shithub: opus

Info  •  Files  •  Log  •  Branches

Makes dual_inner_prod() more generic to increase its use

--- a/celt/bands.c

+++ b/celt/bands.c

@@ -41,6 +41,7 @@

 #include "mathops.h"

 #include "rate.h"

 #include "quant_bands.h"

+#include "pitch.h"

 int hysteresis_decision(opus_val16 val, const opus_val16 *thresholds, const opus_val16 *hysteresis, int N, int prev)

 {

@@ -383,11 +384,7 @@

    opus_val32 t, lgain, rgain;

    /* Compute the norm of X+Y and X-Y as |X|^2 + |Y|^2 +/- sum(xy) */

-   for (j=0;j<N;j++)

-   {

-      xp = MAC16_16(xp, X[j], Y[j]);

-      side = MAC16_16(side, Y[j], Y[j]);

-   }

+   dual_inner_prod(Y, X, Y, N, &xp, &side);

    /* Compensating for the mid normalization */

    xp = MULT16_32_Q15(mid, xp);

    /* mid and side are in Q15, not Q14 like X and Y */

--- a/celt/pitch.c

+++ b/celt/pitch.c

@@ -394,20 +394,6 @@

    RESTORE_STACK;

 }

-#ifndef OVERRIDE_DUAL_INNER_PROD

-static opus_val32 dual_inner_prod(opus_val16 *x, opus_val16 *y01, opus_val16 *y02, int N)

-{

-   int i;

-   opus_val32 xy=0;

-   for (i=0;i<N;i++)

-   {

-      xy = MAC16_16(xy, x[i], y01[i]);

-      xy = MAC16_16(xy, x[i], y02[i]);

-   }

-   return xy;

-}

-#endif

-

 static const int second_check[16] = {0, 0, 3, 2, 3, 2, 5, 2, 3, 2, 3, 2, 5, 2, 3, 2};

 opus_val16 remove_doubling(opus_val16 *x, int maxperiod, int minperiod,

       int N, int *T0_, int prev_period, opus_val16 prev_gain)

@@ -415,7 +401,7 @@

    int k, i, T, T0;

    opus_val16 g, g0;

    opus_val16 pg;

-   opus_val32 xy,xx,yy;

+   opus_val32 xy,xx,yy,xy2;

    opus_val32 xcorr[3];

    opus_val32 best_xy, best_yy;

    int offset;

@@ -435,12 +421,7 @@

    T = T0 = *T0_;

    ALLOC(yy_lookup, maxperiod+1, opus_val32);

-   xy=xx=0;

-   for (i=0;i<N;i++)

-   {

-      xx = MAC16_16(xx, x[i], x[i]);

-      xy = MAC16_16(xy, x[i], x[i-T0]);

-   }

+   dual_inner_prod(x, x, x-T0, N, &xx, &xy);

    yy_lookup[0] = xx;

    yy=xx;

    for (i=1;i<=maxperiod;i++)

@@ -484,7 +465,8 @@

       {

          T1b = (2*second_check[k]*T0+k)/(2*k);

       }

-      xy = dual_inner_prod(x, &x[-T1], &x[-T1b], N);

+      dual_inner_prod(x, &x[-T1], &x[-T1b], N, &xy, &xy2);

+      xy += xy2;

       yy = yy_lookup[T1] + yy_lookup[T1b];

 #ifdef FIXED_POINT

       {

--- a/celt/pitch.h

+++ b/celt/pitch.h

@@ -118,6 +118,23 @@

 }

 #endif /* OVERRIDE_XCORR_KERNEL */

+#ifndef OVERRIDE_DUAL_INNER_PROD

+static inline void dual_inner_prod(const opus_val16 *x, const opus_val16 *y01, const opus_val16 *y02,

+      int N, opus_val32 *xy1, opus_val32 *xy2)

+{

+   int i;

+   opus_val32 xy01=0;

+   opus_val32 xy02=0;

+   for (i=0;i<N;i++)

+   {

+      xy01 = MAC16_16(xy01, x[i], y01[i]);

+      xy02 = MAC16_16(xy02, x[i], y02[i]);

+   }

+   *xy1 = xy01;

+   *xy2 = xy02;

+}

+#endif

+

 #ifdef FIXED_POINT

 opus_val32

 #else

--- a/celt/x86/pitch_sse.h

+++ b/celt/x86/pitch_sse.h

@@ -72,11 +72,11 @@

 }

 #define OVERRIDE_DUAL_INNER_PROD

-static inline opus_val32 dual_inner_prod(const opus_val16 *x, const opus_val16 *y01, const opus_val16 *y02, int N)

+static inline void dual_inner_prod(const opus_val16 *x, const opus_val16 *y01, const opus_val16 *y02,

+      int N, opus_val32 *xy1, opus_val32 *xy2)

 {

    int i;

    __m128 xsum1, xsum2;

-   opus_val32 xy=0;

    xsum1 = _mm_setzero_ps();

    xsum2 = _mm_setzero_ps();

    for (i=0;i<N-3;i+=4)

@@ -87,17 +87,18 @@

       xsum1 = _mm_add_ps(xsum1,_mm_mul_ps(xi, y1i));

       xsum2 = _mm_add_ps(xsum2,_mm_mul_ps(xi, y2i));

    }

-   xsum1 = _mm_add_ps(xsum1,xsum2);

    /* Horizontal sum */

    xsum1 = _mm_add_ps(xsum1, _mm_movehl_ps(xsum1, xsum1));

    xsum1 = _mm_add_ss(xsum1, _mm_shuffle_ps(xsum1, xsum1, 0x55));

-   _mm_store_ss(&xy, xsum1);

+   _mm_store_ss(xy1, xsum1);

+   xsum2 = _mm_add_ps(xsum2, _mm_movehl_ps(xsum2, xsum2));

+   xsum2 = _mm_add_ss(xsum2, _mm_shuffle_ps(xsum2, xsum2, 0x55));

+   _mm_store_ss(xy2, xsum2);

    for (;i<N;i++)

    {

-      xy = MAC16_16(xy, x[i], y01[i]);

-      xy = MAC16_16(xy, x[i], y02[i]);

+      *xy1 = MAC16_16(*xy1, x[i], y01[i]);

+      *xy2 = MAC16_16(*xy2, x[i], y02[i]);

    }

-   return xy;

 }

 #define OVERRIDE_COMB_FILTER_CONST