IN5050 - ARM Resources & FAQ
?
A page for resources and frequently asked questions for the Jetson AGX Xavier machines. If you have any other questions, please send an email to?in5050@ifi.uio.no
The following table gives an overview of the status of the?ARM machines at IFI.?
Remember that you might need to SSH into login.ifi.uio.no to access the IN5050 login machine (roa.sinlab.eu). Students from a previous semester created this guide on setting up SSH and SSHFS on external machines. The guide is still valid. However, you might have to change some hostnames/IP addresses.
A username and password have been provided to all groups. The following table gives an overview of the status of the ARM machines with a GPU at IFI:
| Computer | CPU | Memory | SIMD Version | Status |
| tegra-1 | NVIDIA Tegra Xavier | 16GB | ARMv8.2 / NEON | Up |
| tegra-2 | NVIDIA Tegra Xavier | 16GB | ARMv8.2 / NEON | Up |
| tegra-3 | NVIDIA Tegra Xavier | 32GB | ARMv8.2 / NEON | Up |
| tegra-4 | NVIDIA Tegra Xavier | 32GB | ARMv8.2 / NEON | Up |
| tegra-5 | NVIDIA Tegra Xavier | 16GB | ARMv8.2 / NEON | Down |
?
ARM Programming?Resources
?
Frequently Asked Questions
Q: Do we have any video source files to test with??
A: The video source files are stored in /mnt/sdcard.
?
Q: My video is broken. Are there any tools to analyze the video?
A: Yes, there are! Try out YUView, an open-source and cross-platform YUV player and analysis tool.
?
Q: Can you install extra tools and packages on the lab machines?
A: Yes, talk to us on Mattermost, and give us the packages you need.
?
Q: We have trouble enabling FP16 SIMD extensions on the Tegra machines.
A: FP16 is optional on ARMv8 processors, so you have to compile your program with the flag: -march=armv8.2-a+fp16
?
Q:?How about compiler optimizations?
A:?For this assignment, you will use no automatic vectorization and only O1 optimization.
?
Q: What tools are available for profiling my code?
A: Have a look at gprof, kcachegrind
?
Q: Can we copy/paste a fast DCT implementation optimized for NEON, which we found online, into our code?
A: No, you may not.?
?
Q:?What kind of speed improvements are you expecting from our code??
A:?We do not expect a certain percentage of improvement. We want you to experiment with instructions that exploit the built-in parallelism of the ARM architecture, and todescribe in your report what you have tried and achieved. What part of the code have you investigated? What did you try? How did it go? This is the most important. It is positive if you can document a pattern in the improvements. There is no need to speculate about how the processor parallelizes, since we lack documentation to evaluate whether the speculation is correct.
?