vllm - 💡(How to fix) Fix [Bug]: [Regression] MiniMax-M2.5-int4 fails with cudaErrorPeerAccessUnsupported since PR #43410

vllm2026-05-29 16:35:43

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

Running Intel/Minimax-M2.5-int4-AutoRound on vLLM nightly fails with cudaErrorPeerAccessUnsupported: 217 during model loading at MiniMaxText01RMSNormTP.__init__. Version 0.21.1rc1.dev268+gd0a100c87 works correctly.

Tagging the original author, @jeejeelee for visibility. Full details are included above, but for brevity, I am running 8 RTX 3090 on a single node, all on PCIe. No NVLink.

GLM suggests the following potential cause: PR #43410 ("[Kernel] Porting fuse_minimax_qk_norm to manual fusion", merged May 26 2026) moved the LamportWorkspace initialisation from lazy (first forward pass) to eager (init time). The IpcBuffer class in lamport_workspace.py unconditionally calls cudaIpcOpenMemHandle with cudaIpcMemLazyEnablePeerAccess, which fails on GPU topologies that do not support peer-to-peer access (e.g., PCIe-only systems without NVLink).

Working version: 0.21.1rc1.dev268+gd0a100c87 Broken version: 0.21.1rc1.dev433+g0585b5ba2

Error Message

(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] (Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] WorkerProc failed to start. Traceback (most recent call last): File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/executor/multiproc_executor.py", line 841, in worker_main worker = WorkerProc(*args, **kwargs) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper return func(*args, **kwargs) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/executor/multiproc_executor.py", line 620, in init self.worker.load_model() ~~~~^^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/worker/gpu_worker.py", line 345, in load_model self.model_runner.load_model(load_dummy_weights=load_dummy_weights) ~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper return func(*args, **kwargs) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/worker/gpu_model_runner.py", line 5052, in load_model self.model = model_loader.load_model( ~~~~~^ vllm_config=self.vllm_config, model_config=self.model_config ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper return func(*args, **kwargs) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/model_loader/base_loader.py", line 55, in load_model model = initialize_model( vllm_config=vllm_config, model_config=model_config, prefix=prefix, ) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper return func(*args, **kwargs) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/model_loader/utils.py", line 61, in initialize_model model = model_class(vllm_config=vllm_config, prefix=prefix) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 537, in init self.model = MiniMaxM2Model( ~~~~^ vllm_config=vllm_config, prefix=maybe_prefix(prefix, "model") ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/compilation/decorators.py", line 383, in init old_init(self, *args, **kwargs) ~~~~^^^^^^^^^^^^^^^^^^^^^^^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 351, in init self.start_layer, self.end_layer, self.layers = make_layers( ~~~~~^ config.num_hidden_layers, ^^^^^^^^^^^^^^^^^^^^^^^^^ ...<7 lines>... prefix=f"{prefix}.layers", ^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/utils.py", line 643, in make_layers + get_offloader().wrap_modules( ~~~~~~^ layer_fn(prefix=f"{prefix}.{idx}") for idx in range(start_layer, end_layer) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/offloader/base.py", line 104, in wrap_modules return list(modules_generator) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/utils.py", line 644, in <genexpr> layer_fn(prefix=f"{prefix}.{idx}") for idx in range(start_layer, end_layer) ~~^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 353, in <lambda> lambda prefix: MiniMaxM2DecoderLayer( ~~~~~^ config, ^^^^^^^ ...<3 lines>... quant_config=quant_config, ^^^^^^^^^^^^^^^^^^^^^^^^^^ ), ^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 276, in init self.self_attn = MiniMaxM2Attention( ~~~~^ hidden_size=self.hidden_size, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ...<10 lines>... prefix=f"{prefix}.self_attn", ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 221, in init self.q_norm = MiniMaxText01RMSNormTP( ~~~~^ self.head_dim * self.total_num_heads, eps=rms_norm_eps ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/rms_norm_tp.py", line 146, in init self.workspace = get_allreduce_workspace( ~~~~~^ rank=self.tp_rank, ^^^^^^^^^^^^^^^^^^ ...<2 lines>... process_group=get_tp_group().cpu_group, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ ) ^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 300, in get_allreduce_workspace ws = LamportWorkspace(rank, world_size, comm_size, process_group) File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 185, in init self._lamport = IpcBuffer(rank, world_size, lamport_total, process_group) ~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 100, in init _check(err) ~~~~^^^^^ File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 30, in _check raise RuntimeError(f"CUDA runtime error: {error}") RuntimeError: CUDA runtime error: <cudaError_t.cudaErrorPeerAccessUnsupported: 217>

Root Cause

Tagging the original author, @jeejeelee for visibility. Full details are included above, but for brevity, I am running 8 RTX 3090 on a single node, all on PCIe. No NVLink.

Working version: 0.21.1rc1.dev268+gd0a100c87 Broken version: 0.21.1rc1.dev433+g0585b5ba2

Fix Action

Fix / Workaround

============================== 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): 48 On-line CPU(s) list: 0-47 Vendor ID: AuthenticAMD Model name: AMD EPYC 7443 24-Core Processor CPU family: 25 Model: 1 Thread(s) per core: 2 Core(s) per socket: 24 Socket(s): 1 Stepping: 1 Microcode version: 0xa0011de Frequency boost: enabled CPU(s) scaling MHz: 47% CPU max MHz: 4035.6440 CPU min MHz: 1500.0000 BogoMIPS: 5699.72 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 nopl xtopology nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid 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 ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local user_shstk clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin brs arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold v_vmsave_vmload vgif v_spec_ctrl umip pku ospke vaes vpclmulqdq rdpid overflow_recov succor smca fsrm debug_swap Virtualization: AMD-V L1d cache: 768 KiB (24 instances) L1i cache: 768 KiB (24 instances) L2 cache: 12 MiB (24 instances) L3 cache: 128 MiB (4 instances) NUMA node(s): 1 NUMA node0 CPU(s): 0-47 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 Old microcode: 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; Retpolines; IBPB conditional; IBRS_FW; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected Vulnerability Srbds: Not affected Vulnerability Tsa: Mitigation; Clear CPU buffers Vulnerability Tsx async abort: Not affected Vulnerability Vmscape: Mitigation; IBPB before exit to userspace

Code Example

Collecting environment information...
==============================
        System Info
==============================
OS                           : Arch Linux (x86_64)
GCC version                  : (GCC) 16.1.1 20260430
Clang version                : 22.1.5
CMake version                : version 4.3.3
Libc version                 : glibc-2.43

==============================
       PyTorch Info
==============================
PyTorch version              : 2.11.0+cu130
Is debug build               : False
CUDA used to build PyTorch   : 13.0
ROCM used to build PyTorch   : N/A
XPU used to build PyTorch    : N/A

==============================
      Python Environment
==============================
Python version               : 3.13.13 (main, Apr 14 2026, 14:28:56) [Clang 22.1.3 ] (64-bit runtime)
Python platform              : Linux-7.0.9-arch2-1-x86_64-with-glibc2.43

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 13.1.115
CUDA_MODULE_LOADING set to   :
GPU models and configuration :
GPU 0: NVIDIA GeForce RTX 3090
GPU 1: NVIDIA GeForce RTX 3090
GPU 2: NVIDIA GeForce RTX 3090
GPU 3: NVIDIA GeForce RTX 3090
GPU 4: NVIDIA GeForce RTX 3090
GPU 5: NVIDIA GeForce RTX 3090
GPU 6: NVIDIA GeForce RTX 3090
GPU 7: NVIDIA GeForce RTX 3090

Nvidia driver version        : 595.71.05
cuDNN version                : Could not collect
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):                                  48
On-line CPU(s) list:                     0-47
Vendor ID:                               AuthenticAMD
Model name:                              AMD EPYC 7443 24-Core Processor
CPU family:                              25
Model:                                   1
Thread(s) per core:                      2
Core(s) per socket:                      24
Socket(s):                               1
Stepping:                                1
Microcode version:                       0xa0011de
Frequency boost:                         enabled
CPU(s) scaling MHz:                      47%
CPU max MHz:                             4035.6440
CPU min MHz:                             1500.0000
BogoMIPS:                                5699.72
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 nopl xtopology nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid 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 ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local user_shstk clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin brs arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold v_vmsave_vmload vgif v_spec_ctrl umip pku ospke vaes vpclmulqdq rdpid overflow_recov succor smca fsrm debug_swap
Virtualization:                          AMD-V
L1d cache:                               768 KiB (24 instances)
L1i cache:                               768 KiB (24 instances)
L2 cache:                                12 MiB (24 instances)
L3 cache:                                128 MiB (4 instances)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-47
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 Old microcode:             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; Retpolines; IBPB conditional; IBRS_FW; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Mitigation; Clear CPU buffers
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Mitigation; IBPB before exit to userspace

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.11.post3
[pip3] numpy==2.4.6
[pip3] nvidia-cublas==13.1.0.3
[pip3] nvidia-cuda-cccl==13.3.3.3.1
[pip3] nvidia-cuda-crt==13.3.33
[pip3] nvidia-cuda-cupti==13.0.85
[pip3] nvidia-cuda-nvcc==13.3.33
[pip3] nvidia-cuda-nvrtc==13.0.88
[pip3] nvidia-cuda-runtime==13.0.96
[pip3] nvidia-cudnn-cu13==9.19.0.56
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft==12.0.0.61
[pip3] nvidia-cufile==1.15.1.6
[pip3] nvidia-curand==10.4.0.35
[pip3] nvidia-cusolver==12.0.4.66
[pip3] nvidia-cusparse==12.6.3.3
[pip3] nvidia-cusparselt-cu13==0.8.0
[pip3] nvidia-cutlass-dsl==4.5.2
[pip3] nvidia-cutlass-dsl-libs-base==4.5.2
[pip3] nvidia-cutlass-dsl-libs-cu13==4.5.2
[pip3] nvidia-ml-py==13.595.45
[pip3] nvidia-nccl-cu13==2.28.9
[pip3] nvidia-nvjitlink==13.0.88
[pip3] nvidia-nvshmem-cu13==3.4.5
[pip3] nvidia-nvtx==13.0.85
[pip3] nvidia-nvvm==13.3.33
[pip3] pyzmq==27.1.0
[pip3] tokenspeed-triton==3.7.10.post20260505
[pip3] torch==2.11.0+cu130
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.11.0+cu130
[pip3] torchvision==0.26.0+cu130
[pip3] transformers==5.9.0
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.21.1rc1.dev433+g0585b5ba2 (git sha: 0585b5ba2)
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled; XPU: Disabled
GPU Topology:
  	GPU0	GPU1	GPU2	GPU3	GPU4	GPU5	GPU6	GPU7	CPU Affinity	NUMA Affinity	GPU NUMA ID
GPU0	 X 	NODE	NODE	NODE	NODE	NODE	NODE	NODE	0-47	0		N/A
GPU1	NODE	 X 	PHB	NODE	NODE	NODE	NODE	NODE	0-47	0		N/A
GPU2	NODE	PHB	 X 	NODE	NODE	NODE	NODE	NODE	0-47	0		N/A
GPU3	NODE	NODE	NODE	 X 	PHB	PHB	NODE	NODE	0-47	0		N/A
GPU4	NODE	NODE	NODE	PHB	 X 	PHB	NODE	NODE	0-47	0		N/A
GPU5	NODE	NODE	NODE	PHB	PHB	 X 	NODE	NODE	0-47	0		N/A
GPU6	NODE	NODE	NODE	NODE	NODE	NODE	 X 	PHB	0-47	0		N/A
GPU7	NODE	NODE	NODE	NODE	NODE	NODE	PHB	 X 	0-47	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
==============================
CUDA_PATH=/opt/cuda
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_hubby

---

uv run vllm serve Intel/MiniMax-M2.5-int4-AutoRound
--served-model-name minimax-m2.5
--tensor-parallel-size 8
--enable-expert-parallel
--max-model-len auto
--gpu-memory-utilization 0.93
--attention-backend flashinfer
--enable-flashinfer-autotune
--reasoning-parser minimax_m2
--tool-call-parser minimax_m2
--enable-auto-tool-choice
--max-num-seqs 4
--max-num-batched-tokens 4096
--enable-prefix-caching
--enable-chunked-prefill
--trust-remote-code
--host 0.0.0.0
--port ${PORT}
--seed 3407
--disable-custom-all-reduce

---

(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] WorkerProc failed to start.
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] Traceback (most recent call last):
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/executor/multiproc_executor.py", line 841, in worker_main
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     worker = WorkerProc(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return func(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/executor/multiproc_executor.py", line 620, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.worker.load_model()
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~~~^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/worker/gpu_worker.py", line 345, in load_model
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.model_runner.load_model(load_dummy_weights=load_dummy_weights)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return func(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/worker/gpu_model_runner.py", line 5052, in load_model
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.model = model_loader.load_model(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                  ~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         vllm_config=self.vllm_config, model_config=self.model_config
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return func(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/model_loader/base_loader.py", line 55, in load_model
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     model = initialize_model(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         vllm_config=vllm_config,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         model_config=model_config,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         prefix=prefix,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return func(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/model_loader/utils.py", line 61, in initialize_model
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     model = model_class(vllm_config=vllm_config, prefix=prefix)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 537, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.model = MiniMaxM2Model(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                  ~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         vllm_config=vllm_config, prefix=maybe_prefix(prefix, "model")
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/compilation/decorators.py", line 383, in __init__
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     old_init(self, *args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~~~^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 351, in __init__
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.start_layer, self.end_layer, self.layers = make_layers(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                                                     ~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         config.num_hidden_layers,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ...<7 lines>...
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         prefix=f"{prefix}.layers",
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/utils.py", line 643, in make_layers
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     + get_offloader().wrap_modules(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]       ~~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         layer_fn(prefix=f"{prefix}.{idx}") for idx in range(start_layer, end_layer)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/offloader/base.py", line 104, in wrap_modules
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return list(modules_generator)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/utils.py", line 644, in <genexpr>
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     layer_fn(prefix=f"{prefix}.{idx}") for idx in range(start_layer, end_layer)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 353, in <lambda>
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     lambda prefix: MiniMaxM2DecoderLayer(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                    ~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         config,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ...<3 lines>...
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         quant_config=quant_config,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ),
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 276, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.self_attn = MiniMaxM2Attention(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                      ~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         hidden_size=self.hidden_size,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ...<10 lines>...
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         prefix=f"{prefix}.self_attn",
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 221, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.q_norm = MiniMaxText01RMSNormTP(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                   ~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         self.head_dim * self.total_num_heads, eps=rms_norm_eps
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/rms_norm_tp.py", line 146, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.workspace = get_allreduce_workspace(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                      ~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         rank=self.tp_rank,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ...<2 lines>...
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         process_group=get_tp_group().cpu_group,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 300, in get_allreduce_workspace
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ws = LamportWorkspace(rank, world_size, comm_size, process_group)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 185, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self._lamport = IpcBuffer(rank, world_size, lamport_total, process_group)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                     ~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 100, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     _check(err)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~~~^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 30, in _check
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     raise RuntimeError(f"CUDA runtime error: {error}")
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] RuntimeError: CUDA runtime error: <cudaError_t.cudaErrorPeerAccessUnsupported: 217>

RAW_BUFFERClick to expand / collapse

Your current environment

<details> <summary>The output of <code>python collect_env.py</code></summary>

Collecting environment information...
==============================
        System Info
==============================
OS                           : Arch Linux (x86_64)
GCC version                  : (GCC) 16.1.1 20260430
Clang version                : 22.1.5
CMake version                : version 4.3.3
Libc version                 : glibc-2.43

==============================
       PyTorch Info
==============================
PyTorch version              : 2.11.0+cu130
Is debug build               : False
CUDA used to build PyTorch   : 13.0
ROCM used to build PyTorch   : N/A
XPU used to build PyTorch    : N/A

==============================
      Python Environment
==============================
Python version               : 3.13.13 (main, Apr 14 2026, 14:28:56) [Clang 22.1.3 ] (64-bit runtime)
Python platform              : Linux-7.0.9-arch2-1-x86_64-with-glibc2.43

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 13.1.115
CUDA_MODULE_LOADING set to   :
GPU models and configuration :
GPU 0: NVIDIA GeForce RTX 3090
GPU 1: NVIDIA GeForce RTX 3090
GPU 2: NVIDIA GeForce RTX 3090
GPU 3: NVIDIA GeForce RTX 3090
GPU 4: NVIDIA GeForce RTX 3090
GPU 5: NVIDIA GeForce RTX 3090
GPU 6: NVIDIA GeForce RTX 3090
GPU 7: NVIDIA GeForce RTX 3090

Nvidia driver version        : 595.71.05
cuDNN version                : Could not collect
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):                                  48
On-line CPU(s) list:                     0-47
Vendor ID:                               AuthenticAMD
Model name:                              AMD EPYC 7443 24-Core Processor
CPU family:                              25
Model:                                   1
Thread(s) per core:                      2
Core(s) per socket:                      24
Socket(s):                               1
Stepping:                                1
Microcode version:                       0xa0011de
Frequency boost:                         enabled
CPU(s) scaling MHz:                      47%
CPU max MHz:                             4035.6440
CPU min MHz:                             1500.0000
BogoMIPS:                                5699.72
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 nopl xtopology nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 pcid 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 ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local user_shstk clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin brs arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold v_vmsave_vmload vgif v_spec_ctrl umip pku ospke vaes vpclmulqdq rdpid overflow_recov succor smca fsrm debug_swap
Virtualization:                          AMD-V
L1d cache:                               768 KiB (24 instances)
L1i cache:                               768 KiB (24 instances)
L2 cache:                                12 MiB (24 instances)
L3 cache:                                128 MiB (4 instances)
NUMA node(s):                            1
NUMA node0 CPU(s):                       0-47
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 Old microcode:             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; Retpolines; IBPB conditional; IBRS_FW; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                     Not affected
Vulnerability Tsa:                       Mitigation; Clear CPU buffers
Vulnerability Tsx async abort:           Not affected
Vulnerability Vmscape:                   Mitigation; IBPB before exit to userspace

==============================
Versions of relevant libraries
==============================
[pip3] flashinfer-python==0.6.11.post3
[pip3] numpy==2.4.6
[pip3] nvidia-cublas==13.1.0.3
[pip3] nvidia-cuda-cccl==13.3.3.3.1
[pip3] nvidia-cuda-crt==13.3.33
[pip3] nvidia-cuda-cupti==13.0.85
[pip3] nvidia-cuda-nvcc==13.3.33
[pip3] nvidia-cuda-nvrtc==13.0.88
[pip3] nvidia-cuda-runtime==13.0.96
[pip3] nvidia-cudnn-cu13==9.19.0.56
[pip3] nvidia-cudnn-frontend==1.18.0
[pip3] nvidia-cufft==12.0.0.61
[pip3] nvidia-cufile==1.15.1.6
[pip3] nvidia-curand==10.4.0.35
[pip3] nvidia-cusolver==12.0.4.66
[pip3] nvidia-cusparse==12.6.3.3
[pip3] nvidia-cusparselt-cu13==0.8.0
[pip3] nvidia-cutlass-dsl==4.5.2
[pip3] nvidia-cutlass-dsl-libs-base==4.5.2
[pip3] nvidia-cutlass-dsl-libs-cu13==4.5.2
[pip3] nvidia-ml-py==13.595.45
[pip3] nvidia-nccl-cu13==2.28.9
[pip3] nvidia-nvjitlink==13.0.88
[pip3] nvidia-nvshmem-cu13==3.4.5
[pip3] nvidia-nvtx==13.0.85
[pip3] nvidia-nvvm==13.3.33
[pip3] pyzmq==27.1.0
[pip3] tokenspeed-triton==3.7.10.post20260505
[pip3] torch==2.11.0+cu130
[pip3] torch_c_dlpack_ext==0.1.5
[pip3] torchaudio==2.11.0+cu130
[pip3] torchvision==0.26.0+cu130
[pip3] transformers==5.9.0
[pip3] triton==3.6.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.21.1rc1.dev433+g0585b5ba2 (git sha: 0585b5ba2)
vLLM Build Flags:
  CUDA Archs: Not Set; ROCm: Disabled; XPU: Disabled
GPU Topology:
  	GPU0	GPU1	GPU2	GPU3	GPU4	GPU5	GPU6	GPU7	CPU Affinity	NUMA Affinity	GPU NUMA ID
GPU0	 X 	NODE	NODE	NODE	NODE	NODE	NODE	NODE	0-47	0		N/A
GPU1	NODE	 X 	PHB	NODE	NODE	NODE	NODE	NODE	0-47	0		N/A
GPU2	NODE	PHB	 X 	NODE	NODE	NODE	NODE	NODE	0-47	0		N/A
GPU3	NODE	NODE	NODE	 X 	PHB	PHB	NODE	NODE	0-47	0		N/A
GPU4	NODE	NODE	NODE	PHB	 X 	PHB	NODE	NODE	0-47	0		N/A
GPU5	NODE	NODE	NODE	PHB	PHB	 X 	NODE	NODE	0-47	0		N/A
GPU6	NODE	NODE	NODE	NODE	NODE	NODE	 X 	PHB	0-47	0		N/A
GPU7	NODE	NODE	NODE	NODE	NODE	NODE	PHB	 X 	0-47	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
==============================
CUDA_PATH=/opt/cuda
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
TORCHINDUCTOR_CACHE_DIR=/tmp/torchinductor_hubby

</details>

🐛 Describe the bug

Description

Tagging the original author, @jeejeelee for visibility. Full details are included above, but for brevity, I am running 8 RTX 3090 on a single node, all on PCIe. No NVLink.

Working version: 0.21.1rc1.dev268+gd0a100c87 Broken version: 0.21.1rc1.dev433+g0585b5ba2

Expected behavior

The model should load successfully, falling back to NCCL-based AllReduce when P2P access is unavailable, as it did prior to PR #43410.

Reproduction

Environment variables

OMP_NUM_THREADS=1
CUDA_DEVICE_ORDER=PCI_BUS_ID
CUDA_VISIBLE_DEVICES=0,3,6,7,2,4,5,1
HF_TOKEN=hf_token
HF_XET_HIGH_PERFORMANCE=1
CUDA_HOME=/opt/cuda
VLLM_MARLIN_USE_ATOMIC_ADD=1
VLLM_MEMORY_PROFILER_ESTIMATE_CUDAGRAPHS=1
VLLM_FLOAT32_MATMUL_PRECISION=high
VLLM_USE_NCCL_SYMM_MEM=0
SAFETENSORS_FAST_GPU=1

Start command

uv run vllm serve Intel/MiniMax-M2.5-int4-AutoRound
--served-model-name minimax-m2.5
--tensor-parallel-size 8
--enable-expert-parallel
--max-model-len auto
--gpu-memory-utilization 0.93
--attention-backend flashinfer
--enable-flashinfer-autotune
--reasoning-parser minimax_m2
--tool-call-parser minimax_m2
--enable-auto-tool-choice
--max-num-seqs 4
--max-num-batched-tokens 4096
--enable-prefix-caching
--enable-chunked-prefill
--trust-remote-code
--host 0.0.0.0
--port ${PORT}
--seed 3407
--disable-custom-all-reduce

Error

(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] WorkerProc failed to start.
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] Traceback (most recent call last):
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/executor/multiproc_executor.py", line 841, in worker_main
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     worker = WorkerProc(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return func(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/executor/multiproc_executor.py", line 620, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.worker.load_model()
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~~~^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/worker/gpu_worker.py", line 345, in load_model
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.model_runner.load_model(load_dummy_weights=load_dummy_weights)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return func(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/v1/worker/gpu_model_runner.py", line 5052, in load_model
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.model = model_loader.load_model(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                  ~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         vllm_config=self.vllm_config, model_config=self.model_config
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return func(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/model_loader/base_loader.py", line 55, in load_model
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     model = initialize_model(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         vllm_config=vllm_config,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         model_config=model_config,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         prefix=prefix,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/tracing/otel.py", line 178, in sync_wrapper
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return func(*args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/model_loader/utils.py", line 61, in initialize_model
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     model = model_class(vllm_config=vllm_config, prefix=prefix)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 537, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.model = MiniMaxM2Model(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                  ~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         vllm_config=vllm_config, prefix=maybe_prefix(prefix, "model")
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/compilation/decorators.py", line 383, in __init__
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     old_init(self, *args, **kwargs)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~~~^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 351, in __init__
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.start_layer, self.end_layer, self.layers = make_layers(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                                                     ~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         config.num_hidden_layers,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ...<7 lines>...
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         prefix=f"{prefix}.layers",
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/utils.py", line 643, in make_layers
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     + get_offloader().wrap_modules(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]       ~~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         layer_fn(prefix=f"{prefix}.{idx}") for idx in range(start_layer, end_layer)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/offloader/base.py", line 104, in wrap_modules
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     return list(modules_generator)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/utils.py", line 644, in <genexpr>
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     layer_fn(prefix=f"{prefix}.{idx}") for idx in range(start_layer, end_layer)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 353, in <lambda>
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     lambda prefix: MiniMaxM2DecoderLayer(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                    ~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         config,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ...<3 lines>...
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         quant_config=quant_config,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ),
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 276, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.self_attn = MiniMaxM2Attention(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                      ~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         hidden_size=self.hidden_size,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ...<10 lines>...
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         prefix=f"{prefix}.self_attn",
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/models/minimax_m2.py", line 221, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.q_norm = MiniMaxText01RMSNormTP(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                   ~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         self.head_dim * self.total_num_heads, eps=rms_norm_eps
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/rms_norm_tp.py", line 146, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self.workspace = get_allreduce_workspace(
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                      ~~~~~^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         rank=self.tp_rank,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ...<2 lines>...
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         process_group=get_tp_group().cpu_group,
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     )
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 300, in get_allreduce_workspace
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ws = LamportWorkspace(rank, world_size, comm_size, process_group)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 185, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     self._lamport = IpcBuffer(rank, world_size, lamport_total, process_group)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]                     ~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 100, in init
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     _check(err)
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     ~~~~^^^^^
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]   File "/home/hubby/.venv/lib/python3.13/site-packages/vllm/model_executor/layers/minimax_rms_norm/lamport_workspace.py", line 30, in _check
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874]     raise RuntimeError(f"CUDA runtime error: {error}")
(Worker_TP1_EP1 pid=912664) ERROR 05-29 16:44:33 [multiproc_executor.py:874] RuntimeError: CUDA runtime error: <cudaError_t.cudaErrorPeerAccessUnsupported: 217>

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.

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

FAQ

Expected behavior

The model should load successfully, falling back to NCCL-based AllReduce when P2P access is unavailable, as it did prior to PR #43410.

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering

vllm - 💡(How to fix) Fix [Bug]: [Regression] MiniMax-M2.5-int4 fails with cudaErrorPeerAccessUnsupported since PR #43410

Recommended Tools

GitHub issue graph ai analysis

Error Message

Root Cause

Fix Action

Fix / Workaround

============================== CPU Info

Code Example

Your current environment

🐛 Describe the bug

Description

Expected behavior

Reproduction

Environment variables

Start command

Error

Before submitting a new issue...

FAQ

Expected behavior

Still need to ship something?

TRENDING