| /aosp_15_r20/external/pytorch/aten/src/ATen/native/quantized/cpu/qnnpack/src/q8avgpool/ |
| H A D | mp8x9p8q-neon.c | 192 const uint16x8_t vsum0123 = vaddq_u16(vsum01, vsum23); in pytorch_q8avgpool_ukernel_mp8x9p8q__neon() local
242 const uint16x8_t vsum0123 = vaddq_u16(vsum01, vsum23); in pytorch_q8avgpool_ukernel_mp8x9p8q__neon() local
316 const int16x8_t vsum0123 = vaddq_s16(vsum01, vsum23); in pytorch_q8avgpool_ukernel_mp8x9p8q__neon() local
392 const int16x8_t vsum0123 = vaddq_s16(vsum01, vsum23); in pytorch_q8avgpool_ukernel_mp8x9p8q__neon() local
|
| H A D | mp8x9p8q-sse2.c | 208 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in pytorch_q8avgpool_ukernel_mp8x9p8q__sse2() local
266 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in pytorch_q8avgpool_ukernel_mp8x9p8q__sse2() local
347 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in pytorch_q8avgpool_ukernel_mp8x9p8q__sse2() local
425 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in pytorch_q8avgpool_ukernel_mp8x9p8q__sse2() local
|
| /aosp_15_r20/external/XNNPACK/src/f32-gavgpool/ |
| H A D | 7p7x-minmax-neon-c4.c | 84 const float32x4_t vsum0123 = vaddq_f32(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__neon_c4() local
138 const float32x4_t vsum0123 = vaddq_f32(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__neon_c4() local
166 const float32x4_t vsum0123 = vaddq_f32(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__neon_c4() local
|
| H A D | 7p7x-minmax-wasmsimd-x86-c4.c | 99 const v128_t vsum0123 = wasm_f32x4_add(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__wasmsimd_x86_c4() local
161 const v128_t vsum0123 = wasm_f32x4_add(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__wasmsimd_x86_c4() local
190 const v128_t vsum0123 = wasm_f32x4_add(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__wasmsimd_x86_c4() local
|
| H A D | 7p7x-minmax-wasmsimd-arm-c4.c | 99 const v128_t vsum0123 = wasm_f32x4_add(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__wasmsimd_arm_c4() local
161 const v128_t vsum0123 = wasm_f32x4_add(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__wasmsimd_arm_c4() local
190 const v128_t vsum0123 = wasm_f32x4_add(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__wasmsimd_arm_c4() local
|
| H A D | 7p7x-minmax-sse-c4.c | 98 const __m128 vsum0123 = _mm_add_ps(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__sse_c4() local
160 const __m128 vsum0123 = _mm_add_ps(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__sse_c4() local
189 const __m128 vsum0123 = _mm_add_ps(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__sse_c4() local
|
| H A D | 7p7x-minmax-wasm-c1.c | 83 const float vsum0123 = vsum01 + vsum23; in xnn_f32_gavgpool_minmax_ukernel_7p7x__wasm_c1() local
137 const float vsum0123 = vsum01 + vsum23; in xnn_f32_gavgpool_minmax_ukernel_7p7x__wasm_c1() local
|
| H A D | 7p7x-minmax-scalar-c1.c | 83 const float vsum0123 = vsum01 + vsum23; in xnn_f32_gavgpool_minmax_ukernel_7p7x__scalar_c1() local
137 const float vsum0123 = vsum01 + vsum23; in xnn_f32_gavgpool_minmax_ukernel_7p7x__scalar_c1() local
|
| /aosp_15_r20/external/XNNPACK/src/qu8-avgpool/ |
| H A D | 9p8x-minmax-neon-c8.c | 183 const uint16x8_t vsum0123 = vaddq_u16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8() local
278 const int16x8_t vsum0123 = vaddq_s16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8() local
352 const int16x8_t vsum0123 = vaddq_s16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__neon_c8() local
|
| H A D | 9p8x-minmax-sse2-c8.c | 198 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__sse2_c8() local
304 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__sse2_c8() local
379 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__sse2_c8() local
|
| H A D | 9p8x-minmax-scalar-c1.c | 176 const uint32_t vsum0123 = vsum01 + vsum23; in xnn_qu8_avgpool_minmax_ukernel_9p8x__scalar_c1() local
268 const uint32_t vsum0123 = vsum01 + vsum23; in xnn_qu8_avgpool_minmax_ukernel_9p8x__scalar_c1() local
|
| /aosp_15_r20/external/pytorch/aten/src/ATen/native/quantized/cpu/qnnpack/src/q8gemm/ |
| H A D | 4x-sumrows-neon.c | 140 int32x4_t vsum0123 = vmulq_n_s32(vreinterpretq_s32_u32(vacc0123), multiplier); in pytorch_q8sumrows_ukernel_4x__neon() local
|
| /aosp_15_r20/external/XNNPACK/src/f16-gavgpool-cw/ |
| H A D | neonfp16arith-x8.c | 78 const float16x8_t vsum0123 = vpaddq_f16(vsum01, vsum23); in xnn_f16_gavgpool_cw_ukernel__neonfp16arith_x8() local
|
| /aosp_15_r20/external/XNNPACK/src/amalgam/ |
| H A D | sse.c | 4360 const __m128 vsum0123 = _mm_add_ps(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__sse_c4() local
4422 const __m128 vsum0123 = _mm_add_ps(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__sse_c4() local
4451 const __m128 vsum0123 = _mm_add_ps(vsum01, vsum23); in xnn_f32_gavgpool_minmax_ukernel_7p7x__sse_c4() local
|
| H A D | sse2.c | 7974 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__sse2_c8() local
8080 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__sse2_c8() local
8155 const __m128i vsum0123 = _mm_add_epi16(vsum01, vsum23); in xnn_qu8_avgpool_minmax_ukernel_9p8x__sse2_c8() local
|