| ... | @@ -34,6 +34,7 @@ Other speed optimizations: |
... | @@ -34,6 +34,7 @@ Other speed optimizations: |
|
|
- faster seeking w/ frame-mt. Right now, a flush means pictures being decoded in frame-mt are marked as "discarded" but their decoding continues. It would be better to have a quick abort check in each sbrow or so, and then return early so the post-seek can continue earlier.
|
|
- faster seeking w/ frame-mt. Right now, a flush means pictures being decoded in frame-mt are marked as "discarded" but their decoding continues. It would be better to have a quick abort check in each sbrow or so, and then return early so the post-seek can continue earlier.
|
|
|
|
|
|
|
|
Cleanups:
|
|
Cleanups:
|
|
|
|
- internal buffers in `decode.c` are all allocated as if the stream is 10-bit 4:4:4, so changing these buffers could reduce memory usage somewhat, especially the palette and coef buffers;
|
|
|
- LR/MC intermediate 2d buffers in C dsp can be reduced by doing windowed like in SIMD;
|
|
- LR/MC intermediate 2d buffers in C dsp can be reduced by doing windowed like in SIMD;
|
|
|
- cdef: noskip_mask resolution can be 8x8;
|
|
- cdef: noskip_mask resolution can be 8x8;
|
|
|
- ref_mvs: non-cur frame MVs can be at 8x8 resolution, only direct neighbours need to be 4x4;
|
|
- ref_mvs: non-cur frame MVs can be at 8x8 resolution, only direct neighbours need to be 4x4;
|
| ... | |
... | |
| ... | | ... | |