[CINN] Adjust tile config parameters based on model benchmarks #69363
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Outdated suggestions cannot be applied.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
CINN
PR Types
Performance
Description
Adjust two key parameters in TileConfig based on recent end-to-end model benchmark results.
Change 1:
Add one more condition for using WarpReduce. This improves performance for huge-S-small-R cases, such as:
Change 2:
Set tighter limit on spatial inner loop number (from 32 to 4). This prevents performance degradation in Transpose and certain element-wise operations.
Note: this change doesn't imply that 32 is inherently a worse choice than 4; rather, it is due to the lack of support for scheduling Transpose, and 4 is currently the optimal choice for nvcc to do just acceptable optimization.
Pcard-85711