Skip to content
This repository has been archived by the owner on Mar 21, 2024. It is now read-only.

Cleanup CTK version checks #630

Merged
merged 3 commits into from
Mar 6, 2023

Conversation

gevtushenko
Copy link
Collaborator

This PR fixes the int128 usage for CTK < 11.5 and also cleans up a few outdated checks (CTK < 11.0).

@gevtushenko gevtushenko linked an issue Feb 18, 2023 that may be closed by this pull request
@gevtushenko gevtushenko added the type: bug: functional Does not work as intended. label Feb 18, 2023
gevtushenko added a commit to gevtushenko/thrust that referenced this pull request Feb 18, 2023
@gevtushenko gevtushenko added the testing: gpuCI in progress Started gpuCI testing. label Feb 18, 2023
#if (__CUDACC_VER_MAJOR__ >= 11 || CUDA_VERSION >= 11000) && !_NVHPC_CUDA && \
!defined(CUB_DISABLE_BF16_SUPPORT)
#include <cuda_bf16.h>
#if !_NVHPC_CUDA && !defined(CUB_DISABLE_BF16_SUPPORT)
Copy link
Collaborator

Choose a reason for hiding this comment

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

Does bf16 not work on nvc++?

If so, would it make sense to just add the _NVHPC_CUDA check into CUB_DISABLE_BF16_SUPPORT?

Copy link
Collaborator

@jrhemstad jrhemstad left a comment

Choose a reason for hiding this comment

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

I left some ideas about simplifying things even more, but they are entirely optional.

@jrhemstad jrhemstad changed the base branch from main to 2.1.x February 27, 2023 18:06
gevtushenko added a commit to gevtushenko/thrust that referenced this pull request Mar 5, 2023
@gevtushenko gevtushenko merged commit b4f7e23 into NVIDIA:2.1.x Mar 6, 2023
@gevtushenko gevtushenko added testing: gpuCI passed Passed gpuCI testing. and removed testing: gpuCI in progress Started gpuCI testing. labels Mar 6, 2023
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
testing: gpuCI passed Passed gpuCI testing. type: bug: functional Does not work as intended.
Projects
Archived in project
Development

Successfully merging this pull request may close these issues.

Compiler error with CUDA <11.5 due to int128
3 participants