fixing out of bound access for nll_loss #1752
Open
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
for more information, see https://pre-commit.ci
IvanYashchuk
reviewed
Feb 7, 2025
thunder/torch/__init__.py
Outdated
| @@ -5111,12 +5111,18 @@ def _nll_loss_helper( | |||
| bcast_weight = reshape(weight, [num_class] + [1 for _ in range(2, a.ndim)]) | |||
| out = out * bcast_weight | |||
|
|
|||
| assert isinstance(ignore_index, Number) | |||
There was a problem hiding this comment.
If ignore_index is a NumberProxy are the constraints created as expected for >= 0 and < num_class comparisons with the symbolic values cache mode?
There was a problem hiding this comment.
For today with constant constraints, I think that's right. But then we could have the same out of bound access issue with NumberProxy for ignore_index.
If we do need to support that, I think for now we need to use < 0 or >= num_class to handle it.
There was a problem hiding this comment.
BTW, added the support.
For some reason thunderfx does indeed hit give ignore_index as NumberProxy instead, you probably know this better than I do.
jjsjann123
changed the title
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Outdated suggestions cannot be applied.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.
Fixes #1744
See torch.nn.functional.nll_loss definition here: https://pytorch.org/docs/stable/generated/torch.nn.functional.nll_loss.html
Because we use
prims.take_along_axis, which doesn't cap out-of-bound access. For ignore_index that's out side of [0, C), we need to:target(so we are not going to be accessing arbitrary memory);