Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | mdct15: add inverse transform postrotation SIMD | Rostislav Pehlivanov | 2017-07-30 | 1 | -15/+23 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | 2.5ms frames: Before (c): 2638 decicycles in postrotate, 2097040 runs, 112 skips After (sse3): 1467 decicycles in postrotate, 2097083 runs, 69 skips After (avx2): 1244 decicycles in postrotate, 2097085 runs, 67 skips 5ms frames: Before (c): 4987 decicycles in postrotate, 1048371 runs, 205 skips After (sse3): 2644 decicycles in postrotate, 1048509 runs, 67 skips After (avx2): 2031 decicycles in postrotate, 1048523 runs, 53 skips 10ms frames: Before (c): 9153 decicycles in postrotate, 523575 runs, 713 skips After (sse3): 5110 decicycles in postrotate, 523726 runs, 562 skips After (avx2): 3738 decicycles in postrotate, 524223 runs, 65 skips 20ms frames: Before (c): 17857 decicycles in postrotate, 261866 runs, 278 skips After (sse3): 10041 decicycles in postrotate, 261746 runs, 398 skips After (avx2): 7050 decicycles in postrotate, 262116 runs, 28 skips Improves total decoding performance for real world content by 9% with avx2. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com> | ||||
* | mdct15: simplify prereindexing and forward transform postrotation | Rostislav Pehlivanov | 2017-07-25 | 1 | -19/+18 |
| | | | | | | Equivalent. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com> | ||||
* | mdct15: remove redundant scale argument to imdct_half | Rostislav Pehlivanov | 2017-07-11 | 1 | -8/+3 |
| | | | | | | | The only use of that argument was for Opus downmixing which is very rare and better done after the mdcts. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com> | ||||
* | mdct15: add assembly optimizations for the 15-point FFT | Rostislav Pehlivanov | 2017-06-23 | 1 | -92/+90 |
| | | | | | | | c: 1802 decicycles in fft15,16774635 runs, 2581 skips avx: 865 decicycles in fft15,16776378 runs, 838 skips Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com> | ||||
* | mdct15: use existing complex multiplication macros | Rostislav Pehlivanov | 2017-04-08 | 1 | -21/+15 |
| | | | | Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com> | ||||
* | mdct15: fix left shift of a negative value | Rostislav Pehlivanov | 2017-03-22 | 1 | -1/+1 |
| | | | | | | Should fix Chromium issue 704064. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com> | ||||
* | imdct15: rename to mdct15 and add a forward transform | Rostislav Pehlivanov | 2017-02-14 | 1 | -0/+335 |
Handles strides (needed for Opus transients), does pre-reindexing and folding without needing a copy. Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com> |