[FXML-4770] TosaToLinalg: fix tosa.cast INT_MAX overflow constant #206
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
mgehre-amd
reviewed
Jun 21, 2024
|
Please open the same PR to the release branch |
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Outdated suggestions cannot be applied.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 lowering a tosa.cast op from float to signed int, the float is compared against INT_MAX to check for overflow. The 1 + INT_MAX(width) constant is converted to floating-point, and comparison with it is done in FP.
However, an intermediate
int64_ttype is used, which overflows and turns 1 + INT_MAX(64) negative when the destination type isi64. This PR usesuint64_tinstead (withgetZExtValue) so that 1 + INT_MAX stays in bounds and is correctly represented as floating-point.