/external/XNNPACK/src/f32-raddextexp/gen/ |
D | avx512f-p5-scalef-x128-acc2.c | 155 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e0); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x128_acc2() local 167 vaccv0 = _mm512_add_ps(vaccv0, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x128_acc2()
|
D | avx512f-p5-scalef-x128.c | 152 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e0); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x128() local 163 vaccv0 = _mm512_add_ps(vaccv0, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x128()
|
D | avx512f-p5-scalef-x144.c | 162 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e0); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x144() local 174 vaccv0 = _mm512_add_ps(vaccv0, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x144()
|
D | avx512f-p5-scalef-x144-acc3.c | 168 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e2); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x144_acc3() local 182 vaccv2 = _mm512_add_ps(vaccv2, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x144_acc3()
|
D | avx512f-p5-scalef-x160.c | 172 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e0); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x160() local 185 vaccv0 = _mm512_add_ps(vaccv0, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x160()
|
D | avx512f-p5-scalef-x160-acc2.c | 175 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e0); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x160_acc2() local 189 vaccv0 = _mm512_add_ps(vaccv0, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x160_acc2()
|
D | avx512f-p5-scalef-x128-acc4.c | 161 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e2); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x128_acc4() local 175 vaccv2 = _mm512_add_ps(vaccv2, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x128_acc4()
|
D | avx2-p5-x64.c | 160 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e0), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x64() local 175 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x64()
|
D | avx2-p5-x72.c | 170 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e0), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x72() local 186 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x72()
|
D | avx512f-p5-scalef-x192.c | 192 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e0); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x192() local 207 vaccv0 = _mm512_add_ps(vaccv0, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x192()
|
D | avx512f-p5-scalef-x192-acc2.c | 195 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e0); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x192_acc2() local 211 vaccv0 = _mm512_add_ps(vaccv0, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x192_acc2()
|
D | avx512f-p5-scalef-x160-acc5.c | 184 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e2); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x160_acc5() local 201 vaccv2 = _mm512_add_ps(vaccv2, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x160_acc5()
|
D | avx2-p5-x64-acc2.c | 163 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e0), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc2() local 179 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc2()
|
D | avx2-p5-x64-acc4.c | 169 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e2), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4() local 187 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x64_acc4()
|
D | avx2-p5-x80-acc2.c | 183 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e0), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc2() local 201 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc2()
|
D | avx512f-p5-scalef-x192-acc3.c | 198 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e2); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x192_acc3() local 215 vaccv2 = _mm512_add_ps(vaccv2, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x192_acc3()
|
D | avx2-p5-x72-acc3.c | 176 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e2), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3() local 194 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x72_acc3()
|
D | avx2-p5-x80.c | 180 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e0), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x80() local 197 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x80()
|
D | avx512f-p5-scalef-x192-acc6.c | 207 const __m512 vdelta_e2 = _mm512_sub_ps(vn2, vmax_e2); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x192_acc6() local 227 vaccv2 = _mm512_add_ps(vaccv2, _mm512_scalef_ps(vp2, vdelta_e2)); in xnn_f32_raddextexp_ukernel__avx512f_p5_scalef_x192_acc6()
|
D | avx2-p5-x96.c | 200 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e0), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x96() local 219 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x96()
|
D | avx2-p5-x80-acc5.c | 192 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e2), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5() local 213 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x80_acc5()
|
D | avx2-p5-x96-acc3.c | 206 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e2), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc3() local 227 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc3()
|
D | avx2-p5-x96-acc2.c | 203 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e0), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc2() local 223 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc2()
|
D | avx2-p5-x96-acc6.c | 215 const __m256 vdelta_e2 = _mm256_max_ps(_mm256_sub_ps(vn2, vmax_e2), vmin_exponent); in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6() local 239 …_mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(vdelta_e2, vmagic_bias)), … in xnn_f32_raddextexp_ukernel__avx2_p5_x96_acc6()
|