| /aosp_15_r20/external/XNNPACK/src/f32-velu/gen/ |
| H A D | velu-avx2-rr1-lut8-p4-perm-x24.c | 66 __m256 vs2 = _mm256_castsi256_ps(_mm256_add_epi32(vl2, ven2)); in xnn_f32_velu_ukernel__avx2_rr1_lut8_p4_perm_x24() local
|
| H A D | velu-avx-rr2-p6-x24.c | 70 __m256 vs2 = _mm256_insertf128_ps(_mm256_castps128_ps256(vs2_lo), vs2_hi, 1); in xnn_f32_velu_ukernel__avx_rr2_p6_x24() local
|
| H A D | velu-avx2-rr1-lut4-p4-perm-x24.c | 66 __m256 vs2 = _mm256_castsi256_ps(_mm256_add_epi32(vl2, ven2)); in xnn_f32_velu_ukernel__avx2_rr1_lut4_p4_perm_x24() local
|
| H A D | velu-avx512f-rr1-p6-x80.c | 65 __m512 vs2 = _mm512_castsi512_ps(_mm512_slli_epi32(_mm512_castps_si512(vn2), 23)); in xnn_f32_velu_ukernel__avx512f_rr1_p6_x80() local
|
| H A D | velu-avx2-rr1-lut16-p3-gather-x24.c | 71 __m256 vs2 = _mm256_castsi256_ps(_mm256_add_epi32(vl2, ven2)); in xnn_f32_velu_ukernel__avx2_rr1_lut16_p3_gather_x24() local
|
| H A D | velu-avx2-rr1-p6-x40.c | 63 __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn2), 23)); in xnn_f32_velu_ukernel__avx2_rr1_p6_x40() local
|
| H A D | velu-wasm-rr2-p6-x6.c | 68 float vs2 = uint32_as_float(float_as_uint32(vn2) << 23); in xnn_f32_velu_ukernel__wasm_rr2_p6_x6() local
|
| H A D | velu-scalar-rr2-p6-x6.c | 68 float vs2 = uint32_as_float(float_as_uint32(vn2) << 23); in xnn_f32_velu_ukernel__scalar_rr2_p6_x6() local
|
| H A D | velu-avx512f-rr1-lut16-p3-perm-x64.c | 69 __m512 vs2 = _mm512_castsi512_ps(_mm512_add_epi32(vl2, ven2)); in xnn_f32_velu_ukernel__avx512f_rr1_lut16_p3_perm_x64() local
|
| H A D | velu-avx2-rr1-lut8-p4-perm-x32.c | 72 __m256 vs2 = _mm256_castsi256_ps(_mm256_add_epi32(vl2, ven2)); in xnn_f32_velu_ukernel__avx2_rr1_lut8_p4_perm_x32() local
|
| /aosp_15_r20/external/XNNPACK/src/f16-raddstoreexpminusmax/gen/ |
| H A D | neonfp16arith-rr2-p2-x32.c | 58 const float16x8_t vs2 = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn2), 10)); in xnn_f16_raddstoreexpminusmax_ukernel__neonfp16arith_rr2_p2_x32() local
|
| H A D | neonfp16arith-rr2-p2-x32-acc2.c | 59 const float16x8_t vs2 = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn2), 10)); in xnn_f16_raddstoreexpminusmax_ukernel__neonfp16arith_rr2_p2_x32_acc2() local
|
| H A D | avx2-rr1-p2-x40.c | 61 const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn2), 23)); in xnn_f16_raddstoreexpminusmax_ukernel__avx2_rr1_p2_x40() local
|
| H A D | avx2-rr1-p2-x40-acc2.c | 62 const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn2), 23)); in xnn_f16_raddstoreexpminusmax_ukernel__avx2_rr1_p2_x40_acc2() local
|
| H A D | avx2-rr1-p2-x40-acc5.c | 65 const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn2), 23)); in xnn_f16_raddstoreexpminusmax_ukernel__avx2_rr1_p2_x40_acc5() local
|
| H A D | neonfp16arith-rr2-p2-x32-acc4.c | 61 const float16x8_t vs2 = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn2), 10)); in xnn_f16_raddstoreexpminusmax_ukernel__neonfp16arith_rr2_p2_x32_acc4() local
|
| /aosp_15_r20/external/XNNPACK/src/f32-vsigmoid/gen/ |
| H A D | vsigmoid-scalar-rr2-lut64-p2-div-x4.c | 65 const float vs2 = uint32_as_float(xnn_table_exp2minus_k_over_64[vidx2] + ve2); in xnn_f32_vsigmoid_ukernel__scalar_rr2_lut64_p2_div_x4() local
|
| H A D | vsigmoid-scalar-rr2-lut2048-p1-div-x4.c | 65 const float vs2 = uint32_as_float(xnn_table_exp2minus_k_over_2048[vidx2] + ve2); in xnn_f32_vsigmoid_ukernel__scalar_rr2_lut2048_p1_div_x4() local
|
| H A D | vsigmoid-avx2-rr1-p5-nr1fma-x24.c | 54 const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn2), 23)); in xnn_f32_vsigmoid_ukernel__avx2_rr1_p5_nr1fma_x24() local
|
| H A D | vsigmoid-avx2-rr1-p5-nr2fma-x24.c | 54 const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn2), 23)); in xnn_f32_vsigmoid_ukernel__avx2_rr1_p5_nr2fma_x24() local
|
| H A D | vsigmoid-avx-rr2-p5-div-x24.c | 61 const __m256 vs2 = _mm256_insertf128_ps(_mm256_castps128_ps256(vs_lo2), vs_hi2, 1); in xnn_f32_vsigmoid_ukernel__avx_rr2_p5_div_x24() local
|
| H A D | vsigmoid-avx2-rr1-p5-div-x32.c | 57 const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn2), 23)); in xnn_f32_vsigmoid_ukernel__avx2_rr1_p5_div_x32() local
|
| /aosp_15_r20/external/XNNPACK/src/f16-vsigmoid/gen/ |
| H A D | vsigmoid-avx2-rr1-p2-div-x32.c | 57 const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn2), 23)); in xnn_f16_vsigmoid_ukernel__avx2_rr1_p2_div_x32() local
|
| H A D | vsigmoid-neonfp16arith-rr2-p2-nr1recps-x24.c | 52 const float16x8_t vs2 = vreinterpretq_f16_s16(vshlq_n_s16(vreinterpretq_s16_f16(vn2), 10)); in xnn_f16_vsigmoid_ukernel__neonfp16arith_rr2_p2_nr1recps_x24() local
|
| /aosp_15_r20/external/XNNPACK/src/f32-vscaleextexp/gen/ |
| H A D | avx2-p5-x48.c | 154 …const __m256 vs2 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve2, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x48() local
|