Skip to content
This repository was archived by the owner on May 9, 2024. It is now read-only.

Add min/max implementations for L0 path #487

Merged
merged 1 commit into from
May 23, 2023
Merged

Conversation

kurapov-peter
Copy link
Contributor

The cas and xchg are there to be used by get_value_group implementation that will come later.
Fixes Select.GroupByPerfectHash and most of the Select.GroupByBaselineHash.

@kurapov-peter kurapov-peter force-pushed the pakurapo/shared-min-max branch from 89fde8d to be9257d Compare May 22, 2023 11:05
Copy link
Contributor

@alexbaden alexbaden left a comment

Choose a reason for hiding this comment

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

I have to say, I like the CUDA C++ definition of specialized device functions (like atomics) better. Is that a possibility for the genx.ll file in the future, either in DPC++ or some C/C++ API?

@kurapov-peter
Copy link
Contributor Author

Sure, we can try using, for example, these https://github.com/oneapi-src/DPCPP_Reference/blob/dpcpp/reference/headers/atomicoperations.h once all the deps conflicts with dpcpp are resolved.

@kurapov-peter kurapov-peter merged commit 61dac61 into main May 23, 2023
@kurapov-peter kurapov-peter deleted the pakurapo/shared-min-max branch May 23, 2023 11:41
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants