Skip to content

fix complile errors.#4289

Closed
awan-10 wants to merge 1 commit intomasterfrom
amawa/fix-amd-rocm
Closed

fix complile errors.#4289
awan-10 wants to merge 1 commit intomasterfrom
amawa/fix-amd-rocm

Conversation

@awan-10
Copy link
Contributor

@awan-10 awan-10 commented Sep 7, 2023

No description provided.



unsigned meta_group_rank = (threadIdx.x + (threadIdx.y + threadIdx.z * blockDim.y) * blockDim.x) % hw_warp_size;
unsigned meta_group_size = (blockDim.x * blockDim.y * blockDim.z) / hw_warp_size;
Copy link
Contributor

@RezaYazdaniAminabadi RezaYazdaniAminabadi Sep 7, 2023

Choose a reason for hiding this comment

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

I think this (meta_group_size ) is just hw_warp_size!

@awan-10
Copy link
Contributor Author

awan-10 commented Sep 7, 2023

@RezaYazdaniAminabadi - closing this PR in favor of Connor's PR (#3180). Both have similar solutions but Connor's one is working for end to end generation.

@awan-10 awan-10 closed this Sep 7, 2023
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