Commit a81e821f authored by Luca Barbato's avatar Luca Barbato Committed by James Zern
Browse files

Namespace the idct/iad symbols

Make linking to libvpx and libaom at the same time possible.

Change-Id: I7bab8527a32e446e3d564e6fa5d94ccd056bc63f
parent b1c3bb57
...@@ -92,7 +92,7 @@ void aom_iwht4x4_1_add_c(const tran_low_t *in, uint8_t *dest, int dest_stride) { ...@@ -92,7 +92,7 @@ void aom_iwht4x4_1_add_c(const tran_low_t *in, uint8_t *dest, int dest_stride) {
} }
} }
void idct4_c(const tran_low_t *input, tran_low_t *output) { void aom_idct4_c(const tran_low_t *input, tran_low_t *output) {
tran_low_t step[4]; tran_low_t step[4];
tran_high_t temp1, temp2; tran_high_t temp1, temp2;
// stage 1 // stage 1
...@@ -120,7 +120,7 @@ void aom_idct4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -120,7 +120,7 @@ void aom_idct4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
// Rows // Rows
for (i = 0; i < 4; ++i) { for (i = 0; i < 4; ++i) {
idct4_c(input, outptr); aom_idct4_c(input, outptr);
input += 4; input += 4;
outptr += 4; outptr += 4;
} }
...@@ -128,7 +128,7 @@ void aom_idct4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -128,7 +128,7 @@ void aom_idct4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
// Columns // Columns
for (i = 0; i < 4; ++i) { for (i = 0; i < 4; ++i) {
for (j = 0; j < 4; ++j) temp_in[j] = out[j * 4 + i]; for (j = 0; j < 4; ++j) temp_in[j] = out[j * 4 + i];
idct4_c(temp_in, temp_out); aom_idct4_c(temp_in, temp_out);
for (j = 0; j < 4; ++j) { for (j = 0; j < 4; ++j) {
dest[j * stride + i] = clip_pixel_add(dest[j * stride + i], dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
ROUND_POWER_OF_TWO(temp_out[j], 4)); ROUND_POWER_OF_TWO(temp_out[j], 4));
...@@ -153,7 +153,7 @@ void aom_idct4x4_1_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -153,7 +153,7 @@ void aom_idct4x4_1_add_c(const tran_low_t *input, uint8_t *dest,
} }
} }
void idct8_c(const tran_low_t *input, tran_low_t *output) { void aom_idct8_c(const tran_low_t *input, tran_low_t *output) {
tran_low_t step1[8], step2[8]; tran_low_t step1[8], step2[8];
tran_high_t temp1, temp2; tran_high_t temp1, temp2;
// stage 1 // stage 1
...@@ -215,7 +215,7 @@ void aom_idct8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -215,7 +215,7 @@ void aom_idct8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
// First transform rows // First transform rows
for (i = 0; i < 8; ++i) { for (i = 0; i < 8; ++i) {
idct8_c(input, outptr); aom_idct8_c(input, outptr);
input += 8; input += 8;
outptr += 8; outptr += 8;
} }
...@@ -223,7 +223,7 @@ void aom_idct8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -223,7 +223,7 @@ void aom_idct8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
// Then transform columns // Then transform columns
for (i = 0; i < 8; ++i) { for (i = 0; i < 8; ++i) {
for (j = 0; j < 8; ++j) temp_in[j] = out[j * 8 + i]; for (j = 0; j < 8; ++j) temp_in[j] = out[j * 8 + i];
idct8_c(temp_in, temp_out); aom_idct8_c(temp_in, temp_out);
for (j = 0; j < 8; ++j) { for (j = 0; j < 8; ++j) {
dest[j * stride + i] = clip_pixel_add(dest[j * stride + i], dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
ROUND_POWER_OF_TWO(temp_out[j], 5)); ROUND_POWER_OF_TWO(temp_out[j], 5));
...@@ -243,7 +243,7 @@ void aom_idct8x8_1_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -243,7 +243,7 @@ void aom_idct8x8_1_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
} }
} }
void iadst4_c(const tran_low_t *input, tran_low_t *output) { void aom_iadst4_c(const tran_low_t *input, tran_low_t *output) {
tran_high_t s0, s1, s2, s3, s4, s5, s6, s7; tran_high_t s0, s1, s2, s3, s4, s5, s6, s7;
tran_low_t x0 = input[0]; tran_low_t x0 = input[0];
...@@ -280,7 +280,7 @@ void iadst4_c(const tran_low_t *input, tran_low_t *output) { ...@@ -280,7 +280,7 @@ void iadst4_c(const tran_low_t *input, tran_low_t *output) {
output[3] = WRAPLOW(dct_const_round_shift(s0 + s1 - s3), 8); output[3] = WRAPLOW(dct_const_round_shift(s0 + s1 - s3), 8);
} }
void iadst8_c(const tran_low_t *input, tran_low_t *output) { void aom_iadst8_c(const tran_low_t *input, tran_low_t *output) {
int s0, s1, s2, s3, s4, s5, s6, s7; int s0, s1, s2, s3, s4, s5, s6, s7;
tran_high_t x0 = input[7]; tran_high_t x0 = input[7];
...@@ -366,7 +366,7 @@ void aom_idct8x8_12_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -366,7 +366,7 @@ void aom_idct8x8_12_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
// First transform rows // First transform rows
// only first 4 row has non-zero coefs // only first 4 row has non-zero coefs
for (i = 0; i < 4; ++i) { for (i = 0; i < 4; ++i) {
idct8_c(input, outptr); aom_idct8_c(input, outptr);
input += 8; input += 8;
outptr += 8; outptr += 8;
} }
...@@ -374,7 +374,7 @@ void aom_idct8x8_12_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -374,7 +374,7 @@ void aom_idct8x8_12_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
// Then transform columns // Then transform columns
for (i = 0; i < 8; ++i) { for (i = 0; i < 8; ++i) {
for (j = 0; j < 8; ++j) temp_in[j] = out[j * 8 + i]; for (j = 0; j < 8; ++j) temp_in[j] = out[j * 8 + i];
idct8_c(temp_in, temp_out); aom_idct8_c(temp_in, temp_out);
for (j = 0; j < 8; ++j) { for (j = 0; j < 8; ++j) {
dest[j * stride + i] = clip_pixel_add(dest[j * stride + i], dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
ROUND_POWER_OF_TWO(temp_out[j], 5)); ROUND_POWER_OF_TWO(temp_out[j], 5));
...@@ -382,7 +382,7 @@ void aom_idct8x8_12_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -382,7 +382,7 @@ void aom_idct8x8_12_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
} }
} }
void idct16_c(const tran_low_t *input, tran_low_t *output) { void aom_idct16_c(const tran_low_t *input, tran_low_t *output) {
tran_low_t step1[16], step2[16]; tran_low_t step1[16], step2[16];
tran_high_t temp1, temp2; tran_high_t temp1, temp2;
...@@ -556,7 +556,7 @@ void aom_idct16x16_256_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -556,7 +556,7 @@ void aom_idct16x16_256_add_c(const tran_low_t *input, uint8_t *dest,
// First transform rows // First transform rows
for (i = 0; i < 16; ++i) { for (i = 0; i < 16; ++i) {
idct16_c(input, outptr); aom_idct16_c(input, outptr);
input += 16; input += 16;
outptr += 16; outptr += 16;
} }
...@@ -564,7 +564,7 @@ void aom_idct16x16_256_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -564,7 +564,7 @@ void aom_idct16x16_256_add_c(const tran_low_t *input, uint8_t *dest,
// Then transform columns // Then transform columns
for (i = 0; i < 16; ++i) { for (i = 0; i < 16; ++i) {
for (j = 0; j < 16; ++j) temp_in[j] = out[j * 16 + i]; for (j = 0; j < 16; ++j) temp_in[j] = out[j * 16 + i];
idct16_c(temp_in, temp_out); aom_idct16_c(temp_in, temp_out);
for (j = 0; j < 16; ++j) { for (j = 0; j < 16; ++j) {
dest[j * stride + i] = clip_pixel_add(dest[j * stride + i], dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
ROUND_POWER_OF_TWO(temp_out[j], 6)); ROUND_POWER_OF_TWO(temp_out[j], 6));
...@@ -572,7 +572,7 @@ void aom_idct16x16_256_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -572,7 +572,7 @@ void aom_idct16x16_256_add_c(const tran_low_t *input, uint8_t *dest,
} }
} }
void iadst16_c(const tran_low_t *input, tran_low_t *output) { void aom_iadst16_c(const tran_low_t *input, tran_low_t *output) {
tran_high_t s0, s1, s2, s3, s4, s5, s6, s7, s8; tran_high_t s0, s1, s2, s3, s4, s5, s6, s7, s8;
tran_high_t s9, s10, s11, s12, s13, s14, s15; tran_high_t s9, s10, s11, s12, s13, s14, s15;
...@@ -753,7 +753,7 @@ void aom_idct16x16_10_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -753,7 +753,7 @@ void aom_idct16x16_10_add_c(const tran_low_t *input, uint8_t *dest,
// First transform rows. Since all non-zero dct coefficients are in // First transform rows. Since all non-zero dct coefficients are in
// upper-left 4x4 area, we only need to calculate first 4 rows here. // upper-left 4x4 area, we only need to calculate first 4 rows here.
for (i = 0; i < 4; ++i) { for (i = 0; i < 4; ++i) {
idct16_c(input, outptr); aom_idct16_c(input, outptr);
input += 16; input += 16;
outptr += 16; outptr += 16;
} }
...@@ -761,7 +761,7 @@ void aom_idct16x16_10_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -761,7 +761,7 @@ void aom_idct16x16_10_add_c(const tran_low_t *input, uint8_t *dest,
// Then transform columns // Then transform columns
for (i = 0; i < 16; ++i) { for (i = 0; i < 16; ++i) {
for (j = 0; j < 16; ++j) temp_in[j] = out[j * 16 + i]; for (j = 0; j < 16; ++j) temp_in[j] = out[j * 16 + i];
idct16_c(temp_in, temp_out); aom_idct16_c(temp_in, temp_out);
for (j = 0; j < 16; ++j) { for (j = 0; j < 16; ++j) {
dest[j * stride + i] = clip_pixel_add(dest[j * stride + i], dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
ROUND_POWER_OF_TWO(temp_out[j], 6)); ROUND_POWER_OF_TWO(temp_out[j], 6));
...@@ -781,7 +781,7 @@ void aom_idct16x16_1_add_c(const tran_low_t *input, uint8_t *dest, int stride) { ...@@ -781,7 +781,7 @@ void aom_idct16x16_1_add_c(const tran_low_t *input, uint8_t *dest, int stride) {
} }
} }
void idct32_c(const tran_low_t *input, tran_low_t *output) { void aom_idct32_c(const tran_low_t *input, tran_low_t *output) {
tran_low_t step1[32], step2[32]; tran_low_t step1[32], step2[32];
tran_high_t temp1, temp2; tran_high_t temp1, temp2;
...@@ -1167,7 +1167,7 @@ void aom_idct32x32_1024_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -1167,7 +1167,7 @@ void aom_idct32x32_1024_add_c(const tran_low_t *input, uint8_t *dest,
zero_coeff[j] = zero_coeff[2 * j] | zero_coeff[2 * j + 1]; zero_coeff[j] = zero_coeff[2 * j] | zero_coeff[2 * j + 1];
if (zero_coeff[0] | zero_coeff[1]) if (zero_coeff[0] | zero_coeff[1])
idct32_c(input, outptr); aom_idct32_c(input, outptr);
else else
memset(outptr, 0, sizeof(tran_low_t) * 32); memset(outptr, 0, sizeof(tran_low_t) * 32);
input += 32; input += 32;
...@@ -1177,7 +1177,7 @@ void aom_idct32x32_1024_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -1177,7 +1177,7 @@ void aom_idct32x32_1024_add_c(const tran_low_t *input, uint8_t *dest,
// Columns // Columns
for (i = 0; i < 32; ++i) { for (i = 0; i < 32; ++i) {
for (j = 0; j < 32; ++j) temp_in[j] = out[j * 32 + i]; for (j = 0; j < 32; ++j) temp_in[j] = out[j * 32 + i];
idct32_c(temp_in, temp_out); aom_idct32_c(temp_in, temp_out);
for (j = 0; j < 32; ++j) { for (j = 0; j < 32; ++j) {
dest[j * stride + i] = clip_pixel_add(dest[j * stride + i], dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
ROUND_POWER_OF_TWO(temp_out[j], 6)); ROUND_POWER_OF_TWO(temp_out[j], 6));
...@@ -1195,7 +1195,7 @@ void aom_idct32x32_135_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -1195,7 +1195,7 @@ void aom_idct32x32_135_add_c(const tran_low_t *input, uint8_t *dest,
// Rows // Rows
// only upper-left 16x16 has non-zero coeff // only upper-left 16x16 has non-zero coeff
for (i = 0; i < 16; ++i) { for (i = 0; i < 16; ++i) {
idct32_c(input, outptr); aom_idct32_c(input, outptr);
input += 32; input += 32;
outptr += 32; outptr += 32;
} }
...@@ -1203,7 +1203,7 @@ void aom_idct32x32_135_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -1203,7 +1203,7 @@ void aom_idct32x32_135_add_c(const tran_low_t *input, uint8_t *dest,
// Columns // Columns
for (i = 0; i < 32; ++i) { for (i = 0; i < 32; ++i) {
for (j = 0; j < 32; ++j) temp_in[j] = out[j * 32 + i]; for (j = 0; j < 32; ++j) temp_in[j] = out[j * 32 + i];
idct32_c(temp_in, temp_out); aom_idct32_c(temp_in, temp_out);
for (j = 0; j < 32; ++j) { for (j = 0; j < 32; ++j) {
dest[j * stride + i] = clip_pixel_add(dest[j * stride + i], dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
ROUND_POWER_OF_TWO(temp_out[j], 6)); ROUND_POWER_OF_TWO(temp_out[j], 6));
...@@ -1221,7 +1221,7 @@ void aom_idct32x32_34_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -1221,7 +1221,7 @@ void aom_idct32x32_34_add_c(const tran_low_t *input, uint8_t *dest,
// Rows // Rows
// only upper-left 8x8 has non-zero coeff // only upper-left 8x8 has non-zero coeff
for (i = 0; i < 8; ++i) { for (i = 0; i < 8; ++i) {
idct32_c(input, outptr); aom_idct32_c(input, outptr);
input += 32; input += 32;
outptr += 32; outptr += 32;
} }
...@@ -1229,7 +1229,7 @@ void aom_idct32x32_34_add_c(const tran_low_t *input, uint8_t *dest, ...@@ -1229,7 +1229,7 @@ void aom_idct32x32_34_add_c(const tran_low_t *input, uint8_t *dest,
// Columns // Columns
for (i = 0; i < 32; ++i) { for (i = 0; i < 32; ++i) {
for (j = 0; j < 32; ++j) temp_in[j] = out[j * 32 + i]; for (j = 0; j < 32; ++j) temp_in[j] = out[j * 32 + i];
idct32_c(temp_in, temp_out); aom_idct32_c(temp_in, temp_out);
for (j = 0; j < 32; ++j) { for (j = 0; j < 32; ++j) {
dest[j * stride + i] = clip_pixel_add(dest[j * stride + i], dest[j * stride + i] = clip_pixel_add(dest[j * stride + i],
ROUND_POWER_OF_TWO(temp_out[j], 6)); ROUND_POWER_OF_TWO(temp_out[j], 6));
......
...@@ -88,13 +88,13 @@ static INLINE tran_low_t highbd_dct_const_round_shift(tran_high_t input, ...@@ -88,13 +88,13 @@ static INLINE tran_low_t highbd_dct_const_round_shift(tran_high_t input,
#define WRAPLOW(x, bd) ((int32_t)(x)) #define WRAPLOW(x, bd) ((int32_t)(x))
#endif // CONFIG_EMULATE_HARDWARE #endif // CONFIG_EMULATE_HARDWARE
void idct4_c(const tran_low_t *input, tran_low_t *output); void aom_idct4_c(const tran_low_t *input, tran_low_t *output);
void idct8_c(const tran_low_t *input, tran_low_t *output); void aom_idct8_c(const tran_low_t *input, tran_low_t *output);
void idct16_c(const tran_low_t *input, tran_low_t *output); void aom_idct16_c(const tran_low_t *input, tran_low_t *output);
void idct32_c(const tran_low_t *input, tran_low_t *output); void aom_idct32_c(const tran_low_t *input, tran_low_t *output);
void iadst4_c(const tran_low_t *input, tran_low_t *output); void aom_iadst4_c(const tran_low_t *input, tran_low_t *output);
void iadst8_c(const tran_low_t *input, tran_low_t *output); void aom_iadst8_c(const tran_low_t *input, tran_low_t *output);
void iadst16_c(const tran_low_t *input, tran_low_t *output); void aom_iadst16_c(const tran_low_t *input, tran_low_t *output);
#if CONFIG_AOM_HIGHBITDEPTH #if CONFIG_AOM_HIGHBITDEPTH
void aom_highbd_idct4_c(const tran_low_t *input, tran_low_t *output, int bd); void aom_highbd_idct4_c(const tran_low_t *input, tran_low_t *output, int bd);
......
...@@ -179,7 +179,7 @@ static INLINE void transpose_4x4(__m128i *res) { ...@@ -179,7 +179,7 @@ static INLINE void transpose_4x4(__m128i *res) {
res[1] = _mm_unpackhi_epi16(tr0_0, tr0_1); res[1] = _mm_unpackhi_epi16(tr0_0, tr0_1);
} }
void idct4_sse2(__m128i *in) { void aom_idct4_sse2(__m128i *in) {
const __m128i k__cospi_p16_p16 = pair_set_epi16(cospi_16_64, cospi_16_64); const __m128i k__cospi_p16_p16 = pair_set_epi16(cospi_16_64, cospi_16_64);
const __m128i k__cospi_p16_m16 = pair_set_epi16(cospi_16_64, -cospi_16_64); const __m128i k__cospi_p16_m16 = pair_set_epi16(cospi_16_64, -cospi_16_64);
const __m128i k__cospi_p24_m08 = pair_set_epi16(cospi_24_64, -cospi_8_64); const __m128i k__cospi_p24_m08 = pair_set_epi16(cospi_24_64, -cospi_8_64);
...@@ -215,7 +215,7 @@ void idct4_sse2(__m128i *in) { ...@@ -215,7 +215,7 @@ void idct4_sse2(__m128i *in) {
in[1] = _mm_shuffle_epi32(in[1], 0x4E); in[1] = _mm_shuffle_epi32(in[1], 0x4E);
} }
void iadst4_sse2(__m128i *in) { void aom_iadst4_sse2(__m128i *in) {
const __m128i k__sinpi_p01_p04 = pair_set_epi16(sinpi_1_9, sinpi_4_9); const __m128i k__sinpi_p01_p04 = pair_set_epi16(sinpi_1_9, sinpi_4_9);
const __m128i k__sinpi_p03_p02 = pair_set_epi16(sinpi_3_9, sinpi_2_9); const __m128i k__sinpi_p03_p02 = pair_set_epi16(sinpi_3_9, sinpi_2_9);
const __m128i k__sinpi_p02_m01 = pair_set_epi16(sinpi_2_9, -sinpi_1_9); const __m128i k__sinpi_p02_m01 = pair_set_epi16(sinpi_2_9, -sinpi_1_9);
...@@ -541,7 +541,7 @@ void aom_idct8x8_1_add_sse2(const tran_low_t *input, uint8_t *dest, ...@@ -541,7 +541,7 @@ void aom_idct8x8_1_add_sse2(const tran_low_t *input, uint8_t *dest,
RECON_AND_STORE(dest + 7 * stride, dc_value); RECON_AND_STORE(dest + 7 * stride, dc_value);
} }
void idct8_sse2(__m128i *in) { void aom_idct8_sse2(__m128i *in) {
const __m128i rounding = _mm_set1_epi32(DCT_CONST_ROUNDING); const __m128i rounding = _mm_set1_epi32(DCT_CONST_ROUNDING);
const __m128i stg1_0 = pair_set_epi16(cospi_28_64, -cospi_4_64); const __m128i stg1_0 = pair_set_epi16(cospi_28_64, -cospi_4_64);
const __m128i stg1_1 = pair_set_epi16(cospi_4_64, cospi_28_64); const __m128i stg1_1 = pair_set_epi16(cospi_4_64, cospi_28_64);
...@@ -566,7 +566,7 @@ void idct8_sse2(__m128i *in) { ...@@ -566,7 +566,7 @@ void idct8_sse2(__m128i *in) {
in[4], in[5], in[6], in[7]); in[4], in[5], in[6], in[7]);
} }
void iadst8_sse2(__m128i *in) { void aom_iadst8_sse2(__m128i *in) {
const __m128i k__cospi_p02_p30 = pair_set_epi16(cospi_2_64, cospi_30_64); const __m128i k__cospi_p02_p30 = pair_set_epi16(cospi_2_64, cospi_30_64);
const __m128i k__cospi_p30_m02 = pair_set_epi16(cospi_30_64, -cospi_2_64); const __m128i k__cospi_p30_m02 = pair_set_epi16(cospi_30_64, -cospi_2_64);
const __m128i k__cospi_p10_p22 = pair_set_epi16(cospi_10_64, cospi_22_64); const __m128i k__cospi_p10_p22 = pair_set_epi16(cospi_10_64, cospi_22_64);
...@@ -2136,13 +2136,13 @@ static void idct16_8col(__m128i *in) { ...@@ -2136,13 +2136,13 @@ static void idct16_8col(__m128i *in) {
in[15] = _mm_sub_epi16(s[0], s[15]); in[15] = _mm_sub_epi16(s[0], s[15]);
} }
void idct16_sse2(__m128i *in0, __m128i *in1) { void aom_idct16_sse2(__m128i *in0, __m128i *in1) {
array_transpose_16x16(in0, in1); array_transpose_16x16(in0, in1);
idct16_8col(in0); idct16_8col(in0);
idct16_8col(in1); idct16_8col(in1);
} }
void iadst16_sse2(__m128i *in0, __m128i *in1) { void aom_iadst16_sse2(__m128i *in0, __m128i *in1) {
array_transpose_16x16(in0, in1); array_transpose_16x16(in0, in1);
iadst16_8col(in0); iadst16_8col(in0);
iadst16_8col(in1); iadst16_8col(in1);
...@@ -3517,7 +3517,7 @@ void aom_highbd_idct4x4_16_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3517,7 +3517,7 @@ void aom_highbd_idct4x4_16_add_sse2(const tran_low_t *input, uint8_t *dest8,
if (!test) { if (!test) {
// Do the row transform // Do the row transform
idct4_sse2(inptr); aom_idct4_sse2(inptr);
// Check the min & max values // Check the min & max values
max_input = _mm_max_epi16(inptr[0], inptr[1]); max_input = _mm_max_epi16(inptr[0], inptr[1]);
...@@ -3553,7 +3553,7 @@ void aom_highbd_idct4x4_16_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3553,7 +3553,7 @@ void aom_highbd_idct4x4_16_add_sse2(const tran_low_t *input, uint8_t *dest8,
} }
if (optimised_cols) { if (optimised_cols) {
idct4_sse2(inptr); aom_idct4_sse2(inptr);
// Final round and shift // Final round and shift
inptr[0] = _mm_add_epi16(inptr[0], eight); inptr[0] = _mm_add_epi16(inptr[0], eight);
...@@ -3633,7 +3633,7 @@ void aom_highbd_idct8x8_64_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3633,7 +3633,7 @@ void aom_highbd_idct8x8_64_add_sse2(const tran_low_t *input, uint8_t *dest8,
if (!test) { if (!test) {
// Do the row transform // Do the row transform
idct8_sse2(inptr); aom_idct8_sse2(inptr);
// Find the min & max for the column transform // Find the min & max for the column transform
max_input = _mm_max_epi16(inptr[0], inptr[1]); max_input = _mm_max_epi16(inptr[0], inptr[1]);
...@@ -3670,7 +3670,7 @@ void aom_highbd_idct8x8_64_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3670,7 +3670,7 @@ void aom_highbd_idct8x8_64_add_sse2(const tran_low_t *input, uint8_t *dest8,
} }
if (optimised_cols) { if (optimised_cols) {
idct8_sse2(inptr); aom_idct8_sse2(inptr);
// Final round & shift and Reconstruction and Store // Final round & shift and Reconstruction and Store
{ {
...@@ -3734,7 +3734,7 @@ void aom_highbd_idct8x8_10_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3734,7 +3734,7 @@ void aom_highbd_idct8x8_10_add_sse2(const tran_low_t *input, uint8_t *dest8,
if (!test) { if (!test) {
// Do the row transform // Do the row transform
idct8_sse2(inptr); aom_idct8_sse2(inptr);
// Find the min & max for the column transform // Find the min & max for the column transform
// N.B. Only first 4 cols contain non-zero coeffs // N.B. Only first 4 cols contain non-zero coeffs
...@@ -3773,7 +3773,7 @@ void aom_highbd_idct8x8_10_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3773,7 +3773,7 @@ void aom_highbd_idct8x8_10_add_sse2(const tran_low_t *input, uint8_t *dest8,
} }
if (optimised_cols) { if (optimised_cols) {
idct8_sse2(inptr); aom_idct8_sse2(inptr);
// Final round & shift and Reconstruction and Store // Final round & shift and Reconstruction and Store
{ {
...@@ -3839,7 +3839,7 @@ void aom_highbd_idct16x16_256_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3839,7 +3839,7 @@ void aom_highbd_idct16x16_256_add_sse2(const tran_low_t *input, uint8_t *dest8,
if (!test) { if (!test) {
// Do the row transform // Do the row transform
idct16_sse2(inptr, inptr + 16); aom_idct16_sse2(inptr, inptr + 16);
// Find the min & max for the column transform // Find the min & max for the column transform
max_input = _mm_max_epi16(inptr[0], inptr[1]); max_input = _mm_max_epi16(inptr[0], inptr[1]);
...@@ -3881,7 +3881,7 @@ void aom_highbd_idct16x16_256_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3881,7 +3881,7 @@ void aom_highbd_idct16x16_256_add_sse2(const tran_low_t *input, uint8_t *dest8,
} }
if (optimised_cols) { if (optimised_cols) {
idct16_sse2(inptr, inptr + 16); aom_idct16_sse2(inptr, inptr + 16);
// Final round & shift and Reconstruction and Store // Final round & shift and Reconstruction and Store
{ {
...@@ -3954,7 +3954,7 @@ void aom_highbd_idct16x16_10_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3954,7 +3954,7 @@ void aom_highbd_idct16x16_10_add_sse2(const tran_low_t *input, uint8_t *dest8,
if (!test) { if (!test) {
// Do the row transform (N.B. This transposes inptr) // Do the row transform (N.B. This transposes inptr)
idct16_sse2(inptr, inptr + 16); aom_idct16_sse2(inptr, inptr + 16);
// Find the min & max for the column transform // Find the min & max for the column transform
// N.B. Only first 4 cols contain non-zero coeffs // N.B. Only first 4 cols contain non-zero coeffs
...@@ -3999,7 +3999,7 @@ void aom_highbd_idct16x16_10_add_sse2(const tran_low_t *input, uint8_t *dest8, ...@@ -3999,7 +3999,7 @@ void aom_highbd_idct16x16_10_add_sse2(const tran_low_t *input, uint8_t *dest8,
} }
if (optimised_cols) { if (optimised_cols) {
idct16_sse2(inptr, inptr + 16); aom_idct16_sse2(inptr, inptr + 16);
// Final round & shift and Reconstruction and Store // Final round & shift and Reconstruction and Store
{ {
......
...@@ -187,11 +187,11 @@ static INLINE void write_buffer_8x16(uint8_t *dest, __m128i *in, int stride) { ...@@ -187,11 +187,11 @@ static INLINE void write_buffer_8x16(uint8_t *dest, __m128i *in, int stride) {
RECON_AND_STORE(dest + 15 * stride, in[15]); RECON_AND_STORE(dest + 15 * stride, in[15]);
} }
void idct4_sse2(__m128i *in); void aom_idct4_sse2(__m128i *in);
void idct8_sse2(__m128i *in); void aom_idct8_sse2(__m128i *in);
void idct16_sse2(__m128i *in0, __m128i *in1); void aom_idct16_sse2(__m128i *in0, __m128i *in1);
void iadst4_sse2(__m128i *in); void aom_iadst4_sse2(__m128i *in);
void iadst8_sse2(__m128i *in); void aom_iadst8_sse2(__m128i *in);
void iadst16_sse2(__m128i *in0, __m128i *in1); void aom_iadst16_sse2(__m128i *in0, __m128i *in1);
#endif // AOM_DSP_X86_INV_TXFM_SSE2_H_ #endif // AOM_DSP_X86_INV_TXFM_SSE2_H_
...@@ -21,10 +21,10 @@ ...@@ -21,10 +21,10 @@
void av1_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride, void av1_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride,
int tx_type) { int tx_type) {
const transform_2d IHT_4[] = { const transform_2d IHT_4[] = {
{ idct4_c, idct4_c }, // DCT_DCT = 0 { aom_idct4_c, aom_idct4_c }, // DCT_DCT = 0
{ iadst4_c, idct4_c }, // ADST_DCT = 1 { aom_iadst4_c, aom_idct4_c }, // ADST_DCT = 1
{ idct4_c, iadst4_c }, // DCT_ADST = 2 { aom_idct4_c, aom_iadst4_c }, // DCT_ADST = 2
{ iadst4_c, iadst4_c } // ADST_ADST = 3 { aom_iadst4_c, aom_iadst4_c } // ADST_ADST = 3
}; };
int i, j; int i, j;
...@@ -51,10 +51,10 @@ void av1_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride, ...@@ -51,10 +51,10 @@ void av1_iht4x4_16_add_c(const tran_low_t *input, uint8_t *dest, int stride,
} }
static const transform_2d IHT_8[] = { static const transform_2d IHT_8[] = {
{ idct8_c, idct8_c }, // DCT_DCT = 0 { aom_idct8_c, aom_idct8_c }, // DCT_DCT = 0
{ iadst8_c, idct8_c }, // ADST_DCT = 1 { aom_iadst8_c, aom_idct8_c }, // ADST_DCT = 1
{ idct8_c, iadst8_c }, // DCT_ADST = 2 { aom_idct8_c, aom_iadst8_c }, // DCT_ADST = 2
{ iadst8_c, iadst8_c } // ADST_ADST = 3 { aom_iadst8_c, aom_iadst8_c } // ADST_ADST = 3
}; };
void av1_iht8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride, void av1_iht8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride,
...@@ -84,10 +84,10 @@ void av1_iht8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride, ...@@ -84,10 +84,10 @@ void av1_iht8x8_64_add_c(const tran_low_t *input, uint8_t *dest, int stride,
} }