[Bug]: CPU silently doesn't support prompt adapter

[wallashss]

Your current environment

The output of `python collect_env.py`

Collecting environment information...
INFO 09-13 19:13:45 importing.py:10] Triton not installed; certain GPU-related functions will not be available.
PyTorch version: 2.4.0+cpu
Is debug build: False
CUDA used to build PyTorch: Could not collect
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.4 LTS (x86_64)
GCC version: (Ubuntu 12.3.0-1ubuntu1~22.04) 12.3.0
Clang version: Could not collect
CMake version: version 3.30.2
Libc version: glibc-2.35

Python version: 3.10.12 (main, Jul 29 2024, 16:56:48) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-4.18.0-372.46.1.el8_6.x86_64-x86_64-with-glibc2.35
Is CUDA available: False
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: 
GPU 0: NVIDIA A100-SXM4-80GB
  MIG 3g.40gb     Device  0:

Nvidia driver version: 535.104.05
cuDNN version: Could not collect
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, 57 bits virtual
Byte Order:                      Little Endian
CPU(s):                          80
On-line CPU(s) list:             0-79
Vendor ID:                       GenuineIntel
Model name:                      Intel Xeon Processor (Icelake)
CPU family:                      6
Model:                           134
Thread(s) per core:              2
Core(s) per socket:              20
Socket(s):                       2
Stepping:                        0
BogoMIPS:                        5600.03
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 cpuid tsc_known_freq pni pclmulqdq vmx ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch cpuid_fault invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves wbnoinvd arat avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid fsrm md_clear arch_capabilities
Virtualization:                  VT-x
Hypervisor vendor:               KVM
Virtualization type:             full
L1d cache:                       2.5 MiB (80 instances)
L1i cache:                       2.5 MiB (80 instances)
L2 cache:                        160 MiB (40 instances)
L3 cache:                        32 MiB (2 instances)
NUMA node(s):                    2
NUMA node0 CPU(s):               0-39
NUMA node1 CPU(s):               40-79
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Mmio stale data:   Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
Vulnerability Retbleed:          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 IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected

Versions of relevant libraries:
[pip3] intel_extension_for_pytorch==2.4.0+gitfbaa4bc
[pip3] numpy==1.26.4
[pip3] pyzmq==26.2.0
[pip3] torch==2.4.0+cpu
[pip3] torchvision==0.19.0+cpu
[pip3] transformers==4.44.2
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.6.0@36bf8150cc3a048d69d9d2196128462014b9599d
vLLM Build Flags:
CUDA Archs: Not Set; ROCm: Disabled; Neuron: Disabled
GPU Topology:
�[4mGPU0	NIC0	CPU Affinity	NUMA Affinity	GPU NUMA ID�[0m
GPU0	 X 	SYS	40-79	1		N/A
NIC0	SYS	 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

Model Input Dumps

No response

🐛 Describe the bug

This sample script does a simple test using prompt adapter:

from vllm import LLM, SamplingParams
from vllm.prompt_adapter.request import PromptAdapterRequest

MODEL_PATH = "bigscience/bloomz-560m"
PA_PATH = 'stevhliu/bloomz-560m_PROMPT_TUNING_CAUSAL_LM'

llm = LLM(MODEL_PATH,
                   enforce_eager=True,
                   enable_prompt_adapter=True,
                   max_prompt_adapter_token=8)


sampling_params = SamplingParams(temperature=0.0,
                                          max_tokens=3,
                                          stop_token_ids=[3])

pa_name = "twitter_pa"
pa_id = 1

prompts = [
    "Tweet text : @nationalgridus I have no water and the bill is \
    current and paid. Can you do something about this? Label : ",
    "Tweet text : @nationalgridus Looks good thanks! Label : "
]
outputs = llm.generate(prompts, 
                       sampling_params=sampling_params,
                       prompt_adapter_request=PromptAdapterRequest(
                               pa_name, pa_id, PA_PATH, 8) if pa_id else None)

for o in outputs:
    print('_________')
    print('### Text')
    print('_________')
    for o2 in o.outputs:
        print(o2.text)

For standard installation using NVIDIA gpu I get following output:

_________
### Text
_________
complaint
_________
### Text
_________
no complaint

Running using CPU backend I get:

_________
### Text
_________
 _
_________
### Text
_________
 🌍

Although vLLM behavior can differ on different implementation due to numeric instability I checked the hugging face cache dirs and vllm didn't even download the adapter model (did this in a clean container). Also checking the code cpu_model_runner.py it receives the prompt_adapter_config but it does nothing with that. Therefore, I am pretty sure the feature is not supported on CPU backend.

Expectation

My proposal is to add a verification for cpu backend when users tried to use prompt adapter and raise an exception or log an warning to inform the user that this feature is not supported.

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.