[CUTLASS][Cherry-pick] Introduce several features of cutlass profiler #15573
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
- allow Conv2d using different alignment factors for input and epilogue, which can influence performance - store the profiler cache on disk, reducing CUTLASS profiler overhead across different runs - use the same set of default tile configurations as CUTLASS for sm80 https://github.com/NVIDIA/cutlass/blob/master/tools/library/scripts/generator.py#L1881
Collaborator
|
Thanks for contributing to TVM! Please refer to the contributing guidelines https://tvm.apache.org/docs/contribute/ for useful information and tips. Please request code reviews from Reviewers by @-ing them in a comment.
Generated by tvm-bot |
Hzfengsy
approved these changes
Aug 17, 2023
Member
Could you please also patch it to unity branch to address the potential conflict? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 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Outdated suggestions cannot be applied.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.
#14275 was sent and merged to
unityfor no good reason, so I'm sending this tomainnow.Also the PR makes the conv2d profiling time 4x slower by expanding the search space over the output alignments. In most cases (
profile_all_alignments = False, the default), we just want to pick the largest-possible alignment. To prevent the conv2d profiling time from blowing up in the default path, this PR adds a fix on top of #14275. Due to this difference, expect merge conflict on the nextunity+mainmerge.@junrushao @spectrometerHBH @Hzfengsy @jwfromm