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
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
|
/*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software Foundation,
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*/
#pragma once
/** \file
* \ingroup bli
*
* A virtual array is a data structure that behaves similar to an array, but its elements are
* accessed through virtual methods. This improves the decoupling of a function from its callers,
* because it does not have to know exactly how the data is laid out in memory, or if it is stored
* in memory at all. It could just as well be computed on the fly.
*
* Taking a virtual array as parameter instead of a more specific non-virtual type has some
* tradeoffs. Access to individual elements of the individual elements is higher due to function
* call overhead. On the other hand, potential callers don't have to convert the data into the
* specific format required for the function. This can be a costly conversion if only few of the
* elements are accessed in the end.
*
* Functions taking a virtual array as input can still optimize for different data layouts. For
* example, they can check if the array is stored as an array internally or if it is the same
* element for all indices. Whether it is worth to optimize for different data layouts in a
* function has to be decided on a case by case basis. One should always do some benchmarking to
* see of the increased compile time and binary size is worth it.
*/
#include "BLI_array.hh"
#include "BLI_index_mask.hh"
#include "BLI_span.hh"
namespace blender {
/* An immutable virtual array. */
template<typename T> class VArray {
protected:
int64_t size_;
public:
VArray(const int64_t size) : size_(size)
{
BLI_assert(size_ >= 0);
}
virtual ~VArray() = default;
T get(const int64_t index) const
{
BLI_assert(index >= 0);
BLI_assert(index < size_);
return this->get_impl(index);
}
int64_t size() const
{
return size_;
}
bool is_empty() const
{
return size_ == 0;
}
IndexRange index_range() const
{
return IndexRange(size_);
}
/* Returns true when the virtual array is stored as a span internally. */
bool is_span() const
{
if (size_ == 0) {
return true;
}
return this->is_span_impl();
}
/* Returns the internally used span of the virtual array. This invokes undefined behavior is the
* virtual array is not stored as a span internally. */
Span<T> get_internal_span() const
{
BLI_assert(this->is_span());
if (size_ == 0) {
return {};
}
return this->get_internal_span_impl();
}
/* Returns true when the virtual array returns the same value for every index. */
bool is_single() const
{
if (size_ == 1) {
return true;
}
return this->is_single_impl();
}
/* Returns the value that is returned for every index. This invokes undefined behavior if the
* virtual array would not return the same value for every index. */
T get_internal_single() const
{
BLI_assert(this->is_single());
if (size_ == 1) {
return this->get(0);
}
return this->get_internal_single_impl();
}
/* Get the element at a specific index. Note that this operator cannot be used to assign values
* to an index, because the return value is not a reference. */
T operator[](const int64_t index) const
{
return this->get(index);
}
/* Copy the entire virtual array into a span. */
void materialize(MutableSpan<T> r_span) const
{
this->materialize(IndexMask(size_), r_span);
}
/* Copy some indices of the virtual array into a span. */
void materialize(IndexMask mask, MutableSpan<T> r_span) const
{
BLI_assert(mask.min_array_size() <= size_);
this->materialize_impl(mask, r_span);
}
void materialize_to_uninitialized(MutableSpan<T> r_span) const
{
this->materialize_to_uninitialized(IndexMask(size_), r_span);
}
void materialize_to_uninitialized(IndexMask mask, MutableSpan<T> r_span) const
{
BLI_assert(mask.min_array_size() <= size_);
this->materialize_to_uninitialized_impl(mask, r_span);
}
protected:
virtual T get_impl(const int64_t index) const = 0;
virtual bool is_span_impl() const
{
return false;
}
virtual Span<T> get_internal_span_impl() const
{
BLI_assert_unreachable();
return {};
}
virtual bool is_single_impl() const
{
return false;
}
virtual T get_internal_single_impl() const
{
/* Provide a default implementation, so that subclasses don't have to provide it. This method
* should never be called because `is_single_impl` returns false by default. */
BLI_assert_unreachable();
return T();
}
virtual void materialize_impl(IndexMask mask, MutableSpan<T> r_span) const
{
T *dst = r_span.data();
if (this->is_span()) {
const T *src = this->get_internal_span().data();
mask.foreach_index([&](const int64_t i) { dst[i] = src[i]; });
}
else if (this->is_single()) {
const T single = this->get_internal_single();
mask.foreach_index([&](const int64_t i) { dst[i] = single; });
}
else {
mask.foreach_index([&](const int64_t i) { dst[i] = this->get(i); });
}
}
virtual void materialize_to_uninitialized_impl(IndexMask mask, MutableSpan<T> r_span) const
{
T *dst = r_span.data();
if (this->is_span()) {
const T *src = this->get_internal_span().data();
mask.foreach_index([&](const int64_t i) { new (dst + i) T(src[i]); });
}
else if (this->is_single()) {
const T single = this->get_internal_single();
mask.foreach_index([&](const int64_t i) { new (dst + i) T(single); });
}
else {
mask.foreach_index([&](const int64_t i) { new (dst + i) T(this->get(i)); });
}
}
};
/* Similar to VArray, but the elements are mutable. */
template<typename T> class VMutableArray : public VArray<T> {
public:
VMutableArray(const int64_t size) : VArray<T>(size)
{
}
void set(const int64_t index, T value)
{
BLI_assert(index >= 0);
BLI_assert(index < this->size_);
this->set_impl(index, std::move(value));
}
/* Copy the values from the source span to all elements in the virtual array. */
void set_all(Span<T> src)
{
BLI_assert(src.size() == this->size_);
this->set_all_impl(src);
}
MutableSpan<T> get_internal_span()
{
BLI_assert(this->is_span());
Span<T> span = static_cast<const VArray<T> *>(this)->get_internal_span();
return MutableSpan<T>(const_cast<T *>(span.data()), span.size());
}
protected:
virtual void set_impl(const int64_t index, T value) = 0;
virtual void set_all_impl(Span<T> src)
{
if (this->is_span()) {
const MutableSpan<T> span = this->get_internal_span();
initialized_copy_n(src.data(), this->size_, span.data());
}
else {
const int64_t size = this->size_;
for (int64_t i = 0; i < size; i++) {
this->set(i, src[i]);
}
}
}
};
template<typename T> using VArrayPtr = std::unique_ptr<VArray<T>>;
template<typename T> using VMutableArrayPtr = std::unique_ptr<VMutableArray<T>>;
/**
* A virtual array implementation for a span. Methods in this class are final so that it can be
* devirtualized by the compiler in some cases (e.g. when #devirtualize_varray is used).
*/
template<typename T> class VArray_For_Span : public VArray<T> {
protected:
const T *data_ = nullptr;
public:
VArray_For_Span(const Span<T> data) : VArray<T>(data.size()), data_(data.data())
{
}
protected:
VArray_For_Span(const int64_t size) : VArray<T>(size)
{
}
T get_impl(const int64_t index) const final
{
return data_[index];
}
bool is_span_impl() const final
{
return true;
}
Span<T> get_internal_span_impl() const final
{
return Span<T>(data_, this->size_);
}
};
template<typename T> class VMutableArray_For_MutableSpan : public VMutableArray<T> {
protected:
T *data_ = nullptr;
public:
VMutableArray_For_MutableSpan(const MutableSpan<T> data)
: VMutableArray<T>(data.size()), data_(data.data())
{
}
protected:
VMutableArray_For_MutableSpan(const int64_t size) : VMutableArray<T>(size)
{
}
T get_impl(const int64_t index) const final
{
return data_[index];
}
void set_impl(const int64_t index, T value) final
{
data_[index] = value;
}
bool is_span_impl() const override
{
return true;
}
Span<T> get_internal_span_impl() const override
{
return Span<T>(data_, this->size_);
}
};
/**
* A variant of `VArray_For_Span` that owns the underlying data.
* The `Container` type has to implement a `size()` and `data()` method.
* The `data()` method has to return a pointer to the first element in the continuous array of
* elements.
*/
template<typename Container, typename T = typename Container::value_type>
class VArray_For_ArrayContainer : public VArray_For_Span<T> {
private:
Container container_;
public:
VArray_For_ArrayContainer(Container container)
: VArray_For_Span<T>((int64_t)container.size()), container_(std::move(container))
{
this->data_ = container_.data();
}
};
/**
* A virtual array implementation that returns the same value for every index. This class is final
* so that it can be devirtualized by the compiler in some cases (e.g. when #devirtualize_varray is
* used).
*/
template<typename T> class VArray_For_Single final : public VArray<T> {
private:
T value_;
public:
VArray_For_Single(T value, const int64_t size) : VArray<T>(size), value_(std::move(value))
{
}
protected:
T get_impl(const int64_t UNUSED(index)) const override
{
return value_;
}
bool is_span_impl() const override
{
return this->size_ == 1;
}
Span<T> get_internal_span_impl() const override
{
return Span<T>(&value_, 1);
}
bool is_single_impl() const override
{
return true;
}
T get_internal_single_impl() const override
{
return value_;
}
};
/**
* In many cases a virtual array is a span internally. In those cases, access to individual could
* be much more efficient than calling a virtual method. When the underlying virtual array is not a
* span, this class allocates a new array and copies the values over.
*
* This should be used in those cases:
* - All elements in the virtual array are accessed multiple times.
* - In most cases, the underlying virtual array is a span, so no copy is necessary to benefit
* from faster access.
* - An API is called, that does not accept virtual arrays, but only spans.
*/
template<typename T> class VArray_Span final : public Span<T> {
private:
const VArray<T> &varray_;
Array<T> owned_data_;
public:
VArray_Span(const VArray<T> &varray) : Span<T>(), varray_(varray)
{
this->size_ = varray_.size();
if (varray_.is_span()) {
this->data_ = varray_.get_internal_span().data();
}
else {
owned_data_.~Array();
new (&owned_data_) Array<T>(varray_.size(), NoInitialization{});
varray_.materialize_to_uninitialized(owned_data_);
this->data_ = owned_data_.data();
}
}
};
/**
* Same as VArray_Span, but for a mutable span.
* The important thing to note is that when changing this span, the results might not be
* immediately reflected in the underlying virtual array (only when the virtual array is a span
* internally). The #save method can be used to write all changes to the underlying virtual array,
* if necessary.
*/
template<typename T> class VMutableArray_Span final : public MutableSpan<T> {
private:
VMutableArray<T> &varray_;
Array<T> owned_data_;
bool save_has_been_called_ = false;
bool show_not_saved_warning_ = true;
public:
/* Create a span for any virtual array. This is cheap when the virtual array is a span itself. If
* not, a new array has to be allocated as a wrapper for the underlying virtual array. */
VMutableArray_Span(VMutableArray<T> &varray, const bool copy_values_to_span = true)
: MutableSpan<T>(), varray_(varray)
{
this->size_ = varray_.size();
if (varray_.is_span()) {
this->data_ = varray_.get_internal_span().data();
}
else {
if (copy_values_to_span) {
owned_data_.~Array();
new (&owned_data_) Array<T>(varray_.size(), NoInitialization{});
varray_.materialize_to_uninitialized(owned_data_);
}
else {
owned_data_.reinitialize(varray_.size());
}
this->data_ = owned_data_.data();
}
}
~VMutableArray_Span()
{
if (show_not_saved_warning_) {
if (!save_has_been_called_) {
std::cout << "Warning: Call `save()` to make sure that changes persist in all cases.\n";
}
}
}
/* Write back all values from a temporary allocated array to the underlying virtual array. */
void save()
{
save_has_been_called_ = true;
if (this->data_ != owned_data_.data()) {
return;
}
varray_.set_all(owned_data_);
}
void disable_not_applied_warning()
{
show_not_saved_warning_ = false;
}
};
/**
* This class makes it easy to create a virtual array for an existing function or lambda. The
* `GetFunc` should take a single `index` argument and return the value at that index.
*/
template<typename T, typename GetFunc> class VArray_For_Func final : public VArray<T> {
private:
GetFunc get_func_;
public:
VArray_For_Func(const int64_t size, GetFunc get_func)
: VArray<T>(size), get_func_(std::move(get_func))
{
}
private:
T get_impl(const int64_t index) const override
{
return get_func_(index);
}
void materialize_impl(IndexMask mask, MutableSpan<T> r_span) const override
{
T *dst = r_span.data();
mask.foreach_index([&](const int64_t i) { dst[i] = get_func_(i); });
}
void materialize_to_uninitialized_impl(IndexMask mask, MutableSpan<T> r_span) const override
{
T *dst = r_span.data();
mask.foreach_index([&](const int64_t i) { new (dst + i) T(get_func_(i)); });
}
};
template<typename StructT, typename ElemT, ElemT (*GetFunc)(const StructT &)>
class VArray_For_DerivedSpan : public VArray<ElemT> {
private:
const StructT *data_;
public:
VArray_For_DerivedSpan(const Span<StructT> data) : VArray<ElemT>(data.size()), data_(data.data())
{
}
private:
ElemT get_impl(const int64_t index) const override
{
return GetFunc(data_[index]);
}
void materialize_impl(IndexMask mask, MutableSpan<ElemT> r_span) const override
{
ElemT *dst = r_span.data();
mask.foreach_index([&](const int64_t i) { dst[i] = GetFunc(data_[i]); });
}
void materialize_to_uninitialized_impl(IndexMask mask, MutableSpan<ElemT> r_span) const override
{
ElemT *dst = r_span.data();
mask.foreach_index([&](const int64_t i) { new (dst + i) ElemT(GetFunc(data_[i])); });
}
};
template<typename StructT,
typename ElemT,
ElemT (*GetFunc)(const StructT &),
void (*SetFunc)(StructT &, ElemT)>
class VMutableArray_For_DerivedSpan : public VMutableArray<ElemT> {
private:
StructT *data_;
public:
VMutableArray_For_DerivedSpan(const MutableSpan<StructT> data)
: VMutableArray<ElemT>(data.size()), data_(data.data())
{
}
private:
ElemT get_impl(const int64_t index) const override
{
return GetFunc(data_[index]);
}
void set_impl(const int64_t index, ElemT value) override
{
SetFunc(data_[index], std::move(value));
}
void materialize_impl(IndexMask mask, MutableSpan<ElemT> r_span) const override
{
ElemT *dst = r_span.data();
mask.foreach_index([&](const int64_t i) { dst[i] = GetFunc(data_[i]); });
}
void materialize_to_uninitialized_impl(IndexMask mask, MutableSpan<ElemT> r_span) const override
{
ElemT *dst = r_span.data();
mask.foreach_index([&](const int64_t i) { new (dst + i) ElemT(GetFunc(data_[i])); });
}
};
/**
* Generate multiple versions of the given function optimized for different virtual arrays.
* One has to be careful with nesting multiple devirtualizations, because that results in an
* exponential number of function instantiations (increasing compile time and binary size).
*
* Generally, this function should only be used when the virtual method call overhead to get an
* element from a virtual array is significant.
*/
template<typename T, typename Func>
inline void devirtualize_varray(const VArray<T> &varray, const Func &func, bool enable = true)
{
/* Support disabling the devirtualization to simplify benchmarking. */
if (enable) {
if (varray.is_single()) {
/* `VArray_For_Single` can be used for devirtualization, because it is declared `final`. */
const VArray_For_Single<T> varray_single{varray.get_internal_single(), varray.size()};
func(varray_single);
return;
}
if (varray.is_span()) {
/* `VArray_For_Span` can be used for devirtualization, because it is declared `final`. */
const VArray_For_Span<T> varray_span{varray.get_internal_span()};
func(varray_span);
return;
}
}
func(varray);
}
/**
* Same as `devirtualize_varray`, but devirtualizes two virtual arrays at the same time.
* This is better than nesting two calls to `devirtualize_varray`, because it instantiates fewer
* cases.
*/
template<typename T1, typename T2, typename Func>
inline void devirtualize_varray2(const VArray<T1> &varray1,
const VArray<T2> &varray2,
const Func &func,
bool enable = true)
{
devirtualize_varray(
varray1,
[&](const auto &varray1) {
devirtualize_varray(
varray2, [&](const auto &varray2) { func(varray1, varray2); }, enable);
},
enable);
// /* Support disabling the devirtualization to simplify benchmarking. */
// if (enable) {
// const bool is_span1 = varray1.is_span();
// const bool is_span2 = varray2.is_span();
// const bool is_single1 = varray1.is_single();
// const bool is_single2 = varray2.is_single();
// if (is_span1 && is_span2) {
// const VArray_For_Span<T1> varray1_span{varray1.get_internal_span()};
// const VArray_For_Span<T2> varray2_span{varray2.get_internal_span()};
// func(varray1_span, varray2_span);
// return;
// }
// if (is_span1 && is_single2) {
// const VArray_For_Span<T1> varray1_span{varray1.get_internal_span()};
// const VArray_For_Single<T2> varray2_single{varray2.get_internal_single(), varray2.size()};
// func(varray1_span, varray2_single);
// return;
// }
// if (is_single1 && is_span2) {
// const VArray_For_Single<T1> varray1_single{varray1.get_internal_single(), varray1.size()};
// const VArray_For_Span<T2> varray2_span{varray2.get_internal_span()};
// func(varray1_single, varray2_span);
// return;
// }
// if (is_single1 && is_single2) {
// const VArray_For_Single<T1> varray1_single{varray1.get_internal_single(), varray1.size()};
// const VArray_For_Single<T2> varray2_single{varray2.get_internal_single(), varray2.size()};
// func(varray1_single, varray2_single);
// return;
// }
// }
// /* This fallback is used even when one of the inputs could be optimized. It's probably not
// worth
// * it to optimize just one of the inputs, because then the compiler still has to call into
// * unknown code, which inhibits many compiler optimizations. */
// func(varray1, varray2);
}
} // namespace blender
|