jit : update how nn.Parameters are detected and tagged with STATIC_MEMORY_LOCATION tag #1611
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
t-vi
reviewed
Jan 7, 2025
Co-authored-by: Thomas Viehmann <[email protected]>
kshitij12345
changed the title
t-vi
approved these changes
Jan 7, 2025
Do you have another fix in mind? Or is it the fact that we can't see nn.Module via thunderfx path that is bad? |
|
Tagging parameters makes a lot of sense to me. They're parameters for a reason. In the future we may also consider what happens if someone calls thunderfx on a module. We could detect that a module was passed to the call before passing it to torch.compile. However, there's no requirement that someone pass the module itself for optimization. |
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Outdated suggestions cannot be applied.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.
Related #1575
On thunderfx path, we don't see the
nn.Modulebut we can seenn.Parameter. In this PR, we tag the proxy with STATIC_MEMORY_LOCATION if the original tensor is an instance ofnn.Parameter.NOTE - This PR doesn't fix determining the buffers on thunderfx path.