Parametrized magnetic field in alpaka

[lecriste]

PR description:

Preliminary PR to access the magnetic field approximation in alpaka.
No changes are expected in the output.

Triggered by last bullet in https://docs.google.com/presentation/d/1WVF6F0CHImtcGzj-Pnyvo6nEBkTJokDsJx1JyxwLtJo/edit#slide=id.g2c0e7808fad_23_360

[cmsbuild]

cms-bot internal usage

[cmsbuild]

-code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44360/39400

• This PR adds an extra 20KB to repository

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

• code-format:
  https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44360/39400/code-format.patch
  e.g. curl -k https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44360/39400/code-format.patch | patch -p1
  You can also run scram build code-format to apply code format directly

[jfernan2]

assign heterogeneous

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44360/39401

• This PR adds an extra 20KB to repository

[cmsbuild]

New categories assigned: heterogeneous

@fwyzard,@makortel you have been requested to review this Pull request/Issue and eventually sign? Thanks

[cmsbuild]

A new Pull Request was created by @lecriste for master.

It involves the following packages:

• MagneticField/ParametrizedEngine (reconstruction)

@jfernan2, @fwyzard, @cmsbuild, @makortel, @mandrenguyen can you please review it and eventually sign? Thanks.
@namapane this is something you requested to watch as well.
@sextonkennedy, @antoniovilela, @rappoccio you are the release manager for this.

cms-bot commands are listed here

[namapane]

I'd like to add a warning here. This class provides a fast, rough approximation that is intended to be used within the tracker region only. The nominal region of validity is defined here:

https://github.com/cms-sw/cmssw/blob/master/MagneticField/ParametrizedEngine/src/ParabolicParametrizedMagneticField.cc#L44

It is generally a bad idea to use this approximation outside this region.

Also, please note that the header is placed in src on purpose, because this class is not intended to be used directly by other packages - this is just part of the implementation of the plugin that the package implements.
The MF is a service, which is served depending on the conditions (magnet current). Hardcoding a piece of implementation elsewhere will break this dependence, and one may at some point have a very hard time to figure out that some piece of code is using the wrong field for, e.g., runs at B=0 or if we ever have to take data at reduced magnet currents.

I would really advise to use the MF in the way it is designed and used in the rest of CMSSW, and not to make some pieces available in the public interface of the package.

[makortel]

@namapane

served depending on the conditions (magnet current). Hardcoding a piece of implementation elsewhere will break this dependence, and one may at some point have a very hard time to figure out that some piece of code is using the wrong field for, e.g., runs at B=0 or if we ever have to take data at reduced magnet currents.

Just to make sure I've understood correctly, are all of the conditions effectively encoded in the four parameters (c1, b0, b2, a) of the ParabolicParametrizedMagneticField?

Would this concern be satisfied if the parameters would have to be obtained from the magnetic field object in the EventSetup?

I would really advise to use the MF in the way it is designed and used in the rest of CMSSW, and not to make some pieces available in the public interface of the package.

The challenge is that the existing magnetic field infrastructure is not usable in GPU, which is what I believe @lecriste et al would need to do (in one way or another).

[lecriste]

The challenge is that the existing magnetic field infrastructure is not usable in GPU, which is what I believe @lecriste et al would need to do (in one way or another).

Right, and we verified that for the use-case of the e/gamma seeding, the parabolic approximation is suitable to be adopted at the ECAL boundary.

[cmsbuild]

-code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44360/39402

• This PR adds an extra 20KB to repository

• Found files with invalid states:

  • MagneticField/ParametrizedEngine/plugins/alpaka/ParabolicParametrizedMagneticField.h:
    • Added: 877e601
    • Modified: b138ff1
    • Deleted: 0146963

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

• code-format:
  https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44360/39402/code-format.patch
  e.g. curl -k https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44360/39402/code-format.patch | patch -p1
  You can also run scram build code-format to apply code format directly

[cmsbuild]

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-44360/41361

• There are other open Pull requests which might conflict with changes you have proposed:
  • File Configuration/Geometry/README.md modified in PR(s): Clean up Phase-2 Geometry D86, D88, D91, D92, D93, D94, D97 #45370
  • File Configuration/Geometry/python/dict2026Geometry.py modified in PR(s): Clean up Phase-2 Geometry D86, D88, D91, D92, D93, D94, D97 #45370
  • File Configuration/PyReleaseValidation/README.md modified in PR(s): Add run3 tracking low pu era #33532, CMSSW Integration of LST #45117, TICLv5 : Superclustering DNN #45333, [NOT TO MERGE] PR test DD4hep in 14_1 #45516, Add Phase2 PbPb reco process modifier #45683, Update to relval_gpu.pyfor Alpaka and Pixel Triplets Alpaka Workflows #45694
  • File Configuration/PyReleaseValidation/python/relval_2026.py modified in PR(s): CMSSW Integration of LST #45117, Clean up Phase-2 Geometry D86, D88, D91, D92, D93, D94, D97 #45370, Add Phase2 PbPb reco process modifier #45683
  • File Configuration/PyReleaseValidation/python/relval_gpu.py modified in PR(s): Update to relval_gpu.pyfor Alpaka and Pixel Triplets Alpaka Workflows #45694
  • File Configuration/PyReleaseValidation/python/upgradeWorkflowComponents.py modified in PR(s): Add run3 tracking low pu era #33532, CMSSW Integration of LST #45117, TICLv5 : Superclustering DNN #45333, Clean up Phase-2 Geometry D86, D88, D91, D92, D93, D94, D97 #45370, Allowing Jumping BPix2->FPix2 For Pixel Doublets (and Alpaka Pixel Triplets *.406 wf) #45478, Update Pixel GPU DQM online client #45666, Add Phase2 PbPb reco process modifier #45683, Update to relval_gpu.pyfor Alpaka and Pixel Triplets Alpaka Workflows #45694, Having Upgrade Matrix Failing for Duplicate Suffixes #45715
  • File Configuration/PyReleaseValidation/scripts/runTheMatrix.py modified in PR(s): Clean up Phase-2 Geometry D86, D88, D91, D92, D93, D94, D97 #45370
  • File Configuration/StandardSequences/python/GeometryConf.py modified in PR(s): Clean up Phase-2 Geometry D86, D88, D91, D92, D93, D94, D97 #45370
  • File Configuration/StandardSequences/python/Reconstruction_cff.py modified in PR(s): Add Phase2 PbPb reco process modifier #45683
  • File DQMOffline/Configuration/python/DQMOffline_cff.py modified in PR(s): Tau DQM miniAOD validation for DATA #44029
  • File DataFormats/Portable/interface/PortableCollection.h modified in PR(s): [RFC] Add tests and rudimentary protections for default-constructed PortableCollections #44844
  • File DataFormats/Portable/interface/PortableDeviceCollection.h modified in PR(s): [RFC] Add tests and rudimentary protections for default-constructed PortableCollections #44844
  • File DataFormats/Portable/interface/PortableHostCollection.h modified in PR(s): [RFC] Add tests and rudimentary protections for default-constructed PortableCollections #44844
  • File DataFormats/TrackingRecHitSoA/interface/TrackingRecHitsDevice.h modified in PR(s): Add specialisations of make_device_view with a Queue argument #45572
  • File FWCore/ParameterSet/src/Entry.cc modified in PR(s): Introduce vFloat and float types in ParameterSets #45216
  • File Geometry/CommonTopologies/interface/SimplePixelTopology.h modified in PR(s): Allowing Jumping BPix2->FPix2 For Pixel Doublets (and Alpaka Pixel Triplets *.406 wf) #45478
  • File HLTrigger/Configuration/python/HLT_75e33/modules/hgcalDigisL1Seeded_cfi.py modified in PR(s): add hlt (HLT) label to Phase-2 HLT modules (sequences) #45636
  • File HLTrigger/Configuration/python/HLT_75e33/modules/hltHpsPFTauDeepTauProducer_cfi.py modified in PR(s): add hlt (HLT) label to Phase-2 HLT modules (sequences) #45636
  • File HLTrigger/Configuration/python/HLT_75e33/modules/hltRechitInRegionsHGCAL_cfi.py modified in PR(s): add hlt (HLT) label to Phase-2 HLT modules (sequences) #45636
  • File HLTrigger/Configuration/python/HLT_75e33/modules/ticlSeedingL1_cfi.py modified in PR(s): add hlt (HLT) label to Phase-2 HLT modules (sequences) #45636
  • File HLTrigger/Configuration/python/HLT_75e33_cff.py modified in PR(s): add hlt (HLT) label to Phase-2 HLT modules (sequences) #45636
  • File HLTrigger/Configuration/python/customizeHLTforCMSSW.py modified in PR(s): Allowing Jumping BPix2->FPix2 For Pixel Doublets (and Alpaka Pixel Triplets *.406 wf) #45478
  • File HeterogeneousCore/AlpakaInterface/interface/memory.h modified in PR(s): Add specialisations of make_device_view with a Queue argument #45572
  • File HeterogeneousCore/AlpakaTest/plugins/alpaka/TestAlgo.dev.cc modified in PR(s): [RFC] Add {Copy,Move}ToDeviceCache<T> class templates and moveToDeviceAsync function template #43969
  • File HeterogeneousCore/AlpakaTest/plugins/alpaka/TestAlgo.h modified in PR(s): [RFC] Add {Copy,Move}ToDeviceCache<T> class templates and moveToDeviceAsync function template #43969
  • File L1Trigger/L1TCaloLayer1/plugins/L1TCaloSummary.cc modified in PR(s): Add L1TCaloLayer1Summary and necessary changes to emulator #45605
  • File L1Trigger/L1TCaloLayer1/python/simCaloStage2Layer1Summary_cfi.py modified in PR(s): Add L1TCaloLayer1Summary and necessary changes to emulator #45605
  • File L1Trigger/L1TGlobal/python/simGtStage2Digis_cfi.py modified in PR(s): Add ZDC EtSums to Stage-2 L1T emulation #45712
  • File L1Trigger/L1TTrackMatch/plugins/L1TrackJetEmulatorProducer.cc modified in PR(s): Phase 2 Bugfix: add the TkJetWord ValidBit to match Firmware #45065
  • File PhysicsTools/NanoAOD/plugins/rntuple/NanoAODRNTupleOutputModule.cc modified in PR(s): [RFC] descriptions.addDefault -> descriptions.addWithDefaultLabel #44372
  • File RecoBTag/FeatureTools/plugins/DeepBoostedJetTagInfoProducer.cc modified in PR(s): update PackedCandidate::dzError to return what it should in track parameterization #45612
  • File RecoHGCal/Configuration/python/RecoHGCal_EventContent_cff.py modified in PR(s): TICLv5 : Superclustering DNN #45333, add hlt (HLT) label to Phase-2 HLT modules (sequences) #45636
  • File RecoLocalTracker/SiPixelClusterizer/plugins/SiPixelRawToClusterGPUKernel.cu modified in PR(s): [DO NOT MERGE] remove duplicate pixels #37359
  • File RecoLocalTracker/SiPixelClusterizer/plugins/alpaka/SiPixelPhase2DigiToCluster.cc modified in PR(s): [RFC] Add tests and rudimentary protections for default-constructed PortableCollections #44844
  • File RecoLocalTracker/SiPixelClusterizer/plugins/alpaka/SiPixelRawToClusterKernel.dev.cc modified in PR(s): Add specialisations of make_device_view with a Queue argument #45572
  • File RecoLocalTracker/SiPixelClusterizer/plugins/gpuClusterChargeCut.h modified in PR(s): [DO NOT MERGE] remove duplicate pixels #37359
  • File RecoTauTag/RecoTau/interface/DeepTauIdBase.h modified in PR(s): update PackedCandidate::dzError to return what it should in track parameterization #45612
  • File RecoTracker/PixelSeeding/plugins/alpaka/CAHitNtupletGeneratorKernels.dev.cc modified in PR(s): Add specialisations of make_device_view with a Queue argument #45572
  • File Validation/Configuration/python/hgcalSimValid_cff.py modified in PR(s): TICLv5 : Superclustering DNN #45333
  • File Validation/SiTrackerPhase2V/plugins/Phase2OTHarvestTrackingParticles.cc modified in PR(s): Outer Tracker Phase2 DQM: additional stub validation plots #45475
  • File Validation/SiTrackerPhase2V/plugins/Phase2OTValidateTTStub.cc modified in PR(s): Outer Tracker Phase2 DQM: additional stub validation plots #45475
  • File Validation/SiTrackerPhase2V/plugins/Phase2OTValidateTrackingParticles.cc modified in PR(s): Outer Tracker Phase2 DQM: additional stub validation plots #45475
  • File Validation/SiTrackerPhase2V/python/Phase2TrackerValidationFirstStep_cff.py modified in PR(s): Outer Tracker Phase2 DQM: additional stub validation plots #45475

[lecriste]

Found files with invalid states:

• MagneticField/ParametrizedEngine/plugins/alpaka/ParabolicParametrizedMagneticField.h:

  • Added: 877e601
  • Modified: b138ff1
  • Deleted: 90c6128

@lecriste Would it be feasible to squash the commit history into one commit?

Done.

[cmsbuild]

Pull request #44360 was updated. @cmsbuild, @fwyzard, @jfernan2, @makortel, @mandrenguyen can you please check and sign again.

[makortel]

@cmsbuild, please test

[cmsbuild]

+1

Size: This PR adds an extra 12KB to repository
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-1b3cf3/40964/summary.html
COMMIT: 0828b85
CMSSW: CMSSW_14_1_X_2024-08-15-2300/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/44360/40964/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

• You potentially removed 3 lines from the logs
• Reco comparison results: 5 differences found in the comparisons
• DQMHistoTests: Total files compared: 45
• DQMHistoTests: Total histograms compared: 3422510
• DQMHistoTests: Total failures: 6
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 3422484
• DQMHistoTests: Total skipped: 20
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 44 files compared)
• Checked 196 log files, 165 edm output root files, 45 DQM output files
• TriggerResults: no differences found

[fwyzard]

+heterogeneous

[jfernan2]

+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 now be reviewed by the release team before it's merged. @rappoccio, @sextonkennedy, @mandrenguyen, @antoniovilela (and backports should be raised in the release meeting by the corresponding L2)

[mandrenguyen]

+1