Robust behavior when finalizers resurrect GC data (fixes #124) #184
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
This reserves another bit in roots to store whether the node has been finalized. In a GC cycle, unmarked nodes are finalized at most once, and a second mark phase collects nodes that have not been ressurected.
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Outdated suggestions cannot be applied.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 is implemented through using a persistent bit that is set when a node's finalizer is run, and by running sweep() only on finalized nodes. During a GC cycle, the first mark() is used only to determine which nodes to finalize. The second mark() determines which nodes to reclaim.
Since the GC already did two mark phases, I thought it appropriate to keep a second mark phase for this implementation. It may be possible to defer reclamation until another GC cycle entirely, but that would require an equivalent to
needs_dropfor finalization to be efficient for empty finalizers, which appears to be non-trivial (or even impossible) without language support.I also added a test that resurrects a GC in a finalizer and passes miri with these changes. I'm not aware of an easy way to produce a segfault in this scenario.