/external/llvm-project/llvm/test/CodeGen/Hexagon/ |
D | vect-vd0.ll | 13 %v1 = call <16 x i32> @llvm.hexagon.V6.vd0() 19 declare <16 x i32> @llvm.hexagon.V6.vd0() #1
|
D | hvx-vzero.ll | 12 %v1 = tail call <16 x i32> @llvm.hexagon.V6.vd0() 18 declare <16 x i32> @llvm.hexagon.V6.vd0() #1
|
D | hvx-double-vzero.ll | 11 %v1 = tail call <32 x i32> @llvm.hexagon.V6.vd0.128B() 17 declare <32 x i32> @llvm.hexagon.V6.vd0.128B() #1
|
D | v6-unaligned-spill.ll | 26 %v2 = tail call <32 x i32> @llvm.hexagon.V6.vd0.128B() #2 43 declare <32 x i32> @llvm.hexagon.V6.vd0.128B() #1
|
/external/XNNPACK/src/f32-sigmoid/gen/ |
D | avx2-rr1-p5-nr2fma-x16.c | 78 const __m256 vd0 = _mm256_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x16() local 81 __m256 vr0 = _mm256_rcp_ps(vd0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x16() 84 vr0 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x16() 87 vr0 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x16()
|
D | scalar-lut2048-p1-div-x2.c | 73 const float vd0 = vy0 + vone; in xnn_f32_sigmoid_ukernel__scalar_lut2048_p1_div_x2() local 76 float vf0 = vy0 / vd0; in xnn_f32_sigmoid_ukernel__scalar_lut2048_p1_div_x2()
|
D | avx2-rr1-p5-nr2fma-x24.c | 90 const __m256 vd0 = _mm256_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x24() local 94 __m256 vr0 = _mm256_rcp_ps(vd0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x24() 98 vr0 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x24() 102 vr0 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x24()
|
D | avx-rr2-p5-nr2-x16.c | 87 const __m256 vd0 = _mm256_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x16() local 90 __m256 vr0 = _mm256_rcp_ps(vd0); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x16() 93 vr0 = _mm256_mul_ps(vr0, _mm256_sub_ps(vtwo, _mm256_mul_ps(vr0, vd0))); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x16() 94 vr0 = _mm256_mul_ps(vr0, _mm256_sub_ps(vtwo, _mm256_mul_ps(vr0, vd0))); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x16()
|
D | avx512f-rr1-lut16-p3-perm-scalef-nr1fma-x32.c | 75 const __m512 vd0 = _mm512_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx512f_rr1_lut16_p3_perm_scalef_nr1fma_x32() local 78 __m512 vr0 = _mm512_rcp14_ps(vd0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_lut16_p3_perm_scalef_nr1fma_x32() 81 vr0 = _mm512_fmadd_ps(_mm512_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_lut16_p3_perm_scalef_nr1fma_x32()
|
D | avx512f-rr1-p5-scalef-nr1fma-x32.c | 72 const __m512 vd0 = _mm512_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x32() local 75 __m512 vr0 = _mm512_rcp14_ps(vd0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x32() 78 vr0 = _mm512_fmadd_ps(_mm512_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x32()
|
D | avx512f-rr2-lut32-p2-perm2-scalef-nr1fma-x32.c | 81 const __m512 vd0 = _mm512_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx512f_rr2_lut32_p2_perm2_scalef_nr1fma_x32() local 84 __m512 vr0 = _mm512_rcp14_ps(vd0); in xnn_f32_sigmoid_ukernel__avx512f_rr2_lut32_p2_perm2_scalef_nr1fma_x32() 87 vr0 = _mm512_fmadd_ps(_mm512_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx512f_rr2_lut32_p2_perm2_scalef_nr1fma_x32()
|
D | scalar-p5-div-x2.c | 80 const float vd0 = ve0 + vone; in xnn_f32_sigmoid_ukernel__scalar_p5_div_x2() local 83 float vf0 = ve0 / vd0; in xnn_f32_sigmoid_ukernel__scalar_p5_div_x2()
|
D | scalar-lut64-p2-div-x2.c | 76 const float vd0 = vy0 + vone; in xnn_f32_sigmoid_ukernel__scalar_lut64_p2_div_x2() local 79 float vf0 = vy0 / vd0; in xnn_f32_sigmoid_ukernel__scalar_lut64_p2_div_x2()
|
D | avx2-rr1-p5-nr1fma-x16.c | 78 const __m256 vd0 = _mm256_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x16() local 81 __m256 vr0 = _mm256_rcp_ps(vd0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x16() 84 vr0 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x16()
|
D | avx2-rr1-p5-nr2fma-x32.c | 102 const __m256 vd0 = _mm256_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x32() local 107 __m256 vr0 = _mm256_rcp_ps(vd0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x32() 112 vr0 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x32() 117 vr0 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr2fma_x32()
|
D | avx-rr2-p5-nr2-x24.c | 102 const __m256 vd0 = _mm256_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x24() local 106 __m256 vr0 = _mm256_rcp_ps(vd0); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x24() 110 vr0 = _mm256_mul_ps(vr0, _mm256_sub_ps(vtwo, _mm256_mul_ps(vr0, vd0))); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x24() 111 vr0 = _mm256_mul_ps(vr0, _mm256_sub_ps(vtwo, _mm256_mul_ps(vr0, vd0))); in xnn_f32_sigmoid_ukernel__avx_rr2_p5_nr2_x24()
|
D | avx512f-rr2-lut32-p2-perm2-scalef-nr1fma-x48.c | 92 const __m512 vd0 = _mm512_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx512f_rr2_lut32_p2_perm2_scalef_nr1fma_x48() local 96 __m512 vr0 = _mm512_rcp14_ps(vd0); in xnn_f32_sigmoid_ukernel__avx512f_rr2_lut32_p2_perm2_scalef_nr1fma_x48() 100 vr0 = _mm512_fmadd_ps(_mm512_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx512f_rr2_lut32_p2_perm2_scalef_nr1fma_x48()
|
D | avx512f-rr1-p5-scalef-nr1fma-x48.c | 83 const __m512 vd0 = _mm512_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x48() local 87 __m512 vr0 = _mm512_rcp14_ps(vd0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x48() 91 vr0 = _mm512_fmadd_ps(_mm512_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x48()
|
D | avx512f-rr1-lut16-p3-perm-scalef-nr1fma-x48.c | 86 const __m512 vd0 = _mm512_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx512f_rr1_lut16_p3_perm_scalef_nr1fma_x48() local 90 __m512 vr0 = _mm512_rcp14_ps(vd0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_lut16_p3_perm_scalef_nr1fma_x48() 94 vr0 = _mm512_fmadd_ps(_mm512_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_lut16_p3_perm_scalef_nr1fma_x48()
|
D | avx2-rr1-p5-nr1fma-x24.c | 90 const __m256 vd0 = _mm256_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x24() local 94 __m256 vr0 = _mm256_rcp_ps(vd0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x24() 98 vr0 = _mm256_fmadd_ps(_mm256_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx2_rr1_p5_nr1fma_x24()
|
D | avx512f-rr1-p5-scalef-nr1fma-x64.c | 94 const __m512 vd0 = _mm512_add_ps(ve0, vone); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x64() local 99 __m512 vr0 = _mm512_rcp14_ps(vd0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x64() 104 vr0 = _mm512_fmadd_ps(_mm512_fnmadd_ps(vr0, vd0, vone), vr0, vr0); in xnn_f32_sigmoid_ukernel__avx512f_rr1_p5_scalef_nr1fma_x64()
|
/external/XNNPACK/src/f32-ibilinear/gen/ |
D | scalar-c2.c | 64 const float vd0 = vb0 - vt0; in xnn_f32_ibilinear_ukernel__scalar_c2() local 67 const float vo0 = vt0 + vd0 * valphav; in xnn_f32_ibilinear_ukernel__scalar_c2()
|
D | scalar-c4.c | 80 const float vd0 = vb0 - vt0; in xnn_f32_ibilinear_ukernel__scalar_c4() local 85 const float vo0 = vt0 + vd0 * valphav; in xnn_f32_ibilinear_ukernel__scalar_c4()
|
/external/XNNPACK/src/f32-ibilinear-chw/gen/ |
D | scalar-p2.c | 66 const float vd0 = vb0 - vt0; in xnn_f32_ibilinear_chw_ukernel__scalar_p2() local 69 const float vo0 = vt0 + vd0 * valphav0; in xnn_f32_ibilinear_chw_ukernel__scalar_p2()
|
D | scalar-p4.c | 90 const float vd0 = vb0 - vt0; in xnn_f32_ibilinear_chw_ukernel__scalar_p4() local 95 const float vo0 = vt0 + vd0 * valphav0; in xnn_f32_ibilinear_chw_ukernel__scalar_p4()
|