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This repository has been archived by the owner on Mar 21, 2024.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

Fix main/GitHub/warp reduce #516

Merged
merged 3 commits into from
Jul 26, 2022

Conversation

gevtushenko
Copy link
Collaborator

This PR fixes the issue of single-threaded warp reduces. It also addresses the issue which requested hardware acceleration for reductions on ampere.

@gevtushenko gevtushenko requested a review from alliepiper June 24, 2022 22:31
@gevtushenko gevtushenko added this to the 2.0.0 milestone Jun 24, 2022
gevtushenko added a commit to gevtushenko/thrust that referenced this pull request Jun 24, 2022
gevtushenko added a commit to gevtushenko/thrust that referenced this pull request Jun 25, 2022
@gevtushenko gevtushenko added the testing: gpuCI passed Passed gpuCI testing. label Jun 25, 2022
Copy link
Collaborator

@brycelelbach brycelelbach left a comment

Choose a reason for hiding this comment

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

Looks good to me.

cub/warp/specializations/warp_reduce_shfl.cuh Outdated Show resolved Hide resolved
cub/warp/warp_reduce.cuh Outdated Show resolved Hide resolved
@gevtushenko gevtushenko force-pushed the fix-main/github/warp_reduce branch from 738be7d to 91963c8 Compare July 25, 2022 15:29
gevtushenko added a commit to gevtushenko/thrust that referenced this pull request Jul 25, 2022
@gevtushenko gevtushenko merged commit 728a2a2 into NVIDIA:main Jul 26, 2022
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
testing: gpuCI passed Passed gpuCI testing.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Single thread warp reduce Abstraction for Compute 8.0 warp reduce functions in CUB
2 participants