Avoid index prefetch fallback noise when disabled

The missing index prefetcher state is expected when MLA/index prefetch is disabled. Logging it as an index-prefetch fallback made the normal synchronous materialize path look like a failed fast path and created noisy reports. This keeps consume-miss warnings for enabled prefetchers, but suppresses missing-prefetcher warnings when the prefetch feature gate is off. Constraint: Index prefetch is controlled by the same MLA prefetch enable gate in the current pipeline. Rejected: Suppress all index prefetch fallback warnings | consume misses still indicate an enabled prefetch pipeline failed to serve a requested buffer. Confidence: high Scope-risk: narrow Directive: Missing prefetcher is not a fallback when SGLANG_CP_SHARED_KV_ENABLE_MLA_PREFETCH=0; do not re-add warning noise without a separate enable gate. Tested: Local py_compile for nsa_indexer.py and test_cp_shared_kv_runtime.py. Tested: Remote g0034 targeted regression set passed 4 tests: disabled sync materialize no warning, consume miss warning, first-layer no-warning, enabled create-skip warning. Not-tested: Full CUDA ETE run with logs.
2026-06-02 23:52:46 +08:00
parent 401de0f8ce
commit 937f89ef89
3 changed files with 101 additions and 19 deletions
@@ -3658,6 +3658,50 @@ class TestCpSharedKVTaiMaterializeIntegration(unittest.TestCase):
        )
        self.assertIn("cuda_unavailable_or_stream_capturing", logger.call_args.args[1])

+    def test_index_sync_materialize_does_not_warn_when_prefetch_disabled(self):
+        from sglang.srt.environ import envs
+        from sglang.srt.layers.attention.nsa import nsa_indexer
+
+        class FakePool:
+            page_size = 4
+
+            def get_index_k_with_scale_buffer(self, layer_id):
+                return torch.arange(0, 12, dtype=torch.float32).view(4, 3)
+
+        class FakeLayout:
+            page_size = 4
+            cp_size = 1
+            cp_rank = 0
+
+        fallback_buffer = torch.full((3, 3), 9.0)
+        fallback_pages = torch.tensor([[1, 2]], dtype=torch.int32)
+        forward_batch = SimpleNamespace(
+            token_to_kv_pool=FakePool(),
+            uses_cp_shared_kv=True,
+            cp_shared_kv_layout=FakeLayout(),
+            cp_shared_kv_index_prefetcher=None,
+        )
+        logical_pages = torch.tensor([[1, 2]], dtype=torch.int32)
+        indexer = object.__new__(nsa_indexer.Indexer)
+
+        with envs.SGLANG_CP_SHARED_KV_ENABLE_MLA_PREFETCH.override(False), patch.object(
+            nsa_indexer,
+            "materialize_shared_paged_buffer",
+            return_value=(fallback_buffer, fallback_pages),
+        ), patch(
+            "sglang.srt.layers.attention.nsa.nsa_indexer.logger",
+            create=True,
+        ) as logger:
+            dense_buffer, dense_pages = indexer._maybe_materialize_shared_index_buffer(
+                forward_batch,
+                layer_id=7,
+                logical_page_table=logical_pages,
+            )
+
+        self.assertIs(dense_buffer, fallback_buffer)
+        self.assertIs(dense_pages, fallback_pages)
+        logger.warning.assert_not_called()
+

 if __name__ == "__main__":
    unittest.main()