Skip to content

Use ck::half_t for Host Reduction#195

Merged
asroy merged 3 commits into
developfrom
uniform_half
Apr 21, 2022
Merged

Use ck::half_t for Host Reduction#195
asroy merged 3 commits into
developfrom
uniform_half

Conversation

@qianfengz
Copy link
Copy Markdown
Contributor

This P.R is to remove the using of half_float::half in Host Reduction so that ck::half_t is used by Reduction on both the kernel and host. Some host math functions are added to support ck::half on host.

@qianfengz qianfengz requested a review from asroy April 20, 2022 16:05
@@ -245,7 +222,7 @@ binop_with_nan_check2(std::function<void(AccDataType&, AccDataType, bool&)> opRe
int& accuIndex,
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.

Is this tensor index? If so, index_t should be used.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

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

Yes, need correction

@@ -245,7 +222,7 @@ binop_with_nan_check2(std::function<void(AccDataType&, AccDataType, bool&)> opRe
int& accuIndex,
int currIndex)
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.

same here

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants