[move-only] When adding implicit liveness uses for class field/global accesses, do not use the terminator, use the end access. #67088
Conversation
This file contains hidden or 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
|
@swift-ci test |
jckarter
approved these changes
Jul 3, 2023
|
I found the problem. We were not properly converting an assignable_but_not_consumable to initable_but_not_consumable in DefiniteInitialization. |
…r to bisect on variables that we are checking. These are only on in NDEBUG mode. It just makes it easier to quickly triage which variable is causing a checking problem.
… accesses, do not use the terminator, use the end access. Previously, when doing this I could just use the terminator both in cases of inout and for class field/global accesses... but after some recent changes to field sensitive pruned liveness, this seems to no longer work. So just do the right thing and use the field access. The specific bug was that we would introduce a destroy_addr along one of the paths where the value wasn't defined resulting in a dominance error. I added a SIL test that shows this fixes the issue, a swift test, and also an Interpreter test that validates the correctness. rdar://111659649
…ble_but_not_consumable -> initable_but_not_consumable. I fixed this for assign but missed a place where we needed to do the same thing for copy_addr. rdar://111709236
Both of the previous fixes in this PR were found via this test.
gottesmm
force-pushed
the
pr-f9b416b5281022981483e4d42a150a0ff0aec2e3
branch
from
29ca777 to
268cd76
Compare
|
@swift-ci test |
jckarter
approved these changes
Jul 3, 2023
|
Mishal just validated for me on the bot, that the test passes now. |
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.
Previously, when doing this I could just use the terminator both in cases of
inout and for class field/global accesses... but after some recent changes to
field sensitive pruned liveness, this seems to no longer work. So just do the
right thing and use the field access.
The specific bug was that we would introduce a destroy_addr along one of the
paths where the value wasn't defined resulting in a dominance error.
I added a SIL test that shows this fixes the issue, a swift test, and also an
Interpreter test that validates the correctness.
rdar://111659649