Fix reversed conditions in ObjectOptimizer::calculateCacheKey()
#15909
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
clonker
previously approved these changes
Mar 3, 2025
cameel
force-pushed
the
fix-reversed-conditions-in-object-optimizer-cache-key
branch
from
4ccf599 to
45add19
Compare
clonker
previously approved these changes
Mar 3, 2025
clonker
approved these changes
Mar 3, 2025
- This fixes the error caused by accessing an empty optional. - Switch to static_cast
cameel
force-pushed
the
fix-reversed-conditions-in-object-optimizer-cache-key
branch
from
1842cfa to
c9aa6f4
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Outdated suggestions cannot be applied.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.
Fixes #15871.
The cache key calculation was using reversed values for boolean flags. This was unintended, but wasn't really breaking anything on its own since they can't be read back and it only matters that they are unique. Actual problems started in #15467, where a new entry was based on the incorrect condition, putting code dereferencing an
optionalin the wrong branch.This started failing a debug assertion in STL, but fortunately does not seem like it could lead to broken output. If the value that ends up getting stored is zero, it will result in EOF and non-EOF compilation getting the same cache key, but it's currently not possible to mix them. If the value is some random garbage from memory, it will lead to identical objects getting different keys, bypassing the cache, but we haven't observed this kind of degradation in benchmarks so far.