| /aosp_15_r20/external/XNNPACK/src/f32-vscaleextexp/gen/ |
| H A D | avx512f-p5-scalef-x16.c | 75 const __m512 ve0 = _mm512_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x16() local
|
| H A D | avx512f-p5-scalef-x32.c | 85 const __m512 ve0 = _mm512_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x32() local
|
| H A D | avx2-p5-x8.c | 81 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x8() local
|
| H A D | avx2-p5-x16.c | 91 __m256 ve0 = _mm256_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x16() local
|
| H A D | avx512f-p5-scalef-x48.c | 95 const __m512 ve0 = _mm512_add_ps(vn0, vscalee); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x48() local
|
| /aosp_15_r20/external/XNNPACK/src/f32-velu/gen/ |
| H A D | velu-scalar-rr2-p6-x2.c | 95 const float ve0 = (vp0 + vs0) * valpha; in xnn_f32_velu_ukernel__scalar_rr2_p6_x2() local
|
| H A D | velu-wasm-rr2-p6-x2.c | 87 const float ve0 = (vp0 + vs0) * valpha; in xnn_f32_velu_ukernel__wasm_rr2_p6_x2() local
|
| H A D | velu-wasm-rr2-lut16-p3-x2.c | 81 const float ve0 = (vp0 + vs0) * valpha; in xnn_f32_velu_ukernel__wasm_rr2_lut16_p3_x2() local
|
| H A D | velu-scalar-rr2-lut16-p3-x2.c | 89 const float ve0 = (vp0 + vs0) * valpha; in xnn_f32_velu_ukernel__scalar_rr2_lut16_p3_x2() local
|
| H A D | velu-scalar-rr2-p6-x3.c | 114 const float ve0 = (vp0 + vs0) * valpha; in xnn_f32_velu_ukernel__scalar_rr2_p6_x3() local
|
| H A D | velu-avx2-rr1-p6-x16.c | 80 const __m256 ve0 = _mm256_fmadd_ps(vp0, valpha, vs0); in xnn_f32_velu_ukernel__avx2_rr1_p6_x16() local
|
| H A D | velu-wasm-rr2-p6-x3.c | 102 const float ve0 = (vp0 + vs0) * valpha; in xnn_f32_velu_ukernel__wasm_rr2_p6_x3() local
|
| /aosp_15_r20/external/XNNPACK/src/f32-vsigmoid/gen/ |
| H A D | vsigmoid-scalar-rr2-p5-div-x2.c | 76 const float ve0 = vt0 * vp0 + vs0; in xnn_f32_vsigmoid_ukernel__scalar_rr2_p5_div_x2() local
|
| H A D | vsigmoid-avx512f-rr1-p5-scalef-div-x32.c | 69 const __m512 ve0 = _mm512_scalef_ps(vp0, vn0); in xnn_f32_vsigmoid_ukernel__avx512f_rr1_p5_scalef_div_x32() local
|
| H A D | vsigmoid-scalar-rr2-lut64-p2-div-x2.c | 49 const uint32_t ve0 = float_as_uint32(vn0) << 17; in xnn_f32_vsigmoid_ukernel__scalar_rr2_lut64_p2_div_x2() local
|
| H A D | vsigmoid-scalar-rr2-lut2048-p1-div-x2.c | 49 const uint32_t ve0 = float_as_uint32(vn0) << 12; in xnn_f32_vsigmoid_ukernel__scalar_rr2_lut2048_p1_div_x2() local
|
| H A D | vsigmoid-avx512f-rr1-lut16-p3-perm-scalef-div-x32.c | 68 const __m512 ve0 = _mm512_scalef_ps(vp0, vn0); in xnn_f32_vsigmoid_ukernel__avx512f_rr1_lut16_p3_perm_scalef_div_x32() local
|
| H A D | vsigmoid-avx512f-rr1-p5-scalef-nr1fma-x32.c | 69 const __m512 ve0 = _mm512_scalef_ps(vp0, vn0); in xnn_f32_vsigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x32() local
|
| H A D | vsigmoid-avx512f-rr2-lut32-p2-perm2-scalef-div-x32.c | 70 const __m512 ve0 = _mm512_scalef_ps(vp0, vn0); in xnn_f32_vsigmoid_ukernel__avx512f_rr2_lut32_p2_perm2_scalef_div_x32() local
|
| H A D | vsigmoid-avx512f-rr1-p5-scalef-div-x48.c | 79 const __m512 ve0 = _mm512_scalef_ps(vp0, vn0); in xnn_f32_vsigmoid_ukernel__avx512f_rr1_p5_scalef_div_x48() local
|
| /aosp_15_r20/external/XNNPACK/src/f32-raddstoreexpminusmax/gen/ |
| H A D | scalar-rr2-lut64-p2-x2.c | 70 const uint32_t ve0 = (float_as_uint32(vn0) & UINT32_C(0xFFFFFFC0)) << 17; in xnn_f32_raddstoreexpminusmax_ukernel__scalar_rr2_lut64_p2_x2() local
|
| H A D | scalar-rr2-lut64-p2-x2-acc2.c | 71 const uint32_t ve0 = (float_as_uint32(vn0) & UINT32_C(0xFFFFFFC0)) << 17; in xnn_f32_raddstoreexpminusmax_ukernel__scalar_rr2_lut64_p2_x2_acc2() local
|
| /aosp_15_r20/external/XNNPACK/src/f16-velu/gen/ |
| H A D | velu-avx2-rr1-p3-x16.c | 72 const __m256 ve0 = _mm256_fmadd_ps(vp0, vt0, vs0); in xnn_f16_velu_ukernel__avx2_rr1_p3_x16() local
|
| H A D | velu-neonfp16arith-rr1-p3-x16.c | 73 float16x8_t ve0 = vfmsq_f16(vs0, vp0, vminus_alpha); in xnn_f16_velu_ukernel__neonfp16arith_rr1_p3_x16() local
|
| /aosp_15_r20/external/XNNPACK/src/f16-vsigmoid/gen/ |
| H A D | vsigmoid-avx2-rr1-p2-div-x16.c | 64 const __m256 ve0 = _mm256_fmadd_ps(vt0, vp0, vs0); in xnn_f16_vsigmoid_ukernel__avx2_rr1_p2_div_x16() local
|