我正在尝试在装有 Intel 和 Nvidia GeForce 820M 显卡的戴尔笔记本电脑上安装 CUDA 7.5。最初,我从 .run 文件安装它,遇到了很多问题,例如启动时黑屏等。解决这些问题后,当我运行 deviceQuery 时,出现错误:未检测到支持 CUDA 的设备。错误是正确的,因为我在任何地方都找不到 /dev/nvidia*。
这是我的inxi -bG
输出:
System: Host: shubh-Inspiron-3542
Kernel: 3.13.0-93-generic x86_64 (64 bit)
Desktop: Gnome 3.10.4
Distro: Ubuntu 14.04 trusty
Machine: System: Dell (portable) product: Inspiron 3542
Mobo: Dell model: 0926J6 version: A03
Bios: Dell version: A03 date: 05/27/2014
CPU: Dual core Intel Core i5-4210U CPU (-HT-MCP-) clocked at 782.00 MHz
Graphics: Card: Intel Haswell-ULT Integrated Graphics Controller
X.Org: 1.15.1 drivers: intel (unloaded: fbdev,vesa) Resolution: [email protected]
GLX Renderer: Mesa DRI Intel Haswell Mobile GLX Version: 3.0 Mesa 10.1.3
Network: Card-1: Qualcomm Atheros QCA9565 / AR9565 Wireless Network Adapter driver: ath9k
Card-2: Realtek RTL8101/2/6E PCI Express Fast/Gigabit Ethernet controller driver: r8169
Card-3: Atheros
Drives: HDD Total Size: 1000.2GB (6.0% used)
Info: Processes: 207 Uptime: 5 min Memory: 1077.5/3857.5MB Client: Shell
我在这里找不到 NVIDIA 卡;
但是当我跑步时lspci | grep -i nvidia
,我得到了:
08:00.0 3D controller: NVIDIA Corporation GF117M [GeForce 610M/710M/810M/820M / GT 620M/625M/630M/720M] (rev a1)
同样对于运行dpkg -l | grep nvidia
,我得到了输出:
rc bumblebee-nvidia 3.2.1-9~gpu14.04.1 amd64 NVIDIA Optimus support using the proprietary NVIDIA driver
rc nvidia-304 304.131-0ubuntu0.14.04.2 amd64 NVIDIA legacy binary driver - version 304.131
ii nvidia-340 340.96-0ubuntu0.14.04.1 amd64 NVIDIA binary driver - version 340.96
ii nvidia-common 1:0.2.91.11 amd64 transitional package for ubuntu-drivers-common
rc nvidia-opencl-icd-304 304.131-0ubuntu0.14.04.2 amd64 NVIDIA OpenCL ICD
ii nvidia-opencl-icd-340 340.96-0ubuntu0.14.04.1 amd64 NVIDIA OpenCL ICD
ii nvidia-prime 0.6.2 amd64 Tools to enable NVIDIA's Prime
ii nvidia-settings 331.20-0ubuntu8 amd64 Tool for configuring the NVIDIA graphics driver
目前,我已通过安装旧版 nvidia 驱动程序 340 恢复了系统,并安装了 CUDA 6.5,因为我正在尝试使用 nvidia 340 驱动程序。但是,如果可能的话,我希望在我的计算机上安装 CUDA 7.5。我受够了重新安装 nvidia 驱动程序,因为启动时会出现黑屏。
答案1
如果你愿意升级,我成功地在 16.04 版中将 CUDA 部署在了不受限制的 NVIDIA GPU 上。只要你模仿有关文件的说明,这些步骤甚至可以在 14.x 中.run
工作