Skip to content

[CUB] Fix DeviceAdjacentDifference for cuda::device_buffer iterators #9861

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
Noperi0r wants to merge 4 commits into
base: main
Choose a base branch
from
+18 −1
Open

[CUB] Fix DeviceAdjacentDifference for cuda::device_buffer iterators #9861

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1b058db
Fix DeviceAdjacentDifference for device_buffer iterators [skip-docs]
Noperi0r Jul 14, 2026
facc0d5
Move DeviceAdjacentDifference device_buffer test [skip-docs]
Noperi0r Jul 15, 2026
7aed3a4
Make AdjacentDifference output pointer binding const [skip-docs]
Noperi0r Jul 15, 2026
8c1fa4f
Use launch wrapper in device_buffer regression test [skip-docs]
Noperi0r Jul 15, 2026
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
2 changes: 1 addition & 1 deletion cub/cub/agent/agent_adjacent_difference.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -105,7 +105,7 @@ struct AgentDifference
OffsetT num_items)
: temp_storage(temp_storage.Alias())
, input_it(input_it)
, load_it(LoadIt(input_it))
, load_it(try_make_cache_modified_iterator<Policy::LOAD_MODIFIER>(input_it))
, first_tile_previous(first_tile_previous)
, result(result)
, difference_op(difference_op)
Expand Down
35 changes: 35 additions & 0 deletions cub/test/catch2_test_device_adjacent_difference_api.cu
View file
Open in desktop

@bernhardmgruber bernhardmgruber Jul 14, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Important: Please move the new test into the file catch2_test_device_adjacent_difference_substract_left.cu. The _api.cu files cover tested API examples for the documentation.

@Noperi0r Noperi0r Jul 15, 2026

Copy link
Copy Markdown
Author

Choose a reason for hiding this comment

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

Moved the regression test to
catch2_test_device_adjacent_difference_substract_left.cu
and restored catch2_test_device_adjacent_difference_api.cu to its previous scope.

The relocated focused test, target build, and targeted pre-commit checks pass locally.

Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,12 @@

#include <cub/device/device_adjacent_difference.cuh>

#include <cuda/buffer>
#include <cuda/devices>
#include <cuda/std/functional>

#include <cstdint>

#include <c2h/catch2_test_helper.h>

// Guard: the legacy memory-size query call with all defaults (no explicit difference_op,
Expand Down Expand Up @@ -51,4 +57,33 @@ C2H_TEST("DeviceAdjacentDifference::SubtractRight legacy size-query is unambiguo
REQUIRE(cudaSuccess == cub::DeviceAdjacentDifference::SubtractRight(nullptr, bytes, d_in, n));
}

C2H_TEST("DeviceAdjacentDifference::SubtractLeftCopy accepts cuda::device_buffer input",
"[adjacent_difference][device]")
{
using type = std::int32_t;

cuda::stream stream{cuda::devices[0]};
auto input = cuda::make_device_buffer<type>(stream, cuda::devices[0], {2, 5, 9, 14, 20});
c2h::device_vector<type> output(input.size());

@bernhardmgruber bernhardmgruber Jul 14, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Suggested change
c2h::device_vector<type> output(input.size());
c2h::device_vector<type> output(input.size(), thrust::no_init);

@Noperi0r Noperi0r Jul 15, 2026

Copy link
Copy Markdown
Author

Choose a reason for hiding this comment

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

Applied thrust::no_init to the output allocation in the relocated test.

auto output_it = thrust::raw_pointer_cast(output.data());

void* temporary_storage = nullptr;
std::size_t temporary_bytes = 0;
REQUIRE(
cudaSuccess
== cub::DeviceAdjacentDifference::SubtractLeftCopy(
temporary_storage, temporary_bytes, input.begin(), output_it, input.size(), cuda::std::minus<>{}, stream.get()));

c2h::device_vector<std::uint8_t> temporary_buffer(temporary_bytes);

@bernhardmgruber bernhardmgruber Jul 14, 2026

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Suggested change
c2h::device_vector<std::uint8_t> temporary_buffer(temporary_bytes);
c2h::device_vector<std::uint8_t> temporary_buffer(temporary_bytes, thrust::no_init);

@Noperi0r Noperi0r Jul 15, 2026

Copy link
Copy Markdown
Author

Choose a reason for hiding this comment

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

Applied thrust::no_init to the temporary-storage allocation in the relocated test.

temporary_storage = thrust::raw_pointer_cast(temporary_buffer.data());
REQUIRE(
cudaSuccess
== cub::DeviceAdjacentDifference::SubtractLeftCopy(
temporary_storage, temporary_bytes, input.begin(), output_it, input.size(), cuda::std::minus<>{}, stream.get()));
stream.sync();

const c2h::host_vector<type> expected{2, 3, 4, 5, 6};
REQUIRE(output == expected);
}

// todo(giannis): extract examples from the docs to literalinclude extracts here
Loading