| /external/XNNPACK/src/f32-sigmoid/gen/ |
| D | avx2-rr1-p5-nr2fma-x72.c | 170 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x72() local
180 __m256 vr8 = _mm256_rcp_ps(vd8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x72()
190 vr8 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr8, vd8, vone), vr8, vr8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x72()
200 vr8 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr8, vd8, vone), vr8, vr8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x72()
|
| D | avx2-rr1-p5-nr2fma-x80.c | 182 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x80() local
193 __m256 vr8 = _mm256_rcp_ps(vd8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x80()
204 vr8 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr8, vd8, vone), vr8, vr8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x80()
215 vr8 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr8, vd8, vone), vr8, vr8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x80()
|
| D | avx-rr2-p5-nr2-x72.c | 200 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x72() local
210 __m256 vr8 = _mm256_rcp_ps(vd8); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x72()
228 vr8 = _mm256_mul_ps(vr8, _mm256_sub_ps(vtwo, _mm256_mul_ps(vr8, vd8))); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x72()
229 vr8 = _mm256_mul_ps(vr8, _mm256_sub_ps(vtwo, _mm256_mul_ps(vr8, vd8))); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x72()
|
| D | avx2-rr1-p5-nr1fma-x72.c | 170 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x72() local
180 __m256 vr8 = _mm256_rcp_ps(vd8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x72()
190 vr8 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr8, vd8, vone), vr8, vr8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x72()
|
| D | avx2-rr1-p5-nr1fma-x80.c | 182 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x80() local
193 __m256 vr8 = _mm256_rcp_ps(vd8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x80()
204 vr8 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr8, vd8, vone), vr8, vr8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x80()
|
| D | avx-rr2-p5-nr2-x80.c | 215 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x80() local
226 __m256 vr8 = _mm256_rcp_ps(vd8); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x80()
245 vr8 = _mm256_mul_ps(vr8, _mm256_sub_ps(vtwo, _mm256_mul_ps(vr8, vd8))); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x80()
246 vr8 = _mm256_mul_ps(vr8, _mm256_sub_ps(vtwo, _mm256_mul_ps(vr8, vd8))); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x80()
|
| D | avx2-rr1-p5-div-x72.c | 170 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x72() local
180 __m256 vf8 = _mm256_div_ps(ve8, vd8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x72()
|
| D | avx2-rr1-p5-div-x80.c | 182 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80() local
193 __m256 vf8 = _mm256_div_ps(ve8, vd8); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_div_x80()
|
| D | avx-rr2-p5-div-x72.c | 199 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_div_x72() local
209 __m256 vf8 = _mm256_div_ps(ve8, vd8); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_div_x72()
|
| D | avx-rr2-p5-div-x80.c | 214 const __m256 vd8 = _mm256_add_ps(ve8, vone); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_div_x80() local
225 __m256 vf8 = _mm256_div_ps(ve8, vd8); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_div_x80()
|
| /external/toolchain-utils/android_bench_suite/panorama_input/ |
| D | test_008.ppm | 2298 C)0 �� �����C0)bTB�vd8*(1"
|