[vLLM IR] 1/N Implement IR skeleton and rms_norm op (#33825)
Signed-off-by: Luka Govedič <lgovedic@redhat.com> Signed-off-by: Xinyu Chen <xinyu1.chen@intel.com> Signed-off-by: chzhang <chaojun.zhang@intel.com> Signed-off-by: Luka Govedic <luka.govedic@gmail.com> Co-authored-by: Xinyu Chen <xinyu1.chen@intel.com> Co-authored-by: Chaojun Zhang <chaojun.zhang@intel.com> Co-authored-by: Luka Govedič <ProExpertProg@h100-01.nemg-001.lab.rdu2.dc.redhat.com>
This commit is contained in:
co-authored by
Xinyu Chen
Chaojun Zhang
Luka Govedič
parent
0fab52f0aa
commit
40bb175027
@@ -10,6 +10,7 @@ from typing import Any
|
||||
import torch
|
||||
|
||||
import vllm.envs as envs
|
||||
import vllm.ir
|
||||
from vllm.config import CUDAGraphMode, ParallelConfig, VllmConfig
|
||||
from vllm.logger import init_logger
|
||||
from vllm.platforms import current_platform
|
||||
@@ -378,7 +379,13 @@ def set_forward_context(
|
||||
)
|
||||
|
||||
try:
|
||||
with override_forward_context(forward_context):
|
||||
with (
|
||||
override_forward_context(forward_context),
|
||||
vllm_config.kernel_config.ir_op_priority.set_priority(),
|
||||
vllm.ir.enable_torch_wrap(
|
||||
vllm_config.compilation_config.ir_enable_torch_wrap
|
||||
),
|
||||
):
|
||||
yield
|
||||
finally:
|
||||
global last_logging_time, batchsize_logging_interval
|
||||
|
||||
Reference in New Issue
Block a user