Skip to content

[cudax->libcu++] Move the hierarchy type from cudax to libcu++#6611

Merged
pciolkosz merged 5 commits intoNVIDIA:mainfrom
pciolkosz:move_hierarchy_to_libcudacxx
Nov 17, 2025
Merged

[cudax->libcu++] Move the hierarchy type from cudax to libcu++#6611
pciolkosz merged 5 commits intoNVIDIA:mainfrom
pciolkosz:move_hierarchy_to_libcudacxx

Conversation

@pciolkosz
Copy link
Contributor

The usual moving of APIs. The hierarchy and launch testing is a bit overlapped, so I had to temporarily comment out some tests.

@pciolkosz pciolkosz requested review from a team as code owners November 13, 2025 02:32
@pciolkosz pciolkosz requested a review from fbusato November 13, 2025 02:32
@github-project-automation github-project-automation bot moved this to Todo in CCCL Nov 13, 2025
@cccl-authenticator-app cccl-authenticator-app bot moved this from Todo to In Review in CCCL Nov 13, 2025
@github-actions

This comment has been minimized.

@pciolkosz pciolkosz force-pushed the move_hierarchy_to_libcudacxx branch from cae6dfb to 9b1a990 Compare November 14, 2025 00:31
@pciolkosz pciolkosz requested a review from a team as a code owner November 14, 2025 00:31
@pciolkosz pciolkosz requested a review from alliepiper November 14, 2025 00:31
@pciolkosz pciolkosz requested a review from a team as a code owner November 14, 2025 02:05
@github-actions

This comment has been minimized.

*/
_CCCL_BEGIN_NAMESPACE_CUDA

namespace __detail
Copy link
Contributor

Choose a reason for hiding this comment

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

I'm surprised that we have a __detail namespace in libcu++

Copy link
Contributor

Choose a reason for hiding this comment

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

We commonly do not use that, I would prefer to drop it if possible, but nothing critical


namespace cuda::experimental
{
_CCCL_BEGIN_NAMESPACE_CUDA
Copy link
Contributor

Choose a reason for hiding this comment

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

the review process is hard. The files have been moved and GitHub doesn't mark many lines as under review. I will add the comments below.

Copy link
Contributor

Choose a reason for hiding this comment

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

dim3 should be ::dim3 and fully qualify uint32_t (and similar)

Copy link
Contributor

Choose a reason for hiding this comment

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

_CCCL_HOST_DEVICE should be _CCCL_API

Copy link
Contributor

Choose a reason for hiding this comment

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

noexcept is missing in multiple points

Copy link
Contributor Author

Choose a reason for hiding this comment

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

We will be doing a refactor of the hierarchy files later on, so I would not review it too deeply now

Copy link
Contributor

Choose a reason for hiding this comment

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

that's entirely fine.
Reviewing in this way is hard. Github should allow reviewing moved files...

@github-project-automation github-project-automation bot moved this from In Review to In Progress in CCCL Nov 14, 2025
@pciolkosz pciolkosz force-pushed the move_hierarchy_to_libcudacxx branch 2 times, most recently from 636245b to 85e87fc Compare November 15, 2025 00:10
@pciolkosz pciolkosz force-pushed the move_hierarchy_to_libcudacxx branch from 85e87fc to 73a666d Compare November 15, 2025 00:16
@github-actions
Copy link
Contributor

🥳 CI Workflow Results

🟩 Finished in 1h 53m: Pass: 100%/124 | Total: 1d 01h | Max: 1h 47m | Hits: 97%/230572

See results here.


if ("MSVC" STREQUAL "${CMAKE_CXX_COMPILER_ID}")
# libcudacxx requires dim3 to be usable from a constexpr context, and the CUDART headers require
# __cplusplus to be defined for this to work:
Copy link
Member

Choose a reason for hiding this comment

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

We should file a bug for them to fix that and use the appropriate MSVC dialect detection.

Copy link
Contributor

Choose a reason for hiding this comment

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

I believe that is already fixed

Copy link
Member

@wmaxey wmaxey left a comment

Choose a reason for hiding this comment

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

Approved for the CMake changes.

@github-project-automation github-project-automation bot moved this from In Progress to In Review in CCCL Nov 17, 2025
@pciolkosz pciolkosz merged commit c5a7406 into NVIDIA:main Nov 17, 2025
139 checks passed
@github-project-automation github-project-automation bot moved this from In Review to Done in CCCL Nov 17, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: Done

Development

Successfully merging this pull request may close these issues.

5 participants