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

[V1] Enable custom ops with piecewise CUDA graphs (#10228)

Browse Source 
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
This commit is contained in:
Woosuk Kwon
2024-11-11 11:58:07 -08:00
committed by GitHub
parent 8a7fe47d32
commit 9d5b4e4dea
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/v1/worker/gpu_model_runner.py
View File

@@ -1,3 +1,4 @@
import os
import time import time
from dataclasses import dataclass from dataclasses import dataclass
from typing import TYPE_CHECKING, Dict, List, Optional, Set from typing import TYPE_CHECKING, Dict, List, Optional, Set
@@ -405,6 +406,7 @@ class GPUModelRunner:
if self.use_cuda_graph: if self.use_cuda_graph:
# FIXME(woosuk): Currently, we do not use inductor to reduce the # FIXME(woosuk): Currently, we do not use inductor to reduce the
# compilation time and any potential issues with the inductor. # compilation time and any potential issues with the inductor.
os.environ["VLLM_CUSTOM_OPS"] = "all"
set_compilation_config( set_compilation_config(
CompilationConfig( CompilationConfig(
use_cudagraph=True, use_cudagraph=True,