[MLIR] Emit error instead of segfault in current scatter lowering #1217
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
erick-xanadu
force-pushed
the
eochoa/2024-10-18/exit-early-instead-of-segfault
branch
from
dd276c9
to
953f334
Compare
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Outdated suggestions cannot be applied.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.
Context: The current scatter lowering makes some assumptions and only handles a subset of the inputs. However, it is too optimistic and will attempt to convert inputs that it can't handle. In particular if there are more than one input update and result tensors, it will segfault. This condition may never happen but it is valid IR. It may never happen if another pass runs before that canonicalizes mhlo.scatter. However, it is good to emit an error in case the canonicalization was not run before.
Description of the Change: Emit error instead of segfaulting
Benefits: No segfaults
Possible Drawbacks: None
Related GitHub Issues:
Depends on #1216