| /external/XNNPACK/src/f32-vscaleextexp/gen/ |
| D | avx2-p5-x80.c | 180 __m256 ve9 = _mm256_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80() local
195 ve9 = _mm256_max_ps(ve9, vmin_exponent); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
209 …const __m256 vs9 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve9, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x80()
|
| D | avx2-p5-x88.c | 190 __m256 ve9 = _mm256_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88() local
206 ve9 = _mm256_max_ps(ve9, vmin_exponent); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
221 …const __m256 vs9 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve9, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x88()
|
| D | avx2-p5-x96.c | 200 __m256 ve9 = _mm256_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96() local
217 ve9 = _mm256_max_ps(ve9, vmin_exponent); in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
233 …const __m256 vs9 = _mm256_castsi256_ps(_mm256_slli_epi32(_mm256_castps_si256(_mm256_add_ps(ve9, vm… in xnn_f32_vscaleextexp_ukernel__avx2_p5_x96()
|
| D | avx512f-p5-scalef-x160.c | 174 const __m512 ve9 = _mm512_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x160() local
186 vf9 = _mm512_scalef_ps(vf9, ve9); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x160()
|
| D | avx512f-p5-scalef-x176.c | 184 const __m512 ve9 = _mm512_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x176() local
197 vf9 = _mm512_scalef_ps(vf9, ve9); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x176()
|
| D | avx512f-p5-scalef-x192.c | 194 const __m512 ve9 = _mm512_add_ps(vn9, vscalee); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x192() local
208 vf9 = _mm512_scalef_ps(vf9, ve9); in xnn_f32_vscaleextexp_ukernel__avx512f_p5_scalef_x192()
|
| /external/XNNPACK/src/f32-sigmoid/gen/ |
| D | avx2-rr1-p5-div-x80.c | 172 const __m256 ve9 = _mm256_fmadd_ps(vt9, vp9, vs9); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80() local
183 const __m256 vd9 = _mm256_add_ps(ve9, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
194 __m256 vf9 = _mm256_div_ps(ve9, vd9); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
|
| D | avx2-rr1-p5-nr1fma-x80.c | 172 const __m256 ve9 = _mm256_fmadd_ps(vt9, vp9, vs9); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x80() local
183 const __m256 vd9 = _mm256_add_ps(ve9, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x80()
217 __m256 vf9 = _mm256_mul_ps(ve9, vr9); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x80()
|
| D | avx2-rr1-p5-nr2fma-x80.c | 172 const __m256 ve9 = _mm256_fmadd_ps(vt9, vp9, vs9); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x80() local
183 const __m256 vd9 = _mm256_add_ps(ve9, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x80()
227 __m256 vf9 = _mm256_mul_ps(ve9, vr9); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x80()
|
| D | avx-rr2-p5-div-x80.c | 204 const __m256 ve9 = _mm256_add_ps(_mm256_mul_ps(vt9, vp9), vs9); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_div_x80() local
215 const __m256 vd9 = _mm256_add_ps(ve9, vone); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_div_x80()
226 __m256 vf9 = _mm256_div_ps(ve9, vd9); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_div_x80()
|
| D | avx-rr2-p5-nr2-x80.c | 205 const __m256 ve9 = _mm256_add_ps(_mm256_mul_ps(vt9, vp9), vs9); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x80() local
216 const __m256 vd9 = _mm256_add_ps(ve9, vone); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x80()
259 __m256 vf9 = _mm256_mul_ps(ve9, vr9); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x80()
|
| /external/XNNPACK/src/f32-velu/gen/ |
| D | velu-avx2-rr1-lut4-p4-perm-x80.c | 212 const __m256 ve9 = _mm256_fmadd_ps(vp9, valpha, vs9); in xnn_f32_velu_ukernel__avx2_rr1_lut4_p4_perm_x80() local
224 const __m256 vy9 = _mm256_blendv_ps(vx9, ve9, vx9); in xnn_f32_velu_ukernel__avx2_rr1_lut4_p4_perm_x80()
|
| D | velu-avx2-rr1-lut16-p3-gather-x80.c | 211 const __m256 ve9 = _mm256_fmadd_ps(vp9, valpha, vs9); in xnn_f32_velu_ukernel__avx2_rr1_lut16_p3_gather_x80() local
223 const __m256 vy9 = _mm256_blendv_ps(vx9, ve9, vx9); in xnn_f32_velu_ukernel__avx2_rr1_lut16_p3_gather_x80()
|
| D | velu-avx2-rr1-p6-x80.c | 213 const __m256 ve9 = _mm256_fmadd_ps(vp9, valpha, vs9); in xnn_f32_velu_ukernel__avx2_rr1_p6_x80() local
225 const __m256 vy9 = _mm256_blendv_ps(vx9, ve9, vx9); in xnn_f32_velu_ukernel__avx2_rr1_p6_x80()
|
| D | velu-avx2-rr1-lut8-p4-perm-x80.c | 211 const __m256 ve9 = _mm256_fmadd_ps(vp9, valpha, vs9); in xnn_f32_velu_ukernel__avx2_rr1_lut8_p4_perm_x80() local
223 const __m256 vy9 = _mm256_blendv_ps(vx9, ve9, vx9); in xnn_f32_velu_ukernel__avx2_rr1_lut8_p4_perm_x80()
|