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

[XPU] avoid triton import for xpu (#9440)

Browse Source 
Co-authored-by: Cyrus Leung <tlleungac@connect.ust.hk>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
This commit is contained in:
Yan Ma 2024-10-24 13:14:00 +08:00 committed by GitHub
parent 33bab41060
commit 056a68c7db
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
vllm/triton_utils/importing.py
View File

@ -5,10 +5,12 @@ from vllm.platforms import current_platform
logger = init_logger(__name__) logger = init_logger(__name__)
# neuron has too old torch HAS_TRITON = (
HAS_TRITON = find_spec( find_spec("triton") is not None
"triton") is not None and not current_platform.is_neuron() and not current_platform.is_xpu() # Not compatible
and not current_platform.is_neuron() # neuron has too old torch
)
if not HAS_TRITON: if not HAS_TRITON:
logger.info("Triton not installed; certain GPU-related functions" logger.info("Triton not installed or not compatible; certain GPU-related"
" will not be available.") " functions will not be available.")