Skip to content

Conversation

@abhilash1910
Copy link
Contributor

@abhilash1910 abhilash1910 commented Sep 26, 2024

Used in llm.c
warp.meta_group_size()
cc @ShengchenJ @zhimingwang36 @yihanwg

@abhilash1910 abhilash1910 requested a review from a team as a code owner September 26, 2024 09:07
Copy link
Contributor

@ShengchenJ ShengchenJ left a comment

Choose a reason for hiding this comment

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

LGTM

@abhilash1910 abhilash1910 reopened this Nov 25, 2024
cg::tiled_partition<2>(block).meta_group_size();
// CHECK: catile2.get_local_linear_range();
catile2.meta_group_size();
// CHECK: atile2.get_local_linear_range();
Copy link
Contributor

Choose a reason for hiding this comment

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

the meta_group_size means the group number after the partition in the CUDA docs(https://docs.nvidia.com/cuda/cuda-c-programming-guide/#coalesced-groups) but get_local_linear_range means the size of the work items in partitioned group.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Yes mapping is changed. Please review , thanks

@zhimingwang36 zhimingwang36 merged commit cbbd427 into oneapi-src:SYCLomatic Nov 28, 2024
1 check passed
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.

3 participants