This is a collection of results for tests run with kdevops on different kernels. Each user has their own dedicated namespace.
This git tree leverages git lfs, what this allow is for you to download only a smaller version of the repository without the actual tarballs of the results, you then can decide when and if you need the tarballs by selectively downloading them with "git lfs fetch --include". As it stands without LFS if you clone this git tree you'd have to download 218M, if you use GIT_LFS_SKIP_SMUDGE=1 before the clone to avoid downloading all the files you'd just get a 2.5M directory, for a 98.85% of space savings.
To leverage LFS with an example:
GIT_LFS_SKIP_SMUDGE=1 git clone https://github.com/linux-kdevops/kdevops-results-archive.git
cd kdevops-results-archive
git lfs fetch --include "fstests/gh/linux-ext4-kpd/20250415/0001/linux-6-15-rc2/8ffd015db85f.xz
git lfs checkout "fstests/gh/linux-ext4-kpd/20250415/0001/linux-6-15-rc2/8ffd015db85f.xzkdevops will automatically make use of the archive if you have it present. To be clear, you should have both the kdevops and kdevops-results-archive directory in the same parent directory. Do not put kdevops-results-archive inside the kdevops directory.
Please ensure you use *.xz tarballs, so, if you're going to push
results, we have a sanity check to ensure you don't push directories
with more than 5 files. To ensure this you can use:
git config core.hooksPath hooksCommits pushed to kdevops-results-archive automatically trigger re-generation of the entire kdevops dashboard:
The dashboard code was generated by an AI agent based on target descriptions of what we wanted. We have better things to do than to write dashboard code and so this is meant to be super simple. The dashboard therefore leverages results inside each commit. So if you want to process results, add them to the commit log. The entire generation of the dashboard takes about 1-2 seconds to run today, and given we rotate the archive on epochs, it means we'll sustain this speed over time to make this scalable.
If you want to try to enhance the dashboard, consider using either claude-ai or github codepilot, they are proving to be the best agents to easily modify this code. You can just describe what you want to change, and try to demo it before pushing as follows:
rm -rf dashboard;
time ./bin/gen-dashboard.py -s $(git rev-list --max-parents=0 HEAD)
That's it, point your we browser to your local dashboard/index.html and if it looks like an enhancements just push.
To see contents you can use something like:
tar -tOJf fstests/mcgrof/xfs/libvirt-qemu/20240505-0001/6.9.0-rc6.xz 2>&1 | grep xfs | grep 033
To see fstests results you can use something like this:
tar -xOJf fstests/mcgrof/xfs/libvirt-qemu/20240505-0001/6.9.0-rc6.xz 6.9.0-rc6/xfs_reflink_1024/xfs/033.out.bad
tar -xOJf fstests/mcgrof/xfs/libvirt-qemu/20240505-0001/6.9.0-rc6.xz 6.9.0-rc6/xfs_reflink_1024/xfs/033.dmesg