CentOS 7.x 安装CUDA
<ol><li>
<p>当前环境</p>
<pre><code class="language-python"># cat /etc/redhat-release
CentOS Linux release 7.9.2009 (Core
</code></pre>
</li>
<li>
<p>首先需要确认自己的GPU是否支持CUDA。查看支持CUDA的GPU列表</p>
<pre><code class="language-python"># lspci | grep -i nvidia
af:00.0 3D controller: NVIDIA Corporation GA100 (rev a1)
</code></pre>
</li>
<li>
<p>验证是否有GCC编译环境。</p>
<pre><code class="language-python"># gcc -v
</code></pre>
</li>
<li>
<p>验证是否安装了内核头文件和开发包。</p>
<pre><code class="language-python"># yum install kernel-devel-$(uname -r) kernel-headers-$(uname -r)
</code></pre>
</li>
<li>
<p>查看本机是否有装nouveau驱动，如果有就禁用掉（因为它和NVIDIA官方的驱动有冲突，所以安装NVIDIA驱动和CUDA前应该禁用它）。如果没有就跳过本步骤。</p>
<pre><code class="language-python"># lsmod | grep nouveau # 如果有结果就执行下面操作禁用掉，没有就忽略。
# vim /etc/modprobe.d/blacklist-nouveau.conf
blacklist nouveau
options nouveau modeset=0
# mv /boot/initramfs-$(uname -r).img{,.bak}
# dracut /boot/initramfs-$(uname -r).img $(uname -r)
## 重启系统并验证。
# shutdown -r now
# lsmod | grep nouveau
</code></pre>
</li>
<li>
<p>下载驱动程序，CUDA历史包下载地址。</p>
<pre><code class="language-python"># wget https://developer.download.nvidia.com/compute/cuda/11.2.0/local_installers/cuda-repo-rhel7-11-2-local-11.2.0_460.27.04-1.x86_64.rpm
# yum -y localinstall cuda-repo-rhel7-11-2-local-11.2.0_460.27.04-1.x86_64.rpm
# yum -y install nvidia-driver-latest-dkms cuda cuda-drivers
</code></pre>
</li>
<li>
<p>验证。</p>
<pre><code class="language-python"># cd /usr/local/cuda/samples/1_Utilities/deviceQuery
# make
# ./deviceQuery
</code></pre>
</li>
</ol>
<p> <br>
</p>
<p><font color="blue" size="4"><em><strong>写作不易，转载请注明出处，谢谢~~</strong></em></font></p>
<p> </p><br><br>
来源：https://www.cnblogs.com/ccbloom/p/15323176.html
頁:
[1]