diff options
| author | Lynne <dev@lynne.ee> | 2022-09-19 23:48:53 +0200 |
|---|---|---|
| committer | Lynne <dev@lynne.ee> | 2022-09-21 07:12:39 +0200 |
| commit | 3ade6a8644ab519fcd7caa7ef457dd406162bc14 (patch) | |
| tree | 80135c60b8bdc5d4975f986f68ae44ef80449335 /tests/checkasm/checkasm.h | |
| parent | cc367a9b8aa1e473bf60c5dc5e03431a7bbcd125 (diff) | |
| download | ffmpeg-3ade6a8644ab519fcd7caa7ef457dd406162bc14.tar.gz | |
x86/lpc: implement a new Welch windowing function
Old one was written with the assumption only even inputs would be given.
This very messy replacement supports even and odd inputs, and supports
AVX2 for extra speed. The buffers given are usually quite big (4k samples),
so the speedup is worth it.
The new SSE version is still faster than the old inline asm version by 33%.
Also checkasm is provided to make sure this monstrosity works.
This fixes some FATE tests.
Diffstat (limited to 'tests/checkasm/checkasm.h')
| -rw-r--r-- | tests/checkasm/checkasm.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/tests/checkasm/checkasm.h b/tests/checkasm/checkasm.h index 171dd06b47..97e909170f 100644 --- a/tests/checkasm/checkasm.h +++ b/tests/checkasm/checkasm.h @@ -68,6 +68,7 @@ void checkasm_check_idctdsp(void); void checkasm_check_jpeg2000dsp(void); void checkasm_check_llviddsp(void); void checkasm_check_llviddspenc(void); +void checkasm_check_lpc(void); void checkasm_check_motion(void); void checkasm_check_nlmeans(void); void checkasm_check_opusdsp(void); |