Add sglang sglang launch log summary

[ghangz]

Summary

• Adds a focused sglang launch log summary improvement for MetaX-MACA/sglang.
• The change targets MetaX MACA development and validation workflows, with emphasis on earlier diagnostics, reproducible logs, or safer benchmark tooling.
• Existing default behavior is kept compatible; the new logic is scoped to explicit checks, helper tools, or validation metadata.

Validation

• Verified on Gitee.AI MetaX GPU resources: sglang_SGLang image batch, 12/12 PASS; PyTorch-MACA batch also covered SGLang tooling.
• Branch validation command: python tools/maca_launch_log_summary.py --self-test
• Pull request text is intentionally ASCII-only to avoid encoding issues on web forms and API clients.

Review notes

• Source branch: ghangz:mengz/sglang-launch-log-summary
• Target branch: MetaX-MACA/sglang:main
• Maintainers can modify this branch if follow-up adjustments are needed.

[gemini-code-assist]

Code Review

This pull request introduces a new Python script, tools/maca_launch_log_summary.py, which parses and summarizes service, benchmark, or distributed runtime logs into JSON format by extracting metrics and errors. Feedback suggests reading the log files line-by-line rather than loading entire files into memory to avoid potential Out-Of-Memory (OOM) errors, and wrapping file operations in a try-except block to gracefully handle I/O errors.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

Reading the entire log file into memory using path.read_text().splitlines() can lead to high memory consumption or Out-Of-Memory (OOM) errors if the log files are extremely large. Additionally, if any of the specified log files do not exist or cannot be read, the script will crash with an unhandled exception.

It is highly recommended to open the file and iterate over it line-by-line. Wrapping the file operations in a try-except block to catch OSError will also make the parser much more robust by reporting the error in the output JSON instead of crashing.

Suggested change

	def parse(path: Path) -> dict[str, object]:
	errors: list[dict[str, object]] = []
	values: list[dict[str, object]] = []
	for lineno, line in enumerate(path.read_text(encoding="utf-8", errors="replace").splitlines(), 1):
	if ERROR_RE.search(line):
	errors.append({"line": lineno, "text": line.strip()})
	for match in NUMBER_RE.finditer(line):
	key = match.group("key").lower()
	if not METRICS or any(token in key for token in METRICS):
	values.append({"line": lineno, "metric": key, "value": float(match.group("value"))})
	return {"path": str(path), "metric_count": len(values), "error_count": len(errors), "metrics": values, "errors": errors}
	def parse(path: Path) -> dict[str, object]:
	errors: list[dict[str, object]] = []
	values: list[dict[str, object]] = []
	try:
	with path.open(encoding="utf-8", errors="replace") as f:
	for lineno, line in enumerate(f, 1):
	if ERROR_RE.search(line):
	errors.append({"line": lineno, "text": line.strip()})
	for match in NUMBER_RE.finditer(line):
	key = match.group("key").lower()
	if not METRICS or any(token in key for token in METRICS):
	values.append({"line": lineno, "metric": key, "value": float(match.group("value"))})
	except OSError as e:
	errors.append({"line": 0, "text": f"Failed to read file: {e}"})
	return {"path": str(path), "metric_count": len(values), "error_count": len(errors), "metrics": values, "errors": errors}

References

1. Avoid loading entire files into memory at once when processing line-by-line to prevent high memory usage and potential OOM errors. Always handle potential I/O errors (like FileNotFoundError or PermissionError) gracefully.