Bound CP prefill batching by estimated temp memory

CP shared-KV bs>1 batching was only bounded by request count, extend tokens, and cached tokens. That left temporary GPU buffers such as MLA/index materialization, remap metadata, logits windows, and transfer descriptors implicit, and raw extend-token limits could exceed the active chunked-prefill budget.\n\nThis adds an explicit max-buffer-size admission gate with a CPU-only stream-aware estimator, wires it through PrefillAdder/Scheduler, performs a startup CUDA smoke allocation when configured, and reports the estimate in the scheduler admission benchmark. When chunked prefill is active, the effective CP extend-token limit is capped by the current chunk budget so the CP path does not advertise unreachable batch capacity or lift max-prefill-tokens too far.\n\nConstraint: Admission estimation must stay CPU-only on the scheduler hot path; CUDA allocation is limited to startup smoke checking.\nConstraint: Single oversized requests must still be allowed to run alone to avoid scheduler deadlock.\nRejected: Rely only on --max-prefill-tokens | it does not reliably bound the first oversized request and does not model cache-hit/load-back pressure.\nRejected: Let CP extend limit exceed chunked-prefill size | it creates an unreachable effective capacity and misleading budget lift.\nConfidence: medium\nScope-risk: moderate\nDirective: If bs>1 L1 prefetch is enabled later, update CPSharedKVPrefillBufferEstimatorContext.bs_gt1_l1_prefetch_enabled and include the live prefetch dense buffers in overlap windows.\nTested: local py_compile for touched files\nTested: local PYTHONPATH=python pytest -q test/registered/unit/managers/test_cp_shared_kv_prefill_buffer_estimator.py (4 passed)\nTested: remote cjy-glm5-new targeted pytest for new server_args, PrefillAdder, estimator, and benchmark cases (10 passed)\nTested: remote cjy-glm5-new PYTHONPATH=python pytest -q test/registered/unit/managers/test_cp_shared_kv_prefill_buffer_estimator.py test/registered/unit/managers/test_prefill_adder.py test/registered/unit/managers/test_prefill_scheduler_admission_bench.py (29 passed before chunk cap, then test_prefill_adder.py 21 passed after chunk cap)\nNot-tested: full server_args suite because existing TestPrepareServerArgs tries to reach HuggingFace and fails under container DNS/network\nNot-tested: GLM5 ETE smoke with --cp-shared-kv-prefill-max-buffer-size

test/registered/unit/server_args/test_server_args.py

@@ -3,6 +3,8 @@ import tempfile
 import unittest
 from unittest.mock import MagicMock, patch
 
+import pytest
+
 from sglang.srt.server_args import PortArgs, ServerArgs, prepare_server_args
 from sglang.test.ci.ci_register import register_cpu_ci
 from sglang.test.test_utils import (
@@ -75,6 +77,60 @@ def test_cp_shared_kv_prefill_bs_gt1_parser_limits():
     assert args.cp_shared_kv_prefill_max_total_extend_tokens == 8192
 
 
+def test_cp_shared_kv_prefill_max_buffer_size_defaults_to_gb():
+    import argparse
+
+    parser = argparse.ArgumentParser()
+    ServerArgs.add_cli_args(parser)
+    raw_args = parser.parse_args(
+        [
+            "--model-path",
+            "dummy",
+            "--enable-cp-shared-kv-prefill-bs-gt1",
+            "--cp-shared-kv-prefill-max-buffer-size",
+            "8",
+        ]
+    )
+    args = ServerArgs.from_cli_args(raw_args)
+    assert args.cp_shared_kv_prefill_max_buffer_size == 8_000_000_000
+
+
+def test_cp_shared_kv_prefill_max_buffer_size_accepts_iec_suffix():
+    import argparse
+
+    parser = argparse.ArgumentParser()
+    ServerArgs.add_cli_args(parser)
+    raw_args = parser.parse_args(
+        [
+            "--model-path",
+            "dummy",
+            "--enable-cp-shared-kv-prefill-bs-gt1",
+            "--cp-shared-kv-prefill-max-buffer-size",
+            "8Gi",
+        ]
+    )
+    args = ServerArgs.from_cli_args(raw_args)
+    assert args.cp_shared_kv_prefill_max_buffer_size == 8 * 2**30
+
+
+def test_cp_shared_kv_prefill_max_buffer_size_rejects_non_positive_value():
+    import argparse
+
+    parser = argparse.ArgumentParser()
+    ServerArgs.add_cli_args(parser)
+    raw_args = parser.parse_args(
+        [
+            "--model-path",
+            "dummy",
+            "--enable-cp-shared-kv-prefill-bs-gt1",
+            "--cp-shared-kv-prefill-max-buffer-size",
+            "0",
+        ]
+    )
+    with pytest.raises(ValueError, match="cp_shared_kv_prefill_max_buffer_size"):
+        ServerArgs.from_cli_args(raw_args)
+
+
 def test_hicache_mem_layout_parser_accepts_layer_page_first():
     import argparse