Mercurial > libavcodec.hg
diff dsputil.h @ 5737:efa3c1f9259a libavcodec
sse2 version of compute_autocorr().
4x faster than c (somehow, even though doubles only allow 2x simd).
overal flac encoding: 15-50% faster on core2, 4-11% on k8, 3-13% on p4.
author | lorenm |
---|---|
date | Sat, 29 Sep 2007 22:31:18 +0000 |
parents | d7970c9e3049 |
children | 09f99af1db40 |
line wrap: on
line diff
--- a/dsputil.h Sat Sep 29 15:20:22 2007 +0000 +++ b/dsputil.h Sat Sep 29 22:31:18 2007 +0000 @@ -328,6 +328,8 @@ /* assume len is a multiple of 4, and arrays are 16-byte aligned */ void (*vorbis_inverse_coupling)(float *mag, float *ang, int blocksize); + /* no alignment needed */ + void (*flac_compute_autocorr)(const int32_t *data, int len, int lag, double *autoc); /* assume len is a multiple of 8, and arrays are 16-byte aligned */ void (*vector_fmul)(float *dst, const float *src, int len); void (*vector_fmul_reverse)(float *dst, const float *src0, const float *src1, int len);