Closed JoshWolper closed 3 years ago
Oh and I should mention I upgraded through using: python3 -m pip install taichi --upgrade
nvidia-smi not available: Command '['nvidia-smi']' returned non-zero exit status 255.
It's a bit weird to see this error. I think it indicates an NVidia driver issue. What's your CUDA version?
Hmm weird, the last version of taichi worked fine with this CUDA driver. Here's the output of nvcc --version:
josh@jg9:~$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Fri_Feb__8_19:08:17_PST_2019
Cuda compilation tools, release 10.1, V10.1.105
RuntimeError: [cuda_driver.h:operator()@80] CUDA Error CUDA_ERROR_COMPAT_NOT_SUPPORTED_ON_DEVICE: forward compatibility was attempted on non supported HW while calling init (cuInit)
Maybe https://github.com/pytorch/pytorch/issues/40671#issuecomment-650712854?
Oh wow yep, a reboot fixed the issue!! Thanks for this, I'll reboot after updating taichi from now on!!
Describe the bug When I try to run code on GPU it no longer works. These files worked perfectly on GPU before the upgrade.
To Reproduce This is the same code file from my other issue about the indexing, now the error that comes up is related to GPU though...
Log/Screenshots Please post the full log of the program (instead of just a few lines around the error message, unless the log is > 1000 lines). This will help us diagnose what's happening. For example:
Additional comments If possible, please also consider attaching the output of command
ti diagnose
. This produces the detailed environment information and hopefully helps us diagnose faster.Here's the output of ti diagonose (I notice the same CUDA error in there but nothing has changed on my system except for taichi):
`[Taichi] mode=release [Taichi] preparing sandbox at /tmp/taichi-h8f13ll9 [Taichi] version 0.7.12, llvm 10.0.0, commit 8d452801, linux, python 3.6.9
Taichi Programming Language
Docs: https://taichi.rtfd.io/en/stable GitHub: https://github.com/taichi-dev/taichi Forum: https://forum.taichi.graphics
Taichi system diagnose:
python: 3.6.9 (default, Oct 8 2020, 12:12:24) [GCC 8.4.0] system: linux executable: /usr/bin/python3 platform: Linux-5.4.0-53-generic-x86_64-with-Ubuntu-18.04-bionic architecture: 64bit ELF uname: uname_result(system='Linux', node='jg9', release='5.4.0-53-generic', version='#59~18.04.1-Ubuntu SMP Wed Oct 21 12:14:56 UTC 2020', machine='x86_64', processor='x86_64') locale: en_US.UTF-8 PATH: /usr/local/cuda-10.1/bin/:/home/josh/.local/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/usr/local/games:/snap/bin PYTHONPATH: ['/home/josh/.local/bin', '/home/josh/Desktop/FastIPC', '/home/josh/Desktop/FastIPC/projects/brittle/referenceCode', '/usr/lib/python36.zip', '/usr/lib/python3.6', '/usr/lib/python3.6/lib-dynload', '/home/josh/.local/lib/python3.6/site-packages', '/usr/local/lib/python3.6/dist-packages', '/usr/lib/python3/dist-packages', '/home/josh/.local/lib/python3.6/site-packages/taichi/core/../lib']
No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 18.04.4 LTS Release: 18.04 Codename: bionic
TAICHI_REPO_DIR=
import: <module 'taichi' from '/home/josh/.local/lib/python3.6/site-packages/taichi/init.py'>
cc: True cpu: True metal: False opengl: False cuda: True
glewinfo
not available: [Errno 2] No such file or directory: 'glewinfo': 'glewinfo'nvidia-smi
not available: Command '['nvidia-smi']' returned non-zero exit status 255. [Taichi] mode=release [Taichi] preparing sandbox at /tmp/taichi-lrr8036_ [Taichi] version 0.7.12, llvm 10.0.0, commit 8d452801, linux, python 3.6.9[Taichi] mode=release [Taichi] preparing sandbox at /tmp/taichi-ojdqcu8n [Taichi] version 0.7.12, llvm 10.0.0, commit 8d452801, linux, python 3.6.9 [Taichi] Starting on arch=x64
[W 01/12/21 12:03:20.175] [init.py:adaptive_arch_select@574] Arch=[<Arch.opengl: 6>] is not supported, falling back to CPU [Taichi] mode=release [Taichi] preparing sandbox at /tmp/taichi-dri9tsr9 [Taichi] version 0.7.12, llvm 10.0.0, commit 8d452801, linux, python 3.6.9 [Taichi] Starting on arch=x64
Traceback (most recent call last): File "", line 1, in
File "/home/josh/.local/lib/python3.6/site-packages/taichi/lang/init.py", line 216, in init
ti.get_runtime().create_program()
File "/home/josh/.local/lib/python3.6/site-packages/taichi/lang/impl.py", line 215, in create_program
self.prog = taichi_lang_core.Program()
RuntimeError: [cuda_driver.h:operator()@80] CUDA Error CUDA_ERROR_COMPAT_NOT_SUPPORTED_ON_DEVICE: forward compatibility was attempted on non supported HW while calling init (cuInit)
Taichi CUDA test failed: Command '['/usr/bin/python3', '-c', 'import taichi as ti; ti.init(arch=ti.cuda)']' returned non-zero exit status 1.
[Taichi] mode=release
[Taichi] preparing sandbox at /tmp/taichi-tdc2kanu
[Taichi] version 0.7.12, llvm 10.0.0, commit 8d452801, linux, python 3.6.9
Taichi Programming Language
Docs: https://taichi.rtfd.io/en/stable GitHub: https://github.com/taichi-dev/taichi Forum: https://forum.taichi.graphics
Running example minimal ... [Taichi] Starting on arch=x64 [Taichi] materializing...
Consider attaching this log when maintainers ask about system information.