xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 11:06:15 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Doc]: fixing typos in various files. (#29717)

Browse Source 
Signed-off-by: Didier Durand <durand.didier@gmail.com>
This commit is contained in:
Didier Durand 2025-11-29 10:15:39 +01:00 committed by GitHub
parent 6afc0ffaf6
commit 04a797cd0e
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
5 changed files with 5 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/distributed/device_communicators/pynccl_allocator.py
View File

@ -157,7 +157,7 @@ class nccl_symm_mem_context:
if self.disabled: if self.disabled:
return self return self
assert self.pynccl_comm is not None, ( assert self.pynccl_comm is not None, (
"Symmetric memory requires pynccl to be initalized" "Symmetric memory requires pynccl to be initialized"
) )
assert self.pynccl_comm.nccl_version >= 22703, ( assert self.pynccl_comm.nccl_version >= 22703, (
"NCCL version 2.27.3 or higher is required for NCCL symmetric memory" "NCCL version 2.27.3 or higher is required for NCCL symmetric memory"

2
vllm/distributed/parallel_state.py
View File

@ -1583,7 +1583,7 @@ def destroy_distributed_environment():
def cleanup_dist_env_and_memory(shutdown_ray: bool = False): def cleanup_dist_env_and_memory(shutdown_ray: bool = False):
# Ensure all objects are not freezed before cleanup # Ensure all objects are not frozen before cleanup
gc.unfreeze() gc.unfreeze()
destroy_model_parallel() destroy_model_parallel()

2
vllm/entrypoints/openai/serving_models.py
View File

@ -150,7 +150,7 @@ class OpenAIServingModels:
lora_request.base_model_name = base_model_name lora_request.base_model_name = base_model_name
# Validate that the adapter can be loaded into the engine # Validate that the adapter can be loaded into the engine
# This will also pre-load it for incoming requests # This will also preload it for incoming requests
try: try:
await self.engine_client.add_lora(lora_request) await self.engine_client.add_lora(lora_request)
except Exception as e: except Exception as e:

2
vllm/model_executor/layers/quantization/quark/schemes/quark_ocp_mx.py
View File

@ -37,7 +37,7 @@ logger = init_logger(__name__)
# use `rocm_aiter_ops.is_asm_fp4_gemm_dynamic_quant_enabled()` # use `rocm_aiter_ops.is_asm_fp4_gemm_dynamic_quant_enabled()`
# for envs checks which does not require @cache anymore. # for envs checks which does not require @cache anymore.
# triton kernel is torch compile compatible. # triton kernel is torch compile compatible.
# does not require direct registeration. # does not require direct registration.
# use `rocm_aiter_ops.triton_fp4_gemm_dynamic_qaunt`. # use `rocm_aiter_ops.triton_fp4_gemm_dynamic_qaunt`.
@cache @cache
def is_rocm_aiter_fp4_asm_gemm_enabled() -> bool: def is_rocm_aiter_fp4_asm_gemm_enabled() -> bool: