A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
[Misc] refactor function name (#23029)
Signed-off-by: Andy Xie <andy.xning@gmail.com>
Ning Xie committed
Aug 18, 2025 at 13:16 UTC
08d5f7113a024818b2867782c2539794b7aa162b
2 files changed
+2
-2
vllm/platforms/cpu.py
+1
-1
@@ -268,7 +268,7 @@ class CpuPlatform(Platform):
268
DEFAULT_MAX_NUM_BATCHED_TOKENS)
269
270
@classmethod
271
- def get_allowed_cpu_memory_node_list(
271
+ def get_allowed_cpu_core_node_list(
272
cls) -> tuple[list[int], list[LogicalCPUInfo]]:
273
assert platform.system() == "Linux"
274
vllm/v1/worker/cpu_worker.py
+1
-1
@@ -132,7 +132,7 @@ class CPUWorker(Worker):
132
"""
133
134
allowed_numa_nodes, logical_cpu_list = \
135
- CpuPlatform.get_allowed_cpu_memory_node_list()
135
+ CpuPlatform.get_allowed_cpu_core_node_list()
136
assert len(allowed_numa_nodes) >= self.parallel_config.world_size, (
137
f"No enough allowed NUMA nodes to bind threads of "
138
f"{self.parallel_config.world_size} CPUWorkers. "