[Bug]: Embedding model not working with tensor parallel

[Vincent-Li-9701]

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 20.04.6 LTS (x86_64)
GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0
Clang version: Could not collect
CMake version: version 3.29.3
Libc version: glibc-2.31

Python version: 3.9.6 (default, Aug 18 2021, 19:38:01)  [GCC 7.5.0] (64-bit runtime)
Python platform: Linux-5.10.192-183.736.amzn2.x86_64-x86_64-with-glibc2.31
Is CUDA available: True
CUDA runtime version: 12.4.131
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: 
GPU 0: NVIDIA A100-SXM4-40GB
GPU 1: NVIDIA A100-SXM4-40GB
GPU 2: NVIDIA A100-SXM4-40GB
GPU 3: NVIDIA A100-SXM4-40GB
GPU 4: NVIDIA A100-SXM4-40GB
GPU 5: NVIDIA A100-SXM4-40GB
GPU 6: NVIDIA A100-SXM4-40GB
GPU 7: NVIDIA A100-SXM4-40GB

Nvidia driver version: 535.129.03
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
Byte Order:                         Little Endian
Address sizes:                      46 bits physical, 48 bits virtual
CPU(s):                             96
On-line CPU(s) list:                0-95
Thread(s) per core:                 2
Core(s) per socket:                 24
Socket(s):                          2
NUMA node(s):                       2
Vendor ID:                          GenuineIntel
CPU family:                         6
Model:                              85
Model name:                         Intel(R) Xeon(R) Platinum 8275CL CPU @ 3.00GHz
Stepping:                           7
CPU MHz:                            3598.393
BogoMIPS:                           5999.99
Hypervisor vendor:                  KVM
Virtualization type:                full
L1d cache:                          1.5 MiB
L1i cache:                          1.5 MiB
L2 cache:                           48 MiB
L3 cache:                           71.5 MiB
NUMA node0 CPU(s):                  0-23,48-71
NUMA node1 CPU(s):                  24-47,72-95
Vulnerability Gather data sampling: Unknown: Dependent on hypervisor status
Vulnerability Itlb multihit:        KVM: Mitigation: VMX unsupported
Vulnerability L1tf:                 Mitigation; PTE Inversion
Vulnerability Mds:                  Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
Vulnerability Meltdown:             Mitigation; PTI
Vulnerability Mmio stale data:      Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
Vulnerability Retbleed:             Vulnerable
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass:    Vulnerable
Vulnerability Spectre v1:           Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:           Mitigation; Retpolines, STIBP disabled, RSB filling
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected
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 arch_perfmon rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves ida arat pku ospke

Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] nvidia-nccl-cu12==2.20.5
[pip3] torch==2.3.0+default
[pip3] triton==2.3.0
[pip3] vllm_nccl_cu12==2.18.1.0.4.0
[conda] numpy                     1.26.4                   pypi_0    pypi
[conda] nvidia-nccl-cu12          2.20.5                   pypi_0    pypi
[conda] torch                     2.3.0+default            pypi_0    pypi
[conda] triton                    2.3.0                    pypi_0    pypi
[conda] vllm-nccl-cu12            2.18.1.0.4.0             pypi_0    pypiROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: 0.4.2
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      NV12    NV12    NV12    NV12    NV12    NV12    NV12    0-23,48-71      0               N/A
GPU1    NV12     X      NV12    NV12    NV12    NV12    NV12    NV12    0-23,48-71      0               N/A
GPU2    NV12    NV12     X      NV12    NV12    NV12    NV12    NV12    0-23,48-71      0               N/A
GPU3    NV12    NV12    NV12     X      NV12    NV12    NV12    NV12    0-23,48-71      0               N/A
GPU4    NV12    NV12    NV12    NV12     X      NV12    NV12    NV12    24-47,72-95     1               N/A
GPU5    NV12    NV12    NV12    NV12    NV12     X      NV12    NV12    24-47,72-95     1               N/A
GPU6    NV12    NV12    NV12    NV12    NV12    NV12     X      NV12    24-47,72-95     1               N/A
GPU7    NV12    NV12    NV12    NV12    NV12    NV12    NV12     X      24-47,72-95     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

I installed via the source code and try to use the embedding model. Are the embedding models not intended to run with tensor parallel?
Slightly modify the example script used in here

# Sample prompts.
prompts = [
    "Hello, my name is",
    "The president of the United States is",
    "The capital of France is",
    "The future of AI is",
]

# Create an LLM.
model = LLM(model="intfloat/e5-mistral-7b-instruct", enforce_eager=True, tensor_parallel_size=8)
# Generate embedding. The output is a list of EmbeddingRequestOutputs.
outputs = model.encode(prompts)
# Print the outputs.
for output in outputs:
    print(output.outputs.embedding)  # list of 4096 floats

[rank0]:   File "/mnt/task_runtime/paraphrase.py", line 14, in <module>
[rank0]:     outputs = model.encode(prompts)
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/entrypoints/llm.py", line 238, in encode
[rank0]:     return self._run_engine(use_tqdm)
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/entrypoints/llm.py", line 341, in _run_engine
[rank0]:     step_outputs = self.llm_engine.step()
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/engine/llm_engine.py", line 681, in step
[rank0]:     output = self.model_executor.execute_model(
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/executor/ray_gpu_executor.py", line 177, in execute_model
[rank0]:     all_outputs = self._run_workers(
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/executor/ray_gpu_executor.py", line 259, in _run_workers
[rank0]:     ray_worker_outputs = ray.get(ray_worker_outputs)
[rank0]:   File "/miniconda/lib/python3.9/site-packages/ray/_private/auto_init_hook.py", line 21, in auto_init_wrapper
[rank0]:     return fn(*args, **kwargs)
[rank0]:   File "/miniconda/lib/python3.9/site-packages/ray/_private/client_mode_hook.py", line 103, in wrapper
[rank0]:     return func(*args, **kwargs)
[rank0]:   File "/miniconda/lib/python3.9/site-packages/ray/_private/worker.py", line 2623, in get
[rank0]:     values, debugger_breakpoint = worker.get_objects(object_refs, timeout=timeout)
[rank0]:   File "/miniconda/lib/python3.9/site-packages/ray/_private/worker.py", line 861, in get_objects
[rank0]:     raise value.as_instanceof_cause()
[rank0]: ray.exceptions.RayTaskError(TypeError): ray::RayWorkerWrapper.execute_method() (pid=912273, ip=240.57.235.115, actor_id=4d3c7a4a4aba230a2a32f79b01000000, repr=<vllm.executor.ray_utils.RayWorkerWrapper object at 0x7fb10f0fdcd0>)
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/worker/worker_base.py", line 146, in execute_method
[rank0]:     raise e
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/worker/worker_base.py", line 137, in execute_method
[rank0]:     return executor(*args, **kwargs)
[rank0]:   File "/miniconda/lib/python3.9/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
[rank0]:     return func(*args, **kwargs)
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/worker/worker.py", line 278, in execute_model
[rank0]:     output = self.model_runner.execute_model(seq_group_metadata_list,
[rank0]:   File "/miniconda/lib/python3.9/site-packages/torch/utils/_contextlib.py", line 115, in decorate_context
[rank0]:     return func(*args, **kwargs)
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/worker/embedding_model_runner.py", line 82, in execute_model
[rank0]:     return self.model.pooler(hidden_states=hidden_states,
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/model_executor/models/llama_embedding.py", line 49, in pooler
[rank0]:     return self._pooler(hidden_states, pooling_metadata)
[rank0]:   File "/miniconda/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1532, in _wrapped_call_impl
[rank0]:     return self._call_impl(*args, **kwargs)
[rank0]:   File "/miniconda/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1541, in _call_impl
[rank0]:     return forward_call(*args, **kwargs)
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/model_executor/layers/pooler.py", line 40, in forward
[rank0]:     prompt_lens = PoolingTensors.from_pooling_metadata(
[rank0]:   File "/mnt/task_runtime/local_packages/vllm/vllm/model_executor/pooling_metadata.py", line 61, in from_pooling_metadata
[rank0]:     prompt_lens_t = torch.tensor(
[rank0]: TypeError: an integer is required (got type NoneType)```

[Vincent-Li-9701]

Could the author help clarify? Really appreciate that. @CatherineSue