###### tags: `lab` `sys`
# Teslaマシン環境構築
Boot meneuはF11
## Teslaへのインストール
一般的なGTX系のドライバと同じ方法でインストールできる。注意したいのはNvidia-Driverを誤ったものを入れるとログインループになります。
### 手順
1. Ubuntuのセットアップ
- これはGTX系の資料を参考に
2. X serverを殺す
- `init 3`で殺すと確実
3. `Tesla Driver for Linux x64`のインストール
- [Tesla Driver for Linux](https://www.nvidia.com/Download/index.aspx?lang=en-us)
- ==CUDAとnvidia-Driverの互換性==
- :::info
一般的にnvidia-driver384はCUDA7, 8, 9番台をサポートしている。今回のTeslaのdriverバージョンは最も古いものでv384である。つまり、公式ではTesla V100はCUDA9しかサポートしていないが、nvidia-driverに互換性のあるCUDAなら7でも8でも実行可能である可能性が高い
:::
## 注意事項
Teslaには専用のドライバがある。
```
chmod +x NVIDIA-Linux-x86_64-384.183.run
./NVIDIA-Linux-x86_64-384.183.run --no-opengl-files --no-libglx-indirect --dkms
```
そう、384なんである。なので
Driverはこのスクリプト経由
Cudaの本体はtoolkitから焼くことで、cuda8.0環境が使えるのではないかと考えている
https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#package-manager-metas
| Meta Package | Purpose |
| ----------------------- | ------------------------------------------------------------ |
| cuda | Installs all CUDA Toolkit and Driver packages. Handles upgrading to the next version of the cuda package when it's released. |
| cuda-10-2 | Installs all CUDA Toolkit and Driver packages. Remains at version 10.2 until an additional version of CUDA is installed. |
| cuda-toolkit-10-2 | Installs all CUDA Toolkit packages required to develop CUDA applications. ==Does not include the driver.== |
| cuda-tools-10-2 | Installs all CUDA command line and visual tools. |
| cuda-runtime-10-2 | Installs all CUDA Toolkit packages required to run CUDA applications, as well as the Driver packages. |
| cuda-compiler-10-2 | Installs all CUDA compiler packages. |
| cuda-libraries-10-2 | Installs all runtime CUDA Library packages. |
| cuda-libraries-dev-10-2 | Installs all development CUDA Library packages. |
| cuda-drivers | Installs all Driver packages. Handles upgrading to the next version of the Driver packages when they're released. |
## 参考サイト
[CUDA Toolkit／GPUカードドライバー導入手順](https://help.sakura.ad.jp/115000122721/#03)
## 選択肢
- > The distribution-provided pre-install script failed! Are you sure you want to continue?
- Continue installation ☑
- ②
- Abort installation
- instalalation is falid
- > ②Would you like to register the kernel module sources with DKMS? This will allow DKMS to automatically build a new module, if you
install a different kernel later.
- yse ☑
- no
``
WARNING: Unable to find a suitable destination to install 32-bit compatibility libraries. Your system may not be set up for
32-bit compatibility. 32-bit compatibility files will not be installed; if you wish to install them, re-run the
installation and set a valid directory with the --compat32-libdir option.
```
- OK
## 地獄のChainer2
実際のところこれで行けるかわからない...
### cuDNNのインストール
```
tar xzvf cudnn-8.0-linux-x64-v5.1.tgz
sudo cp -a cuda/lib64/* /usr/local/cuda-8.0/lib64/
sudo cp -a cuda/include/* /usr/local/cuda-8.0/include/
sudo ldconfig
```
https://qiita.com/JeJeNeNo/items/05e148a325192004e2cd
- chainer v1.18.0
- cudnn 5.1 インストール可能だが動作しない
- cudnn 6.0 インストール失敗
- chainer v1.24.0
- cudnn 6.0 インストール可能
```
cupy-cuda80 6.3.0
ito@FRONTIER:~$ pip uninstall cupy-cuda80
```
OK?
https://stackoverflow.com/questions/51283708/python-pip-package-requestsdependencywarning-when-installing-elastic-search-cura
これと同じエラーが出たので同じ対処法をした
成功
とりあえず、これでretry ファイル回してみるわ。
```
1epoch: 0%| | 0/11464 [00:00<?, ?it/s]Traceback (most recent call last):
File "Main.py", line 50, in <module>
TrainNetwork.TrainCNNModel(dataidx["train"], i)
File "/home/ito.t/work/naka2018_31/TrainNetwork.py", line 84, in TrainCNNModel
network = DNN.TrainCNN(X,T,epoch_cnt=60, fold_cnt=fold_cnt)
File "/home/ito.t/work/naka2018_31/DNN.py", line 514, in TrainCNN
opt.update(model,x,t)
File "/home/ito.t/.local/lib/python2.7/site-packages/chainer/optimizer.py", line 411, in update
loss = lossfun(*args, **kwds)
File "/home/ito.t/.local/lib/python2.7/site-packages/chainer/links/model/classifier.py", line 67, in __call__
self.y = self.predictor(*x)
File "/home/ito.t/work/naka2018_31/DNN.py", line 216, in __call__
x = getattr(self, name)(x, self.tarin)
File "/home/ito.t/work/naka2018_31/DNN.py", line 113, in __call__
x = getattr(self, name)(x, train)
File "/home/ito.t/work/naka2018_31/DNN.py", line 90, in __call__
x = F.concat((x, tmp))
File "/home/ito.t/.local/lib/python2.7/site-packages/chainer/functions/array/concat.py", line 87, in concat
return Concat(axis=axis)(*xs)
File "/home/ito.t/.local/lib/python2.7/site-packages/chainer/function.py", line 199, in __call__
outputs = self.forward(in_data)
File "/home/ito.t/.local/lib/python2.7/site-packages/chainer/functions/array/concat.py", line 43, in forward
return xp.concatenate(xs, axis=self.axis),
File "/home/ito.t/.local/lib/python2.7/site-packages/cupy/manipulation/join.py", line 49, in concatenate
return core.concatenate_method(tup, axis)
File "cupy/core/core.pyx", line 2072, in cupy.core.core.concatenate_method (cupy/core/core.cpp:63848)
File "cupy/core/core.pyx", line 2115, in cupy.core.core.concatenate_method (cupy/core/core.cpp:63748)
File "cupy/core/core.pyx", line 2167, in cupy.core.core.concatenate (cupy/core/core.cpp:64932)
File "cupy/core/core.pyx", line 1298, in cupy.core.core.ndarray.__setitem__ (cupy/core/core.cpp:27914)
File "cupy/core/core.pyx", line 2734, in cupy.core.core._scatter_op (cupy/core/core.cpp:71428)
File "cupy/core/core.pyx", line 1541, in cupy.core.core.elementwise_copy (cupy/core/core.cpp:57352)
File "cupy/core/elementwise.pxi", line 776, in cupy.core.core.ufunc.__call__ (cupy/core/core.cpp:47765)
File "cupy/util.pyx", line 39, in cupy.util.memoize.decorator.ret (cupy/util.cpp:1480)
File "cupy/core/elementwise.pxi", line 584, in cupy.core.core._get_ufunc_kernel (cupy/core/core.cpp:44235)
File "cupy/core/elementwise.pxi", line 32, in cupy.core.core._get_simple_elementwise_kernel (cupy/core/core.cpp:33912)
File "cupy/core/carray.pxi", line 86, in cupy.core.core.compile_with_cache (cupy/core/core.cpp:33567)
File "/home/ito.t/.local/lib/python2.7/site-packages/cupy/cuda/compiler.py", line 156, in compile_with_cache
cubin = nvcc(source, options, arch)
File "/home/ito.t/.local/lib/python2.7/site-packages/cupy/cuda/compiler.py", line 78, in nvcc
_run_nvcc(cmd, root_dir)
File "/home/ito.t/.local/lib/python2.7/site-packages/cupy/cuda/compiler.py", line 56, in _run_nvcc
raise RuntimeError(msg)
RuntimeError: `nvcc` command returns non-zero exit status.
command: ['nvcc', '--cubin', '-arch', 'sm_70', '/tmp/tmpCODN17/kern.cu']
return-code: 1
stdout/stderr:
nvcc fatal : Value 'sm_70' is not defined for option 'gpu-architecture'
```
Sign in with Wallet
Connect another wallet