VoxelForge/Standalone/Research/WorldAssemblerM3/Linux104/RunL104ReadinessPairs.py

#!/usr/bin/env python3
"""Interleaved L103/L104 near and whole-view readiness at W8/W12."""
from __future__ import annotations

import json
import re
import statistics
import subprocess
import time
from pathlib import Path

ROOT = Path(__file__).resolve().parents[4]
OUT = ROOT / "Saved/Linux/l104/readiness-paired"
DATA = ROOT / "Standalone/data/owner_world.json"
BINARIES = {
    "before": ROOT / "Saved/Linux/l101/bin/VoxelForgeCoreHeadless-l101-clang",
    "after": ROOT / "Saved/Linux/l104/bin/VoxelForgeCoreHeadless-l104-clang",
}
STAMPS = {
    "before": "c5f49a45f51856b857a1ee8877b7d777208e67daf64281caefc7ef84e8006410",
    "after": "b23900c9d2ee10b0ee29bec1b1b7f74ee3fa697f5c66682e808b19ad9c334780",
}
STRATA = {
    "SW": {"camera": (8.0, 16.0, -372.2626953125), "near": 1408, "whole": 2850},
    "FI": {"camera": (25.8369140625, 3.19921875, -737.0), "near": 1216, "whole": 2249},
    "TN": {"camera": (19.166015625, -14.2001953125, -753.3740234375), "near": 1344, "whole": 2233},
    "VS": {"camera": (-20.0625, -5.068359375, -900.353515625), "near": 1344, "whole": 2329},
}
NEAR = re.compile(r"^\[WorldAssemblerNearLODReady\] lods=0-2 tiles=(\d+)/(\d+) cold_ready_s=([0-9.]+)$", re.M)
PUBLISHED = re.compile(r"^\[WorldAssemblerNearLODPublished\] lods=0-2 tiles=(\d+)/(\d+) cold_ready_s=([0-9.]+)$", re.M)
PIPELINE_READY = re.compile(r"^\[WorldAssemblerNearLODPipelineReady\] lods=0-2 tiles=(\d+)/(\d+) cold_ready_s=([0-9.]+)$", re.M)
PIPELINE = re.compile(r"^\[StreamPipeline\] (.*)$", re.M)
CONFIG = re.compile(r"^\[HeadlessConfig\] (.*)$", re.M)
STAMP = re.compile(r"^\[VoxelForgeSourceStamp\] sha256=(\S+)$", re.M)
FIELDS = re.compile(r"([a-zA-Z0-9_]+)=([^ ]+)")


def run(stratum: str, workers: int, build: str, phase: str, replicate: int) -> dict:
    OUT.mkdir(parents=True, exist_ok=True)
    suffix = f"-r{replicate}" if phase == "measured" else ""
    log = OUT / f"{stratum.lower()}-w{workers}-{build}-{phase}{suffix}.log"
    camera = ",".join(f"{value:.12f}" for value in STRATA[stratum]["camera"]) + ",0,0"
    command = [
        "taskset", "-c", "0-11", "prlimit", "--as=6442450944", "--", "env",
        f"VOXELFORGE_L104_FAR_HIT_CACHE={0 if build == 'before' else 1}",
        "timeout", "1740s", str(BINARIES[build]), str(DATA),
        "--world-assembler-prototype", "--stream-benchmark", "--stream-pipeline",
        f"--workers={workers}", "--clip-radius=3", "--max-clip-level=8", "--max-level=8",
        f"--camera={camera}", "--profile-lod-work", "--stream-scheduler-policy=3",
    ]
    started = time.monotonic()
    with log.open("w", encoding="utf-8") as stream:
        result = subprocess.run(command, cwd=ROOT, stdout=stream, stderr=subprocess.STDOUT,
                                check=False, timeout=1740)
    content = log.read_text(encoding="utf-8", errors="replace")
    near, published, ready = NEAR.search(content), PUBLISHED.search(content), PIPELINE_READY.search(content)
    pipeline, config, stamp = PIPELINE.search(content), CONFIG.search(content), STAMP.search(content)
    if result.returncode or not all((near, published, ready, pipeline, config, stamp)):
        raise RuntimeError(f"readiness run failed or markers missing: {log}; exit={result.returncode}")
    if stamp.group(1) != STAMPS[build]:
        raise RuntimeError(f"source stamp mismatch: {log}: {stamp.group(1)}")
    spec = STRATA[stratum]
    config_fields = dict(FIELDS.findall(config.group(1)))
    if int(config_fields["workers"]) != workers:
        raise RuntimeError(f"worker mismatch: {log}")
    for marker in (near, published, ready):
        if (int(marker.group(1)), int(marker.group(2))) != (spec["near"], spec["near"]):
            raise RuntimeError(f"incomplete near set: {log}: {marker.group(1)}/{marker.group(2)}")
    pipeline_fields = dict(FIELDS.findall(pipeline.group(1)))
    whole_done, whole_total = (int(value) for value in pipeline_fields["whole_tiles"].split("/"))
    if (whole_done, whole_total) != (spec["whole"], spec["whole"]):
        raise RuntimeError(f"incomplete whole set: {log}: {whole_done}/{whole_total}")
    row = {
        "stratum": stratum, "workers": workers, "build": build, "phase": phase,
        "replicate": replicate, "exit_code": result.returncode,
        "source_stamp": stamp.group(1), "near_tiles": spec["near"], "whole_tiles": whole_done,
        "lod02_ready_s": float(near.group(3)),
        "lod02_published_s": float(published.group(3)),
        "lod02_pipeline_ready_s": float(ready.group(3)),
        "whole_view_ready_s": float(pipeline_fields["whole_view_ready"].removesuffix("s")),
        "shape_s": float(pipeline_fields["shape"].removesuffix("s")),
        "far_shape_s": float(pipeline_fields["far_shape"].removesuffix("s")),
        "wall_s": round(time.monotonic() - started, 3),
        "log": str(log.relative_to(ROOT)),
    }
    print("[L104ReadinessRun] " + json.dumps(row, sort_keys=True), flush=True)
    return row


def main() -> int:
    for binary in BINARIES.values():
        if not binary.is_file():
            raise SystemExit(f"missing binary: {binary}")
    if not DATA.is_file():
        raise SystemExit(f"missing owner input: {DATA}")
    cells = [(stratum, workers) for stratum in ("VS", "TN", "FI", "SW")
             for workers in (12, 8)]
    for stratum, workers in cells:
        for build in ("before", "after"):
            run(stratum, workers, build, "warmup", 0)
    orders = (("before", "after"), ("after", "before"), ("before", "after"))
    rows = []
    for stratum, workers in cells:
        for replicate, order in enumerate(orders, 1):
            for build in order:
                rows.append(run(stratum, workers, build, "measured", replicate))
                (OUT / "runs-partial.json").write_text(json.dumps(rows, indent=2) + "\n",
                                                       encoding="utf-8")
    summary = []
    for stratum in STRATA:
        for workers in (8, 12):
            group = [row for row in rows if row["stratum"] == stratum and row["workers"] == workers]
            item = {"stratum": stratum, "workers": workers, "measured_runs_per_build": 3}
            for build in ("before", "after"):
                subset = [row for row in group if row["build"] == build]
                if len(subset) != 3:
                    raise RuntimeError(f"expected 3 measured runs: {stratum} W{workers} {build}")
                for metric in ("lod02_ready_s", "whole_view_ready_s"):
                    values = [row[metric] for row in subset]
                    item[f"{build}_{metric}"] = values
                    item[f"{build}_{metric}_median"] = statistics.median(values)
            for metric in ("lod02_ready_s", "whole_view_ready_s"):
                item[f"{metric}_delta_pct"] = 100 * (
                    item[f"after_{metric}_median"] / item[f"before_{metric}_median"] - 1)
            summary.append(item)
    result = {"runs": rows, "summary": summary}
    (OUT / "summary.json").write_text(json.dumps(result, indent=2) + "\n", encoding="utf-8")
    return 0


if __name__ == "__main__":
    raise SystemExit(main())