Here is the scenario: You just installed Visual Studio 2017 Community Edition You just installed CUDA Toolkit 10 You create a CUDA project in Visual Studio You are trying to build the project with the default code and get an […]
In order to get Nsight Eclipse launched without any errors there are two things that need to happen: Make sure the java-1_8_0-openjdk package is installed on your system https://www.digitalocean.com/community/tutorials/how-to-install-java-with-apt-on-ubuntu-18-04 Launch nsight with the following command: nsight -vm /usr/lib/jvm/java-8-openjdk-amd64/jre/bin/java Here […]
If you would like to run CUDA Samples on Jetson Xavier: Open a terminal in the sample you would like to run. For example, on my machine, open a terminal in Home/NVIDIA_CUDA-10.0_Samples/5_Simulations/oceanFFT Make sure the terminal is in that directory. […]
DO NOT USE A VIRTUAL MACHINE! I am having problems running code written on the host machine (VM Ubuntu 18.04 LTS) on the Jetson Xavier. I keep getting a “Exec Format Error” whenever I run remotely on the Jetson Xavier. […]
Use your NVIDIA GPU to make your computer vision project run faster! Below is a link to a detailed video walkthrough of adding the CUDA module to OpenCV on a Windows-based machine. The video walkthrough is 1 hour and 30 […]
A video walkthrough (59 minutes) of using CUDA dynamic parallelism to achieve some an objective. Includes source code. Click the link below.
Learn how to use CUDA Pinned Memory to make your applications run faster. You can’t process if you don’t have the data! Pinned memory is used give your GPU data faster so it can keep busy processing. This is a […]
When I try to run CUDA code that takes a long time to process on the GPU, I would always get an error such as the following: Error: C:/kernel.cu:170, code: 4, reason: unspecified launch failure After spending many sleepless nights […]
NVIDIA CEO explains Dynamic Parallelism
Learn how to use cooperative groups to make your parallel processing code more organized and manageable. The video walkthrough is 32+ minutes long and includes example source code.