chore(deps): updated vihaco dependency to 0.4.0

[robpatterson13]

Summary

Migrates ppvm-vihaco and vihaco-circuit-isa from vihaco 0.1.1 to vihaco 0.4.0.

Changes

• Migrates PPVM to vihaco 0.4.0 surface/runtime instruction APIs.
• Replaces the legacy parser pipeline with v0.4 parsing and resolution.
• Generates the circuit instruction set with vihaco::component!.
• Enforces one-to-one source-to-runtime instruction lowering.
• Updates .sst syntax, including:

 const.u64 0 → cpu::cpu.const u64, 0
eq.u32 → cpu::cpu.eq u32
circuit.h → circuit::circuit.h
ret → cpu::cpu.ret 0
@label: → cpu::cpu.label @label

• Resolves branch targets and function calls to runtime addresses.
• Adds bytecode v2 serialization for function tables, labels, signatures, and main_function.
• Updates CLI, TUI, examples, embedded programs, and test fixtures.
• Adds and updates migration documentation in docs/changelog/vihaco-0.4.md.

Verification

• cargo test --workspace
• cargo check -p ppvm-vihaco
• cargo test -p ppvm-vihaco
• cargo test -p vihaco-circuit-isa
• cargo test -p ppvm-cli
• cargo test -p ppvm-tui

[github-actions]

👋 Thanks for opening your first pull request against PPVM!

A quick note on contribution terms: by submitting this PR you
agree that your contribution is licensed under the
Apache License 2.0
and that you accept the
PPVM Contributor License Agreement.
Please skim those before a maintainer reviews — opening this PR
counts as your acceptance.

A few things that will speed up review:

• Read CONTRIBUTING.md
  for the workflow, build commands, and style notes.
• Run prek run --all-files locally; CI runs the same checks.
• Use Conventional Commits
  for commit messages.

We'll get to your PR as soon as we can. Thanks for contributing!

[david-pl]

@robpatterson13 overall looks good, just some cosmetic comments. Agent review found two issues (which probably weren't introduced here, but still):

1. Symbolic function references such as @helper fail because FunctionRef operands are parsed only as numeric IDs.
2. main_function is stored as a code address instead of a function ID, and execution always starts at PC 0 rather than at @main.

Here are some failing tests (run as part of tests in ppvm-vihaco/src/lib.rs):

#[test]fnsymbolic_function_ref_calls_named_function(){let src = "device circuit.n_qubits 1;\n\ fn @main() {\n\ cpu::cpu.const fn_ref, @helper\n\ cpu::cpu.call_indirect\n\ cpu::cpu.ret 0\n\ }\n\ fn @helper() {\n\ cpu::cpu.const u64, 0\n\ circuit::circuit.measure\n\ cpu::cpu.ret 0\n\ }\n";let machine = run_program(src).expect("symbolic function reference should resolve");assert_eq!(machine.measurement_record().len(),1);}#[test]fnexecution_starts_at_main_when_helper_is_declared_first(){let src = "device circuit.n_qubits 1;\n\ fn @helper() {\n\ cpu::cpu.const u64, 0\n\ circuit::circuit.x\n\ cpu::cpu.ret 0\n\ }\n\ fn @main() {\n\ cpu::cpu.const u64, 0\n\ circuit::circuit.measure\n\ cpu::cpu.ret 0\n\ }\n";let module = compile_program(src).expect("program should compile");letmut machine = PPVM::default();
machine.load(&module).expect("module should load");
machine.run().expect("program should run");assert_eq!((module.main_function, machine.measurement_record().len()),(Some(1),1),"main_function should be a function-table ID and execution should begin at @main");}

Also, CI is blocked because the pre-commit hooks fail in an unrelated crate (not sure why). Would you mind fixing that so CI actually runs on the PR?

I'll approve the changes, feel free to merge once the above things are cleared up.

[github-actions]

PR Preview Action v1.8.1
Preview removed because the pull request was closed.
2026-08-27 16:30 UTC