Post B6FI0HJLP2wxRNgLXk by azonenberg@ioc.exchange
(DIR) More posts by azonenberg@ioc.exchange
(DIR) Post #B6FFYkUV7SWtvLe652 by azonenberg@ioc.exchange
0 likes, 0 repeats
Progress! CI jobs for two consecutive commits dispatched by GitHub commit hook.First one had some minor issues, I fixed the CI script, then pushed a new commit before the first build had finished running.If everything works, we should get a full build of both commits soon.I've started work on the infrastructure for artifact uploading but need another day or two to get that all wrapped up.
(DIR) Post #B6FGgA8gG6R9S2XRc8 by kmlyyll@mastodon.social
0 likes, 0 repeats
@azonenberg Very nice progress. Getting consecutive CI runs working reliably is a huge milestone, and fixing issues during active builds is always a good sign the system is maturing quickly. Excited to see the artifact upload infrastructure come together too.
(DIR) Post #B6FGqrYTh41102huN6 by azonenberg@ioc.exchange
0 likes, 0 repeats
@kmlyyll Still a bunch more work to do. Eventually I want to separate building and testing so I don't have to lock up a GPU during a compile (only during the test).And I need to set up hardware-in-loop testing on real oscilloscopes etc.But once I have it at least matching the feature set of the original GitHub Actions pipeline, I can shut that down for good
(DIR) Post #B6FH4hzQ3RNHO3uXU8 by kmlyyll@mastodon.social
0 likes, 0 repeats
@azonenberg Separating compile and GPU test stages sounds like the right move. And hardware-in-loop testing with real oscilloscopes is seriously cool territory for CI infrastructure. (Wwhich I have no experience :P)Reaching feature parity with the existing GitHub Actions pipeline already sounds like a huge achievement to me!Well done!
(DIR) Post #B6FI0HJLP2wxRNgLXk by azonenberg@ioc.exchange
0 likes, 0 repeats
I need to get some sleep, I'm already up pretty late considering how early I'm getting up.But I wanted to wait until at least one job had run from both commits (debian 12 nvidia).The dashboard shows the builds in order of when the runner VM started (not when the jobs were dispatched) so a few are slightly out of order due to SLURM's scheduling.One thing I still need to figure out is how to summarize all of the builds from a single commit to get a go/no go status.