gen-zig CSE hoists count + 1 above var count and across its reassignment #8197

Re-filed from #6292. An attempt claimed that issue on 2026-10-05 and left an empty branch, and a dispatch row that has spent its retry ceiling keeps its issue for good - so no bee could take it again, however much the brief improved. The work below is carried over unchanged.

Refs #6063

The reference path (t27c gen + zig test) fails to compile two specs because the Zig backend's common-subexpression pass hoists an expression above the declaration of the mutable local it reads.

specs/fpga/apb_bridge.t27 line 169:

fn apb_port_count(cfg: ApbConfig) -> u32 {
    var count : u32 = 0;
    count = count + 1;
    count = count + 1;
    ...

t27c gen (master 47621f4) emits:

fn apb_port_count(cfg: ApbConfig) u32 {
    const _cse1 = count + 1;
    const _cse6 = count + 1;
    const _cse7 = count + 1;
    var count: u32 = 0;

and zig test answers use of undeclared identifier 'count'. Same in specs/fpga/axi4.t27. Besides the order, count + 1 is not a common subexpression at all: count is reassigned between the uses, so even after the declaration the CSE would change the result.

Found while landing #6291: t27b now passes both specs (all tests), but they stay outside the honest denominator until the reference compiles them. (specs/fpga/linker.t27 is the same situation through .align = 4, already #4807.)

Boundary

• specs/fpga/apb_bridge.t27
• specs/fpga/axi4.t27
• specs/fpga/linker.t27

User Scenarios

• Given a spec with a mutable local variable that is reassigned between uses, when the Zig backend runs its common-subexpression elimination pass, then the pass must not hoist expressions that read the variable above its declaration.
• Given a spec where the same expression count + 1 appears multiple times but count is reassigned between appearances, when the CSE pass runs, then it must not treat those appearances as a common subexpression.

Requirements

• FR-001: The Zig backend's CSE pass MUST ensure that any hoisted expression only references variables that are already declared in the current scope at the hoist point.
• FR-002: The Zig backend's CSE pass MUST NOT combine expressions that read a mutable variable if that variable is reassigned between the expression occurrences.
• FR-003: The compiler MUST emit valid Zig code that compiles with zig test for all specs in the specs/fpga/ directory.

Success Criteria

• t27c gen specs/fpga/apb_bridge.t27 produces Zig code that compiles with zig test and exits 0
• t27c gen specs/fpga/axi4.t27 produces Zig code that compiles with zig test and exits 0
• t27c gen specs/fpga/linker.t27 produces Zig code that compiles with zig test and exits 0
• All three specs pass the full test suite (3 specs × their respective test counts)

The instruments

Every one of these reads and prints; none of them writes. Run them from the repository root, with <spec> replaced by the file in ## Boundary:

.t27 is not Rust and not Zig, and a bee fluent in either writes one by
accident. Measured 2026-09-20: a bee filled eight bodies in
specs/file/watcher.t27 with return Ok(());, Err(FileError::WatcherNotFound)
and for i in 0..watchers.length. The parse ratchet refused the whole file --
Unexpected token in expression: RParen, line 123 -- and a spec that does not
parse generates nothing, so every test it already carried stopped running.

At the top level the parser accepts exactly these eight forms, with an optional
pub, and no others:

const   var   fn   enum   struct   test   invariant   bench

There is no trait, no impl, no type X = ..., no generics, no macro. What
a bee reaches for from another language, and what happens:

• Rust's result sugar (Ok(()), Err(E::V), ?) is not a construct here.
• Rust macros (println!, format!, vec!) are not constructs here.
• A range loop (for i in 0..n) is Rust and Zig, not this language.
• Zig builtins (catch unreachable, @intCast, @constCast) are not constructs here.
• Cross-module reuse (use other::fn;) parses and then generates a comment and an unqualified call, so the Zig fails with use of undeclared identifier (use a::b; generates no import, so no spec can reuse a function from another spec #4298). It does not work yet.

t27c parse <spec> answers in one line whether what you wrote is the language.
Run it before you report, every time.

• t27c spec-status <spec> - the compiler's one-word verdict: IMPLEMENTED, PARTIAL, UNWRITTEN, NOPARSE, NOFN. Exit code is 0 whatever it says, so read the word.
• t27c symbols <spec> - every name the file declares, with its kind. Answers "does this already exist here?" before you add it.
• t27c outline <spec> - per function: its locals, what it calls, what it returns. The contract you are implementing against.
• t27c coverage <spec> - which functions have a test and which do not. The issue asks for tests; this is how you check you wrote them.
• t27c lint <spec> - style and shape warnings. It OVER-REPORTS has no test or invariant: measured 2026-09-20 over 60 specs it printed 677 of those where coverage found 190 untested functions, disagreeing on 59 of the 60 - it warns about bit_to_trit_pair in specs/base/ternary_encoding.t27, which test bit_to_trit_pair_zero calls on line 249. Read it as a hint; coverage is the answer.
• t27c typecheck <spec> - types, before generation. Prints Typecheck OK (0 errors, 0 warnings) or the errors.
• t27c test-report <spec> - builds this spec and runs its own tests, which is what the oracle does. BLOCKED means the generated Zig does not compile, with the error beside it.
• python3 tools/dupe_scan.py --name <function> - where that function already lives, if it does. 576 of 4021 bodies here are byte-identical copies.

The rest of the toolbelt is in docs/BEE_TOOLBELT.md.