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

[Log] Optimize duplicate startup log (#33944)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye
2026-02-06 12:49:56 -05:00
committed by GitHub
parent 7bec435130
commit 67a746e87f
3 changed files with 10 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
vllm/compilation/backends.py
View File

@@ -263,11 +263,12 @@ class CompilerManager:
now = time.time() now = time.time()
elapsed = now - compilation_start_time elapsed = now - compilation_start_time
compilation_config.compilation_time += elapsed compilation_config.compilation_time += elapsed
logger.info( logger.info_once(
"Directly load the compiled graph(s) for compile range %s " "Directly load the compiled graph(s) for compile range %s "
"from the cache, took %.3f s", "from the cache, took %.3f s",
str(compile_range), str(compile_range),
elapsed, elapsed,
scope="local",
) )
return compiled_graph return compiled_graph

8
vllm/utils/deep_gemm.py
View File

@@ -91,14 +91,16 @@ def is_deep_gemm_e8m0_used() -> bool:
_lazy_init() _lazy_init()
if _fp8_gemm_nt_impl is None: if _fp8_gemm_nt_impl is None:
logger.info_once("DeepGEMM E8M0 disabled: _fp8_gemm_nt_impl not found") logger.info_once(
"DeepGEMM E8M0 disabled: _fp8_gemm_nt_impl not found", scope="local"
)
return False return False
if envs.VLLM_USE_DEEP_GEMM_E8M0: if envs.VLLM_USE_DEEP_GEMM_E8M0:
logger.info_once("DeepGEMM E8M0 enabled on current platform.") logger.info_once("DeepGEMM E8M0 enabled on current platform.", scope="local")
return True return True
logger.info_once("DeepGEMM E8M0 disabled on current configuration.") logger.info_once("DeepGEMM E8M0 disabled on current configuration.", scope="local")
return False return False

6
vllm/v1/worker/gpu_worker.py
View File

@@ -119,9 +119,6 @@ class Worker(WorkerBase):
self.use_v2_model_runner = envs.VLLM_USE_V2_MODEL_RUNNER self.use_v2_model_runner = envs.VLLM_USE_V2_MODEL_RUNNER
if self.use_v2_model_runner:
logger.info_once("Using V2 Model Runner", scope="global")
def sleep(self, level: int = 1) -> None: def sleep(self, level: int = 1) -> None:
from vllm.device_allocator.cumem import CuMemAllocator from vllm.device_allocator.cumem import CuMemAllocator
@@ -240,6 +237,9 @@ class Worker(WorkerBase):
current_platform.dist_backend, current_platform.dist_backend,
) )
if self.use_v2_model_runner:
logger.info_once("Using V2 Model Runner", scope="local")
# Set random seed. # Set random seed.
set_random_seed(self.model_config.seed) set_random_seed(self.model_config.seed)