Skip to content

[None][perf] eliminate KDA recurrent-state and metadata copies #2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
jiaganc wants to merge 1 commit into from
Closed

[None][perf] eliminate KDA recurrent-state and metadata copies #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 12 additions & 3 deletions tensorrt_llm/_torch/attention_backend/interface.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -43,6 +43,13 @@
ALLOWED_LAYER_TYPES as ALLOWED_ATTENTION_LAYER_TYPES


def _default_mamba_metadata_cls() -> Type[Any]:
"""Import lazily to avoid an interface/Mamba metadata import cycle."""
from ..modules.mamba.mamba2_metadata import Mamba2Metadata

return Mamba2Metadata


@dataclass
class AttentionRuntimeFeatures:
chunked_prefill: bool = False
Expand Down Expand Up @@ -173,6 +180,8 @@ class AttentionMetadata:

mamba_metadata: Optional[Any] = None
mamba_chunk_size: int = 128
mamba_metadata_cls: Type[Any] = field(
default_factory=_default_mamba_metadata_cls, init=False, repr=False)

# The number of tokens in the padded sequence.
padded_num_tokens: Optional[int] = None
Expand Down Expand Up @@ -347,9 +356,9 @@ def _prepare_mamba_metadata(self):

if self.mamba_metadata is None:
if isinstance(self.kv_cache_manager, BaseMambaCacheManager):
from ..modules.mamba.mamba2_metadata import Mamba2Metadata
self.mamba_metadata = Mamba2Metadata(self.max_num_requests,
self.mamba_chunk_size)
metadata_cls = self.mamba_metadata_cls
self.mamba_metadata = metadata_cls(self.max_num_requests,
self.mamba_chunk_size)
else:
self.mamba_metadata = False
return
Expand Down
Loading
Loading