vllm/vllm/attention/backends/abstract.py

133 lines
3.8 KiB
Python

from abc import ABC, abstractmethod
from dataclasses import dataclass, fields
from typing import (Any, Dict, Generic, List, Optional, Set, Tuple, Type,
TypeVar)
import torch
class AttentionBackend(ABC):
"""Abstract class for attention backends."""
@staticmethod
@abstractmethod
def get_name() -> str:
raise NotImplementedError
@staticmethod
@abstractmethod
def get_impl_cls() -> Type["AttentionImpl"]:
raise NotImplementedError
@staticmethod
@abstractmethod
def make_metadata(*args, **kwargs) -> "AttentionMetadataPerStage":
raise NotImplementedError
@staticmethod
@abstractmethod
def get_kv_cache_shape(
num_blocks: int,
block_size: int,
num_kv_heads: int,
head_size: int,
) -> Tuple[int, ...]:
raise NotImplementedError
@staticmethod
@abstractmethod
def swap_blocks(
src_kv_cache: torch.Tensor,
dst_kv_cache: torch.Tensor,
src_to_dst: torch.Tensor,
) -> None:
raise NotImplementedError
@staticmethod
@abstractmethod
def copy_blocks(
kv_caches: List[torch.Tensor],
src_to_dists: torch.Tensor,
) -> None:
raise NotImplementedError
@dataclass
class AttentionMetadataPerStage:
"""Attention metadata for a specific stage. I.e., prefill or decode."""
def asdict_zerocopy(self,
skip_fields: Optional[Set[str]] = None
) -> Dict[str, Any]:
"""Similar to dataclasses.asdict, but avoids deepcopying."""
if skip_fields is None:
skip_fields = set()
# Note that if we add dataclasses as fields, they will need
# similar handling.
return {
field.name: getattr(self, field.name)
for field in fields(self) if field.name not in skip_fields
}
T = TypeVar("T", bound=AttentionMetadataPerStage)
@dataclass
class AttentionMetadata(Generic[T]):
"""Attention metadata for prefill and decode batched together."""
# Total number of prefill requests.
num_prefills: int
# Number of prefill tokens.
num_prefill_tokens: int
# Number of decode tokens. Note that it is equivalent to the number of
# decode requests.
num_decode_tokens: int
# The attention metadata for prefill requests in a batch.
# None if there's no prefill requests in a batch.
prefill_metadata: Optional[T]
# The attention metadata for decode requests in a batch.
# None if there's no decode requests in a batch.
decode_metadata: Optional[T]
# (num_tokens,). The indices of the token slots that input tokens will be
# stored into. E.g., if `slot_mapping` is [35, 2, 17] and the block size
# is 16, the three tokens are stored in the 3rd slot in block 2, 2nd slot
# in block 0, and 1st slot in block 1, respectively.
slot_mapping: torch.Tensor
# The kv cache's data type.
kv_cache_dtype: str
def __post_init__(self):
if self.num_prefill_tokens > 0:
assert self.num_prefills > 0
assert self.prefill_metadata is not None
if self.num_decode_tokens > 0:
assert self.decode_metadata is not None
class AttentionImpl(ABC):
@abstractmethod
def __init__(
self,
num_heads: int,
head_size: int,
scale: float,
num_kv_heads: Optional[int] = None,
alibi_slopes: Optional[List[float]] = None,
sliding_window: Optional[int] = None,
) -> None:
raise NotImplementedError
@abstractmethod
def forward(
self,
query: torch.Tensor,
key: torch.Tensor,
value: torch.Tensor,
kv_cache: torch.Tensor,
attn_metadata: AttentionMetadata,
kv_scale: float,
) -> torch.Tensor:
raise NotImplementedError