Skip to content

[Bug]: Consistent vLLM crash when trying to reuse previous torch compilation with DeepSeek R1 on B200 #24915

Closed
#25093
Closed
[Bug]: Consistent vLLM crash when trying to reuse previous torch compilation with DeepSeek R1 on B200#24915
#25093
Labels
bugSomething isn't workingtorch.compile
@smarterclayton

Description

@smarterclayton
smarterclayton
opened on Sep 15, 2025

Your current environment

The output of python collect_env.py for the prefill worker 
Collecting environment information...
==============================
        System Info
==============================
OS                           : Ubuntu 22.04.5 LTS (x86_64)
GCC version                  : (Ubuntu 11.4.0-1ubuntu1~22.04.2) 11.4.0
Clang version                : 14.0.0-1ubuntu1.1
CMake version                : version 4.1.0
Libc version                 : glibc-2.35

==============================
       PyTorch Info
==============================
PyTorch version              : 2.8.0+cu128
Is debug build               : False
CUDA used to build PyTorch   : 12.8
ROCM used to build PyTorch   : N/A

==============================
      Python Environment
==============================
Python version               : 3.12.11 (main, Jun  4 2025, 08:56:18) [GCC 11.4.0] (64-bit runtime)
Python platform              : Linux-6.6.93+-x86_64-with-glibc2.35

==============================
       CUDA / GPU Info
==============================
Is CUDA available            : True
CUDA runtime version         : 12.8.93
CUDA_MODULE_LOADING set to   : LAZY
GPU models and configuration : 
GPU 0: NVIDIA B200
GPU 1: NVIDIA B200
GPU 2: NVIDIA B200
GPU 3: NVIDIA B200
GPU 4: NVIDIA B200
GPU 5: NVIDIA B200
GPU 6: NVIDIA B200
GPU 7: NVIDIA B200

Nvidia driver version        : 575.57.08
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:                           52 bits physical, 57 bits virtual
Byte Order:                              Little Endian
CPU(s):                                  224
On-line CPU(s) list:                     0-223
Vendor ID:                               GenuineIntel
Model name:                              INTEL(R) XEON(R) PLATINUM 8581C CPU @ 2.10GHz
CPU family:                              6
Model:                                   207
Thread(s) per core:                      2
Core(s) per socket:                      56
Socket(s):                               2
Stepping:                                2
BogoMIPS:                                4200.00
Flags:                                   fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc rep_good nopl xtopology nonstop_tsc cpuid tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch ssbd ibrs ibpb stibp ibrs_enhanced fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves avx_vnni avx512_bf16 wbnoinvd arat avx512vbmi umip avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid cldemote movdiri movdir64b fsrm md_clear serialize tsxldtrk amx_bf16 avx512_fp16 amx_tile amx_int8 arch_capabilities
Hypervisor vendor:                       KVM
Virtualization type:                     full
L1d cache:                               5.3 MiB (112 instances)
L1i cache:                               3.5 MiB (112 instances)
L2 cache:                                224 MiB (112 instances)
L3 cache:                                520 MiB (2 instances)
NUMA node(s):                            2
NUMA node0 CPU(s):                       0-55,112-167
NUMA node1 CPU(s):                       56-111,168-223
Vulnerability Gather data sampling:      Not affected
Vulnerability Indirect target selection: Mitigation; Aligned branch/return thunks
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:      Not affected
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 disabled; PBRSB-eIBRS SW sequence; BHI SW loop, KVM SW loop
Vulnerability Srbds:                     Not affected
Vulnerability Tsx async abort:           Not affected

==============================
Versions of relevant libraries
==============================
[pip3] numpy==2.2.6
[pip3] nvidia-cublas-cu12==12.8.4.1
[pip3] nvidia-cuda-cupti-cu12==12.8.90
[pip3] nvidia-cuda-nvrtc-cu12==12.8.93
[pip3] nvidia-cuda-runtime-cu12==12.8.90
[pip3] nvidia-cudnn-cu12==9.10.2.21
[pip3] nvidia-cufft-cu12==11.3.3.83
[pip3] nvidia-cufile-cu12==1.13.1.3
[pip3] nvidia-curand-cu12==10.3.9.90
[pip3] nvidia-cusolver-cu12==11.7.3.90
[pip3] nvidia-cusparse-cu12==12.5.8.93
[pip3] nvidia-cusparselt-cu12==0.7.1
[pip3] nvidia-nccl-cu12==2.27.3
[pip3] nvidia-nvjitlink-cu12==12.8.93
[pip3] nvidia-nvtx-cu12==12.8.90
[pip3] pyzmq==27.1.0
[pip3] torch==2.8.0+cu128
[pip3] torchaudio==2.8.0+cu128
[pip3] torchvision==0.23.0+cu128
[pip3] transformers==4.56.1
[pip3] triton==3.4.0
[conda] Could not collect

==============================
         vLLM Info
==============================
ROCM Version                 : Could not collect
vLLM Version                 : 0.1.dev1+g0e219cd50 (git sha: 0e219cd50)
vLLM Build Flags:
  CUDA Archs: 9.0a 10.0; ROCm: Disabled
GPU Topology:
  	GPU0	GPU1	GPU2	GPU3	GPU4	GPU5	GPU6	GPU7	NIC0	NIC1	NIC2	NIC3	NIC4	NIC5	NIC6	NIC7	CPU Affinity	NUMA Affinity	GPU NUMA ID
GPU0	 X 	NV18	NV18	NV18	NV18	NV18	NV18	NV18	PIX	PIX	NODE	NODE	SYS	SYS	SYS	SYS	0-55,112-167	0		N/A
GPU1	NV18	 X 	NV18	NV18	NV18	NV18	NV18	NV18	PIX	PIX	NODE	NODE	SYS	SYS	SYS	SYS	0-55,112-167	0		N/A
GPU2	NV18	NV18	 X 	NV18	NV18	NV18	NV18	NV18	NODE	NODE	PIX	PIX	SYS	SYS	SYS	SYS	0-55,112-167	0		N/A
GPU3	NV18	NV18	NV18	 X 	NV18	NV18	NV18	NV18	NODE	NODE	PIX	PIX	SYS	SYS	SYS	SYS	0-55,112-167	0		N/A
GPU4	NV18	NV18	NV18	NV18	 X 	NV18	NV18	NV18	SYS	SYS	SYS	SYS	PIX	PIX	NODE	NODE	56-111,168-223	1		N/A
GPU5	NV18	NV18	NV18	NV18	NV18	 X 	NV18	NV18	SYS	SYS	SYS	SYS	PIX	PIX	NODE	NODE	56-111,168-223	1		N/A
GPU6	NV18	NV18	NV18	NV18	NV18	NV18	 X 	NV18	SYS	SYS	SYS	SYS	NODE	NODE	PIX	PIX	56-111,168-223	1		N/A
GPU7	NV18	NV18	NV18	NV18	NV18	NV18	NV18	 X 	SYS	SYS	SYS	SYS	NODE	NODE	PIX	PIX	56-111,168-223	1		N/A
NIC0	PIX	PIX	NODE	NODE	SYS	SYS	SYS	SYS	 X 	PIX	NODE	NODE	SYS	SYS	SYS	SYS				
NIC1	PIX	PIX	NODE	NODE	SYS	SYS	SYS	SYS	PIX	 X 	NODE	NODE	SYS	SYS	SYS	SYS				
NIC2	NODE	NODE	PIX	PIX	SYS	SYS	SYS	SYS	NODE	NODE	 X 	PIX	SYS	SYS	SYS	SYS				
NIC3	NODE	NODE	PIX	PIX	SYS	SYS	SYS	SYS	NODE	NODE	PIX	 X 	SYS	SYS	SYS	SYS				
NIC4	SYS	SYS	SYS	SYS	PIX	PIX	NODE	NODE	SYS	SYS	SYS	SYS	 X 	PIX	NODE	NODE				
NIC5	SYS	SYS	SYS	SYS	PIX	PIX	NODE	NODE	SYS	SYS	SYS	SYS	PIX	 X 	NODE	NODE				
NIC6	SYS	SYS	SYS	SYS	NODE	NODE	PIX	PIX	SYS	SYS	SYS	SYS	NODE	NODE	 X 	PIX				
NIC7	SYS	SYS	SYS	SYS	NODE	NODE	PIX	PIX	SYS	SYS	SYS	SYS	NODE	NODE	PIX	 X 				

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

NIC Legend:

  NIC0: mlx5_0
  NIC1: mlx5_1
  NIC2: mlx5_2
  NIC3: mlx5_3
  NIC4: mlx5_4
  NIC5: mlx5_5
  NIC6: mlx5_6
  NIC7: mlx5_7

==============================
     Environment Variables
==============================
CUDA_HOME=/usr/local/cuda/
CUDA_HOME=/usr/local/cuda/
CUDA_MAJOR=12
CUDA_VERSION=12.8.1
LD_LIBRARY_PATH=/usr/local/nvshmem/lib:/usr/local/nixl/lib/x86_64-linux-gnu:/opt/ucx/lib:/usr/local/lib:/usr/local/gib/lib64:/usr/local/nvidia/lib64:/usr/local/cuda/lib64:
NCCL_CROSS_NIC=0
NCCL_DEBUG=INFO
NCCL_IB_ADAPTIVE_ROUTING=1
NCCL_IB_FIFO_TC=84
NCCL_IB_QPS_PER_CONNECTION=4
NCCL_IB_TC=52
NCCL_NET=gIB
NCCL_NET_GDR_LEVEL=PIX
NCCL_NVLS_CHUNKSIZE=524288
NCCL_P2P_NET_CHUNKSIZE=131072
NCCL_TUNER_CONFIG_PATH=/usr/local/gib/configs/tuner_config_a4.txtpb
NCCL_VERSION=2.25.1-1
NVIDIA_DRIVER_CAPABILITIES=compute,utility
NVIDIA_GDRCOPY=disabled
NVIDIA_PRODUCT_NAME=CUDA
NVIDIA_REQUIRE_CUDA=cuda>=12.8 brand=unknown,driver>=470,driver<471 brand=grid,driver>=470,driver<471 brand=tesla,driver>=470,driver<471 brand=nvidia,driver>=470,driver<471 brand=quadro,driver>=470,driver<471 brand=quadrortx,driver>=470,driver<471 brand=nvidiartx,driver>=470,driver<471 brand=vapps,driver>=470,driver<471 brand=vpc,driver>=470,driver<471 brand=vcs,driver>=470,driver<471 brand=vws,driver>=470,driver<471 brand=cloudgaming,driver>=470,driver<471 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=vapps,driver>=535,driver<536 brand=vpc,driver>=535,driver<536 brand=vcs,driver>=535,driver<536 brand=vws,driver>=535,driver<536 brand=cloudgaming,driver>=535,driver<536 brand=unknown,driver>=550,driver<551 brand=grid,driver>=550,driver<551 brand=tesla,driver>=550,driver<551 brand=nvidia,driver>=550,driver<551 brand=quadro,driver>=550,driver<551 brand=quadrortx,driver>=550,driver<551 brand=nvidiartx,driver>=550,driver<551 brand=vapps,driver>=550,driver<551 brand=vpc,driver>=550,driver<551 brand=vcs,driver>=550,driver<551 brand=vws,driver>=550,driver<551 brand=cloudgaming,driver>=550,driver<551 brand=unknown,driver>=560,driver<561 brand=grid,driver>=560,driver<561 brand=tesla,driver>=560,driver<561 brand=nvidia,driver>=560,driver<561 brand=quadro,driver>=560,driver<561 brand=quadrortx,driver>=560,driver<561 brand=nvidiartx,driver>=560,driver<561 brand=vapps,driver>=560,driver<561 brand=vpc,driver>=560,driver<561 brand=vcs,driver>=560,driver<561 brand=vws,driver>=560,driver<561 brand=cloudgaming,driver>=560,driver<561 brand=unknown,driver>=565,driver<566 brand=grid,driver>=565,driver<566 brand=tesla,driver>=565,driver<566 brand=nvidia,driver>=565,driver<566 brand=quadro,driver>=565,driver<566 brand=quadrortx,driver>=565,driver<566 brand=nvidiartx,driver>=565,driver<566 brand=vapps,driver>=565,driver<566 brand=vpc,driver>=565,driver<566 brand=vcs,driver>=565,driver<566 brand=vws,driver>=565,driver<566 brand=cloudgaming,driver>=565,driver<566
NVIDIA_VISIBLE_DEVICES=all
TORCH_CUDA_ARCH_LIST=9.0a 10.0
VLLM_ALL2ALL_BACKEND=deepep_low_latency
VLLM_ATTENTION_BACKEND=CUTLASS_MLA
VLLM_CACHE_ROOT=/root/.nv/ComputeCache/.vllm
VLLM_LOGGING_LEVEL=DEBUG
VLLM_MOE_DP_CHUNK_SIZE=512
VLLM_NIXL_SIDE_CHANNEL_HOST=10.4.2.177
VLLM_NIXL_SIDE_CHANNEL_PORT=6555
VLLM_RANDOMIZE_DP_DUMMY_INPUTS=1
VLLM_TORCH_PROFILER_DIR=/code/traces
VLLM_USE_DEEP_GEMM=1
PYTORCH_NVML_BASED_CUDA_CHECK=1
TORCHINDUCTOR_COMPILE_THREADS=1
CUDA_MODULE_LOADING=LAZY

🐛 Describe the bug

When deploying deepseek R1 in a 1p1d config with the following config for the prefill node https://github.com/smarterclayton/ig-load/blob/pd_3/config/deploy_base_deepseek_r1_ep_pd.yaml#L85, I experience a crash when I try to reuse the torch compilation cache from a previous successful startup.

Scenario:

  1. VLLM_CACHE_ROOT is set to a directory
  2. The decode and prefill instances start correctly the first time
  3. If prefill restarts for any reason (crash, manual restart), subsequent prefill server starts fails with the included crash below
  4. If I rm -rf $VLLM_CACHE_ROOT/torch_compile_cache, the next startup succeeds, but subsequent prefill starts continue to fail
(EngineCore_DP5 pid=953) DEBUG 09-15 16:35:12 [compilation/backends.py:124] Directly load the 118-th graph for dynamic shape from inductor_standalone via handle ('artifact_shape_None_subgraph_118', '/root/.nv/ComputeCache/.vllm/torch_compile_cache/3f2fad2cbf/rank_0_5/backbone/artifact_shape_None_subgraph_118')
(EngineCore_DP5 pid=953) DEBUG 09-15 16:35:12 [compilation/backends.py:124] Directly load the 119-th graph for dynamic shape from inductor_standalone via handle ('artifact_shape_None_subgraph_119', '/root/.nv/ComputeCache/.vllm/torch_compile_cache/3f2fad2cbf/rank_0_5/backbone/artifact_shape_None_subgraph_119')
(EngineCore_DP5 pid=953) INFO 09-15 16:35:12 [compilation/backends.py:161] Directly load the compiled graph(s) for dynamic shape from the cache, took 6.271 s
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712] EngineCore failed to start.
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712] Traceback (most recent call last):
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/engine/core.py", line 699, in run_engine_core
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     engine_core = DPEngineCoreProc(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/engine/core.py", line 969, in __init__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     super().__init__(vllm_config, local_client, handshake_address,
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/engine/core.py", line 502, in __init__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     super().__init__(vllm_config, executor_class, log_stats,
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/engine/core.py", line 90, in __init__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     self._initialize_kv_caches(vllm_config)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/engine/core.py", line 188, in _initialize_kv_caches
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     self.model_executor.determine_available_memory())
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/executor/abstract.py", line 85, in determine_available_memory
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return self.collective_rpc("determine_available_memory")
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/executor/uniproc_executor.py", line 83, in collective_rpc
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return [run_method(self.driver_worker, method, args, kwargs)]
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/utils/__init__.py", line 3067, in run_method
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return func(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 120, in decorate_context
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return func(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/worker/gpu_worker.py", line 263, in determine_available_memory
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     self.model_runner.profile_run()
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/worker/gpu_model_runner.py", line 3047, in profile_run
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     = self._dummy_run(self.max_num_tokens, is_profile=True)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/utils/_contextlib.py", line 120, in decorate_context
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return func(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/v1/worker/gpu_model_runner.py", line 2825, in _dummy_run
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     outputs = self.model(
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]               ^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return self._call_impl(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1784, in _call_impl
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return forward_call(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/model_executor/models/deepseek_v2.py", line 907, in forward
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     hidden_states = self.model(input_ids, positions, intermediate_tensors,
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]                     ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/compilation/decorators.py", line 305, in __call__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     output = self.compiled_callable(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/_dynamo/eval_frame.py", line 736, in compile_wrapper
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return fn(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/model_executor/models/deepseek_v2.py", line 767, in forward
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     def forward(
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/_dynamo/eval_frame.py", line 375, in __call__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return super().__call__(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return self._call_impl(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1784, in _call_impl
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return forward_call(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/_dynamo/eval_frame.py", line 929, in _fn
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return fn(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/fx/graph_module.py", line 848, in call_wrapped
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return self._wrapped_call(self, *args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/fx/graph_module.py", line 424, in __call__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     raise e
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/fx/graph_module.py", line 411, in __call__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return super(self.cls, obj).__call__(*args, **kwargs)  # type: ignore[misc]
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1773, in _wrapped_call_impl
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return self._call_impl(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/nn/modules/module.py", line 1784, in _call_impl
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return forward_call(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "<eval_with_key>.240", line 723, in forward
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     submod_10 = self.submod_10(submod_9, s72, getitem_22, l_self_modules_layers_modules_4_modules_input_layernorm_parameters_weight_, l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_fused_qkv_a_proj_parameters_weight_, l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_fused_qkv_a_proj_parameters_weight_scale_inv_, l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_q_a_layernorm_parameters_weight_, l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_q_b_proj_parameters_weight_, l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_q_b_proj_parameters_weight_scale_inv_, l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_kv_a_layernorm_parameters_weight_, l_self_modules_layers_modules_0_modules_self_attn_modules_mla_attn_modules_rotary_emb_buffers_cos_sin_cache_, l_positions_);  submod_9 = getitem_22 = l_self_modules_layers_modules_4_modules_input_layernorm_parameters_weight_ = l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_fused_qkv_a_proj_parameters_weight_ = l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_fused_qkv_a_proj_parameters_weight_scale_inv_ = l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_q_a_layernorm_parameters_weight_ = l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_q_b_proj_parameters_weight_ = l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_q_b_proj_parameters_weight_scale_inv_ = l_self_modules_layers_modules_4_modules_self_attn_modules_mla_attn_modules_kv_a_layernorm_parameters_weight_ = None
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/compilation/cuda_graph.py", line 119, in __call__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return self.runnable(*args, **kwargs)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/compilation/cuda_piecewise_backend.py", line 90, in __call__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return self.compiled_graph_for_general_shape(*args)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/vllm/vllm/compilation/compiler_interface.py", line 237, in compiled_graph_wrapper
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     graph_output = inductor_compiled_graph(*args)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/_inductor/standalone_compile.py", line 62, in __call__
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return self._compiled_fn(*args)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]            ^^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/_inductor/standalone_compile.py", line 182, in <lambda>
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     return CompiledArtifact(lambda *args: compiled_fn(list(args)), None)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]                                           ^^^^^^^^^^^^^^^^^^^^^^^
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 353, in runtime_wrapper
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     torch.autograd.graph.increment_version(mutated_args)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]   File "/app/venv/lib/python3.12/site-packages/torch/autograd/graph.py", line 235, in increment_version
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712]     torch._C._increment_version(tensor)
(EngineCore_DP5 pid=953) ERROR 09-15 16:35:15 [v1/engine/core.py:712] RuntimeError: increment_version expects each element of the iterable to be a tensor

This does not occur on the decode configuration.

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.

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't workingtorch.compile

    Type

    No type

    Projects

    DeepSeek V3/R1

    Status

    Done
    torch.compile integration

    Status

    Done

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions