Hlt2 pp commissioning settings

Settings for running hlt2 on pp commissioning data with dd4hep. Goes with MooreOnline!174 (merged)

Superseded by !1961 (merged)

marked this merge request as draft

added RTA label

Indentation needs to be fixed here. On top, I'm reluctant to add code without a test checking it's not broken. You may say the test is in MooreOnline but then I would move this file there too.

marked this merge request as ready

mentioned in issue DaVinci#81 (closed)

added 44 commits

• e600041e...e499b772 - 43 commits from branch master
• f8b5c3b0 - settings for running hlt2 on pp commissioning data with dd4hep

Compare with previous version

added 3 commits

• f8b5c3b0...ae7cf082 - 2 commits from branch master
• d754f475 - settings for running hlt2 on pp commissioning data with dd4hep

Compare with previous version

added 1 commit

• 8c92b64a - settings for running hlt2 on pp commissioning data with dd4hep

Compare with previous version

mentioned in merge request MooreOnline!174 (merged)

mentioned in merge request !1928 (merged)

Where are we with this ? Should it be retested ? Was a test added ? Should it go back to Draft ?

marked this merge request as draft

changed the description

mentioned in merge request !1961 (merged)

added 109 commits

• 8c92b64a...2db843a7 - 108 commits from branch master
• cc7b356f - settings for running hlt2 on pp commissioning data with dd4hep

Compare with previous version

marked this merge request as ready

changed the description

resolved all threads

mentioned in issue #504 (closed)

/ci-test MooreOnline!174 (merged)

Action failed with

"lhcb/MooreOnline!174" cannot be merged

This needs a rebase

I was 15 minutes too late with the update :)

/ci-test MooreOnline!174 (merged)

Started integration test build. Once done, check the results or the comparison to a reference build.

Throughput Test Moore_hlt2_fastest_reco: 496.6 Events/s -- change of -0.16% vs. reference

Throughput Test Moore_hlt2_pp_thor: 242.5 Events/s -- change of -0.46% vs. reference

Throughput Test Moore_spruce_all_lines: 150.1 Events/s -- change of -0.59% vs. reference . Total bandwidth 2687.0 MB/s -- change of 0.00% vs. reference

Throughput Test Moore_hlt1_pp_default: 26629.2 Events/s -- change of 0.17% vs. reference

Looks rather good, but for one Alignment test.

I don't see a connection of this MR to failures in that particular alignment test. @shollitt or @freiss can you please take a look? https://lhcb-nightlies.web.cern.ch/nightly/lhcb-master-mr/6598/Alignment/x86_64_v2-centos7-gcc11-opt/tests#Escher.align-ft-modules

For this MR, I've moved the streaming settings to MooreOnline, since they'll break the tests introduced in !1961 (merged) .

I agree that the failure of this alignment test seems unrelated to the changes here

Indeed, and it was confirmed by @gunther that we've missed it in an MR merged yesterday. And it's there is master too : https://lhcb-nightlies.web.cern.ch/nightly/lhcb-master/1880/Alignment/x86_64_v2-centos7-gcc11-opt/tests

So this looks good then, let's try to merge.

But wait, this is only adding a .py file that is completely unused. What's the point ? If it's used in MooreOnline, then it should probably be moved there. If it will be used in the future in Moore, then let's merge it together with its future user.

It will be used in !1961 (merged) (which has already cherry-picked this MR so I could test it locally). I think this could be closed in favour of !1961 (merged).

added ci-test-triggered label

• [2022-12-16 00:04] Validation started with lhcb-master-mr#6598

added 8 commits

• cc7b356f...e17e2e83 - 6 commits from branch master
• 67545c96 - settings for running hlt2 on pp commissioning data with dd4hep
• e1f780e6 - move stream settings to MooreOnline

Compare with previous version

closed

changed the description