blender
/
blender
136
406
Fork 589
Code Issues 6.2k Pull Requests 758 Projects 19 Wiki Activity

Cleanup: Slightly improve virtual array implementation consistency 

Previously the base virtual array implementation optimized for
common cases where data is stored as spans or single values.
However, that didn't make sense when there are already
sub-classes that handle those cases specifically. Instead,
implement the faster materialize methods for each class.
Now, if the base class is reached, it means no optimizations
for avoiding virtual function call overhead are used.

Differential Revision: https://developer.blender.org/D15549
This commit is contained in:
Hans Goudey 2022-08-28 14:40:49 -05:00
parent 67f3259c54
commit e0414070d9
1 changed files with 29 additions and 43 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

72
source/blender/blenlib/BLI_virtual_array.hh
View File

@ -108,25 +108,7 @@ template<typename T> class VArrayImpl {
*/
virtual void materialize(IndexMask mask, MutableSpan<T> r_span) const
{
T *dst = r_span.data();
/* Optimize for a few different common cases. */
const CommonVArrayInfo info = this->common_info();
switch (info.type) {
case CommonVArrayInfo::Type::Any: {
mask.foreach_index([&](const int64_t i) { dst[i] = this->get(i); });
break;
}
case CommonVArrayInfo::Type::Span: {
const T *src = static_cast<const T *>(info.data);
mask.foreach_index([&](const int64_t i) { dst[i] = src[i]; });
break;
}
case CommonVArrayInfo::Type::Single: {
const T single = *static_cast<const T *>(info.data);
mask.foreach_index([&](const int64_t i) { dst[i] = single; });
break;
}
}
mask.foreach_index([&](const int64_t i) { r_span[i] = this->get(i); });
}
/**
@ -135,24 +117,7 @@ template<typename T> class VArrayImpl {
virtual void materialize_to_uninitialized(IndexMask mask, MutableSpan<T> r_span) const
{
T *dst = r_span.data();
/* Optimize for a few different common cases. */
const CommonVArrayInfo info = this->common_info();
switch (info.type) {
case CommonVArrayInfo::Type::Any: {
mask.foreach_index([&](const int64_t i) { new (dst + i) T(this->get(i)); });
break;
}
case CommonVArrayInfo::Type::Span: {
const T *src = static_cast<const T *>(info.data);
mask.foreach_index([&](const int64_t i) { new (dst + i) T(src[i]); });
break;
}
case CommonVArrayInfo::Type::Single: {
const T single = *static_cast<const T *>(info.data);
mask.foreach_index([&](const int64_t i) { new (dst + i) T(single); });
break;
}
}
mask.foreach_index([&](const int64_t i) { new (dst + i) T(this->get(i)); });
}
/**
@ -288,8 +253,20 @@ template<typename T> class VArrayImpl_For_Span : public VMutableArrayImpl<T> {
return data_ == static_cast<const T *>(other_info.data);
}
void materialize(IndexMask mask, MutableSpan<T> r_span) const override
{
mask.foreach_index([&](const int64_t i) { r_span[i] = data_[i]; });
}
void materialize_to_uninitialized(IndexMask mask, MutableSpan<T> r_span) const override
{
T *dst = r_span.data();
mask.foreach_index([&](const int64_t i) { new (dst + i) T(data_[i]); });
}
void materialize_compressed(IndexMask mask, MutableSpan<T> r_span) const override
{
BLI_assert(mask.size() == r_span.size());
mask.to_best_mask_type([&](auto best_mask) {
for (const int64_t i : IndexRange(best_mask.size())) {
r_span[i] = data_[best_mask[i]];
@ -300,6 +277,7 @@ template<typename T> class VArrayImpl_For_Span : public VMutableArrayImpl<T> {
void materialize_compressed_to_uninitialized(IndexMask mask,
MutableSpan<T> r_span) const override
{
BLI_assert(mask.size() == r_span.size());
T *dst = r_span.data();
mask.to_best_mask_type([&](auto best_mask) {
for (const int64_t i : IndexRange(best_mask.size())) {
@ -378,6 +356,17 @@ template<typename T> class VArrayImpl_For_Single final : public VArrayImpl<T> {
return CommonVArrayInfo(CommonVArrayInfo::Type::Single, true, &value_);
}
void materialize(IndexMask mask, MutableSpan<T> r_span) const override
{
r_span.fill_indices(mask, value_);
}
void materialize_to_uninitialized(IndexMask mask, MutableSpan<T> r_span) const override
{
T *dst = r_span.data();
mask.foreach_index([&](const int64_t i) { new (dst + i) T(value_); });
}
void materialize_compressed(IndexMask mask, MutableSpan<T> r_span) const override
{
BLI_assert(mask.size() == r_span.size());
@ -887,12 +876,9 @@ template<typename T> class VMutableArray;
* construct the virtual array first and then move it into the vector.
*/
namespace varray_tag {
struct span {
};
struct single_ref {
};
struct single {
};
struct span {};
struct single_ref {};
struct single {};
} // namespace varray_tag
/**