[bugfix][DCP] fix block_size of hash in DCP prefix caching (#26296)

Signed-off-by: Chen Zhang <zhangch99@outlook.com>
This commit is contained in:
Chen Zhang
2025-10-10 03:02:49 -07:00
committed by GitHub
parent 720d3cd0f0
commit 606b00e80f
5 changed files with 12 additions and 10 deletions
+2 -8
View File
@@ -45,6 +45,7 @@ class Scheduler(SchedulerInterface):
vllm_config: VllmConfig,
kv_cache_config: KVCacheConfig,
structured_output_manager: StructuredOutputManager,
block_size: int,
mm_registry: MultiModalRegistry = MULTIMODAL_REGISTRY,
include_finished_set: bool = False,
log_stats: bool = False,
@@ -101,15 +102,8 @@ class Scheduler(SchedulerInterface):
num_gpu_blocks = self.cache_config.num_gpu_blocks
assert num_gpu_blocks is not None and num_gpu_blocks > 0
self.block_size = self.cache_config.block_size
self.block_size = block_size
self.dcp_world_size = vllm_config.parallel_config.decode_context_parallel_size
# Note(hc): The scheduler’s block_size must be multiplied
# by dcp_world_size, since block hashes are computed on the
# original full token sequence at a granularity of
# original_block_size × dcp_world_size.
if self.dcp_world_size > 1:
self.block_size *= self.dcp_world_size
# req_id -> Request
self.requests: dict[str, Request] = {}