[TPU][Pallas]Fix example/cross_entropy.py on Pallas TPU#2002
Open
yarongmu-google wants to merge 9 commits intopytorch:mainfrom
Open
[TPU][Pallas]Fix example/cross_entropy.py on Pallas TPU#2002yarongmu-google wants to merge 9 commits intopytorch:mainfrom
yarongmu-google wants to merge 9 commits intopytorch:mainfrom
Conversation
…errors and fix zero division in block size calculation
…py.py to avoid unaligned HBM gather This optimizes the cross_entropy kernel to be hardware agnostic. By calculating the target logits via a boolean mask over the streaming dense block, it stays entirely within TensorCore/VMEM boundaries on TPU and perfectly coalesced on GPU, eliminating the unaligned 1D HBM gather which Pallas TC kernels do not natively support without SC DMA staging.
meta-cla
bot
added
the
CLA Signed
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The kernels currently has 2 common issues that need support:
Add CI workflow #2 is the bigger fix here.
After this PR:
Benchmark Results
Implementation Time (ms) Speedup
helion 0.3826 1.10x
torch 0.4208 1.00x (ref)