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
tests/v1/attention/test_mla_backends.py
View File

@@ -19,12 +19,12 @@ from tests.v1.attention.utils import (
)
from vllm import _custom_ops as ops
from vllm.config.vllm import set_current_vllm_config
from vllm.model_executor.layers.attention.mla_attention import QueryLenSupport
from vllm.model_executor.layers.attention_layer_base import AttentionLayerBase
from vllm.utils.math_utils import cdiv
from vllm.utils.torch_utils import STR_DTYPE_TO_TORCH_DTYPE
from vllm.v1.attention.backend import CommonAttentionMetadata
from vllm.v1.attention.backends.fa_utils import flash_attn_supports_mla
from vllm.v1.attention.backends.mla.common import QueryLenSupport
from vllm.v1.attention.backends.registry import AttentionBackendEnum
from vllm.v1.attention.ops.flashmla import is_flashmla_dense_supported
from vllm.v1.kv_cache_interface import FullAttentionSpec