[Hardware][Intel CPU] Adding intel openmp tunings in Docker file (#6008)

Signed-off-by: Yuan Zhou <yuan.zhou@intel.com>
This commit is contained in:
Yuan
2024-07-04 15:22:12 -07:00
committed by GitHub
parent 27902d42be
commit 81d7a50f24
4 changed files with 37 additions and 5 deletions
+4 -1
View File
@@ -13,7 +13,7 @@ from vllm.distributed import (ensure_model_parallel_initialized,
from vllm.logger import init_logger
from vllm.model_executor import set_random_seed
from vllm.sequence import ExecuteModelRequest
from vllm.utils import STR_DTYPE_TO_TORCH_DTYPE
from vllm.utils import STR_DTYPE_TO_TORCH_DTYPE, init_kmp_env
from vllm.worker.cpu_model_runner import CPUModelRunner
from vllm.worker.worker_base import (LocalOrDistributedWorkerBase,
LoraNotSupportedWorkerBase, WorkerInput)
@@ -150,6 +150,9 @@ class CPUWorker(LoraNotSupportedWorkerBase, LocalOrDistributedWorkerBase):
if self.is_driver_worker:
assert self.rank == 0, "The driver worker must have rank 0."
# try to initialize intel openmp optimized tunings
init_kmp_env()
if self.model_config.trust_remote_code:
# note: lazy import to avoid importing torch before initializing
from vllm.utils import init_cached_hf_modules