/dports/math/blis/blis-0.8.1/kernels/zen/1f/ |
H A D | bli_dotxf_zen_int_8.c | 143 const dim_t n_iter_unroll = 1; in bli_sdotxf_zen_int_8() local 264 const dim_t n_iter_unroll = 4; in bli_sdotxf_zen_int_8() local 302 x0 += incx * n_iter_unroll; in bli_sdotxf_zen_int_8() 303 a0 += inca * n_iter_unroll; in bli_sdotxf_zen_int_8() 322 m -= n_iter_unroll * m_viter; in bli_sdotxf_zen_int_8() 323 a += n_iter_unroll * m_viter * inca; in bli_sdotxf_zen_int_8() 533 const dim_t n_iter_unroll = 1; in bli_ddotxf_zen_int_8() local 638 const dim_t n_iter_unroll = 3; in bli_ddotxf_zen_int_8() local 682 x0 += incx * n_iter_unroll; in bli_ddotxf_zen_int_8() 683 a0 += inca * n_iter_unroll; in bli_ddotxf_zen_int_8() [all …]
|
H A D | bli_axpyf_zen_int_8.c | 73 const dim_t n_iter_unroll = 1; in bli_saxpyf_zen_int_8() local 137 m_viter = ( m ) / ( n_elem_per_reg * n_iter_unroll ); in bli_saxpyf_zen_int_8() 138 m_left = ( m ) % ( n_elem_per_reg * n_iter_unroll ); in bli_saxpyf_zen_int_8() 282 const dim_t n_iter_unroll = 1; in bli_daxpyf_zen_int_8() local 346 m_viter = ( m ) / ( n_elem_per_reg * n_iter_unroll ); in bli_daxpyf_zen_int_8() 347 m_left = ( m ) % ( n_elem_per_reg * n_iter_unroll ); in bli_daxpyf_zen_int_8()
|
/dports/math/blis/blis-0.8.1/kernels/zen/1/ |
H A D | bli_axpyv_zen_int.c | 69 const dim_t n_iter_unroll = 4; in bli_saxpyv_zen_int() local 87 n_viter = ( n ) / ( n_elem_per_reg * n_iter_unroll ); in bli_saxpyv_zen_int() 88 n_left = ( n ) % ( n_elem_per_reg * n_iter_unroll ); in bli_saxpyv_zen_int() 135 x0 += n_elem_per_reg * n_iter_unroll; in bli_saxpyv_zen_int() 136 y0 += n_elem_per_reg * n_iter_unroll; in bli_saxpyv_zen_int() 173 const dim_t n_iter_unroll = 4; in bli_daxpyv_zen_int() local 191 n_viter = ( n ) / ( n_elem_per_reg * n_iter_unroll ); in bli_daxpyv_zen_int() 192 n_left = ( n ) % ( n_elem_per_reg * n_iter_unroll ); in bli_daxpyv_zen_int() 239 x0 += n_elem_per_reg * n_iter_unroll; in bli_daxpyv_zen_int() 240 y0 += n_elem_per_reg * n_iter_unroll; in bli_daxpyv_zen_int()
|
H A D | bli_dotxv_zen_int.c | 71 const dim_t n_iter_unroll = 4; in bli_sdotxv_zen_int() local 103 n_viter = ( n ) / ( n_elem_per_reg * n_iter_unroll ); in bli_sdotxv_zen_int() 104 n_left = ( n ) % ( n_elem_per_reg * n_iter_unroll ); in bli_sdotxv_zen_int() 147 x0 += ( n_elem_per_reg * n_iter_unroll ); in bli_sdotxv_zen_int() 148 y0 += ( n_elem_per_reg * n_iter_unroll ); in bli_sdotxv_zen_int() 199 const dim_t n_iter_unroll = 4; in bli_ddotxv_zen_int() local 231 n_viter = ( n ) / ( n_elem_per_reg * n_iter_unroll ); in bli_ddotxv_zen_int() 232 n_left = ( n ) % ( n_elem_per_reg * n_iter_unroll ); in bli_ddotxv_zen_int() 275 x0 += ( n_elem_per_reg * n_iter_unroll ); in bli_ddotxv_zen_int() 276 y0 += ( n_elem_per_reg * n_iter_unroll ); in bli_ddotxv_zen_int()
|
H A D | bli_dotv_zen_int.c | 69 const dim_t n_iter_unroll = 4; in bli_sdotv_zen_int() local 94 n_viter = ( n ) / ( n_elem_per_reg * n_iter_unroll ); in bli_sdotv_zen_int() 95 n_left = ( n ) % ( n_elem_per_reg * n_iter_unroll ); in bli_sdotv_zen_int() 141 x0 += ( n_elem_per_reg * n_iter_unroll ); in bli_sdotv_zen_int() 142 y0 += ( n_elem_per_reg * n_iter_unroll ); in bli_sdotv_zen_int() 191 const dim_t n_iter_unroll = 4; in bli_ddotv_zen_int() local 216 n_viter = ( n ) / ( n_elem_per_reg * n_iter_unroll ); in bli_ddotv_zen_int() 217 n_left = ( n ) % ( n_elem_per_reg * n_iter_unroll ); in bli_ddotv_zen_int() 263 x0 += ( n_elem_per_reg * n_iter_unroll ); in bli_ddotv_zen_int() 264 y0 += ( n_elem_per_reg * n_iter_unroll ); in bli_ddotv_zen_int()
|
H A D | bli_scalv_zen_int.c | 68 const dim_t n_iter_unroll = 4; in bli_sscalv_zen_int() local 101 n_viter = ( n ) / ( n_elem_per_reg * n_iter_unroll ); in bli_sscalv_zen_int() 102 n_left = ( n ) % ( n_elem_per_reg * n_iter_unroll ); in bli_sscalv_zen_int() 141 x0 += n_elem_per_reg * n_iter_unroll; in bli_sscalv_zen_int() 167 const dim_t n_iter_unroll = 4; in bli_dscalv_zen_int() local 200 n_viter = ( n ) / ( n_elem_per_reg * n_iter_unroll ); in bli_dscalv_zen_int() 201 n_left = ( n ) % ( n_elem_per_reg * n_iter_unroll ); in bli_dscalv_zen_int() 240 x0 += n_elem_per_reg * n_iter_unroll; in bli_dscalv_zen_int()
|
/dports/math/blis/blis-0.8.1/kernels/penryn/1f/ |
H A D | bli_axpyf_penryn_int.c | 66 const dim_t n_iter_unroll = 2; in bli_daxpyf_penryn_int() local 136 m_run = ( m - m_pre ) / ( n_elem_per_reg * n_iter_unroll ); in bli_daxpyf_penryn_int() 137 m_left = ( m - m_pre ) % ( n_elem_per_reg * n_iter_unroll ); in bli_daxpyf_penryn_int() 215 a0 += n_elem_per_reg * n_iter_unroll; in bli_daxpyf_penryn_int() 216 a1 += n_elem_per_reg * n_iter_unroll; in bli_daxpyf_penryn_int() 217 a2 += n_elem_per_reg * n_iter_unroll; in bli_daxpyf_penryn_int() 218 a3 += n_elem_per_reg * n_iter_unroll; in bli_daxpyf_penryn_int() 219 y0 += n_elem_per_reg * n_iter_unroll; in bli_daxpyf_penryn_int()
|
H A D | bli_axpy2v_penryn_int.c | 67 const dim_t n_iter_unroll = 4; in bli_daxpy2v_penryn_int() local 131 n_run = ( n - n_pre ) / ( n_elem_per_reg * n_iter_unroll ); in bli_daxpy2v_penryn_int() 132 n_left = ( n - n_pre ) % ( n_elem_per_reg * n_iter_unroll ); in bli_daxpy2v_penryn_int() 235 x1 += n_elem_per_reg * n_iter_unroll; in bli_daxpy2v_penryn_int() 236 y1 += n_elem_per_reg * n_iter_unroll; in bli_daxpy2v_penryn_int() 237 z1 += n_elem_per_reg * n_iter_unroll; in bli_daxpy2v_penryn_int()
|
H A D | bli_dotxf_penryn_int.c | 68 const dim_t n_iter_unroll = 4; in bli_ddotxf_penryn_int() local 155 m_run = ( m - m_pre ) / ( n_elem_per_reg * n_iter_unroll ); in bli_ddotxf_penryn_int() 156 m_left = ( m - m_pre ) % ( n_elem_per_reg * n_iter_unroll ); in bli_ddotxf_penryn_int() 241 x0 += n_elem_per_reg * n_iter_unroll; in bli_ddotxf_penryn_int() 242 x1 += n_elem_per_reg * n_iter_unroll; in bli_ddotxf_penryn_int() 243 x2 += n_elem_per_reg * n_iter_unroll; in bli_ddotxf_penryn_int() 244 x3 += n_elem_per_reg * n_iter_unroll; in bli_ddotxf_penryn_int() 245 y0 += n_elem_per_reg * n_iter_unroll; in bli_ddotxf_penryn_int()
|
H A D | bli_dotxaxpyf_penryn_int.c | 74 const dim_t n_iter_unroll = 2; in bli_ddotxaxpyf_penryn_int() local 174 m_run = ( m - m_pre ) / ( n_elem_per_reg * n_iter_unroll ); in bli_ddotxaxpyf_penryn_int() 175 m_left = ( m - m_pre ) % ( n_elem_per_reg * n_iter_unroll ); in bli_ddotxaxpyf_penryn_int() 296 a0 += n_elem_per_reg * n_iter_unroll; in bli_ddotxaxpyf_penryn_int() 298 a2 += n_elem_per_reg * n_iter_unroll; in bli_ddotxaxpyf_penryn_int() 300 w1 += n_elem_per_reg * n_iter_unroll; in bli_ddotxaxpyf_penryn_int() 301 z1 += n_elem_per_reg * n_iter_unroll; in bli_ddotxaxpyf_penryn_int()
|
/dports/math/blis/blis-0.8.1/kernels/penryn/1/ |
H A D | bli_axpyv_penryn_int.c | 62 const dim_t n_iter_unroll = 4; in bli_daxpyv_penryn_int() local 120 n_run = ( n - n_pre ) / ( n_elem_per_reg * n_iter_unroll ); in bli_daxpyv_penryn_int() 121 n_left = ( n - n_pre ) % ( n_elem_per_reg * n_iter_unroll ); in bli_daxpyv_penryn_int() 171 x1 += n_elem_per_reg * n_iter_unroll; in bli_daxpyv_penryn_int() 172 y1 += n_elem_per_reg * n_iter_unroll; in bli_daxpyv_penryn_int()
|
/dports/math/blis/blis-0.8.1/kernels/zen2/1f/ |
H A D | bli_axpyf_zen_int_5.c | 71 const dim_t n_iter_unroll = 2; in bli_saxpyf_zen_int_5() local 230 y0 += n_iter_unroll * n_elem_per_reg; in bli_saxpyf_zen_int_5() 231 a0 += n_iter_unroll * n_elem_per_reg; in bli_saxpyf_zen_int_5() 232 a1 += n_iter_unroll * n_elem_per_reg; in bli_saxpyf_zen_int_5() 233 a2 += n_iter_unroll * n_elem_per_reg; in bli_saxpyf_zen_int_5() 234 a3 += n_iter_unroll * n_elem_per_reg; in bli_saxpyf_zen_int_5() 235 a4 += n_iter_unroll * n_elem_per_reg; in bli_saxpyf_zen_int_5() 345 const dim_t n_iter_unroll = 2; in bli_daxpyf_zen_int_5() local 504 y0 += n_iter_unroll * n_elem_per_reg; in bli_daxpyf_zen_int_5() 505 a0 += n_iter_unroll * n_elem_per_reg; in bli_daxpyf_zen_int_5() [all …]
|
/dports/math/blis/blis-0.8.1/config/template/kernels/1/ |
H A D | bli_axpyv_template_noopt_var1.c | 100 const dim_t n_iter_unroll = 1; in bli_zaxpyv_template_noopt() local 102 const dim_t n_elem_per_iter = n_elem_per_reg * n_iter_unroll; in bli_zaxpyv_template_noopt()
|
H A D | bli_dotv_template_noopt_var1.c | 106 const dim_t n_iter_unroll = 1; in bli_zdotv_template_noopt() local 108 const dim_t n_elem_per_iter = n_elem_per_reg * n_iter_unroll; in bli_zdotv_template_noopt()
|
/dports/math/blis/blis-0.8.1/config/template/kernels/1f/ |
H A D | bli_axpyf_template_noopt_var1.c | 118 const dim_t n_iter_unroll = 1; in bli_zaxpyf_template_noopt() local 120 const dim_t n_elem_per_iter = n_elem_per_reg * n_iter_unroll; in bli_zaxpyf_template_noopt()
|
H A D | bli_dotxf_template_noopt_var1.c | 121 const dim_t n_iter_unroll = 1; in bli_zdotxf_template_noopt() local 123 const dim_t n_elem_per_iter = n_elem_per_reg * n_iter_unroll; in bli_zdotxf_template_noopt()
|
H A D | bli_axpy2v_template_noopt_var1.c | 110 const dim_t n_iter_unroll = 1; in bli_zaxpy2v_template_noopt() local 112 const dim_t n_elem_per_iter = n_elem_per_reg * n_iter_unroll; in bli_zaxpy2v_template_noopt()
|
H A D | bli_dotaxpyv_template_noopt_var1.c | 116 const dim_t n_iter_unroll = 1; in bli_zdotaxpyv_template_noopt() local 118 const dim_t n_elem_per_iter = n_elem_per_reg * n_iter_unroll; in bli_zdotaxpyv_template_noopt()
|
H A D | bli_dotxaxpyf_template_noopt_var1.c | 134 const dim_t n_iter_unroll = 1; in bli_zdotxaxpyf_template_noopt() local 136 const dim_t n_elem_per_iter = n_elem_per_reg * n_iter_unroll; in bli_zdotxaxpyf_template_noopt()
|