[Attention] Added caching for flashinfer binaries during JIT #17730
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
Comment on lines
68
to
73
| if hash_file.exists(): | ||
| with open(hash_file, "r") as f: | ||
| cached_hash = f.read().strip() | ||
| if cached_hash == system_lib_hash_value: | ||
| # Cache hit: return all object files in build_directory | ||
| return list(build_directory.glob("*.o")) |
There was a problem hiding this comment.
For cache hit, we also need to make sure all .o files exist and are not modified. If any of the .o files is missing or is modified since the latest_object_mtime, we also need to recompile. So let's add this into the hash key as well.
There was a problem hiding this comment.
I think its good to check for existance of the object file, but wouldn't the second condition will be never true(because of the definition of the latest_object_mtime). I think we should be pretty safe after the deletion check.
MasterJH5574
changed the title
|
@MasterJH5574 I have addressed the PR reviews. |
MasterJH5574
approved these changes
Mar 13, 2025
ShiboXing
pushed a commit
to ShiboXing/tvm
that referenced
this pull request
Aug 10, 2025
…17730) In this PR I have added support for caching the flashinfer binaries during the JIT flow in TVM.
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Outdated suggestions cannot be applied.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.
In this PR I have added support for caching the flash infer binaries during the JIT flow in TVM.