[CUDA] Remove htanh from unsupported math ops for CUDA 12.8 #17639
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This PR removes htanh from the list of unsupported CUDA half operators, as it is started to be supported since CUDA 12.8. Specifically, we added a CUDA version check in the generated CUDA code, so that when the CUDA version is older than 12.8, htanh will still be treated as an unsupported operator and fall back to the packed operation. While for newer CUDA versions, we directly use the function that is defined in `cuda_fp16.h`.
MasterJH5574
force-pushed
the
tvm-dev/2025-02-11-cuda-htanh
branch
from
3a52974 to
8226c81
Compare
tqchen
approved these changes
Feb 11, 2025
ShiboXing
pushed a commit
to ShiboXing/tvm
that referenced
this pull request
Aug 10, 2025
…7639) This PR removes htanh from the list of unsupported CUDA half operators, as it is started to be supported since CUDA 12.8. Specifically, we added a CUDA version check in the generated CUDA code, so that when the CUDA version is older than 12.8, htanh will still be treated as an unsupported operator and fall back to the packed operation. While for newer CUDA versions, we directly use the function that is defined in `cuda_fp16.h`.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.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.
This PR removes htanh from the list of unsupported CUDA half operators, as it is started to be supported since CUDA 12.8 (https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html#cuda-math-release-12-8).
Specifically, we added a CUDA version check in the generated CUDA code, so that when the CUDA version is older than 12.8, htanh will still be treated as an unsupported operator and fall back to the packed operation. While for newer CUDA versions, we directly use the function that is defined in
cuda_fp16.h.