# round 16 track b — decode wave zip and table barrier guards

## experiment contract
- owner: `Window::layer` dispatch scheduling and gemv kernel table initialization. only the assigned decode block and `gemv.metal` change, in `/tmp/sol_r16_dec/strata`.
- hypothesis: zip independent resident moe and shared expert work into three waves rather than four, preserving all data dependencies and fallback dispatch order; remove table-init barriers only where the compile-time table has one dummy element.
- constraints: no live server interaction or full-model runs; each standalone test invocation has a 4.9s deadline. no new tracked vendor edits. gemv_x retains its activation-staging barrier.
- evaluator: baseline/current scratch kernels on identical seeded inputs, memcmp of output bytes, plus short interleaved gpu timings. six non-table formats × ntok 1..4, gemv and cat. table formats and pair covered where existing tests permit. require exact output equality; speed differences below 3% treated as noise without replication.
- scheduling proof: first wave moe_gu + shared gate/up/gate-inp; second moe_down + swiglu_prep; third shared down. fallback sequence/barriers must remain identical. kernel unit tests cannot establish full-model scheduling parity; report this limitation.
- budget: local short unit invocations only, serial gpu measurements, up to three repeated comparison passes. build cpu work may run alongside inspection. stop or narrow tests that exceed 4.9s; do not silently describe timed-out tests as passing.

## implementation

patch: `/tmp/sol_r16_dec/sol_r16_dec.patch` (sha256 `602c139238113e75e1ac528109215f040645d617b8590051b2610fc3cdbae09d`). it applies with `git -C vendor/strata apply --check`. no tracked vendor file was edited. base checkout HEAD: `296b59d2f73c8f04a805df29c2fa97655adace78`; the checkout was already dirty, so exact before/after file hashes are recorded in `/tmp/sol_r16_dec/metadata.json`. baseline is the copied working tree, not pristine HEAD.

- two shared-work lambdas let the split path emit resident gu beside shared gate/up/gate-inp, resident down beside shared prep, then shared down. phase 2 remains after the host wait. this is **one fewer internal buffer barrier per layer**, hence **48 fewer for a 48-layer decode window**, by schedule construction, not an end-to-end timing result.
- `!sigs_` / `!esplit` retain the original dispatch/barrier order. signal and cut positions retain their dispatch indices (same number of dispatches).
- gemv/cat only initialize and synchronize real tables. pair synchronizes if either format needs a table. gemv_x only guards the no-op table load: its activation staging barrier stays unconditional. the KS>1 reduction barrier is untouched.
- no extra buffers, allocations, public interfaces, persistent state, or new feature flags.

## build and correctness

machine: apple m5, macos 27.0 build 26a428, safe metal math (default). every standalone execution had `timeout=4.9`; cpu builds were not subject to the test deadline.

build succeeded:

```text
cmake -S /tmp/sol_r16_dec/strata/metal -B /tmp/sol_r16_dec/strata/build-metal -G Ninja -DCMAKE_BUILD_TYPE=Release -DMETAL_CPP_DIR=/tmp/sol_r16_dec/strata/third_party/metal-cpp -DSTRATA_GGML_DIR=/Users/dawn/proj/swift1.5-qwen3.8-next-flash-ps-scr-qat-persona/vendor/llama.cpp
cmake --build /tmp/sol_r16_dec/strata/build-metal --target strata_window_cli gemv_parity route_parity fuse_parity expert_parity pair_parity -j 4
```

- `route_parity`: pass, k=512 and k=256+moce, 200 windows each, zero mismatches; 2.406s.
- `expert_parity`: pass, default q2_0/q2_0, ntok=1; h relative l2 1.81e-7, y 6.13e-8; 0.477s.
- `pair_parity`: all pass, including table/non-table pairs and raw q2_0 down; 3.338s.
- unmodified `gemv_parity` and `fuse_parity`: **deadline exceeded at 4.9s**, not passes. logs retained. no `moe_parity` target/source exists; an explicit build request reports unknown target. expert parity and the zip test below supply relevant coverage, but are not a fictitious pass of that missing target.
- gemv suite sharded **outside the strata tree**, changing only its format/token iteration lists: 39/39 invocations pass (13 formats × ntok 1,3,8); original shapes, tile list, double reference, and token-wise bitwise comparisons remain. maximum invocation 2.828s. source: `/tmp/sol_r16_dec/gemv_shard.cpp`; logs `gemv_shard_*.log`.
- fuse suite sharded outside strata only by token count: ntok 2..8 pass, including cat, bf16/f32 fused router, and combined norm variants. **ntok=1 still hits the 4.9s deadline**, including a line-buffered diagnostic attempt; no diagnostic device/result line was emitted before termination. do not treat the full fuse suite as passing. logs `fuse_shard_*.log` and `fuse_nt1_diagnostic.log`.
- actual owned scheduling block extracted unchanged from baseline and candidate into an external mock fixture: **32/32 configurations pass** (split flag off/on × signal off/on × router fusion off/on × streaming off/on × predictor off/on). fallback pipeline order/barrier flags and signal/cut indices match. split path explicitly asserts the seven dispatches and `[true,false,false,false,true,false,true]` flags, with 3→2 internal barriers. fixture mocks payloads; this is not numerical full-model validation. source `/tmp/sol_r16_dec/schedule_probe.cpp`, logs `schedule_0.log`, `schedule_1.log`.
- real metal zip parity: ntok 1..4 × zero/mixed/all resident sets, 8 repetitions each, comparing **8 full intermediate/output buffers** (`h`, `ye`, shared g/u/gate-inp, prepared shared h, shared down, final combine). **768/768 buffer comparisons equal byte-for-byte**. uses arena addressing, phase-1 cached slots, phase-2 missing experts, indirect plans, concurrent encoders, and real q2_0 kernels at 2560/640 dimensions. both schedules use identical guarded kernels, isolating scheduling. max invocation 3.557s. source `/tmp/sol_r16_dec/zip_probe.cpp`, logs `zip_probe_1.log` through `_4.log`.

## guard-only benchmark and exact parity

external harness `/tmp/sol_r16_dec/guard_probe.cpp` compiles baseline and candidate kernel function bodies unchanged, with reduced template instantiations to fit the short invocation budget. same seeded weights/inputs, safe math, prepared activations, and paired processes. gemv/cat/x use tile `16,1,4,1` (cat/x have no KS); pair uses `32,4`. shape 2560→640; cat is 640+640 rows, not the production 48+48 alpha/beta shape. six requested formats × ntok 1..4 × 3 independent invocations = **72/72 successful invocations**, maximum wall time **0.785s**. each compares gemv, cat (both outputs), x, and same-format pair to the baseline with `memcmp`: **100% exact**.

each kernel warms both variants together for 50ms, then runs five paired batches of 64 dispatches, alternating baseline/candidate order; gpu time per dispatch is the metric. table below reports median base and candidate times across the three invocations. raw logs retain the paired ratios; medians of times and ratios need not divide identically.

| format | ntok | gemv base → guard (µs) | cat base → guard (µs) |
|---|---:|---:|---:|
| Q6_K | 1 | 7.027 → 6.981 | 18.323 → 18.270 |
| Q6_K | 2 | 9.395 → 9.395 | 19.953 → 20.047 |
| Q6_K | 3 | 13.047 → 13.023 | 25.964 → 26.168 |
| Q6_K | 4 | 15.890 → 15.896 | 30.842 → 31.098 |
| Q8_0 | 1 | 6.771 → 6.738 | 22.783 → 22.660 |
| Q8_0 | 2 | 9.251 → 9.216 | 23.932 → 23.532 |
| Q8_0 | 3 | 12.055 → 12.013 | 26.114 → 26.232 |
| Q8_0 | 4 | 14.654 → 14.684 | 29.785 → 30.182 |
| Q2_0 | 1 | 5.027 → 4.956 | 8.148 → 8.139 |
| Q2_0 | 2 | 6.869 → 6.835 | 11.895 → 11.855 |
| Q2_0 | 3 | 9.073 → 9.010 | 15.979 → 15.977 |
| Q2_0 | 4 | 11.242 → 11.211 | 20.301 → 20.246 |
| BF16 | 1 | 21.738 → 21.574 | 42.350 → 42.387 |
| BF16 | 2 | 21.796 → 21.764 | 43.016 → 43.252 |
| BF16 | 3 | 21.828 → 21.744 | 43.326 → 43.423 |
| BF16 | 4 | 21.953 → 21.965 | 44.823 → 45.260 |
| Q4_K | 1 | 6.943 → 7.012 | 11.160 → 11.105 |
| Q4_K | 2 | 8.878 → 8.891 | 14.748 → 14.701 |
| Q4_K | 3 | 13.403 → 13.368 | 23.375 → 23.287 |
| Q4_K | 4 | 16.223 → 16.186 | 29.035 → 28.949 |
| Q5_K | 1 | 7.529 → 7.443 | 12.816 → 12.797 |
| Q5_K | 2 | 9.283 → 9.304 | 16.802 → 16.758 |
| Q5_K | 3 | 13.953 → 13.938 | 25.884 → 25.851 |
| Q5_K | 4 | 16.850 → 16.785 | 31.630 → 31.506 |


**no credible guard-only speedup established.** geometric mean of per-case median paired base/candidate ratios: gemv **1.0026×**, cat **0.9998×**, x **0.9988×**, pair **0.9985×**. gemv case ratios range 0.9838–1.0156; cat 0.9843–1.0117. all lie inside the predeclared 3% noise threshold after replication. this is a correctness-validated barrier cleanup, not evidence for a throughput claim. compiler elimination of redundant barriers is plausible, but not proven by these measurements.

## artifacts and limits

all large artifacts are under `/tmp/sol_r16_dec/`: `build.log`, `metadata.json`, `probe_runs.json`, `probe_summary.json`, `probe_*.log`, harness sources/binaries, and baseline/candidate reduced msl. exact full baseline files can be recovered from the scratch git index (`git -C /tmp/sol_r16_dec/strata show :metal/kernels/gemv.metal`, likewise window.cpp).

reproduce probes using `guard_probe <format-id> <ntok>` (14,8,42,30,12,13), `zip_probe <ntok>`, `schedule_probe` with `STRATA_METAL_EXPERT_SPLIT=0/1`, `gemv_shard` with `TEST_TYPE`/`TEST_NTOK`, and `fuse_shard` with `TEST_NTOK`; always enforce the 4.9s deadline. harnesses link scratch `build-metal/libmtl_rt.a`, include scratch `metal/src`, and use `-std=c++20 -O2 -framework Metal -framework Foundation`.

reviewed the complete production diff and surrounding code with `view_diff`/`view_code`; existing moe table guards provide the matching idiom. checked patch application and file/line scope. no live server request, port 8080 operation, full-model run, or full-model performance claim. synthetic parity does not establish real-model answer parity or host-overlap wall-time benefit. the uncompleted fuse ntok=1 check remains a validation limitation.