/external/XNNPACK/src/f32-raddexpminusmax/gen/ |
D | avx2-p5-x64.c | 84 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64() local 164 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64() 173 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64()
|
D | avx2-p5-x64-acc2.c | 85 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2() local 165 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2() 174 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc2()
|
D | avx2-p5-x64-acc4.c | 87 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc4() local 167 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc4() 176 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x64_acc4()
|
D | avx2-p5-x72.c | 87 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72() local 175 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72() 185 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72()
|
D | avx2-p5-x72-acc3.c | 89 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3() local 177 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3() 187 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x72_acc3()
|
D | avx2-p5-x80.c | 90 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80() local 186 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80() 197 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80()
|
D | avx2-p5-x80-acc2.c | 91 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2() local 187 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2() 198 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc2()
|
D | avx2-p5-x80-acc5.c | 94 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5() local 190 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5() 201 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x80_acc5()
|
D | avx2-p5-x96-acc3.c | 98 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3() local 210 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3() 223 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddexpminusmax_ukernel__avx2_p5_x96_acc3()
|
/external/XNNPACK/src/f32-vscaleexpminusmax/gen/ |
D | avx2-p5-x64.c | 86 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64() local 166 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64() 175 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x64()
|
D | avx2-p5-x72.c | 89 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72() local 177 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72() 187 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x72()
|
D | avx2-p5-x80.c | 92 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80() local 188 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80() 199 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x80()
|
D | avx2-p5-x88.c | 95 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88() local 199 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88() 211 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_vscaleexpminusmax_ukernel__avx2_p5_x88()
|
/external/XNNPACK/src/f32-raddstoreexpminusmax/gen/ |
D | avx2-p5-x64-acc2.c | 86 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2() local 166 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2() 175 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc2()
|
D | avx2-p5-x64.c | 85 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64() local 165 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64() 174 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64()
|
D | avx2-p5-x64-acc4.c | 88 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc4() local 168 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc4() 177 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x64_acc4()
|
D | avx2-p5-x72.c | 88 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x72() local 176 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x72() 186 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x72()
|
D | avx2-p5-x72-acc3.c | 90 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x72_acc3() local 178 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x72_acc3() 188 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x72_acc3()
|
D | avx2-p5-x80-acc5.c | 95 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5() local 191 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5() 202 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc5()
|
D | avx2-p5-x80.c | 91 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80() local 187 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80() 198 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80()
|
D | avx2-p5-x80-acc2.c | 92 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2() local 188 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2() 199 __m256 vf7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_raddstoreexpminusmax_ukernel__avx2_p5_x80_acc2()
|
/external/XNNPACK/src/f32-sigmoid/gen/ |
D | avx2-rr1-p5-div-x64.c | 94 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x64() local 164 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x64() 173 const __m256 ve7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x64()
|
D | avx2-rr1-p5-div-x80.c | 100 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80() local 184 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80() 195 const __m256 ve7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
|
D | avx2-rr1-p5-div-x72.c | 97 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x72() local 174 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x72() 184 const __m256 ve7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x72()
|
D | avx2-rr1-p5-nr1fma-x64.c | 94 const __m256 vs7 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(vn7), 23)); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x64() local 164 vt7 = _mm256_mul_ps(vt7, vs7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x64() 173 const __m256 ve7 = _mm256_fmadd_ps(vt7, vp7, vs7); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x64()
|