Fix permute_multi_embedding kernel #3227
Closed
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
X-link: https://github.com/facebookresearch/FBGEMM/pull/325
Looks like a typo to use
permute_id = threadIdx.y + blockIdx.x * blockDim.x
which should beblockDim.y
. This doesn't affect Nvidia because blockDim.x and y are both 32 (32 threads per warp + 32 warps). For AMD GPU, blockDim.x is 64 and blockDim.y is 16, causing numerical issues.Reviewed By: leitian, jianyuh, joebos
Differential Revision: D63936776