ref: f7d1486f4890431a49802817818a5ae0a4c3c8dc
parent: 1814463864a338b87305099f13d4938a8b3db822
author: Johann <johannkoenig@google.com>
date: Mon May 1 10:10:54 EDT 2017
neon variance: process 16 values at a time Read in a Q register. Works on blocks of 16 and larger. Improvement of about 20% for 64x64. The smaller blocks are faster, but don't have quite the same level of improvement. 16x32 is only about 5% BUG=webm:1422 Change-Id: Ie11a877c7b839e66690a48117a46657b2ac82d4b
--- a/vpx_dsp/arm/variance_neon.c
+++ b/vpx_dsp/arm/variance_neon.c
@@ -32,34 +32,47 @@
}
// w * h must be less than 2048 or sum_s16 may overflow.
-// Process a block of any size where the width is divisible by 8.
-// TODO(johannkoenig): bump this up to 16 at a time.
-static void variance_neon_w8(const uint8_t *a, int a_stride, const uint8_t *b,
- int b_stride, int w, int h, uint32_t *sse,
- int *sum) {+// Process a block of any size where the width is divisible by 16.
+static void variance_neon_w16(const uint8_t *a, int a_stride, const uint8_t *b,
+ int b_stride, int w, int h, uint32_t *sse,
+ int *sum) {int i, j;
- int16x8_t v_sum = vdupq_n_s16(0);
- int32x4_t v_sse_lo = vdupq_n_s32(0);
- int32x4_t v_sse_hi = vdupq_n_s32(0);
+ int16x8_t sum_s16 = vdupq_n_s16(0);
+ int32x4_t sse_lo_s32 = vdupq_n_s32(0);
+ int32x4_t sse_hi_s32 = vdupq_n_s32(0);
for (i = 0; i < h; ++i) {- for (j = 0; j < w; j += 8) {- const uint8x8_t v_a = vld1_u8(&a[j]);
- const uint8x8_t v_b = vld1_u8(&b[j]);
- const uint16x8_t v_diff = vsubl_u8(v_a, v_b);
- const int16x8_t sv_diff = vreinterpretq_s16_u16(v_diff);
- v_sum = vaddq_s16(v_sum, sv_diff);
- v_sse_lo =
- vmlal_s16(v_sse_lo, vget_low_s16(sv_diff), vget_low_s16(sv_diff));
- v_sse_hi =
- vmlal_s16(v_sse_hi, vget_high_s16(sv_diff), vget_high_s16(sv_diff));
+ for (j = 0; j < w; j += 16) {+ const uint8x16_t a_u8 = vld1q_u8(a + j);
+ const uint8x16_t b_u8 = vld1q_u8(b + j);
+
+ const uint16x8_t diff_lo_u16 =
+ vsubl_u8(vget_low_u8(a_u8), vget_low_u8(b_u8));
+ const uint16x8_t diff_hi_u16 =
+ vsubl_u8(vget_high_u8(a_u8), vget_high_u8(b_u8));
+
+ const int16x8_t diff_lo_s16 = vreinterpretq_s16_u16(diff_lo_u16);
+ const int16x8_t diff_hi_s16 = vreinterpretq_s16_u16(diff_hi_u16);
+
+ sum_s16 = vaddq_s16(sum_s16, diff_lo_s16);
+ sum_s16 = vaddq_s16(sum_s16, diff_hi_s16);
+
+ sse_lo_s32 = vmlal_s16(sse_lo_s32, vget_low_s16(diff_lo_s16),
+ vget_low_s16(diff_lo_s16));
+ sse_lo_s32 = vmlal_s16(sse_lo_s32, vget_high_s16(diff_lo_s16),
+ vget_high_s16(diff_lo_s16));
+
+ sse_hi_s32 = vmlal_s16(sse_hi_s32, vget_low_s16(diff_hi_s16),
+ vget_low_s16(diff_hi_s16));
+ sse_hi_s32 = vmlal_s16(sse_hi_s32, vget_high_s16(diff_hi_s16),
+ vget_high_s16(diff_hi_s16));
}
a += a_stride;
b += b_stride;
}
- *sum = horizontal_add_s16x8(v_sum);
- *sse = (unsigned int)horizontal_add_s32x4(vaddq_s32(v_sse_lo, v_sse_hi));
+ *sum = horizontal_add_s16x8(sum_s16);
+ *sse = (unsigned int)horizontal_add_s32x4(vaddq_s32(sse_lo_s32, sse_hi_s32));
}
// w * h must be less than 2048 or sum_s16 may overflow.
@@ -106,7 +119,7 @@
void vpx_get16x16var_neon(const uint8_t *a, int a_stride, const uint8_t *b,
int b_stride, unsigned int *sse, int *sum) {- variance_neon_w8(a, a_stride, b, b_stride, 16, 16, sse, sum);
+ variance_neon_w16(a, a_stride, b, b_stride, 16, 16, sse, sum);
}
#define varianceNxM(n, m, shift) \
@@ -117,7 +130,7 @@
if (n == 8) \
variance_neon_w8x2(a, a_stride, b, b_stride, m, sse, &sum); \
else \
- variance_neon_w8(a, a_stride, b, b_stride, n, m, sse, &sum); \
+ variance_neon_w16(a, a_stride, b, b_stride, n, m, sse, &sum); \
if (n * m < 16 * 16) \
return *sse - ((sum * sum) >> shift); \
else \
@@ -138,9 +151,9 @@
unsigned int *sse) {int sum1, sum2;
uint32_t sse1, sse2;
- variance_neon_w8(a, a_stride, b, b_stride, 32, 32, &sse1, &sum1);
- variance_neon_w8(a + (32 * a_stride), a_stride, b + (32 * b_stride), b_stride,
- 32, 32, &sse2, &sum2);
+ variance_neon_w16(a, a_stride, b, b_stride, 32, 32, &sse1, &sum1);
+ variance_neon_w16(a + (32 * a_stride), a_stride, b + (32 * b_stride),
+ b_stride, 32, 32, &sse2, &sum2);
*sse = sse1 + sse2;
sum1 += sum2;
return *sse - (unsigned int)(((int64_t)sum1 * sum1) >> 11);
@@ -151,9 +164,9 @@
unsigned int *sse) {int sum1, sum2;
uint32_t sse1, sse2;
- variance_neon_w8(a, a_stride, b, b_stride, 64, 16, &sse1, &sum1);
- variance_neon_w8(a + (16 * a_stride), a_stride, b + (16 * b_stride), b_stride,
- 64, 16, &sse2, &sum2);
+ variance_neon_w16(a, a_stride, b, b_stride, 64, 16, &sse1, &sum1);
+ variance_neon_w16(a + (16 * a_stride), a_stride, b + (16 * b_stride),
+ b_stride, 64, 16, &sse2, &sum2);
*sse = sse1 + sse2;
sum1 += sum2;
return *sse - (unsigned int)(((int64_t)sum1 * sum1) >> 11);
@@ -165,19 +178,19 @@
int sum1, sum2;
uint32_t sse1, sse2;
- variance_neon_w8(a, a_stride, b, b_stride, 64, 16, &sse1, &sum1);
- variance_neon_w8(a + (16 * a_stride), a_stride, b + (16 * b_stride), b_stride,
- 64, 16, &sse2, &sum2);
+ variance_neon_w16(a, a_stride, b, b_stride, 64, 16, &sse1, &sum1);
+ variance_neon_w16(a + (16 * a_stride), a_stride, b + (16 * b_stride),
+ b_stride, 64, 16, &sse2, &sum2);
sse1 += sse2;
sum1 += sum2;
- variance_neon_w8(a + (16 * 2 * a_stride), a_stride, b + (16 * 2 * b_stride),
- b_stride, 64, 16, &sse2, &sum2);
+ variance_neon_w16(a + (16 * 2 * a_stride), a_stride, b + (16 * 2 * b_stride),
+ b_stride, 64, 16, &sse2, &sum2);
sse1 += sse2;
sum1 += sum2;
- variance_neon_w8(a + (16 * 3 * a_stride), a_stride, b + (16 * 3 * b_stride),
- b_stride, 64, 16, &sse2, &sum2);
+ variance_neon_w16(a + (16 * 3 * a_stride), a_stride, b + (16 * 3 * b_stride),
+ b_stride, 64, 16, &sse2, &sum2);
*sse = sse1 + sse2;
sum1 += sum2;
return *sse - (unsigned int)(((int64_t)sum1 * sum1) >> 12);