Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Not running Apple M1: third_party/cuda/bin/ptxas --version died with <Signals.SIGTRAP: 5>. #30

Closed
AIWintermuteAI opened this issue Oct 23, 2023 · 1 comment

Comments

@AIWintermuteAI
Copy link

It looks like the default version (and cpu too) do not run on M1. I tried specifying - EXTRA_LAUNCH_ARGS="--listen --verbose --loader llama.cpp", that however did not have any effect.
It would be nice to have it fixed, since llama.cpp actually runs pretty well with smaller models on M1.
Any pointers?

[+] Building 0.0s (0/0)                                                                                         
[+] Running 2/0
 ✔ Container text-generation-webui                                                                                                                                             Recreated0.1s 
 ! text-generation-webui-docker The requested image's platform (linux/amd64) does not match the detected host platform (linux/arm64/v8) and no specific platform was requested 0.0s 
Attaching to text-generation-webui
text-generation-webui  | === Running text-generation-webui variant: 'DEFAULT' snapshot-2023-10-15 ===
text-generation-webui  | === (This version is 11 commits behind origin main) ===
text-generation-webui  | === Image build date: 2023-10-18 11:30:52 ===
text-generation-webui  | 2023-10-23 18:27:36 WARNING:
text-generation-webui  | You are potentially exposing the web UI to the entire internet without any access password.
text-generation-webui  | You can create one with the "--gradio-auth" flag like this:
text-generation-webui  | 
text-generation-webui  | --gradio-auth username:password
text-generation-webui  | 
text-generation-webui  | Make sure to replace username:password with your own.
text-generation-webui  | /venv/lib/python3.10/site-packages/bitsandbytes/cextension.py:34: UserWarning: The installed version of bitsandbytes was compiled without GPU support. 8-bit optimizers, 8-bit multiplication, and GPU quantization are unavailable.
text-generation-webui  |   warn("The installed version of bitsandbytes was compiled without GPU support. "
text-generation-webui  | /venv/lib/python3.10/site-packages/bitsandbytes/libbitsandbytes_cpu.so: undefined symbol: cadam32bit_grad_fp32
text-generation-webui  | Traceback (most recent call last):
text-generation-webui  |   File "/app/server.py", line 31, in <module>
text-generation-webui  |     from modules import (
text-generation-webui  |   File "/app/modules/training.py", line 21, in <module>
text-generation-webui  |     from peft import (
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/peft/__init__.py", line 22, in <module>
text-generation-webui  |     from .auto import (
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/peft/auto.py", line 31, in <module>
text-generation-webui  |     from .mapping import MODEL_TYPE_TO_PEFT_MODEL_MAPPING
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/peft/mapping.py", line 23, in <module>
text-generation-webui  |     from .peft_model import (
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/peft/peft_model.py", line 38, in <module>
text-generation-webui  |     from .tuners import (
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/peft/tuners/__init__.py", line 21, in <module>
text-generation-webui  |     from .lora import LoraConfig, LoraModel
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/peft/tuners/lora.py", line 45, in <module>
text-generation-webui  |     import bitsandbytes as bnb
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/bitsandbytes/__init__.py", line 16, in <module>
text-generation-webui  |     from .nn import modules
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/bitsandbytes/nn/__init__.py", line 6, in <module>
text-generation-webui  |     from .triton_based_modules import SwitchBackLinear, SwitchBackLinearGlobal, SwitchBackLinearVectorwise, StandardLinear
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/bitsandbytes/nn/triton_based_modules.py", line 8, in <module>
text-generation-webui  |     from bitsandbytes.triton.dequantize_rowwise import dequantize_rowwise
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/bitsandbytes/triton/dequantize_rowwise.py", line 10, in <module>
text-generation-webui  |     import triton
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/__init__.py", line 20, in <module>
text-generation-webui  |     from .compiler import compile, CompilationError
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/compiler/__init__.py", line 1, in <module>
text-generation-webui  |     from .compiler import CompiledKernel, compile, instance_descriptor
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/compiler/compiler.py", line 27, in <module>
text-generation-webui  |     from .code_generator import ast_to_ttir
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/compiler/code_generator.py", line 8, in <module>
text-generation-webui  |     from .. import language
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/language/__init__.py", line 4, in <module>
text-generation-webui  |     from . import math
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/language/math.py", line 4, in <module>
text-generation-webui  |     from . import core
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/language/core.py", line 1376, in <module>
text-generation-webui  |     def minimum(x, y):
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/runtime/jit.py", line 542, in jit
text-generation-webui  |     return decorator(fn)
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/runtime/jit.py", line 534, in decorator
text-generation-webui  |     return JITFunction(
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/runtime/jit.py", line 433, in __init__
text-generation-webui  |     self.run = self._make_launcher()
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/runtime/jit.py", line 388, in _make_launcher
text-generation-webui  |     scope = {"version_key": version_key(),
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/runtime/jit.py", line 120, in version_key
text-generation-webui  |     ptxas = path_to_ptxas()[0]
text-generation-webui  |   File "/venv/lib/python3.10/site-packages/triton/common/backend.py", line 114, in path_to_ptxas
text-generation-webui  |     result = subprocess.check_output([ptxas_bin, "--version"], stderr=subprocess.STDOUT)
text-generation-webui  |   File "/usr/lib/python3.10/subprocess.py", line 421, in check_output
text-generation-webui  |     return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
text-generation-webui  |   File "/usr/lib/python3.10/subprocess.py", line 526, in run
text-generation-webui  |     raise CalledProcessError(retcode, process.args,
text-generation-webui  | subprocess.CalledProcessError: Command '['/venv/lib/python3.10/site-packages/triton/common/../third_party/cuda/bin/ptxas', '--version']' died with <Signals.SIGTRAP: 5>.
text-generation-webui exited with code 1
@AIWintermuteAI AIWintermuteAI changed the title Not running M1: third_party/cuda/bin/ptxas --version died with <Signals.SIGTRAP: 5>. Not running Apple M1: third_party/cuda/bin/ptxas --version died with <Signals.SIGTRAP: 5>. Oct 23, 2023
@Atinoda
Copy link
Owner

Atinoda commented Oct 23, 2023

I cannot properly support Mac OS because I don't have any Apple Silicon at the moment. I absolutely intend to support it in future though, because the unified memory architecture is monster for LLM inference work!

Can you please check #22 and also leave a comment there if you get things working? I'm going to close this issue and direct the Mac stuff to that thread - hopefully it'll be useful to you and other people as a central reference for now.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants