fix: replace numeric_limits::infinity() with literal to fix CUDA 12.8 NVRTC compilation#327
Open
kuishou68 wants to merge 1 commit into
Open
fix: replace numeric_limits::infinity() with literal to fix CUDA 12.8 NVRTC compilation#327kuishou68 wants to merge 1 commit into
kuishou68 wants to merge 1 commit into
Conversation
… NVRTC compilation - cute::numeric_limits<T>::infinity() is not constexpr in CUDA 12.8 NVRTC - Using -1e38f literal as negative infinity workaround (mathematically equivalent for softmax masking) - Fixes JIT compilation failure in smxx_clean_logits kernel on CUDA 12.8 Closes deepseek-ai#295
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
On CUDA 12.8, NVRTC fails to compile
smxx_clean_logits.cuhwith:cuda::std::numeric_limits<T>::infinity()is notconstexprin CUDA 12.8 NVRTC environment.Solution
Replace with raw float literal
-1e38f, which is a true compile-time constant and mathematically equivalent to negative infinity for softmax masking purposes.Testing
Closes #295