Skip to content

test(runtime): clear C5a per-key vetting sets in the inline-guard test reset #6805

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
proggeramlug merged 2 commits into from
Jul 23, 2026
+11 −0
Merged

test(runtime): clear C5a per-key vetting sets in the inline-guard test reset #6805

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c9bac17
test(runtime): clear the C5a per-key vetting sets in the inline-guard…
Jul 22, 2026
c3e4042
docs(changelog): changeset for test-reset follow-up
Jul 22, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions changelog.d/6805-test-reset-c5a-sets.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
test(runtime): the class-field inline-guard test reset also clears the C5a per-key vetting hash sets, removing cross-test order dependence for reused key names (CodeRabbit follow-up on #6802).
10 changes: 10 additions & 0 deletions crates/perry-runtime/src/object/descriptor_state.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -147,6 +147,16 @@ pub(crate) fn class_field_inline_guard_enabled() -> bool {
#[cfg(test)]
pub(crate) fn test_reset_class_field_inline_guard() {
PERRY_CLASS_FIELD_INLINE_GUARD_DISABLED.store(0, Ordering::Relaxed);
// Also clear the C5a per-key vetting sets (production-monotonic, so
// without this a key name reused across tests in one process would
// inherit an earlier test's declared-field / installed-key state and
// make the disable decision order-dependent (CodeRabbit on #6802).
if let Ok(mut guard) = DECLARED_FIELD_NAME_HASHES.write() {
guard.take();
}
if let Ok(mut guard) = PROTO_DESCRIPTOR_KEY_HASHES.write() {
guard.take();
}
Comment on lines +154 to +159

@coderabbitai coderabbitai Bot Jul 22, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win

Do not silently skip a poisoned reset lock.

If a test panics while holding either lock, write() returns Err and this helper leaves the corresponding hash set populated, reintroducing order-dependent test state. Recover the poisoned guard with into_inner() (or fail explicitly) so the reset remains deterministic.

🤖 Prompt for AI Agents 
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@crates/perry-runtime/src/object/descriptor_state.rs` around lines 154 - 159,
Update the reset helper’s write-lock handling for DECLARED_FIELD_NAME_HASHES and
PROTO_DESCRIPTOR_KEY_HASHES so poisoned locks are recovered with into_inner() or
cause an explicit failure, rather than being silently skipped. Ensure
guard.take() runs for both hash sets even after a prior test panic, preserving
deterministic reset behavior.

}

/// #5654: flip the process-wide inline gate only when the descriptor target can
Expand Down
Loading