Fix UB-introducing rewrite in FindIntrinsics #8539
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
FindIntrinsics was rewriting i8(rounding_shift_left(i16(foo_i8), 11)) to rounding_shift_left(foo_i8, 11) I.e. it decided it could do the shift in the narrower type. However this isn't correct, because 11 is a valid shift for a 16-bit int, but not for an 8-bit int. The former is zero, the latter gets turned into a poison value because we lower it to llvm's shl. This was discovered by a random failure in test/correctness/lossless_cast.cpp in another PR for seed 826708018. This PR fixes this case by adding a compile-time check that the shift is in-range. For the examples in test/correctness/intrinsics.cpp the shift amount ends up in a let, so making this work on those cases required handling a TODO: tracking the constant integer bounds of variables in scope in FindIntrinsics and therefore also in lossless_cast.
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Outdated suggestions cannot be applied.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.
FindIntrinsics was rewriting
i8(rounding_shift_left(i16(foo_i8), 11))to
rounding_shift_left(foo_i8, 11)I.e. it decided it could do the shift in the narrower type. However this isn't correct, because 11 is a valid shift for a 16-bit int, but not for an 8-bit int. The former is zero, the latter gets turned into a poison value because we lower it to llvm's shl. This was discovered by a random failure in test/correctness/lossless_cast.cpp in another PR for seed 826708018.
This PR fixes this case by adding a compile-time check that the shift is in-range. For the examples in test/correctness/intrinsics.cpp the shift amount ends up in a let, so making this work on those cases required handling a TODO: tracking the constant integer bounds of variables in scope in FindIntrinsics and therefore also in lossless_cast.