Will iteratively add documentation here.

Notes:

• Job restriction plugin is in place (related: Evaluate job restrictions plugin to improve build/test/(trestle?) machine isolation infrastructure#3379)
• https://ci.adoptium.net/job/Trestle_Pipeline/ as placeholder / front end to triggering build/test pipeline

I notice that this is running the same platform specific build pipelines underneath the top level ones so it is using the _temurin suffixed ones. Has any consideration been made to using the _hotspot ones instead for trestle? Clearly there are various implications of this (retention disk space being one) but it may be useful to have that separation, particularly since the trestle pipelines have to put in additional build options that would not be required for a "VARIANT=hotspot" build.
I ran a jdk8u pipeline earlier and that fell over due to the Temurin variant builds bedding our extra company name option which doesn't exist on a normal openjdk fork.

Yes, I would prefer to use the _hotspot variant but hit build failure upon running it, so shifted back to _temurin variant in order to build successfully.

Hmmm yeah I just did a quick test with one of the hotspot pipelines equivalent to a _temurin one I ran yesterday) and it fell over with - LMK if this looks similar to what you hit:

Loading library openjdk-jenkins-helper@null
Examining [adoptium/jenkins-helper](https://github.com/adoptium/jenkins-helper)
Attempting to resolve null as a branch
Attempting to resolve null as a tag
ERROR: Could not resolve null
[Pipeline] }
[Pipeline] // node
[Pipeline] End of Pipeline
ERROR: No version null found for library openjdk-jenkins-helper
/home/jenkins/.jenkins/jobs/build-scripts/jobs/jobs/jobs/jdk8u/workspace/jdk8u-aix-ppc64-hotspot@tmp/jfrog/1202/.jfrog deleted
Finished: FAILURE

ERROR: Could not resolve null seems familiar, yes

Yes, I would prefer to use the _hotspot variant but hit build failure upon running it, so shifted back to _temurin variant in order to build successfully.

Interesting - would definitely be good to fix that. My initial though is that it might be problematic if the jobs haven't been regenerated in ages so are out of sync...