Warning: Do not install the display driver! (At least it didn't work on my Thinkpad T460p) CUDA 9.0 for tf $ nvidia-smi | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. Next step: Install cuDNN CUDA 9.2 $ nvidia-smi Install CUDA requirements (see section below).Download CUDA for Ubuntu 17.10 (runfile local) - Tensorflow recommends CUDA 9.0 - CUDA 9.2 seems not to work with tf.Sudo ln -s /usr/bin/g++-6 /usr/local/cuda/bin/g++ echo 'export PATH=/usr/local/cuda-9.1/bin:$PATH' > ~/.bashrcĮcho 'export LD_LIBRARY_PATH=/usr/local/cuda-9.1/lib64:$LD_LIBRARY_PATH' > ~/.bash.rcįinally you have to set up simlinks to gcc-6 and g++-6 or you will get a warning on compiling your own code.(Required) sudo ln -s /usr/bin/gcc-6 /usr/local/cuda/bin/gcc Also the nvidia documentation will tell you what to export. The run file will give you a reminder, too. It will ask you a bunch of questions, and it will want them answered immediately.Īnswer yes to the unsupported confguration.Īfter that is installed then make sure you set up your paths. It will install it using the new gcc compilers. Mark the file as executable(I right click on the file in desktop). I ended up installing the 9.1 run file for Ubuntu 17.04. (I would probably skip this one, but go to it if other stuff is giving issues) sudo apt install nvidia-cuda-toolkit You can try using nvidia-cuda-toolkit, but the paths to the libraries wasn't familiar to me. As of today nvidia driver 396 is available. I use the GUI updater for the most part called Software & Updates, in the Additional Drivers tab. sudo add-apt-repository ppa:graphics-drivers/ppa (I would say this is required before installing, unless you want to risk some login loop of death). There might be added extra stuff that I could have probably not had to do, but I am going to include it anyways.įirst get the ppa repository drivers. The second method has the downside that it's not as easy to upgrade or remove. Make sure you don’t agree to install the new driver. Accept the terms and conditions, say yes to installing with an unsupported configuration, and no to “Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 384.81?”.make the downloaded file executable with sudo chmod +x.After installing the proprietary NVIDIA driver, download CUDA 9 installation from their site (get Ubuntu the 17.04 runfile version).CUDA requires gcc6, use update-alternatives to maintain both gcc7 and gcc6 as explained here.Īlternatively you can follow Taylor's instructions:.sudo apt update & sudo apt install nvidia-cuda-toolkit, or install it from the ubuntu software center.In the Additional Drivers tab in software & updates select the NVIDIA proprietary driver (390 for CUDA 9).In software & updates, select the restricted and multiverse repositories. | 0 NVIDIA GeForce … Off | 00000000:01:00.Enable multiverse repository, install nvidia drivers and nvidia-cuda-toolkit and gcc6 (preferably using update-alternatives to easily switch versions): PS: Here is what I get when invoke nvidia-smi So, I find myself stuck with incompatibility between the driver and Cuda. The only toolkit option I see at nVIDIA’s download site appears to be for driver 520.61.05. When I run nvidia-smi I see CUDA version as 11.8 but I don’t know how to install it. The display driver I was able to install and run without any issues is 520.56.06. I just got a new desktop containing RTX 4090 running Ubuntu 22.04 LTS.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |