search

PygmalionAI / aphrodite-engine

Large-scale LLM inference engine
https://aphrodite.pygmalion.chat
GNU Affero General Public License v3.0
952 stars 104 forks source link

[Bug]: FP Quantizer Error when loading using --quantization deepspeedfp, Triton version related #620

Open Inktomi93 opened 2 weeks ago

Inktomi93 commented 2 weeks ago

Your current environment

Collecting environment information...
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 24.04 LTS (x86_64)
GCC version: (Ubuntu 13.2.0-23ubuntu4) 13.2.0
Clang version: Could not collect 
CMake version: version 3.30.2
Libc version: glibc-2.39
Python version: 3.11.9 (main, Apr 19 2024, 16:48:06) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-6.8.0-41-generic-x86_64-with-glibc2.39
Is CUDA available: True
CUDA runtime version: 12.6.20
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: 
GPU 0: NVIDIA RTX A6000
GPU 1: NVIDIA RTX A6000

Nvidia driver version: 560.35.03
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
Address sizes:                        46 bits physical, 48 bits virtual
Byte Order:                           Little Endian
CPU(s):                               24
On-line CPU(s) list:                  0-23
Vendor ID:                            GenuineIntel
Model name:                           13th Gen Intel(R) Core(TM) i7-13700K
CPU family:                           6
Model:                                183
Thread(s) per core:                   2
Core(s) per socket:                   16
Socket(s):                            1
Stepping:                             1
CPU(s) scaling MHz:                   27%
CPU max MHz:                          5400.0000
CPU min MHz:                          800.0000
BogoMIPS:                             6835.20
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 tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushopt clwb intel_pt sha_ni xsaveopt xsavec xgetbv1 xsaves split_lock_detect user_shstk avx_vnni dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req hfi vnmi umip pku ospke waitpkg gfni vaes vpclmulqdq tme rdpid movdiri movdir64b fsrm md_clear serialize pconfig arch_lbr ibt flush_l1d arch_capabilities
Virtualization:                       VT-x
L1d cache:                            640 KiB (16 instances)
L1i cache:                            768 KiB (16 instances)
L2 cache:                             24 MiB (10 instances)
L3 cache:                             30 MiB (1 instance)
NUMA node(s):                         1
NUMA node0 CPU(s):                    0-23
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 Reg file data sampling: Mitigation; Clear Register File
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 conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected
Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] torch==2.4.0
[pip3] torchvision==0.19.0
[pip3] triton==3.0.0
[conda] numpy                     1.26.4                   pypi_0    pypi
[conda] torch                     2.4.0                    pypi_0    pypi
[conda] torchvision               0.19.0                   pypi_0    pypi
[conda] triton                    3.0.0                    pypi_0    pypiROCM Version: Could not collect 
Aphrodite Version: 0.6.0
Aphrodite Build Flags:
CUDA Archs: Not Set; ROCm: Disabled

🐛 Describe the bug

Traceback (most recent call last):
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/multiprocessing/process.py", line 314, in _bootstrap
    self.run()
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/multiprocessing/process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/endpoints/openai/rpc/server.py", line 204, in run_rpc_server
    server = AsyncEngineRPCServer(async_engine_args, port)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/endpoints/openai/rpc/server.py", line 23, in __init__
    self.engine = AsyncAphrodite.from_engine_args(async_engine_args)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/engine/async_aphrodite.py", line 470, in from_engine_args
    engine = cls(
             ^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/engine/async_aphrodite.py", line 379, in __init__
    self.engine = self._init_engine(*args, **kwargs)
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/engine/async_aphrodite.py", line 550, in _init_engine
    return engine_class(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/engine/aphrodite_engine.py", line 243, in __init__
    self.model_executor = executor_class(
                          ^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/executor/multiproc_gpu_executor.py", line 212, in __init__
    super().__init__(*args, **kwargs)
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/executor/distributed_gpu_executor.py", line 24, in __init__
    super().__init__(*args, **kwargs)
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/executor/executor_base.py", line 47, in __init__
    self._init_executor()
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/executor/multiproc_gpu_executor.py", line 137, in _init_executor
    self._run_workers("load_model",
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/executor/multiproc_gpu_executor.py", line 189, in _run_workers
    driver_worker_output = driver_worker_method(*args, **kwargs)
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/task_handler/worker.py", line 146, in load_model
    self.model_runner.load_model()
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/task_handler/model_runner.py", line 730, in load_model
    self.model = get_model(model_config=self.model_config,
                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/model_loader/__init__.py", line 21, in get_model
    return loader.load_model(model_config=model_config,
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/model_loader/loader.py", line 323, in load_model
    model = _initialize_model(model_config, self.load_config,
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/model_loader/loader.py", line 153, in _initialize_model
    return model_class(config=model_config.hf_config,
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/models/commandr.py", line 342, in __init__
    self.model = CohereModel(config,
                 ^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/models/commandr.py", line 277, in __init__
    self.layers = nn.ModuleList([
                                ^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/models/commandr.py", line 278, in <listcomp>
    CohereDecoderLayer(config, cache_config, quant_config=quant_config)
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/models/commandr.py", line 228, in __init__
    self.self_attn = CohereAttention(config,
                     ^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/models/commandr.py", line 157, in __init__
    self.qkv_proj = QKVParallelLinear(
                    ^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/layers/linear.py", line 634, in __init__
    super().__init__(input_size=input_size,
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/modeling/layers/linear.py", line 302, in __init__
    self.quant_method.create_weights(
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/quantization/deepspeedfp.py", line 101, in create_weights
    weight = DeepSpeedFPParameter(
             ^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/aphrodite/quantization/deepspeedfp.py", line 161, in __new__
    self.fp_quantizer = FP_Quantize(group_size=quant_config.group_size)
                        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/deepspeed/ops/fp_quantizer/quantize.py", line 49, in __init__
    fp_quant_module = FPQuantizerBuilder().load()
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/deepspeed/ops/op_builder/builder.py", line 530, in load
    return self.jit_load(verbose)
           ^^^^^^^^^^^^^^^^^^^^^^
  File "/home/inktomi/miniconda3/envs/aphrodite/lib/python3.11/site-packages/deepspeed/ops/op_builder/builder.py", line 534, in jit_load
    raise RuntimeError(
RuntimeError: Unable to JIT load the fp_quantizer op due to it not being compatible due to hardware/software issue. FP Quantizer is using an untested triton version (3.0.0), only 2.3.0 and 2.3.1 are known to be compatible with these kernels

Did a fresh install in a new conda environment and installed Deepspeed using the documented instruction: pip install deepspeed>=0.14.2 Behavior persisted between multiple attempts and even tried building from source with the same result.

AlpinDale commented 13 hours ago

As of #755, it's recommended to use -q fpX instead, where X is a number between 2 and 7.