nb_kernel310_x86_64_sse2.s
来自「最著名最快的分子模拟软件」· S 代码 · 共 1,320 行 · 第 1/3 页
S
1,320 行
#### $Id: nb_kernel310_x86_64_sse2.s,v 1.4.2.3 2006/09/22 08:40:36 lindahl Exp $#### Gromacs 4.0 Copyright (c) 1991-2003 ## David van der Spoel, Erik Lindahl#### This program is free software; you can redistribute it and/or## modify it under the terms of the GNU General Public License## as published by the Free Software Foundation; either version 2## of the License, or (at your option) any later version.#### To help us fund GROMACS development, we humbly ask that you cite## the research papers on the package. Check out http://www.gromacs.org## ## And Hey:## Gnomes, ROck Monsters And Chili Sauce##.globl nb_kernel310_x86_64_sse2.globl _nb_kernel310_x86_64_sse2nb_kernel310_x86_64_sse2: _nb_kernel310_x86_64_sse2: ## Room for return address and rbp (16 bytes).set nb310_fshift, 16.set nb310_gid, 24.set nb310_pos, 32.set nb310_faction, 40.set nb310_charge, 48.set nb310_p_facel, 56.set nb310_argkrf, 64.set nb310_argcrf, 72.set nb310_Vc, 80.set nb310_type, 88.set nb310_p_ntype, 96.set nb310_vdwparam, 104.set nb310_Vvdw, 112.set nb310_p_tabscale, 120.set nb310_VFtab, 128.set nb310_invsqrta, 136.set nb310_dvda, 144.set nb310_p_gbtabscale, 152.set nb310_GBtab, 160.set nb310_p_nthreads, 168.set nb310_count, 176.set nb310_mtx, 184.set nb310_outeriter, 192.set nb310_inneriter, 200.set nb310_work, 208 ## stack offsets for local variables ## bottom of stack is cache-aligned for sse2 use .set nb310_ix, 0.set nb310_iy, 16.set nb310_iz, 32.set nb310_iq, 48.set nb310_dx, 64.set nb310_dy, 80.set nb310_dz, 96.set nb310_two, 112.set nb310_six, 128.set nb310_twelve, 144.set nb310_tsc, 160.set nb310_qq, 176.set nb310_c6, 192.set nb310_c12, 208.set nb310_eps, 224.set nb310_vctot, 240.set nb310_Vvdwtot, 256.set nb310_fix, 272.set nb310_fiy, 288.set nb310_fiz, 304.set nb310_half, 320.set nb310_three, 336.set nb310_nri, 352.set nb310_iinr, 360.set nb310_jindex, 368.set nb310_jjnr, 376.set nb310_shift, 384.set nb310_shiftvec, 392.set nb310_facel, 400.set nb310_innerjjnr, 408.set nb310_is3, 416.set nb310_ii3, 420.set nb310_ntia, 424.set nb310_innerk, 428.set nb310_n, 432.set nb310_nn1, 436.set nb310_ntype, 440.set nb310_nouter, 444.set nb310_ninner, 448 push %rbp movq %rsp,%rbp push %rbx emms push %r12 push %r13 push %r14 push %r15 subq $472,%rsp ## local variable stack space (n*16+8) ## zero 32-bit iteration counters movl $0,%eax movl %eax,nb310_nouter(%rsp) movl %eax,nb310_ninner(%rsp) movl (%rdi),%edi movl %edi,nb310_nri(%rsp) movq %rsi,nb310_iinr(%rsp) movq %rdx,nb310_jindex(%rsp) movq %rcx,nb310_jjnr(%rsp) movq %r8,nb310_shift(%rsp) movq %r9,nb310_shiftvec(%rsp) movq nb310_p_ntype(%rbp),%rdi movl (%rdi),%edi movl %edi,nb310_ntype(%rsp) movq nb310_p_facel(%rbp),%rsi movsd (%rsi),%xmm0 movsd %xmm0,nb310_facel(%rsp) movq nb310_p_tabscale(%rbp),%rax movsd (%rax),%xmm3 shufpd $0,%xmm3,%xmm3 movapd %xmm3,nb310_tsc(%rsp) ## create constant floating-point factors on stack movl $0x00000000,%eax ## lower half of double half IEEE (hex) movl $0x3fe00000,%ebx movl %eax,nb310_half(%rsp) movl %ebx,nb310_half+4(%rsp) movsd nb310_half(%rsp),%xmm1 shufpd $0,%xmm1,%xmm1 ## splat to all elements movapd %xmm1,%xmm3 addpd %xmm3,%xmm3 ## one movapd %xmm3,%xmm2 addpd %xmm2,%xmm2 ## two addpd %xmm2,%xmm3 ## three movapd %xmm3,%xmm4 addpd %xmm4,%xmm4 ## six movapd %xmm4,%xmm5 addpd %xmm5,%xmm5 ## twelve movapd %xmm1,nb310_half(%rsp) movapd %xmm2,nb310_two(%rsp) movapd %xmm3,nb310_three(%rsp) movapd %xmm4,nb310_six(%rsp) movapd %xmm5,nb310_twelve(%rsp)_nb_kernel310_x86_64_sse2.nb310_threadloop: movq nb310_count(%rbp),%rsi ## pointer to sync counter movl (%rsi),%eax_nb_kernel310_x86_64_sse2.nb310_spinlock: movl %eax,%ebx ## ebx=*count=nn0 addl $1,%ebx ## ebx=nn1=nn0+10 lock cmpxchgl %ebx,(%rsi) ## write nn1 to *counter, ## if it hasnt changed. ## or reread *counter to eax. pause ## -> better p4 performance jnz _nb_kernel310_x86_64_sse2.nb310_spinlock ## if(nn1>nri) nn1=nri movl nb310_nri(%rsp),%ecx movl %ecx,%edx subl %ebx,%ecx cmovlel %edx,%ebx ## if(nn1>nri) nn1=nri ## Cleared the spinlock if we got here. ## eax contains nn0, ebx contains nn1. movl %eax,nb310_n(%rsp) movl %ebx,nb310_nn1(%rsp) subl %eax,%ebx ## calc number of outer lists movl %eax,%esi ## copy n to esi jg _nb_kernel310_x86_64_sse2.nb310_outerstart jmp _nb_kernel310_x86_64_sse2.nb310_end_nb_kernel310_x86_64_sse2.nb310_outerstart: ## ebx contains number of outer iterations addl nb310_nouter(%rsp),%ebx movl %ebx,nb310_nouter(%rsp)_nb_kernel310_x86_64_sse2.nb310_outer: movq nb310_shift(%rsp),%rax ## rax = pointer into shift[] movl (%rax,%rsi,4),%ebx ## rbx=shift[n] lea (%rbx,%rbx,2),%rbx ## rbx=3*is movl %ebx,nb310_is3(%rsp) ## store is3 movq nb310_shiftvec(%rsp),%rax ## rax = base of shiftvec[] movsd (%rax,%rbx,8),%xmm0 movsd 8(%rax,%rbx,8),%xmm1 movsd 16(%rax,%rbx,8),%xmm2 movq nb310_iinr(%rsp),%rcx ## rcx = pointer into iinr[] movl (%rcx,%rsi,4),%ebx ## ebx =ii movq nb310_charge(%rbp),%rdx movsd (%rdx,%rbx,8),%xmm3 mulsd nb310_facel(%rsp),%xmm3 shufpd $0,%xmm3,%xmm3 movq nb310_type(%rbp),%rdx movl (%rdx,%rbx,4),%edx imull nb310_ntype(%rsp),%edx shll %edx movl %edx,nb310_ntia(%rsp) lea (%rbx,%rbx,2),%rbx ## rbx = 3*ii=ii3 movq nb310_pos(%rbp),%rax ## rax = base of pos[] addsd (%rax,%rbx,8),%xmm0 addsd 8(%rax,%rbx,8),%xmm1 addsd 16(%rax,%rbx,8),%xmm2 movapd %xmm3,nb310_iq(%rsp) shufpd $0,%xmm0,%xmm0 shufpd $0,%xmm1,%xmm1 shufpd $0,%xmm2,%xmm2 movapd %xmm0,nb310_ix(%rsp) movapd %xmm1,nb310_iy(%rsp) movapd %xmm2,nb310_iz(%rsp) movl %ebx,nb310_ii3(%rsp) ## clear vctot and i forces xorpd %xmm15,%xmm15 movapd %xmm15,nb310_vctot(%rsp) movapd %xmm15,nb310_Vvdwtot(%rsp) movapd %xmm15,%xmm14 movapd %xmm15,%xmm13 movq nb310_jindex(%rsp),%rax movl (%rax,%rsi,4),%ecx ## jindex[n] movl 4(%rax,%rsi,4),%edx ## jindex[n+1] subl %ecx,%edx ## number of innerloop atoms movq nb310_pos(%rbp),%rsi movq nb310_faction(%rbp),%rdi movq nb310_jjnr(%rsp),%rax shll $2,%ecx addq %rcx,%rax movq %rax,nb310_innerjjnr(%rsp) ## pointer to jjnr[nj0] movl %edx,%ecx subl $2,%edx addl nb310_ninner(%rsp),%ecx movl %ecx,nb310_ninner(%rsp) addl $0,%edx movl %edx,nb310_innerk(%rsp) ## number of innerloop atoms jge _nb_kernel310_x86_64_sse2.nb310_unroll_loop jmp _nb_kernel310_x86_64_sse2.nb310_checksingle_nb_kernel310_x86_64_sse2.nb310_unroll_loop: ## twice unrolled innerloop here movq nb310_innerjjnr(%rsp),%rdx ## pointer to jjnr[k] movl (%rdx),%r10d movl 4(%rdx),%r11d addq $8,nb310_innerjjnr(%rsp) ## advance pointer (unrolled 2) movq nb310_pos(%rbp),%rsi ## base of pos[] lea (%r10,%r10,2),%rax ## replace jnr with j3 lea (%r11,%r11,2),%rbx ## move two coordinates to xmm4-xmm6 movlpd (%rsi,%rax,8),%xmm4 movlpd 8(%rsi,%rax,8),%xmm5 movlpd 16(%rsi,%rax,8),%xmm6 movhpd (%rsi,%rbx,8),%xmm4 movhpd 8(%rsi,%rbx,8),%xmm5 movhpd 16(%rsi,%rbx,8),%xmm6 ## calc dr subpd nb310_ix(%rsp),%xmm4 subpd nb310_iy(%rsp),%xmm5 subpd nb310_iz(%rsp),%xmm6 ## store dr movapd %xmm4,%xmm9 movapd %xmm5,%xmm10 movapd %xmm6,%xmm11 movq nb310_charge(%rbp),%rsi ## base of charge[] ## square it mulpd %xmm4,%xmm4 mulpd %xmm5,%xmm5 mulpd %xmm6,%xmm6 addpd %xmm5,%xmm4 addpd %xmm6,%xmm4 movlpd (%rsi,%r10,8),%xmm3 cvtpd2ps %xmm4,%xmm5 rsqrtps %xmm5,%xmm5 cvtps2pd %xmm5,%xmm2 ## lu in low xmm2 movhpd (%rsi,%r11,8),%xmm3 ## lookup seed in xmm2 movapd %xmm2,%xmm5 ## copy of lu mulpd %xmm2,%xmm2 ## lu*lu movapd nb310_three(%rsp),%xmm1 mulpd %xmm4,%xmm2 ## rsq*lu*lu movapd nb310_half(%rsp),%xmm0 subpd %xmm2,%xmm1 ## 30-rsq*lu*lu mulpd %xmm5,%xmm1 mulpd %xmm0,%xmm1 ## xmm0=iter1 of rinv (new lu) mulpd nb310_iq(%rsp),%xmm3 movapd %xmm3,nb310_qq(%rsp) movq nb310_type(%rbp),%rsi movapd %xmm1,%xmm5 ## copy of lu mulpd %xmm1,%xmm1 ## lu*lu movapd nb310_three(%rsp),%xmm2 mulpd %xmm4,%xmm1 ## rsq*lu*lu movapd nb310_half(%rsp),%xmm0 subpd %xmm1,%xmm2 ## 30-rsq*lu*lu mulpd %xmm5,%xmm2 mulpd %xmm2,%xmm0 ## xmm0=rinv movapd %xmm0,%xmm1 ## xmm1=rinv movl (%rsi,%r10,4),%r12d movl (%rsi,%r11,4),%r13d mulpd %xmm0,%xmm4 ## xmm4=r mulpd nb310_tsc(%rsp),%xmm4 cvttpd2pi %xmm4,%mm6 ## mm6 = lu idx cvtpi2pd %mm6,%xmm5 subpd %xmm5,%xmm4 movapd %xmm4,nb310_eps(%rsp) shll %r12d shll %r13d pslld $2,%mm6 ## idx *= 4 movq nb310_VFtab(%rbp),%rsi movd %mm6,%r8d psrlq $32,%mm6 movd %mm6,%r9d movl nb310_ntia(%rsp),%edi addl %edi,%r12d addl %edi,%r13d movq nb310_vdwparam(%rbp),%rdi movapd (%rsi,%r8,8),%xmm4 ## Y1 F1 movapd (%rsi,%r9,8),%xmm8 ## Y2 F2 movapd %xmm4,%xmm5 unpcklpd %xmm8,%xmm4 ## Y1 Y2 unpckhpd %xmm8,%xmm5 ## F1 F2 movlpd (%rdi,%r12,8),%xmm3 movlpd 8(%rdi,%r12,8),%xmm7 movapd %xmm1,%xmm0 ## rinv mulpd %xmm0,%xmm0 ## rinvsq movapd %xmm0,%xmm2 ## rinvsq mulpd %xmm2,%xmm2 ## rinv4 mulpd %xmm0,%xmm2 ## rinv6 movapd %xmm2,%xmm12 mulpd %xmm12,%xmm12 ## rinv12 movhpd (%rdi,%r13,8),%xmm3 movhpd 8(%rdi,%r13,8),%xmm7 movapd 16(%rsi,%r8,8),%xmm6 ## G1 H1 movapd 16(%rsi,%r9,8),%xmm8 ## G2 H2 mulpd %xmm3,%xmm2 ## vvdw6=c6*rinv6 mulpd %xmm7,%xmm12 ## vvdw12=c12*rinv12 movapd %xmm12,%xmm0 subpd %xmm2,%xmm12 ## Vvdw=Vvdw12-Vvdw6 ## add potential to vvdwtot addpd nb310_Vvdwtot(%rsp),%xmm12 movapd %xmm12,nb310_Vvdwtot(%rsp) movapd %xmm6,%xmm7 unpcklpd %xmm8,%xmm6 ## G1 G2 unpckhpd %xmm8,%xmm7 ## H1 H2 ## coulomb table ready, in xmm4-xmm7 movapd nb310_eps(%rsp),%xmm3 mulpd %xmm3,%xmm7 ## Heps mulpd %xmm3,%xmm6 ## Geps mulpd %xmm3,%xmm7 ## Heps2 addpd %xmm6,%xmm5 ## F+Geps addpd %xmm7,%xmm5 ## F+Geps+Heps2 = Fp addpd %xmm7,%xmm7 ## 2*Heps2 addpd %xmm6,%xmm7 ## 2*Heps2+Geps addpd %xmm5,%xmm7 ## FF = Fp + 2*Heps2 + Geps mulpd %xmm3,%xmm5 ## eps*Fp addpd %xmm4,%xmm5 ## VV mulpd nb310_qq(%rsp),%xmm5 ## VV*qq=vcoul mulpd nb310_qq(%rsp),%xmm7 ## FF*qq=fijC ## the fj's - start by accumulating forces from memory movq nb310_faction(%rbp),%rdi movlpd (%rdi,%rax,8),%xmm3 movlpd 8(%rdi,%rax,8),%xmm4 movlpd 16(%rdi,%rax,8),%xmm6 movhpd (%rdi,%rbx,8),%xmm3 movhpd 8(%rdi,%rbx,8),%xmm4 movhpd 16(%rdi,%rbx,8),%xmm6 ## LJ forces mulpd nb310_six(%rsp),%xmm2 mulpd nb310_twelve(%rsp),%xmm0 subpd %xmm2,%xmm0 mulpd %xmm1,%xmm0 ## (12*vnb12-6*vnb6)*rinv ## add potential to vctot addpd nb310_vctot(%rsp),%xmm5 movapd %xmm5,nb310_vctot(%rsp) mulpd nb310_tsc(%rsp),%xmm7 subpd %xmm7,%xmm0 mulpd %xmm1,%xmm0 ## fscal ## calculate scalar force by multiplying dx/dy/dz with fscal mulpd %xmm0,%xmm9 mulpd %xmm0,%xmm10 mulpd %xmm0,%xmm11 addpd %xmm9,%xmm3 addpd %xmm10,%xmm4 addpd %xmm11,%xmm6 ## now update f_i ⌨️ 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?