[FFmpeg-cvslog] opusenc: use float_dsp for non-transient windowing

[Rostislav Pehlivanov]

ffmpeg | branch: master | Rostislav Pehlivanov <atomnuker at gmail.com> | Fri Jul 14 04:50:07 2017 +0100| [91b27b83939a73b7313440f79512549b5b8ff7c1] | committer: Rostislav Pehlivanov

opusenc: use float_dsp for non-transient windowing

Also fixes transient windowing

Signed-off-by: Rostislav Pehlivanov <atomnuker at gmail.com>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=91b27b83939a73b7313440f79512549b5b8ff7c1
---

 libavcodec/opusenc.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/libavcodec/opusenc.c b/libavcodec/opusenc.c
index 6cefd33884..547c62cf6b 100644
--- a/libavcodec/opusenc.c
+++ b/libavcodec/opusenc.c
@@ -208,7 +208,7 @@ static void celt_apply_preemph_filter(OpusEncContext *s, CeltFrame *f)
 static void celt_frame_mdct(OpusEncContext *s, CeltFrame *f)
 {
     int i, t, ch;
-    float *win = s->scratch;
+    float *win = s->scratch, *temp = s->scratch + 1920;
 
     if (f->transient) {
         for (ch = 0; ch < f->channels; ch++) {
@@ -216,9 +216,9 @@ static void celt_frame_mdct(OpusEncContext *s, CeltFrame *f)
             float *src1 = b->overlap;
             for (t = 0; t < f->blocks; t++) {
                 float *src2 = &b->samples[CELT_OVERLAP*t];
-                s->dsp->vector_fmul(win, src1, ff_celt_window, CELT_OVERLAP);
+                s->dsp->vector_fmul(win, src1, ff_celt_window, 128);
                 s->dsp->vector_fmul_reverse(&win[CELT_OVERLAP], src2,
-                                            ff_celt_window - 8, CELT_OVERLAP + 8);
+                                            ff_celt_window - 8, 128);
                 src1 = src2;
                 s->mdct[0]->mdct(s->mdct[0], b->coeffs + t, win, f->blocks);
             }
@@ -226,21 +226,21 @@ static void celt_frame_mdct(OpusEncContext *s, CeltFrame *f)
     } else {
         int blk_len = OPUS_BLOCK_SIZE(f->size), wlen = OPUS_BLOCK_SIZE(f->size + 1);
         int rwin = blk_len - CELT_OVERLAP, lap_dst = (wlen - blk_len - CELT_OVERLAP) >> 1;
+        memset(win, 0, wlen*sizeof(float));
         for (ch = 0; ch < f->channels; ch++) {
             CeltBlock *b = &f->block[ch];
 
-            memset(win, 0, wlen*sizeof(float));
+            /* Overlap */
+            s->dsp->vector_fmul(temp, b->overlap, ff_celt_window, 128);
+            memcpy(win + lap_dst, temp, CELT_OVERLAP*sizeof(float));
 
+            /* Samples, flat top window */
             memcpy(&win[lap_dst + CELT_OVERLAP], b->samples, rwin*sizeof(float));
 
-            /* Alignment fucks me over */
-            //s->dsp->vector_fmul(&dst[lap_dst], b->overlap, ff_celt_window, CELT_OVERLAP);
-            //s->dsp->vector_fmul_reverse(&dst[lap_dst + blk_len - CELT_OVERLAP], b->samples, ff_celt_window, CELT_OVERLAP);
-
-            for (i = 0; i < CELT_OVERLAP; i++) {
-                win[lap_dst           + i] = b->overlap[i]       *ff_celt_window[i];
-                win[lap_dst + blk_len + i] = b->samples[rwin + i]*ff_celt_window[CELT_OVERLAP - i - 1];
-            }
+            /* Samples, windowed */
+            s->dsp->vector_fmul_reverse(temp, b->samples + rwin,
+                                        ff_celt_window - 8, 128);
+            memcpy(win + lap_dst + blk_len, temp, CELT_OVERLAP*sizeof(float));
 
             s->mdct[f->size]->mdct(s->mdct[f->size], b->coeffs, win, 1);
         }