Allow profiling workflow(s) to be specified with PR multiline comment

[gartung]

Adding multiline comment

test parameters:
workflows_profiling=11834.21

should override the default
PR_TEST_MATRIX_EXTRAS_PROFILING=23434.21,11834.21

[cmsbuild]

A new Pull Request was created by @gartung (Patrick Gartung) for branch master.

@cmsbuild, @smuzaffar, @mrodozov can you please review it and eventually sign? Thanks.
cms-bot commands are listed here

[gartung]

@smuzaffar

[cmsbuild]

Pull request #1545 was updated.

[slava77]

should this PR still be restricted to just what's possible/tested in the context of https://github.com/cms-cmpwg/profiling/blob/master/Gen_tool/Gen.sh and what follow from it (4 steps etc) ?

I'm guessing that I can't ask for some random wf, like 136.88811 which we often use for data PAT/miniAOD testing.

[smuzaffar]

@slava77 , yes you can not ask for any random workflow. Workflow should the one defined here https://github.com/cms-sw/cms-bot/blob/master/cmssw-pr-test-config#L3

[cmsbuild]

Pull request #1545 was updated.

[smuzaffar]

@gartung , can we trigger an IB profiling job if PR profiling workflow is not part of https://github.com/cms-sw/cms-bot/blob/master/cmssw-pr-test-config#L3 ? I think from the pr-schedule-tests we can check this and ask jenkins to start such job

[gartung]

The profiling script expects a 4 step simulation workflow with simulation, pileup mixing, reconstruction, and miniAOD production. I am trying to figure out a way to validate the workflow against PR_TEST_MATRIX_EXTRAS_PROFILING

[cmsbuild]

Pull request #1545 was updated.

[cmsbuild]

Pull request #1545 was updated.

[gartung]

enable profiling

[gartung]

test parameters:
workflows_profiling=11834.21

[gartung]

please test

[gartung]

abort

[smuzaffar]

@gartung There is no way we can test this as it is. Either we merge it or make two PRs one for cms-bot so that it can properly pass the new profiling parameter and other to use it in profiling job.
I suggest to just merge it :-)

[gartung]

OK Thanks.