Fix pf candidate scaling

[kdlong]

PR description:

Fix candidate scaling in cases where momentum is a huge number, where the mass-aware scaling hits numeric issues. The fix is simple, just calculate E in a way without ratios involved. This should be a fix for #41397

PR validation:

Checked that the crash and nan candidate are removed. The mass scaling works properly now.

If this PR is a backport please specify the original PR and why you need to backport that PR. If this PR will be backported please specify to which release cycle the backport is meant for:

Backport of #41550

[cmsbuild]

A new Pull Request was created by @kdlong (Kenneth Long) for CMSSW_13_0_X.

It involves the following packages:

• DataFormats/ParticleFlowCandidate (reconstruction)
• RecoParticleFlow/PFClusterTools (reconstruction)
• RecoParticleFlow/PFProducer (reconstruction)

@cmsbuild, @mandrenguyen, @clacaputo can you please review it and eventually sign? Thanks.
@mmarionncern, @rovere, @lgray, @missirol, @hatakeyamak, @seemasharmafnal 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

• Backported from Fix numeric issues in PFCand scaling, add some debug output #41550

[mmusich]

type pf

[mmusich]

type bug-fix

[mmusich]

backport of #41550

• not sure the bot will acknowledge this from me, though

[cmsbuild]

Pull request #41551 was updated. @cmsbuild, @mandrenguyen, @clacaputo can you please check and sign again.

[cmsbuild]

Pull request #41551 was updated. @cmsbuild, @mandrenguyen, @clacaputo can you please check and sign again.

[cmsbuild]

Pull request #41551 was updated. @cmsbuild, @mandrenguyen, @clacaputo can you please check and sign again.

[mandrenguyen]

please test

[cmsbuild]

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-296aea/32653/summary.html
COMMIT: a88ac6f
CMSSW: CMSSW_13_0_X_2023-05-15-1100/el8_amd64_gcc11
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/41551/32653/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

• You potentially added 13 lines to the logs
• Reco comparison results: 27 differences found in the comparisons
• DQMHistoTests: Total files compared: 49
• DQMHistoTests: Total histograms compared: 3555563
• DQMHistoTests: Total failures: 101
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 3555440
• DQMHistoTests: Total skipped: 22
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 48 files compared)
• Checked 213 log files, 164 edm output root files, 49 DQM output files
• TriggerResults: no differences found

[mandrenguyen]

+1

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_13_0_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_13_2_X is complete. This pull request will now be reviewed by the release team before it's merged. @perrotta, @dpiparo, @rappoccio (and backports should be raised in the release meeting by the corresponding L2)

[perrotta]

We should have discussed this backport today at the ORP meeting, but we did not address it...

The (dirty) fix for the segmentation violation in prompt reco was already merged with #41474

This is actually a proper fix, but it would (slightly) change the miniAOD outputs for the pf packed canditates, which propagate down to slimmed jets. As such, it should not get propagated onto a closed release already used for the prompt reconstruction.

Therefore my suggestion is that we do not backport this PR into the datataking relase 13_0_X. However, I'm open to suggestions. In the meanwhile, I put it on hold.

[perrotta]

hold

[cmsbuild]

Pull request has been put on hold by @perrotta
They need to issue an unhold command to remove the hold state or L1 can unhold it for all

[perrotta]

At the ORP of 23/05/23 we agreed with @cms-sw/reconstruction-l2 that this should not get backported to 13_0_X, due to the changes in the miniAOD outputs. In any case, the events to be fixed are very rare nonetheless

[kdlong]

Thanks @perrotta, it makes sense. Should I close the PR?

[perrotta]

Thanks @perrotta, it makes sense. Should I close the PR?

Yes, please