Skip to content

reduce attr bitset memory #5

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
LHT129 wants to merge 1 commit into
base: main
Choose a base branch
from
Open

reduce attr bitset memory #5

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
54 changes: 33 additions & 21 deletions src/attr/multi_bitset_manager.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,8 +20,11 @@ namespace vsag {
MultiBitsetManager::MultiBitsetManager(Allocator* allocator,
uint64_t count,
ComputableBitsetType bitset_type)
: allocator_(allocator), count_(count), bitsets_(allocator), bitset_type_(bitset_type) {
bitsets_.resize(count, nullptr);
: allocator_(allocator),
count_(count),
bitsets_(allocator),
bitset_map_(count, -1, allocator),
bitset_type_(bitset_type) {
}

MultiBitsetManager::MultiBitsetManager(Allocator* allocator, uint64_t count)
Expand All @@ -43,53 +46,62 @@ MultiBitsetManager::SetNewCount(uint64_t new_count) {
return;
}
this->count_ = new_count;
this->bitsets_.resize(new_count, nullptr);
this->bitset_map_.resize(new_count, -1);
}

ComputableBitset*
MultiBitsetManager::GetOneBitset(uint64_t id) const {
if (id >= count_) {
return nullptr;
}
return this->bitsets_[id];
auto inner_id = bitset_map_[id];
if (inner_id == -1) {
return nullptr;
}
return this->bitsets_[inner_id];
}

void
MultiBitsetManager::InsertValue(uint64_t id, uint64_t offset, bool value) {
if (id >= count_) {
this->SetNewCount(id + 1);
}
if (this->bitsets_[id] == nullptr) {
this->bitsets_[id] =
ComputableBitset::MakeRawInstance(this->bitset_type_, this->allocator_);
auto inner_id = bitset_map_[id];
if (inner_id == -1) {
inner_id = static_cast<int16_t>(bitsets_.size());
bitset_map_[id] = inner_id;
bitsets_.emplace_back(
Comment on lines +71 to +73
Copy link

@sourcery-ai sourcery-ai bot Jul 30, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

issue (bug_risk): Potential for int16_t overflow in inner_id assignment.

Casting bitsets_.size() to int16_t may overflow if the number of bitsets exceeds int16_t's range. Use a larger integer type or add an overflow check.

Comment on lines +71 to +73
Copy link

@sourcery-ai sourcery-ai bot Jul 30, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

issue (bug_risk): Potential for int16_t overflow if bitsets_ grows large.

If creating more than 32,767 bitsets is possible, switch to a larger integer type for bitset_map_ to prevent overflow issues.

ComputableBitset::MakeRawInstance(this->bitset_type_, this->allocator_));
}
this->bitsets_[id]->Set(static_cast<int64_t>(offset), value);
this->bitsets_[inner_id]->Set(static_cast<int64_t>(offset), value);
}

void
MultiBitsetManager::Serialize(StreamWriter& writer) {
StreamWriter::WriteObj(writer, count_);
for (auto id : bitset_map_) {
StreamWriter::WriteObj(writer, id);
}
uint64_t size = bitsets_.size();
StreamWriter::WriteObj(writer, size);
for (auto* bitset : bitsets_) {
if (bitset == nullptr) {
StreamWriter::WriteObj(writer, false);
} else {
StreamWriter::WriteObj(writer, true);
bitset->Serialize(writer);
}
bitset->Serialize(writer);
}
}

void
MultiBitsetManager::Deserialize(lvalue_or_rvalue<StreamReader> reader) {
StreamReader::ReadObj(reader, count_);
this->bitsets_.resize(count_, nullptr);
this->bitset_map_.resize(count_, -1);
for (uint64_t i = 0; i < count_; i++) {
bool has_bitset;
StreamReader::ReadObj(reader, has_bitset);
if (has_bitset) {
bitsets_[i] = ComputableBitset::MakeRawInstance(bitset_type_, allocator_);
bitsets_[i]->Deserialize(reader);
}
StreamReader::ReadObj(reader, this->bitset_map_[i]);
}
uint64_t size;
StreamReader::ReadObj(reader, size);
bitsets_.resize(size, nullptr);
for (uint64_t i = 0; i < size; i++) {
bitsets_[i] = ComputableBitset::MakeRawInstance(bitset_type_, allocator_);
bitsets_[i]->Deserialize(reader);
}
}

Expand Down
3 changes: 3 additions & 0 deletions src/attr/multi_bitset_manager.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -116,6 +116,9 @@ class MultiBitsetManager {
/// A vector containing pointers to ComputableBitset instances.
Vector<ComputableBitset*> bitsets_;

/// map origin id to inner id (avoiding nullptr)
Vector<int16_t> bitset_map_;

/// The current number of ComputableBitset instances managed by this object. Defaults to 1.
uint64_t count_{1};

Expand Down