... | ... | @@ -29,7 +29,7 @@ Performance optimizations: |
|
|
- make palette typed `pixel` instead of `uint16_t` so we don't need one extra `packuswb` in each 8-bit simd function call;
|
|
|
- order_palette() to dsp for simd;
|
|
|
- change coef contexting (hi/lo_ctx) to be diagonal-oriented for dsp/simd;
|
|
|
- change multi-symbol coding `read_symbol()` symbol discovery loop to be simd'ed;
|
|
|
- change multi-symbol coding `read_symbol()` symbol discovery loop and adaptivity to be simd'ed;
|
|
|
- postfilter threading;
|
|
|
- threading can become a generic worker queue (one tile_sbrow symbol parsing/recon, one sbrow postfilter(s)) and then use a generic single threadpool instead of separate tile/frame[/postfilter?] ones.
|
|
|
|
... | ... | |