Installing CUDA 9.1 on Fedora 28: Taking the Easy Way Out

So, yesterday morning I ran into a bit of a problem on my desktop which was still running Fedora 27. I started up the computer and the greeter/login screen failed to load. The mouse cursor was visible and responsive, but the screen was otherwise black. I was able to get to a terminal, login and launch the X-Server, so I did some basic trouble shooting, but when none of the usual things things worked, I decided to just back up my data and upgrade to Fedora 28.

Installing Nvidia’s CUDA 9.1 on Fedora 27

 

If you’re like me, you to want to make sure that you’re using an up-to-date version of your operating system. This is not a problem for the vast majority of software, but when it comes to CUDA, it seems that NVidia is always a bit behind the times and it seems that only end-of-life versions of Fedora are ever officially supported. This doesn’t prevent you from installing CUDA on non-supported versions of Fedora, it just means you have to jump through a few more hoops.

Below you’ll find how I installed CUDA 9.1 on my Lenovo Ideapad Y700 laptop with NVidia Optimus technology. In the near future I’ll update this post to include desktop instructions, though it seems like there shouldn’t be any major issues there.

Update: A few days ago, I finally updated my desktop to Fedora 27 and installed CUDA 9.1 there. The install went quite smoothly, and I was able to build all the samples by simple going into the main sample directory and typing ‘make’. On my laptop, this worked to a degree, but since I had to use bumblebee, it couldn’t locate an NVidia library for one of the samples. On my desktop, this was not an issue. Below, I’ve updated the text to be a bit more general.

Is it time for me to say goodbye to Fedora?

The default KDE desktop in Manjaro after a fresh install on my HP Pavilion dv6-6135dx.

Way back in the year of 2012, I made a decision that would change the way that I used a computer from that day forward. Just like many other people, I had always used a version of Microsoft Windows as the operating system on my computers, even the ones I built myself. Using Windows for doing work, however, was becoming increasingly frustrating. It was a huge pain to do development work, especially for a fledgling scientist who had only taken a single programming course and had to figure out everything for myself. The code I had written was slower than molasses on an Alaskan winter day, but the part that was the slowest could be done in parallel. I already had used OpenMP to parallelize the for loop, but it was still going to take forever to run. I learned about GPGPU, OpenCL and CUDA that day. CUDA was the most appealing as it used simple C-like syntax, and I was a C++ programmer. To use CUDA on Windows, I would have needed to purchase Visual Studio, not easy to do on a graduate student’s salary, but I learned you could use it for free on Linux. That was it. The decision was made. That day I installed Fedora 17 on my work desktop, and never looked back.

How Rust Can be A Good Thing

Last September I saw this article about a programming language I hadn’t heard about before. That language was Rust. At the time, I played around with it a tiny bit but saw no real compelling reason for me to invest the time into learning another language. Then, earlier this week I saw this article on arXiv, and decided to give Rust another look.

Installing Nvidia’s CUDA 8.0 on Fedora 25

UPDATE (December 21, 2017): Fedora 25 is now end-of-life and if you still wish to use CUDA on Fedora 25, CUDA 9.1 includes a repository which should make installation trivial. For Fedora 27 see this new post.

UPDATE (March 28, 2017): I have found a way to use at least some of the C++11 features. See the end of the post for the changes.

It’s been a long time since I’ve posted anything. Since my last post I had to update the operating system on my work desktop as Fedora 23 went end-of-life. While it is possible to simply use dnf to upgrade your system, I have the proprietary Nvidia driver installed on my system for two reason. First, with my graphics cards (GeForce GTX 650’s), the Nouveau driver doesn’t seem to really work. Without fail, after about 30 minutes the screen no longer refreshes, making it seem as though the system is locked up. Second, I do some GPU programming with CUDA which requires the proprietary driver. With the proprietary driver installed, it’s a bit more difficult to upgrade, so I tend to just back everything up and do a clean install.

Having done that recently, I find myself this morning needing to re-install CUDA as I have a computational problem which could benefit from some massive parallelism. I figured I’d go ahead and post the procedure here for my future reference and in case anyone else might benefit from it.

Microsoft Windows 10 is Malware

Many of you may be aware that Microsoft came under fire not too long ago for violating its own guidelines for what the little x in the top right hand corner of a dialog box or window should do, close the window and cancel any process associated with that window. Instead, they decided that in a last ditch effort to force people to upgrade to Windows 10, clicking the x on the notification that your Windows 10 update was ready (after it had downloaded itself onto your computer without your permission or by disguising itself as a different update) would essentially be you consenting to the update. I could link a bunch of articles about this, but it would probably be easier for you to just google “Is Windows 10 malware?” and you can see for yourself. After the outcry, Microsoft said they would adjust their update policy, but that’s not what this post is about. This is about Windows 10 intentionally harming peoples computers who dual boot with a Linux distribution.