1080ti+ubuntu14.04

Source: Internet
Author: User

I toss back and forth for a few days, from the installation of several systems, and then asked my classmates only to know that the original driver version of the problem, alas, the first time to run to Nvidia to see their online doc. I am quite depressed, dare not make simple point ah, I am a computer small white ah, hardware is ignorant ah. The overall steps are actually and http://www.cnblogs.com/gaowengang/p/6068788.html, and here, http://blog.csdn.net/masa_fish/article/details/51882183.

The important thing is to see the NVIDIA documentation. In fact, the installation procedures in the document are written. Also if driver wrong, sudo nvidia-uninstall, and then re-install can also.

Environmental Systems Ubuntu 14.04, gpu:1080ti

Be sure to look at the GPU before installing The drive supported by the card is the series, and Cuda toolkit The best way to bring your own series and compatibility is to check the official documentation, although I don't see (ashamed) myself.

Installation steps:

    1. Install ubuntu14.04, the specific can search online.
    2. Download 1080Ti Driver,nvidia official website download is nvidia-linux-x86_64-384.90.run, the number meaning of the Linux system, x86,64 bit system, 384 refers to the driver version number.
    3. Download cuda8.0. I was looking for someone, the version is Cuda_8.0.61_375.26_linux.run, the number, 8.0.61 is CUDA Toolkit version number, we commonly known as cuda8.0,375 is the driver serial number, Linux is the operating system. My classmates at Nvidia told me that driver had to be backwards compatible, meaning you could have a 384 drive, but it was a 8.0 compatibility with Cuda. And from the official website download driver version number, 1080Ti needs 384 driver, relatively new, so the direct use of cuda8.0 self-matching driver is not possible.
    4. To create a file:sudo vim /etc/modprobe.d/blacklist-nouveau.conf。并添加以下内容:

Blacklist nouveau

Options Nouveau modeset=0

5. Regenerate kernel Initramfs, which is to update the kernel. Execute the following command:

sudo update-initramfs–u

6. Restart system, after the new boot into the login interface, Ctrl + Alt + F1 into the command line mode.

7. After login, execute sudo service LIGHTDM stop to turn off LIGHTDM, that is, turn off desktop services.

8. sudo./nvidia-linux-x86_64-384.90.run--no-opengl-files, install driver driver. In this process, the beginning of the preinstall may not be successful, no tube, said to be thousand old problems, continue to install, and then basically is yes, I also prompted the installation process is not compatible with 32-bit, no tube. Then when I hit Nvidia x configuration, I chose no because my Nvidia card was just used for compute. There's nothing else to pay attention to.

9. After installing driver, continue to install Cuda. sudo./cuda_8.0.61_375.26_linux.run–no-opengl-libs. During the installation process, do not install driver, prompting you to select No when installing. Others can choose Yes and the default path.

10. After the installation is complete, sudo service LIGHTDM start//restart Desktop Services

Nvidia-smi can see the video card situation. You can also go to the samples path, make a bit, and then go to the subdirectory, then makes, to see the devicequery situation.

12. Set environment variables and so on. sudo gedit /etc/profilethen enter at the end of the text:

Export Path=/usr/local/cuda-7.5/bin: $PATH

Export ld_library_path=/usr/local/cuda7.5/lib64

13. Set sudo gedit ~/.BASHRC, enter the above two sentences at the end of the text. Run sudo ldconfig.

1080ti+ubuntu14.04

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.