Introduce procModifier for parabolic magnetic field in track building for default iterative tracking

[mmasciov]

PR description:

As per title, this PR is introducing a process modifier to enable the usage of parabolic (rather than full) magnetic field for track building, for the default iterative tracking.
It follows a presentation at the TRK POG.

The process modifier is NOT enabled by default.
To enable process modifier, the following addition to cmsDriver.py command for step3 is sufficient:
--procModifiers=trackingParabolicMf

For documentation: the usage of the parabolic magnetic field in track building was (unintentionally) reverted in CMSSW 71X, as detailed in issue #41396.

In addition, commit eaf188f:

• renames RecoTracker.TrackProducer.TrackProducerIterativeDefault_cfi.TrackProducer to RecoTracker.TrackProducer.TrackProducerIterativeDefault_cfi.TrackProducerIterativeDefault (as suggested in comment);
• uses a safer syntax in the corresponding clone (as suggested in comment from PR Use parabolic magnetic field in track final fit and modify min N(hits) condition in track duplicate merge #39578).

PR validation:

As is, this is a technical PR: by default, the process modifier is NOT enabled.
To summarize, when the process modifier is enabled:

• For tracking physics performance: impact on efficiency and fake rate is negligible; impact on resolution is negligible; impact on pT residual at low pT is visible, although small.
• For tracking timing performance: overall track building (full tracking) time is reduced by less than 5% (~2%).

More details can be found in TRK POG presentation, or in the linked MTV results for TTbar events with PU.

FYI, @mmusich, @slava77

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-41557/35431

• This PR adds an extra 60KB to repository

• There are other open Pull requests which might conflict with changes you have proposed:

  • File RecoTracker/IterativeTracking/python/JetCoreRegionalStep_cff.py modified in PR(s): Update DeepCore training for Run-3 #38169

[cmsbuild]

A new Pull Request was created by @mmasciov (Mario Masciovecchio) for master.

It involves the following packages:

• Configuration/ProcessModifiers (operations)
• RecoTracker/CkfPattern (reconstruction)
• RecoTracker/IterativeTracking (reconstruction)
• RecoTracker/TkNavigation (reconstruction)

@perrotta, @rappoccio, @clacaputo, @cmsbuild, @mandrenguyen, @fabiocos, @davidlange6 can you please review it and eventually sign? Thanks.
@fabiocos, @VourMa, @makortel, @felicepantaleo, @GiacomoSguazzoni, @JanFSchulte, @rovere, @VinInn, @Martin-Grunewald, @missirol, @ebrondol, @gpetruc, @mmusich, @mtosi, @dgulhan this is something you requested to watch as well.
@perrotta, @dpiparo, @rappoccio you are the release manager for this.

cms-bot commands are listed here

[cmsbuild]

-code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-41557/35435

• This PR adds an extra 52KB to repository

• There are other open Pull requests which might conflict with changes you have proposed:

  • File RecoTracker/IterativeTracking/python/JetCoreRegionalStep_cff.py modified in PR(s): Update DeepCore training for Run-3 #38169

Code check has found code style and quality issues which could be resolved by applying following patch(s)

• code-checks:
  https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-41557/35435/code-checks.patch
  e.g. curl -k https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-41557/35435/code-checks.patch | patch -p1
  You can also run scram build code-checks to apply code checks directly

[slava77]

I don't see changes in DisplacedGeneralStep_cff.py and DisplacedRegionalStep_cff.py.
I guess we missed them for fitting as well

[mmasciov]

I don't see changes in DisplacedGeneralStep_cff.py and DisplacedRegionalStep_cff.py. I guess we missed them for fitting as well

Yes, no change in those so far. Can be propagated if wished; however, those configurations are not run by default, and all these and previous changes only targeted default iterations. Thus, will not do unless requested.

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-41557/35436

• This PR adds an extra 52KB to repository

• There are other open Pull requests which might conflict with changes you have proposed:

  • File RecoTracker/IterativeTracking/python/JetCoreRegionalStep_cff.py modified in PR(s): Update DeepCore training for Run-3 #38169

[cmsbuild]

Pull request #41557 was updated. @perrotta, @rappoccio, @clacaputo, @cmsbuild, @mandrenguyen, @fabiocos, @davidlange6 can you please check and sign again.

[mmasciov]

I don't see changes in DisplacedGeneralStep_cff.py and DisplacedRegionalStep_cff.py. I guess we missed them for fitting as well

Yes, no change in those so far. Can be propagated if wished; however, those configurations are not run by default, and all these and previous changes only targeted default iterations. Thus, will not do unless requested.

As a change was necessary in DisplacedRegionalStep_cff.py, I ended up propagating changes to both.

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-41557/35467

• This PR adds an extra 60KB to repository

• There are other open Pull requests which might conflict with changes you have proposed:

  • File RecoTracker/IterativeTracking/python/JetCoreRegionalStep_cff.py modified in PR(s): Update DeepCore training for Run-3 #38169

[cmsbuild]

Pull request #41557 was updated. @perrotta, @rappoccio, @clacaputo, @cmsbuild, @mandrenguyen, @fabiocos, @davidlange6 can you please check and sign again.

[slava77]

@cmsbuild please test

[cmsbuild]

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-0227aa/32474/summary.html
COMMIT: 1bc9f4d
CMSSW: CMSSW_13_2_X_2023-05-08-1100/el8_amd64_gcc11
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/41557/32474/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

• You potentially added 9 lines to the logs
• Reco comparison results: 4 differences found in the comparisons
• DQMHistoTests: Total files compared: 48
• DQMHistoTests: Total histograms compared: 3460836
• DQMHistoTests: Total failures: 3
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 3460811
• DQMHistoTests: Total skipped: 22
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 47 files compared)
• Checked 207 log files, 159 edm output root files, 48 DQM output files
• TriggerResults: no differences found

[clacaputo]

+reconstruction

[rappoccio]

+1

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next master IBs (tests are also fine). This pull request will be automatically merged.