GeezerVison

crates/vm/src/frame.rs (1)

6835-6846: Deduplicate the FOR_ITER exhaustion target adjustment.

for_iter_jump_on_exhausted() now re-implements the same END_FOR skip logic as for_iter_jump_target(). Reusing the helper would keep this control-flow rule in one place.

♻️ Proposed cleanup

     /// Handle iterator exhaustion in specialized FOR_ITER handlers.
     /// Skips END_FOR if present at target and jumps.
     fn for_iter_jump_on_exhausted(&mut self, target: bytecode::Label) {
-        let target_idx = target.as_usize();
-        let jump_target = if let Some(unit) = self.code.instructions.get(target_idx) {
-            if matches!(
-                unit.op,
-                bytecode::Instruction::EndFor | bytecode::Instruction::InstrumentedEndFor
-            ) {
-                bytecode::Label::new(target.as_u32() + 1)
-            } else {
-                target
-            }
-        } else {
-            target
-        };
-        self.jump(jump_target);
+        self.jump(self.for_iter_jump_target(target));
     }

Also applies to: 8817-8832

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@crates/vm/src/frame.rs` around lines 6835 - 6846, The FOR_ITER exhaustion
target adjustment logic is duplicated between for_iter_jump_target and
for_iter_jump_on_exhausted; refactor so the rule lives in one place by making
for_iter_jump_on_exhausted call for_iter_jump_target (or extract a shared helper
used by both). Locate the duplicate logic in functions for_iter_jump_target and
for_iter_jump_on_exhausted and remove the repeated END_FOR/InstrumentedEndFor
check from for_iter_jump_on_exhausted, delegating to for_iter_jump_target to
perform the END_FOR skip and return the adjusted bytecode::Label.

crates/compiler-core/src/bytecode/oparg.rs (1)

702-705: From<$name> for usize bakes in a pointer-width assumption.

std intentionally does not provide From<u32> for usize; this macro reintroduces the same potentially truncating cast for every oparg wrapper. I'd keep as_usize() as the explicit conversion and only add a trait impl here if you want a checked TryFrom.

Suggested change

-        impl From<$name> for usize {
-            fn from(value: $name) -> Self {
-                value.as_usize()
-            }
-        }

🤖 Prompt for AI Agents

Verify each finding against the current code and only fix it if needed.

In `@crates/compiler-core/src/bytecode/oparg.rs` around lines 702 - 705, The impl
From<$name> for usize bakes in a pointer-width assumption and can truncate on
some platforms; remove that impl and keep callers using the explicit as_usize()
method on the wrapper ($name.as_usize()). If you need a fallible/checked
conversion, add a TryFrom<$name> for usize implementation instead (implement
TryFrom::try_from for $name that performs a safe/checked conversion and returns
a Result), but do not reintroduce the infallible From<$name> for usize.