nb_kernel234_ia32_sse.s

来自「最著名最快的分子模拟软件」· S 代码 · 共 1,980 行 · 第 1/5 页

S
1,980 行
字号
12345
    movhps 8(%esi,%edx,4),%xmm3    ## other half of table      movaps %xmm7,%xmm6    shufps $136,%xmm3,%xmm6 ## constant 10001000    shufps $221,%xmm3,%xmm7 ## constant 11011101    ## dispersion table ready, in xmm4-xmm7     mulps  %xmm1,%xmm6      ## xmm6=Geps     mulps  %xmm2,%xmm7      ## xmm7=Heps2     addps  %xmm6,%xmm5    addps  %xmm7,%xmm5      ## xmm5=Fp     mulps  nb234_two(%esp),%xmm7         ## two*Heps2     addps  %xmm6,%xmm7    addps  %xmm5,%xmm7 ## xmm7=FF     mulps  %xmm1,%xmm5 ## xmm5=eps*Fp     addps  %xmm4,%xmm5 ## xmm5=VV     movaps nb234_c6(%esp),%xmm4    mulps  %xmm4,%xmm7   ## fijD     mulps  %xmm4,%xmm5   ## Vvdw6     movaps  %xmm7,nb234_fstmp(%esp)    addps  nb234_Vvdwtot(%esp),%xmm5    movaps %xmm5,nb234_Vvdwtot(%esp)    ## repulsion     movlps 16(%esi,%eax,4),%xmm5    movlps 16(%esi,%ecx,4),%xmm7    movhps 16(%esi,%ebx,4),%xmm5    movhps 16(%esi,%edx,4),%xmm7    ## got half table     movaps %xmm5,%xmm4    shufps $136,%xmm7,%xmm4 ## constant 10001000    shufps $221,%xmm7,%xmm5 ## constant 11011101    movlps 24(%esi,%eax,4),%xmm7    movlps 24(%esi,%ecx,4),%xmm3    movhps 24(%esi,%ebx,4),%xmm7    movhps 24(%esi,%edx,4),%xmm3    ## other half of table      movaps %xmm7,%xmm6    shufps $136,%xmm3,%xmm6 ## constant 10001000    shufps $221,%xmm3,%xmm7 ## constant 11011101    ## repulsion table ready, in xmm4-xmm7     mulps  %xmm1,%xmm6      ## xmm6=Geps     mulps  %xmm2,%xmm7      ## xmm7=Heps2     addps  %xmm6,%xmm5    addps  %xmm7,%xmm5      ## xmm5=Fp     mulps  nb234_two(%esp),%xmm7         ## two*Heps2     addps  %xmm6,%xmm7    addps  %xmm5,%xmm7 ## xmm7=FF     mulps  %xmm1,%xmm5 ## xmm5=eps*Fp     addps  %xmm4,%xmm5 ## xmm5=VV     movaps nb234_c12(%esp),%xmm4    mulps  %xmm4,%xmm7 ## fijR     mulps  %xmm4,%xmm5 ## Vvdw12     addps  nb234_fstmp(%esp),%xmm7    mulps nb234_tsc(%esp),%xmm7    addps  nb234_Vvdwtot(%esp),%xmm5    movaps %xmm5,nb234_Vvdwtot(%esp)        xorps %xmm2,%xmm2        mulps  %xmm0,%xmm7        subps %xmm7,%xmm2        movaps %xmm2,%xmm0        movaps %xmm2,%xmm1    movd  %mm0,%eax    movd  %mm1,%ebx    movd  %mm2,%ecx    movd  %mm3,%edx        xorps %xmm3,%xmm3        movaps %xmm3,%xmm4        movaps %xmm3,%xmm5        mulps nb234_dxOO(%esp),%xmm0        mulps nb234_dyOO(%esp),%xmm1        mulps nb234_dzOO(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixO(%esp),%xmm0        addps nb234_fiyO(%esp),%xmm1        addps nb234_fizO(%esp),%xmm2        movaps %xmm3,nb234_fjxO(%esp)        movaps %xmm4,nb234_fjyO(%esp)        movaps %xmm5,nb234_fjzO(%esp)        movaps %xmm0,nb234_fixO(%esp)        movaps %xmm1,nb234_fiyO(%esp)        movaps %xmm2,nb234_fizO(%esp)        ## Coulomb interactions         ## start with H1-H1 interaction         movaps nb234_rinvH1H1(%esp),%xmm0        movaps %xmm0,%xmm7      ## xmm7=rinv         movaps nb234_krf(%esp),%xmm5        mulps  %xmm0,%xmm0      ## xmm0=rinvsq         mulps  nb234_rsqH1H1(%esp),%xmm5   ## xmm5=krsq         movaps %xmm5,%xmm6        addps  %xmm7,%xmm6      ## xmm6=rinv+ krsq         subps  nb234_crf(%esp),%xmm6        mulps  nb234_qqHH(%esp),%xmm6   ## xmm6=voul=qq*(rinv+ krsq-crf)         mulps nb234_two(%esp),%xmm5        subps  %xmm5,%xmm7      ## xmm7=rinv-2*krsq         mulps  nb234_qqHH(%esp),%xmm7   ## xmm7 = coul part of fscal         addps  nb234_vctot(%esp),%xmm6   ## local vctot summation variable         mulps  %xmm7,%xmm0        movaps %xmm0,%xmm1        movaps %xmm0,%xmm2        xorps %xmm3,%xmm3        movaps %xmm3,%xmm4        movaps %xmm3,%xmm5        mulps nb234_dxH1H1(%esp),%xmm0        mulps nb234_dyH1H1(%esp),%xmm1        mulps nb234_dzH1H1(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixH1(%esp),%xmm0        addps nb234_fiyH1(%esp),%xmm1        addps nb234_fizH1(%esp),%xmm2        movaps %xmm3,nb234_fjxH1(%esp)        movaps %xmm4,nb234_fjyH1(%esp)        movaps %xmm5,nb234_fjzH1(%esp)        movaps %xmm0,nb234_fixH1(%esp)        movaps %xmm1,nb234_fiyH1(%esp)        movaps %xmm2,nb234_fizH1(%esp)        ## H1-H2 interaction         movaps nb234_rinvH1H2(%esp),%xmm0        movaps %xmm0,%xmm7      ## xmm7=rinv         movaps nb234_krf(%esp),%xmm5        movaps %xmm0,%xmm1        mulps  nb234_rsqH1H2(%esp),%xmm5   ## xmm5=krsq         movaps %xmm5,%xmm4        addps  %xmm7,%xmm4      ## xmm4=r inv+ krsq         subps  nb234_crf(%esp),%xmm4        mulps  %xmm0,%xmm0        mulps  nb234_qqHH(%esp),%xmm4   ## xmm4=voul=qq*(rinv+ krsq-crf)         mulps  nb234_two(%esp),%xmm5        subps  %xmm5,%xmm7      ## xmm7=rinv-2*krsq         mulps  nb234_qqHH(%esp),%xmm7   ## xmm7 = coul part of fscal         addps  %xmm4,%xmm6      ## add to local vctot         mulps %xmm7,%xmm0       ## fsOH1          movaps %xmm0,%xmm1        movaps %xmm0,%xmm2        xorps %xmm3,%xmm3        movaps %xmm3,%xmm4        movaps %xmm3,%xmm5        mulps nb234_dxH1H2(%esp),%xmm0        mulps nb234_dyH1H2(%esp),%xmm1        mulps nb234_dzH1H2(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixH1(%esp),%xmm0        addps nb234_fiyH1(%esp),%xmm1        addps nb234_fizH1(%esp),%xmm2        movaps %xmm3,nb234_fjxH2(%esp)        movaps %xmm4,nb234_fjyH2(%esp)        movaps %xmm5,nb234_fjzH2(%esp)        movaps %xmm0,nb234_fixH1(%esp)        movaps %xmm1,nb234_fiyH1(%esp)        movaps %xmm2,nb234_fizH1(%esp)        ## H1-M interaction          movaps nb234_rinvH1M(%esp),%xmm0        movaps %xmm0,%xmm7      ## xmm7=Rinv         movaps nb234_krf(%esp),%xmm5        movaps %xmm0,%xmm1        mulps  nb234_rsqH1M(%esp),%xmm5   ## xmm5=krsq         movaps %xmm5,%xmm4        addps  %xmm7,%xmm4      ## xmm4=r inv+ krsq         subps  nb234_crf(%esp),%xmm4        mulps %xmm0,%xmm0        mulps  nb234_qqMH(%esp),%xmm4   ## xmm4=voul=qq*(rinv+ krsq-crf)         mulps  nb234_two(%esp),%xmm5        subps  %xmm5,%xmm7      ## xmm7=rinv-2*krsq         mulps  nb234_qqMH(%esp),%xmm7   ## xmm7 = coul part of fscal         addps  %xmm4,%xmm6      ## add to local vctot         mulps %xmm7,%xmm0       ## fsOH2         movaps %xmm0,%xmm1        movaps %xmm0,%xmm2        xorps %xmm3,%xmm3        movaps %xmm3,%xmm4        movaps %xmm3,%xmm5        mulps nb234_dxH1M(%esp),%xmm0        mulps nb234_dyH1M(%esp),%xmm1        mulps nb234_dzH1M(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixH1(%esp),%xmm0        addps nb234_fiyH1(%esp),%xmm1        addps nb234_fizH1(%esp),%xmm2        movaps %xmm3,nb234_fjxM(%esp)        movaps %xmm4,nb234_fjyM(%esp)        movaps %xmm5,nb234_fjzM(%esp)        movaps %xmm0,nb234_fixH1(%esp)        movaps %xmm1,nb234_fiyH1(%esp)        movaps %xmm2,nb234_fizH1(%esp)        ## H2-H1 interaction         movaps nb234_rinvH2H1(%esp),%xmm0        movaps %xmm0,%xmm7      ## xmm7=rinv         movaps nb234_krf(%esp),%xmm5        movaps %xmm0,%xmm1        mulps  nb234_rsqH2H1(%esp),%xmm5   ## xmm5=krsq         movaps %xmm5,%xmm4        addps  %xmm7,%xmm4      ## xmm4=r inv+ krsq         subps  nb234_crf(%esp),%xmm4        mulps %xmm0,%xmm0        mulps  nb234_qqHH(%esp),%xmm4   ## xmm4=voul=qq*(rinv+ krsq-crf)         mulps  nb234_two(%esp),%xmm5        subps  %xmm5,%xmm7      ## xmm7=rinv-2*krsq         mulps  nb234_qqHH(%esp),%xmm7   ## xmm7 = coul part of fscal         addps  %xmm4,%xmm6      ## add to local vctot         mulps %xmm7,%xmm0       ## fsOH2         movaps %xmm0,%xmm1        movaps %xmm0,%xmm2        movaps nb234_fjxH1(%esp),%xmm3        movaps nb234_fjyH1(%esp),%xmm4        movaps nb234_fjzH1(%esp),%xmm5        mulps nb234_dxH2H1(%esp),%xmm0        mulps nb234_dyH2H1(%esp),%xmm1        mulps nb234_dzH2H1(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixH2(%esp),%xmm0        addps nb234_fiyH2(%esp),%xmm1        addps nb234_fizH2(%esp),%xmm2        movaps %xmm3,nb234_fjxH1(%esp)        movaps %xmm4,nb234_fjyH1(%esp)        movaps %xmm5,nb234_fjzH1(%esp)        movaps %xmm0,nb234_fixH2(%esp)        movaps %xmm1,nb234_fiyH2(%esp)        movaps %xmm2,nb234_fizH2(%esp)        ## H2-H2 interaction         movaps nb234_rinvH2H2(%esp),%xmm0        movaps %xmm0,%xmm7      ## xmm7=rinv         movaps nb234_krf(%esp),%xmm5        movaps %xmm0,%xmm1        mulps  nb234_rsqH2H2(%esp),%xmm5   ## xmm5=krsq         movaps %xmm5,%xmm4        addps  %xmm7,%xmm4      ## xmm4=r inv+ krsq         subps  nb234_crf(%esp),%xmm4        mulps %xmm0,%xmm0        mulps  nb234_qqHH(%esp),%xmm4   ## xmm4=voul=qq*(rinv+ krsq-crf)         mulps  nb234_two(%esp),%xmm5        subps  %xmm5,%xmm7      ## xmm7=rinv-2*krsq         mulps  nb234_qqHH(%esp),%xmm7   ## xmm7 = coul part of fscal         addps  %xmm4,%xmm6      ## add to local vctot         mulps %xmm7,%xmm0       ## fsOH2         movaps %xmm0,%xmm1        movaps %xmm0,%xmm2        movaps nb234_fjxH2(%esp),%xmm3        movaps nb234_fjyH2(%esp),%xmm4        movaps nb234_fjzH2(%esp),%xmm5        mulps nb234_dxH2H2(%esp),%xmm0        mulps nb234_dyH2H2(%esp),%xmm1        mulps nb234_dzH2H2(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixH2(%esp),%xmm0        addps nb234_fiyH2(%esp),%xmm1        addps nb234_fizH2(%esp),%xmm2        movaps %xmm3,nb234_fjxH2(%esp)        movaps %xmm4,nb234_fjyH2(%esp)        movaps %xmm5,nb234_fjzH2(%esp)        movaps %xmm0,nb234_fixH2(%esp)        movaps %xmm1,nb234_fiyH2(%esp)        movaps %xmm2,nb234_fizH2(%esp)        ## H2-M interaction         movaps nb234_rinvH2M(%esp),%xmm0        movaps %xmm0,%xmm7      ## xmm7=rinv         movaps nb234_krf(%esp),%xmm5        movaps %xmm0,%xmm1        mulps  nb234_rsqH2M(%esp),%xmm5   ## xmm5=krsq         movaps %xmm5,%xmm4        addps  %xmm7,%xmm4      ## xmm4=r inv+ krsq         subps  nb234_crf(%esp),%xmm4        mulps %xmm0,%xmm0        mulps  nb234_qqMH(%esp),%xmm4   ## xmm4=voul=qq*(rinv+ krsq-crf)         mulps  nb234_two(%esp),%xmm5        subps  %xmm5,%xmm7      ## xmm7=rinv-2*krsq         mulps  nb234_qqMH(%esp),%xmm7   ## xmm7 = coul part of fscal         addps  %xmm4,%xmm6      ## add to local vctot         mulps %xmm7,%xmm0       ## fsOH2         movaps %xmm0,%xmm1        movaps %xmm0,%xmm2        movaps nb234_fjxM(%esp),%xmm3        movaps nb234_fjyM(%esp),%xmm4        movaps nb234_fjzM(%esp),%xmm5        mulps nb234_dxH2M(%esp),%xmm0        mulps nb234_dyH2M(%esp),%xmm1        mulps nb234_dzH2M(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixH2(%esp),%xmm0        addps nb234_fiyH2(%esp),%xmm1        addps nb234_fizH2(%esp),%xmm2        movaps %xmm3,nb234_fjxM(%esp)        movaps %xmm4,nb234_fjyM(%esp)        movaps %xmm5,nb234_fjzM(%esp)        movaps %xmm0,nb234_fixH2(%esp)        movaps %xmm1,nb234_fiyH2(%esp)        movaps %xmm2,nb234_fizH2(%esp)        ## M-H1 interaction         movaps nb234_rinvMH1(%esp),%xmm0        movaps %xmm0,%xmm7      ## xmm7=rinv         movaps nb234_krf(%esp),%xmm5        movaps %xmm0,%xmm1        mulps  nb234_rsqMH1(%esp),%xmm5   ## xmm5=krsq         movaps %xmm5,%xmm4        addps  %xmm7,%xmm4      ## xmm4=r inv+ krsq         subps  nb234_crf(%esp),%xmm4        mulps %xmm0,%xmm0        mulps  nb234_qqMH(%esp),%xmm4   ## xmm4=voul=qq*(rinv+ krsq-crf)         mulps  nb234_two(%esp),%xmm5        subps  %xmm5,%xmm7      ## xmm7=rinv-2*krsq         mulps  nb234_qqMH(%esp),%xmm7   ## xmm7 = coul part of fscal         addps  %xmm4,%xmm6      ## add to local vctot         mulps %xmm7,%xmm0       ## fsOH2         movaps %xmm0,%xmm1        movaps %xmm0,%xmm2        movaps nb234_fjxH1(%esp),%xmm3        movaps nb234_fjyH1(%esp),%xmm4        movaps nb234_fjzH1(%esp),%xmm5        mulps nb234_dxMH1(%esp),%xmm0        mulps nb234_dyMH1(%esp),%xmm1        mulps nb234_dzMH1(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixM(%esp),%xmm0        addps nb234_fiyM(%esp),%xmm1        addps nb234_fizM(%esp),%xmm2        movaps %xmm3,nb234_fjxH1(%esp)        movaps %xmm4,nb234_fjyH1(%esp)        movaps %xmm5,nb234_fjzH1(%esp)        movaps %xmm0,nb234_fixM(%esp)        movaps %xmm1,nb234_fiyM(%esp)        movaps %xmm2,nb234_fizM(%esp)        ## M-H2 interaction         movaps nb234_rinvMH2(%esp),%xmm0        movaps %xmm0,%xmm7      ## xmm7=rinv         movaps nb234_krf(%esp),%xmm5        movaps %xmm0,%xmm1        mulps  nb234_rsqMH2(%esp),%xmm5   ## xmm5=krsq         movaps %xmm5,%xmm4        addps  %xmm7,%xmm4      ## xmm4=r inv+ krsq         subps  nb234_crf(%esp),%xmm4        mulps %xmm0,%xmm0        mulps  nb234_qqMH(%esp),%xmm4   ## xmm4=voul=qq*(rinv+ krsq-crf)         mulps  nb234_two(%esp),%xmm5        subps  %xmm5,%xmm7      ## xmm7=rinv-2*krsq         mulps  nb234_qqMH(%esp),%xmm7   ## xmm7 = coul part of fscal         addps  %xmm4,%xmm6      ## add to local vctot         mulps %xmm7,%xmm0       ## fsOH2         movaps %xmm0,%xmm1        movaps %xmm0,%xmm2        movaps nb234_fjxH2(%esp),%xmm3        movaps nb234_fjyH2(%esp),%xmm4        movaps nb234_fjzH2(%esp),%xmm5        mulps nb234_dxMH2(%esp),%xmm0        mulps nb234_dyMH2(%esp),%xmm1        mulps nb234_dzMH2(%esp),%xmm2        subps %xmm0,%xmm3        subps %xmm1,%xmm4        subps %xmm2,%xmm5        addps nb234_fixM(%esp),%xmm0        addps nb234_fiyM(%esp),%xmm1        addps nb234_fizM(%esp),%xmm2        movaps %xmm3,nb234_fjxH2(%esp)        movaps %xmm4,nb234_fjyH2(%esp)        movaps %xmm5,nb234_fjzH2(%esp)        movaps %xmm0,nb234_fixM(%esp)        movaps %xmm1,nb234_fiyM(%esp)
12345

nb_kernel234_ia32_sse.s - 源码说明

本页面展示了「最著名最快的分子模拟软件」中的 nb_kernel234_ia32_sse.s 源码文件，采用 S 编程语言编写，共 1,980 行代码。您可以在线阅读完整代码内容，也可以返回资源详情页下载完整源码包进行本地学习和开发。

虫虫开发者社区收录了大量与分子模拟相关的技术资源，包括源代码、技术文档、电路图等，是电子工程师和嵌入式开发者的专业学习平台。

⌨️ 快捷键说明

复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?