diff options
| author | Loren Merritt <lorenm@u.washington.edu> | 2011-02-12 11:48:16 +0000 |
|---|---|---|
| committer | Janne Grunau <janne-ffmpeg@jannau.net> | 2011-02-13 15:36:39 +0100 |
| commit | e6b1ed693ae4098e6b9eabf938fc31ec0b09b120 (patch) | |
| tree | 8afa645b29c35cde5a286b29cbab58fbbed4c1ed /libavcodec/fft.c | |
| parent | 3c33c0e26325f189dbf7f212f8e9042feb83bfb0 (diff) | |
FFT: factor a shuffle out of the inner loop and merge it into fft_permute.
6% faster SSE FFT on Conroe, 2.5% on Penryn.
Signed-off-by: Janne Grunau <janne-ffmpeg@jannau.net>
Diffstat (limited to 'libavcodec/fft.c')
| -rw-r--r-- | libavcodec/fft.c | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/libavcodec/fft.c b/libavcodec/fft.c index 3fd4d279cb..eade76a516 100644 --- a/libavcodec/fft.c +++ b/libavcodec/fft.c @@ -97,6 +97,7 @@ av_cold int ff_fft_init(FFTContext *s, int nbits, int inverse) if (!s->tmp_buf) goto fail; s->inverse = inverse; + s->fft_permutation = FF_FFT_PERM_DEFAULT; s->fft_permute = ff_fft_permute_c; s->fft_calc = ff_fft_calc_c; @@ -113,8 +114,12 @@ av_cold int ff_fft_init(FFTContext *s, int nbits, int inverse) for(j=4; j<=nbits; j++) { ff_init_ff_cos_tabs(j); } - for(i=0; i<n; i++) - s->revtab[-split_radix_permutation(i, n, s->inverse) & (n-1)] = i; + for(i=0; i<n; i++) { + int j = i; + if (s->fft_permutation == FF_FFT_PERM_SWAP_LSBS) + j = (j&~3) | ((j>>1)&1) | ((j<<1)&2); + s->revtab[-split_radix_permutation(i, n, s->inverse) & (n-1)] = j; + } return 0; fail: |