improve asm sucessor to deal with arbritary jumps

[xunilrj]

Description

Checklist

• I have linked to any relevant issues.
• I have commented my code, particularly in hard-to-understand areas.
• I have updated the documentation where relevant (API docs, the reference, and the Sway book).
  • If my change requires substantial documentation changes, I have requested support from the DevRel team
• I have added tests that prove my fix is effective or that my feature works.
• I have added (or requested a maintainer to add) the necessary Breaking* or New Feature labels where relevant.
• I have done my best to ensure that my PR adheres to the Fuel Labs Code Review Standards.
• I have requested a review from the relevant team or maintainers.

[cursor]

PR Summary

Medium Risk
Touches CFG successor computation used by liveness analysis and reachability-based optimizations, which can impact register allocation and dead-code elimination correctness. Risk is mitigated by conservatively bailing out of optimizations when successor sets are Unknown (e.g., JumpToAddr, potential ECAL).

Overview
Introduces a new InstructionSuccessor enum and changes Op::successors() (and virtual/control-flow successor helpers) to return explicit successor semantics instead of a raw Vec<usize>, distinguishing Unknown, ReturnFromCall, and Many successor sets.

Updates liveness analysis and CFG reachability simplification to consume this new API; reachability now conservatively disables CFG simplification when successors are Unknown (and similarly bails early on JumpToAddr). Snapshot output for main_args_various_types is updated to reflect resulting register allocation changes.

^{Written by Cursor Bugbot for commit 1327cf7. This will update automatically on new commits. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Off-by-one in assertion allows invalid index
  • Changed the successors assertion to index < ops.len() so invalid end/empty-slice indices are rejected before successor computation.

Or push these changes by commenting:

@cursor push ce836f34a4

Preview (ce836f34a4)

diff --git a/sway-core/src/asm_lang/virtual_ops.rs b/sway-core/src/asm_lang/virtual_ops.rs
--- a/sway-core/src/asm_lang/virtual_ops.rs
+++ b/sway-core/src/asm_lang/virtual_ops.rs
@@ -963,7 +963,7 @@
     /// ECAL can do pretty much anything, but it executes the next instruction, if "$pc"
     /// is not manipulated.
     pub(crate) fn successors(&self, index: usize, ops: &[Op]) -> InstructionSuccessor {
-        assert!(index <= ops.len());
+        assert!(index < ops.len());
         match self {
             VirtualOp::RET(..) | VirtualOp::RETD(..) | VirtualOp::RVRT(..) => {
                 InstructionSuccessor::Many(vec![])

_{This Bugbot Autofix run was free. To enable autofix for future PRs, go to the Cursor dashboard.}

[cursor]

Off-by-one in assertion allows invalid index

Low Severity

The assertion assert!(index <= ops.len()) uses <= instead of <. Since index represents a valid instruction index into ops, the invariant is index < ops.len(). Using <= allows index == ops.len() to pass the assert silently. Worse, if ops is empty and index == 0, the assert passes but ops.len() - 1 on line 974 causes a usize underflow panic. With <, the assert would correctly catch both cases.

 

[codspeed-hq]

Merging this PR will not alter performance

✅ 25 untouched benchmarks

---

_{Comparing xunilrj/improve-asm-successor (1327cf7) with master (cf207d2)}