Home
last modified time | relevance | path

Searched refs:_mm256_andnot_ps (Results 1 – 25 of 86) sorted by relevance

1234

/external/XNNPACK/src/f32-vscaleexpminusmax/gen/
Davx2-p5-x88.c218 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
219 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
220 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
221 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
222 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
223 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
224 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
225 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
226 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
227 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
[all …]
Davx2-p5-x96.c231 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
232 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
233 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
234 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
235 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
236 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
237 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
238 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
239 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
240 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x96()
[all …]
Davx2-p5-x80.c205 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
206 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
207 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
208 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
209 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
210 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
211 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
212 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
213 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
214 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
[all …]
Davx2-p5-x72.c192 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
193 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
194 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
195 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
196 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
197 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
198 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
199 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
200 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
263 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
[all …]
Davx2-p5-x64.c179 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
180 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
181 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
182 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
183 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
184 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
185 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
186 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
247 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
297 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
/external/XNNPACK/src/f32-raddexpminusmax/gen/
Davx2-p5-x96-acc3.c231 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
232 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
233 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
234 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
235 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
236 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
237 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
238 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
239 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
240 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
[all …]
Davx2-p5-x96-acc6.c234 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
235 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
236 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
237 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
238 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
239 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
240 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
241 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
242 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
243 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc6()
[all …]
Davx2-p5-x96.c229 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
230 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
231 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
232 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
233 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
234 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
235 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
236 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
237 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
238 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96()
[all …]
Davx2-p5-x96-acc2.c230 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
231 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
232 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
233 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
234 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
235 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
236 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
237 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
238 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
239 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc2()
[all …]
Davx2-p5-x80.c203 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
204 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
205 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
206 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
207 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
208 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
209 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
210 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
211 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
212 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
[all …]
Davx2-p5-x80-acc2.c204 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
205 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
206 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
207 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
208 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
209 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
210 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
211 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
212 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
213 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
[all …]
Davx2-p5-x72.c190 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
191 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
192 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
193 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
194 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
195 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
196 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
197 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
198 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
251 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
[all …]
Davx2-p5-x72-acc3.c192 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
193 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
194 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
195 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
196 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
197 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
198 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
199 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
200 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
256 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
[all …]
Davx2-p5-x80-acc5.c207 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
208 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
209 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
210 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
211 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
212 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
213 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
214 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
215 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
216 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
[all …]
Davx2-p5-x64.c177 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
178 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
179 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
180 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
181 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
182 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
183 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
184 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
236 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
282 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
Davx2-p5-x64-acc2.c178 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
179 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
180 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
181 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
182 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
183 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
184 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
185 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
239 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
285 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
/external/XNNPACK/src/f32-raddstoreexpminusmax/gen/
Davx2-p5-x96-acc2.c231 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
232 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
233 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
234 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
235 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
236 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
237 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
238 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
239 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
240 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc2()
[all …]
Davx2-p5-x96-acc3.c232 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
233 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
234 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
235 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
236 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
237 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
238 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
239 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
240 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
241 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc3()
[all …]
Davx2-p5-x96-acc6.c235 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
236 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
237 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
238 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
239 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
240 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
241 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
242 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
243 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
244 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96_acc6()
[all …]
Davx2-p5-x96.c230 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
231 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
232 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
233 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
234 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
235 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
236 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
237 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
238 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
239 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x96()
[all …]
Davx2-p5-x80-acc5.c208 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
209 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
210 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
211 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
212 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
213 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
214 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
215 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
216 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
217 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
[all …]
Davx2-p5-x80.c204 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
205 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
206 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
207 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
208 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
209 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
210 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
211 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
212 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
213 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
[all …]
Davx2-p5-x80-acc2.c205 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
206 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
207 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
208 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
209 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
210 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
211 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
212 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
213 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vx8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
214 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vx9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
[all …]
Davx2-p5-x64-acc2.c179 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vx0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
180 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vx1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
181 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vx2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
182 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vx3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
183 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vx4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
184 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vx5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
185 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vx6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
186 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vx7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
251 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
301 vf = _mm256_andnot_ps(_mm256_cmp_ps(vx, vdenorm_cutoff, _CMP_LT_OS), vf); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
/external/XNNPACK/src/f32-sigmoid/gen/
Davx2-rr1-p5-div-x80.c225 vf0 = _mm256_andnot_ps(_mm256_cmp_ps(vz0, vdenorm_cutoff, _CMP_LT_OS), vf0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
226 vf1 = _mm256_andnot_ps(_mm256_cmp_ps(vz1, vdenorm_cutoff, _CMP_LT_OS), vf1); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
227 vf2 = _mm256_andnot_ps(_mm256_cmp_ps(vz2, vdenorm_cutoff, _CMP_LT_OS), vf2); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
228 vf3 = _mm256_andnot_ps(_mm256_cmp_ps(vz3, vdenorm_cutoff, _CMP_LT_OS), vf3); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
229 vf4 = _mm256_andnot_ps(_mm256_cmp_ps(vz4, vdenorm_cutoff, _CMP_LT_OS), vf4); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
230 vf5 = _mm256_andnot_ps(_mm256_cmp_ps(vz5, vdenorm_cutoff, _CMP_LT_OS), vf5); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
231 vf6 = _mm256_andnot_ps(_mm256_cmp_ps(vz6, vdenorm_cutoff, _CMP_LT_OS), vf6); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
232 vf7 = _mm256_andnot_ps(_mm256_cmp_ps(vz7, vdenorm_cutoff, _CMP_LT_OS), vf7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
233 vf8 = _mm256_andnot_ps(_mm256_cmp_ps(vz8, vdenorm_cutoff, _CMP_LT_OS), vf8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
234 vf9 = _mm256_andnot_ps(_mm256_cmp_ps(vz9, vdenorm_cutoff, _CMP_LT_OS), vf9); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
[all …]

1234