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

[P/D] NIXL Integration (#17751)

Browse Source 
Signed-off-by: ApostaC <yihua98@uchicago.edu>
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
Signed-off-by: rshaw@neuralmagic.com <robertgshaw2@gmail.com>
Signed-off-by: Robert Shaw <rshaw@neuralmagic.com>
Signed-off-by: mgoin <mgoin64@gmail.com>
Signed-off-by: Nick Hill <nhill@redhat.com>
Signed-off-by: Brent Salisbury <bsalisbu@redhat.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
Co-authored-by: ApostaC <yihua98@uchicago.edu>
Co-authored-by: Robert Shaw <rshaw@neuralmagic.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Tyler Michael Smith <tysmith@redhat.com>
Co-authored-by: Brent Salisbury <bsalisbu@redhat.com>
This commit is contained in:
Robert Shaw
2025-05-12 12:46:16 -04:00
committed by GitHub
parent 05a4324f8e
commit d19110204c
34 changed files with 2723 additions and 108 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
vllm/v1/request.py
View File

@@ -1,8 +1,9 @@
# SPDX-License-Identifier: Apache-2.0
import enum
from typing import TYPE_CHECKING, Optional, Union
from typing import TYPE_CHECKING, Any, Optional, Union
from vllm.distributed.kv_transfer.kv_connector.v1 import KVTransferParams
from vllm.multimodal.inputs import MultiModalKwargs, PlaceholderRange
from vllm.sampling_params import SamplingParams
from vllm.utils import is_list_of
@@ -61,6 +62,15 @@ class Request:
self.num_encoder_inputs = len(self.mm_inputs)
self.has_encoder_inputs = self.num_encoder_inputs > 0
# P/D: KV transfer parameters (raw and parsed).
raw_params = (None if sampling_params.extra_args is None
else sampling_params.extra_args.get(
"kv_transfer_params", None))
self.raw_kv_transfer_params: Optional[dict[str, Any]] = raw_params
# Each connector parses the raw dictionary and sets this
# attr the first time that the request is processed.
self.kv_transfer_params: Optional[KVTransferParams] = None
# Sanity check
assert len(self.mm_inputs) == len(self.mm_positions)
if self.mm_hashes:
@@ -150,6 +160,7 @@ class RequestStatus(enum.IntEnum):
"""Status of a request."""
WAITING = enum.auto()
WAITING_FOR_FSM = enum.auto()
WAITING_FOR_REMOTE_KVS = enum.auto()
RUNNING = enum.auto()
PREEMPTED = enum.auto()
# Note: anything after PREEMPTED will be considered