Skip to content
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

GH-38090: [C++][Emscripten] acero/swiss_join: Suppress shorten-64-to-32 warnings #38094

Closed
wants to merge 1 commit into from
Closed

GH-38090: [C++][Emscripten] acero/swiss_join: Suppress shorten-64-to-32 warnings #38094

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
65 changes: 38 additions & 27 deletions cpp/src/arrow/acero/swiss_join.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -405,7 +405,7 @@ void RowArray::DebugPrintToFile(const char* filename, bool print_sorted) const {
return;
}
std::vector<char> buffer;
buffer.resize(sb.st_size);
buffer.resize(static_cast<size_t>(sb.st_size));
std::vector<std::string> lines;
FILE* fin;
#if defined(_MSC_VER) && _MSC_VER >= 1400
Expand Down Expand Up @@ -538,7 +538,7 @@ void RowArrayMerge::CopyFixedLength(RowTableImpl* target, const RowTableImpl& so
//
if (!source_rows_permutation) {
memcpy(target->mutable_data(1) + fixed_length * first_target_row_id, source.data(1),
fixed_length * num_source_rows);
static_cast<size_t>(fixed_length * num_source_rows));
} else {
// Row length must be a multiple of 64-bits due to enforced alignment.
// Loop for each output row copying a fixed number of 64-bit words.
Expand Down Expand Up @@ -617,7 +617,8 @@ void RowArrayMerge::CopyNulls(RowTableImpl* target, const RowTableImpl& source,
int num_bytes_per_row = target->metadata().null_masks_bytes_per_row;
uint8_t* target_nulls = target->null_masks() + num_bytes_per_row * first_target_row_id;
if (!source_rows_permutation) {
memcpy(target_nulls, source.null_masks(), num_bytes_per_row * num_source_rows);
memcpy(target_nulls, source.null_masks(),
static_cast<size_t>(num_bytes_per_row * num_source_rows));
} else {
for (int64_t i = 0; i < num_source_rows; ++i) {
int64_t source_row_id = source_rows_permutation[i];
Expand Down Expand Up @@ -873,7 +874,7 @@ void SwissTableWithKeys::EqualCallback(int num_keys, const uint16_t* selection_m
auto selection_to_use_buf =
arrow::util::TempVectorHolder<uint16_t>(in->temp_stack, num_keys);
ARROW_DCHECK(in->temp_group_ids);
in->temp_group_ids->resize(in->batch->length);
in->temp_group_ids->resize(static_cast<size_t>(in->batch->length));

if (selection_maybe_null) {
for (int i = 0; i < num_keys; ++i) {
Expand Down Expand Up @@ -1069,14 +1070,16 @@ uint8_t* SwissTableForJoin::local_has_match(int64_t thread_id) {
return nullptr;
}

ThreadLocalState& local_state = local_states_[thread_id];
ThreadLocalState& local_state = local_states_[static_cast<size_t>(thread_id)];
if (local_state.has_match.empty() && num_rows_hash_table > 0) {
local_state.has_match.resize(bit_util::BytesForBits(num_rows_hash_table) +
sizeof(uint64_t));
memset(local_state.has_match.data(), 0, bit_util::BytesForBits(num_rows_hash_table));
local_state.has_match.resize(
static_cast<size_t>(bit_util::BytesForBits(num_rows_hash_table)) +
sizeof(uint64_t));
memset(local_state.has_match.data(), 0,
static_cast<size_t>(bit_util::BytesForBits(num_rows_hash_table)));
}

return local_states_[thread_id].has_match.data();
return local_states_[static_cast<size_t>(thread_id)].has_match.data();
}

void SwissTableForJoin::UpdateHasMatchForKeys(int64_t thread_id, int num_ids,
Expand All @@ -1098,8 +1101,10 @@ void SwissTableForJoin::MergeHasMatch() {
return;
}

has_match_.resize(bit_util::BytesForBits(num_rows_hash_table) + sizeof(uint64_t));
memset(has_match_.data(), 0, bit_util::BytesForBits(num_rows_hash_table));
has_match_.resize(static_cast<size_t>(bit_util::BytesForBits(num_rows_hash_table)) +
sizeof(uint64_t));
memset(has_match_.data(), 0,
static_cast<size_t>(bit_util::BytesForBits(num_rows_hash_table)));

for (size_t tid = 0; tid < local_states_.size(); ++tid) {
if (!local_states_[tid].has_match.empty()) {
Expand Down Expand Up @@ -1202,11 +1207,11 @@ Status SwissTableForJoinBuild::PushNextBatch(int64_t thread_id,
const ExecBatch* payload_batch_maybe_null,
arrow::util::TempVectorStack* temp_stack) {
ARROW_DCHECK(thread_id < dop_);
ThreadState& locals = thread_states_[thread_id];
ThreadState& locals = thread_states_[static_cast<size_t>(thread_id)];

// Compute hash
//
locals.batch_hashes.resize(key_batch.length);
locals.batch_hashes.resize(static_cast<size_t>(key_batch.length));
RETURN_NOT_OK(Hashing32::HashBatch(
key_batch, locals.batch_hashes.data(), locals.temp_column_arrays, hardware_flags_,
temp_stack, /*start_row=*/0, static_cast<int>(key_batch.length)));
Expand Down Expand Up @@ -1234,7 +1239,8 @@ Status SwissTableForJoinBuild::PushNextBatch(int64_t thread_id,
// We want each partition to correspond to a range of block indices,
// so we also partition on the highest bits of the hash.
//
return locals.batch_hashes[i] >> (31 - log_num_prtns_) >> 1;
return locals.batch_hashes[static_cast<size_t>(i)] >> (31 - log_num_prtns_) >>
1;
},
[&locals](int64_t i, int pos) {
locals.batch_prtn_row_ids[pos] = static_cast<uint16_t>(i);
Expand All @@ -1255,7 +1261,7 @@ Status SwissTableForJoinBuild::PushNextBatch(int64_t thread_id,
locals.temp_prtn_ids.resize(num_prtns_);

RETURN_NOT_OK(prtn_locks_.ForEachPartition(
thread_id, locals.temp_prtn_ids.data(),
static_cast<size_t>(thread_id), locals.temp_prtn_ids.data(),
/*is_prtn_empty_fn=*/
[&](int prtn_id) {
return locals.batch_prtn_ranges[prtn_id + 1] == locals.batch_prtn_ranges[prtn_id];
Expand All @@ -1275,7 +1281,7 @@ Status SwissTableForJoinBuild::ProcessPartition(int64_t thread_id,
arrow::util::TempVectorStack* temp_stack,
int prtn_id) {
ARROW_DCHECK(thread_id < dop_);
ThreadState& locals = thread_states_[thread_id];
ThreadState& locals = thread_states_[static_cast<size_t>(thread_id)];

int num_rows_new =
locals.batch_prtn_ranges[prtn_id + 1] - locals.batch_prtn_ranges[prtn_id];
Expand Down Expand Up @@ -1370,14 +1376,16 @@ Status SwissTableForJoinBuild::PreparePrtnMerge() {
//
target_->no_duplicate_keys_ = no_duplicate_keys;
if (!no_duplicate_keys) {
target_->row_offset_for_key_.resize(num_keys + 1);
target_->row_offset_for_key_.resize(static_cast<size_t>(num_keys + 1));
int64_t num_rows = 0;
for (int i = 0; i < num_prtns_; ++i) {
for (size_t i = 0; i < static_cast<size_t>(num_prtns_); ++i) {
int64_t first_key = partition_keys_first_row_id_[i];
target_->row_offset_for_key_[first_key] = static_cast<uint32_t>(num_rows);
target_->row_offset_for_key_[static_cast<size_t>(first_key)] =
static_cast<uint32_t>(num_rows);
num_rows += static_cast<int64_t>(prtn_states_[i].key_ids.size());
}
target_->row_offset_for_key_[num_keys] = static_cast<uint32_t>(num_rows);
target_->row_offset_for_key_[static_cast<size_t>(num_keys)] =
static_cast<uint32_t>(num_rows);
}

return Status::OK();
Expand Down Expand Up @@ -1449,7 +1457,7 @@ void SwissTableForJoinBuild::PrtnMerge(int prtn_id) {
for (size_t i = 0; i < prtn_state.key_ids.size(); ++i) {
uint32_t key_id = prtn_state.key_ids[i];
int64_t position = --counters[key_id];
source_payload_ids[position] = static_cast<int64_t>(i);
source_payload_ids[static_cast<size_t>(position)] = static_cast<int64_t>(i);
}
// Add base payload id to all of the counters.
//
Expand Down Expand Up @@ -1764,7 +1772,8 @@ void JoinNullFilter::Filter(const ExecBatch& key_batch, int batch_start_row,
// Filter out nulls for this column
//
if (!is_output_initialized) {
memset(inout_bit_vector, 0xff, bit_util::BytesForBits(num_batch_rows));
memset(inout_bit_vector, 0xff,
static_cast<size_t>(bit_util::BytesForBits(num_batch_rows)));
is_output_initialized = true;
}
arrow::internal::BitmapAnd(inout_bit_vector, 0, non_null_buffer, offset,
Expand Down Expand Up @@ -1950,7 +1959,7 @@ Status JoinProbeProcessor::OnNextBatch(int64_t thread_id,
static_cast<uint16_t>(minibatch_start);
}

RETURN_NOT_OK(materialize_[thread_id]->AppendProbeOnly(
RETURN_NOT_OK(materialize_[static_cast<size_t>(thread_id)]->AppendProbeOnly(
keypayload_batch, num_passing_ids, materialize_batch_ids_buf.mutable_data(),
[&](ExecBatch batch) {
return output_batch_fn_(thread_id, std::move(batch));
Expand Down Expand Up @@ -1990,7 +1999,7 @@ Status JoinProbeProcessor::OnNextBatch(int64_t thread_id,
// Call materialize for resulting id tuples pointing to matching pairs
// of rows.
//
RETURN_NOT_OK(materialize_[thread_id]->Append(
RETURN_NOT_OK(materialize_[static_cast<size_t>(thread_id)]->Append(
keypayload_batch, num_matches_next, materialize_batch_ids,
materialize_key_ids, materialize_payload_ids, [&](ExecBatch batch) {
return output_batch_fn_(thread_id, std::move(batch));
Expand All @@ -2016,7 +2025,7 @@ Status JoinProbeProcessor::OnNextBatch(int64_t thread_id,
static_cast<uint16_t>(minibatch_start);
}

RETURN_NOT_OK(materialize_[thread_id]->AppendProbeOnly(
RETURN_NOT_OK(materialize_[static_cast<size_t>(thread_id)]->AppendProbeOnly(
keypayload_batch, num_passing_ids, materialize_batch_ids_buf.mutable_data(),
[&](ExecBatch batch) {
return output_batch_fn_(thread_id, std::move(batch));
Expand Down Expand Up @@ -2219,7 +2228,9 @@ class SwissJoin : public HashJoinImpl {

ExecBatch input_batch;
ARROW_ASSIGN_OR_RAISE(
input_batch, KeyPayloadFromInput(/*side=*/1, &build_side_batches_[batch_id]));
input_batch,
KeyPayloadFromInput(/*side=*/1,
&build_side_batches_[static_cast<size_t>(batch_id)]));

if (input_batch.length == 0) {
return Status::OK();
Expand Down Expand Up @@ -2302,7 +2313,7 @@ class SwissJoin : public HashJoinImpl {
}
hash_table_ready_.store(true);

return build_finished_callback_(thread_id);
return build_finished_callback_(static_cast<size_t>(thread_id));
}

Status StartScanHashTable(int64_t thread_id) {
Expand Down
3 changes: 2 additions & 1 deletion cpp/src/arrow/acero/swiss_join_internal.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -381,7 +381,8 @@ class SwissTableForJoin {
}
int64_t num_keys() const { return map_.keys()->num_rows(); }
int64_t num_rows() const {
return no_duplicate_keys_ ? num_keys() : row_offset_for_key_[num_keys()];
return no_duplicate_keys_ ? num_keys()
: row_offset_for_key_[static_cast<size_t>(num_keys())];
}

uint32_t payload_id_to_key_id(uint32_t payload_id) const;
Expand Down
Loading