diff options
| author | Rostislav Pehlivanov <atomnuker@gmail.com> | 2016-10-08 17:59:14 +0300 |
|---|---|---|
| committer | Rostislav Pehlivanov <atomnuker@gmail.com> | 2016-10-18 23:41:18 +0300 |
| commit | d2ae5f77c61a29c3c63cc4c41c74ccfca4167649 (patch) | |
| tree | bd615f22e6807e025da052c411e12e9a7be0852e /libavcodec/aacenc_pred.c | |
| parent | 3b02f6dd7be880fd6c1bcaf2fd0c1314dcee7aa6 (diff) | |
aacenc: add SIMD optimizations for abs_pow34 and quantization
Performance improvements:
quant_bands:
with: 681 decicycles in quant_bands, 8388453 runs, 155 skips
without: 1190 decicycles in quant_bands, 8388386 runs, 222 skips
Around 42% for the function
Twoloop coder:
abs_pow34:
with/without: 7.82s/8.17s
Around 4% for the entire encoder
Both:
with/without: 7.15s/8.17s
Around 12% for the entire encoder
Fast coder:
abs_pow34:
with/without: 3.40s/3.77s
Around 10% for the entire encoder
Both:
with/without: 3.02s/3.77s
Around 20% faster for the entire encoder
Signed-off-by: Rostislav Pehlivanov <atomnuker@gmail.com>
Tested-by: Michael Niedermayer <michael@niedermayer.cc>
Reviewed-by: James Almer <jamrial@gmail.com>
Diffstat (limited to 'libavcodec/aacenc_pred.c')
| -rw-r--r-- | libavcodec/aacenc_pred.c | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/libavcodec/aacenc_pred.c b/libavcodec/aacenc_pred.c index e77a3de987..d111192f06 100644 --- a/libavcodec/aacenc_pred.c +++ b/libavcodec/aacenc_pred.c @@ -270,7 +270,7 @@ void ff_aac_search_for_pred(AACEncContext *s, SingleChannelElement *sce) continue; /* Normal coefficients */ - abs_pow34_v(O34, &sce->coeffs[start_coef], num_coeffs); + s->abs_pow34(O34, &sce->coeffs[start_coef], num_coeffs); dist1 = quantize_and_encode_band_cost(s, NULL, &sce->coeffs[start_coef], NULL, O34, num_coeffs, sce->sf_idx[sfb], cb_n, s->lambda / band->threshold, INFINITY, &cost1, NULL, 0); @@ -279,7 +279,7 @@ void ff_aac_search_for_pred(AACEncContext *s, SingleChannelElement *sce) /* Encoded coefficients - needed for #bits, band type and quant. error */ for (i = 0; i < num_coeffs; i++) SENT[i] = sce->coeffs[start_coef + i] - sce->prcoeffs[start_coef + i]; - abs_pow34_v(S34, SENT, num_coeffs); + s->abs_pow34(S34, SENT, num_coeffs); if (cb_n < RESERVED_BT) cb_p = av_clip(find_min_book(find_max_val(1, num_coeffs, S34), sce->sf_idx[sfb]), cb_min, cb_max); else @@ -291,7 +291,7 @@ void ff_aac_search_for_pred(AACEncContext *s, SingleChannelElement *sce) /* Reconstructed coefficients - needed for distortion measurements */ for (i = 0; i < num_coeffs; i++) sce->prcoeffs[start_coef + i] += QERR[i] != 0.0f ? (sce->prcoeffs[start_coef + i] - QERR[i]) : 0.0f; - abs_pow34_v(P34, &sce->prcoeffs[start_coef], num_coeffs); + s->abs_pow34(P34, &sce->prcoeffs[start_coef], num_coeffs); if (cb_n < RESERVED_BT) cb_p = av_clip(find_min_book(find_max_val(1, num_coeffs, P34), sce->sf_idx[sfb]), cb_min, cb_max); else |