AVX2 implementation of av1_highbd_convolve_x_sr_c
AVX2 implementation of av1_highbd_convolve_x_sr_c has been added.
In the function av1_highbd_convolve_2d_sr_avx2:
aligned store instruction is changed to unaligned store.
BUG Fix: Rounding by (2*FILTER_BITS - round_0 - round_1) is included.
Change-Id: I50821518be23d14f7b185ac160ec96b85df5a481
diff --git a/aom_dsp/x86/highbd_convolve_avx2.c b/aom_dsp/x86/highbd_convolve_avx2.c
index 35b0b4f..2f6a820 100644
--- a/aom_dsp/x86/highbd_convolve_avx2.c
+++ b/aom_dsp/x86/highbd_convolve_avx2.c
@@ -13,6 +13,8 @@
#include "./aom_dsp_rtcd.h"
#include "aom_dsp/x86/convolve.h"
+#include "aom_dsp/x86/convolve_avx2.h"
+#include "aom_dsp/x86/synonyms.h"
// -----------------------------------------------------------------------------
// Copy and average
@@ -192,6 +194,104 @@
}
}
+void av1_highbd_convolve_x_sr_avx2(const uint16_t *src, int src_stride,
+ uint16_t *dst, int dst_stride, int w, int h,
+ InterpFilterParams *filter_params_x,
+ InterpFilterParams *filter_params_y,
+ const int subpel_x_q4, const int subpel_y_q4,
+ ConvolveParams *conv_params, int bd) {
+ int i, j;
+ const int fo_horiz = filter_params_x->taps / 2 - 1;
+ const uint16_t *const src_ptr = src - fo_horiz;
+ (void)subpel_y_q4;
+ (void)filter_params_y;
+
+ // Check that, even with 12-bit input, the intermediate values will fit
+ // into an unsigned 16-bit intermediate array.
+ assert(bd + FILTER_BITS + 2 - conv_params->round_0 <= 16);
+
+ __m256i s[4], coeffs_x[4];
+
+ const __m256i round_const_x =
+ _mm256_set1_epi32(((1 << conv_params->round_0) >> 1));
+ const __m128i round_shift_x = _mm_cvtsi32_si128(conv_params->round_0);
+
+ const int bits = FILTER_BITS - conv_params->round_0;
+ const __m128i round_shift_bits = _mm_cvtsi32_si128(bits);
+ const __m256i round_const_bits = _mm256_set1_epi32((1 << bits) >> 1);
+ const __m256i clip_pixel =
+ _mm256_set1_epi16(bd == 10 ? 1023 : (bd == 12 ? 4095 : 255));
+ const __m256i zero = _mm256_setzero_si256();
+
+ assert(bits >= 0);
+ assert((FILTER_BITS - conv_params->round_1) >= 0 ||
+ ((conv_params->round_0 + conv_params->round_1) == 2 * FILTER_BITS));
+
+ prepare_coeffs(filter_params_x, subpel_x_q4, coeffs_x);
+
+ for (j = 0; j < w; j += 8) {
+ /* Horizontal filter */
+ for (i = 0; i < h; i += 2) {
+ const __m256i row0 =
+ _mm256_loadu_si256((__m256i *)&src_ptr[i * src_stride + j]);
+ __m256i row1 =
+ _mm256_loadu_si256((__m256i *)&src_ptr[(i + 1) * src_stride + j]);
+
+ const __m256i r0 = _mm256_permute2x128_si256(row0, row1, 0x20);
+ const __m256i r1 = _mm256_permute2x128_si256(row0, row1, 0x31);
+
+ // even pixels
+ s[0] = _mm256_alignr_epi8(r1, r0, 0);
+ s[1] = _mm256_alignr_epi8(r1, r0, 4);
+ s[2] = _mm256_alignr_epi8(r1, r0, 8);
+ s[3] = _mm256_alignr_epi8(r1, r0, 12);
+
+ __m256i res_even = convolve(s, coeffs_x);
+ res_even = _mm256_sra_epi32(_mm256_add_epi32(res_even, round_const_x),
+ round_shift_x);
+
+ // odd pixels
+ s[0] = _mm256_alignr_epi8(r1, r0, 2);
+ s[1] = _mm256_alignr_epi8(r1, r0, 6);
+ s[2] = _mm256_alignr_epi8(r1, r0, 10);
+ s[3] = _mm256_alignr_epi8(r1, r0, 14);
+
+ __m256i res_odd = convolve(s, coeffs_x);
+ res_odd = _mm256_sra_epi32(_mm256_add_epi32(res_odd, round_const_x),
+ round_shift_x);
+
+ res_even = _mm256_sra_epi32(_mm256_add_epi32(res_even, round_const_bits),
+ round_shift_bits);
+ res_odd = _mm256_sra_epi32(_mm256_add_epi32(res_odd, round_const_bits),
+ round_shift_bits);
+
+ __m256i res_even1 = _mm256_packs_epi32(res_even, res_even);
+ __m256i res_odd1 = _mm256_packs_epi32(res_odd, res_odd);
+
+ __m256i res = _mm256_unpacklo_epi16(res_even1, res_odd1);
+ res = _mm256_min_epi16(res, clip_pixel);
+ res = _mm256_max_epi16(res, zero);
+
+ if (w - j > 4) {
+ _mm_storeu_si128((__m128i *)&dst[i * dst_stride + j],
+ _mm256_castsi256_si128(res));
+ _mm_storeu_si128((__m128i *)&dst[i * dst_stride + j + dst_stride],
+ _mm256_extracti128_si256(res, 1));
+ } else if (w == 4) {
+ _mm_storel_epi64((__m128i *)&dst[i * dst_stride + j],
+ _mm256_castsi256_si128(res));
+ _mm_storel_epi64((__m128i *)&dst[i * dst_stride + j + dst_stride],
+ _mm256_extracti128_si256(res, 1));
+ } else {
+ xx_storel_32((__m128i *)&dst[i * dst_stride + j],
+ _mm256_castsi256_si128(res));
+ xx_storel_32((__m128i *)&dst[i * dst_stride + j + dst_stride],
+ _mm256_extracti128_si256(res, 1));
+ }
+ }
+ }
+}
+
#define CONV8_ROUNDING_BITS (7)
// -----------------------------------------------------------------------------
diff --git a/av1/common/av1_rtcd_defs.pl b/av1/common/av1_rtcd_defs.pl
index 94990ea..6199796 100755
--- a/av1/common/av1_rtcd_defs.pl
+++ b/av1/common/av1_rtcd_defs.pl
@@ -524,6 +524,7 @@
add_proto qw/void av1_highbd_convolve_x/, "const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, InterpFilterParams *filter_params_x, InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd";
add_proto qw/void av1_highbd_convolve_x_sr/, "const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, InterpFilterParams *filter_params_x, InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd";
+specialize qw/av1_highbd_convolve_x_sr c avx2/;
add_proto qw/void av1_highbd_convolve_y/, "const uint16_t *src, int src_stride, uint16_t *dst, int dst_stride, int w, int h, InterpFilterParams *filter_params_x, InterpFilterParams *filter_params_y, const int subpel_x_q4, const int subpel_y_q4, ConvolveParams *conv_params, int bd";
diff --git a/av1/common/x86/highbd_convolve_2d_avx2.c b/av1/common/x86/highbd_convolve_2d_avx2.c
index 73c2bea..750d4a8 100644
--- a/av1/common/x86/highbd_convolve_2d_avx2.c
+++ b/av1/common/x86/highbd_convolve_2d_avx2.c
@@ -50,8 +50,10 @@
(1 << (bd + 2 * FILTER_BITS - conv_params->round_0 - 1)));
const __m128i round_shift_y = _mm_cvtsi32_si128(conv_params->round_1);
- const __m128i bits = _mm_cvtsi32_si128(
- FILTER_BITS * 2 - conv_params->round_0 - conv_params->round_1);
+ const int bits =
+ FILTER_BITS * 2 - conv_params->round_0 - conv_params->round_1;
+ const __m128i round_shift_bits = _mm_cvtsi32_si128(bits);
+ const __m256i round_const_bits = _mm256_set1_epi32((1 << bits) >> 1);
const __m256i clip_pixel =
_mm256_set1_epi16(bd == 10 ? 1023 : (bd == 12 ? 4095 : 255));
const __m256i zero = _mm256_setzero_si256();
@@ -132,33 +134,40 @@
const __m256i res_a = convolve(s, coeffs_y);
__m256i res_a_round = _mm256_sra_epi32(
_mm256_add_epi32(res_a, round_const_y), round_shift_y);
- res_a_round = _mm256_sra_epi32(res_a_round, bits);
- res_a_round = _mm256_min_epi16(res_a_round, clip_pixel);
- res_a_round = _mm256_max_epi16(res_a_round, zero);
+
+ res_a_round = _mm256_sra_epi32(
+ _mm256_add_epi32(res_a_round, round_const_bits), round_shift_bits);
if (w - j > 4) {
const __m256i res_b = convolve(s + 4, coeffs_y);
__m256i res_b_round = _mm256_sra_epi32(
_mm256_add_epi32(res_b, round_const_y), round_shift_y);
- res_b_round = _mm256_sra_epi32(res_b_round, bits);
- res_b_round = _mm256_min_epi16(res_b_round, clip_pixel);
- res_b_round = _mm256_max_epi16(res_b_round, zero);
+ res_b_round =
+ _mm256_sra_epi32(_mm256_add_epi32(res_b_round, round_const_bits),
+ round_shift_bits);
- const __m256i res_16bit =
- _mm256_packs_epi32(res_a_round, res_b_round);
+ __m256i res_16bit = _mm256_packs_epi32(res_a_round, res_b_round);
+ res_16bit = _mm256_min_epi16(res_16bit, clip_pixel);
+ res_16bit = _mm256_max_epi16(res_16bit, zero);
- _mm_store_si128((__m128i *)&dst[i * dst_stride + j],
- _mm256_castsi256_si128(res_16bit));
- _mm_store_si128((__m128i *)&dst[i * dst_stride + j + dst_stride],
- _mm256_extracti128_si256(res_16bit, 1));
+ _mm_storeu_si128((__m128i *)&dst[i * dst_stride + j],
+ _mm256_castsi256_si128(res_16bit));
+ _mm_storeu_si128((__m128i *)&dst[i * dst_stride + j + dst_stride],
+ _mm256_extracti128_si256(res_16bit, 1));
} else if (w == 4) {
res_a_round = _mm256_packs_epi32(res_a_round, res_a_round);
+ res_a_round = _mm256_min_epi16(res_a_round, clip_pixel);
+ res_a_round = _mm256_max_epi16(res_a_round, zero);
+
_mm_storel_epi64((__m128i *)&dst[i * dst_stride + j],
_mm256_castsi256_si128(res_a_round));
_mm_storel_epi64((__m128i *)&dst[i * dst_stride + j + dst_stride],
_mm256_extracti128_si256(res_a_round, 1));
} else {
res_a_round = _mm256_packs_epi32(res_a_round, res_a_round);
+ res_a_round = _mm256_min_epi16(res_a_round, clip_pixel);
+ res_a_round = _mm256_max_epi16(res_a_round, zero);
+
xx_storel_32((__m128i *)&dst[i * dst_stride + j],
_mm256_castsi256_si128(res_a_round));
xx_storel_32((__m128i *)&dst[i * dst_stride + j + dst_stride],
diff --git a/test/av1_convolve_2d_test.cc b/test/av1_convolve_2d_test.cc
index 9d04418..fc03c16 100644
--- a/test/av1_convolve_2d_test.cc
+++ b/test/av1_convolve_2d_test.cc
@@ -173,6 +173,9 @@
INSTANTIATE_TEST_CASE_P(AVX2, AV1HighbdConvolve2DSrTest,
libaom_test::AV1HighbdConvolve2D::BuildParams(
av1_highbd_convolve_2d_sr_avx2, 1, 1, 0));
+INSTANTIATE_TEST_CASE_P(AVX2_X, AV1HighbdConvolve2DSrTest,
+ libaom_test::AV1HighbdConvolve2D::BuildParams(
+ av1_highbd_convolve_x_sr_avx2, 1, 0, 0));
#endif
#if CONFIG_JNT_COMP && HAVE_SSE4_1
TEST_P(AV1HighbdJntConvolve2DTest, CheckOutput) {
