[PHI] Fix performance issue in bilinear interpolation's backward kernel. #68541
Conversation
This file contains 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
|
你的PR提交成功，感谢你对开源项目的贡献! |
|
Sorry to inform you that c4a214d's CIs have passed for more than 7 days. To prevent PR conflicts, you need to re-run all CIs manually. |
phlrain
approved these changes
Oct 10, 2024
Hongqing-work
approved these changes
Oct 10, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
PR Category
Performance Optimization
PR Types
Performance
Description
Due to the excessively dense atomic operations used in the original backpropagation operator for bilinear interpolation, the performance deteriorates drastically under high parallelism caused by atomic operation contention. Through extensive quantitative analysis, this work designs a high-performance backpropagation operator upon the base of original operator that circumvents atomic operations and activates it under a certain threshold.
In typical cases, the optimized operator performs more than 3x better than the original operator, and its speedup ratio increases with the interpolation factor.
pcard-76996