Closed
Description
Your current environment
The output of python collect_env.py
(vllm-dev) mathieu@sophia:vllm $ python collect_env.py
INFO 06-02 20:48:21 [__init__.py:243] Automatically detected platform cuda.
Collecting environment information...
==============================
System Info
==============================
OS : Fedora Linux 42 (Workstation Edition) (x86_64)
GCC version : (GCC) 15.1.1 20250521 (Red Hat 15.1.1-2)
Clang version : 20.1.5 (Fedora 20.1.5-1.fc42)
CMake version : version 3.31.6
Libc version : glibc-2.41
==============================
PyTorch Info
==============================
PyTorch version : 2.7.0+cu126
Is debug build : False
CUDA used to build PyTorch : 12.6
ROCM used to build PyTorch : N/A
==============================
Python Environment
==============================
Python version : 3.12.10 | packaged by conda-forge | (main, Apr 10 2025, 22:21:13) [GCC 13.3.0] (64-bit runtime)
Python platform : Linux-6.14.9-300.fc42.x86_64-x86_64-with-glibc2.41
==============================
CUDA / GPU Info
==============================
Is CUDA available : True
CUDA runtime version : Could not collect
CUDA_MODULE_LOADING set to : LAZY
GPU models and configuration :
GPU 0: NVIDIA GeForce RTX 3090
GPU 1: NVIDIA GeForce RTX 3090
Nvidia driver version : 570.153.02
cuDNN version : Probably one of the following:
/usr/lib64/libcudnn.so.9.10.0
/usr/lib64/libcudnn_adv.so.9.10.0
/usr/lib64/libcudnn_cnn.so.9.10.0
/usr/lib64/libcudnn_engines_precompiled.so.9.10.0
/usr/lib64/libcudnn_engines_runtime_compiled.so.9.10.0
/usr/lib64/libcudnn_graph.so.9.10.0
/usr/lib64/libcudnn_heuristic.so.9.10.0
/usr/lib64/libcudnn_ops.so.9.10.0
HIP runtime version : N/A
MIOpen runtime version : N/A
Is XNNPACK available : True
==============================
CPU Info
==============================
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 48 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 24
On-line CPU(s) list: 0-23
Vendor ID: AuthenticAMD
Model name: AMD Ryzen 9 7900 12-Core Processor
CPU family: 25
Model: 97
Thread(s) per core: 2
Core(s) per socket: 12
Socket(s): 1
Stepping: 2
Frequency boost: enabled
CPU(s) scaling MHz: 85%
CPU max MHz: 5485,0000
CPU min MHz: 545,0000
BogoMIPS: 7399,81
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good amd_lbr_v2 nopl xtopology nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba perfmon_v2 ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local user_shstk avx512_bf16 clzero irperf xsaveerptr rdpru wbnoinvd cppc arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic vgif x2avic v_spec_ctrl vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid overflow_recov succor smca fsrm flush_l1d amd_lbr_pmc_freeze
Virtualization: AMD-V
L1d cache: 384 KiB (12 instances)
L1i cache: 384 KiB (12 instances)
L2 cache: 12 MiB (12 instances)
L3 cache: 64 MiB (2 instances)
NUMA node(s): 1
NUMA node0 CPU(s): 0-23
Vulnerability Gather data sampling: Not affected
Vulnerability Ghostwrite: Not affected
Vulnerability Indirect target selection: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Mitigation; Safe RET
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP always-on; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
==============================
Versions of relevant libraries
==============================
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.6.4.1
[pip3] nvidia-cuda-cupti-cu12==12.6.80
[pip3] nvidia-cuda-nvrtc-cu12==12.6.77
[pip3] nvidia-cuda-runtime-cu12==12.6.77
[pip3] nvidia-cudnn-cu12==9.5.1.17
[pip3] nvidia-cufft-cu12==11.3.0.4
[pip3] nvidia-cufile-cu12==1.11.1.6
[pip3] nvidia-curand-cu12==10.3.7.77
[pip3] nvidia-cusolver-cu12==11.7.1.2
[pip3] nvidia-cusparse-cu12==12.5.4.2
[pip3] nvidia-cusparselt-cu12==0.6.3
[pip3] nvidia-nccl-cu12==2.26.2
[pip3] nvidia-nvjitlink-cu12==12.6.85
[pip3] nvidia-nvtx-cu12==12.6.77
[pip3] pyzmq==26.4.0
[pip3] torch==2.7.0
[pip3] torchaudio==2.7.0
[pip3] torchvision==0.22.0
[pip3] transformers==4.52.4
[pip3] triton==3.3.0
[conda] Could not collect
==============================
vLLM Info
==============================
ROCM Version : 6.3.42133-0
Neuron SDK Version : N/A
vLLM Version : 0.9.1.dev243+gca2f6b9c3 (git sha: ca2f6b9c3)
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0 GPU1 CPU Affinity NUMA Affinity GPU NUMA ID
GPU0 X PHB 0-23 0 N/A
GPU1 PHB X 0-23 0 N/A
Legend:
X = Self
SYS = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
PHB = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
PXB = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
PIX = Connection traversing at most a single PCIe bridge
NV# = Connection traversing a bonded set of # NVLinks
==============================
Environment Variables
==============================
NCCL_CUMEM_ENABLE=0
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY
🐛 Describe the bug
Running vllm serve "deepseek-ai/DeepSeek-R1-0528-Qwen3-8B" --tensor-parallel-size 2 results in 2 python processes taking up 100% each of a CPU while vllm is idle.
py-spy top output of one of the offending processes
GIL: 49.00%, Active: 100.00%, Threads: 5
%Own %Total OwnTime TotalTime Function (filename)
54.00% 54.00% 5.03s 5.03s sched_yield (vllm/distributed/utils.py)
14.00% 100.00% 1.09s 8.95s acquire_read (vllm/distributed/device_communicators/shm_broadcast.py)
12.00% 13.00% 1.04s 1.09s get_metadata (vllm/distributed/device_communicators/shm_broadcast.py)
5.00% 5.00% 0.590s 0.670s __init__ (contextlib.py)
6.00% 100.00% 0.410s 8.95s __enter__ (contextlib.py)
3.00% 4.00% 0.380s 0.400s __exit__ (contextlib.py)
5.00% 10.00% 0.360s 1.03s helper (contextlib.py)
1.00% 1.00% 0.050s 0.050s buf (multiprocessing/shared_memory.py)
0.00% 100.00% 0.000s 8.95s dequeue (vllm/distributed/device_communicators/shm_broadcast.py)
0.00% 100.00% 0.000s 8.95s worker_main (vllm/v1/executor/multiproc_executor.py)
0.00% 100.00% 0.000s 8.95s worker_busy_loop (vllm/v1/executor/multiproc_executor.py)
0.00% 100.00% 0.000s 8.95s <module> (<string>)
0.00% 100.00% 0.000s 8.95s run (multiprocessing/process.py)
0.00% 100.00% 0.000s 8.95s _main (multiprocessing/spawn.py)
0.00% 100.00% 0.000s 8.95s _bootstrap (multiprocessing/process.py)
It looks like sched_yield tries to yield the CPU, but there's no other process that wants the CPU, so the loop in acquire_read runs again and sched_yield is called again, pegging the CPU.
Replacing the implementation of sched_yield with just a time.sleep(0.0001) call decreases the CPU usage to something like 2% on my system, but that implementation might be too naive?
Before submitting a new issue...
- Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.