File: | gromacs/gmxlib/nonbonded/nb_kernel_sse4_1_single/nb_kernel_ElecEw_VdwCSTab_GeomW4P1_sse4_1_single.c |
Location: | line 845, column 22 |
Description: | Value stored to 'one_twelfth' during its initialization is never read |
1 | /* |
2 | * This file is part of the GROMACS molecular simulation package. |
3 | * |
4 | * Copyright (c) 2012,2013,2014, by the GROMACS development team, led by |
5 | * Mark Abraham, David van der Spoel, Berk Hess, and Erik Lindahl, |
6 | * and including many others, as listed in the AUTHORS file in the |
7 | * top-level source directory and at http://www.gromacs.org. |
8 | * |
9 | * GROMACS is free software; you can redistribute it and/or |
10 | * modify it under the terms of the GNU Lesser General Public License |
11 | * as published by the Free Software Foundation; either version 2.1 |
12 | * of the License, or (at your option) any later version. |
13 | * |
14 | * GROMACS is distributed in the hope that it will be useful, |
15 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
16 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
17 | * Lesser General Public License for more details. |
18 | * |
19 | * You should have received a copy of the GNU Lesser General Public |
20 | * License along with GROMACS; if not, see |
21 | * http://www.gnu.org/licenses, or write to the Free Software Foundation, |
22 | * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. |
23 | * |
24 | * If you want to redistribute modifications to GROMACS, please |
25 | * consider that scientific software is very special. Version |
26 | * control is crucial - bugs must be traceable. We will be happy to |
27 | * consider code for inclusion in the official distribution, but |
28 | * derived work must not be called official GROMACS. Details are found |
29 | * in the README & COPYING files - if they are missing, get the |
30 | * official version at http://www.gromacs.org. |
31 | * |
32 | * To help us fund GROMACS development, we humbly ask that you cite |
33 | * the research papers on the package. Check out http://www.gromacs.org. |
34 | */ |
35 | /* |
36 | * Note: this file was generated by the GROMACS sse4_1_single kernel generator. |
37 | */ |
38 | #ifdef HAVE_CONFIG_H1 |
39 | #include <config.h> |
40 | #endif |
41 | |
42 | #include <math.h> |
43 | |
44 | #include "../nb_kernel.h" |
45 | #include "types/simple.h" |
46 | #include "gromacs/math/vec.h" |
47 | #include "nrnb.h" |
48 | |
49 | #include "gromacs/simd/math_x86_sse4_1_single.h" |
50 | #include "kernelutil_x86_sse4_1_single.h" |
51 | |
52 | /* |
53 | * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomW4P1_VF_sse4_1_single |
54 | * Electrostatics interaction: Ewald |
55 | * VdW interaction: CubicSplineTable |
56 | * Geometry: Water4-Particle |
57 | * Calculate force/pot: PotentialAndForce |
58 | */ |
59 | void |
60 | nb_kernel_ElecEw_VdwCSTab_GeomW4P1_VF_sse4_1_single |
61 | (t_nblist * gmx_restrict nlist, |
62 | rvec * gmx_restrict xx, |
63 | rvec * gmx_restrict ff, |
64 | t_forcerec * gmx_restrict fr, |
65 | t_mdatoms * gmx_restrict mdatoms, |
66 | nb_kernel_data_t gmx_unused__attribute__ ((unused)) * gmx_restrict kernel_data, |
67 | t_nrnb * gmx_restrict nrnb) |
68 | { |
69 | /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or |
70 | * just 0 for non-waters. |
71 | * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different |
72 | * jnr indices corresponding to data put in the four positions in the SIMD register. |
73 | */ |
74 | int i_shift_offset,i_coord_offset,outeriter,inneriter; |
75 | int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx; |
76 | int jnrA,jnrB,jnrC,jnrD; |
77 | int jnrlistA,jnrlistB,jnrlistC,jnrlistD; |
78 | int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD; |
79 | int *iinr,*jindex,*jjnr,*shiftidx,*gid; |
80 | real rcutoff_scalar; |
81 | real *shiftvec,*fshift,*x,*f; |
82 | real *fjptrA,*fjptrB,*fjptrC,*fjptrD; |
83 | real scratch[4*DIM3]; |
84 | __m128 tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall; |
85 | int vdwioffset0; |
86 | __m128 ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0; |
87 | int vdwioffset1; |
88 | __m128 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1; |
89 | int vdwioffset2; |
90 | __m128 ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2; |
91 | int vdwioffset3; |
92 | __m128 ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3; |
93 | int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D; |
94 | __m128 jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0; |
95 | __m128 dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00; |
96 | __m128 dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10; |
97 | __m128 dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20; |
98 | __m128 dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30; |
99 | __m128 velec,felec,velecsum,facel,crf,krf,krf2; |
100 | real *charge; |
101 | int nvdwtype; |
102 | __m128 rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6; |
103 | int *vdwtype; |
104 | real *vdwparam; |
105 | __m128 one_sixth = _mm_set1_ps(1.0/6.0); |
106 | __m128 one_twelfth = _mm_set1_ps(1.0/12.0); |
107 | __m128i vfitab; |
108 | __m128i ifour = _mm_set1_epi32(4); |
109 | __m128 rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF; |
110 | real *vftab; |
111 | __m128i ewitab; |
112 | __m128 ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV; |
113 | real *ewtab; |
114 | __m128 dummy_mask,cutoff_mask; |
115 | __m128 signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) ); |
116 | __m128 one = _mm_set1_ps(1.0); |
117 | __m128 two = _mm_set1_ps(2.0); |
118 | x = xx[0]; |
119 | f = ff[0]; |
120 | |
121 | nri = nlist->nri; |
122 | iinr = nlist->iinr; |
123 | jindex = nlist->jindex; |
124 | jjnr = nlist->jjnr; |
125 | shiftidx = nlist->shift; |
126 | gid = nlist->gid; |
127 | shiftvec = fr->shift_vec[0]; |
128 | fshift = fr->fshift[0]; |
129 | facel = _mm_set1_ps(fr->epsfac); |
130 | charge = mdatoms->chargeA; |
131 | nvdwtype = fr->ntype; |
132 | vdwparam = fr->nbfp; |
133 | vdwtype = mdatoms->typeA; |
134 | |
135 | vftab = kernel_data->table_vdw->data; |
136 | vftabscale = _mm_set1_ps(kernel_data->table_vdw->scale); |
137 | |
138 | sh_ewald = _mm_set1_ps(fr->ic->sh_ewald); |
139 | ewtab = fr->ic->tabq_coul_FDV0; |
140 | ewtabscale = _mm_set1_ps(fr->ic->tabq_scale); |
141 | ewtabhalfspace = _mm_set1_ps(0.5/fr->ic->tabq_scale); |
142 | |
143 | /* Setup water-specific parameters */ |
144 | inr = nlist->iinr[0]; |
145 | iq1 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1])); |
146 | iq2 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2])); |
147 | iq3 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3])); |
148 | vdwioffset0 = 2*nvdwtype*vdwtype[inr+0]; |
149 | |
150 | /* Avoid stupid compiler warnings */ |
151 | jnrA = jnrB = jnrC = jnrD = 0; |
152 | j_coord_offsetA = 0; |
153 | j_coord_offsetB = 0; |
154 | j_coord_offsetC = 0; |
155 | j_coord_offsetD = 0; |
156 | |
157 | outeriter = 0; |
158 | inneriter = 0; |
159 | |
160 | for(iidx=0;iidx<4*DIM3;iidx++) |
161 | { |
162 | scratch[iidx] = 0.0; |
163 | } |
164 | |
165 | /* Start outer loop over neighborlists */ |
166 | for(iidx=0; iidx<nri; iidx++) |
167 | { |
168 | /* Load shift vector for this list */ |
169 | i_shift_offset = DIM3*shiftidx[iidx]; |
170 | |
171 | /* Load limits for loop over neighbors */ |
172 | j_index_start = jindex[iidx]; |
173 | j_index_end = jindex[iidx+1]; |
174 | |
175 | /* Get outer coordinate index */ |
176 | inr = iinr[iidx]; |
177 | i_coord_offset = DIM3*inr; |
178 | |
179 | /* Load i particle coords and add shift vector */ |
180 | gmx_mm_load_shift_and_4rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset, |
181 | &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3); |
182 | |
183 | fix0 = _mm_setzero_ps(); |
184 | fiy0 = _mm_setzero_ps(); |
185 | fiz0 = _mm_setzero_ps(); |
186 | fix1 = _mm_setzero_ps(); |
187 | fiy1 = _mm_setzero_ps(); |
188 | fiz1 = _mm_setzero_ps(); |
189 | fix2 = _mm_setzero_ps(); |
190 | fiy2 = _mm_setzero_ps(); |
191 | fiz2 = _mm_setzero_ps(); |
192 | fix3 = _mm_setzero_ps(); |
193 | fiy3 = _mm_setzero_ps(); |
194 | fiz3 = _mm_setzero_ps(); |
195 | |
196 | /* Reset potential sums */ |
197 | velecsum = _mm_setzero_ps(); |
198 | vvdwsum = _mm_setzero_ps(); |
199 | |
200 | /* Start inner kernel loop */ |
201 | for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4) |
202 | { |
203 | |
204 | /* Get j neighbor index, and coordinate index */ |
205 | jnrA = jjnr[jidx]; |
206 | jnrB = jjnr[jidx+1]; |
207 | jnrC = jjnr[jidx+2]; |
208 | jnrD = jjnr[jidx+3]; |
209 | j_coord_offsetA = DIM3*jnrA; |
210 | j_coord_offsetB = DIM3*jnrB; |
211 | j_coord_offsetC = DIM3*jnrC; |
212 | j_coord_offsetD = DIM3*jnrD; |
213 | |
214 | /* load j atom coordinates */ |
215 | gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB, |
216 | x+j_coord_offsetC,x+j_coord_offsetD, |
217 | &jx0,&jy0,&jz0); |
218 | |
219 | /* Calculate displacement vector */ |
220 | dx00 = _mm_sub_ps(ix0,jx0); |
221 | dy00 = _mm_sub_ps(iy0,jy0); |
222 | dz00 = _mm_sub_ps(iz0,jz0); |
223 | dx10 = _mm_sub_ps(ix1,jx0); |
224 | dy10 = _mm_sub_ps(iy1,jy0); |
225 | dz10 = _mm_sub_ps(iz1,jz0); |
226 | dx20 = _mm_sub_ps(ix2,jx0); |
227 | dy20 = _mm_sub_ps(iy2,jy0); |
228 | dz20 = _mm_sub_ps(iz2,jz0); |
229 | dx30 = _mm_sub_ps(ix3,jx0); |
230 | dy30 = _mm_sub_ps(iy3,jy0); |
231 | dz30 = _mm_sub_ps(iz3,jz0); |
232 | |
233 | /* Calculate squared distance and things based on it */ |
234 | rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00); |
235 | rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10); |
236 | rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20); |
237 | rsq30 = gmx_mm_calc_rsq_ps(dx30,dy30,dz30); |
238 | |
239 | rinv00 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq00); |
240 | rinv10 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq10); |
241 | rinv20 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq20); |
242 | rinv30 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq30); |
243 | |
244 | rinvsq10 = _mm_mul_ps(rinv10,rinv10); |
245 | rinvsq20 = _mm_mul_ps(rinv20,rinv20); |
246 | rinvsq30 = _mm_mul_ps(rinv30,rinv30); |
247 | |
248 | /* Load parameters for j particles */ |
249 | jq0 = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0, |
250 | charge+jnrC+0,charge+jnrD+0); |
251 | vdwjidx0A = 2*vdwtype[jnrA+0]; |
252 | vdwjidx0B = 2*vdwtype[jnrB+0]; |
253 | vdwjidx0C = 2*vdwtype[jnrC+0]; |
254 | vdwjidx0D = 2*vdwtype[jnrD+0]; |
255 | |
256 | fjx0 = _mm_setzero_ps(); |
257 | fjy0 = _mm_setzero_ps(); |
258 | fjz0 = _mm_setzero_ps(); |
259 | |
260 | /************************** |
261 | * CALCULATE INTERACTIONS * |
262 | **************************/ |
263 | |
264 | r00 = _mm_mul_ps(rsq00,rinv00); |
265 | |
266 | /* Compute parameters for interactions between i and j atoms */ |
267 | gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A, |
268 | vdwparam+vdwioffset0+vdwjidx0B, |
269 | vdwparam+vdwioffset0+vdwjidx0C, |
270 | vdwparam+vdwioffset0+vdwjidx0D, |
271 | &c6_00,&c12_00); |
272 | |
273 | /* Calculate table index by multiplying r with table scale and truncate to integer */ |
274 | rt = _mm_mul_ps(r00,vftabscale); |
275 | vfitab = _mm_cvttps_epi32(rt); |
276 | vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (rt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
277 | vfitab = _mm_slli_epi32(vfitab,3); |
278 | |
279 | /* CUBIC SPLINE TABLE DISPERSION */ |
280 | Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(0) & 3];})) ); |
281 | F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(1) & 3];})) ); |
282 | G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(2) & 3];})) ); |
283 | H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(3) & 3];})) ); |
284 | _MM_TRANSPOSE4_PS(Y,F,G,H)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((Y ), (F)); tmp2 = _mm_unpacklo_ps((G), (H)); tmp1 = _mm_unpackhi_ps ((Y), (F)); tmp3 = _mm_unpackhi_ps((G), (H)); (Y) = _mm_movelh_ps (tmp0, tmp2); (F) = _mm_movehl_ps(tmp2, tmp0); (G) = _mm_movelh_ps (tmp1, tmp3); (H) = _mm_movehl_ps(tmp3, tmp1); } while (0); |
285 | Heps = _mm_mul_ps(vfeps,H); |
286 | Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps))); |
287 | VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp)); |
288 | vvdw6 = _mm_mul_ps(c6_00,VV); |
289 | FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps)))); |
290 | fvdw6 = _mm_mul_ps(c6_00,FF); |
291 | |
292 | /* CUBIC SPLINE TABLE REPULSION */ |
293 | vfitab = _mm_add_epi32(vfitab,ifour); |
294 | Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(0) & 3];})) ); |
295 | F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(1) & 3];})) ); |
296 | G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(2) & 3];})) ); |
297 | H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(3) & 3];})) ); |
298 | _MM_TRANSPOSE4_PS(Y,F,G,H)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((Y ), (F)); tmp2 = _mm_unpacklo_ps((G), (H)); tmp1 = _mm_unpackhi_ps ((Y), (F)); tmp3 = _mm_unpackhi_ps((G), (H)); (Y) = _mm_movelh_ps (tmp0, tmp2); (F) = _mm_movehl_ps(tmp2, tmp0); (G) = _mm_movelh_ps (tmp1, tmp3); (H) = _mm_movehl_ps(tmp3, tmp1); } while (0); |
299 | Heps = _mm_mul_ps(vfeps,H); |
300 | Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps))); |
301 | VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp)); |
302 | vvdw12 = _mm_mul_ps(c12_00,VV); |
303 | FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps)))); |
304 | fvdw12 = _mm_mul_ps(c12_00,FF); |
305 | vvdw = _mm_add_ps(vvdw12,vvdw6); |
306 | fvdw = _mm_xor_ps(signbit,_mm_mul_ps(_mm_add_ps(fvdw6,fvdw12),_mm_mul_ps(vftabscale,rinv00))); |
307 | |
308 | /* Update potential sum for this i atom from the interaction with this j atom. */ |
309 | vvdwsum = _mm_add_ps(vvdwsum,vvdw); |
310 | |
311 | fscal = fvdw; |
312 | |
313 | /* Calculate temporary vectorial force */ |
314 | tx = _mm_mul_ps(fscal,dx00); |
315 | ty = _mm_mul_ps(fscal,dy00); |
316 | tz = _mm_mul_ps(fscal,dz00); |
317 | |
318 | /* Update vectorial force */ |
319 | fix0 = _mm_add_ps(fix0,tx); |
320 | fiy0 = _mm_add_ps(fiy0,ty); |
321 | fiz0 = _mm_add_ps(fiz0,tz); |
322 | |
323 | fjx0 = _mm_add_ps(fjx0,tx); |
324 | fjy0 = _mm_add_ps(fjy0,ty); |
325 | fjz0 = _mm_add_ps(fjz0,tz); |
326 | |
327 | /************************** |
328 | * CALCULATE INTERACTIONS * |
329 | **************************/ |
330 | |
331 | r10 = _mm_mul_ps(rsq10,rinv10); |
332 | |
333 | /* Compute parameters for interactions between i and j atoms */ |
334 | qq10 = _mm_mul_ps(iq1,jq0); |
335 | |
336 | /* EWALD ELECTROSTATICS */ |
337 | |
338 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
339 | ewrt = _mm_mul_ps(r10,ewtabscale); |
340 | ewitab = _mm_cvttps_epi32(ewrt); |
341 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
342 | ewitab = _mm_slli_epi32(ewitab,2); |
343 | ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})) ); |
344 | ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})) ); |
345 | ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})) ); |
346 | ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})) ); |
347 | _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((ewtabF ), (ewtabD)); tmp2 = _mm_unpacklo_ps((ewtabV), (ewtabFn)); tmp1 = _mm_unpackhi_ps((ewtabF), (ewtabD)); tmp3 = _mm_unpackhi_ps ((ewtabV), (ewtabFn)); (ewtabF) = _mm_movelh_ps(tmp0, tmp2); ( ewtabD) = _mm_movehl_ps(tmp2, tmp0); (ewtabV) = _mm_movelh_ps (tmp1, tmp3); (ewtabFn) = _mm_movehl_ps(tmp3, tmp1); } while ( 0); |
348 | felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD)); |
349 | velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec))); |
350 | velec = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec)); |
351 | felec = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec)); |
352 | |
353 | /* Update potential sum for this i atom from the interaction with this j atom. */ |
354 | velecsum = _mm_add_ps(velecsum,velec); |
355 | |
356 | fscal = felec; |
357 | |
358 | /* Calculate temporary vectorial force */ |
359 | tx = _mm_mul_ps(fscal,dx10); |
360 | ty = _mm_mul_ps(fscal,dy10); |
361 | tz = _mm_mul_ps(fscal,dz10); |
362 | |
363 | /* Update vectorial force */ |
364 | fix1 = _mm_add_ps(fix1,tx); |
365 | fiy1 = _mm_add_ps(fiy1,ty); |
366 | fiz1 = _mm_add_ps(fiz1,tz); |
367 | |
368 | fjx0 = _mm_add_ps(fjx0,tx); |
369 | fjy0 = _mm_add_ps(fjy0,ty); |
370 | fjz0 = _mm_add_ps(fjz0,tz); |
371 | |
372 | /************************** |
373 | * CALCULATE INTERACTIONS * |
374 | **************************/ |
375 | |
376 | r20 = _mm_mul_ps(rsq20,rinv20); |
377 | |
378 | /* Compute parameters for interactions between i and j atoms */ |
379 | qq20 = _mm_mul_ps(iq2,jq0); |
380 | |
381 | /* EWALD ELECTROSTATICS */ |
382 | |
383 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
384 | ewrt = _mm_mul_ps(r20,ewtabscale); |
385 | ewitab = _mm_cvttps_epi32(ewrt); |
386 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
387 | ewitab = _mm_slli_epi32(ewitab,2); |
388 | ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})) ); |
389 | ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})) ); |
390 | ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})) ); |
391 | ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})) ); |
392 | _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((ewtabF ), (ewtabD)); tmp2 = _mm_unpacklo_ps((ewtabV), (ewtabFn)); tmp1 = _mm_unpackhi_ps((ewtabF), (ewtabD)); tmp3 = _mm_unpackhi_ps ((ewtabV), (ewtabFn)); (ewtabF) = _mm_movelh_ps(tmp0, tmp2); ( ewtabD) = _mm_movehl_ps(tmp2, tmp0); (ewtabV) = _mm_movelh_ps (tmp1, tmp3); (ewtabFn) = _mm_movehl_ps(tmp3, tmp1); } while ( 0); |
393 | felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD)); |
394 | velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec))); |
395 | velec = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec)); |
396 | felec = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec)); |
397 | |
398 | /* Update potential sum for this i atom from the interaction with this j atom. */ |
399 | velecsum = _mm_add_ps(velecsum,velec); |
400 | |
401 | fscal = felec; |
402 | |
403 | /* Calculate temporary vectorial force */ |
404 | tx = _mm_mul_ps(fscal,dx20); |
405 | ty = _mm_mul_ps(fscal,dy20); |
406 | tz = _mm_mul_ps(fscal,dz20); |
407 | |
408 | /* Update vectorial force */ |
409 | fix2 = _mm_add_ps(fix2,tx); |
410 | fiy2 = _mm_add_ps(fiy2,ty); |
411 | fiz2 = _mm_add_ps(fiz2,tz); |
412 | |
413 | fjx0 = _mm_add_ps(fjx0,tx); |
414 | fjy0 = _mm_add_ps(fjy0,ty); |
415 | fjz0 = _mm_add_ps(fjz0,tz); |
416 | |
417 | /************************** |
418 | * CALCULATE INTERACTIONS * |
419 | **************************/ |
420 | |
421 | r30 = _mm_mul_ps(rsq30,rinv30); |
422 | |
423 | /* Compute parameters for interactions between i and j atoms */ |
424 | qq30 = _mm_mul_ps(iq3,jq0); |
425 | |
426 | /* EWALD ELECTROSTATICS */ |
427 | |
428 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
429 | ewrt = _mm_mul_ps(r30,ewtabscale); |
430 | ewitab = _mm_cvttps_epi32(ewrt); |
431 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
432 | ewitab = _mm_slli_epi32(ewitab,2); |
433 | ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})) ); |
434 | ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})) ); |
435 | ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})) ); |
436 | ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})) ); |
437 | _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((ewtabF ), (ewtabD)); tmp2 = _mm_unpacklo_ps((ewtabV), (ewtabFn)); tmp1 = _mm_unpackhi_ps((ewtabF), (ewtabD)); tmp3 = _mm_unpackhi_ps ((ewtabV), (ewtabFn)); (ewtabF) = _mm_movelh_ps(tmp0, tmp2); ( ewtabD) = _mm_movehl_ps(tmp2, tmp0); (ewtabV) = _mm_movelh_ps (tmp1, tmp3); (ewtabFn) = _mm_movehl_ps(tmp3, tmp1); } while ( 0); |
438 | felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD)); |
439 | velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec))); |
440 | velec = _mm_mul_ps(qq30,_mm_sub_ps(rinv30,velec)); |
441 | felec = _mm_mul_ps(_mm_mul_ps(qq30,rinv30),_mm_sub_ps(rinvsq30,felec)); |
442 | |
443 | /* Update potential sum for this i atom from the interaction with this j atom. */ |
444 | velecsum = _mm_add_ps(velecsum,velec); |
445 | |
446 | fscal = felec; |
447 | |
448 | /* Calculate temporary vectorial force */ |
449 | tx = _mm_mul_ps(fscal,dx30); |
450 | ty = _mm_mul_ps(fscal,dy30); |
451 | tz = _mm_mul_ps(fscal,dz30); |
452 | |
453 | /* Update vectorial force */ |
454 | fix3 = _mm_add_ps(fix3,tx); |
455 | fiy3 = _mm_add_ps(fiy3,ty); |
456 | fiz3 = _mm_add_ps(fiz3,tz); |
457 | |
458 | fjx0 = _mm_add_ps(fjx0,tx); |
459 | fjy0 = _mm_add_ps(fjy0,ty); |
460 | fjz0 = _mm_add_ps(fjz0,tz); |
461 | |
462 | fjptrA = f+j_coord_offsetA; |
463 | fjptrB = f+j_coord_offsetB; |
464 | fjptrC = f+j_coord_offsetC; |
465 | fjptrD = f+j_coord_offsetD; |
466 | |
467 | gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0); |
468 | |
469 | /* Inner loop uses 179 flops */ |
470 | } |
471 | |
472 | if(jidx<j_index_end) |
473 | { |
474 | |
475 | /* Get j neighbor index, and coordinate index */ |
476 | jnrlistA = jjnr[jidx]; |
477 | jnrlistB = jjnr[jidx+1]; |
478 | jnrlistC = jjnr[jidx+2]; |
479 | jnrlistD = jjnr[jidx+3]; |
480 | /* Sign of each element will be negative for non-real atoms. |
481 | * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones, |
482 | * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries. |
483 | */ |
484 | dummy_mask = gmx_mm_castsi128_ps_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128())); |
485 | jnrA = (jnrlistA>=0) ? jnrlistA : 0; |
486 | jnrB = (jnrlistB>=0) ? jnrlistB : 0; |
487 | jnrC = (jnrlistC>=0) ? jnrlistC : 0; |
488 | jnrD = (jnrlistD>=0) ? jnrlistD : 0; |
489 | j_coord_offsetA = DIM3*jnrA; |
490 | j_coord_offsetB = DIM3*jnrB; |
491 | j_coord_offsetC = DIM3*jnrC; |
492 | j_coord_offsetD = DIM3*jnrD; |
493 | |
494 | /* load j atom coordinates */ |
495 | gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB, |
496 | x+j_coord_offsetC,x+j_coord_offsetD, |
497 | &jx0,&jy0,&jz0); |
498 | |
499 | /* Calculate displacement vector */ |
500 | dx00 = _mm_sub_ps(ix0,jx0); |
501 | dy00 = _mm_sub_ps(iy0,jy0); |
502 | dz00 = _mm_sub_ps(iz0,jz0); |
503 | dx10 = _mm_sub_ps(ix1,jx0); |
504 | dy10 = _mm_sub_ps(iy1,jy0); |
505 | dz10 = _mm_sub_ps(iz1,jz0); |
506 | dx20 = _mm_sub_ps(ix2,jx0); |
507 | dy20 = _mm_sub_ps(iy2,jy0); |
508 | dz20 = _mm_sub_ps(iz2,jz0); |
509 | dx30 = _mm_sub_ps(ix3,jx0); |
510 | dy30 = _mm_sub_ps(iy3,jy0); |
511 | dz30 = _mm_sub_ps(iz3,jz0); |
512 | |
513 | /* Calculate squared distance and things based on it */ |
514 | rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00); |
515 | rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10); |
516 | rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20); |
517 | rsq30 = gmx_mm_calc_rsq_ps(dx30,dy30,dz30); |
518 | |
519 | rinv00 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq00); |
520 | rinv10 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq10); |
521 | rinv20 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq20); |
522 | rinv30 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq30); |
523 | |
524 | rinvsq10 = _mm_mul_ps(rinv10,rinv10); |
525 | rinvsq20 = _mm_mul_ps(rinv20,rinv20); |
526 | rinvsq30 = _mm_mul_ps(rinv30,rinv30); |
527 | |
528 | /* Load parameters for j particles */ |
529 | jq0 = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0, |
530 | charge+jnrC+0,charge+jnrD+0); |
531 | vdwjidx0A = 2*vdwtype[jnrA+0]; |
532 | vdwjidx0B = 2*vdwtype[jnrB+0]; |
533 | vdwjidx0C = 2*vdwtype[jnrC+0]; |
534 | vdwjidx0D = 2*vdwtype[jnrD+0]; |
535 | |
536 | fjx0 = _mm_setzero_ps(); |
537 | fjy0 = _mm_setzero_ps(); |
538 | fjz0 = _mm_setzero_ps(); |
539 | |
540 | /************************** |
541 | * CALCULATE INTERACTIONS * |
542 | **************************/ |
543 | |
544 | r00 = _mm_mul_ps(rsq00,rinv00); |
545 | r00 = _mm_andnot_ps(dummy_mask,r00); |
546 | |
547 | /* Compute parameters for interactions between i and j atoms */ |
548 | gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A, |
549 | vdwparam+vdwioffset0+vdwjidx0B, |
550 | vdwparam+vdwioffset0+vdwjidx0C, |
551 | vdwparam+vdwioffset0+vdwjidx0D, |
552 | &c6_00,&c12_00); |
553 | |
554 | /* Calculate table index by multiplying r with table scale and truncate to integer */ |
555 | rt = _mm_mul_ps(r00,vftabscale); |
556 | vfitab = _mm_cvttps_epi32(rt); |
557 | vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (rt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
558 | vfitab = _mm_slli_epi32(vfitab,3); |
559 | |
560 | /* CUBIC SPLINE TABLE DISPERSION */ |
561 | Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(0) & 3];})) ); |
562 | F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(1) & 3];})) ); |
563 | G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(2) & 3];})) ); |
564 | H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(3) & 3];})) ); |
565 | _MM_TRANSPOSE4_PS(Y,F,G,H)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((Y ), (F)); tmp2 = _mm_unpacklo_ps((G), (H)); tmp1 = _mm_unpackhi_ps ((Y), (F)); tmp3 = _mm_unpackhi_ps((G), (H)); (Y) = _mm_movelh_ps (tmp0, tmp2); (F) = _mm_movehl_ps(tmp2, tmp0); (G) = _mm_movelh_ps (tmp1, tmp3); (H) = _mm_movehl_ps(tmp3, tmp1); } while (0); |
566 | Heps = _mm_mul_ps(vfeps,H); |
567 | Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps))); |
568 | VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp)); |
569 | vvdw6 = _mm_mul_ps(c6_00,VV); |
570 | FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps)))); |
571 | fvdw6 = _mm_mul_ps(c6_00,FF); |
572 | |
573 | /* CUBIC SPLINE TABLE REPULSION */ |
574 | vfitab = _mm_add_epi32(vfitab,ifour); |
575 | Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(0) & 3];})) ); |
576 | F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(1) & 3];})) ); |
577 | G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(2) & 3];})) ); |
578 | H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(3) & 3];})) ); |
579 | _MM_TRANSPOSE4_PS(Y,F,G,H)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((Y ), (F)); tmp2 = _mm_unpacklo_ps((G), (H)); tmp1 = _mm_unpackhi_ps ((Y), (F)); tmp3 = _mm_unpackhi_ps((G), (H)); (Y) = _mm_movelh_ps (tmp0, tmp2); (F) = _mm_movehl_ps(tmp2, tmp0); (G) = _mm_movelh_ps (tmp1, tmp3); (H) = _mm_movehl_ps(tmp3, tmp1); } while (0); |
580 | Heps = _mm_mul_ps(vfeps,H); |
581 | Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps))); |
582 | VV = _mm_add_ps(Y,_mm_mul_ps(vfeps,Fp)); |
583 | vvdw12 = _mm_mul_ps(c12_00,VV); |
584 | FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps)))); |
585 | fvdw12 = _mm_mul_ps(c12_00,FF); |
586 | vvdw = _mm_add_ps(vvdw12,vvdw6); |
587 | fvdw = _mm_xor_ps(signbit,_mm_mul_ps(_mm_add_ps(fvdw6,fvdw12),_mm_mul_ps(vftabscale,rinv00))); |
588 | |
589 | /* Update potential sum for this i atom from the interaction with this j atom. */ |
590 | vvdw = _mm_andnot_ps(dummy_mask,vvdw); |
591 | vvdwsum = _mm_add_ps(vvdwsum,vvdw); |
592 | |
593 | fscal = fvdw; |
594 | |
595 | fscal = _mm_andnot_ps(dummy_mask,fscal); |
596 | |
597 | /* Calculate temporary vectorial force */ |
598 | tx = _mm_mul_ps(fscal,dx00); |
599 | ty = _mm_mul_ps(fscal,dy00); |
600 | tz = _mm_mul_ps(fscal,dz00); |
601 | |
602 | /* Update vectorial force */ |
603 | fix0 = _mm_add_ps(fix0,tx); |
604 | fiy0 = _mm_add_ps(fiy0,ty); |
605 | fiz0 = _mm_add_ps(fiz0,tz); |
606 | |
607 | fjx0 = _mm_add_ps(fjx0,tx); |
608 | fjy0 = _mm_add_ps(fjy0,ty); |
609 | fjz0 = _mm_add_ps(fjz0,tz); |
610 | |
611 | /************************** |
612 | * CALCULATE INTERACTIONS * |
613 | **************************/ |
614 | |
615 | r10 = _mm_mul_ps(rsq10,rinv10); |
616 | r10 = _mm_andnot_ps(dummy_mask,r10); |
617 | |
618 | /* Compute parameters for interactions between i and j atoms */ |
619 | qq10 = _mm_mul_ps(iq1,jq0); |
620 | |
621 | /* EWALD ELECTROSTATICS */ |
622 | |
623 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
624 | ewrt = _mm_mul_ps(r10,ewtabscale); |
625 | ewitab = _mm_cvttps_epi32(ewrt); |
626 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
627 | ewitab = _mm_slli_epi32(ewitab,2); |
628 | ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})) ); |
629 | ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})) ); |
630 | ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})) ); |
631 | ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})) ); |
632 | _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((ewtabF ), (ewtabD)); tmp2 = _mm_unpacklo_ps((ewtabV), (ewtabFn)); tmp1 = _mm_unpackhi_ps((ewtabF), (ewtabD)); tmp3 = _mm_unpackhi_ps ((ewtabV), (ewtabFn)); (ewtabF) = _mm_movelh_ps(tmp0, tmp2); ( ewtabD) = _mm_movehl_ps(tmp2, tmp0); (ewtabV) = _mm_movelh_ps (tmp1, tmp3); (ewtabFn) = _mm_movehl_ps(tmp3, tmp1); } while ( 0); |
633 | felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD)); |
634 | velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec))); |
635 | velec = _mm_mul_ps(qq10,_mm_sub_ps(rinv10,velec)); |
636 | felec = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec)); |
637 | |
638 | /* Update potential sum for this i atom from the interaction with this j atom. */ |
639 | velec = _mm_andnot_ps(dummy_mask,velec); |
640 | velecsum = _mm_add_ps(velecsum,velec); |
641 | |
642 | fscal = felec; |
643 | |
644 | fscal = _mm_andnot_ps(dummy_mask,fscal); |
645 | |
646 | /* Calculate temporary vectorial force */ |
647 | tx = _mm_mul_ps(fscal,dx10); |
648 | ty = _mm_mul_ps(fscal,dy10); |
649 | tz = _mm_mul_ps(fscal,dz10); |
650 | |
651 | /* Update vectorial force */ |
652 | fix1 = _mm_add_ps(fix1,tx); |
653 | fiy1 = _mm_add_ps(fiy1,ty); |
654 | fiz1 = _mm_add_ps(fiz1,tz); |
655 | |
656 | fjx0 = _mm_add_ps(fjx0,tx); |
657 | fjy0 = _mm_add_ps(fjy0,ty); |
658 | fjz0 = _mm_add_ps(fjz0,tz); |
659 | |
660 | /************************** |
661 | * CALCULATE INTERACTIONS * |
662 | **************************/ |
663 | |
664 | r20 = _mm_mul_ps(rsq20,rinv20); |
665 | r20 = _mm_andnot_ps(dummy_mask,r20); |
666 | |
667 | /* Compute parameters for interactions between i and j atoms */ |
668 | qq20 = _mm_mul_ps(iq2,jq0); |
669 | |
670 | /* EWALD ELECTROSTATICS */ |
671 | |
672 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
673 | ewrt = _mm_mul_ps(r20,ewtabscale); |
674 | ewitab = _mm_cvttps_epi32(ewrt); |
675 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
676 | ewitab = _mm_slli_epi32(ewitab,2); |
677 | ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})) ); |
678 | ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})) ); |
679 | ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})) ); |
680 | ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})) ); |
681 | _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((ewtabF ), (ewtabD)); tmp2 = _mm_unpacklo_ps((ewtabV), (ewtabFn)); tmp1 = _mm_unpackhi_ps((ewtabF), (ewtabD)); tmp3 = _mm_unpackhi_ps ((ewtabV), (ewtabFn)); (ewtabF) = _mm_movelh_ps(tmp0, tmp2); ( ewtabD) = _mm_movehl_ps(tmp2, tmp0); (ewtabV) = _mm_movelh_ps (tmp1, tmp3); (ewtabFn) = _mm_movehl_ps(tmp3, tmp1); } while ( 0); |
682 | felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD)); |
683 | velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec))); |
684 | velec = _mm_mul_ps(qq20,_mm_sub_ps(rinv20,velec)); |
685 | felec = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec)); |
686 | |
687 | /* Update potential sum for this i atom from the interaction with this j atom. */ |
688 | velec = _mm_andnot_ps(dummy_mask,velec); |
689 | velecsum = _mm_add_ps(velecsum,velec); |
690 | |
691 | fscal = felec; |
692 | |
693 | fscal = _mm_andnot_ps(dummy_mask,fscal); |
694 | |
695 | /* Calculate temporary vectorial force */ |
696 | tx = _mm_mul_ps(fscal,dx20); |
697 | ty = _mm_mul_ps(fscal,dy20); |
698 | tz = _mm_mul_ps(fscal,dz20); |
699 | |
700 | /* Update vectorial force */ |
701 | fix2 = _mm_add_ps(fix2,tx); |
702 | fiy2 = _mm_add_ps(fiy2,ty); |
703 | fiz2 = _mm_add_ps(fiz2,tz); |
704 | |
705 | fjx0 = _mm_add_ps(fjx0,tx); |
706 | fjy0 = _mm_add_ps(fjy0,ty); |
707 | fjz0 = _mm_add_ps(fjz0,tz); |
708 | |
709 | /************************** |
710 | * CALCULATE INTERACTIONS * |
711 | **************************/ |
712 | |
713 | r30 = _mm_mul_ps(rsq30,rinv30); |
714 | r30 = _mm_andnot_ps(dummy_mask,r30); |
715 | |
716 | /* Compute parameters for interactions between i and j atoms */ |
717 | qq30 = _mm_mul_ps(iq3,jq0); |
718 | |
719 | /* EWALD ELECTROSTATICS */ |
720 | |
721 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
722 | ewrt = _mm_mul_ps(r30,ewtabscale); |
723 | ewitab = _mm_cvttps_epi32(ewrt); |
724 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
725 | ewitab = _mm_slli_epi32(ewitab,2); |
726 | ewtabF = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})) ); |
727 | ewtabD = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})) ); |
728 | ewtabV = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})) ); |
729 | ewtabFn = _mm_load_ps( ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})) ); |
730 | _MM_TRANSPOSE4_PS(ewtabF,ewtabD,ewtabV,ewtabFn)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((ewtabF ), (ewtabD)); tmp2 = _mm_unpacklo_ps((ewtabV), (ewtabFn)); tmp1 = _mm_unpackhi_ps((ewtabF), (ewtabD)); tmp3 = _mm_unpackhi_ps ((ewtabV), (ewtabFn)); (ewtabF) = _mm_movelh_ps(tmp0, tmp2); ( ewtabD) = _mm_movehl_ps(tmp2, tmp0); (ewtabV) = _mm_movelh_ps (tmp1, tmp3); (ewtabFn) = _mm_movehl_ps(tmp3, tmp1); } while ( 0); |
731 | felec = _mm_add_ps(ewtabF,_mm_mul_ps(eweps,ewtabD)); |
732 | velec = _mm_sub_ps(ewtabV,_mm_mul_ps(_mm_mul_ps(ewtabhalfspace,eweps),_mm_add_ps(ewtabF,felec))); |
733 | velec = _mm_mul_ps(qq30,_mm_sub_ps(rinv30,velec)); |
734 | felec = _mm_mul_ps(_mm_mul_ps(qq30,rinv30),_mm_sub_ps(rinvsq30,felec)); |
735 | |
736 | /* Update potential sum for this i atom from the interaction with this j atom. */ |
737 | velec = _mm_andnot_ps(dummy_mask,velec); |
738 | velecsum = _mm_add_ps(velecsum,velec); |
739 | |
740 | fscal = felec; |
741 | |
742 | fscal = _mm_andnot_ps(dummy_mask,fscal); |
743 | |
744 | /* Calculate temporary vectorial force */ |
745 | tx = _mm_mul_ps(fscal,dx30); |
746 | ty = _mm_mul_ps(fscal,dy30); |
747 | tz = _mm_mul_ps(fscal,dz30); |
748 | |
749 | /* Update vectorial force */ |
750 | fix3 = _mm_add_ps(fix3,tx); |
751 | fiy3 = _mm_add_ps(fiy3,ty); |
752 | fiz3 = _mm_add_ps(fiz3,tz); |
753 | |
754 | fjx0 = _mm_add_ps(fjx0,tx); |
755 | fjy0 = _mm_add_ps(fjy0,ty); |
756 | fjz0 = _mm_add_ps(fjz0,tz); |
757 | |
758 | fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch; |
759 | fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch; |
760 | fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch; |
761 | fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch; |
762 | |
763 | gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0); |
764 | |
765 | /* Inner loop uses 183 flops */ |
766 | } |
767 | |
768 | /* End of innermost loop */ |
769 | |
770 | gmx_mm_update_iforce_4atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3, |
771 | f+i_coord_offset,fshift+i_shift_offset); |
772 | |
773 | ggid = gid[iidx]; |
774 | /* Update potential energies */ |
775 | gmx_mm_update_1pot_ps(velecsum,kernel_data->energygrp_elec+ggid); |
776 | gmx_mm_update_1pot_ps(vvdwsum,kernel_data->energygrp_vdw+ggid); |
777 | |
778 | /* Increment number of inner iterations */ |
779 | inneriter += j_index_end - j_index_start; |
780 | |
781 | /* Outer loop uses 26 flops */ |
782 | } |
783 | |
784 | /* Increment number of outer iterations */ |
785 | outeriter += nri; |
786 | |
787 | /* Update outer/inner flops */ |
788 | |
789 | inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_VF,outeriter*26 + inneriter*183)(nrnb)->n[eNR_NBKERNEL_ELEC_VDW_W4_VF] += outeriter*26 + inneriter *183; |
790 | } |
791 | /* |
792 | * Gromacs nonbonded kernel: nb_kernel_ElecEw_VdwCSTab_GeomW4P1_F_sse4_1_single |
793 | * Electrostatics interaction: Ewald |
794 | * VdW interaction: CubicSplineTable |
795 | * Geometry: Water4-Particle |
796 | * Calculate force/pot: Force |
797 | */ |
798 | void |
799 | nb_kernel_ElecEw_VdwCSTab_GeomW4P1_F_sse4_1_single |
800 | (t_nblist * gmx_restrict nlist, |
801 | rvec * gmx_restrict xx, |
802 | rvec * gmx_restrict ff, |
803 | t_forcerec * gmx_restrict fr, |
804 | t_mdatoms * gmx_restrict mdatoms, |
805 | nb_kernel_data_t gmx_unused__attribute__ ((unused)) * gmx_restrict kernel_data, |
806 | t_nrnb * gmx_restrict nrnb) |
807 | { |
808 | /* Suffixes 0,1,2,3 refer to particle indices for waters in the inner or outer loop, or |
809 | * just 0 for non-waters. |
810 | * Suffixes A,B,C,D refer to j loop unrolling done with SSE, e.g. for the four different |
811 | * jnr indices corresponding to data put in the four positions in the SIMD register. |
812 | */ |
813 | int i_shift_offset,i_coord_offset,outeriter,inneriter; |
814 | int j_index_start,j_index_end,jidx,nri,inr,ggid,iidx; |
815 | int jnrA,jnrB,jnrC,jnrD; |
816 | int jnrlistA,jnrlistB,jnrlistC,jnrlistD; |
817 | int j_coord_offsetA,j_coord_offsetB,j_coord_offsetC,j_coord_offsetD; |
818 | int *iinr,*jindex,*jjnr,*shiftidx,*gid; |
819 | real rcutoff_scalar; |
820 | real *shiftvec,*fshift,*x,*f; |
821 | real *fjptrA,*fjptrB,*fjptrC,*fjptrD; |
822 | real scratch[4*DIM3]; |
823 | __m128 tx,ty,tz,fscal,rcutoff,rcutoff2,jidxall; |
824 | int vdwioffset0; |
825 | __m128 ix0,iy0,iz0,fix0,fiy0,fiz0,iq0,isai0; |
826 | int vdwioffset1; |
827 | __m128 ix1,iy1,iz1,fix1,fiy1,fiz1,iq1,isai1; |
828 | int vdwioffset2; |
829 | __m128 ix2,iy2,iz2,fix2,fiy2,fiz2,iq2,isai2; |
830 | int vdwioffset3; |
831 | __m128 ix3,iy3,iz3,fix3,fiy3,fiz3,iq3,isai3; |
832 | int vdwjidx0A,vdwjidx0B,vdwjidx0C,vdwjidx0D; |
833 | __m128 jx0,jy0,jz0,fjx0,fjy0,fjz0,jq0,isaj0; |
834 | __m128 dx00,dy00,dz00,rsq00,rinv00,rinvsq00,r00,qq00,c6_00,c12_00; |
835 | __m128 dx10,dy10,dz10,rsq10,rinv10,rinvsq10,r10,qq10,c6_10,c12_10; |
836 | __m128 dx20,dy20,dz20,rsq20,rinv20,rinvsq20,r20,qq20,c6_20,c12_20; |
837 | __m128 dx30,dy30,dz30,rsq30,rinv30,rinvsq30,r30,qq30,c6_30,c12_30; |
838 | __m128 velec,felec,velecsum,facel,crf,krf,krf2; |
839 | real *charge; |
840 | int nvdwtype; |
841 | __m128 rinvsix,rvdw,vvdw,vvdw6,vvdw12,fvdw,fvdw6,fvdw12,vvdwsum,sh_vdw_invrcut6; |
842 | int *vdwtype; |
843 | real *vdwparam; |
844 | __m128 one_sixth = _mm_set1_ps(1.0/6.0); |
845 | __m128 one_twelfth = _mm_set1_ps(1.0/12.0); |
Value stored to 'one_twelfth' during its initialization is never read | |
846 | __m128i vfitab; |
847 | __m128i ifour = _mm_set1_epi32(4); |
848 | __m128 rt,vfeps,vftabscale,Y,F,G,H,Heps,Fp,VV,FF; |
849 | real *vftab; |
850 | __m128i ewitab; |
851 | __m128 ewtabscale,eweps,sh_ewald,ewrt,ewtabhalfspace,ewtabF,ewtabFn,ewtabD,ewtabV; |
852 | real *ewtab; |
853 | __m128 dummy_mask,cutoff_mask; |
854 | __m128 signbit = _mm_castsi128_ps( _mm_set1_epi32(0x80000000) ); |
855 | __m128 one = _mm_set1_ps(1.0); |
856 | __m128 two = _mm_set1_ps(2.0); |
857 | x = xx[0]; |
858 | f = ff[0]; |
859 | |
860 | nri = nlist->nri; |
861 | iinr = nlist->iinr; |
862 | jindex = nlist->jindex; |
863 | jjnr = nlist->jjnr; |
864 | shiftidx = nlist->shift; |
865 | gid = nlist->gid; |
866 | shiftvec = fr->shift_vec[0]; |
867 | fshift = fr->fshift[0]; |
868 | facel = _mm_set1_ps(fr->epsfac); |
869 | charge = mdatoms->chargeA; |
870 | nvdwtype = fr->ntype; |
871 | vdwparam = fr->nbfp; |
872 | vdwtype = mdatoms->typeA; |
873 | |
874 | vftab = kernel_data->table_vdw->data; |
875 | vftabscale = _mm_set1_ps(kernel_data->table_vdw->scale); |
876 | |
877 | sh_ewald = _mm_set1_ps(fr->ic->sh_ewald); |
878 | ewtab = fr->ic->tabq_coul_F; |
879 | ewtabscale = _mm_set1_ps(fr->ic->tabq_scale); |
880 | ewtabhalfspace = _mm_set1_ps(0.5/fr->ic->tabq_scale); |
881 | |
882 | /* Setup water-specific parameters */ |
883 | inr = nlist->iinr[0]; |
884 | iq1 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+1])); |
885 | iq2 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+2])); |
886 | iq3 = _mm_mul_ps(facel,_mm_set1_ps(charge[inr+3])); |
887 | vdwioffset0 = 2*nvdwtype*vdwtype[inr+0]; |
888 | |
889 | /* Avoid stupid compiler warnings */ |
890 | jnrA = jnrB = jnrC = jnrD = 0; |
891 | j_coord_offsetA = 0; |
892 | j_coord_offsetB = 0; |
893 | j_coord_offsetC = 0; |
894 | j_coord_offsetD = 0; |
895 | |
896 | outeriter = 0; |
897 | inneriter = 0; |
898 | |
899 | for(iidx=0;iidx<4*DIM3;iidx++) |
900 | { |
901 | scratch[iidx] = 0.0; |
902 | } |
903 | |
904 | /* Start outer loop over neighborlists */ |
905 | for(iidx=0; iidx<nri; iidx++) |
906 | { |
907 | /* Load shift vector for this list */ |
908 | i_shift_offset = DIM3*shiftidx[iidx]; |
909 | |
910 | /* Load limits for loop over neighbors */ |
911 | j_index_start = jindex[iidx]; |
912 | j_index_end = jindex[iidx+1]; |
913 | |
914 | /* Get outer coordinate index */ |
915 | inr = iinr[iidx]; |
916 | i_coord_offset = DIM3*inr; |
917 | |
918 | /* Load i particle coords and add shift vector */ |
919 | gmx_mm_load_shift_and_4rvec_broadcast_ps(shiftvec+i_shift_offset,x+i_coord_offset, |
920 | &ix0,&iy0,&iz0,&ix1,&iy1,&iz1,&ix2,&iy2,&iz2,&ix3,&iy3,&iz3); |
921 | |
922 | fix0 = _mm_setzero_ps(); |
923 | fiy0 = _mm_setzero_ps(); |
924 | fiz0 = _mm_setzero_ps(); |
925 | fix1 = _mm_setzero_ps(); |
926 | fiy1 = _mm_setzero_ps(); |
927 | fiz1 = _mm_setzero_ps(); |
928 | fix2 = _mm_setzero_ps(); |
929 | fiy2 = _mm_setzero_ps(); |
930 | fiz2 = _mm_setzero_ps(); |
931 | fix3 = _mm_setzero_ps(); |
932 | fiy3 = _mm_setzero_ps(); |
933 | fiz3 = _mm_setzero_ps(); |
934 | |
935 | /* Start inner kernel loop */ |
936 | for(jidx=j_index_start; jidx<j_index_end && jjnr[jidx+3]>=0; jidx+=4) |
937 | { |
938 | |
939 | /* Get j neighbor index, and coordinate index */ |
940 | jnrA = jjnr[jidx]; |
941 | jnrB = jjnr[jidx+1]; |
942 | jnrC = jjnr[jidx+2]; |
943 | jnrD = jjnr[jidx+3]; |
944 | j_coord_offsetA = DIM3*jnrA; |
945 | j_coord_offsetB = DIM3*jnrB; |
946 | j_coord_offsetC = DIM3*jnrC; |
947 | j_coord_offsetD = DIM3*jnrD; |
948 | |
949 | /* load j atom coordinates */ |
950 | gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB, |
951 | x+j_coord_offsetC,x+j_coord_offsetD, |
952 | &jx0,&jy0,&jz0); |
953 | |
954 | /* Calculate displacement vector */ |
955 | dx00 = _mm_sub_ps(ix0,jx0); |
956 | dy00 = _mm_sub_ps(iy0,jy0); |
957 | dz00 = _mm_sub_ps(iz0,jz0); |
958 | dx10 = _mm_sub_ps(ix1,jx0); |
959 | dy10 = _mm_sub_ps(iy1,jy0); |
960 | dz10 = _mm_sub_ps(iz1,jz0); |
961 | dx20 = _mm_sub_ps(ix2,jx0); |
962 | dy20 = _mm_sub_ps(iy2,jy0); |
963 | dz20 = _mm_sub_ps(iz2,jz0); |
964 | dx30 = _mm_sub_ps(ix3,jx0); |
965 | dy30 = _mm_sub_ps(iy3,jy0); |
966 | dz30 = _mm_sub_ps(iz3,jz0); |
967 | |
968 | /* Calculate squared distance and things based on it */ |
969 | rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00); |
970 | rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10); |
971 | rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20); |
972 | rsq30 = gmx_mm_calc_rsq_ps(dx30,dy30,dz30); |
973 | |
974 | rinv00 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq00); |
975 | rinv10 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq10); |
976 | rinv20 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq20); |
977 | rinv30 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq30); |
978 | |
979 | rinvsq10 = _mm_mul_ps(rinv10,rinv10); |
980 | rinvsq20 = _mm_mul_ps(rinv20,rinv20); |
981 | rinvsq30 = _mm_mul_ps(rinv30,rinv30); |
982 | |
983 | /* Load parameters for j particles */ |
984 | jq0 = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0, |
985 | charge+jnrC+0,charge+jnrD+0); |
986 | vdwjidx0A = 2*vdwtype[jnrA+0]; |
987 | vdwjidx0B = 2*vdwtype[jnrB+0]; |
988 | vdwjidx0C = 2*vdwtype[jnrC+0]; |
989 | vdwjidx0D = 2*vdwtype[jnrD+0]; |
990 | |
991 | fjx0 = _mm_setzero_ps(); |
992 | fjy0 = _mm_setzero_ps(); |
993 | fjz0 = _mm_setzero_ps(); |
994 | |
995 | /************************** |
996 | * CALCULATE INTERACTIONS * |
997 | **************************/ |
998 | |
999 | r00 = _mm_mul_ps(rsq00,rinv00); |
1000 | |
1001 | /* Compute parameters for interactions between i and j atoms */ |
1002 | gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A, |
1003 | vdwparam+vdwioffset0+vdwjidx0B, |
1004 | vdwparam+vdwioffset0+vdwjidx0C, |
1005 | vdwparam+vdwioffset0+vdwjidx0D, |
1006 | &c6_00,&c12_00); |
1007 | |
1008 | /* Calculate table index by multiplying r with table scale and truncate to integer */ |
1009 | rt = _mm_mul_ps(r00,vftabscale); |
1010 | vfitab = _mm_cvttps_epi32(rt); |
1011 | vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (rt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
1012 | vfitab = _mm_slli_epi32(vfitab,3); |
1013 | |
1014 | /* CUBIC SPLINE TABLE DISPERSION */ |
1015 | Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(0) & 3];})) ); |
1016 | F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(1) & 3];})) ); |
1017 | G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(2) & 3];})) ); |
1018 | H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(3) & 3];})) ); |
1019 | _MM_TRANSPOSE4_PS(Y,F,G,H)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((Y ), (F)); tmp2 = _mm_unpacklo_ps((G), (H)); tmp1 = _mm_unpackhi_ps ((Y), (F)); tmp3 = _mm_unpackhi_ps((G), (H)); (Y) = _mm_movelh_ps (tmp0, tmp2); (F) = _mm_movehl_ps(tmp2, tmp0); (G) = _mm_movelh_ps (tmp1, tmp3); (H) = _mm_movehl_ps(tmp3, tmp1); } while (0); |
1020 | Heps = _mm_mul_ps(vfeps,H); |
1021 | Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps))); |
1022 | FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps)))); |
1023 | fvdw6 = _mm_mul_ps(c6_00,FF); |
1024 | |
1025 | /* CUBIC SPLINE TABLE REPULSION */ |
1026 | vfitab = _mm_add_epi32(vfitab,ifour); |
1027 | Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(0) & 3];})) ); |
1028 | F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(1) & 3];})) ); |
1029 | G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(2) & 3];})) ); |
1030 | H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(3) & 3];})) ); |
1031 | _MM_TRANSPOSE4_PS(Y,F,G,H)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((Y ), (F)); tmp2 = _mm_unpacklo_ps((G), (H)); tmp1 = _mm_unpackhi_ps ((Y), (F)); tmp3 = _mm_unpackhi_ps((G), (H)); (Y) = _mm_movelh_ps (tmp0, tmp2); (F) = _mm_movehl_ps(tmp2, tmp0); (G) = _mm_movelh_ps (tmp1, tmp3); (H) = _mm_movehl_ps(tmp3, tmp1); } while (0); |
1032 | Heps = _mm_mul_ps(vfeps,H); |
1033 | Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps))); |
1034 | FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps)))); |
1035 | fvdw12 = _mm_mul_ps(c12_00,FF); |
1036 | fvdw = _mm_xor_ps(signbit,_mm_mul_ps(_mm_add_ps(fvdw6,fvdw12),_mm_mul_ps(vftabscale,rinv00))); |
1037 | |
1038 | fscal = fvdw; |
1039 | |
1040 | /* Calculate temporary vectorial force */ |
1041 | tx = _mm_mul_ps(fscal,dx00); |
1042 | ty = _mm_mul_ps(fscal,dy00); |
1043 | tz = _mm_mul_ps(fscal,dz00); |
1044 | |
1045 | /* Update vectorial force */ |
1046 | fix0 = _mm_add_ps(fix0,tx); |
1047 | fiy0 = _mm_add_ps(fiy0,ty); |
1048 | fiz0 = _mm_add_ps(fiz0,tz); |
1049 | |
1050 | fjx0 = _mm_add_ps(fjx0,tx); |
1051 | fjy0 = _mm_add_ps(fjy0,ty); |
1052 | fjz0 = _mm_add_ps(fjz0,tz); |
1053 | |
1054 | /************************** |
1055 | * CALCULATE INTERACTIONS * |
1056 | **************************/ |
1057 | |
1058 | r10 = _mm_mul_ps(rsq10,rinv10); |
1059 | |
1060 | /* Compute parameters for interactions between i and j atoms */ |
1061 | qq10 = _mm_mul_ps(iq1,jq0); |
1062 | |
1063 | /* EWALD ELECTROSTATICS */ |
1064 | |
1065 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
1066 | ewrt = _mm_mul_ps(r10,ewtabscale); |
1067 | ewitab = _mm_cvttps_epi32(ewrt); |
1068 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
1069 | gmx_mm_load_4pair_swizzle_ps(ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})), |
1070 | ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})), |
1071 | &ewtabF,&ewtabFn); |
1072 | felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn)); |
1073 | felec = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec)); |
1074 | |
1075 | fscal = felec; |
1076 | |
1077 | /* Calculate temporary vectorial force */ |
1078 | tx = _mm_mul_ps(fscal,dx10); |
1079 | ty = _mm_mul_ps(fscal,dy10); |
1080 | tz = _mm_mul_ps(fscal,dz10); |
1081 | |
1082 | /* Update vectorial force */ |
1083 | fix1 = _mm_add_ps(fix1,tx); |
1084 | fiy1 = _mm_add_ps(fiy1,ty); |
1085 | fiz1 = _mm_add_ps(fiz1,tz); |
1086 | |
1087 | fjx0 = _mm_add_ps(fjx0,tx); |
1088 | fjy0 = _mm_add_ps(fjy0,ty); |
1089 | fjz0 = _mm_add_ps(fjz0,tz); |
1090 | |
1091 | /************************** |
1092 | * CALCULATE INTERACTIONS * |
1093 | **************************/ |
1094 | |
1095 | r20 = _mm_mul_ps(rsq20,rinv20); |
1096 | |
1097 | /* Compute parameters for interactions between i and j atoms */ |
1098 | qq20 = _mm_mul_ps(iq2,jq0); |
1099 | |
1100 | /* EWALD ELECTROSTATICS */ |
1101 | |
1102 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
1103 | ewrt = _mm_mul_ps(r20,ewtabscale); |
1104 | ewitab = _mm_cvttps_epi32(ewrt); |
1105 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
1106 | gmx_mm_load_4pair_swizzle_ps(ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})), |
1107 | ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})), |
1108 | &ewtabF,&ewtabFn); |
1109 | felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn)); |
1110 | felec = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec)); |
1111 | |
1112 | fscal = felec; |
1113 | |
1114 | /* Calculate temporary vectorial force */ |
1115 | tx = _mm_mul_ps(fscal,dx20); |
1116 | ty = _mm_mul_ps(fscal,dy20); |
1117 | tz = _mm_mul_ps(fscal,dz20); |
1118 | |
1119 | /* Update vectorial force */ |
1120 | fix2 = _mm_add_ps(fix2,tx); |
1121 | fiy2 = _mm_add_ps(fiy2,ty); |
1122 | fiz2 = _mm_add_ps(fiz2,tz); |
1123 | |
1124 | fjx0 = _mm_add_ps(fjx0,tx); |
1125 | fjy0 = _mm_add_ps(fjy0,ty); |
1126 | fjz0 = _mm_add_ps(fjz0,tz); |
1127 | |
1128 | /************************** |
1129 | * CALCULATE INTERACTIONS * |
1130 | **************************/ |
1131 | |
1132 | r30 = _mm_mul_ps(rsq30,rinv30); |
1133 | |
1134 | /* Compute parameters for interactions between i and j atoms */ |
1135 | qq30 = _mm_mul_ps(iq3,jq0); |
1136 | |
1137 | /* EWALD ELECTROSTATICS */ |
1138 | |
1139 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
1140 | ewrt = _mm_mul_ps(r30,ewtabscale); |
1141 | ewitab = _mm_cvttps_epi32(ewrt); |
1142 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
1143 | gmx_mm_load_4pair_swizzle_ps(ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})), |
1144 | ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})), |
1145 | &ewtabF,&ewtabFn); |
1146 | felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn)); |
1147 | felec = _mm_mul_ps(_mm_mul_ps(qq30,rinv30),_mm_sub_ps(rinvsq30,felec)); |
1148 | |
1149 | fscal = felec; |
1150 | |
1151 | /* Calculate temporary vectorial force */ |
1152 | tx = _mm_mul_ps(fscal,dx30); |
1153 | ty = _mm_mul_ps(fscal,dy30); |
1154 | tz = _mm_mul_ps(fscal,dz30); |
1155 | |
1156 | /* Update vectorial force */ |
1157 | fix3 = _mm_add_ps(fix3,tx); |
1158 | fiy3 = _mm_add_ps(fiy3,ty); |
1159 | fiz3 = _mm_add_ps(fiz3,tz); |
1160 | |
1161 | fjx0 = _mm_add_ps(fjx0,tx); |
1162 | fjy0 = _mm_add_ps(fjy0,ty); |
1163 | fjz0 = _mm_add_ps(fjz0,tz); |
1164 | |
1165 | fjptrA = f+j_coord_offsetA; |
1166 | fjptrB = f+j_coord_offsetB; |
1167 | fjptrC = f+j_coord_offsetC; |
1168 | fjptrD = f+j_coord_offsetD; |
1169 | |
1170 | gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0); |
1171 | |
1172 | /* Inner loop uses 156 flops */ |
1173 | } |
1174 | |
1175 | if(jidx<j_index_end) |
1176 | { |
1177 | |
1178 | /* Get j neighbor index, and coordinate index */ |
1179 | jnrlistA = jjnr[jidx]; |
1180 | jnrlistB = jjnr[jidx+1]; |
1181 | jnrlistC = jjnr[jidx+2]; |
1182 | jnrlistD = jjnr[jidx+3]; |
1183 | /* Sign of each element will be negative for non-real atoms. |
1184 | * This mask will be 0xFFFFFFFF for dummy entries and 0x0 for real ones, |
1185 | * so use it as val = _mm_andnot_ps(mask,val) to clear dummy entries. |
1186 | */ |
1187 | dummy_mask = gmx_mm_castsi128_ps_mm_castsi128_ps(_mm_cmplt_epi32(_mm_loadu_si128((const __m128i *)(jjnr+jidx)),_mm_setzero_si128())); |
1188 | jnrA = (jnrlistA>=0) ? jnrlistA : 0; |
1189 | jnrB = (jnrlistB>=0) ? jnrlistB : 0; |
1190 | jnrC = (jnrlistC>=0) ? jnrlistC : 0; |
1191 | jnrD = (jnrlistD>=0) ? jnrlistD : 0; |
1192 | j_coord_offsetA = DIM3*jnrA; |
1193 | j_coord_offsetB = DIM3*jnrB; |
1194 | j_coord_offsetC = DIM3*jnrC; |
1195 | j_coord_offsetD = DIM3*jnrD; |
1196 | |
1197 | /* load j atom coordinates */ |
1198 | gmx_mm_load_1rvec_4ptr_swizzle_ps(x+j_coord_offsetA,x+j_coord_offsetB, |
1199 | x+j_coord_offsetC,x+j_coord_offsetD, |
1200 | &jx0,&jy0,&jz0); |
1201 | |
1202 | /* Calculate displacement vector */ |
1203 | dx00 = _mm_sub_ps(ix0,jx0); |
1204 | dy00 = _mm_sub_ps(iy0,jy0); |
1205 | dz00 = _mm_sub_ps(iz0,jz0); |
1206 | dx10 = _mm_sub_ps(ix1,jx0); |
1207 | dy10 = _mm_sub_ps(iy1,jy0); |
1208 | dz10 = _mm_sub_ps(iz1,jz0); |
1209 | dx20 = _mm_sub_ps(ix2,jx0); |
1210 | dy20 = _mm_sub_ps(iy2,jy0); |
1211 | dz20 = _mm_sub_ps(iz2,jz0); |
1212 | dx30 = _mm_sub_ps(ix3,jx0); |
1213 | dy30 = _mm_sub_ps(iy3,jy0); |
1214 | dz30 = _mm_sub_ps(iz3,jz0); |
1215 | |
1216 | /* Calculate squared distance and things based on it */ |
1217 | rsq00 = gmx_mm_calc_rsq_ps(dx00,dy00,dz00); |
1218 | rsq10 = gmx_mm_calc_rsq_ps(dx10,dy10,dz10); |
1219 | rsq20 = gmx_mm_calc_rsq_ps(dx20,dy20,dz20); |
1220 | rsq30 = gmx_mm_calc_rsq_ps(dx30,dy30,dz30); |
1221 | |
1222 | rinv00 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq00); |
1223 | rinv10 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq10); |
1224 | rinv20 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq20); |
1225 | rinv30 = gmx_mm_invsqrt_psgmx_simd_invsqrt_f(rsq30); |
1226 | |
1227 | rinvsq10 = _mm_mul_ps(rinv10,rinv10); |
1228 | rinvsq20 = _mm_mul_ps(rinv20,rinv20); |
1229 | rinvsq30 = _mm_mul_ps(rinv30,rinv30); |
1230 | |
1231 | /* Load parameters for j particles */ |
1232 | jq0 = gmx_mm_load_4real_swizzle_ps(charge+jnrA+0,charge+jnrB+0, |
1233 | charge+jnrC+0,charge+jnrD+0); |
1234 | vdwjidx0A = 2*vdwtype[jnrA+0]; |
1235 | vdwjidx0B = 2*vdwtype[jnrB+0]; |
1236 | vdwjidx0C = 2*vdwtype[jnrC+0]; |
1237 | vdwjidx0D = 2*vdwtype[jnrD+0]; |
1238 | |
1239 | fjx0 = _mm_setzero_ps(); |
1240 | fjy0 = _mm_setzero_ps(); |
1241 | fjz0 = _mm_setzero_ps(); |
1242 | |
1243 | /************************** |
1244 | * CALCULATE INTERACTIONS * |
1245 | **************************/ |
1246 | |
1247 | r00 = _mm_mul_ps(rsq00,rinv00); |
1248 | r00 = _mm_andnot_ps(dummy_mask,r00); |
1249 | |
1250 | /* Compute parameters for interactions between i and j atoms */ |
1251 | gmx_mm_load_4pair_swizzle_ps(vdwparam+vdwioffset0+vdwjidx0A, |
1252 | vdwparam+vdwioffset0+vdwjidx0B, |
1253 | vdwparam+vdwioffset0+vdwjidx0C, |
1254 | vdwparam+vdwioffset0+vdwjidx0D, |
1255 | &c6_00,&c12_00); |
1256 | |
1257 | /* Calculate table index by multiplying r with table scale and truncate to integer */ |
1258 | rt = _mm_mul_ps(r00,vftabscale); |
1259 | vfitab = _mm_cvttps_epi32(rt); |
1260 | vfeps = _mm_sub_ps(rt,_mm_round_ps(rt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (rt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
1261 | vfitab = _mm_slli_epi32(vfitab,3); |
1262 | |
1263 | /* CUBIC SPLINE TABLE DISPERSION */ |
1264 | Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(0) & 3];})) ); |
1265 | F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(1) & 3];})) ); |
1266 | G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(2) & 3];})) ); |
1267 | H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(3) & 3];})) ); |
1268 | _MM_TRANSPOSE4_PS(Y,F,G,H)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((Y ), (F)); tmp2 = _mm_unpacklo_ps((G), (H)); tmp1 = _mm_unpackhi_ps ((Y), (F)); tmp3 = _mm_unpackhi_ps((G), (H)); (Y) = _mm_movelh_ps (tmp0, tmp2); (F) = _mm_movehl_ps(tmp2, tmp0); (G) = _mm_movelh_ps (tmp1, tmp3); (H) = _mm_movehl_ps(tmp3, tmp1); } while (0); |
1269 | Heps = _mm_mul_ps(vfeps,H); |
1270 | Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps))); |
1271 | FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps)))); |
1272 | fvdw6 = _mm_mul_ps(c6_00,FF); |
1273 | |
1274 | /* CUBIC SPLINE TABLE REPULSION */ |
1275 | vfitab = _mm_add_epi32(vfitab,ifour); |
1276 | Y = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,0)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(0) & 3];})) ); |
1277 | F = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,1)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(1) & 3];})) ); |
1278 | G = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,2)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(2) & 3];})) ); |
1279 | H = _mm_load_ps( vftab + gmx_mm_extract_epi32(vfitab,3)(__extension__ ({ __v4si __a = (__v4si)(vfitab); __a[(3) & 3];})) ); |
1280 | _MM_TRANSPOSE4_PS(Y,F,G,H)do { __m128 tmp3, tmp2, tmp1, tmp0; tmp0 = _mm_unpacklo_ps((Y ), (F)); tmp2 = _mm_unpacklo_ps((G), (H)); tmp1 = _mm_unpackhi_ps ((Y), (F)); tmp3 = _mm_unpackhi_ps((G), (H)); (Y) = _mm_movelh_ps (tmp0, tmp2); (F) = _mm_movehl_ps(tmp2, tmp0); (G) = _mm_movelh_ps (tmp1, tmp3); (H) = _mm_movehl_ps(tmp3, tmp1); } while (0); |
1281 | Heps = _mm_mul_ps(vfeps,H); |
1282 | Fp = _mm_add_ps(F,_mm_mul_ps(vfeps,_mm_add_ps(G,Heps))); |
1283 | FF = _mm_add_ps(Fp,_mm_mul_ps(vfeps,_mm_add_ps(G,_mm_add_ps(Heps,Heps)))); |
1284 | fvdw12 = _mm_mul_ps(c12_00,FF); |
1285 | fvdw = _mm_xor_ps(signbit,_mm_mul_ps(_mm_add_ps(fvdw6,fvdw12),_mm_mul_ps(vftabscale,rinv00))); |
1286 | |
1287 | fscal = fvdw; |
1288 | |
1289 | fscal = _mm_andnot_ps(dummy_mask,fscal); |
1290 | |
1291 | /* Calculate temporary vectorial force */ |
1292 | tx = _mm_mul_ps(fscal,dx00); |
1293 | ty = _mm_mul_ps(fscal,dy00); |
1294 | tz = _mm_mul_ps(fscal,dz00); |
1295 | |
1296 | /* Update vectorial force */ |
1297 | fix0 = _mm_add_ps(fix0,tx); |
1298 | fiy0 = _mm_add_ps(fiy0,ty); |
1299 | fiz0 = _mm_add_ps(fiz0,tz); |
1300 | |
1301 | fjx0 = _mm_add_ps(fjx0,tx); |
1302 | fjy0 = _mm_add_ps(fjy0,ty); |
1303 | fjz0 = _mm_add_ps(fjz0,tz); |
1304 | |
1305 | /************************** |
1306 | * CALCULATE INTERACTIONS * |
1307 | **************************/ |
1308 | |
1309 | r10 = _mm_mul_ps(rsq10,rinv10); |
1310 | r10 = _mm_andnot_ps(dummy_mask,r10); |
1311 | |
1312 | /* Compute parameters for interactions between i and j atoms */ |
1313 | qq10 = _mm_mul_ps(iq1,jq0); |
1314 | |
1315 | /* EWALD ELECTROSTATICS */ |
1316 | |
1317 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
1318 | ewrt = _mm_mul_ps(r10,ewtabscale); |
1319 | ewitab = _mm_cvttps_epi32(ewrt); |
1320 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
1321 | gmx_mm_load_4pair_swizzle_ps(ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})), |
1322 | ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})), |
1323 | &ewtabF,&ewtabFn); |
1324 | felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn)); |
1325 | felec = _mm_mul_ps(_mm_mul_ps(qq10,rinv10),_mm_sub_ps(rinvsq10,felec)); |
1326 | |
1327 | fscal = felec; |
1328 | |
1329 | fscal = _mm_andnot_ps(dummy_mask,fscal); |
1330 | |
1331 | /* Calculate temporary vectorial force */ |
1332 | tx = _mm_mul_ps(fscal,dx10); |
1333 | ty = _mm_mul_ps(fscal,dy10); |
1334 | tz = _mm_mul_ps(fscal,dz10); |
1335 | |
1336 | /* Update vectorial force */ |
1337 | fix1 = _mm_add_ps(fix1,tx); |
1338 | fiy1 = _mm_add_ps(fiy1,ty); |
1339 | fiz1 = _mm_add_ps(fiz1,tz); |
1340 | |
1341 | fjx0 = _mm_add_ps(fjx0,tx); |
1342 | fjy0 = _mm_add_ps(fjy0,ty); |
1343 | fjz0 = _mm_add_ps(fjz0,tz); |
1344 | |
1345 | /************************** |
1346 | * CALCULATE INTERACTIONS * |
1347 | **************************/ |
1348 | |
1349 | r20 = _mm_mul_ps(rsq20,rinv20); |
1350 | r20 = _mm_andnot_ps(dummy_mask,r20); |
1351 | |
1352 | /* Compute parameters for interactions between i and j atoms */ |
1353 | qq20 = _mm_mul_ps(iq2,jq0); |
1354 | |
1355 | /* EWALD ELECTROSTATICS */ |
1356 | |
1357 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
1358 | ewrt = _mm_mul_ps(r20,ewtabscale); |
1359 | ewitab = _mm_cvttps_epi32(ewrt); |
1360 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
1361 | gmx_mm_load_4pair_swizzle_ps(ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})), |
1362 | ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})), |
1363 | &ewtabF,&ewtabFn); |
1364 | felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn)); |
1365 | felec = _mm_mul_ps(_mm_mul_ps(qq20,rinv20),_mm_sub_ps(rinvsq20,felec)); |
1366 | |
1367 | fscal = felec; |
1368 | |
1369 | fscal = _mm_andnot_ps(dummy_mask,fscal); |
1370 | |
1371 | /* Calculate temporary vectorial force */ |
1372 | tx = _mm_mul_ps(fscal,dx20); |
1373 | ty = _mm_mul_ps(fscal,dy20); |
1374 | tz = _mm_mul_ps(fscal,dz20); |
1375 | |
1376 | /* Update vectorial force */ |
1377 | fix2 = _mm_add_ps(fix2,tx); |
1378 | fiy2 = _mm_add_ps(fiy2,ty); |
1379 | fiz2 = _mm_add_ps(fiz2,tz); |
1380 | |
1381 | fjx0 = _mm_add_ps(fjx0,tx); |
1382 | fjy0 = _mm_add_ps(fjy0,ty); |
1383 | fjz0 = _mm_add_ps(fjz0,tz); |
1384 | |
1385 | /************************** |
1386 | * CALCULATE INTERACTIONS * |
1387 | **************************/ |
1388 | |
1389 | r30 = _mm_mul_ps(rsq30,rinv30); |
1390 | r30 = _mm_andnot_ps(dummy_mask,r30); |
1391 | |
1392 | /* Compute parameters for interactions between i and j atoms */ |
1393 | qq30 = _mm_mul_ps(iq3,jq0); |
1394 | |
1395 | /* EWALD ELECTROSTATICS */ |
1396 | |
1397 | /* Calculate Ewald table index by multiplying r with scale and truncate to integer */ |
1398 | ewrt = _mm_mul_ps(r30,ewtabscale); |
1399 | ewitab = _mm_cvttps_epi32(ewrt); |
1400 | eweps = _mm_sub_ps(ewrt,_mm_round_ps(ewrt, _MM_FROUND_FLOOR)__extension__ ({ __m128 __X = (ewrt); (__m128) __builtin_ia32_roundps ((__v4sf)__X, ((0x00 | 0x01))); })); |
1401 | gmx_mm_load_4pair_swizzle_ps(ewtab + gmx_mm_extract_epi32(ewitab,0)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(0) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,1)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(1) & 3];})), |
1402 | ewtab + gmx_mm_extract_epi32(ewitab,2)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(2) & 3];})),ewtab + gmx_mm_extract_epi32(ewitab,3)(__extension__ ({ __v4si __a = (__v4si)(ewitab); __a[(3) & 3];})), |
1403 | &ewtabF,&ewtabFn); |
1404 | felec = _mm_add_ps(_mm_mul_ps( _mm_sub_ps(one,eweps),ewtabF),_mm_mul_ps(eweps,ewtabFn)); |
1405 | felec = _mm_mul_ps(_mm_mul_ps(qq30,rinv30),_mm_sub_ps(rinvsq30,felec)); |
1406 | |
1407 | fscal = felec; |
1408 | |
1409 | fscal = _mm_andnot_ps(dummy_mask,fscal); |
1410 | |
1411 | /* Calculate temporary vectorial force */ |
1412 | tx = _mm_mul_ps(fscal,dx30); |
1413 | ty = _mm_mul_ps(fscal,dy30); |
1414 | tz = _mm_mul_ps(fscal,dz30); |
1415 | |
1416 | /* Update vectorial force */ |
1417 | fix3 = _mm_add_ps(fix3,tx); |
1418 | fiy3 = _mm_add_ps(fiy3,ty); |
1419 | fiz3 = _mm_add_ps(fiz3,tz); |
1420 | |
1421 | fjx0 = _mm_add_ps(fjx0,tx); |
1422 | fjy0 = _mm_add_ps(fjy0,ty); |
1423 | fjz0 = _mm_add_ps(fjz0,tz); |
1424 | |
1425 | fjptrA = (jnrlistA>=0) ? f+j_coord_offsetA : scratch; |
1426 | fjptrB = (jnrlistB>=0) ? f+j_coord_offsetB : scratch; |
1427 | fjptrC = (jnrlistC>=0) ? f+j_coord_offsetC : scratch; |
1428 | fjptrD = (jnrlistD>=0) ? f+j_coord_offsetD : scratch; |
1429 | |
1430 | gmx_mm_decrement_1rvec_4ptr_swizzle_ps(fjptrA,fjptrB,fjptrC,fjptrD,fjx0,fjy0,fjz0); |
1431 | |
1432 | /* Inner loop uses 160 flops */ |
1433 | } |
1434 | |
1435 | /* End of innermost loop */ |
1436 | |
1437 | gmx_mm_update_iforce_4atom_swizzle_ps(fix0,fiy0,fiz0,fix1,fiy1,fiz1,fix2,fiy2,fiz2,fix3,fiy3,fiz3, |
1438 | f+i_coord_offset,fshift+i_shift_offset); |
1439 | |
1440 | /* Increment number of inner iterations */ |
1441 | inneriter += j_index_end - j_index_start; |
1442 | |
1443 | /* Outer loop uses 24 flops */ |
1444 | } |
1445 | |
1446 | /* Increment number of outer iterations */ |
1447 | outeriter += nri; |
1448 | |
1449 | /* Update outer/inner flops */ |
1450 | |
1451 | inc_nrnb(nrnb,eNR_NBKERNEL_ELEC_VDW_W4_F,outeriter*24 + inneriter*160)(nrnb)->n[eNR_NBKERNEL_ELEC_VDW_W4_F] += outeriter*24 + inneriter *160; |
1452 | } |