/
solver_2d.hpp
412 lines (364 loc) · 15.5 KB
/
solver_2d.hpp
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
/** @file
* @copyright University of Warsaw
* @section LICENSE
* GPLv3+ (see the COPYING file or http://www.gnu.org/licenses/)
*/
#pragma once
#include <libmpdata++/solvers/detail/solver_common.hpp>
namespace libmpdataxx
{
namespace solvers
{
namespace detail
{
using namespace libmpdataxx::arakawa_c;
template<typename ct_params_t, int n_tlev, int minhalo>
class solver<
ct_params_t,
n_tlev,
minhalo,
typename std::enable_if<ct_params_t::n_dims == 2 >::type
> : public solver_common<ct_params_t, n_tlev, minhalo>
{
using parent_t = solver_common<ct_params_t, n_tlev, minhalo>;
public:
using real_t = typename ct_params_t::real_t;
protected:
const rng_t i, j; // TODO: to be removed
// generic field used for various statistics (currently Courant number and divergence)
typename parent_t::arr_t &stat_field; // TODO: should be in solver common but cannot be allocated there ?
virtual void xchng_sclr(typename parent_t::arr_t &arr,
const idx_t<2> &range_ijk,
const int ext = 0,
const bool deriv = false
) final // for a given array
{
const auto range_ijk_0__ext = this->extend_range(range_ijk[0], ext);
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->fill_halos_sclr(arr, range_ijk[1]^ext, deriv);
for (auto &bc : this->bcs[1]) bc->fill_halos_sclr(arr, range_ijk_0__ext, deriv);
this->mem->barrier();
}
void xchng(int e) final
{
this->xchng_sclr(this->mem->psi[e][ this->n[e]], this->ijk, this->halo);
}
void xchng_vctr_alng(arrvec_t<typename parent_t::arr_t> &arrvec, const bool ad = false, const bool cyclic = false) final
{
this->mem->barrier();
if (!cyclic)
{
for (auto &bc : this->bcs[0]) bc->fill_halos_vctr_alng(arrvec, j, ad);
for (auto &bc : this->bcs[1]) bc->fill_halos_vctr_alng(arrvec, i, ad);
}
else
{
for (auto &bc : this->bcs[0]) bc->fill_halos_vctr_alng_cyclic(arrvec, j, ad);
for (auto &bc : this->bcs[1]) bc->fill_halos_vctr_alng_cyclic(arrvec, i, ad);
}
// TODO: open bc nust be last!!!
this->mem->barrier();
}
virtual void xchng_flux(arrvec_t<typename parent_t::arr_t> &arrvec) final
{
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->fill_halos_flux(arrvec, j);
for (auto &bc : this->bcs[1]) bc->fill_halos_flux(arrvec, i);
this->mem->barrier();
}
virtual void xchng_sgs_div(
typename parent_t::arr_t &arr,
const idx_t<2> &range_ijk
) final
{
this->mem->barrier();
for (auto &bc : this->bcs[1]) bc->fill_halos_sgs_div_stgr(arr, range_ijk[0]); // vip_div is staggered in vertical
for (auto &bc : this->bcs[0]) bc->fill_halos_sgs_div(arr, range_ijk[1]^h);
this->mem->barrier();
}
virtual void xchng_sgs_vctr(arrvec_t<typename parent_t::arr_t> &av,
const typename parent_t::arr_t &b,
const idx_t<2> &range_ijk
) final
{
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->fill_halos_sgs_vctr(av, b, range_ijk[1]);
for (auto &bc : this->bcs[1]) bc->fill_halos_sgs_vctr(av, b, range_ijk[0]);
this->mem->barrier();
}
virtual void xchng_sgs_tnsr_diag(arrvec_t<typename parent_t::arr_t> &av,
const typename parent_t::arr_t &w,
const typename parent_t::arr_t &vip_div,
const idx_t<2> &range_ijk
) final
{
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->fill_halos_sgs_tnsr(av, w, vip_div, range_ijk[1], this->dijk[0]);
for (auto &bc : this->bcs[1]) bc->fill_halos_sgs_tnsr(av, w, vip_div, range_ijk[0], this->dijk[1]);
this->mem->barrier();
}
virtual void xchng_sgs_tnsr_offdiag(arrvec_t<typename parent_t::arr_t> &av,
const arrvec_t<typename parent_t::arr_t> &bv,
const idx_t<2> &range_ijk,
const idx_t<2> &range_ijkm
) final
{
// off-diagonal components of stress tensor are treated the same as a vector
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->fill_halos_sgs_vctr(av, bv[0], range_ijkm[1], 2);
for (auto &bc : this->bcs[1]) bc->fill_halos_sgs_vctr(av, bv[0], range_ijkm[0], 1);
this->mem->barrier();
}
virtual void xchng_vctr_nrml(
arrvec_t<typename parent_t::arr_t> &arrvec,
const idx_t<2> &range_ijk,
const int ext = 0,
const bool cyclic = false
) final
{
const auto range_ijk_0__ext_h = this->extend_range(range_ijk[0], ext, h);
this->mem->barrier();
if (!cyclic)
{
for (auto &bc : this->bcs[1]) bc->fill_halos_vctr_nrml(arrvec[0], range_ijk_0__ext_h);
for (auto &bc : this->bcs[0]) bc->fill_halos_vctr_nrml(arrvec[1], range_ijk[1]^ext^h);
}
else
{
for (auto &bc : this->bcs[1]) bc->fill_halos_vctr_nrml_cyclic(arrvec[0], range_ijk_0__ext_h);
for (auto &bc : this->bcs[0]) bc->fill_halos_vctr_nrml_cyclic(arrvec[1], range_ijk[1]^ext^h);
}
this->mem->barrier();
}
virtual void xchng_pres(
typename parent_t::arr_t &arr,
const idx_t<2> &range_ijk,
const int ext = 0
) final
{
const auto range_ijk_0__ext = this->extend_range(range_ijk[0], ext);
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->fill_halos_pres(arr, range_ijk[1]^ext);
for (auto &bc : this->bcs[1]) bc->fill_halos_pres(arr, range_ijk_0__ext);
this->mem->barrier();
}
virtual void set_edges(
arrvec_t<typename parent_t::arr_t> &av,
const idx_t<2> &range_ijk,
const int &sign
) final
{
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->set_edge_pres(av[0], range_ijk[1], sign);
for (auto &bc : this->bcs[1]) bc->set_edge_pres(av[1], range_ijk[0], sign);
this->mem->barrier();
}
virtual void save_edges(
const arrvec_t<typename parent_t::arr_t> &av,
const idx_t<2> &range_ijk
) final
{
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->save_edge_vel(av[0], range_ijk[1]);
for (auto &bc : this->bcs[1]) bc->save_edge_vel(av[1], range_ijk[0]);
this->mem->barrier();
}
virtual void avg_edge_sclr(typename parent_t::arr_t &arr,
const idx_t<2> &range_ijk
) final
{
this->mem->barrier();
for (auto &bc : this->bcs[0]) bc->copy_edge_sclr_to_halo1_cyclic(arr, range_ijk[1]);
for (auto &bc : this->bcs[1]) bc->copy_edge_sclr_to_halo1_cyclic(arr, range_ijk[0]);
this->mem->barrier(); // wait for all threads to copy edge to halo before modifying edge. note: shmem decomposition in x
for (auto &bc : this->bcs[0]) bc->avg_edge_and_halo1_sclr_cyclic(arr, range_ijk[1]);
for (auto &bc : this->bcs[1]) bc->avg_edge_and_halo1_sclr_cyclic(arr, range_ijk[0]);
this->mem->barrier();
}
// TODO: ref in argument...
void hook_ante_loop(const typename parent_t::advance_arg_t nt) // TODO: this nt conflicts in fact with multiple-advance()-call logic!
{
parent_t::hook_ante_loop(nt);
// sanity check for non-divergence of the initial Courant number field
// (including compatibility with the initial condition)
// TODO: same in 1D
if (!opts::isset(ct_params_t::opts, opts::dfl))
{
real_t max_abs_div = max_abs_vctr_div(this->mem->GC);
if (max_abs_div > this->max_abs_div_eps)
throw std::runtime_error("libmpdata++: initial advector field is divergent");
}
}
real_t courant_number(const arrvec_t<typename parent_t::arr_t> &arrvec) final
{
stat_field(this->ijk) = real_t(0.5) * (
abs(arrvec[0](i+h, j) + arrvec[0](i-h, j))
+ abs(arrvec[1](i, j+h) + arrvec[1](i, j-h))
) / formulae::G<ct_params_t::opts, 0>(*this->mem->G, i, j);
return this->mem->max(this->rank, stat_field(this->ijk));
}
real_t max_abs_vctr_div(const arrvec_t<typename parent_t::arr_t> &arrvec) final
{
stat_field(this->ijk) = abs(
(arrvec[0](i+h, j) - arrvec[0](i-h, j))
+ (arrvec[1](i, j+h) - arrvec[1](i, j-h))
) / formulae::G<ct_params_t::opts, 0>(*this->mem->G, i, j);
return this->mem->max(this->rank, stat_field(this->ijk));
}
void scale_gc(const real_t time,
const real_t cur_dt,
const real_t old_dt) final
{
this->mem->GC[0](rng_t(i.first(), i.last()-1)^h, j) *= cur_dt / old_dt;
this->mem->GC[1](i, rng_t(j.first(), j.last()-1)^h) *= cur_dt / old_dt;
this->xchng_vctr_alng(this->mem->GC);
auto ex = this->halo - 1;
this->xchng_vctr_nrml(this->mem->GC, this->ijk, ex);
}
public:
struct ctor_args_t
{
// <TODO> these should be common for 1D,2D,3D
int rank;
typename parent_t::mem_t *mem;
// </TODO>
typename parent_t::bcp_t &bcxl, &bcxr, &bcyl, &bcyr;
const rng_t &i, &j;
};
struct rt_params_t : parent_t::rt_params_t
{
real_t di = 0, dj = 0;
};
protected:
// ctor
solver(
ctor_args_t args,
const rt_params_t &p
) :
parent_t(
args.rank,
args.mem,
p,
idx_t<parent_t::n_dims>({args.i, args.j})
),
i(args.i),
j(args.j),
stat_field(args.mem->tmp[__FILE__][0][0])
{
this->di = p.di;
this->dj = p.dj;
this->dijk = {p.di, p.dj};
this->set_bcs(0, args.bcxl, args.bcxr);
this->set_bcs(1, args.bcyl, args.bcyr);
}
// memory allocation logic using static methods
public:
static void alloc(
typename parent_t::mem_t *mem,
const int &n_iters
) {
// psi
mem->psi.resize(parent_t::n_eqns);
for (int e = 0; e < parent_t::n_eqns; ++e) // equations
for (int n = 0; n < n_tlev; ++n) // time levels
mem->psi[e].push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_sclr(mem->grid_size[0]),
parent_t::rng_sclr(mem->grid_size[1])
)));
// Courant field components (Arakawa-C grid)
mem->GC.push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_vctr(mem->grid_size[0]),
parent_t::rng_sclr(mem->grid_size[1])
)));
mem->GC.push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_sclr(mem->grid_size[0]),
parent_t::rng_vctr(mem->grid_size[1])
)));
// fully third-order accurate mpdata needs also time derivatives of
// the Courant field
if (opts::isset(ct_params_t::opts, opts::div_3rd) ||
opts::isset(ct_params_t::opts, opts::div_3rd_dt))
{
// TODO: why for (auto f : {mem->ndt_GC, mem->ndtt_GC}) doesn't work ?
mem->ndt_GC.push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_vctr(mem->grid_size[0]),
parent_t::rng_sclr(mem->grid_size[1])
)));
mem->ndt_GC.push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_sclr(mem->grid_size[0]),
parent_t::rng_vctr(mem->grid_size[1])
)));
mem->ndtt_GC.push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_vctr(mem->grid_size[0]),
parent_t::rng_sclr(mem->grid_size[1])
)));
mem->ndtt_GC.push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_sclr(mem->grid_size[0]),
parent_t::rng_vctr(mem->grid_size[1])
)));
}
// allocate G
if (opts::isset(ct_params_t::opts, opts::nug))
mem->G.reset(mem->old(new typename parent_t::arr_t(
parent_t::rng_sclr(mem->grid_size[0]),
parent_t::rng_sclr(mem->grid_size[1])
)));
// allocate Kahan summation temporary vars
if (opts::isset(ct_params_t::opts, opts::khn))
for (int n = 0; n < 3; ++n)
mem->khn_tmp.push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_sclr(mem->grid_size[0]),
parent_t::rng_sclr(mem->grid_size[1])
)));
// courant field
alloc_tmp_sclr(mem, __FILE__, 1);
}
protected:
// helper method to allocate a temporary space composed of arbitrarily staggered arrays
static void alloc_tmp_stgr(
typename parent_t::mem_t *mem,
const char * __file__,
const int n_arr,
const std::vector<std::vector<bool>> &stgr,
bool srfc = false
)
{
mem->tmp[__file__].push_back(new arrvec_t<typename parent_t::arr_t>());
for (int n = 0; n < n_arr; ++n)
{
mem->tmp[__file__].back().push_back(mem->old(new typename parent_t::arr_t(
stgr[n][0] ? parent_t::rng_vctr(mem->grid_size[0]) : parent_t::rng_sclr(mem->grid_size[0]),
srfc ? rng_t(0, 0) :
stgr[n][1] ? parent_t::rng_vctr(mem->grid_size[1]) :
parent_t::rng_sclr(mem->grid_size[1])
)));
}
}
// helper method to allocate a temporary space composed of vector-component arrays
static void alloc_tmp_vctr(
typename parent_t::mem_t *mem,
const char * __file__
)
{
alloc_tmp_stgr(mem, __file__, 2, {{true, false}, {false, true}});
}
// helper method to allocate n_arr scalar temporary arrays
static void alloc_tmp_sclr(
typename parent_t::mem_t *mem,
const char * __file__, const int n_arr,
std::string name = "",
bool srfc = false
)
{
mem->tmp[__file__].push_back(new arrvec_t<typename parent_t::arr_t>());
if (!name.empty()) mem->avail_tmp[name] = std::make_pair(__file__, mem->tmp[__file__].size() - 1);
for (int n = 0; n < n_arr; ++n)
mem->tmp[__file__].back().push_back(mem->old(new typename parent_t::arr_t(
parent_t::rng_sclr(mem->grid_size[0]),
srfc ? rng_t(0, 0) : parent_t::rng_sclr(mem->grid_size[1])
)));
}
};
} // namespace detail
} // namespace solvers
} // namespace libmpdataxx