1 #include "simint/boys/boys.h"
2 #include "simint/ostei/gen/ostei_generated.h"
3 #include "simint/vectorization/vectorization.h"
4 #include <math.h>
5 #include <string.h>
6
7
ostei_i_i_d_p(struct simint_multi_shellpair const P,struct simint_multi_shellpair const Q,double screen_tol,double * const restrict work,double * const restrict INT__i_i_d_p)8 int ostei_i_i_d_p(struct simint_multi_shellpair const P,
9 struct simint_multi_shellpair const Q,
10 double screen_tol,
11 double * const restrict work,
12 double * const restrict INT__i_i_d_p)
13 {
14
15 SIMINT_ASSUME_ALIGN_DBL(work);
16 SIMINT_ASSUME_ALIGN_DBL(INT__i_i_d_p);
17 int ab, cd, abcd;
18 int istart, jstart;
19 int iprimcd, nprim_icd, icd;
20 const int check_screen = (screen_tol > 0.0);
21 int i, j;
22 int n;
23 int not_screened;
24 int real_abcd;
25 int iket;
26 int ibra;
27
28 // partition workspace
29 double * const INT__i_s_d_s = work + (SIMINT_NSHELL_SIMD * 0);
30 double * const INT__i_s_f_s = work + (SIMINT_NSHELL_SIMD * 168);
31 double * const INT__k_s_d_s = work + (SIMINT_NSHELL_SIMD * 448);
32 double * const INT__k_s_f_s = work + (SIMINT_NSHELL_SIMD * 664);
33 double * const INT__l_s_d_s = work + (SIMINT_NSHELL_SIMD * 1024);
34 double * const INT__l_s_f_s = work + (SIMINT_NSHELL_SIMD * 1294);
35 double * const INT__m_s_d_s = work + (SIMINT_NSHELL_SIMD * 1744);
36 double * const INT__m_s_f_s = work + (SIMINT_NSHELL_SIMD * 2074);
37 double * const INT__n_s_d_s = work + (SIMINT_NSHELL_SIMD * 2624);
38 double * const INT__n_s_f_s = work + (SIMINT_NSHELL_SIMD * 3020);
39 double * const INT__o_s_d_s = work + (SIMINT_NSHELL_SIMD * 3680);
40 double * const INT__o_s_f_s = work + (SIMINT_NSHELL_SIMD * 4148);
41 double * const INT__q_s_d_s = work + (SIMINT_NSHELL_SIMD * 4928);
42 double * const INT__q_s_f_s = work + (SIMINT_NSHELL_SIMD * 5474);
43 SIMINT_DBLTYPE * const primwork = (SIMINT_DBLTYPE *)(work + SIMINT_NSHELL_SIMD*6384);
44 SIMINT_DBLTYPE * const restrict PRIM_INT__s_s_s_s = primwork + 0;
45 SIMINT_DBLTYPE * const restrict PRIM_INT__p_s_s_s = primwork + 16;
46 SIMINT_DBLTYPE * const restrict PRIM_INT__d_s_s_s = primwork + 61;
47 SIMINT_DBLTYPE * const restrict PRIM_INT__f_s_s_s = primwork + 145;
48 SIMINT_DBLTYPE * const restrict PRIM_INT__g_s_s_s = primwork + 275;
49 SIMINT_DBLTYPE * const restrict PRIM_INT__g_s_p_s = primwork + 455;
50 SIMINT_DBLTYPE * const restrict PRIM_INT__h_s_s_s = primwork + 590;
51 SIMINT_DBLTYPE * const restrict PRIM_INT__h_s_p_s = primwork + 821;
52 SIMINT_DBLTYPE * const restrict PRIM_INT__h_s_d_s = primwork + 1010;
53 SIMINT_DBLTYPE * const restrict PRIM_INT__i_s_s_s = primwork + 1262;
54 SIMINT_DBLTYPE * const restrict PRIM_INT__i_s_p_s = primwork + 1542;
55 SIMINT_DBLTYPE * const restrict PRIM_INT__i_s_d_s = primwork + 1794;
56 SIMINT_DBLTYPE * const restrict PRIM_INT__i_s_f_s = primwork + 2130;
57 SIMINT_DBLTYPE * const restrict PRIM_INT__k_s_s_s = primwork + 2410;
58 SIMINT_DBLTYPE * const restrict PRIM_INT__k_s_p_s = primwork + 2734;
59 SIMINT_DBLTYPE * const restrict PRIM_INT__k_s_d_s = primwork + 3058;
60 SIMINT_DBLTYPE * const restrict PRIM_INT__k_s_f_s = primwork + 3490;
61 SIMINT_DBLTYPE * const restrict PRIM_INT__l_s_s_s = primwork + 3850;
62 SIMINT_DBLTYPE * const restrict PRIM_INT__l_s_p_s = primwork + 4210;
63 SIMINT_DBLTYPE * const restrict PRIM_INT__l_s_d_s = primwork + 4615;
64 SIMINT_DBLTYPE * const restrict PRIM_INT__l_s_f_s = primwork + 5155;
65 SIMINT_DBLTYPE * const restrict PRIM_INT__m_s_s_s = primwork + 5605;
66 SIMINT_DBLTYPE * const restrict PRIM_INT__m_s_p_s = primwork + 5990;
67 SIMINT_DBLTYPE * const restrict PRIM_INT__m_s_d_s = primwork + 6485;
68 SIMINT_DBLTYPE * const restrict PRIM_INT__m_s_f_s = primwork + 7145;
69 SIMINT_DBLTYPE * const restrict PRIM_INT__n_s_s_s = primwork + 7695;
70 SIMINT_DBLTYPE * const restrict PRIM_INT__n_s_p_s = primwork + 8091;
71 SIMINT_DBLTYPE * const restrict PRIM_INT__n_s_d_s = primwork + 8685;
72 SIMINT_DBLTYPE * const restrict PRIM_INT__n_s_f_s = primwork + 9477;
73 SIMINT_DBLTYPE * const restrict PRIM_INT__o_s_s_s = primwork + 10137;
74 SIMINT_DBLTYPE * const restrict PRIM_INT__o_s_p_s = primwork + 10527;
75 SIMINT_DBLTYPE * const restrict PRIM_INT__o_s_d_s = primwork + 11229;
76 SIMINT_DBLTYPE * const restrict PRIM_INT__o_s_f_s = primwork + 12165;
77 SIMINT_DBLTYPE * const restrict PRIM_INT__q_s_s_s = primwork + 12945;
78 SIMINT_DBLTYPE * const restrict PRIM_INT__q_s_p_s = primwork + 13309;
79 SIMINT_DBLTYPE * const restrict PRIM_INT__q_s_d_s = primwork + 14128;
80 SIMINT_DBLTYPE * const restrict PRIM_INT__q_s_f_s = primwork + 15220;
81 double * const hrrwork = (double *)(primwork + 16130);
82 double * const HRR_INT__i_p_d_s = hrrwork + 0;
83 double * const HRR_INT__i_p_f_s = hrrwork + 504;
84 double * const HRR_INT__i_d_d_s = hrrwork + 1344;
85 double * const HRR_INT__i_d_f_s = hrrwork + 2352;
86 double * const HRR_INT__i_f_d_s = hrrwork + 4032;
87 double * const HRR_INT__i_f_f_s = hrrwork + 5712;
88 double * const HRR_INT__i_g_d_s = hrrwork + 8512;
89 double * const HRR_INT__i_g_f_s = hrrwork + 11032;
90 double * const HRR_INT__i_h_d_s = hrrwork + 15232;
91 double * const HRR_INT__i_h_f_s = hrrwork + 18760;
92 double * const HRR_INT__i_i_d_s = hrrwork + 24640;
93 double * const HRR_INT__i_i_f_s = hrrwork + 29344;
94 double * const HRR_INT__k_p_d_s = hrrwork + 37184;
95 double * const HRR_INT__k_p_f_s = hrrwork + 37832;
96 double * const HRR_INT__k_d_d_s = hrrwork + 38912;
97 double * const HRR_INT__k_d_f_s = hrrwork + 40208;
98 double * const HRR_INT__k_f_d_s = hrrwork + 42368;
99 double * const HRR_INT__k_f_f_s = hrrwork + 44528;
100 double * const HRR_INT__k_g_d_s = hrrwork + 48128;
101 double * const HRR_INT__k_g_f_s = hrrwork + 51368;
102 double * const HRR_INT__k_h_d_s = hrrwork + 56768;
103 double * const HRR_INT__k_h_f_s = hrrwork + 61304;
104 double * const HRR_INT__l_p_d_s = hrrwork + 68864;
105 double * const HRR_INT__l_p_f_s = hrrwork + 69674;
106 double * const HRR_INT__l_d_d_s = hrrwork + 71024;
107 double * const HRR_INT__l_d_f_s = hrrwork + 72644;
108 double * const HRR_INT__l_f_d_s = hrrwork + 75344;
109 double * const HRR_INT__l_f_f_s = hrrwork + 78044;
110 double * const HRR_INT__l_g_d_s = hrrwork + 82544;
111 double * const HRR_INT__l_g_f_s = hrrwork + 86594;
112 double * const HRR_INT__m_p_d_s = hrrwork + 93344;
113 double * const HRR_INT__m_p_f_s = hrrwork + 94334;
114 double * const HRR_INT__m_d_d_s = hrrwork + 95984;
115 double * const HRR_INT__m_d_f_s = hrrwork + 97964;
116 double * const HRR_INT__m_f_d_s = hrrwork + 101264;
117 double * const HRR_INT__m_f_f_s = hrrwork + 104564;
118 double * const HRR_INT__n_p_d_s = hrrwork + 110064;
119 double * const HRR_INT__n_p_f_s = hrrwork + 111252;
120 double * const HRR_INT__n_d_d_s = hrrwork + 113232;
121 double * const HRR_INT__n_d_f_s = hrrwork + 115608;
122 double * const HRR_INT__o_p_d_s = hrrwork + 119568;
123 double * const HRR_INT__o_p_f_s = hrrwork + 120972;
124
125
126 // Create constants
127 const SIMINT_DBLTYPE const_1 = SIMINT_DBLSET1(1);
128 const SIMINT_DBLTYPE const_10 = SIMINT_DBLSET1(10);
129 const SIMINT_DBLTYPE const_11 = SIMINT_DBLSET1(11);
130 const SIMINT_DBLTYPE const_12 = SIMINT_DBLSET1(12);
131 const SIMINT_DBLTYPE const_2 = SIMINT_DBLSET1(2);
132 const SIMINT_DBLTYPE const_3 = SIMINT_DBLSET1(3);
133 const SIMINT_DBLTYPE const_4 = SIMINT_DBLSET1(4);
134 const SIMINT_DBLTYPE const_5 = SIMINT_DBLSET1(5);
135 const SIMINT_DBLTYPE const_6 = SIMINT_DBLSET1(6);
136 const SIMINT_DBLTYPE const_7 = SIMINT_DBLSET1(7);
137 const SIMINT_DBLTYPE const_8 = SIMINT_DBLSET1(8);
138 const SIMINT_DBLTYPE const_9 = SIMINT_DBLSET1(9);
139 const SIMINT_DBLTYPE one_half = SIMINT_DBLSET1(0.5);
140
141
142 ////////////////////////////////////////
143 // Loop over shells and primitives
144 ////////////////////////////////////////
145
146 real_abcd = 0;
147 istart = 0;
148 for(ab = 0; ab < P.nshell12_clip; ++ab)
149 {
150 const int iend = istart + P.nprim12[ab];
151
152 cd = 0;
153 jstart = 0;
154
155 for(cd = 0; cd < Q.nshell12_clip; cd += SIMINT_NSHELL_SIMD)
156 {
157 const int nshellbatch = ((cd + SIMINT_NSHELL_SIMD) > Q.nshell12_clip) ? Q.nshell12_clip - cd : SIMINT_NSHELL_SIMD;
158 int jend = jstart;
159 for(i = 0; i < nshellbatch; i++)
160 jend += Q.nprim12[cd+i];
161
162 // Clear the beginning of the workspace (where we are accumulating integrals)
163 memset(work, 0, SIMINT_NSHELL_SIMD * 6384 * sizeof(double));
164 abcd = 0;
165
166
167 for(i = istart; i < iend; ++i)
168 {
169 SIMINT_DBLTYPE bra_screen_max; // only used if check_screen
170
171 if(check_screen)
172 {
173 // Skip this whole thing if always insignificant
174 if((P.screen[i] * Q.screen_max) < screen_tol)
175 continue;
176 bra_screen_max = SIMINT_DBLSET1(P.screen[i]);
177 }
178
179 icd = 0;
180 iprimcd = 0;
181 nprim_icd = Q.nprim12[cd];
182 double * restrict PRIM_PTR_INT__i_s_d_s = INT__i_s_d_s + abcd * 168;
183 double * restrict PRIM_PTR_INT__i_s_f_s = INT__i_s_f_s + abcd * 280;
184 double * restrict PRIM_PTR_INT__k_s_d_s = INT__k_s_d_s + abcd * 216;
185 double * restrict PRIM_PTR_INT__k_s_f_s = INT__k_s_f_s + abcd * 360;
186 double * restrict PRIM_PTR_INT__l_s_d_s = INT__l_s_d_s + abcd * 270;
187 double * restrict PRIM_PTR_INT__l_s_f_s = INT__l_s_f_s + abcd * 450;
188 double * restrict PRIM_PTR_INT__m_s_d_s = INT__m_s_d_s + abcd * 330;
189 double * restrict PRIM_PTR_INT__m_s_f_s = INT__m_s_f_s + abcd * 550;
190 double * restrict PRIM_PTR_INT__n_s_d_s = INT__n_s_d_s + abcd * 396;
191 double * restrict PRIM_PTR_INT__n_s_f_s = INT__n_s_f_s + abcd * 660;
192 double * restrict PRIM_PTR_INT__o_s_d_s = INT__o_s_d_s + abcd * 468;
193 double * restrict PRIM_PTR_INT__o_s_f_s = INT__o_s_f_s + abcd * 780;
194 double * restrict PRIM_PTR_INT__q_s_d_s = INT__q_s_d_s + abcd * 546;
195 double * restrict PRIM_PTR_INT__q_s_f_s = INT__q_s_f_s + abcd * 910;
196
197
198
199 // Load these one per loop over i
200 const SIMINT_DBLTYPE P_alpha = SIMINT_DBLSET1(P.alpha[i]);
201 const SIMINT_DBLTYPE P_prefac = SIMINT_DBLSET1(P.prefac[i]);
202 const SIMINT_DBLTYPE Pxyz[3] = { SIMINT_DBLSET1(P.x[i]), SIMINT_DBLSET1(P.y[i]), SIMINT_DBLSET1(P.z[i]) };
203
204 const SIMINT_DBLTYPE P_PA[3] = { SIMINT_DBLSET1(P.PA_x[i]), SIMINT_DBLSET1(P.PA_y[i]), SIMINT_DBLSET1(P.PA_z[i]) };
205
206 for(j = jstart; j < jend; j += SIMINT_SIMD_LEN)
207 {
208 // calculate the shell offsets
209 // these are the offset from the shell pointed to by cd
210 // for each element
211 int shelloffsets[SIMINT_SIMD_LEN] = {0};
212 int lastoffset = 0;
213 const int nlane = ( ((j + SIMINT_SIMD_LEN) < jend) ? SIMINT_SIMD_LEN : (jend - j));
214
215 if((iprimcd + SIMINT_SIMD_LEN) >= nprim_icd)
216 {
217 // Handle if the first element of the vector is a new shell
218 if(iprimcd >= nprim_icd && ((icd+1) < nshellbatch))
219 {
220 nprim_icd += Q.nprim12[cd + (++icd)];
221 PRIM_PTR_INT__i_s_d_s += 168;
222 PRIM_PTR_INT__i_s_f_s += 280;
223 PRIM_PTR_INT__k_s_d_s += 216;
224 PRIM_PTR_INT__k_s_f_s += 360;
225 PRIM_PTR_INT__l_s_d_s += 270;
226 PRIM_PTR_INT__l_s_f_s += 450;
227 PRIM_PTR_INT__m_s_d_s += 330;
228 PRIM_PTR_INT__m_s_f_s += 550;
229 PRIM_PTR_INT__n_s_d_s += 396;
230 PRIM_PTR_INT__n_s_f_s += 660;
231 PRIM_PTR_INT__o_s_d_s += 468;
232 PRIM_PTR_INT__o_s_f_s += 780;
233 PRIM_PTR_INT__q_s_d_s += 546;
234 PRIM_PTR_INT__q_s_f_s += 910;
235 }
236 iprimcd++;
237 for(n = 1; n < SIMINT_SIMD_LEN; ++n)
238 {
239 if(iprimcd >= nprim_icd && ((icd+1) < nshellbatch))
240 {
241 shelloffsets[n] = shelloffsets[n-1] + 1;
242 lastoffset++;
243 nprim_icd += Q.nprim12[cd + (++icd)];
244 }
245 else
246 shelloffsets[n] = shelloffsets[n-1];
247 iprimcd++;
248 }
249 }
250 else
251 iprimcd += SIMINT_SIMD_LEN;
252
253 // Do we have to compute this vector (or has it been screened out)?
254 // (not_screened != 0 means we have to do this vector)
255 if(check_screen)
256 {
257 const double vmax = vector_max(SIMINT_MUL(bra_screen_max, SIMINT_DBLLOAD(Q.screen, j)));
258 if(vmax < screen_tol)
259 {
260 PRIM_PTR_INT__i_s_d_s += lastoffset*168;
261 PRIM_PTR_INT__i_s_f_s += lastoffset*280;
262 PRIM_PTR_INT__k_s_d_s += lastoffset*216;
263 PRIM_PTR_INT__k_s_f_s += lastoffset*360;
264 PRIM_PTR_INT__l_s_d_s += lastoffset*270;
265 PRIM_PTR_INT__l_s_f_s += lastoffset*450;
266 PRIM_PTR_INT__m_s_d_s += lastoffset*330;
267 PRIM_PTR_INT__m_s_f_s += lastoffset*550;
268 PRIM_PTR_INT__n_s_d_s += lastoffset*396;
269 PRIM_PTR_INT__n_s_f_s += lastoffset*660;
270 PRIM_PTR_INT__o_s_d_s += lastoffset*468;
271 PRIM_PTR_INT__o_s_f_s += lastoffset*780;
272 PRIM_PTR_INT__q_s_d_s += lastoffset*546;
273 PRIM_PTR_INT__q_s_f_s += lastoffset*910;
274 continue;
275 }
276 }
277
278 const SIMINT_DBLTYPE Q_alpha = SIMINT_DBLLOAD(Q.alpha, j);
279 const SIMINT_DBLTYPE PQalpha_mul = SIMINT_MUL(P_alpha, Q_alpha);
280 const SIMINT_DBLTYPE PQalpha_sum = SIMINT_ADD(P_alpha, Q_alpha);
281 const SIMINT_DBLTYPE one_over_PQalpha_sum = SIMINT_DIV(const_1, PQalpha_sum);
282
283
284 /* construct R2 = (Px - Qx)**2 + (Py - Qy)**2 + (Pz -Qz)**2 */
285 SIMINT_DBLTYPE PQ[3];
286 PQ[0] = SIMINT_SUB(Pxyz[0], SIMINT_DBLLOAD(Q.x, j));
287 PQ[1] = SIMINT_SUB(Pxyz[1], SIMINT_DBLLOAD(Q.y, j));
288 PQ[2] = SIMINT_SUB(Pxyz[2], SIMINT_DBLLOAD(Q.z, j));
289 SIMINT_DBLTYPE R2 = SIMINT_MUL(PQ[0], PQ[0]);
290 R2 = SIMINT_FMADD(PQ[1], PQ[1], R2);
291 R2 = SIMINT_FMADD(PQ[2], PQ[2], R2);
292
293 const SIMINT_DBLTYPE alpha = SIMINT_MUL(PQalpha_mul, one_over_PQalpha_sum); // alpha from MEST
294 const SIMINT_DBLTYPE one_over_p = SIMINT_DIV(const_1, P_alpha);
295 const SIMINT_DBLTYPE one_over_q = SIMINT_DIV(const_1, Q_alpha);
296 const SIMINT_DBLTYPE one_over_2p = SIMINT_MUL(one_half, one_over_p);
297 const SIMINT_DBLTYPE one_over_2q = SIMINT_MUL(one_half, one_over_q);
298 const SIMINT_DBLTYPE one_over_2pq = SIMINT_MUL(one_half, one_over_PQalpha_sum);
299 const SIMINT_DBLTYPE Q_PA[3] = { SIMINT_DBLLOAD(Q.PA_x, j), SIMINT_DBLLOAD(Q.PA_y, j), SIMINT_DBLLOAD(Q.PA_z, j) };
300
301 // NOTE: Minus sign!
302 const SIMINT_DBLTYPE a_over_p = SIMINT_MUL(SIMINT_NEG(alpha), one_over_p);
303 SIMINT_DBLTYPE aop_PQ[3];
304 aop_PQ[0] = SIMINT_MUL(a_over_p, PQ[0]);
305 aop_PQ[1] = SIMINT_MUL(a_over_p, PQ[1]);
306 aop_PQ[2] = SIMINT_MUL(a_over_p, PQ[2]);
307
308 SIMINT_DBLTYPE a_over_q = SIMINT_MUL(alpha, one_over_q);
309 SIMINT_DBLTYPE aoq_PQ[3];
310 aoq_PQ[0] = SIMINT_MUL(a_over_q, PQ[0]);
311 aoq_PQ[1] = SIMINT_MUL(a_over_q, PQ[1]);
312 aoq_PQ[2] = SIMINT_MUL(a_over_q, PQ[2]);
313 // Put a minus sign here so we don't have to in RR routines
314 a_over_q = SIMINT_NEG(a_over_q);
315
316
317 //////////////////////////////////////////////
318 // Fjt function section
319 // Maximum v value: 15
320 //////////////////////////////////////////////
321 // The parameter to the Fjt function
322 const SIMINT_DBLTYPE F_x = SIMINT_MUL(R2, alpha);
323
324
325 const SIMINT_DBLTYPE Q_prefac = mask_load(nlane, Q.prefac + j);
326
327
328 boys_F_split(PRIM_INT__s_s_s_s, F_x, 15);
329 SIMINT_DBLTYPE prefac = SIMINT_SQRT(one_over_PQalpha_sum);
330 prefac = SIMINT_MUL(SIMINT_MUL(P_prefac, Q_prefac), prefac);
331 for(n = 0; n <= 15; n++)
332 PRIM_INT__s_s_s_s[n] = SIMINT_MUL(PRIM_INT__s_s_s_s[n], prefac);
333
334 //////////////////////////////////////////////
335 // Primitive integrals: Vertical recurrance
336 //////////////////////////////////////////////
337
338 const SIMINT_DBLTYPE vrr_const_1_over_2p = one_over_2p;
339 const SIMINT_DBLTYPE vrr_const_2_over_2p = SIMINT_MUL(const_2, one_over_2p);
340 const SIMINT_DBLTYPE vrr_const_3_over_2p = SIMINT_MUL(const_3, one_over_2p);
341 const SIMINT_DBLTYPE vrr_const_4_over_2p = SIMINT_MUL(const_4, one_over_2p);
342 const SIMINT_DBLTYPE vrr_const_5_over_2p = SIMINT_MUL(const_5, one_over_2p);
343 const SIMINT_DBLTYPE vrr_const_6_over_2p = SIMINT_MUL(const_6, one_over_2p);
344 const SIMINT_DBLTYPE vrr_const_7_over_2p = SIMINT_MUL(const_7, one_over_2p);
345 const SIMINT_DBLTYPE vrr_const_8_over_2p = SIMINT_MUL(const_8, one_over_2p);
346 const SIMINT_DBLTYPE vrr_const_9_over_2p = SIMINT_MUL(const_9, one_over_2p);
347 const SIMINT_DBLTYPE vrr_const_10_over_2p = SIMINT_MUL(const_10, one_over_2p);
348 const SIMINT_DBLTYPE vrr_const_11_over_2p = SIMINT_MUL(const_11, one_over_2p);
349 const SIMINT_DBLTYPE vrr_const_1_over_2q = one_over_2q;
350 const SIMINT_DBLTYPE vrr_const_2_over_2q = SIMINT_MUL(const_2, one_over_2q);
351 const SIMINT_DBLTYPE vrr_const_1_over_2pq = one_over_2pq;
352 const SIMINT_DBLTYPE vrr_const_2_over_2pq = SIMINT_MUL(const_2, one_over_2pq);
353 const SIMINT_DBLTYPE vrr_const_3_over_2pq = SIMINT_MUL(const_3, one_over_2pq);
354 const SIMINT_DBLTYPE vrr_const_4_over_2pq = SIMINT_MUL(const_4, one_over_2pq);
355 const SIMINT_DBLTYPE vrr_const_5_over_2pq = SIMINT_MUL(const_5, one_over_2pq);
356 const SIMINT_DBLTYPE vrr_const_6_over_2pq = SIMINT_MUL(const_6, one_over_2pq);
357 const SIMINT_DBLTYPE vrr_const_7_over_2pq = SIMINT_MUL(const_7, one_over_2pq);
358 const SIMINT_DBLTYPE vrr_const_8_over_2pq = SIMINT_MUL(const_8, one_over_2pq);
359 const SIMINT_DBLTYPE vrr_const_9_over_2pq = SIMINT_MUL(const_9, one_over_2pq);
360 const SIMINT_DBLTYPE vrr_const_10_over_2pq = SIMINT_MUL(const_10, one_over_2pq);
361 const SIMINT_DBLTYPE vrr_const_11_over_2pq = SIMINT_MUL(const_11, one_over_2pq);
362 const SIMINT_DBLTYPE vrr_const_12_over_2pq = SIMINT_MUL(const_12, one_over_2pq);
363
364
365
366 // Forming PRIM_INT__p_s_s_s[15 * 3];
367 for(n = 0; n < 15; ++n) // loop over orders of auxiliary function
368 {
369
370 PRIM_INT__p_s_s_s[n * 3 + 0] = SIMINT_MUL(P_PA[0], PRIM_INT__s_s_s_s[n * 1 + 0]);
371 PRIM_INT__p_s_s_s[n * 3 + 0] = SIMINT_FMADD( aop_PQ[0], PRIM_INT__s_s_s_s[(n+1) * 1 + 0], PRIM_INT__p_s_s_s[n * 3 + 0]);
372
373 PRIM_INT__p_s_s_s[n * 3 + 1] = SIMINT_MUL(P_PA[1], PRIM_INT__s_s_s_s[n * 1 + 0]);
374 PRIM_INT__p_s_s_s[n * 3 + 1] = SIMINT_FMADD( aop_PQ[1], PRIM_INT__s_s_s_s[(n+1) * 1 + 0], PRIM_INT__p_s_s_s[n * 3 + 1]);
375
376 PRIM_INT__p_s_s_s[n * 3 + 2] = SIMINT_MUL(P_PA[2], PRIM_INT__s_s_s_s[n * 1 + 0]);
377 PRIM_INT__p_s_s_s[n * 3 + 2] = SIMINT_FMADD( aop_PQ[2], PRIM_INT__s_s_s_s[(n+1) * 1 + 0], PRIM_INT__p_s_s_s[n * 3 + 2]);
378
379 }
380
381
382
383 // Forming PRIM_INT__d_s_s_s[14 * 6];
384 for(n = 0; n < 14; ++n) // loop over orders of auxiliary function
385 {
386
387 PRIM_INT__d_s_s_s[n * 6 + 0] = SIMINT_MUL(P_PA[0], PRIM_INT__p_s_s_s[n * 3 + 0]);
388 PRIM_INT__d_s_s_s[n * 6 + 0] = SIMINT_FMADD( aop_PQ[0], PRIM_INT__p_s_s_s[(n+1) * 3 + 0], PRIM_INT__d_s_s_s[n * 6 + 0]);
389 PRIM_INT__d_s_s_s[n * 6 + 0] = SIMINT_FMADD( vrr_const_1_over_2p, SIMINT_FMADD(a_over_p, PRIM_INT__s_s_s_s[(n+1) * 1 + 0], PRIM_INT__s_s_s_s[n * 1 + 0]), PRIM_INT__d_s_s_s[n * 6 + 0]);
390
391 PRIM_INT__d_s_s_s[n * 6 + 1] = SIMINT_MUL(P_PA[1], PRIM_INT__p_s_s_s[n * 3 + 0]);
392 PRIM_INT__d_s_s_s[n * 6 + 1] = SIMINT_FMADD( aop_PQ[1], PRIM_INT__p_s_s_s[(n+1) * 3 + 0], PRIM_INT__d_s_s_s[n * 6 + 1]);
393
394 PRIM_INT__d_s_s_s[n * 6 + 3] = SIMINT_MUL(P_PA[1], PRIM_INT__p_s_s_s[n * 3 + 1]);
395 PRIM_INT__d_s_s_s[n * 6 + 3] = SIMINT_FMADD( aop_PQ[1], PRIM_INT__p_s_s_s[(n+1) * 3 + 1], PRIM_INT__d_s_s_s[n * 6 + 3]);
396 PRIM_INT__d_s_s_s[n * 6 + 3] = SIMINT_FMADD( vrr_const_1_over_2p, SIMINT_FMADD(a_over_p, PRIM_INT__s_s_s_s[(n+1) * 1 + 0], PRIM_INT__s_s_s_s[n * 1 + 0]), PRIM_INT__d_s_s_s[n * 6 + 3]);
397
398 PRIM_INT__d_s_s_s[n * 6 + 5] = SIMINT_MUL(P_PA[2], PRIM_INT__p_s_s_s[n * 3 + 2]);
399 PRIM_INT__d_s_s_s[n * 6 + 5] = SIMINT_FMADD( aop_PQ[2], PRIM_INT__p_s_s_s[(n+1) * 3 + 2], PRIM_INT__d_s_s_s[n * 6 + 5]);
400 PRIM_INT__d_s_s_s[n * 6 + 5] = SIMINT_FMADD( vrr_const_1_over_2p, SIMINT_FMADD(a_over_p, PRIM_INT__s_s_s_s[(n+1) * 1 + 0], PRIM_INT__s_s_s_s[n * 1 + 0]), PRIM_INT__d_s_s_s[n * 6 + 5]);
401
402 }
403
404
405
406 // Forming PRIM_INT__f_s_s_s[13 * 10];
407 for(n = 0; n < 13; ++n) // loop over orders of auxiliary function
408 {
409
410 PRIM_INT__f_s_s_s[n * 10 + 0] = SIMINT_MUL(P_PA[0], PRIM_INT__d_s_s_s[n * 6 + 0]);
411 PRIM_INT__f_s_s_s[n * 10 + 0] = SIMINT_FMADD( aop_PQ[0], PRIM_INT__d_s_s_s[(n+1) * 6 + 0], PRIM_INT__f_s_s_s[n * 10 + 0]);
412 PRIM_INT__f_s_s_s[n * 10 + 0] = SIMINT_FMADD( vrr_const_2_over_2p, SIMINT_FMADD(a_over_p, PRIM_INT__p_s_s_s[(n+1) * 3 + 0], PRIM_INT__p_s_s_s[n * 3 + 0]), PRIM_INT__f_s_s_s[n * 10 + 0]);
413
414 PRIM_INT__f_s_s_s[n * 10 + 1] = SIMINT_MUL(P_PA[1], PRIM_INT__d_s_s_s[n * 6 + 0]);
415 PRIM_INT__f_s_s_s[n * 10 + 1] = SIMINT_FMADD( aop_PQ[1], PRIM_INT__d_s_s_s[(n+1) * 6 + 0], PRIM_INT__f_s_s_s[n * 10 + 1]);
416
417 PRIM_INT__f_s_s_s[n * 10 + 2] = SIMINT_MUL(P_PA[2], PRIM_INT__d_s_s_s[n * 6 + 0]);
418 PRIM_INT__f_s_s_s[n * 10 + 2] = SIMINT_FMADD( aop_PQ[2], PRIM_INT__d_s_s_s[(n+1) * 6 + 0], PRIM_INT__f_s_s_s[n * 10 + 2]);
419
420 PRIM_INT__f_s_s_s[n * 10 + 3] = SIMINT_MUL(P_PA[0], PRIM_INT__d_s_s_s[n * 6 + 3]);
421 PRIM_INT__f_s_s_s[n * 10 + 3] = SIMINT_FMADD( aop_PQ[0], PRIM_INT__d_s_s_s[(n+1) * 6 + 3], PRIM_INT__f_s_s_s[n * 10 + 3]);
422
423 PRIM_INT__f_s_s_s[n * 10 + 4] = SIMINT_MUL(P_PA[2], PRIM_INT__d_s_s_s[n * 6 + 1]);
424 PRIM_INT__f_s_s_s[n * 10 + 4] = SIMINT_FMADD( aop_PQ[2], PRIM_INT__d_s_s_s[(n+1) * 6 + 1], PRIM_INT__f_s_s_s[n * 10 + 4]);
425
426 PRIM_INT__f_s_s_s[n * 10 + 5] = SIMINT_MUL(P_PA[0], PRIM_INT__d_s_s_s[n * 6 + 5]);
427 PRIM_INT__f_s_s_s[n * 10 + 5] = SIMINT_FMADD( aop_PQ[0], PRIM_INT__d_s_s_s[(n+1) * 6 + 5], PRIM_INT__f_s_s_s[n * 10 + 5]);
428
429 PRIM_INT__f_s_s_s[n * 10 + 6] = SIMINT_MUL(P_PA[1], PRIM_INT__d_s_s_s[n * 6 + 3]);
430 PRIM_INT__f_s_s_s[n * 10 + 6] = SIMINT_FMADD( aop_PQ[1], PRIM_INT__d_s_s_s[(n+1) * 6 + 3], PRIM_INT__f_s_s_s[n * 10 + 6]);
431 PRIM_INT__f_s_s_s[n * 10 + 6] = SIMINT_FMADD( vrr_const_2_over_2p, SIMINT_FMADD(a_over_p, PRIM_INT__p_s_s_s[(n+1) * 3 + 1], PRIM_INT__p_s_s_s[n * 3 + 1]), PRIM_INT__f_s_s_s[n * 10 + 6]);
432
433 PRIM_INT__f_s_s_s[n * 10 + 7] = SIMINT_MUL(P_PA[2], PRIM_INT__d_s_s_s[n * 6 + 3]);
434 PRIM_INT__f_s_s_s[n * 10 + 7] = SIMINT_FMADD( aop_PQ[2], PRIM_INT__d_s_s_s[(n+1) * 6 + 3], PRIM_INT__f_s_s_s[n * 10 + 7]);
435
436 PRIM_INT__f_s_s_s[n * 10 + 8] = SIMINT_MUL(P_PA[1], PRIM_INT__d_s_s_s[n * 6 + 5]);
437 PRIM_INT__f_s_s_s[n * 10 + 8] = SIMINT_FMADD( aop_PQ[1], PRIM_INT__d_s_s_s[(n+1) * 6 + 5], PRIM_INT__f_s_s_s[n * 10 + 8]);
438
439 PRIM_INT__f_s_s_s[n * 10 + 9] = SIMINT_MUL(P_PA[2], PRIM_INT__d_s_s_s[n * 6 + 5]);
440 PRIM_INT__f_s_s_s[n * 10 + 9] = SIMINT_FMADD( aop_PQ[2], PRIM_INT__d_s_s_s[(n+1) * 6 + 5], PRIM_INT__f_s_s_s[n * 10 + 9]);
441 PRIM_INT__f_s_s_s[n * 10 + 9] = SIMINT_FMADD( vrr_const_2_over_2p, SIMINT_FMADD(a_over_p, PRIM_INT__p_s_s_s[(n+1) * 3 + 2], PRIM_INT__p_s_s_s[n * 3 + 2]), PRIM_INT__f_s_s_s[n * 10 + 9]);
442
443 }
444
445
446 VRR_I_g_s_s_s(
447 PRIM_INT__g_s_s_s,
448 PRIM_INT__f_s_s_s,
449 PRIM_INT__d_s_s_s,
450 P_PA,
451 a_over_p,
452 aop_PQ,
453 one_over_2p,
454 12);
455
456
457 VRR_I_h_s_s_s(
458 PRIM_INT__h_s_s_s,
459 PRIM_INT__g_s_s_s,
460 PRIM_INT__f_s_s_s,
461 P_PA,
462 a_over_p,
463 aop_PQ,
464 one_over_2p,
465 11);
466
467
468 ostei_general_vrr1_I(6, 10,
469 one_over_2p, a_over_p, aop_PQ, P_PA,
470 PRIM_INT__h_s_s_s, PRIM_INT__g_s_s_s, PRIM_INT__i_s_s_s);
471
472
473 ostei_general_vrr_K(6, 0, 1, 0, 3,
474 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
475 PRIM_INT__i_s_s_s, NULL, NULL, PRIM_INT__h_s_s_s, NULL, PRIM_INT__i_s_p_s);
476
477
478 ostei_general_vrr_K(5, 0, 1, 0, 3,
479 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
480 PRIM_INT__h_s_s_s, NULL, NULL, PRIM_INT__g_s_s_s, NULL, PRIM_INT__h_s_p_s);
481
482
483 ostei_general_vrr_K(6, 0, 2, 0, 2,
484 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
485 PRIM_INT__i_s_p_s, PRIM_INT__i_s_s_s, NULL, PRIM_INT__h_s_p_s, NULL, PRIM_INT__i_s_d_s);
486
487
488 VRR_K_g_s_p_s(
489 PRIM_INT__g_s_p_s,
490 PRIM_INT__g_s_s_s,
491 PRIM_INT__f_s_s_s,
492 Q_PA,
493 aoq_PQ,
494 one_over_2pq,
495 3);
496
497
498 ostei_general_vrr_K(5, 0, 2, 0, 2,
499 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
500 PRIM_INT__h_s_p_s, PRIM_INT__h_s_s_s, NULL, PRIM_INT__g_s_p_s, NULL, PRIM_INT__h_s_d_s);
501
502
503 ostei_general_vrr_K(6, 0, 3, 0, 1,
504 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
505 PRIM_INT__i_s_d_s, PRIM_INT__i_s_p_s, NULL, PRIM_INT__h_s_d_s, NULL, PRIM_INT__i_s_f_s);
506
507
508 ostei_general_vrr1_I(7, 9,
509 one_over_2p, a_over_p, aop_PQ, P_PA,
510 PRIM_INT__i_s_s_s, PRIM_INT__h_s_s_s, PRIM_INT__k_s_s_s);
511
512
513 ostei_general_vrr_K(7, 0, 1, 0, 3,
514 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
515 PRIM_INT__k_s_s_s, NULL, NULL, PRIM_INT__i_s_s_s, NULL, PRIM_INT__k_s_p_s);
516
517
518 ostei_general_vrr_K(7, 0, 2, 0, 2,
519 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
520 PRIM_INT__k_s_p_s, PRIM_INT__k_s_s_s, NULL, PRIM_INT__i_s_p_s, NULL, PRIM_INT__k_s_d_s);
521
522
523 ostei_general_vrr_K(7, 0, 3, 0, 1,
524 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
525 PRIM_INT__k_s_d_s, PRIM_INT__k_s_p_s, NULL, PRIM_INT__i_s_d_s, NULL, PRIM_INT__k_s_f_s);
526
527
528 ostei_general_vrr1_I(8, 8,
529 one_over_2p, a_over_p, aop_PQ, P_PA,
530 PRIM_INT__k_s_s_s, PRIM_INT__i_s_s_s, PRIM_INT__l_s_s_s);
531
532
533 ostei_general_vrr_K(8, 0, 1, 0, 3,
534 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
535 PRIM_INT__l_s_s_s, NULL, NULL, PRIM_INT__k_s_s_s, NULL, PRIM_INT__l_s_p_s);
536
537
538 ostei_general_vrr_K(8, 0, 2, 0, 2,
539 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
540 PRIM_INT__l_s_p_s, PRIM_INT__l_s_s_s, NULL, PRIM_INT__k_s_p_s, NULL, PRIM_INT__l_s_d_s);
541
542
543 ostei_general_vrr_K(8, 0, 3, 0, 1,
544 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
545 PRIM_INT__l_s_d_s, PRIM_INT__l_s_p_s, NULL, PRIM_INT__k_s_d_s, NULL, PRIM_INT__l_s_f_s);
546
547
548 ostei_general_vrr1_I(9, 7,
549 one_over_2p, a_over_p, aop_PQ, P_PA,
550 PRIM_INT__l_s_s_s, PRIM_INT__k_s_s_s, PRIM_INT__m_s_s_s);
551
552
553 ostei_general_vrr_K(9, 0, 1, 0, 3,
554 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
555 PRIM_INT__m_s_s_s, NULL, NULL, PRIM_INT__l_s_s_s, NULL, PRIM_INT__m_s_p_s);
556
557
558 ostei_general_vrr_K(9, 0, 2, 0, 2,
559 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
560 PRIM_INT__m_s_p_s, PRIM_INT__m_s_s_s, NULL, PRIM_INT__l_s_p_s, NULL, PRIM_INT__m_s_d_s);
561
562
563 ostei_general_vrr_K(9, 0, 3, 0, 1,
564 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
565 PRIM_INT__m_s_d_s, PRIM_INT__m_s_p_s, NULL, PRIM_INT__l_s_d_s, NULL, PRIM_INT__m_s_f_s);
566
567
568 ostei_general_vrr1_I(10, 6,
569 one_over_2p, a_over_p, aop_PQ, P_PA,
570 PRIM_INT__m_s_s_s, PRIM_INT__l_s_s_s, PRIM_INT__n_s_s_s);
571
572
573 ostei_general_vrr_K(10, 0, 1, 0, 3,
574 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
575 PRIM_INT__n_s_s_s, NULL, NULL, PRIM_INT__m_s_s_s, NULL, PRIM_INT__n_s_p_s);
576
577
578 ostei_general_vrr_K(10, 0, 2, 0, 2,
579 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
580 PRIM_INT__n_s_p_s, PRIM_INT__n_s_s_s, NULL, PRIM_INT__m_s_p_s, NULL, PRIM_INT__n_s_d_s);
581
582
583 ostei_general_vrr_K(10, 0, 3, 0, 1,
584 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
585 PRIM_INT__n_s_d_s, PRIM_INT__n_s_p_s, NULL, PRIM_INT__m_s_d_s, NULL, PRIM_INT__n_s_f_s);
586
587
588 ostei_general_vrr1_I(11, 5,
589 one_over_2p, a_over_p, aop_PQ, P_PA,
590 PRIM_INT__n_s_s_s, PRIM_INT__m_s_s_s, PRIM_INT__o_s_s_s);
591
592
593 ostei_general_vrr_K(11, 0, 1, 0, 3,
594 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
595 PRIM_INT__o_s_s_s, NULL, NULL, PRIM_INT__n_s_s_s, NULL, PRIM_INT__o_s_p_s);
596
597
598 ostei_general_vrr_K(11, 0, 2, 0, 2,
599 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
600 PRIM_INT__o_s_p_s, PRIM_INT__o_s_s_s, NULL, PRIM_INT__n_s_p_s, NULL, PRIM_INT__o_s_d_s);
601
602
603 ostei_general_vrr_K(11, 0, 3, 0, 1,
604 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
605 PRIM_INT__o_s_d_s, PRIM_INT__o_s_p_s, NULL, PRIM_INT__n_s_d_s, NULL, PRIM_INT__o_s_f_s);
606
607
608 ostei_general_vrr1_I(12, 4,
609 one_over_2p, a_over_p, aop_PQ, P_PA,
610 PRIM_INT__o_s_s_s, PRIM_INT__n_s_s_s, PRIM_INT__q_s_s_s);
611
612
613 ostei_general_vrr_K(12, 0, 1, 0, 3,
614 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
615 PRIM_INT__q_s_s_s, NULL, NULL, PRIM_INT__o_s_s_s, NULL, PRIM_INT__q_s_p_s);
616
617
618 ostei_general_vrr_K(12, 0, 2, 0, 2,
619 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
620 PRIM_INT__q_s_p_s, PRIM_INT__q_s_s_s, NULL, PRIM_INT__o_s_p_s, NULL, PRIM_INT__q_s_d_s);
621
622
623 ostei_general_vrr_K(12, 0, 3, 0, 1,
624 one_over_2q, a_over_q, one_over_2pq, aoq_PQ, Q_PA,
625 PRIM_INT__q_s_d_s, PRIM_INT__q_s_p_s, NULL, PRIM_INT__o_s_d_s, NULL, PRIM_INT__q_s_f_s);
626
627
628
629
630 ////////////////////////////////////
631 // Accumulate contracted integrals
632 ////////////////////////////////////
633 if(lastoffset == 0)
634 {
635 contract_all(168, PRIM_INT__i_s_d_s, PRIM_PTR_INT__i_s_d_s);
636 contract_all(280, PRIM_INT__i_s_f_s, PRIM_PTR_INT__i_s_f_s);
637 contract_all(216, PRIM_INT__k_s_d_s, PRIM_PTR_INT__k_s_d_s);
638 contract_all(360, PRIM_INT__k_s_f_s, PRIM_PTR_INT__k_s_f_s);
639 contract_all(270, PRIM_INT__l_s_d_s, PRIM_PTR_INT__l_s_d_s);
640 contract_all(450, PRIM_INT__l_s_f_s, PRIM_PTR_INT__l_s_f_s);
641 contract_all(330, PRIM_INT__m_s_d_s, PRIM_PTR_INT__m_s_d_s);
642 contract_all(550, PRIM_INT__m_s_f_s, PRIM_PTR_INT__m_s_f_s);
643 contract_all(396, PRIM_INT__n_s_d_s, PRIM_PTR_INT__n_s_d_s);
644 contract_all(660, PRIM_INT__n_s_f_s, PRIM_PTR_INT__n_s_f_s);
645 contract_all(468, PRIM_INT__o_s_d_s, PRIM_PTR_INT__o_s_d_s);
646 contract_all(780, PRIM_INT__o_s_f_s, PRIM_PTR_INT__o_s_f_s);
647 contract_all(546, PRIM_INT__q_s_d_s, PRIM_PTR_INT__q_s_d_s);
648 contract_all(910, PRIM_INT__q_s_f_s, PRIM_PTR_INT__q_s_f_s);
649 }
650 else
651 {
652 contract(168, shelloffsets, PRIM_INT__i_s_d_s, PRIM_PTR_INT__i_s_d_s);
653 contract(280, shelloffsets, PRIM_INT__i_s_f_s, PRIM_PTR_INT__i_s_f_s);
654 contract(216, shelloffsets, PRIM_INT__k_s_d_s, PRIM_PTR_INT__k_s_d_s);
655 contract(360, shelloffsets, PRIM_INT__k_s_f_s, PRIM_PTR_INT__k_s_f_s);
656 contract(270, shelloffsets, PRIM_INT__l_s_d_s, PRIM_PTR_INT__l_s_d_s);
657 contract(450, shelloffsets, PRIM_INT__l_s_f_s, PRIM_PTR_INT__l_s_f_s);
658 contract(330, shelloffsets, PRIM_INT__m_s_d_s, PRIM_PTR_INT__m_s_d_s);
659 contract(550, shelloffsets, PRIM_INT__m_s_f_s, PRIM_PTR_INT__m_s_f_s);
660 contract(396, shelloffsets, PRIM_INT__n_s_d_s, PRIM_PTR_INT__n_s_d_s);
661 contract(660, shelloffsets, PRIM_INT__n_s_f_s, PRIM_PTR_INT__n_s_f_s);
662 contract(468, shelloffsets, PRIM_INT__o_s_d_s, PRIM_PTR_INT__o_s_d_s);
663 contract(780, shelloffsets, PRIM_INT__o_s_f_s, PRIM_PTR_INT__o_s_f_s);
664 contract(546, shelloffsets, PRIM_INT__q_s_d_s, PRIM_PTR_INT__q_s_d_s);
665 contract(910, shelloffsets, PRIM_INT__q_s_f_s, PRIM_PTR_INT__q_s_f_s);
666 PRIM_PTR_INT__i_s_d_s += lastoffset*168;
667 PRIM_PTR_INT__i_s_f_s += lastoffset*280;
668 PRIM_PTR_INT__k_s_d_s += lastoffset*216;
669 PRIM_PTR_INT__k_s_f_s += lastoffset*360;
670 PRIM_PTR_INT__l_s_d_s += lastoffset*270;
671 PRIM_PTR_INT__l_s_f_s += lastoffset*450;
672 PRIM_PTR_INT__m_s_d_s += lastoffset*330;
673 PRIM_PTR_INT__m_s_f_s += lastoffset*550;
674 PRIM_PTR_INT__n_s_d_s += lastoffset*396;
675 PRIM_PTR_INT__n_s_f_s += lastoffset*660;
676 PRIM_PTR_INT__o_s_d_s += lastoffset*468;
677 PRIM_PTR_INT__o_s_f_s += lastoffset*780;
678 PRIM_PTR_INT__q_s_d_s += lastoffset*546;
679 PRIM_PTR_INT__q_s_f_s += lastoffset*910;
680 }
681
682 } // close loop over j
683 } // close loop over i
684
685 //Advance to the next batch
686 jstart = SIMINT_SIMD_ROUND(jend);
687
688 //////////////////////////////////////////////
689 // Contracted integrals: Horizontal recurrance
690 //////////////////////////////////////////////
691
692
693 const double hAB[3] = { P.AB_x[ab], P.AB_y[ab], P.AB_z[ab] };
694
695
696 for(abcd = 0; abcd < nshellbatch; ++abcd, ++real_abcd)
697 {
698 const double hCD[3] = { Q.AB_x[cd+abcd], Q.AB_y[cd+abcd], Q.AB_z[cd+abcd] };
699
700 // set up HRR pointers
701 double const * restrict HRR_INT__i_s_d_s = INT__i_s_d_s + abcd * 168;
702 double const * restrict HRR_INT__i_s_f_s = INT__i_s_f_s + abcd * 280;
703 double const * restrict HRR_INT__k_s_d_s = INT__k_s_d_s + abcd * 216;
704 double const * restrict HRR_INT__k_s_f_s = INT__k_s_f_s + abcd * 360;
705 double const * restrict HRR_INT__l_s_d_s = INT__l_s_d_s + abcd * 270;
706 double const * restrict HRR_INT__l_s_f_s = INT__l_s_f_s + abcd * 450;
707 double const * restrict HRR_INT__m_s_d_s = INT__m_s_d_s + abcd * 330;
708 double const * restrict HRR_INT__m_s_f_s = INT__m_s_f_s + abcd * 550;
709 double const * restrict HRR_INT__n_s_d_s = INT__n_s_d_s + abcd * 396;
710 double const * restrict HRR_INT__n_s_f_s = INT__n_s_f_s + abcd * 660;
711 double const * restrict HRR_INT__o_s_d_s = INT__o_s_d_s + abcd * 468;
712 double const * restrict HRR_INT__o_s_f_s = INT__o_s_f_s + abcd * 780;
713 double const * restrict HRR_INT__q_s_d_s = INT__q_s_d_s + abcd * 546;
714 double const * restrict HRR_INT__q_s_f_s = INT__q_s_f_s + abcd * 910;
715 double * restrict HRR_INT__i_i_d_p = INT__i_i_d_p + real_abcd * 14112;
716
717 // form INT__i_p_d_s
718 ostei_general_hrr_J(6, 1, 2, 0, hAB, HRR_INT__k_s_d_s, HRR_INT__i_s_d_s, HRR_INT__i_p_d_s);
719
720 // form INT__i_p_f_s
721 ostei_general_hrr_J(6, 1, 3, 0, hAB, HRR_INT__k_s_f_s, HRR_INT__i_s_f_s, HRR_INT__i_p_f_s);
722
723 // form INT__k_p_d_s
724 ostei_general_hrr_J(7, 1, 2, 0, hAB, HRR_INT__l_s_d_s, HRR_INT__k_s_d_s, HRR_INT__k_p_d_s);
725
726 // form INT__k_p_f_s
727 ostei_general_hrr_J(7, 1, 3, 0, hAB, HRR_INT__l_s_f_s, HRR_INT__k_s_f_s, HRR_INT__k_p_f_s);
728
729 // form INT__l_p_d_s
730 ostei_general_hrr_J(8, 1, 2, 0, hAB, HRR_INT__m_s_d_s, HRR_INT__l_s_d_s, HRR_INT__l_p_d_s);
731
732 // form INT__l_p_f_s
733 ostei_general_hrr_J(8, 1, 3, 0, hAB, HRR_INT__m_s_f_s, HRR_INT__l_s_f_s, HRR_INT__l_p_f_s);
734
735 // form INT__m_p_d_s
736 ostei_general_hrr_J(9, 1, 2, 0, hAB, HRR_INT__n_s_d_s, HRR_INT__m_s_d_s, HRR_INT__m_p_d_s);
737
738 // form INT__m_p_f_s
739 ostei_general_hrr_J(9, 1, 3, 0, hAB, HRR_INT__n_s_f_s, HRR_INT__m_s_f_s, HRR_INT__m_p_f_s);
740
741 // form INT__n_p_d_s
742 ostei_general_hrr_J(10, 1, 2, 0, hAB, HRR_INT__o_s_d_s, HRR_INT__n_s_d_s, HRR_INT__n_p_d_s);
743
744 // form INT__n_p_f_s
745 ostei_general_hrr_J(10, 1, 3, 0, hAB, HRR_INT__o_s_f_s, HRR_INT__n_s_f_s, HRR_INT__n_p_f_s);
746
747 // form INT__o_p_d_s
748 ostei_general_hrr_J(11, 1, 2, 0, hAB, HRR_INT__q_s_d_s, HRR_INT__o_s_d_s, HRR_INT__o_p_d_s);
749
750 // form INT__o_p_f_s
751 ostei_general_hrr_J(11, 1, 3, 0, hAB, HRR_INT__q_s_f_s, HRR_INT__o_s_f_s, HRR_INT__o_p_f_s);
752
753 // form INT__i_d_d_s
754 ostei_general_hrr_J(6, 2, 2, 0, hAB, HRR_INT__k_p_d_s, HRR_INT__i_p_d_s, HRR_INT__i_d_d_s);
755
756 // form INT__i_d_f_s
757 ostei_general_hrr_J(6, 2, 3, 0, hAB, HRR_INT__k_p_f_s, HRR_INT__i_p_f_s, HRR_INT__i_d_f_s);
758
759 // form INT__k_d_d_s
760 ostei_general_hrr_J(7, 2, 2, 0, hAB, HRR_INT__l_p_d_s, HRR_INT__k_p_d_s, HRR_INT__k_d_d_s);
761
762 // form INT__k_d_f_s
763 ostei_general_hrr_J(7, 2, 3, 0, hAB, HRR_INT__l_p_f_s, HRR_INT__k_p_f_s, HRR_INT__k_d_f_s);
764
765 // form INT__l_d_d_s
766 ostei_general_hrr_J(8, 2, 2, 0, hAB, HRR_INT__m_p_d_s, HRR_INT__l_p_d_s, HRR_INT__l_d_d_s);
767
768 // form INT__l_d_f_s
769 ostei_general_hrr_J(8, 2, 3, 0, hAB, HRR_INT__m_p_f_s, HRR_INT__l_p_f_s, HRR_INT__l_d_f_s);
770
771 // form INT__m_d_d_s
772 ostei_general_hrr_J(9, 2, 2, 0, hAB, HRR_INT__n_p_d_s, HRR_INT__m_p_d_s, HRR_INT__m_d_d_s);
773
774 // form INT__m_d_f_s
775 ostei_general_hrr_J(9, 2, 3, 0, hAB, HRR_INT__n_p_f_s, HRR_INT__m_p_f_s, HRR_INT__m_d_f_s);
776
777 // form INT__n_d_d_s
778 ostei_general_hrr_J(10, 2, 2, 0, hAB, HRR_INT__o_p_d_s, HRR_INT__n_p_d_s, HRR_INT__n_d_d_s);
779
780 // form INT__n_d_f_s
781 ostei_general_hrr_J(10, 2, 3, 0, hAB, HRR_INT__o_p_f_s, HRR_INT__n_p_f_s, HRR_INT__n_d_f_s);
782
783 // form INT__i_f_d_s
784 ostei_general_hrr_J(6, 3, 2, 0, hAB, HRR_INT__k_d_d_s, HRR_INT__i_d_d_s, HRR_INT__i_f_d_s);
785
786 // form INT__i_f_f_s
787 ostei_general_hrr_J(6, 3, 3, 0, hAB, HRR_INT__k_d_f_s, HRR_INT__i_d_f_s, HRR_INT__i_f_f_s);
788
789 // form INT__k_f_d_s
790 ostei_general_hrr_J(7, 3, 2, 0, hAB, HRR_INT__l_d_d_s, HRR_INT__k_d_d_s, HRR_INT__k_f_d_s);
791
792 // form INT__k_f_f_s
793 ostei_general_hrr_J(7, 3, 3, 0, hAB, HRR_INT__l_d_f_s, HRR_INT__k_d_f_s, HRR_INT__k_f_f_s);
794
795 // form INT__l_f_d_s
796 ostei_general_hrr_J(8, 3, 2, 0, hAB, HRR_INT__m_d_d_s, HRR_INT__l_d_d_s, HRR_INT__l_f_d_s);
797
798 // form INT__l_f_f_s
799 ostei_general_hrr_J(8, 3, 3, 0, hAB, HRR_INT__m_d_f_s, HRR_INT__l_d_f_s, HRR_INT__l_f_f_s);
800
801 // form INT__m_f_d_s
802 ostei_general_hrr_J(9, 3, 2, 0, hAB, HRR_INT__n_d_d_s, HRR_INT__m_d_d_s, HRR_INT__m_f_d_s);
803
804 // form INT__m_f_f_s
805 ostei_general_hrr_J(9, 3, 3, 0, hAB, HRR_INT__n_d_f_s, HRR_INT__m_d_f_s, HRR_INT__m_f_f_s);
806
807 // form INT__i_g_d_s
808 ostei_general_hrr_J(6, 4, 2, 0, hAB, HRR_INT__k_f_d_s, HRR_INT__i_f_d_s, HRR_INT__i_g_d_s);
809
810 // form INT__i_g_f_s
811 ostei_general_hrr_J(6, 4, 3, 0, hAB, HRR_INT__k_f_f_s, HRR_INT__i_f_f_s, HRR_INT__i_g_f_s);
812
813 // form INT__k_g_d_s
814 ostei_general_hrr_J(7, 4, 2, 0, hAB, HRR_INT__l_f_d_s, HRR_INT__k_f_d_s, HRR_INT__k_g_d_s);
815
816 // form INT__k_g_f_s
817 ostei_general_hrr_J(7, 4, 3, 0, hAB, HRR_INT__l_f_f_s, HRR_INT__k_f_f_s, HRR_INT__k_g_f_s);
818
819 // form INT__l_g_d_s
820 ostei_general_hrr_J(8, 4, 2, 0, hAB, HRR_INT__m_f_d_s, HRR_INT__l_f_d_s, HRR_INT__l_g_d_s);
821
822 // form INT__l_g_f_s
823 ostei_general_hrr_J(8, 4, 3, 0, hAB, HRR_INT__m_f_f_s, HRR_INT__l_f_f_s, HRR_INT__l_g_f_s);
824
825 // form INT__i_h_d_s
826 ostei_general_hrr_J(6, 5, 2, 0, hAB, HRR_INT__k_g_d_s, HRR_INT__i_g_d_s, HRR_INT__i_h_d_s);
827
828 // form INT__i_h_f_s
829 ostei_general_hrr_J(6, 5, 3, 0, hAB, HRR_INT__k_g_f_s, HRR_INT__i_g_f_s, HRR_INT__i_h_f_s);
830
831 // form INT__k_h_d_s
832 ostei_general_hrr_J(7, 5, 2, 0, hAB, HRR_INT__l_g_d_s, HRR_INT__k_g_d_s, HRR_INT__k_h_d_s);
833
834 // form INT__k_h_f_s
835 ostei_general_hrr_J(7, 5, 3, 0, hAB, HRR_INT__l_g_f_s, HRR_INT__k_g_f_s, HRR_INT__k_h_f_s);
836
837 // form INT__i_i_d_s
838 ostei_general_hrr_J(6, 6, 2, 0, hAB, HRR_INT__k_h_d_s, HRR_INT__i_h_d_s, HRR_INT__i_i_d_s);
839
840 // form INT__i_i_f_s
841 ostei_general_hrr_J(6, 6, 3, 0, hAB, HRR_INT__k_h_f_s, HRR_INT__i_h_f_s, HRR_INT__i_i_f_s);
842
843 // form INT__i_i_d_p
844 for(ibra = 0; ibra < 784; ++ibra)
845 {
846 HRR_INT__i_i_d_p[ibra * 18 + 0] = HRR_INT__i_i_f_s[ibra * 10 + 0] + ( hCD[0] * HRR_INT__i_i_d_s[ibra * 6 + 0] );
847
848 HRR_INT__i_i_d_p[ibra * 18 + 1] = HRR_INT__i_i_f_s[ibra * 10 + 1] + ( hCD[1] * HRR_INT__i_i_d_s[ibra * 6 + 0] );
849
850 HRR_INT__i_i_d_p[ibra * 18 + 2] = HRR_INT__i_i_f_s[ibra * 10 + 2] + ( hCD[2] * HRR_INT__i_i_d_s[ibra * 6 + 0] );
851
852 HRR_INT__i_i_d_p[ibra * 18 + 3] = HRR_INT__i_i_f_s[ibra * 10 + 1] + ( hCD[0] * HRR_INT__i_i_d_s[ibra * 6 + 1] );
853
854 HRR_INT__i_i_d_p[ibra * 18 + 4] = HRR_INT__i_i_f_s[ibra * 10 + 3] + ( hCD[1] * HRR_INT__i_i_d_s[ibra * 6 + 1] );
855
856 HRR_INT__i_i_d_p[ibra * 18 + 5] = HRR_INT__i_i_f_s[ibra * 10 + 4] + ( hCD[2] * HRR_INT__i_i_d_s[ibra * 6 + 1] );
857
858 HRR_INT__i_i_d_p[ibra * 18 + 6] = HRR_INT__i_i_f_s[ibra * 10 + 2] + ( hCD[0] * HRR_INT__i_i_d_s[ibra * 6 + 2] );
859
860 HRR_INT__i_i_d_p[ibra * 18 + 7] = HRR_INT__i_i_f_s[ibra * 10 + 4] + ( hCD[1] * HRR_INT__i_i_d_s[ibra * 6 + 2] );
861
862 HRR_INT__i_i_d_p[ibra * 18 + 8] = HRR_INT__i_i_f_s[ibra * 10 + 5] + ( hCD[2] * HRR_INT__i_i_d_s[ibra * 6 + 2] );
863
864 HRR_INT__i_i_d_p[ibra * 18 + 9] = HRR_INT__i_i_f_s[ibra * 10 + 3] + ( hCD[0] * HRR_INT__i_i_d_s[ibra * 6 + 3] );
865
866 HRR_INT__i_i_d_p[ibra * 18 + 10] = HRR_INT__i_i_f_s[ibra * 10 + 6] + ( hCD[1] * HRR_INT__i_i_d_s[ibra * 6 + 3] );
867
868 HRR_INT__i_i_d_p[ibra * 18 + 11] = HRR_INT__i_i_f_s[ibra * 10 + 7] + ( hCD[2] * HRR_INT__i_i_d_s[ibra * 6 + 3] );
869
870 HRR_INT__i_i_d_p[ibra * 18 + 12] = HRR_INT__i_i_f_s[ibra * 10 + 4] + ( hCD[0] * HRR_INT__i_i_d_s[ibra * 6 + 4] );
871
872 HRR_INT__i_i_d_p[ibra * 18 + 13] = HRR_INT__i_i_f_s[ibra * 10 + 7] + ( hCD[1] * HRR_INT__i_i_d_s[ibra * 6 + 4] );
873
874 HRR_INT__i_i_d_p[ibra * 18 + 14] = HRR_INT__i_i_f_s[ibra * 10 + 8] + ( hCD[2] * HRR_INT__i_i_d_s[ibra * 6 + 4] );
875
876 HRR_INT__i_i_d_p[ibra * 18 + 15] = HRR_INT__i_i_f_s[ibra * 10 + 5] + ( hCD[0] * HRR_INT__i_i_d_s[ibra * 6 + 5] );
877
878 HRR_INT__i_i_d_p[ibra * 18 + 16] = HRR_INT__i_i_f_s[ibra * 10 + 8] + ( hCD[1] * HRR_INT__i_i_d_s[ibra * 6 + 5] );
879
880 HRR_INT__i_i_d_p[ibra * 18 + 17] = HRR_INT__i_i_f_s[ibra * 10 + 9] + ( hCD[2] * HRR_INT__i_i_d_s[ibra * 6 + 5] );
881
882 }
883
884
885 } // close HRR loop
886
887
888 } // close loop cdbatch
889
890 istart = iend;
891 } // close loop over ab
892
893 return P.nshell12_clip * Q.nshell12_clip;
894 }
895
ostei_i_i_p_d(struct simint_multi_shellpair const P,struct simint_multi_shellpair const Q,double screen_tol,double * const restrict work,double * const restrict INT__i_i_p_d)896 int ostei_i_i_p_d(struct simint_multi_shellpair const P,
897 struct simint_multi_shellpair const Q,
898 double screen_tol,
899 double * const restrict work,
900 double * const restrict INT__i_i_p_d)
901 {
902 double Q_AB[3*Q.nshell12];
903 struct simint_multi_shellpair Q_tmp = Q;
904 Q_tmp.PA_x = Q.PB_x; Q_tmp.PA_y = Q.PB_y; Q_tmp.PA_z = Q.PB_z;
905 Q_tmp.PB_x = Q.PA_x; Q_tmp.PB_y = Q.PA_y; Q_tmp.PB_z = Q.PA_z;
906 Q_tmp.AB_x = Q_AB;
907 Q_tmp.AB_y = Q_AB + Q.nshell12;
908 Q_tmp.AB_z = Q_AB + 2*Q.nshell12;
909
910 for(int i = 0; i < Q.nshell12; i++)
911 {
912 Q_tmp.AB_x[i] = -Q.AB_x[i];
913 Q_tmp.AB_y[i] = -Q.AB_y[i];
914 Q_tmp.AB_z[i] = -Q.AB_z[i];
915 }
916
917 int ret = ostei_i_i_d_p(P, Q_tmp, screen_tol, work, INT__i_i_p_d);
918 double buffer[14112] SIMINT_ALIGN_ARRAY_DBL;
919
920 for(int q = 0; q < ret; q++)
921 {
922 int idx = 0;
923 for(int a = 0; a < 28; ++a)
924 for(int b = 0; b < 28; ++b)
925 for(int c = 0; c < 3; ++c)
926 for(int d = 0; d < 6; ++d)
927 buffer[idx++] = INT__i_i_p_d[q*14112+a*504+b*18+d*3+c];
928
929 memcpy(INT__i_i_p_d+q*14112, buffer, 14112*sizeof(double));
930 }
931
932 return ret;
933 }
934
935