[NORMATIVE jnt_comp] remove double rounding
Change-Id: Ib325e33bee8aa3a8445a7f61c55adfd3fb210792
diff --git a/av1/common/convolve.c b/av1/common/convolve.c
index 7645653..95458c1 100644
--- a/av1/common/convolve.c
+++ b/av1/common/convolve.c
@@ -703,7 +703,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
@@ -746,7 +745,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
@@ -788,7 +786,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
@@ -826,7 +823,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
@@ -900,7 +896,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
@@ -1065,7 +1060,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
@@ -1138,7 +1132,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
diff --git a/av1/common/reconinter.c b/av1/common/reconinter.c
index c8c58ce..bd3493f 100644
--- a/av1/common/reconinter.c
+++ b/av1/common/reconinter.c
@@ -1106,17 +1106,19 @@
xs, ys, xd);
} // for (ref = 0; ref < 1 + is_compound; ++ref)
if (conv_params.do_post_rounding) {
+ int round_bits = FILTER_BITS * 2 + is_compound - conv_params.round_0 -
+ conv_params.round_1;
+#if CONFIG_JNT_COMP
+ if (conv_params.use_jnt_comp_avg)
+ round_bits += DIST_PRECISION_BITS - 1;
+#endif // CONFIG_JNT_COMP
if (xd->cur_buf->flags & YV12_FLAG_HIGHBITDEPTH)
- av1_highbd_convolve_rounding(
- tmp_dst, tmp_dst_stride, dst, dst_buf->stride, b4_w, b4_h,
- FILTER_BITS * 2 + is_compound - conv_params.round_0 -
- conv_params.round_1,
- xd->bd);
+ av1_highbd_convolve_rounding(tmp_dst, tmp_dst_stride, dst,
+ dst_buf->stride, b4_w, b4_h,
+ round_bits, xd->bd);
else
- av1_convolve_rounding(
- tmp_dst, tmp_dst_stride, dst, dst_buf->stride, b4_w, b4_h,
- FILTER_BITS * 2 + is_compound - conv_params.round_0 -
- conv_params.round_1);
+ av1_convolve_rounding(tmp_dst, tmp_dst_stride, dst, dst_buf->stride,
+ b4_w, b4_h, round_bits);
}
++col;
}
@@ -1250,16 +1252,17 @@
// TODO(angiebird): This part needs optimization
if (conv_params.do_post_rounding) {
+ int round_bits = FILTER_BITS * 2 + is_compound - conv_params.round_0 -
+ conv_params.round_1;
+#if CONFIG_JNT_COMP
+ if (conv_params.use_jnt_comp_avg) round_bits += DIST_PRECISION_BITS - 1;
+#endif // CONFIG_JNT_COMP
if (xd->cur_buf->flags & YV12_FLAG_HIGHBITDEPTH)
- av1_highbd_convolve_rounding(
- tmp_dst, MAX_SB_SIZE, dst, dst_buf->stride, w, h,
- FILTER_BITS * 2 + is_compound - conv_params.round_0 -
- conv_params.round_1,
- xd->bd);
+ av1_highbd_convolve_rounding(tmp_dst, MAX_SB_SIZE, dst, dst_buf->stride,
+ w, h, round_bits, xd->bd);
else
av1_convolve_rounding(tmp_dst, MAX_SB_SIZE, dst, dst_buf->stride, w, h,
- FILTER_BITS * 2 + is_compound -
- conv_params.round_0 - conv_params.round_1);
+ round_bits);
}
}
}
diff --git a/av1/common/warped_motion.c b/av1/common/warped_motion.c
index cf41301..3b3eb8c 100644
--- a/av1/common/warped_motion.c
+++ b/av1/common/warped_motion.c
@@ -516,7 +516,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
*p += sum * conv_params->bck_offset;
- *p >>= (DIST_PRECISION_BITS - 1);
} else {
*p = sum * conv_params->fwd_offset;
}
@@ -820,7 +819,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
*p += sum * conv_params->bck_offset;
- *p >>= (DIST_PRECISION_BITS - 1);
} else {
*p = sum * conv_params->fwd_offset;
}
diff --git a/av1/common/x86/av1_convolve_scale_sse4.c b/av1/common/x86/av1_convolve_scale_sse4.c
index 7323403..e53717d 100644
--- a/av1/common/x86/av1_convolve_scale_sse4.c
+++ b/av1/common/x86/av1_convolve_scale_sse4.c
@@ -314,10 +314,8 @@
__m128i result;
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
- result =
- _mm_srai_epi32(_mm_add_epi32(_mm_loadu_si128((__m128i *)dst_x),
- _mm_mullo_epi32(subbed, bck_offset)),
- DIST_PRECISION_BITS - 1);
+ result = _mm_add_epi32(_mm_loadu_si128((__m128i *)dst_x),
+ _mm_mullo_epi32(subbed, bck_offset));
} else {
result = _mm_mullo_epi32(subbed, fwd_offset);
}
@@ -344,8 +342,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
-
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
@@ -430,10 +426,8 @@
__m128i result;
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
- result =
- _mm_srai_epi32(_mm_add_epi32(_mm_loadu_si128((__m128i *)dst_x),
- _mm_mullo_epi32(subbed, bck_offset)),
- DIST_PRECISION_BITS - 1);
+ result = _mm_add_epi32(_mm_loadu_si128((__m128i *)dst_x),
+ _mm_mullo_epi32(subbed, bck_offset));
} else {
result = _mm_mullo_epi32(subbed, fwd_offset);
}
@@ -460,8 +454,6 @@
if (conv_params->use_jnt_comp_avg) {
if (conv_params->do_average) {
dst[y * dst_stride + x] += res * conv_params->bck_offset;
-
- dst[y * dst_stride + x] >>= (DIST_PRECISION_BITS - 1);
} else {
dst[y * dst_stride + x] = res * conv_params->fwd_offset;
}
diff --git a/av1/common/x86/convolve_2d_sse2.c b/av1/common/x86/convolve_2d_sse2.c
index 111bc85..e1b731a 100644
--- a/av1/common/x86/convolve_2d_sse2.c
+++ b/av1/common/x86/convolve_2d_sse2.c
@@ -695,22 +695,22 @@
__m128i mul = _mm_mullo_epi16(d32_0, wt1);
__m128i weighted_res = _mm_sll_epi32(mul, left_shift);
__m128i sum = _mm_add_epi32(_mm_loadu_si128(p + 0), weighted_res);
- d32_0 = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ d32_0 = sum;
mul = _mm_mullo_epi16(d32_1, wt1);
weighted_res = _mm_sll_epi32(mul, left_shift);
sum = _mm_add_epi32(_mm_loadu_si128(p + 1), weighted_res);
- d32_1 = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ d32_1 = sum;
mul = _mm_mullo_epi16(d32_2, wt1);
weighted_res = _mm_sll_epi32(mul, left_shift);
sum = _mm_add_epi32(_mm_loadu_si128(p + 2), weighted_res);
- d32_2 = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ d32_2 = sum;
mul = _mm_mullo_epi16(d32_3, wt1);
weighted_res = _mm_sll_epi32(mul, left_shift);
sum = _mm_add_epi32(_mm_loadu_si128(p + 3), weighted_res);
- d32_3 = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ d32_3 = sum;
} else {
d32_0 = _mm_sll_epi32(_mm_mullo_epi16(d32_0, wt0), left_shift);
d32_1 = _mm_sll_epi32(_mm_mullo_epi16(d32_1, wt0), left_shift);
@@ -757,12 +757,12 @@
__m128i mul = _mm_mullo_epi16(d32_0, wt1);
__m128i weighted_res = _mm_sll_epi32(mul, left_shift);
__m128i sum = _mm_add_epi32(_mm_loadu_si128(p + 0), weighted_res);
- d32_0 = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ d32_0 = sum;
mul = _mm_mullo_epi16(d32_1, wt1);
weighted_res = _mm_sll_epi32(mul, left_shift);
sum = _mm_add_epi32(_mm_loadu_si128(p + 1), weighted_res);
- d32_1 = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ d32_1 = sum;
} else {
d32_0 = _mm_sll_epi32(_mm_mullo_epi16(d32_0, wt0), left_shift);
d32_1 = _mm_sll_epi32(_mm_mullo_epi16(d32_1, wt0), left_shift);
@@ -798,7 +798,7 @@
__m128i mul = _mm_mullo_epi16(d32_0, wt1);
__m128i weighted_res = _mm_sll_epi32(mul, left_shift);
__m128i sum = _mm_add_epi32(_mm_loadu_si128(p + 0), weighted_res);
- d32_0 = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ d32_0 = sum;
} else {
d32_0 = _mm_sll_epi32(_mm_mullo_epi16(d32_0, wt0), left_shift);
}
@@ -829,7 +829,7 @@
__m128i mul = _mm_mullo_epi16(d32_0, wt1);
__m128i weighted_res = _mm_sll_epi32(mul, left_shift);
__m128i sum = _mm_add_epi32(_mm_loadl_epi64(p), weighted_res);
- d32_0 = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ d32_0 = sum;
} else {
d32_0 = _mm_sll_epi32(_mm_mullo_epi16(d32_0, wt0), left_shift);
}
diff --git a/av1/common/x86/highbd_convolve_2d_sse4.c b/av1/common/x86/highbd_convolve_2d_sse4.c
index 853ecfd..80adce9 100644
--- a/av1/common/x86/highbd_convolve_2d_sse4.c
+++ b/av1/common/x86/highbd_convolve_2d_sse4.c
@@ -200,13 +200,9 @@
_mm_loadu_si128(p + 0), _mm_mullo_epi32(res_lo_round, wt1));
const __m128i jnt_sum_hi = _mm_add_epi32(
_mm_loadu_si128(p + 1), _mm_mullo_epi32(res_hi_round, wt1));
- const __m128i final_lo =
- _mm_srai_epi32(jnt_sum_lo, DIST_PRECISION_BITS - 1);
- const __m128i final_hi =
- _mm_srai_epi32(jnt_sum_hi, DIST_PRECISION_BITS - 1);
- _mm_storeu_si128(p + 0, final_lo);
- _mm_storeu_si128(p + 1, final_hi);
+ _mm_storeu_si128(p + 0, jnt_sum_lo);
+ _mm_storeu_si128(p + 1, jnt_sum_hi);
} else {
_mm_storeu_si128(p + 0, _mm_mullo_epi32(res_lo_round, wt0));
_mm_storeu_si128(p + 1, _mm_mullo_epi32(res_hi_round, wt0));
diff --git a/av1/common/x86/highbd_warp_plane_sse4.c b/av1/common/x86/highbd_warp_plane_sse4.c
index 0d4d17f..0cd438a 100644
--- a/av1/common/x86/highbd_warp_plane_sse4.c
+++ b/av1/common/x86/highbd_warp_plane_sse4.c
@@ -318,7 +318,7 @@
if (comp_avg) {
const __m128i sum = _mm_add_epi32(_mm_loadu_si128(p),
_mm_mullo_epi32(res_lo, wt1));
- res_lo = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ res_lo = sum;
} else {
res_lo = _mm_mullo_epi32(res_lo, wt0);
}
@@ -342,7 +342,7 @@
if (comp_avg) {
const __m128i sum = _mm_add_epi32(_mm_loadu_si128(p + 1),
_mm_mullo_epi32(res_hi, wt1));
- res_hi = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ res_hi = sum;
} else {
res_hi = _mm_mullo_epi32(res_hi, wt0);
}
diff --git a/av1/common/x86/jnt_convolve_sse4.c b/av1/common/x86/jnt_convolve_sse4.c
index 9de2744..bc23365 100644
--- a/av1/common/x86/jnt_convolve_sse4.c
+++ b/av1/common/x86/jnt_convolve_sse4.c
@@ -88,12 +88,11 @@
static INLINE void mult_add_store(CONV_BUF_TYPE *const dst,
const __m128i *const res,
const __m128i *const avg_mask,
- const __m128i *const wt, int shift) {
+ const __m128i *const wt) {
__m128i d;
d = _mm_load_si128((__m128i *)dst);
d = _mm_and_si128(d, *avg_mask);
d = _mm_add_epi32(d, _mm_mullo_epi32(*res, *wt));
- if (shift) d = _mm_srai_epi32(d, DIST_PRECISION_BITS - 1);
_mm_store_si128((__m128i *)dst, d);
}
@@ -157,8 +156,7 @@
res_shift =
_mm_sra_epi32(_mm_add_epi32(res_shift, round_const), round_shift);
if (conv_params->use_jnt_comp_avg)
- mult_add_store(dst, &res_shift, &avg_mask, &wt,
- conv_params->do_average);
+ mult_add_store(dst, &res_shift, &avg_mask, &wt);
else
add_store(dst, &res_shift, &avg_mask);
src_ptr += src_stride;
@@ -169,8 +167,7 @@
res_shift =
_mm_sra_epi32(_mm_add_epi32(res_shift, round_const), round_shift);
if (conv_params->use_jnt_comp_avg)
- mult_add_store(dst, &res_shift, &avg_mask, &wt,
- conv_params->do_average);
+ mult_add_store(dst, &res_shift, &avg_mask, &wt);
else
add_store(dst, &res_shift, &avg_mask);
src_ptr += src_stride;
@@ -229,9 +226,9 @@
round_shift);
if (conv_params->use_jnt_comp_avg) {
mult_add_store(dst + i * dst_stride + j + 0, &res_lo_shift, &avg_mask,
- &wt, conv_params->do_average);
+ &wt);
mult_add_store(dst + i * dst_stride + j + 4, &res_hi_shift, &avg_mask,
- &wt, conv_params->do_average);
+ &wt);
} else {
add_store(dst + i * dst_stride + j + 0, &res_lo_shift, &avg_mask);
add_store(dst + i * dst_stride + j + 4, &res_hi_shift, &avg_mask);
@@ -248,9 +245,9 @@
round_shift);
if (conv_params->use_jnt_comp_avg) {
mult_add_store(dst + i * dst_stride + j + 0, &res_lo_shift, &avg_mask,
- &wt, conv_params->do_average);
+ &wt);
mult_add_store(dst + i * dst_stride + j + 4, &res_hi_shift, &avg_mask,
- &wt, conv_params->do_average);
+ &wt);
} else {
add_store(dst + i * dst_stride + j + 0, &res_lo_shift, &avg_mask);
add_store(dst + i * dst_stride + j + 4, &res_hi_shift, &avg_mask);
@@ -317,8 +314,7 @@
// Accumulate values into the destination buffer
if (conv_params->use_jnt_comp_avg)
- mult_add_store(dst, &res_lo_shift, &avg_mask, &wt,
- conv_params->do_average);
+ mult_add_store(dst, &res_lo_shift, &avg_mask, &wt);
else
add_store(dst, &res_lo_shift, &avg_mask);
src_ptr += src_stride;
@@ -361,9 +357,9 @@
// Accumulate values into the destination buffer
if (conv_params->use_jnt_comp_avg) {
mult_add_store(dst + i * dst_stride + j + 0, &res_lo_shift, &avg_mask,
- &wt, conv_params->do_average);
+ &wt);
mult_add_store(dst + i * dst_stride + j + 4, &res_hi_shift, &avg_mask,
- &wt, conv_params->do_average);
+ &wt);
} else {
add_store(dst + i * dst_stride + j + 0, &res_lo_shift, &avg_mask);
add_store(dst + i * dst_stride + j + 4, &res_hi_shift, &avg_mask);
@@ -557,17 +553,12 @@
// original c function at: av1/common/convolve.c: av1_convolve_2d_c
__m128i *const p = (__m128i *)&dst[i * dst_stride + j];
if (do_average) {
- _mm_storeu_si128(
- p + 0, _mm_srai_epi32(
- _mm_add_epi32(_mm_loadu_si128(p + 0),
- _mm_mullo_epi32(res_lo_round, wt1)),
- DIST_PRECISION_BITS - 1));
-
- _mm_storeu_si128(
- p + 1, _mm_srai_epi32(
- _mm_add_epi32(_mm_loadu_si128(p + 1),
- _mm_mullo_epi32(res_hi_round, wt1)),
- DIST_PRECISION_BITS - 1));
+ _mm_storeu_si128(p + 0,
+ _mm_add_epi32(_mm_loadu_si128(p + 0),
+ _mm_mullo_epi32(res_lo_round, wt1)));
+ _mm_storeu_si128(p + 1,
+ _mm_add_epi32(_mm_loadu_si128(p + 1),
+ _mm_mullo_epi32(res_hi_round, wt1)));
} else {
_mm_storeu_si128(p + 0, _mm_mullo_epi32(res_lo_round, wt0));
_mm_storeu_si128(p + 1, _mm_mullo_epi32(res_hi_round, wt0));
diff --git a/av1/common/x86/warp_plane_sse4.c b/av1/common/x86/warp_plane_sse4.c
index 94b5f09..043e6a7 100644
--- a/av1/common/x86/warp_plane_sse4.c
+++ b/av1/common/x86/warp_plane_sse4.c
@@ -315,7 +315,7 @@
if (comp_avg) {
const __m128i sum = _mm_add_epi32(_mm_loadu_si128(p),
_mm_mullo_epi32(res_lo, wt1));
- res_lo = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ res_lo = sum;
} else {
res_lo = _mm_mullo_epi32(res_lo, wt0);
}
@@ -337,7 +337,7 @@
if (comp_avg) {
const __m128i sum = _mm_add_epi32(_mm_loadu_si128(p + 1),
_mm_mullo_epi32(res_hi, wt1));
- res_hi = _mm_srai_epi32(sum, DIST_PRECISION_BITS - 1);
+ res_hi = sum;
} else {
res_hi = _mm_mullo_epi32(res_hi, wt0);
}
