Move swa memory pool to a seperate file (#16347)

python/sglang/srt/managers/schedule_batch.py

@@ -58,10 +58,7 @@ from sglang.srt.disaggregation.utils import DisaggregationMode
 from sglang.srt.distributed.parallel_state import get_tensor_model_parallel_rank
 from sglang.srt.environ import envs
 from sglang.srt.layers.attention.fla.chunk_delta_h import CHUNK_SIZE as FLA_CHUNK_SIZE
-from sglang.srt.mem_cache.allocator import (
-    BaseTokenToKVPoolAllocator,
-    SWATokenToKVPoolAllocator,
-)
+from sglang.srt.mem_cache.allocator import BaseTokenToKVPoolAllocator
 from sglang.srt.mem_cache.base_prefix_cache import BasePrefixCache
 from sglang.srt.mem_cache.common import (
     alloc_for_decode,
@@ -72,6 +69,7 @@ from sglang.srt.mem_cache.common import (
 from sglang.srt.mem_cache.mamba_radix_cache import MambaRadixCache
 from sglang.srt.mem_cache.memory_pool import ReqToTokenPool
 from sglang.srt.mem_cache.radix_cache import RadixKey
+from sglang.srt.mem_cache.swa_memory_pool import SWATokenToKVPoolAllocator
 from sglang.srt.metrics.collector import (
     DPCooperationInfo,
     SchedulerMetricsCollector,

python/sglang/srt/managers/schedule_policy.py

@@ -28,10 +28,10 @@ import torch
 
 from sglang.srt.layers.attention.nsa.utils import is_nsa_prefill_cp_in_seq_split
 from sglang.srt.managers.schedule_batch import Req, ScheduleBatch
-from sglang.srt.mem_cache.allocator import SWATokenToKVPoolAllocator
 from sglang.srt.mem_cache.base_prefix_cache import BasePrefixCache
 from sglang.srt.mem_cache.mamba_radix_cache import MambaRadixCache
 from sglang.srt.mem_cache.radix_cache import RadixCache, RadixKey, TreeNode
+from sglang.srt.mem_cache.swa_memory_pool import SWATokenToKVPoolAllocator
 from sglang.srt.mem_cache.swa_radix_cache import SWARadixCache
 from sglang.srt.server_args import ServerArgs