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>
Mirrors canary's cvars::avpack default (xam_info.cc:35) and Sylpheed's
accepted set {3,4,6,8} (xam_info.cc:250-251). With KRNBUG-XEX-001 having
flipped the priv-10 gate, XGetAVPack now reaches its caller in
sub_824AB578; returning 0x16 caused Sylpheed to abort the AV/crypto
block before XeCryptSha. Cascade walks one step (canary-only export
list 11 → 10); sub_824ABA98 is the next candidate.
Tests: 589 → 590. Goldens re-baselined (n50m: 50000005→50000004,
imports 407417→407416). Lockstep deterministic across 3 reruns at
-n 100M (instructions=100000010, import_calls=987686 +2.4×, swaps=2).
9-PC producer probe still 0×; parked handles 0x1004/0x100c/0x15e0
still signal_attempts=0.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
observability.rs installs the tracing subscriber stack (env-filter +
JSON file appender + chrome trace + error layer) and the metrics
recorder shared by the workspace. main.rs grows the new CLI surface:
--parallel, --reservations-table, --trace-handles, --analyze=
{rust,sql,both}, xenia dis --json, --ui, plus the wiring that runs
the CPU through the new scheduler, drives the GPU's threaded backend,
and surfaces the framebuffer + HUD via xenia-ui.
Add tests/parallel_stress.rs (#[ignore]-gated long form, short form
runs 20×@5M) and tests/golden/sylpheed_n2m.json — the digest the
lockstep/parallel combos compare against.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
