aarch64: Fix lowering amounts for shifts #3074
Merged
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
github-actions
bot
added
cranelift
afonso360
force-pushed
the
aarch-ishl-i8
branch
2 times, most recently
from
ada4414
to
c2ca3cf
Compare
akirilov-arm
suggested changes
Jul 14, 2021
This commit addresses two issues: * A panic when shifting any non i128 type by i128 amounts (bytecodealliance#3064) * Wrong results when lowering shifts with small types (i8, i16) In these types when shifting for amounts larger than the size of the type, we would not get the wrapping behaviour that we see on i32 and i64. This is because in these larger types, the wrapping behaviour is automatically implemented by using the appropriate instruction, however we do not have i8 and i16 specific instructions, so we have to manually wrap the shift amount with an AND instruction. This issue is also found on x86_64 and s390x, and a separate issue will be filed for those. Closes bytecodealliance#3064
akirilov-arm
approved these changes
Jul 15, 2021
|
Thanks for reviewing @akirilov-arm! |
|
No problem; once the pull request passes the tests, it should be ready for merging. |
|
These look like spurious failures to me. It looks like crates.io is not having a great time right now. My other PR is also failing due to the same reason. Retrying CI later may work out better |
|
Yes, that's true - all active PRs are failing right now with the same cause, including one of mine and another one. |
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Outdated suggestions cannot be applied.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.
This PR addresses two issues:
In these types when shifting for amounts larger than the size of the
type, we would not get the wrapping behaviour that we see on i32 and i64.
This is because in these larger types, the wrapping behaviour is automatically
implemented by using the appropriate instruction, however we do not
have i8 and i16 specific instructions, so we have to manually wrap
the shift amount with an AND instruction.
This issue is also found on x86_64 and s390x, and a separate issue will
be filed for those.
Closes #3064