Also correct autocorrelation for yet undefined SBR_LOW_POWER. Unify FIXED_POINT and FLOAT implementation.

git-svn-id: svn://svn.rockbox.org/rockbox/trunk@28083 a1c6a512-1295-4272-9138-f99709370657

1 files changed, 36 insertions, 102 deletions

diff --git a/apps/codecs/libfaad/sbr_hfgen.c b/apps/codecs/libfaad/sbr_hfgen.c
index 5c346f1..2f9583c 100644
--- a/apps/codecs/libfaad/sbr_hfgen.c
+++ b/apps/codecs/libfaad/sbr_hfgen.c
@@ -196,85 +196,65 @@ static void auto_correlation(sbr_info *sbr, acorr_coef *ac,
     uint8_t offset = sbr->tHFAdj;
 #ifdef FIXED_POINT
     const real_t rel = FRAC_CONST(0.999999); // 1 / (1 + 1e-6f);
-    uint32_t exp = 2;
-#else
-    const real_t rel = 1 / (1 + 1e-6f);
-#endif
-
-
-#ifdef FIXED_POINT
     /* A pre-shift of >>2 is needed to avoid overflow when multiply-adding
      * the FRACT-variables buffer -- FRACT part is 31 bits. After the
      * calculation has been finished the result 'ac.det' needs to be 
      * post-shifted by <<(4*exp). */
+    const uint32_t exp = 2;
+#else
+    const real_t rel = 1 / (1 + 1e-6f);
+    const uint32_t exp = 0;
+#endif
 
     for (j = offset; j < len + offset; j++)
     {
-        real_t buf_j   = (QMF_RE(buffer[j  ][bd]))>>exp);
-        real_t buf_j_1 = (QMF_RE(buffer[j-1][bd]))>>exp);
-        real_t buf_j_2 = (QMF_RE(buffer[j-2][bd]))>>exp);
-
-        /* normalisation with rounding */
-        r01 += MUL_R(buf_j  , buf_j_1);
-        r02 += MUL_R(buf_j  , buf_j_2);
-        r11 += MUL_R(buf_j_1, buf_j_1);
-    }
-    tmp1 = (QMF_RE(buffer[len+offset-1][bd]))>>exp;
-    tmp2 = (QMF_RE(buffer[    offset-1][bd]))>>exp;
-    RE(ac->r12) = r01 - MUL_R(tmp1, tmp1) + MUL_R(tmp2, tmp2);
+        real_t buf_j   = QMF_RE(buffer[j  ][bd]) >> exp;
+        real_t buf_j_1 = QMF_RE(buffer[j-1][bd]) >> exp;
+        real_t buf_j_2 = QMF_RE(buffer[j-2][bd]) >> exp;
 
-    tmp1 = (QMF_RE(buffer[len+offset-2][bd]))>>exp;
-    tmp2 = (QMF_RE(buffer[    offset-2][bd]))>>exp;
-    RE(ac->r22) = r11 - MUL_R(tmp1, tmp1) + MUL_R(tmp2, tmp2);
-#else
-    for (j = offset; j < len + offset; j++)
-    {
-        r01 += QMF_RE(buffer[j  ][bd]) * QMF_RE(buffer[j-1][bd]);
-        r02 += QMF_RE(buffer[j  ][bd]) * QMF_RE(buffer[j-2][bd]);
-        r11 += QMF_RE(buffer[j-1][bd]) * QMF_RE(buffer[j-1][bd]);
+        r01 += MUL_F(buf_j  , buf_j_1);
+        r02 += MUL_F(buf_j  , buf_j_2);
+        r11 += MUL_F(buf_j_1, buf_j_1);
     }
-    tmp1 = (QMF_RE(buffer[len+offset-1][bd]));
-    tmp2 = (QMF_RE(buffer[    offset-1][bd]));
-    RE(ac->r12) = r01 - tmp1*tmp1 + tmp2*tmp2;
+    tmp1 = QMF_RE(buffer[len+offset-1][bd]) >> exp;
+    tmp2 = QMF_RE(buffer[    offset-1][bd]) >> exp;
+    RE(ac->r12) = r01 - MUL_F(tmp1, tmp1) + MUL_F(tmp2, tmp2);
 
-    tmp1 = (QMF_RE(buffer[len+offset-2][bd]));
-    tmp2 = (QMF_RE(buffer[    offset-2][bd]));
-    RE(ac->r22) = r11 - tmp1*tmp1 + tmp2*tmp2;
-#endif
+    tmp1 = QMF_RE(buffer[len+offset-2][bd]) >> exp;
+    tmp2 = QMF_RE(buffer[    offset-2][bd]) >> exp;
+    RE(ac->r22) = r11 - MUL_F(tmp1, tmp1) + MUL_F(tmp2, tmp2);
     RE(ac->r01) = r01;
     RE(ac->r02) = r02;
     RE(ac->r11) = r11;
 
-    ac->det = MUL_R(RE(ac->r11), RE(ac->r22)) - MUL_F(MUL_R(RE(ac->r12), RE(ac->r12)), rel);
-#ifdef FIXED_POINT
-    ac->det <<= (4*exp);
-#endif
+    ac->det = MUL_F(RE(ac->r11), RE(ac->r22)) - MUL_F(MUL_F(RE(ac->r12), RE(ac->r12)), rel);
+    ac->det <<= (4*exp); /* Post-shift as described above. */
 }
 #else
 static void auto_correlation(sbr_info *sbr, acorr_coef *ac, qmf_t buffer[MAX_NTSRHFG][64],
                              uint8_t bd, uint8_t len)
 {
     real_t r01r = 0, r01i = 0, r02r = 0, r02i = 0, r11r = 0;
-    real_t temp1_r, temp1_i, temp2_r, temp2_i, temp3_r, temp3_i, temp4_r, temp4_i, temp5_r, temp5_i;
-#ifdef FIXED_POINT
-    const real_t rel = FRAC_CONST(0.999999); // 1 / (1 + 1e-6f);
-    uint32_t exp = 2;
-#else
-    const real_t rel = 1 / (1 + 1e-6f);
-#endif
+    real_t temp1_r, temp1_i, temp2_r, temp2_i, temp3_r, temp3_i;
+    real_t temp4_r, temp4_i, temp5_r, temp5_i;
     int8_t j;
     uint8_t offset = sbr->tHFAdj;
-
 #ifdef FIXED_POINT
+    const real_t rel = FRAC_CONST(0.999999); // 1 / (1 + 1e-6f);
     /* A pre-shift of >>2 is needed to avoid overflow when multiply-adding
      * the FRACT-variables buffer -- FRACT part is 31 bits. After the
      * calculation has been finished the result 'ac.det' needs to be 
      * post-shifted by <<(4*exp). */
+    const uint32_t exp = 2;
+#else
+    const real_t rel = 1 / (1 + 1e-6f);
+    const uint32_t exp = 0;
+#endif
 
-    temp2_r = (QMF_RE(buffer[offset-2][bd])) >> exp;
-    temp2_i = (QMF_IM(buffer[offset-2][bd])) >> exp;
-    temp3_r = (QMF_RE(buffer[offset-1][bd])) >> exp;
-    temp3_i = (QMF_IM(buffer[offset-1][bd])) >> exp;
+    temp2_r = QMF_RE(buffer[offset-2][bd]) >> exp;
+    temp2_i = QMF_IM(buffer[offset-2][bd]) >> exp;
+    temp3_r = QMF_RE(buffer[offset-1][bd]) >> exp;
+    temp3_i = QMF_IM(buffer[offset-1][bd]) >> exp;
     // Save these because they are needed after loop
     temp4_r = temp2_r;
     temp4_i = temp2_i;
@@ -287,8 +267,8 @@ static void auto_correlation(sbr_info *sbr, acorr_coef *ac, qmf_t buffer[MAX_NTS
         temp1_i = temp2_i;
         temp2_r = temp3_r;
         temp2_i = temp3_i;
-        temp3_r = (QMF_RE(buffer[j][bd])) >> exp;
-        temp3_i = (QMF_IM(buffer[j][bd])) >> exp;
+        temp3_r = QMF_RE(buffer[j][bd]) >> exp;
+        temp3_i = QMF_IM(buffer[j][bd]) >> exp;
         r01r += MUL_F(temp3_r, temp2_r) + MUL_F(temp3_i, temp2_i);
         r01i += MUL_F(temp3_i, temp2_r) - MUL_F(temp3_r, temp2_i);
         r02r += MUL_F(temp3_r, temp1_r) + MUL_F(temp3_i, temp1_i);
@@ -302,49 +282,15 @@ static void auto_correlation(sbr_info *sbr, acorr_coef *ac, qmf_t buffer[MAX_NTS
                          (MUL_F(temp5_i, temp4_r) - MUL_F(temp5_r, temp4_i));
     RE(ac->r22) = r11r - (MUL_F(temp2_r, temp2_r) + MUL_F(temp2_i, temp2_i)) +
                          (MUL_F(temp4_r, temp4_r) + MUL_F(temp4_i, temp4_i));
-#else
-    temp2_r = QMF_RE(buffer[offset-2][bd]);
-    temp2_i = QMF_IM(buffer[offset-2][bd]);
-    temp3_r = QMF_RE(buffer[offset-1][bd]);
-    temp3_i = QMF_IM(buffer[offset-1][bd]);
-    // Save these because they are needed after loop
-    temp4_r = temp2_r;
-    temp4_i = temp2_i;
-    temp5_r = temp3_r;
-    temp5_i = temp3_i;
-
-    for (j = offset; j < len + offset; j++)
-    {
-        temp1_r = temp2_r;
-        temp1_i = temp2_i;
-        temp2_r = temp3_r;
-        temp2_i = temp3_i;
-        temp3_r = QMF_RE(buffer[j][bd]);
-        temp3_i = QMF_IM(buffer[j][bd]);
-        r01r += temp3_r * temp2_r + temp3_i * temp2_i;
-        r01i += temp3_i * temp2_r - temp3_r * temp2_i;
-        r02r += temp3_r * temp1_r + temp3_i * temp1_i;
-        r02i += temp3_i * temp1_r - temp3_r * temp1_i;
-        r11r += temp2_r * temp2_r + temp2_i * temp2_i;
-    }
-
-    RE(ac->r12) = r01r - (temp3_r * temp2_r + temp3_i * temp2_i) +
-                         (temp5_r * temp4_r + temp5_i * temp4_i);
-    IM(ac->r12) = r01i - (temp3_i * temp2_r - temp3_r * temp2_i) +
-                         (temp5_i * temp4_r - temp5_r * temp4_i);
-    RE(ac->r22) = r11r - (temp2_r * temp2_r + temp2_i * temp2_i) +
-                         (temp4_r * temp4_r + temp4_i * temp4_i);
-#endif
     RE(ac->r01) = r01r;
     IM(ac->r01) = r01i;
     RE(ac->r02) = r02r;
     IM(ac->r02) = r02i;
     RE(ac->r11) = r11r;
 
-    ac->det = MUL_F(RE(ac->r11), RE(ac->r22)) - MUL_F(rel, (MUL_F(RE(ac->r12), RE(ac->r12)) + MUL_F(IM(ac->r12), IM(ac->r12))));
-#ifdef FIXED_POINT
-    ac->det <<= (4*exp);
-#endif
+    ac->det = MUL_F(RE(ac->r11), RE(ac->r22)) - MUL_F((MUL_F(RE(ac->r12), RE(ac->r12)) + MUL_F(IM(ac->r12), IM(ac->r12))), rel);
+    ac->det <<= (4*exp); /* Post-shift as described above. */
+
 }
 #endif
 
@@ -363,17 +309,11 @@ static void calc_prediction_coef(sbr_info *sbr, qmf_t Xlow[MAX_NTSRHFG][64],
         RE(alpha_1[k]) = 0;
         IM(alpha_1[k]) = 0;
     } else {
-#ifdef FIXED_POINT
         mul = DIV_R(REAL_CONST(1.0), ac.det);
         tmp = (MUL_R(RE(ac.r01), RE(ac.r12)) - MUL_R(IM(ac.r01), IM(ac.r12)) - MUL_R(RE(ac.r02), RE(ac.r11)));
         RE(alpha_1[k]) = MUL_R(tmp, mul);
         tmp = (MUL_R(IM(ac.r01), RE(ac.r12)) + MUL_R(RE(ac.r01), IM(ac.r12)) - MUL_R(IM(ac.r02), RE(ac.r11)));
         IM(alpha_1[k]) = MUL_R(tmp, mul);
-#else
-        mul = REAL_CONST(1.0) / ac.det;
-        RE(alpha_1[k]) = (MUL_R(RE(ac.r01), RE(ac.r12)) - MUL_R(IM(ac.r01), IM(ac.r12)) - MUL_R(RE(ac.r02), RE(ac.r11))) * mul;
-        IM(alpha_1[k]) = (MUL_R(IM(ac.r01), RE(ac.r12)) + MUL_R(RE(ac.r01), IM(ac.r12)) - MUL_R(IM(ac.r02), RE(ac.r11))) * mul;
-#endif
     }
 
     if (RE(ac.r11) == 0)
@@ -381,17 +321,11 @@ static void calc_prediction_coef(sbr_info *sbr, qmf_t Xlow[MAX_NTSRHFG][64],
         RE(alpha_0[k]) = 0;
         IM(alpha_0[k]) = 0;
     } else {
-#ifdef FIXED_POINT
         mul = DIV_R(REAL_CONST(1.0), RE(ac.r11));
         tmp = -(RE(ac.r01) + MUL_R(RE(alpha_1[k]), RE(ac.r12)) + MUL_R(IM(alpha_1[k]), IM(ac.r12)));
         RE(alpha_0[k]) = MUL_R(tmp, mul);
         tmp = -(IM(ac.r01) + MUL_R(IM(alpha_1[k]), RE(ac.r12)) - MUL_R(RE(alpha_1[k]), IM(ac.r12)));
         IM(alpha_0[k]) = MUL_R(tmp, mul);
-#else
-        tmp = 1.0f / RE(ac.r11);
-        RE(alpha_0[k]) = -(RE(ac.r01) + MUL_R(RE(alpha_1[k]), RE(ac.r12)) + MUL_R(IM(alpha_1[k]), IM(ac.r12))) * tmp;
-        IM(alpha_0[k]) = -(IM(ac.r01) + MUL_R(IM(alpha_1[k]), RE(ac.r12)) - MUL_R(RE(alpha_1[k]), IM(ac.r12))) * tmp;
-#endif
     }
 
     if ((MUL_R(RE(alpha_0[k]),RE(alpha_0[k])) + MUL_R(IM(alpha_0[k]),IM(alpha_0[k])) >= REAL_CONST(16)) ||