/external/XNNPACK/src/f32-avgpool/ |
D | 9x-minmax-sse-c4.c | 133 const __m128 vsum01678 = _mm_add_ps(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9x__sse_c4() local 134 const __m128 vsum = _mm_add_ps(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9x__sse_c4() 161 const __m128 vsum01678 = _mm_add_ps(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9x__sse_c4() local 162 const __m128 vsum = _mm_add_ps(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9x__sse_c4()
|
D | 9x-minmax-neon-c4.c | 124 const float32x4_t vsum01678 = vaddq_f32(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9x__neon_c4() local 125 const float32x4_t vsum = vaddq_f32(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9x__neon_c4() 152 const float32x4_t vsum01678 = vaddq_f32(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9x__neon_c4() local 153 const float32x4_t vsum = vaddq_f32(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9x__neon_c4()
|
D | 9x-minmax-wasmsimd-arm-c4.c | 133 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_arm_c4() local 134 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_arm_c4() 162 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_arm_c4() local 163 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_arm_c4()
|
D | 9x-minmax-wasmsimd-x86-c4.c | 133 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_x86_c4() local 134 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_x86_c4() 164 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_x86_c4() local 165 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_x86_c4()
|
D | 9x-minmax-wasm-c1.c | 123 const float vsum01678 = vsum018 + vsum67; in xnn_f32_avgpool_minmax_ukernel_9x__wasm_c1() local 124 const float vsum = vsum2345 + vsum01678; in xnn_f32_avgpool_minmax_ukernel_9x__wasm_c1()
|
D | 9x-minmax-scalar-c1.c | 123 const float vsum01678 = vsum018 + vsum67; in xnn_f32_avgpool_minmax_ukernel_9x__scalar_c1() local 124 const float vsum = vsum2345 + vsum01678; in xnn_f32_avgpool_minmax_ukernel_9x__scalar_c1()
|
D | 9p8x-minmax-wasm-c1.c | 100 const float vsum01678 = vsum018 + vsum67; in xnn_f32_avgpool_minmax_ukernel_9p8x__wasm_c1() local 101 const float vsum = vsum2345 + vsum01678; in xnn_f32_avgpool_minmax_ukernel_9p8x__wasm_c1()
|
D | 9p8x-minmax-scalar-c1.c | 100 const float vsum01678 = vsum018 + vsum67; in xnn_f32_avgpool_minmax_ukernel_9p8x__scalar_c1() local 101 const float vsum = vsum2345 + vsum01678; in xnn_f32_avgpool_minmax_ukernel_9p8x__scalar_c1()
|
D | 9p8x-minmax-neon-c4.c | 100 const float32x4_t vsum01678 = vaddq_f32(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9p8x__neon_c4() local 101 const float32x4_t vsum = vaddq_f32(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9p8x__neon_c4()
|
D | 9p8x-minmax-wasmsimd-arm-c4.c | 109 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_avgpool_minmax_ukernel_9p8x__wasmsimd_arm_c4() local 110 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_avgpool_minmax_ukernel_9p8x__wasmsimd_arm_c4()
|
/external/XNNPACK/src/f32-pavgpool/ |
D | 9x-minmax-neon-c4.c | 126 const float32x4_t vsum01678 = vaddq_f32(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9x__neon_c4() local 127 const float32x4_t vsum = vaddq_f32(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9x__neon_c4() 154 const float32x4_t vsum01678 = vaddq_f32(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9x__neon_c4() local 155 const float32x4_t vsum = vaddq_f32(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9x__neon_c4()
|
D | 9x-minmax-sse-c4.c | 136 const __m128 vsum01678 = _mm_add_ps(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9x__sse_c4() local 137 const __m128 vsum = _mm_add_ps(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9x__sse_c4() 164 const __m128 vsum01678 = _mm_add_ps(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9x__sse_c4() local 165 const __m128 vsum = _mm_add_ps(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9x__sse_c4()
|
D | 9x-minmax-wasmsimd-arm-c4.c | 136 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_arm_c4() local 137 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_arm_c4() 165 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_arm_c4() local 166 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_arm_c4()
|
D | 9x-minmax-wasmsimd-x86-c4.c | 136 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_x86_c4() local 137 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_x86_c4() 167 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_x86_c4() local 168 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_x86_c4()
|
D | 9x-minmax-scalar-c1.c | 125 const float vsum01678 = vsum018 + vsum67; in xnn_f32_pavgpool_minmax_ukernel_9x__scalar_c1() local 126 const float vsum = vsum2345 + vsum01678; in xnn_f32_pavgpool_minmax_ukernel_9x__scalar_c1()
|
D | 9x-minmax-wasm-c1.c | 125 const float vsum01678 = vsum018 + vsum67; in xnn_f32_pavgpool_minmax_ukernel_9x__wasm_c1() local 126 const float vsum = vsum2345 + vsum01678; in xnn_f32_pavgpool_minmax_ukernel_9x__wasm_c1()
|
D | 9p8x-minmax-wasm-c1.c | 100 const float vsum01678 = vsum018 + vsum67; in xnn_f32_pavgpool_minmax_ukernel_9p8x__wasm_c1() local 101 const float vsum = vsum2345 + vsum01678; in xnn_f32_pavgpool_minmax_ukernel_9p8x__wasm_c1()
|
D | 9p8x-minmax-scalar-c1.c | 100 const float vsum01678 = vsum018 + vsum67; in xnn_f32_pavgpool_minmax_ukernel_9p8x__scalar_c1() local 101 const float vsum = vsum2345 + vsum01678; in xnn_f32_pavgpool_minmax_ukernel_9p8x__scalar_c1()
|
D | 9p8x-minmax-sse-c4.c | 109 const __m128 vsum01678 = _mm_add_ps(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9p8x__sse_c4() local 110 const __m128 vsum = _mm_add_ps(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9p8x__sse_c4()
|
D | 9p8x-minmax-wasmsimd-arm-c4.c | 109 const v128_t vsum01678 = wasm_f32x4_add(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9p8x__wasmsimd_arm_c4() local 110 const v128_t vsum = wasm_f32x4_add(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9p8x__wasmsimd_arm_c4()
|
D | 9p8x-minmax-neon-c4.c | 100 const float32x4_t vsum01678 = vaddq_f32(vsum018, vsum67); in xnn_f32_pavgpool_minmax_ukernel_9p8x__neon_c4() local 101 const float32x4_t vsum = vaddq_f32(vsum2345, vsum01678); in xnn_f32_pavgpool_minmax_ukernel_9p8x__neon_c4()
|
/external/XNNPACK/src/qu8-avgpool/ |
D | 9x-minmax-neon-c8.c | 135 const uint16x8_t vsum01678 = vaddq_u16(vsum018, vsum67); in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8() local 136 const uint16x8_t vsum = vaddq_u16(vsum2345, vsum01678); in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8() 208 const uint16x8_t vsum01678 = vaddq_u16(vsum018, vsum67); in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8() local 209 const uint16x8_t vsum = vaddq_u16(vsum2345, vsum01678); in xnn_qu8_avgpool_minmax_ukernel_9x__neon_c8()
|
D | 9x-minmax-scalar-c1.c | 129 const uint32_t vsum01678 = vsum018 + vsum67; in xnn_qu8_avgpool_minmax_ukernel_9x__scalar_c1() local 130 const uint32_t vsum = vsum2345 + vsum01678; in xnn_qu8_avgpool_minmax_ukernel_9x__scalar_c1()
|
D | 9x-minmax-sse2-c8.c | 139 const __m128i vsum01678 = _mm_add_epi16(vsum018, vsum67); in xnn_qu8_avgpool_minmax_ukernel_9x__sse2_c8() local 140 const __m128i vsum = _mm_add_epi16(vsum2345, vsum01678); in xnn_qu8_avgpool_minmax_ukernel_9x__sse2_c8() 214 const __m128i vsum01678 = _mm_add_epi16(vsum018, vsum67); in xnn_qu8_avgpool_minmax_ukernel_9x__sse2_c8() local 215 const __m128i vsum = _mm_add_epi16(vsum2345, vsum01678); in xnn_qu8_avgpool_minmax_ukernel_9x__sse2_c8()
|
D | 9p8x-minmax-scalar-c1.c | 107 const uint32_t vsum01678 = vsum018 + vsum67; in xnn_qu8_avgpool_minmax_ukernel_9p8x__scalar_c1() local 109 vacc += (int32_t) vsum01678; in xnn_qu8_avgpool_minmax_ukernel_9p8x__scalar_c1()
|