biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[4/N][Attention] Move MLA common to model_executor (#32060)

Browse Source 
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
This commit is contained in:
Matthew Bonanni
2026-01-13 12:08:45 -05:00
committed by GitHub
parent 4f3676e726
commit 2263d44b68
14 changed files with 50 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/distributed/kv_transfer/kv_connector/v1/example_connector.py
View File

@@ -14,9 +14,9 @@ from vllm.distributed.kv_transfer.kv_connector.v1.base import (
KVConnectorRole,
)
from vllm.logger import init_logger
from vllm.model_executor.layers.attention.mla_attention import MLACommonMetadata
from vllm.utils.hashing import safe_hash
from vllm.v1.attention.backend import AttentionMetadata
from vllm.v1.attention.backends.mla.common import MLACommonMetadata
from vllm.v1.core.sched.output import SchedulerOutput
if TYPE_CHECKING:

2
vllm/distributed/kv_transfer/kv_connector/v1/p2p/p2p_nccl_connector.py
View File

@@ -18,8 +18,8 @@ from vllm.distributed.kv_transfer.kv_connector.v1.p2p.p2p_nccl_engine import (
)
from vllm.distributed.parallel_state import get_world_group
from vllm.logger import init_logger
from vllm.model_executor.layers.attention.mla_attention import MLACommonMetadata
from vllm.v1.attention.backend import AttentionMetadata
from vllm.v1.attention.backends.mla.common import MLACommonMetadata
from vllm.v1.core.sched.output import SchedulerOutput
if TYPE_CHECKING: