[L1T] [Phase-2] Update Correlator code (Superseding #40670)

[aloeliger]

PR description:

This PR duplicates and supersedes #40670, as @cecilecaillol has asked me to take it over.

Previous PR description:

This is an update of the correlator code

Includes:

p2l1pfp#92 by @thesps and @slaurila
p2l1pfp#94 by @thomreis and @cerminar (only adds a new feature to the emulator, doesn't change any of the existing emulated values)
p2l1pfp#96 by @thesps
p2l1pfp#97 by @drankincms
p2l1pfp#98 by @drankincms (not used by default)
p2l1pfp#100 by @thesps
p2l1pfp#101 by @gpetruc (not used in production)
p2l1pfp#105 by @gpetruc (not used in production)
p2l1pfp#107 by @gpetruc (transparent)

Local: cms-l1t-offline#1063

I will be updating this PR as requested in the original thread, as soon as I am able.

[srimanob]

By the way, @aloeliger, do you also open a PR for data external? Could you please mention it also in this PR? Thx.

[aloeliger]

@srimanob I have not opened any PR for data externals. I've never actually opened a data PR. Where should that be opened?

[aloeliger]

@aloeliger if one of the (rightful) concern of squashing is giving the right credits to authors, a partial solution could be to:

• drop 0f12464 in the rebase process (the commit is only adding the .*pb file);
• squash everything after a564daf (the commit for the ROOT file) in a single commit;
• have this last commit co-authored by those involved in the original ones (see the guide here)

In this way you would still keep most of the history and would give credits to all the squashed authors.

@thesps @slaurila @thomreis @cerminar @drankincms & @gpetruc I believe i have prepared a version of the branch doing this (Giovanni and Emyr you should be listed as co-authors on squashed commits). Unless I hear from you by tomorrow morning Geneva time, I will for push that version of the branch and work on other comments for this PR.

[thesps]

@aloeliger I'm fine with the squash, thanks for preparing it

[aloeliger]

@thesps @slaurila @thomreis @cerminar @drankincms & @gpetruc Would it be possible for you to provide the root and model file removed from these PRs? Do these need to be in the data externals? Please let me know so I can open a corresponding PR to provide this data.

[thesps]

You can fetch them from https://github.com/p2l1pfp/cmssw/tree/L1PF_12_5_X/L1Trigger/Phase2L1ParticleFlow/data
(Safer than if I just send you some files)

[srimanob]

@srimanob I have not opened any PR for data externals. I've never actually opened a data PR. Where should that be opened?

I think the place to go is to https://github.com/cms-data. I don't see your root file anymore, so I can't say if you already have repository there already.

[aloeliger]

@smuzaffar would it take anything special to get L1Trigger/Phase2L1ParticleFlow/data/modelTT_PUP_Off_dXY_XYCut_Graph.pb & L1Trigger/Phase2L1ParticleFlow/data/jecs/jecs_20220308.root integrated into https://github.com/cms-data? These would be the first of the data-files for L1Trigger-Phase2L1ParticleFlow. Is there anything special about creation of the repository or it's name?

[smuzaffar]

@aloeliger , nothing special, I just need to create the new data repo. if L1Trigger/Phase2L1ParticleFlow is the right package from these data files then I can create data-L1Trigger-Phase2L1ParticleFlow

[gpetruc]

Isn't it already there? https://github.com/cms-data/L1Trigger-Phase2L1ParticleFlow

[smuzaffar]

Isn't it already there? https://github.com/cms-data/L1Trigger-Phase2L1ParticleFlow

oops, you are right :-) @aloeliger repo is already there, you just need to open a PR with new data files there
thanks @gpetruc

[aloeliger]

Dang. That messed up the history. Okay. Let me retry this.

[aloeliger]

My apologies to the combined authors, but I'm done trying to mess around with the git history on this when the PR has other necessary edits. I am simply going to squash the whole thing.

[gpetruc]

Ok, no problem. Just put in the commit log and new PR a link to the old PR to cms-l1t-offline to help trace back history if needed for debugging later. Il Mer 5 Apr 2023, 09:13 Andrew Loeliger ***@***.***> ha scritto:

…

My apologies to the combined authors, but I'm done trying to mess around with the git history on this when the PR has other necessary. I am simply going to squash the whole thing. — Reply to this email directly, view it on GitHub <#41260 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABDNWR4NU5UIQWLTBXUCYPLW7ULQZANCNFSM6AAAAAAWRUEKXU> . You are receiving this because you were mentioned.Message ID: ***@***.***>