[Bug]: VLLM crashes when prefix caching is enabled #7003

m-harmonic · 2024-07-31T20:05:14Z

Your current environment

PyTorch version: 2.3.0+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.4 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.28.3
Libc version: glibc-2.35

Python version: 3.10.12 (main, Nov 20 2023, 15:14:05) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-5.15.0-1008-gcp-tcpx-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.4.99
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration:
GPU 0: NVIDIA H100 80GB HBM3
GPU 1: NVIDIA H100 80GB HBM3
GPU 2: NVIDIA H100 80GB HBM3
GPU 3: NVIDIA H100 80GB HBM3
GPU 4: NVIDIA H100 80GB HBM3
GPU 5: NVIDIA H100 80GB HBM3
GPU 6: NVIDIA H100 80GB HBM3
GPU 7: NVIDIA H100 80GB HBM3

Nvidia driver version: 535.104.05
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.0.0
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
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):                             208
On-line CPU(s) list:                0-207
Vendor ID:                          GenuineIntel
Model name:                         Intel(R) Xeon(R) Platinum 8481C CPU @ 2.70GHz
CPU family:                         6
Model:                              143
Thread(s) per core:                 2
Core(s) per socket:                 52
Socket(s):                          2
Stepping:                           8
BogoMIPS:                           5399.99
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 ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced fsgsbase tsc_adjust bmi1 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 arat avx512vbmi umip avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid cldemote movdiri movdir64b fsrm md_clear serialize amx_bf16 avx512_fp16 amx_tile amx_int8 arch_capabilities
Hypervisor vendor:                  KVM
Virtualization type:                full
L1d cache:                          4.9 MiB (104 instances)
L1i cache:                          3.3 MiB (104 instances)
L2 cache:                           208 MiB (104 instances)
L3 cache:                           210 MiB (2 instances)
NUMA node(s):                       2
NUMA node0 CPU(s):                  0-51,104-155
NUMA node1 CPU(s):                  52-103,156-207
Vulnerability Gather data sampling: 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 Retbleed:             Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass:    Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:           Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:           Mitigation; Enhanced IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected

Versions of relevant libraries:
[pip3] flake8==7.1.0
[pip3] mypy==1.11.0
[pip3] mypy-extensions==1.0.0
[pip3] numpy==1.24.4
[pip3] nvidia-nccl-cu12==2.20.5
[pip3] onnx==1.15.0rc2
[pip3] optree==0.10.0
[pip3] pytorch-lightning==2.3.3
[pip3] pytorch-quantization==2.1.2
[pip3] pytorch-triton==2.2.0+e28a256d7
[pip3] torch==2.3.0
[pip3] torch-tensorrt==2.3.0a0
[pip3] torchdata==0.7.1a0
[pip3] torchmetrics==1.4.0.post0
[pip3] torchtext==0.18.0
[pip3] torchvision==0.18.0
[pip3] transformers==4.40.2
[pip3] triton==2.3.0
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.4.3
vLLM Build Flags:
CUDA Archs: 5.2 6.0 6.1 7.0 7.2 7.5 8.0 8.6 8.7 9.0+PTX; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0	GPU1	GPU2	GPU3	GPU4	GPU5	GPU6	GPU7	CPU Affinity	NUMA Affinity	GPU NUMA ID
GPU0	 X 	NV18	NV18	NV18	NV18	NV18	NV18	NV18	0-51,104-155	0		N/A
GPU1	NV18	 X 	NV18	NV18	NV18	NV18	NV18	NV18	0-51,104-155	0		N/A
GPU2	NV18	NV18	 X 	NV18	NV18	NV18	NV18	NV18	0-51,104-155	0		N/A
GPU3	NV18	NV18	NV18	 X 	NV18	NV18	NV18	NV18	0-51,104-155	0		N/A
GPU4	NV18	NV18	NV18	NV18	 X 	NV18	NV18	NV18	52-103,156-207	1		N/A
GPU5	NV18	NV18	NV18	NV18	NV18	 X 	NV18	NV18	52-103,156-207	1		N/A
GPU6	NV18	NV18	NV18	NV18	NV18	NV18	 X 	NV18	52-103,156-207	1		N/A
GPU7	NV18	NV18	NV18	NV18	NV18	NV18	NV18	 X 	52-103,156-207	1		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

🐛 Describe the bug

VLLM crashes 100% of the time when using an async engine initialized with enable_prefix_caching=True

The stacktrace is:

File "/workspace/src/vllm/vllm/model_executor/layers/sampler.py", line 112, in forward
  prompt_logprobs, sample_logprobs = _get_logprobs(
File "/workspace/src/vllm/vllm/model_executor/layers/sampler.py", line 760, in _get_logprobs
  assert len(next_token_ids) == len(query_indices)
AssertionError

The problem goes away completely when enable_prefix_caching=False. This is on VLLM version 0.4.3.

The text was updated successfully, but these errors were encountered:

tristan279 · 2024-08-01T12:39:44Z

with vllm=0.5.3.post, only thing i changed was enabling prefix caching, crashed with illegal cuda access errors

cjfcsjt · 2024-08-10T14:58:40Z

@zachzzc @raywanb Still facing the same issue when adopt the following model:

self.vlm_model = LLM(
            model="openbmb/MiniCPM-V-2_6",
            max_model_len=4096,
            trust_remote_code=True,
            gpu_memory_utilization=0.5,
            enable_prefix_caching=True
        )

My vllm version is 0.5.4+cu122. Here are my env:

PyTorch version: 2.4.0+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.2 LTS (x86_64)
GCC version: (Ubuntu 9.5.0-1ubuntu1~22.04) 9.5.0
Clang version: Could not collect
CMake version: version 3.30.1
Libc version: glibc-2.35

Python version: 3.10.13 (main, Sep 11 2023, 13:44:35) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.15.0-113-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.2.140
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: 
GPU 0: NVIDIA A100 80GB PCIe
GPU 1: NVIDIA A100 80GB PCIe
GPU 2: NVIDIA A100 80GB PCIe
GPU 3: NVIDIA A100 80GB PCIe
GPU 4: NVIDIA A100 80GB PCIe
GPU 5: NVIDIA A100 80GB PCIe
GPU 6: NVIDIA A100 80GB PCIe
GPU 7: NVIDIA A100 80GB PCIe

Nvidia driver version: 535.183.01
cuDNN version: Probably one of the following:
/usr/local/cuda-12.2/targets/x86_64-linux/lib/libcudnn.so.8.9.7
/usr/local/cuda-12.2/targets/x86_64-linux/lib/libcudnn_adv_infer.so.8.9.7
/usr/local/cuda-12.2/targets/x86_64-linux/lib/libcudnn_adv_train.so.8.9.7
/usr/local/cuda-12.2/targets/x86_64-linux/lib/libcudnn_cnn_infer.so.8.9.7
/usr/local/cuda-12.2/targets/x86_64-linux/lib/libcudnn_cnn_train.so.8.9.7
/usr/local/cuda-12.2/targets/x86_64-linux/lib/libcudnn_ops_infer.so.8.9.7
/usr/local/cuda-12.2/targets/x86_64-linux/lib/libcudnn_ops_train.so.8.9.7
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:                       x86_64
CPU op-mode(s):                     32-bit, 64-bit
Address sizes:                      46 bits physical, 48 bits virtual
Byte Order:                         Little Endian
CPU(s):                             96
On-line CPU(s) list:                0-95
Vendor ID:                          GenuineIntel
Model name:                         Intel(R) Xeon(R) Gold 6248R CPU @ 3.00GHz
CPU family:                         6
Model:                              85
Thread(s) per core:                 2
Core(s) per socket:                 24
Socket(s):                          2
Stepping:                           7
CPU max MHz:                        4000.0000
CPU min MHz:                        1200.0000
BogoMIPS:                           6000.00
Flags:                              fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single intel_ppin ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts pku ospke avx512_vnni md_clear flush_l1d arch_capabilities
Virtualization:                     VT-x
L1d cache:                          1.5 MiB (48 instances)
L1i cache:                          1.5 MiB (48 instances)
L2 cache:                           48 MiB (48 instances)
L3 cache:                           71.5 MiB (2 instances)
NUMA node(s):                       2
NUMA node0 CPU(s):                  0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78,80,82,84,86,88,90,92,94
NUMA node1 CPU(s):                  1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79,81,83,85,87,89,91,93,95
Vulnerability Gather data sampling: Mitigation; Microcode
Vulnerability Itlb multihit:        KVM: Mitigation: VMX disabled
Vulnerability L1tf:                 Not affected
Vulnerability Mds:                  Not affected
Vulnerability Meltdown:             Not affected
Vulnerability Mmio stale data:      Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Retbleed:             Mitigation; Enhanced IBRS
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass:    Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:           Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:           Mitigation; Enhanced IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI Syscall hardening, KVM SW loop
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Mitigation; TSX disabled

Versions of relevant libraries:
[pip3] mypy-extensions==1.0.0
[pip3] numpy==1.26.4
[pip3] nvidia-nccl-cu12==2.20.5
[pip3] open-clip-torch==2.24.0
[pip3] optree==0.12.1
[pip3] pyzmq==26.0.3
[pip3] torch==2.3.1
[pip3] torch-struct==0.5
[pip3] torchaudio==2.3.1+cu121
[pip3] torchvision==0.18.1
[pip3] transformers==4.44.0
[pip3] transformers-stream-generator==0.0.5
[pip3] triton==3.0.0
[pip3] vllm-nccl-cu12==2.18.1.0.4.0
[pip3] zmq==0.0.0
[conda] blas                      1.0                         mkl  
[conda] ffmpeg                    4.3                  hf484d3e_0    pytorch
[conda] libjpeg-turbo             2.0.0                h9bf148f_0    pytorch
[conda] mkl                       2023.1.0         h213fc3f_46344  
[conda] mkl-service               2.4.0           py310h5eee18b_1  
[conda] mkl_fft                   1.3.8           py310h5eee18b_0  
[conda] mkl_random                1.2.4           py310hdb19cb5_0  
[conda] nomkl                     0.0.3                    pypi_0    pypi
[conda] numpy                     1.26.4          py310h5f9d8c6_0  
[conda] numpy-base                1.26.4          py310hb5e798b_0  
[conda] nvidia-nccl-cu12          2.20.5                   pypi_0    pypi
[conda] open-clip-torch           2.24.0                   pypi_0    pypi
[conda] optree                    0.12.1                   pypi_0    pypi
[conda] pytorch-cuda              12.1                 ha16c6d3_5    pytorch
[conda] pytorch-mutex             1.0                         cpu    pytorch
[conda] pyzmq                     26.0.0                   pypi_0    pypi
[conda] torch                     2.3.1                    pypi_0    pypi
[conda] torch-struct              0.5                      pypi_0    pypi
[conda] torchaudio                2.3.1+cu121              pypi_0    pypi
[conda] torchvision               0.19.0                   pypi_0    pypi
[conda] transformers              4.44.0                   pypi_0    pypi
[conda] transformers-stream-generator 0.0.5                    pypi_0    pypi
[conda] triton                    3.0.0                    pypi_0    pypi
[conda] vllm-nccl-cu12            2.18.1.0.4.0             pypi_0    pypi
[conda] zmq                       0.0.0                    pypi_0    pypi
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.5.4
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      PIX     NODE    NODE    SYS     SYS     SYS     SYS     0,2,4,6,8,10    0               N/A
GPU1    PIX      X      NODE    NODE    SYS     SYS     SYS     SYS     0,2,4,6,8,10    0               N/A
GPU2    NODE    NODE     X      PIX     SYS     SYS     SYS     SYS     0,2,4,6,8,10    0               N/A
GPU3    NODE    NODE    PIX      X      SYS     SYS     SYS     SYS     0,2,4,6,8,10    0               N/A
GPU4    SYS     SYS     SYS     SYS      X      PIX     NODE    NODE    1,3,5,7,9,11    1               N/A
GPU5    SYS     SYS     SYS     SYS     PIX      X      NODE    NODE    1,3,5,7,9,11    1               N/A
GPU6    SYS     SYS     SYS     SYS     NODE    NODE     X      PIX     1,3,5,7,9,11    1               N/A
GPU7    SYS     SYS     SYS     SYS     NODE    NODE    PIX      X      1,3,5,7,9,11    1               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

raywanb · 2024-08-12T05:42:50Z

Can you share the input that caused this to error?

cjfcsjt · 2024-08-12T07:05:16Z

Can you share the input that caused this to error?

@raywanb Here is my input:

sampling_params = SamplingParams(temperature=0.5, max_tokens=1, prompt_logprobs= 1, stop=["<|im_start|>", "<|im_end|>"])
messages = [{
        'role': 'user',
        'content': '(<image>./</image>)\n{question + "Yes"}'
}]
prompt = self.policy_tokenizer.apply_chat_template(messages,
                                                    tokenize=False,
                                                    add_generation_prompt=True)
input = {"prompt": prompt,
            "multi_modal_data": {
            "image": som_img  # Any PIL Image with shape 1280*720
        },
}
output = self.vlm_model.generate(input, sampling_params=sampling_params)

This would raise the issue:

assert len(next_token_ids) == len(query_indices)

Jeffwan · 2024-10-11T17:08:39Z

Seem this is multi_modal model? I do seem some related issues. Can someone confirm whether the multi_modal + prefix cache is supported or not?

github-actions · 2025-01-10T02:02:21Z

This issue has been automatically marked as stale because it has not had any activity within 90 days. It will be automatically closed if no further activity occurs within 30 days. Leave a comment if you feel this issue should remain open. Thank you!

tjohnson31415 · 2025-01-29T00:31:28Z

There seems to be a bug that makes prefix caching and prompt_logprobs incompatible.

I was able to reproduce the assert len(next_token_ids) == len(query_indices) AssertionError with meta-llama/Llama-3.1-8B-Instruct on vllm==0.7.0 by sending the request below twice:

curl http://localhost:8000/v1/completions -H "Content-Type: application/json" \
  -d '{"model": "model", "prompt": "1 2 3 4 5 6 7 8 9 10", "max_tokens": 1, "prompt_logprobs": 1}'

m-harmonic added the bug Something isn't working label Jul 31, 2024

zachzzc mentioned this issue Aug 2, 2024

[Bugfix] Fix block table for seqs that have prefix cache hits #7018

Merged

raywanb mentioned this issue Aug 5, 2024

[BugFix] Potential Fix to CUDA Illegal Memory Error #7142

Closed

github-actions bot added the stale label Jan 10, 2025

fxmarty-amd mentioned this issue Jan 21, 2025

Group reqs by context EleutherAI/lm-evaluation-harness#1425

Merged

github-actions bot added unstale and removed stale labels Jan 29, 2025

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Bug]: VLLM crashes when prefix caching is enabled #7003

[Bug]: VLLM crashes when prefix caching is enabled #7003

m-harmonic commented Jul 31, 2024 •

edited

Loading

tristan279 commented Aug 1, 2024 •

edited

Loading

cjfcsjt commented Aug 10, 2024 •

edited

Loading

raywanb commented Aug 12, 2024

cjfcsjt commented Aug 12, 2024 •

edited

Loading

Jeffwan commented Oct 11, 2024

github-actions bot commented Jan 10, 2025

tjohnson31415 commented Jan 29, 2025 •

edited

Loading

[Bug]: VLLM crashes when prefix caching is enabled #7003

[Bug]: VLLM crashes when prefix caching is enabled #7003

Comments

m-harmonic commented Jul 31, 2024 • edited Loading

Your current environment

🐛 Describe the bug

tristan279 commented Aug 1, 2024 • edited Loading

cjfcsjt commented Aug 10, 2024 • edited Loading

raywanb commented Aug 12, 2024

cjfcsjt commented Aug 12, 2024 • edited Loading

Jeffwan commented Oct 11, 2024

github-actions bot commented Jan 10, 2025

tjohnson31415 commented Jan 29, 2025 • edited Loading

m-harmonic commented Jul 31, 2024 •

edited

Loading

tristan279 commented Aug 1, 2024 •

edited

Loading

cjfcsjt commented Aug 10, 2024 •

edited

Loading

cjfcsjt commented Aug 12, 2024 •

edited

Loading

tjohnson31415 commented Jan 29, 2025 •

edited

Loading