summaryrefslogtreecommitdiff
path: root/vp9/encoder/vp9_dct.c
diff options
context:
space:
mode:
Diffstat (limited to 'vp9/encoder/vp9_dct.c')
-rw-r--r--vp9/encoder/vp9_dct.c935
1 files changed, 134 insertions, 801 deletions
diff --git a/vp9/encoder/vp9_dct.c b/vp9/encoder/vp9_dct.c
index f344ddfb7..97794e3f6 100644
--- a/vp9/encoder/vp9_dct.c
+++ b/vp9/encoder/vp9_dct.c
@@ -564,22 +564,6 @@ void vp9_short_fdct8x8_c(short *InputData, short *OutputData, int pitch) {
}
#endif
-void vp9_short_fhaar2x2_c(short *input, short *output, int pitch) {
- /* [1 1; 1 -1] orthogonal transform */
- /* use position: 0,1, 4, 8 */
- int i;
- short *ip1 = input;
- short *op1 = output;
- for (i = 0; i < 16; i++) {
- op1[i] = 0;
- }
-
- op1[0] = (ip1[0] + ip1[1] + ip1[4] + ip1[8] + 1) >> 1;
- op1[1] = (ip1[0] - ip1[1] + ip1[4] - ip1[8]) >> 1;
- op1[4] = (ip1[0] + ip1[1] - ip1[4] - ip1[8]) >> 1;
- op1[8] = (ip1[0] - ip1[1] - ip1[4] + ip1[8]) >> 1;
-}
-
/* For test */
#define TEST_INT 1
#if TEST_INT
@@ -912,87 +896,7 @@ void vp9_short_fdct8x8_c(int16_t *input, int16_t *output, int pitch) {
}
#endif
-void vp9_short_walsh4x4_c(short *input, short *output, int pitch) {
- int i;
- int a1, b1, c1, d1;
- short *ip = input;
- short *op = output;
- int pitch_short = pitch >> 1;
-
- for (i = 0; i < 4; i++) {
- a1 = ip[0 * pitch_short] + ip[3 * pitch_short];
- b1 = ip[1 * pitch_short] + ip[2 * pitch_short];
- c1 = ip[1 * pitch_short] - ip[2 * pitch_short];
- d1 = ip[0 * pitch_short] - ip[3 * pitch_short];
-
- op[0] = (a1 + b1 + 1) >> 1;
- op[4] = (c1 + d1) >> 1;
- op[8] = (a1 - b1) >> 1;
- op[12] = (d1 - c1) >> 1;
-
- ip++;
- op++;
- }
- ip = output;
- op = output;
-
- for (i = 0; i < 4; i++) {
- a1 = ip[0] + ip[3];
- b1 = ip[1] + ip[2];
- c1 = ip[1] - ip[2];
- d1 = ip[0] - ip[3];
-
- op[0] = (a1 + b1 + 1) >> 1;
- op[1] = (c1 + d1) >> 1;
- op[2] = (a1 - b1) >> 1;
- op[3] = (d1 - c1) >> 1;
-
- ip += 4;
- op += 4;
- }
-}
-
#if CONFIG_LOSSLESS
-void vp9_short_walsh4x4_lossless_c(short *input, short *output, int pitch) {
- int i;
- int a1, b1, c1, d1;
- short *ip = input;
- short *op = output;
- int pitch_short = pitch >> 1;
-
- for (i = 0; i < 4; i++) {
- a1 = (ip[0 * pitch_short] + ip[3 * pitch_short]) >> Y2_WHT_UPSCALE_FACTOR;
- b1 = (ip[1 * pitch_short] + ip[2 * pitch_short]) >> Y2_WHT_UPSCALE_FACTOR;
- c1 = (ip[1 * pitch_short] - ip[2 * pitch_short]) >> Y2_WHT_UPSCALE_FACTOR;
- d1 = (ip[0 * pitch_short] - ip[3 * pitch_short]) >> Y2_WHT_UPSCALE_FACTOR;
-
- op[0] = (a1 + b1 + 1) >> 1;
- op[4] = (c1 + d1) >> 1;
- op[8] = (a1 - b1) >> 1;
- op[12] = (d1 - c1) >> 1;
-
- ip++;
- op++;
- }
- ip = output;
- op = output;
-
- for (i = 0; i < 4; i++) {
- a1 = ip[0] + ip[3];
- b1 = ip[1] + ip[2];
- c1 = ip[1] - ip[2];
- d1 = ip[0] - ip[3];
-
- op[0] = ((a1 + b1 + 1) >> 1) << Y2_WHT_UPSCALE_FACTOR;
- op[1] = ((c1 + d1) >> 1) << Y2_WHT_UPSCALE_FACTOR;
- op[2] = ((a1 - b1) >> 1) << Y2_WHT_UPSCALE_FACTOR;
- op[3] = ((d1 - c1) >> 1) << Y2_WHT_UPSCALE_FACTOR;
-
- ip += 4;
- op += 4;
- }
-}
-
void vp9_short_walsh4x4_x8_c(short *input, short *output, int pitch) {
int i;
int a1, b1, c1, d1;
@@ -1250,6 +1154,9 @@ void vp9_short_fdct16x16_c(short *input, short *out, int pitch) {
}
#else
+
+#define NEW_FDCT16 1
+#if !NEW_FDCT16
static const int16_t C1 = 16305;
static const int16_t C2 = 16069;
static const int16_t C3 = 15679;
@@ -1469,9 +1376,138 @@ void vp9_short_fdct16x16_c(int16_t *input, int16_t *out, int pitch) {
}
#undef RIGHT_SHIFT
#undef ROUNDING
-#endif
-#if !CONFIG_DWTDCTHYBRID
+#else
+// Rewrote to use same algorithm as others.
+static void fdct16_1d(int16_t input[16], int16_t output[16]) {
+ int16_t step[16];
+ int temp1, temp2;
+
+ // step 1
+ step[ 0] = input[0] + input[15];
+ step[ 1] = input[1] + input[14];
+ step[ 2] = input[2] + input[13];
+ step[ 3] = input[3] + input[12];
+ step[ 4] = input[4] + input[11];
+ step[ 5] = input[5] + input[10];
+ step[ 6] = input[6] + input[ 9];
+ step[ 7] = input[7] + input[ 8];
+ step[ 8] = input[7] - input[ 8];
+ step[ 9] = input[6] - input[ 9];
+ step[10] = input[5] - input[10];
+ step[11] = input[4] - input[11];
+ step[12] = input[3] - input[12];
+ step[13] = input[2] - input[13];
+ step[14] = input[1] - input[14];
+ step[15] = input[0] - input[15];
+
+ fdct8_1d(step, step);
+
+ // step 2
+ output[8] = step[8];
+ output[9] = step[9];
+ temp1 = (-step[10] + step[13]) * cospi_16_64;
+ temp2 = (-step[11] + step[12]) * cospi_16_64;
+ output[10] = dct_const_round_shift(temp1);
+ output[11] = dct_const_round_shift(temp2);
+ temp1 = (step[11] + step[12]) * cospi_16_64;
+ temp2 = (step[10] + step[13]) * cospi_16_64;
+ output[12] = dct_const_round_shift(temp1);
+ output[13] = dct_const_round_shift(temp2);
+ output[14] = step[14];
+ output[15] = step[15];
+
+ // step 3
+ step[ 8] = output[8] + output[11];
+ step[ 9] = output[9] + output[10];
+ step[ 10] = output[9] - output[10];
+ step[ 11] = output[8] - output[11];
+ step[ 12] = -output[12] + output[15];
+ step[ 13] = -output[13] + output[14];
+ step[ 14] = output[13] + output[14];
+ step[ 15] = output[12] + output[15];
+
+ // step 4
+ output[8] = step[8];
+ temp1 = -step[9] * cospi_8_64 + step[14] * cospi_24_64;
+ temp2 = -step[10] * cospi_24_64 - step[13] * cospi_8_64;
+ output[9] = dct_const_round_shift(temp1);
+ output[10] = dct_const_round_shift(temp2);
+ output[11] = step[11];
+ output[12] = step[12];
+ temp1 = -step[10] * cospi_8_64 + step[13] * cospi_24_64;
+ temp2 = step[9] * cospi_24_64 + step[14] * cospi_8_64;
+ output[13] = dct_const_round_shift(temp1);
+ output[14] = dct_const_round_shift(temp2);
+ output[15] = step[15];
+
+ // step 5
+ step[8] = output[8] + output[9];
+ step[9] = output[8] - output[9];
+ step[10] = -output[10] + output[11];
+ step[11] = output[10] + output[11];
+ step[12] = output[12] + output[13];
+ step[13] = output[12] - output[13];
+ step[14] = -output[14] + output[15];
+ step[15] = output[14] + output[15];
+
+ // step 6
+ output[0] = step[0];
+ output[8] = step[4];
+ output[4] = step[2];
+ output[12] = step[6];
+ output[2] = step[1];
+ output[10] = step[5];
+ output[6] = step[3];
+ output[14] = step[7];
+
+ temp1 = step[8] * cospi_30_64 + step[15] * cospi_2_64;
+ temp2 = step[9] * cospi_14_64 + step[14] * cospi_18_64;
+ output[1] = dct_const_round_shift(temp1);
+ output[9] = dct_const_round_shift(temp2);
+
+ temp1 = step[10] * cospi_22_64 + step[13] * cospi_10_64;
+ temp2 = step[11] * cospi_6_64 + step[12] * cospi_26_64;
+ output[5] = dct_const_round_shift(temp1);
+ output[13] = dct_const_round_shift(temp2);
+
+ temp1 = -step[11] * cospi_26_64 + step[12] * cospi_6_64;
+ temp2 = -step[10] * cospi_10_64 + step[13] * cospi_22_64;
+ output[3] = dct_const_round_shift(temp1);
+ output[11] = dct_const_round_shift(temp2);
+
+ temp1 = -step[9] * cospi_18_64 + step[14] * cospi_14_64;
+ temp2 = -step[8] * cospi_2_64 + step[15] * cospi_30_64;
+ output[7] = dct_const_round_shift(temp1);
+ output[15] = dct_const_round_shift(temp2);
+}
+
+void vp9_short_fdct16x16_c(int16_t *input, int16_t *out, int pitch) {
+ int shortpitch = pitch >> 1;
+ int i, j;
+ int16_t output[256];
+ int16_t temp_in[16], temp_out[16];
+
+ // First transform columns
+ for (i = 0; i < 16; i++) {
+ for (j = 0; j < 16; j++)
+ temp_in[j] = input[j * shortpitch + i];
+ fdct16_1d(temp_in, temp_out);
+ for (j = 0; j < 16; j++)
+ output[j * 16 + i] = temp_out[j];
+ }
+
+ // Then transform rows
+ for (i = 0; i < 16; ++i) {
+ for (j = 0; j < 16; ++j)
+ temp_in[j] = output[j + i * 16];
+ fdct16_1d(temp_in, temp_out);
+ for (j = 0; j < 16; ++j)
+ out[j + i * 16] = temp_out[j];
+ }
+}
+#endif
+#endif
#define TEST_INT_32x32_DCT 1
@@ -2210,706 +2246,3 @@ void vp9_short_fdct32x32_c(int16_t *input, int16_t *out, int pitch) {
}
#endif
-
-#else // CONFIG_DWTDCTHYBRID
-
-#if DWT_TYPE == 53
-
-// Note: block length must be even for this implementation
-static void analysis_53_row(int length, short *x,
- short *lowpass, short *highpass) {
- int n;
- short r, *a, *b;
-
- n = length >> 1;
- b = highpass;
- a = lowpass;
- while (--n) {
- *a++ = (r = *x++) << 1;
- *b++ = *x - ((r + x[1] + 1) >> 1);
- x++;
- }
- *a = (r = *x++) << 1;
- *b = *x - r;
-
- n = length >> 1;
- b = highpass;
- a = lowpass;
- r = *highpass;
- while (n--) {
- *a++ += (r + (*b) + 1) >> 1;
- r = *b++;
- }
-}
-
-static void analysis_53_col(int length, short *x,
- short *lowpass, short *highpass) {
- int n;
- short r, *a, *b;
-
- n = length >> 1;
- b = highpass;
- a = lowpass;
- while (--n) {
- *a++ = (r = *x++);
- *b++ = (((*x) << 1) - (r + x[1]) + 2) >> 2;
- x++;
- }
- *a = (r = *x++);
- *b = (*x - r + 1) >> 1;
-
- n = length >> 1;
- b = highpass;
- a = lowpass;
- r = *highpass;
- while (n--) {
- *a++ += (r + (*b) + 1) >> 1;
- r = *b++;
- }
-}
-
-static void dyadic_analyze_53(int levels, int width, int height,
- short *x, int pitch_x, short *c, int pitch_c) {
- int lv, i, j, nh, nw, hh = height, hw = width;
- short buffer[2 * DWT_MAX_LENGTH];
- for (i = 0; i < height; i++) {
- for (j = 0; j < width; j++) {
- c[i * pitch_c + j] = x[i * pitch_x + j] << DWT_PRECISION_BITS;
- }
- }
- for (lv = 0; lv < levels; lv++) {
- nh = hh;
- hh = (hh + 1) >> 1;
- nw = hw;
- hw = (hw + 1) >> 1;
- if ((nh < 2) || (nw < 2)) return;
- for (i = 0; i < nh; i++) {
- memcpy(buffer, &c[i * pitch_c], nw * sizeof(short));
- analysis_53_row(nw, buffer, &c[i * pitch_c], &c[i * pitch_c] + hw);
- }
- for (j = 0; j < nw; j++) {
- for (i = 0; i < nh; i++)
- buffer[i + nh] = c[i * pitch_c + j];
- analysis_53_col(nh, buffer + nh, buffer, buffer + hh);
- for (i = 0; i < nh; i++)
- c[i * pitch_c + j] = buffer[i];
- }
- }
-}
-
-#elif DWT_TYPE == 26
-
-static void analysis_26_row(int length, short *x,
- short *lowpass, short *highpass) {
- int i, n;
- short r, s, *a, *b;
- a = lowpass;
- b = highpass;
- for (i = length >> 1; i; i--) {
- r = *x++;
- s = *x++;
- *a++ = r + s;
- *b++ = r - s;
- }
- n = length >> 1;
- if (n >= 4) {
- a = lowpass;
- b = highpass;
- r = *lowpass;
- while (--n) {
- *b++ -= (r - a[1] + 4) >> 3;
- r = *a++;
- }
- *b -= (r - *a + 4) >> 3;
- }
-}
-
-static void analysis_26_col(int length, short *x,
- short *lowpass, short *highpass) {
- int i, n;
- short r, s, *a, *b;
- a = lowpass;
- b = highpass;
- for (i = length >> 1; i; i--) {
- r = *x++;
- s = *x++;
- *a++ = (r + s + 1) >> 1;
- *b++ = (r - s + 1) >> 1;
- }
- n = length >> 1;
- if (n >= 4) {
- a = lowpass;
- b = highpass;
- r = *lowpass;
- while (--n) {
- *b++ -= (r - a[1] + 4) >> 3;
- r = *a++;
- }
- *b -= (r - *a + 4) >> 3;
- }
-}
-
-static void dyadic_analyze_26(int levels, int width, int height,
- short *x, int pitch_x, short *c, int pitch_c) {
- int lv, i, j, nh, nw, hh = height, hw = width;
- short buffer[2 * DWT_MAX_LENGTH];
- for (i = 0; i < height; i++) {
- for (j = 0; j < width; j++) {
- c[i * pitch_c + j] = x[i * pitch_x + j] << DWT_PRECISION_BITS;
- }
- }
- for (lv = 0; lv < levels; lv++) {
- nh = hh;
- hh = (hh + 1) >> 1;
- nw = hw;
- hw = (hw + 1) >> 1;
- if ((nh < 2) || (nw < 2)) return;
- for (i = 0; i < nh; i++) {
- memcpy(buffer, &c[i * pitch_c], nw * sizeof(short));
- analysis_26_row(nw, buffer, &c[i * pitch_c], &c[i * pitch_c] + hw);
- }
- for (j = 0; j < nw; j++) {
- for (i = 0; i < nh; i++)
- buffer[i + nh] = c[i * pitch_c + j];
- analysis_26_col(nh, buffer + nh, buffer, buffer + hh);
- for (i = 0; i < nh; i++)
- c[i * pitch_c + j] = buffer[i];
- }
- }
-}
-
-#elif DWT_TYPE == 97
-
-static void analysis_97(int length, double *x,
- double *lowpass, double *highpass) {
- static const double a_predict1 = -1.586134342;
- static const double a_update1 = -0.05298011854;
- static const double a_predict2 = 0.8829110762;
- static const double a_update2 = 0.4435068522;
- static const double s_low = 1.149604398;
- static const double s_high = 1/1.149604398;
- int i;
- double y[DWT_MAX_LENGTH];
- // Predict 1
- for (i = 1; i < length - 2; i += 2) {
- x[i] += a_predict1 * (x[i - 1] + x[i + 1]);
- }
- x[length - 1] += 2 * a_predict1 * x[length - 2];
- // Update 1
- for (i = 2; i < length; i += 2) {
- x[i] += a_update1 * (x[i - 1] + x[i + 1]);
- }
- x[0] += 2 * a_update1 * x[1];
- // Predict 2
- for (i = 1; i < length - 2; i += 2) {
- x[i] += a_predict2 * (x[i - 1] + x[i + 1]);
- }
- x[length - 1] += 2 * a_predict2 * x[length - 2];
- // Update 2
- for (i = 2; i < length; i += 2) {
- x[i] += a_update2 * (x[i - 1] + x[i + 1]);
- }
- x[0] += 2 * a_update2 * x[1];
- memcpy(y, x, sizeof(*y) * length);
- // Scale and pack
- for (i = 0; i < length / 2; i++) {
- lowpass[i] = y[2 * i] * s_low;
- highpass[i] = y[2 * i + 1] * s_high;
- }
-}
-
-static void dyadic_analyze_97(int levels, int width, int height,
- short *x, int pitch_x, short *c, int pitch_c) {
- int lv, i, j, nh, nw, hh = height, hw = width;
- double buffer[2 * DWT_MAX_LENGTH];
- double y[DWT_MAX_LENGTH * DWT_MAX_LENGTH];
- for (i = 0; i < height; i++) {
- for (j = 0; j < width; j++) {
- y[i * DWT_MAX_LENGTH + j] = x[i * pitch_x + j] << DWT_PRECISION_BITS;
- }
- }
- for (lv = 0; lv < levels; lv++) {
- nh = hh;
- hh = (hh + 1) >> 1;
- nw = hw;
- hw = (hw + 1) >> 1;
- if ((nh < 2) || (nw < 2)) return;
- for (i = 0; i < nh; i++) {
- memcpy(buffer, &y[i * DWT_MAX_LENGTH], nw * sizeof(*buffer));
- analysis_97(nw, buffer, &y[i * DWT_MAX_LENGTH],
- &y[i * DWT_MAX_LENGTH] + hw);
- }
- for (j = 0; j < nw; j++) {
- for (i = 0; i < nh; i++)
- buffer[i + nh] = y[i * DWT_MAX_LENGTH + j];
- analysis_97(nh, buffer + nh, buffer, buffer + hh);
- for (i = 0; i < nh; i++)
- c[i * pitch_c + j] = round(buffer[i]);
- }
- }
-}
-
-#endif // DWT_TYPE
-
-// TODO(debargha): Implement the scaling differently so as not to have to
-// use the floating point dct
-static void dct16x16_1d_f(double input[16], double output[16]) {
- static const double C1 = 0.995184726672197;
- static const double C2 = 0.98078528040323;
- static const double C3 = 0.956940335732209;
- static const double C4 = 0.923879532511287;
- static const double C5 = 0.881921264348355;
- static const double C6 = 0.831469612302545;
- static const double C7 = 0.773010453362737;
- static const double C8 = 0.707106781186548;
- static const double C9 = 0.634393284163646;
- static const double C10 = 0.555570233019602;
- static const double C11 = 0.471396736825998;
- static const double C12 = 0.38268343236509;
- static const double C13 = 0.290284677254462;
- static const double C14 = 0.195090322016128;
- static const double C15 = 0.098017140329561;
-
- vp9_clear_system_state(); // Make it simd safe : __asm emms;
- {
- double step[16];
- double intermediate[16];
- double temp1, temp2;
-
- // step 1
- step[ 0] = input[0] + input[15];
- step[ 1] = input[1] + input[14];
- step[ 2] = input[2] + input[13];
- step[ 3] = input[3] + input[12];
- step[ 4] = input[4] + input[11];
- step[ 5] = input[5] + input[10];
- step[ 6] = input[6] + input[ 9];
- step[ 7] = input[7] + input[ 8];
- step[ 8] = input[7] - input[ 8];
- step[ 9] = input[6] - input[ 9];
- step[10] = input[5] - input[10];
- step[11] = input[4] - input[11];
- step[12] = input[3] - input[12];
- step[13] = input[2] - input[13];
- step[14] = input[1] - input[14];
- step[15] = input[0] - input[15];
-
- // step 2
- output[0] = step[0] + step[7];
- output[1] = step[1] + step[6];
- output[2] = step[2] + step[5];
- output[3] = step[3] + step[4];
- output[4] = step[3] - step[4];
- output[5] = step[2] - step[5];
- output[6] = step[1] - step[6];
- output[7] = step[0] - step[7];
-
- temp1 = step[ 8]*C7;
- temp2 = step[15]*C9;
- output[ 8] = temp1 + temp2;
-
- temp1 = step[ 9]*C11;
- temp2 = step[14]*C5;
- output[ 9] = temp1 - temp2;
-
- temp1 = step[10]*C3;
- temp2 = step[13]*C13;
- output[10] = temp1 + temp2;
-
- temp1 = step[11]*C15;
- temp2 = step[12]*C1;
- output[11] = temp1 - temp2;
-
- temp1 = step[11]*C1;
- temp2 = step[12]*C15;
- output[12] = temp2 + temp1;
-
- temp1 = step[10]*C13;
- temp2 = step[13]*C3;
- output[13] = temp2 - temp1;
-
- temp1 = step[ 9]*C5;
- temp2 = step[14]*C11;
- output[14] = temp2 + temp1;
-
- temp1 = step[ 8]*C9;
- temp2 = step[15]*C7;
- output[15] = temp2 - temp1;
-
- // step 3
- step[ 0] = output[0] + output[3];
- step[ 1] = output[1] + output[2];
- step[ 2] = output[1] - output[2];
- step[ 3] = output[0] - output[3];
-
- temp1 = output[4]*C14;
- temp2 = output[7]*C2;
- step[ 4] = temp1 + temp2;
-
- temp1 = output[5]*C10;
- temp2 = output[6]*C6;
- step[ 5] = temp1 + temp2;
-
- temp1 = output[5]*C6;
- temp2 = output[6]*C10;
- step[ 6] = temp2 - temp1;
-
- temp1 = output[4]*C2;
- temp2 = output[7]*C14;
- step[ 7] = temp2 - temp1;
-
- step[ 8] = output[ 8] + output[11];
- step[ 9] = output[ 9] + output[10];
- step[10] = output[ 9] - output[10];
- step[11] = output[ 8] - output[11];
-
- step[12] = output[12] + output[15];
- step[13] = output[13] + output[14];
- step[14] = output[13] - output[14];
- step[15] = output[12] - output[15];
-
- // step 4
- output[ 0] = (step[ 0] + step[ 1]);
- output[ 8] = (step[ 0] - step[ 1]);
-
- temp1 = step[2]*C12;
- temp2 = step[3]*C4;
- temp1 = temp1 + temp2;
- output[ 4] = 2*(temp1*C8);
-
- temp1 = step[2]*C4;
- temp2 = step[3]*C12;
- temp1 = temp2 - temp1;
- output[12] = 2*(temp1*C8);
-
- output[ 2] = 2*((step[4] + step[ 5])*C8);
- output[14] = 2*((step[7] - step[ 6])*C8);
-
- temp1 = step[4] - step[5];
- temp2 = step[6] + step[7];
- output[ 6] = (temp1 + temp2);
- output[10] = (temp1 - temp2);
-
- intermediate[8] = step[8] + step[14];
- intermediate[9] = step[9] + step[15];
-
- temp1 = intermediate[8]*C12;
- temp2 = intermediate[9]*C4;
- temp1 = temp1 - temp2;
- output[3] = 2*(temp1*C8);
-
- temp1 = intermediate[8]*C4;
- temp2 = intermediate[9]*C12;
- temp1 = temp2 + temp1;
- output[13] = 2*(temp1*C8);
-
- output[ 9] = 2*((step[10] + step[11])*C8);
-
- intermediate[11] = step[10] - step[11];
- intermediate[12] = step[12] + step[13];
- intermediate[13] = step[12] - step[13];
- intermediate[14] = step[ 8] - step[14];
- intermediate[15] = step[ 9] - step[15];
-
- output[15] = (intermediate[11] + intermediate[12]);
- output[ 1] = -(intermediate[11] - intermediate[12]);
-
- output[ 7] = 2*(intermediate[13]*C8);
-
- temp1 = intermediate[14]*C12;
- temp2 = intermediate[15]*C4;
- temp1 = temp1 - temp2;
- output[11] = -2*(temp1*C8);
-
- temp1 = intermediate[14]*C4;
- temp2 = intermediate[15]*C12;
- temp1 = temp2 + temp1;
- output[ 5] = 2*(temp1*C8);
- }
- vp9_clear_system_state(); // Make it simd safe : __asm emms;
-}
-
-static void vp9_short_fdct16x16_c_f(short *input, short *out, int pitch,
- int scale) {
- vp9_clear_system_state(); // Make it simd safe : __asm emms;
- {
- int shortpitch = pitch >> 1;
- int i, j;
- double output[256];
- // First transform columns
- for (i = 0; i < 16; i++) {
- double temp_in[16], temp_out[16];
- for (j = 0; j < 16; j++)
- temp_in[j] = input[j*shortpitch + i];
- dct16x16_1d_f(temp_in, temp_out);
- for (j = 0; j < 16; j++)
- output[j*16 + i] = temp_out[j];
- }
- // Then transform rows
- for (i = 0; i < 16; ++i) {
- double temp_in[16], temp_out[16];
- for (j = 0; j < 16; ++j)
- temp_in[j] = output[j + i*16];
- dct16x16_1d_f(temp_in, temp_out);
- for (j = 0; j < 16; ++j)
- output[j + i*16] = temp_out[j];
- }
- // Scale by some magic number
- for (i = 0; i < 256; i++)
- out[i] = (short)round(output[i] / (2 << scale));
- }
- vp9_clear_system_state(); // Make it simd safe : __asm emms;
-}
-
-void vp9_short_fdct8x8_c_f(short *block, short *coefs, int pitch, int scale) {
- int j1, i, j, k;
- float b[8];
- float b1[8];
- float d[8][8];
- float f0 = (float) .7071068;
- float f1 = (float) .4903926;
- float f2 = (float) .4619398;
- float f3 = (float) .4157348;
- float f4 = (float) .3535534;
- float f5 = (float) .2777851;
- float f6 = (float) .1913417;
- float f7 = (float) .0975452;
- pitch = pitch / 2;
- for (i = 0, k = 0; i < 8; i++, k += pitch) {
- for (j = 0; j < 8; j++) {
- b[j] = (float)(block[k + j] << (3 - scale));
- }
- /* Horizontal transform */
- for (j = 0; j < 4; j++) {
- j1 = 7 - j;
- b1[j] = b[j] + b[j1];
- b1[j1] = b[j] - b[j1];
- }
- b[0] = b1[0] + b1[3];
- b[1] = b1[1] + b1[2];
- b[2] = b1[1] - b1[2];
- b[3] = b1[0] - b1[3];
- b[4] = b1[4];
- b[5] = (b1[6] - b1[5]) * f0;
- b[6] = (b1[6] + b1[5]) * f0;
- b[7] = b1[7];
- d[i][0] = (b[0] + b[1]) * f4;
- d[i][4] = (b[0] - b[1]) * f4;
- d[i][2] = b[2] * f6 + b[3] * f2;
- d[i][6] = b[3] * f6 - b[2] * f2;
- b1[4] = b[4] + b[5];
- b1[7] = b[7] + b[6];
- b1[5] = b[4] - b[5];
- b1[6] = b[7] - b[6];
- d[i][1] = b1[4] * f7 + b1[7] * f1;
- d[i][5] = b1[5] * f3 + b1[6] * f5;
- d[i][7] = b1[7] * f7 - b1[4] * f1;
- d[i][3] = b1[6] * f3 - b1[5] * f5;
- }
- /* Vertical transform */
- for (i = 0; i < 8; i++) {
- for (j = 0; j < 4; j++) {
- j1 = 7 - j;
- b1[j] = d[j][i] + d[j1][i];
- b1[j1] = d[j][i] - d[j1][i];
- }
- b[0] = b1[0] + b1[3];
- b[1] = b1[1] + b1[2];
- b[2] = b1[1] - b1[2];
- b[3] = b1[0] - b1[3];
- b[4] = b1[4];
- b[5] = (b1[6] - b1[5]) * f0;
- b[6] = (b1[6] + b1[5]) * f0;
- b[7] = b1[7];
- d[0][i] = (b[0] + b[1]) * f4;
- d[4][i] = (b[0] - b[1]) * f4;
- d[2][i] = b[2] * f6 + b[3] * f2;
- d[6][i] = b[3] * f6 - b[2] * f2;
- b1[4] = b[4] + b[5];
- b1[7] = b[7] + b[6];
- b1[5] = b[4] - b[5];
- b1[6] = b[7] - b[6];
- d[1][i] = b1[4] * f7 + b1[7] * f1;
- d[5][i] = b1[5] * f3 + b1[6] * f5;
- d[7][i] = b1[7] * f7 - b1[4] * f1;
- d[3][i] = b1[6] * f3 - b1[5] * f5;
- }
- for (i = 0; i < 8; i++) {
- for (j = 0; j < 8; j++) {
- *(coefs + j + i * 8) = (short) floor(d[i][j] + 0.5);
- }
- }
- return;
-}
-
-#define divide_bits(d, n) ((n) < 0 ? (d) << (n) : (d) >> (n))
-
-#if DWTDCT_TYPE == DWTDCT16X16_LEAN
-
-void vp9_short_fdct32x32_c(short *input, short *out, int pitch) {
- // assume out is a 32x32 buffer
- short buffer[16 * 16];
- int i, j;
- const int short_pitch = pitch >> 1;
-#if DWT_TYPE == 26
- dyadic_analyze_26(1, 32, 32, input, short_pitch, out, 32);
-#elif DWT_TYPE == 97
- dyadic_analyze_97(1, 32, 32, input, short_pitch, out, 32);
-#elif DWT_TYPE == 53
- dyadic_analyze_53(1, 32, 32, input, short_pitch, out, 32);
-#endif
- // TODO(debargha): Implement more efficiently by adding output pitch
- // argument to the dct16x16 function
- vp9_short_fdct16x16_c_f(out, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 32, buffer + i * 16, sizeof(short) * 16);
- for (i = 0; i < 16; ++i) {
- for (j = 16; j < 32; ++j) {
- out[i * 32 + j] = divide_bits(out[i * 32 + j], DWT_PRECISION_BITS - 2);
- }
- }
- for (i = 16; i < 32; ++i) {
- for (j = 0; j < 32; ++j) {
- out[i * 32 + j] = divide_bits(out[i * 32 + j], DWT_PRECISION_BITS - 2);
- }
- }
-}
-
-#elif DWTDCT_TYPE == DWTDCT16X16
-
-void vp9_short_fdct32x32_c(short *input, short *out, int pitch) {
- // assume out is a 32x32 buffer
- short buffer[16 * 16];
- int i, j;
- const int short_pitch = pitch >> 1;
-#if DWT_TYPE == 26
- dyadic_analyze_26(1, 32, 32, input, short_pitch, out, 32);
-#elif DWT_TYPE == 97
- dyadic_analyze_97(1, 32, 32, input, short_pitch, out, 32);
-#elif DWT_TYPE == 53
- dyadic_analyze_53(1, 32, 32, input, short_pitch, out, 32);
-#endif
- // TODO(debargha): Implement more efficiently by adding output pitch
- // argument to the dct16x16 function
- vp9_short_fdct16x16_c_f(out, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 32, buffer + i * 16, sizeof(short) * 16);
- vp9_short_fdct16x16_c_f(out + 16, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 32 + 16, buffer + i * 16, sizeof(short) * 16);
-
- vp9_short_fdct16x16_c_f(out + 32 * 16, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 32 + 32 * 16, buffer + i * 16, sizeof(short) * 16);
-
- vp9_short_fdct16x16_c_f(out + 33 * 16, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 32 + 33 * 16, buffer + i * 16, sizeof(short) * 16);
-}
-
-#elif DWTDCT_TYPE == DWTDCT8X8
-
-void vp9_short_fdct32x32_c(short *input, short *out, int pitch) {
- // assume out is a 32x32 buffer
- short buffer[8 * 8];
- int i, j;
- const int short_pitch = pitch >> 1;
-#if DWT_TYPE == 26
- dyadic_analyze_26(2, 32, 32, input, short_pitch, out, 32);
-#elif DWT_TYPE == 97
- dyadic_analyze_97(2, 32, 32, input, short_pitch, out, 32);
-#elif DWT_TYPE == 53
- dyadic_analyze_53(2, 32, 32, input, short_pitch, out, 32);
-#endif
- // TODO(debargha): Implement more efficiently by adding output pitch
- // argument to the dct16x16 function
- vp9_short_fdct8x8_c_f(out, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 8; ++i)
- vpx_memcpy(out + i * 32, buffer + i * 8, sizeof(short) * 8);
-
- vp9_short_fdct8x8_c_f(out + 8, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 8; ++i)
- vpx_memcpy(out + i * 32 + 8, buffer + i * 8, sizeof(short) * 8);
-
- vp9_short_fdct8x8_c_f(out + 32 * 8, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 8; ++i)
- vpx_memcpy(out + i * 32 + 32 * 8, buffer + i * 8, sizeof(short) * 8);
-
- vp9_short_fdct8x8_c_f(out + 33 * 8, buffer, 64, 1 + DWT_PRECISION_BITS);
- for (i = 0; i < 8; ++i)
- vpx_memcpy(out + i * 32 + 33 * 8, buffer + i * 8, sizeof(short) * 8);
-
- for (i = 0; i < 16; ++i) {
- for (j = 16; j < 32; ++j) {
- out[i * 32 + j] = divide_bits(out[i * 32 + j], DWT_PRECISION_BITS - 2);
- }
- }
- for (i = 16; i < 32; ++i) {
- for (j = 0; j < 32; ++j) {
- out[i * 32 + j] = divide_bits(out[i * 32 + j], DWT_PRECISION_BITS - 2);
- }
- }
-}
-
-#endif
-
-#if CONFIG_TX64X64
-void vp9_short_fdct64x64_c(short *input, short *out, int pitch) {
- // assume out is a 64x64 buffer
- short buffer[16 * 16];
- int i, j;
- const int short_pitch = pitch >> 1;
-#if DWT_TYPE == 26
- dyadic_analyze_26(2, 64, 64, input, short_pitch, out, 64);
-#elif DWT_TYPE == 97
- dyadic_analyze_97(2, 64, 64, input, short_pitch, out, 64);
-#elif DWT_TYPE == 53
- dyadic_analyze_53(2, 64, 64, input, short_pitch, out, 64);
-#endif
- // TODO(debargha): Implement more efficiently by adding output pitch
- // argument to the dct16x16 function
- vp9_short_fdct16x16_c_f(out, buffer, 128, 2 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 64, buffer + i * 16, sizeof(short) * 16);
-
-#if DWTDCT_TYPE == DWTDCT16X16_LEAN
- for (i = 0; i < 16; ++i) {
- for (j = 16; j < 48; ++j) {
- out[i * 64 + j] = divide_bits(out[i * 64 + j], DWT_PRECISION_BITS - 1);
- }
- }
- for (i = 16; i < 64; ++i) {
- for (j = 0; j < 64; ++j) {
- out[i * 64 + j] = divide_bits(out[i * 64 + j], DWT_PRECISION_BITS - 1);
- }
- }
-#elif DWTDCT_TYPE == DWTDCT16X16
- vp9_short_fdct16x16_c_f(out + 16, buffer, 128, 2 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 64 + 16, buffer + i * 16, sizeof(short) * 16);
-
- vp9_short_fdct16x16_c_f(out + 64 * 16, buffer, 128, 2 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 64 + 64 * 16, buffer + i * 16, sizeof(short) * 16);
-
- vp9_short_fdct16x16_c_f(out + 65 * 16, buffer, 128, 2 + DWT_PRECISION_BITS);
- for (i = 0; i < 16; ++i)
- vpx_memcpy(out + i * 64 + 65 * 16, buffer + i * 16, sizeof(short) * 16);
-
- // There is no dct used on the highest bands for now.
- // Need to scale these coeffs by a factor of 2/2^DWT_PRECISION_BITS
- // TODO(debargha): experiment with turning these coeffs to 0
- for (i = 0; i < 32; ++i) {
- for (j = 32; j < 64; ++j) {
- out[i * 64 + j] = divide_bits(out[i * 64 + j], DWT_PRECISION_BITS - 1);
- }
- }
- for (i = 32; i < 64; ++i) {
- for (j = 0; j < 64; ++j) {
- out[i * 64 + j] = divide_bits(out[i * 64 + j], DWT_PRECISION_BITS - 1);
- }
- }
-#endif // DWTDCT_TYPE
-}
-#endif // CONFIG_TX64X64
-#endif // CONFIG_DWTDCTHYBRID
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-15 15:19:54 +0000