Behavior-neutral prep on top of the inline-load parity milestone:
* JitCache caches mem.fast_mem() (resolved once — the mapping is invariant for
a run) so compiled blocks build their MemEnv without a virtual call per
execution. MemEnv::from_fast(mem, Option<FastMem>) is the non-virtual
constructor. Measured impact is within noise (96.5 -> 96.4 MIPS) but it
removes redundant per-block work.
* FastMem/MemEnv gain mmio_count: a pointer to GuestMemory's monotonic MMIO
access counter. Unused for now; the upcoming superblock-chaining JIT will
sample it across a block boundary to stop chaining on an MMIO touch
(preserving the interpreter's fine-grained MMIO ordering).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Loads no longer call a trampoline on the common path. When memory exposes a flat
mapping and the address is proven non-MMIO and committed, the compiled block
loads directly from `membase + ea` (byte-swapped) with zero calls; otherwise it
falls back to the read trampoline.
Mechanism:
* MemoryAccess::fast_mem() -> Option<FastMem { membase, page_table, mmio_mask,
mmio_value }>. Default None; GuestMemory returns Some (flat 4 GiB mapping,
page-table pointer, MMIO aperture pair). Wrappers that intercept accesses
(the recompiler's speculative OverlayMemory) inherit None, so the fast path
is disabled under the diff harness and their interception is preserved.
* MemEnv carries membase/page_table/mmio_mask/mmio_value; MemEnv::new(mem)
fills them (or nulls when fast_mem() is None).
* emit_inline_load emits a 5-block diamond: membase==0 -> slow (disabled);
(ea & mmio_mask) == mmio_value -> slow (maybe MMIO); page_table[ea>>12]
COMMIT bit (49) clear -> slow (unmapped; the mapped check is mandatory —
unmapped pages are PROT_NONE and a raw load would fault); else fast load
membase+ea, bswap, zero-extend. Wired all integer loads (lbz/lhz/lwz/ld +
x-forms) and FP loads (lfs/lfsx reuse the diamond then bitcast/fpromote;
lfd/lfdx bit-copy). Stores and FP-punt still trampoline.
Validation (three ways, since the diff harness disables the fast path):
* unit test jit_inline_load_matches_interpreter — real GuestMemory, both the
mapped (fast) and unmapped (slow) address bit-match the interpreter.
* e2e XENIA_JIT=1 boot+movie plays, clean exit.
* diff regression: checked 149.2M blocks, MISMATCHES=0 (validates the
unchanged slow/ALU/branch/FP logic).
Measured (2e9 instr, block-exec MIPS): interpreter 98.2 / JIT trampoline-loads
93.8 / JIT inline-loads 96.5 — inline loads recover the trampoline overhead,
bringing the JIT to parity (~1.7% slower, within noise). Beating the interpreter
needs the levers it can't do: block-linking (30% dispatch), host registers.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Foundation for a staged PPC block-recompiler (plan: closure-threaded ->
Cranelift -> block-linking). No codegen yet — proves the integration seam and
the correctness harness before any lowering exists.
New crates/xenia-cpu/src/recompiler.rs:
- run_block: executes a DecodedBlock; M0 falls back to the interpreter's
execute() for every opcode, so it is bit-identical to step_block. Later
stages dispatch lowered ops here and fall back only for uncompiled opcodes.
- gates jit_enabled()/diff_enabled() (XENIA_JIT / XENIA_JIT_DIFF, cached).
- In-process differential harness (diff_step): the interpreter is AUTHORITATIVE
(drives real ctx+mem, so a JIT bug can never corrupt a run); the JIT runs
SPECULATIVELY on a ctx clone against OverlayMemory (writes buffered in a byte
HashMap, reads fall through to real pre-block memory), then registers are
compared. Blocks touching MMIO or sync_sensitive (reservation/barrier) are
skipped — a device callback can't be run twice and reservation state is
shared cross-thread. report_diff_summary() prints checked/skipped/mismatch.
Why in-process: the guest is only COARSELY deterministic — coord_idle_advance
ticks vsync from wall-clock when idle, so two separate runs are not bit-exact
and a cross-run signature compare would measure jitter, not JIT divergence.
Supporting changes: PpcContext #[derive(Clone)] (harness clears the speculative
clone's reservation_table Arc); is_mmio() on the MemoryAccess trait (default
false) + GuestMemory impl via find_mmio; execute() made pub(crate); run_superblock
routes the block body (DIFF->diff_step, JIT->run_block, else step_block).
Validated: full boot+movie XENIA_JIT_DIFF=1 run = checked 218.4M blocks,
skipped(mmio/sync) 511.6K (0.23%), MISMATCHES=0 CLEAN; movie plays (ADVreads=30,
tid25 resumes); diff-mode ~2x slower (test-only path).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Replace the one-basic-block-per-slot-per-round lockstep dispatch with a
SUPERBLOCK runner: each slot-visit chains straight-line blocks through
their terminating branches up to a deterministic instruction budget,
amortizing the per-round (timebase/coord/round_schedule) and per-slot
(worker_prologue) dispatch tax over ~128 instructions instead of ~6.
Yield-points (end the chain, return to the round) are pure functions of
guest state, preserving the lockstep cross-thread interleaving correctness:
- non-Continue step result (Yield/SystemCall/Trap/Unimpl/Halted);
db16cyc Yield is the spin-wait producer hand-off.
- sync-sensitive block: lwarx/ldarx/stwcx./stdcx. or sync/eieio/isync
(new PpcOpcode::is_sync_sensitive, flagged on DecodedBlock at build).
- MMIO touch: new GuestMemory::mmio_access_count() watermark, sampled
per block, keeps GPU/register ordering at one-block granularity.
- next PC leaves ordinary guest code (import thunk / halt sentinel /
unmapped) -> hand to the full worker_prologue next round.
- instruction budget reached.
Instruction-count/clock accounting stays exact: per-block cycle_count
deltas are summed and handed to worker_epilogue once (instruction_count +
decrement_quantum advance by the precise retired count). XENIA_SUPERBLOCK_BUDGET=1
reproduces the old one-block schedule byte-for-byte.
Budget tuned to 128 (env-overridable): boot progression stays healthy up
to 256, sharp cliff at ~384 (a boot producer/consumer handoff starves);
128 is 3x below the cliff. Also scale the inline-GPU per-round fairness
cap with the budget (flat 64 throttled GPU command processing 17x under
superblocks and collapsed the present loop).
PERF (check -n 100M --gpu-inline): 25.3 -> 42.7 MIPS (1.69x); 1B: 26.0 ->
41.4 MIPS (1.59x). Callgrind n=5M: host instructions 2.178B -> 1.507B
(-31%); worker_prologue -90%, coord_pre_round -91%, begin_slot_visit /
round_schedule_into / coord_post_round / update_timestamp_bundle each
~-90%; interpreter execute byte-identical (real work unchanged).
GATES: C1 boot progression 150M draws 7391/swaps 2164 (baseline 7415/2172),
1B draws 88547/swaps 29228 linear no stall, K8888 decode + RTs=2 intact.
C2 determinism: n50m stable digest byte-identical across fresh runs;
golden re-baselined intentionally (pacing-only deltas: imports 333453->243387,
draws 1274->1279). C3 milestone-1 render: texture_decodes/draws/swaps/
present cadence track baseline (3AJ fade-in pacing preserved). C4: 690
tests green (+2 sync_sensitive).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Profile-driven low-risk optimizations attacking the ~48% per-block /
per-round host-bookkeeping tax found by the callgrind profile. Measured
on the bounded headless workload `check -n 100000000 --gpu-inline`:
baseline ~4490 ms (22.3 MIPS) -> ~3700 ms (27.0 MIPS), +21%.
Tier A (determinism-neutral; n50m golden byte-IDENTICAL, exit 0):
1. mem-watch write path: gate capture_mem_watch_old/check_mem_watch
behind one has_mem_watch() predicted branch in write_u8/16/32/64 +
write_bulk so the common (no-watch) store does no out-of-line call.
check_mem_watch (4.8%) gone from the profile.
2. round-schedule alloc churn: add Scheduler::round_schedule_into filling
a reusable [u8; HW_THREAD_COUNT] stack buffer; the lockstep round loop
no longer __rust_alloc/__rust_dealloc a Vec<u8> per round. Identical
ordering/RNG-advance. __rust_alloc/dealloc gone from the profile.
3. probe-firing: hoist a single KernelState::any_probe_active() guard to
worker_prologue so the four fire_*_if_match calls don't happen at all
when no probe is configured (was 4x call overhead/visit). All four
gone from the profile.
4. thunk-map hash: range-reject pc against the registered import-thunk
address band (KernelState::pc_in_thunk_band, two int compares) before
the thunk_map.get(&pc) HashMap lookup. hash_one (4.3%) gone.
Tier B (#5, time-granularity change — LANDED, no re-baseline needed):
5. update_timestamp_bundle: throttle to a 0.25 ms quantum (only re-write
the KeTimeStampBundle when the deterministic clock advanced >= 2500
units). Inclusive cost 8.65% -> 1.08%. The quantum is far below the
1 ms granularity any guest deadline math needs (tick_count stays
fresh; the hub gate is +66 ms; the fade-in is vsync-counter driven per
3AH, not this bundle). VERIFIED: n50m stable digest BYTE-IDENTICAL to
the existing golden (so no re-baseline), 150M boot reaches the splash
(draws=7415, swaps=2172, gpu.texture.decode{K8888}=448, RTs=2 — all
match the post-3AJ baseline), 688 tests green, release n50m oracle ok.
Remaining headroom: interpreter::execute (13%), decrement_quantum (8%),
step_block (7%) are now the top self-costs — the structural superblock/
JIT lever is the next step for the larger gain.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds an opt-in diagnostic that emits one tracing line per guest store
overlapping any armed byte address, naming the writer (tid, pc, lr)
plus old/new u32 lanes. Mirrors the --pc-probe / --branch-probe shape;
pc/lr are stamped from worker_prologue via a thread-local Cell, so
default runs (empty watch set) take a single is_empty() check on each
write. Lockstep digest preserved (instructions=100000003 across reruns,
sylpheed_n50m.json golden byte-identical).
Diagnostic infra only; no functional change. Used to identify producers
of dispatch-state writes for the audit-017 / audit-019 hunt.
`GuestMemory::write_bulk` did the bulk copy via raw `copy_nonoverlapping`
without bumping page_versions for any of the pages it touched. The
per-byte `write_u8/u16/u32` methods all bump page_versions after their
store; downstream caches (texture cache, shader cache) Acquire-load the
slot to invalidate stale entries on guest writes. Without the bulk
bump, a caller like `NtReadFile` writing a texture/shader resource into
guest memory would leave any cache that had already keyed on the prior
version handing back stale decoded bytes.
After the copy, walk every page the write touched and bump it. Cheap:
the typical bulk write spans a few pages (NtReadFile uses 64-128 KB
chunks → 16-32 pages).
Reservation-table invalidation for `lwarx`/`stwcx.` (XMODBUG-001's
sibling) is NOT addressed here — the reservation table lives on
KernelState, not GuestMemory, and plumbing it through requires a wider
change. Callers that bulk-write code-bearing or atomic-bearing memory
should call `kernel.reservations.invalidate_for_write(addr)`
themselves; XEX-loader and NtReadFile are doing data-bearing writes
that don't intersect lwarx targets, so this is acceptable for now.
Verification at -n 100M lockstep:
swaps: 2 → 2 (unchanged)
draws: 0 → 0
texture_cache_entries: 0 → 0 (Sylpheed hasn't issued IM_LOAD yet
— the bump is silent until a cache
keys on a touched page, which won't
happen until Phase F2/F3 unblocks
the resource-loader workers)
packets: ~59M (within noise)
Tests: 16 memory pass.
Closes XMODBUG-002 (P1).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Re-shape MemoryAccess so write methods take &self and rely on interior
mutability (atomics in GuestMemory, Cell in test mocks). This unblocks
the &Arc<KernelState>-only execution model the CPU/HLE crates moved to.
GuestMemory grows: per-4 KiB-page write-version counter (page_version)
that the CPU's decode cache and the texture cache observe via Acquire,
fenced 32-bit/64-bit read/write helpers (Release on writer / Acquire on
reader) that PM4_EVENT_WRITE_SHD and the matching CPU consumers use to
synchronize fence publication, and broader page-table / heap accounting
needed by the new HLE allocators.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Rust reimplementation of the xenia Xbox 360 emulator targeting reverse-
engineering and preservation, initially scoped to Project Sylpheed.
Includes:
- XEX2 loader (LZX decompression, AES decryption, PE parsing)
- XISO / XGD2 disc image VFS
- PPC interpreter with 200+ opcodes and VMX128 decoding
- Static analyzer: functions, cross-references, labels, asm + SQLite output
- HLE kernel covering the xboxkrnl/xam subset used by Sylpheed init
- Debugger with in-memory and SQLite-backed execution tracing
- `xenia-rs` CLI with extract/dis/exec commands that produce cumulative,
superset SQLite databases and opt-in instruction/import/branch traces
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
