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

Jump to bottom

Explicitly compute null count in concatenate APIs #13104

Merged
merged 6 commits into from
Apr 13, 2023
Merged

Explicitly compute null count in concatenate APIs #13104

Show file tree
Hide file tree
Changes from 4 commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
1050bd0
Update concat APIs. Breaks Parquet unexpectedly
vyasr Apr 7, 2023
6ad40a1
Apply suggestions from code review
vyasr Apr 10, 2023
616ba28
Fix style
vyasr Apr 10, 2023
51f5260
Update cpp/src/structs/copying/concatenate.cu
vyasr Apr 10, 2023
f657f6c
Fix style again
vyasr Apr 10, 2023
adbc039
Merge branch 'branch-23.06' into feat/concat_null_count
vyasr Apr 13, 2023
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
8 changes: 7 additions & 1 deletion cpp/src/lists/copying/concatenate.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,7 @@
#include <thrust/transform.h>

#include <memory>
#include <numeric>

namespace cudf {
namespace lists {
Expand Down Expand Up @@ -123,15 +124,20 @@ std::unique_ptr<column> concatenate(host_span<column_view const> columns,
std::any_of(columns.begin(), columns.end(), [](auto const& col) { return col.has_nulls(); });
rmm::device_buffer null_mask = create_null_mask(
total_list_count, has_nulls ? mask_state::UNINITIALIZED : mask_state::UNALLOCATED);
cudf::size_type null_count{0};
if (has_nulls) {
cudf::detail::concatenate_masks(columns, static_cast<bitmask_type*>(null_mask.data()), stream);
null_count =
std::transform_reduce(columns.begin(), columns.end(), 0, std::plus{}, [](auto const& col) {
return col.null_count();
});
}

// assemble into outgoing list column
return make_lists_column(total_list_count,
std::move(offsets),
std::move(data),
has_nulls ? UNKNOWN_NULL_COUNT : 0,
null_count,
std::move(null_mask),
stream,
mr);
Expand Down
17 changes: 10 additions & 7 deletions cpp/src/structs/copying/concatenate.cu
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
/*
* Copyright (c) 2020, NVIDIA CORPORATION.
* Copyright (c) 2020-2023, NVIDIA CORPORATION.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
Expand Down Expand Up @@ -65,17 +65,20 @@ std::unique_ptr<column> concatenate(host_span<column_view const> columns,
std::any_of(columns.begin(), columns.end(), [](auto const& col) { return col.has_nulls(); });
rmm::device_buffer null_mask =
create_null_mask(total_length, has_nulls ? mask_state::UNINITIALIZED : mask_state::UNALLOCATED);
cudf::size_type null_count{0};
if (has_nulls) {
cudf::detail::concatenate_masks(columns, static_cast<bitmask_type*>(null_mask.data()), stream);
null_count = std::transform_reduce(
columns.begin(),
columns.end(),
0,
std::plus{},
[](auto const& col) { return col.null_count(); });
}

// assemble into outgoing list column
return make_structs_column(total_length,
std::move(children),
has_nulls ? UNKNOWN_NULL_COUNT : 0,
std::move(null_mask),
stream,
mr);
return make_structs_column(
total_length, std::move(children), null_count, std::move(null_mask), stream, mr);
}

} // namespace detail
Expand Down