[KV Connector] Support symmetric DCP disagg for hybrid mamba models (#55531)

Signed-off-by: Wei Zhao <weizha@oci-aga-slurm-1-vscode-02.cm.cluster>
Signed-off-by: Wei Zhao <51183510+wzhao18@users.noreply.github.com>
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com>
Co-authored-by: Wei Zhao <weizha@oci-aga-slurm-1-vscode-02.cm.cluster>
Co-authored-by: OpenAI Codex <noreply@openai.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
Wei Zhao
2026-09-10 11:12:15 -07:00
committed by GitHub
co-authored by Wei Zhao OpenAI Codex mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent 983b7e28c9
commit 7cdd9304ae
5 changed files with 31 additions and 12 deletions
+13 -3
View File
@@ -487,7 +487,7 @@ class EngineArgs:
dcp_comm_backend: DCPCommBackend | None = ParallelConfig.dcp_comm_backend
dcp_q_replicate: bool | None = ParallelConfig.dcp_q_replicate
dcp_kv_cache_interleave_size: int = ParallelConfig.dcp_kv_cache_interleave_size
cp_kv_cache_interleave_size: int = ParallelConfig.cp_kv_cache_interleave_size
cp_kv_cache_interleave_size: int | None = None
data_parallel_size: int = ParallelConfig.data_parallel_size
data_parallel_rank: int | None = None
data_parallel_start_rank: int | None = None
@@ -1104,7 +1104,10 @@ class EngineArgs:
)
parallel_group.add_argument(
"--cp-kv-cache-interleave-size",
**parallel_kwargs["cp_kv_cache_interleave_size"],
**{
**parallel_kwargs["cp_kv_cache_interleave_size"],
"default": None,
},
)
parallel_group.add_argument(
"--prefill-context-parallel-size",
@@ -2363,7 +2366,14 @@ class EngineArgs:
dcp_comm_backend=self.dcp_comm_backend,
dcp_q_replicate=self.dcp_q_replicate,
dcp_kv_cache_interleave_size=self.dcp_kv_cache_interleave_size,
cp_kv_cache_interleave_size=self.cp_kv_cache_interleave_size,
cp_kv_cache_interleave_size=(
self.cp_kv_cache_interleave_size
if self.cp_kv_cache_interleave_size is not None
else ParallelConfig.cp_kv_cache_interleave_size
),
_allow_auto_resolve_cp_interleave_size=(
self.cp_kv_cache_interleave_size is None
),
_api_process_count=self._api_process_count,
_api_process_rank=self._api_process_rank,
assigned_physical_gpu_ids=self._resolve_device_ids(),