| /external/XNNPACK/src/f32-avgpool/ |
| D | 9x-minmax-wasmsimd-arm-c4.c | 124 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_arm_c4() local
154 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_arm_c4() local
|
| D | 9x-minmax-sse-c4.c | 124 const __m128 vi8 = _mm_loadu_ps(i8); in xnn_f32_avgpool_minmax_ukernel_9x__sse_c4() local
153 const __m128 vi8 = _mm_loadu_ps(i8); in xnn_f32_avgpool_minmax_ukernel_9x__sse_c4() local
|
| D | 9x-minmax-wasmsimd-x86-c4.c | 124 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_x86_c4() local
154 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_avgpool_minmax_ukernel_9x__wasmsimd_x86_c4() local
|
| D | 9x-minmax-neon-c4.c | 116 const float32x4_t vi8 = vld1q_f32(i8); i8 += 4; in xnn_f32_avgpool_minmax_ukernel_9x__neon_c4() local
144 const float32x4_t vi8 = vld1q_f32(i8); in xnn_f32_avgpool_minmax_ukernel_9x__neon_c4() local
|
| /external/XNNPACK/src/f32-argmaxpool/ |
| D | 9x-wasmsimd-c4.c | 92 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_argmaxpool_ukernel_9x__wasmsimd_c4() local
144 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_argmaxpool_ukernel_9x__wasmsimd_c4() local
|
| D | 9x-sse2-c4.c | 91 const __m128 vi8 = _mm_loadu_ps(i8); in xnn_f32_argmaxpool_ukernel_9x__sse2_c4() local
143 const __m128 vi8 = _mm_loadu_ps(i8); in xnn_f32_argmaxpool_ukernel_9x__sse2_c4() local
|
| D | 9x-neon-c4.c | 83 const float32x4_t vi8 = vld1q_f32(i8); i8 += 4; in xnn_f32_argmaxpool_ukernel_9x__neon_c4() local
132 const float32x4_t vi8 = vld1q_f32(i8); in xnn_f32_argmaxpool_ukernel_9x__neon_c4() local
|
| D | 9x-scalar-c1.c | 82 const float vi8 = *i8++; in xnn_f32_argmaxpool_ukernel_9x__scalar_c1() local
|
| /external/XNNPACK/src/f16-avgpool/ |
| D | 9x-minmax-neonfp16arith-c8.c | 116 const float16x8_t vi8 = vld1q_f16(i8); i8 += 8; in xnn_f16_avgpool_minmax_ukernel_9x__neonfp16arith_c8() local
144 const float16x8_t vi8 = vld1q_f16(i8); in xnn_f16_avgpool_minmax_ukernel_9x__neonfp16arith_c8() local
|
| D | 9x-minmax-f16c-c8.c | 126 const __m256 vi8 = _mm256_cvtph_ps(_mm_loadu_si128((const __m128i*) i8)); in xnn_f16_avgpool_minmax_ukernel_9x__f16c_c8() local
156 const __m256 vi8 = _mm256_cvtph_ps(_mm_loadu_si128((const __m128i*) i8)); in xnn_f16_avgpool_minmax_ukernel_9x__f16c_c8() local
|
| /external/XNNPACK/src/f32-pavgpool/ |
| D | 9x-minmax-wasmsimd-arm-c4.c | 127 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_arm_c4() local
157 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_arm_c4() local
|
| D | 9x-minmax-wasmsimd-x86-c4.c | 127 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_x86_c4() local
157 const v128_t vi8 = wasm_v128_load(i8); in xnn_f32_pavgpool_minmax_ukernel_9x__wasmsimd_x86_c4() local
|
| D | 9x-minmax-neon-c4.c | 118 const float32x4_t vi8 = vld1q_f32(i8); i8 += 4; in xnn_f32_pavgpool_minmax_ukernel_9x__neon_c4() local
146 const float32x4_t vi8 = vld1q_f32(i8); in xnn_f32_pavgpool_minmax_ukernel_9x__neon_c4() local
|
| D | 9x-minmax-sse-c4.c | 127 const __m128 vi8 = _mm_loadu_ps(i8); in xnn_f32_pavgpool_minmax_ukernel_9x__sse_c4() local
156 const __m128 vi8 = _mm_loadu_ps(i8); in xnn_f32_pavgpool_minmax_ukernel_9x__sse_c4() local
|
| /external/pytorch/aten/src/ATen/native/quantized/cpu/qnnpack/src/sdwconv/ |
| D | up4x9-psimd.c | 83 const psimd_f32 vi8 = psimd_load_f32(i8); in pytorch_sdwconv_ukernel_up4x9__psimd() local
141 const psimd_f32 vi8 = psimd_load_f32(i8); in pytorch_sdwconv_ukernel_up4x9__psimd() local
|
| /external/XNNPACK/src/f16-pavgpool/ |
| D | 9x-minmax-neonfp16arith-c8.c | 118 const float16x8_t vi8 = vld1q_f16(i8); i8 += 8; in xnn_f16_pavgpool_minmax_ukernel_9x__neonfp16arith_c8() local
146 const float16x8_t vi8 = vld1q_f16(i8); in xnn_f16_pavgpool_minmax_ukernel_9x__neonfp16arith_c8() local
|
| D | 9x-minmax-avx2-c8.c | 129 const __m256 vi8 = _mm256_cvtph_ps(_mm_loadu_si128((const __m128i*) i8)); in xnn_f16_pavgpool_minmax_ukernel_9x__avx2_c8() local
159 const __m256 vi8 = _mm256_cvtph_ps(_mm_loadu_si128((const __m128i*) i8)); in xnn_f16_pavgpool_minmax_ukernel_9x__avx2_c8() local
|
| /external/XNNPACK/src/f32-maxpool/ |
| D | 9p8x-minmax-scalar-c1.c | 85 const float vi8 = *i8++; in xnn_f32_maxpool_minmax_ukernel_9p8x__scalar_c1() local
153 const float vi8 = *o; in xnn_f32_maxpool_minmax_ukernel_9p8x__scalar_c1() local
|
| D | 9p8x-minmax-wasm-c1.c | 85 const float vi8 = *i8++; in xnn_f32_maxpool_minmax_ukernel_9p8x__wasm_c1() local
153 const float vi8 = *o; in xnn_f32_maxpool_minmax_ukernel_9p8x__wasm_c1() local
|
| D | 9p8x-minmax-sse-c4.c | 94 const __m128 vi8 = _mm_loadu_ps(i8); in xnn_f32_maxpool_minmax_ukernel_9p8x__sse_c4() local
127 const __m128 vi8 = _mm_loadu_ps(i8); in xnn_f32_maxpool_minmax_ukernel_9p8x__sse_c4() local
|
| /external/XNNPACK/src/u8-maxpool/ |
| D | 9p8x-minmax-scalar-c1.c | 85 const uint32_t vi8 = (uint32_t) *i8++; in xnn_u8_maxpool_minmax_ukernel_9p8x__scalar_c1() local
154 const uint32_t vi8 = (uint32_t) *o; in xnn_u8_maxpool_minmax_ukernel_9p8x__scalar_c1() local
|
| /external/XNNPACK/src/s8-maxpool/ |
| D | 9p8x-minmax-scalar-c1.c | 85 const int32_t vi8 = (int32_t) *i8++; in xnn_s8_maxpool_minmax_ukernel_9p8x__scalar_c1() local
154 const int32_t vi8 = (int32_t) *o; in xnn_s8_maxpool_minmax_ukernel_9p8x__scalar_c1() local
|
| /external/pytorch/aten/src/ATen/native/quantized/cpu/qnnpack/src/q8avgpool/ |
| D | up8x9-neon.c | 101 const uint8x8_t vi8 = vld1_u8(i8); in pytorch_q8avgpool_ukernel_up8x9__neon() local
179 const uint8x8_t vi8 = vreinterpret_u8_u64( in pytorch_q8avgpool_ukernel_up8x9__neon() local
|
| D | up8x9-sse2.c | 89 const __m128i vi8 = _mm_loadl_epi64((const __m128i*)i8); in pytorch_q8avgpool_ukernel_up8x9__sse2() local
171 const __m128i vi8 = in pytorch_q8avgpool_ukernel_up8x9__sse2() local
|
| /external/pytorch/aten/src/ATen/native/quantized/cpu/qnnpack/src/q8dwconv/ |
| D | mp8x27-neon.c | 189 const uint8x8_t vi8 = vld1_u8(i8); in pytorch_q8dwconv_ukernel_mp8x27__neon() local
334 const uint8x8_t vi8 = vreinterpret_u8_u64( in pytorch_q8dwconv_ukernel_mp8x27__neon() local
479 const uint8x8_t vi8 = vld1_u8(i8); in pytorch_q8dwconv_ukernel_mp8x27__neon() local
623 const uint8x8_t vi8 = vreinterpret_u8_u64( in pytorch_q8dwconv_ukernel_mp8x27__neon() local
776 const uint8x8_t vi8 = vld1_u8(i8); in pytorch_q8dwconv_ukernel_mp8x27__neon() local
959 const uint8x8_t vi8 = vreinterpret_u8_u64( in pytorch_q8dwconv_ukernel_mp8x27__neon() local
|