chenchenghao/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 11 Packages Projects Releases Wiki Activity

Standardize CP shared-KV runtime fallback markers

Browse Source 
Current reuse, TAI materialize, and TAI fused MLA store are optimized CP shared-KV paths. Their fallback helpers already emitted warnings, but the messages were not consistently grep-able with the standard CP shared-KV fallback marker.

This moves the marker into the helper layer so individual fallback call sites do not have to remember to include it, and keeps the existing per-reason rate limiting.

Constraint: Runtime fallback logs must be visible without enabling debug logging.

Rejected: Prefix only selected call sites | helper-level prefixing avoids future unmarked fallback messages.

Confidence: high

Scope-risk: narrow

Directive: New CP shared-KV runtime fallback helpers should use [CP_SHARED_KV_FALLBACK] and a component name.

Tested: Local py_compile for cp_shared_kv_runtime.py and test_cp_shared_kv_runtime.py.

Tested: Local extracted assertLogs check for all three fallback helper prefixes.

Tested: Remote g0034 pytest exact runtime helper tests: 2 passed, 3 warnings.

Not-tested: Full cp_shared_kv_runtime.py suite in this slice.

Co-authored-by: OmX <omx@oh-my-codex.dev>
This commit is contained in:
laoyao0822
2026-05-29 06:02:13 +08:00
parent e09c8256d7
commit dd69c5970c
3 changed files with 66 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
docs/advanced_features/nsa_prefill_cp_page_aligned_cache_contract.md
View File

@@ -812,6 +812,22 @@ Implemented C11 slice:
- Tightened the existing unit expectation so repeated fallback events are still
emitted, but now require the production-visible fallback prefix.
Additional C11 finding:
- Runtime fallback helpers for current reuse, TAI materialize, and TAI fused MLA
store already log at WARNING, but their messages do not use the standard
`[CP_SHARED_KV_FALLBACK]` marker. These helpers cover intended optimized hot
paths, so they should standardize the marker instead of requiring every caller
to remember the prefix.
Implemented C11 runtime-helper slice:
- Standardized current-reuse, TAI materialize, and TAI fused MLA store fallback
helpers to prefix warnings with `[CP_SHARED_KV_FALLBACK]` and a component
name.
- Added unit expectations for those helper prefixes and tightened the draft
partial-current-reuse fallback check to require the standard marker.
### C12. Verification gaps before claiming completion
Current state: