[HiCache] Support DCP with DSpark (#35221)

Co-authored-by: Cursor <cursoragent@cursor.com>

python/sglang/srt/mem_cache/hybrid_cache/hybrid_pool_assembler.py

@@ -1013,9 +1013,11 @@ def build_full_draft_pools(
     controller = tree_cache.cache_controller
     host_pool_group = controller.mem_pool_host
 
+    # Note(kpham-sgl): DCP x DSpark draft KV is replicated and spans the virtual
+    # loc space, so match the target host's logical_size instead of physical size.
     draft_host_pool = _build_mha_mla_host_pool(
         pool=pool,
-        host_to_device_ratio=host_pool_group.size / pool.size,
+        host_to_device_ratio=host_pool_group.logical_size / pool.size,
         page_size=controller.page_size,
         layout=server_args.hicache_mem_layout,
         allocator_type=_get_allocator_type(server_args),

python/sglang/srt/mem_cache/kv_cache_builder.py

@@ -117,7 +117,7 @@ def _register_legacy_hicache_draft(
     # so that host indices stay 1-to-1 between target and draft KV caches.
     primary_host_pool = tree_cache.cache_controller.mem_pool_host
     host_pool_kwargs = dict(
-        host_to_device_ratio=primary_host_pool.size / pool.size,
+        host_to_device_ratio=primary_host_pool.logical_size / pool.size,
         host_size=0,
         page_size=page_size,
         layout=server_args.hicache_mem_layout,

python/sglang/srt/server_args.py

@@ -7502,11 +7502,11 @@ class ServerArgs:
                 "backup and the storage keys must become dcp_rank-aware "
                 "first. Run HiCache+DCP with L1/L2 only."
             )
-        if self.speculative_algorithm is not None:
+        if self.speculative_algorithm not in (None, "DSPARK"):
             raise NotImplementedError(
-                "HiCache with --dcp-size > 1 does not support speculative "
-                "decoding yet (the draft-model host pool has no DCP index "
-                "translation)."
+                "HiCache with --dcp-size > 1 only supports DSPARK speculative "
+                "decoding; other draft-model host pools have no DCP index "
+                "translation."
             )
         if self.enable_lmcache:
             raise NotImplementedError(

test/registered/unit/mem_cache/test_hybrid_pool_assembler.py

@@ -2,6 +2,7 @@
 
 import unittest
 from types import SimpleNamespace
+from unittest.mock import patch
 
 from sglang.srt.mem_cache.hybrid_cache.hybrid_pool_assembler import (
     _split_hicache_size,
@@ -54,6 +55,39 @@ class TestDraftSidecarPoolDispatch(CustomTestCase):
         self.assertEqual(specs, [])
         self.assertEqual(entries, [])
 
+    def test_full_builder_sizes_sidecar_for_anchor_logical_space(self):
+        draft_kv_pool = SimpleNamespace(layer_num=1, size=800)
+        draft_host_pool = SimpleNamespace(layer_num=1)
+        tree_cache = SimpleNamespace(
+            cache_controller=SimpleNamespace(
+                mem_pool_host=SimpleNamespace(size=100, logical_size=800),
+                page_size=512,
+            )
+        )
+        server_args = SimpleNamespace(hicache_mem_layout="page_first")
+
+        with (
+            patch(
+                "sglang.srt.mem_cache.hybrid_cache.hybrid_pool_assembler."
+                "_build_mha_mla_host_pool",
+                return_value=draft_host_pool,
+            ) as build_host_pool,
+            patch(
+                "sglang.srt.mem_cache.hybrid_cache.hybrid_pool_assembler."
+                "_get_allocator_type",
+                return_value="default",
+            ),
+        ):
+            specs, entries = build_full_draft_pools(
+                draft_kv_pool=draft_kv_pool,
+                tree_cache=tree_cache,
+                server_args=server_args,
+            )
+
+        self.assertEqual(build_host_pool.call_args.kwargs["host_to_device_ratio"], 1.0)
+        self.assertEqual(len(specs), 1)
+        self.assertIs(entries[0].host_pool, draft_host_pool)
+
 
 if __name__ == "__main__":
     unittest.main()