Consolidate exact trainer-serving parity - #57
Conversation
Broly Security ScanNote Summary 4 actionable finding(s) in this PR
All actionable items are in the table below. 1 finding(s) below the No finding is at or above
Dismiss false positivesTick a box to dismiss the finding; untick it to bring the finding back. That is the same as replying
Note Re-scan this PR anytime with
|
b0ad7c3 to
6a62d38
Compare
6a62d38 to
2bf6fe0
Compare
3b0ccb2 to
19c4328
Compare
19c4328 to
8156138
Compare
Round canonical MoE FP64 accumulators directly to low precision and use the current XoRL exact target. Publish HSDP adapter ownership and restore optimizer state safely across coherent replica-size changes.
| if isinstance(cleanup, R3SourceFilesCleanup): | ||
| for path in cleanup.paths: | ||
| try: | ||
| path.unlink(missing_ok=True) | ||
| except Exception as exc: | ||
| logger.warning("Failed to clean SGLang R3 source file %s: %s", path, exc) | ||
| return |
| if rows == 0: | ||
| pieces.append(torch.empty((0, *shape[1:]), dtype=expected_dtype)) | ||
| continue | ||
| storage = torch.from_file(str(path), shared=False, size=path.stat().st_size // 4, dtype=expected_dtype) |
| stub_dir = Path("/tmp/xorl-nvidia-ml-stub") | ||
| stub_dir.mkdir(exist_ok=True) | ||
| soname = stub_dir / "libnvidia-ml.so.1" | ||
| if not soname.exists(): | ||
| soname.symlink_to(stub) |
Summary
This consolidates the exact trainer-side work previously split across the earlier XoRL PRs into one branch based directly on
main. It is paired with togethercomputer/xorl-sglang#21.Highlights
Validation
Companion serving change
Out of scope
Sparse-delta receiver integration is intentionally not included.