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

SERVING -> RUNNING

Browse Source
This commit is contained in:
Woosuk Kwon 2023-02-12 08:25:05 +00:00
parent 46958cf941
commit a2a9869cb7
2 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
cacheflow/master/block_manager.py
View File

@ -95,7 +95,7 @@ class BlockSpaceManager:
# Simple heuristic: If there is at least one free block # Simple heuristic: If there is at least one free block
# for each sequence, we can append. # for each sequence, we can append.
num_free_gpu_blocks = self.gpu_allocator.get_num_free_blocks() num_free_gpu_blocks = self.gpu_allocator.get_num_free_blocks()
num_seqs = seq_group.num_seqs(status=SequenceStatus.SERVING) num_seqs = seq_group.num_seqs(status=SequenceStatus.RUNNING)
return num_seqs <= num_free_gpu_blocks return num_seqs <= num_free_gpu_blocks
def append(self, seq: Sequence) -> Optional[Tuple[int, int]]: def append(self, seq: Sequence) -> Optional[Tuple[int, int]]:

2
cacheflow/sequence.py
View File

@ -7,7 +7,7 @@ from cacheflow.decoding import DecodingParams
class SequenceStatus(enum.Enum): class SequenceStatus(enum.Enum):
PENDING = enum.auto() PENDING = enum.auto()
SERVING = enum.auto() RUNNING = enum.auto()
SWAPPED = enum.auto() SWAPPED = enum.auto()
FINISHED = enum.auto() FINISHED = enum.auto()