 deletions
Download patch file Download diff file Expand all files Collapse all files

4
Userland/Libraries/LibVideo/VP9/Decoder.cpp
View file

@ -1087,12 +1087,12 @@ DecoderErrorOr<void> Decoder::reconstruct(u8 plane, BlockContext const& block_co
auto index = index_from_row_and_column(i, j, block_size);
if (index == 0)
continue;
dequantized[index] = (m_parser->m_tokens[index] * ac_quant) / dq_denominator;
dequantized[index] = (block_context.residual_tokens[index] * ac_quant) / dq_denominator;
}
}
// 2. Dequant[ 0 ][ 0 ] is set equal to ( Tokens[ 0 ] * get_dc_quant( plane ) ) / dqDenom
dequantized[0] = (m_parser->m_tokens[0] * get_dc_quantizer(block_context, plane)) / dq_denominator;
dequantized[0] = (block_context.residual_tokens[0] * get_dc_quantizer(block_context, plane)) / dq_denominator;
// It is a requirement of bitstream conformance that the values written into the Dequant array in steps 1 and 2
// are representable by a signed integer with 8 + BitDepth bits.