Drop __mul_used from IrOp::Mul codegen and __div_used from IrOp::Div
/ IrOp::Mod codegen (modulo reuses the same routine). The linker
skips gen_multiply / gen_divide for programs that never emit the
markers, following the same pattern already used by __audio_used /
__ppu_update_used / __sprite_cycle_used.
The optimizer already rewrites multiplies and divides by constant
powers of two into shifts (and modulo by constant powers of two
into masks), so the markers only fire for genuinely runtime math.
A program like `examples/comparisons.ne` that never multiplies or
divides now reclaims ~56 bytes of PRG; programs that use only one
of the two reclaim the other's share.
Audio goldens flip for every example that uses audio. The .ne
sources are unchanged and the pixel goldens are byte-identical —
the audio stream differs only because removing the math routines
shifts the audio tick's absolute address in PRG by 56 bytes, which
changes which of its internal branches cross 6502 page boundaries
and therefore the per-frame cycle count of a single NMI by 1-5
clocks. Over 180 frames the accumulated drift shifts APU register
write timing enough to render a different digital sample stream
at the same logical wave shape. Expected consequence of ROM-layout
change under cycle-accurate emulation; documented path per
CLAUDE.md "Updating goldens".
https://claude.ai/code/session_016kM6P7PukktBDqTZexrrAN
Every NEScript condition (`if x < N`, `while i < end`, etc.)
lowers in two IR ops: `CmpX(d, a, b)` materializes a 0/1
boolean into temp `d`, and the block's terminator
`Branch(d, t, f)` reads `d` and branches on it. The codegen
faithfully emitted both halves — `LDA / CMP / branch-to-true /
LDA #0 / JMP done / true: LDA #1 / done:`, then later
`LDA d_slot / BNE branch_t / JMP branch_f` — about 14 cycles +
13 bytes per condition.
The 6502's natural pattern is one `CMP` + one branch on the
flags it just set: 8 cycles, no register-clobber, no temp slot.
Detect the canonical pattern in `gen_block` (last op is an 8-bit
`CmpX` whose dest temp is what the terminator branches on, with
no other uses) and emit the fused form directly via a new
`gen_cmp_branch` helper. The temp's allocation, store, load, and
the terminator's branch fall away.
Bookkeeping subtlety: the source temps `a`/`b` must be retired
*after* the fused emit, not before — the original `gen_op` order
is "emit body of op, then `retire_op_sources`". Decrementing
their use counts before the CMP would free their slots while
they were still live; `load_temp(a)` would then re-allocate `a`
to whatever stale slot the free list popped next. Got hit by
this on the first attempt — the SHA-256 example dutifully
returned all-zero hashes until the order was fixed.
Updated `ir_codegen_local_label_suffix_is_bank_namespaced`: the
test was relying on `if x == 0` to emit `__ir_cmp_*` labels for
its bank-namespacing check, which the fusion now collapses into
direct branches. Switched the test source to a shift-by-variable
pattern (`x = x << n`), which always emits `__ir_shift_loop_*`
labels regardless of future cmp/branch optimizations.
Cycle savings: ~6 cycles per condition. The SHA-256 rotate
loops alone account for ~9K cycles per block. Across all
examples the cycle drift shows up as audio-tick phase shifts
in five timing-sensitive ROMs (`audio_demo`, `friendly_assets`,
`noise_triangle_sfx`, `platformer`, `sfx_pitch_envelope`); the
goldens for those are refreshed in this commit, plus
`platformer.gif` (the only demo gif whose bytes actually moved).
Verified: cargo test/clippy/fmt clean on rustc 1.95.0;
emulator harness 34/34; reproducibility diff clean; SHA-256 of
"NES" still computes to AE9145DB…4E0D.
https://claude.ai/code/session_01FRmSBruVWCufm3LsUVMs8v
Programs that put functions in switchable banks can now call across
bank boundaries — `bank A { fun step() { helper() } }` where
`helper` lives in `bank B` used to panic in the IR codegen. Three
small pieces unblock it:
1. **Generic trampoline.** `runtime/gen_bank_trampoline` no longer
takes a `fixed_bank_index` argument. Instead it reads the
caller's current bank from `ZP_BANK_CURRENT`, pushes it on the
hardware stack, switches to the target, JSRs the entry, then
pulls and restores the saved bank. The same per-callee stub
works for fixed→banked and banked→banked direction; nested
trampolines compose because each PHA/PLA pair sits inside its
own JSR/RTS frame. `gen_mapper_init` seeds `ZP_BANK_CURRENT`
with the fixed bank index for any banked mapper so the very
first cross-bank call from the fixed bank still restores to
the fixed bank (matching pre-banked-banked semantics).
2. **Codegen drops the panic.** The `Some(from), Some(to)` arm in
the call-resolution switch now emits `JSR __tramp_<name>` like
the fixed→banked case instead of panicking. Banked→fixed calls
still go direct (the fixed bank is always mapped at $C000).
3. **Bank-namespaced local labels.** Two banks emitting the same
`__ir_cmp_e_8` would trip the linker's discovery-pass duplicate-
label check the moment any banked code generated a comparison.
The new `local_label_suffix` helper prefixes the suffix with the
current bank name when banked code is being emitted, leaving
fixed-bank label generation untouched (so existing examples are
byte-identical apart from the trampoline / init bytes
themselves).
The new `examples/uxrom_banked_to_banked.ne` demonstrates the path
end-to-end: `bank Logic { fun step() { ... clamp() } }` calls
`bank Helpers { fun clamp() { ... } }` once per frame. The harness
golden is committed alongside it. The five existing banked example
ROMs change byte-for-byte because of the new trampoline shape and
the seed-ZP_BANK_CURRENT init, but their emulator goldens still
match exactly — observable behaviour is unchanged.
https://claude.ai/code/session_01KEczoNUX3WmcFLfq6iAQxB
