Updates L1T OMTF for Phase 2 [17_0_X]#51296
Merged
cmsbuild merged 16 commits intoJun 30, 2026
Merged
Conversation
Changes from https://github.com/kbunkow/cmssw/tree/from-CMSSW_14_2_0_pre2_KB_v1/L1Trigger/L1TMuonOverlapPhase1 Merged changes from the CMSSW master L1TMuonOverlapPhase1: OMTFConfiguration.h Renaming and modifying several methods (e.g. for eta and phi conversions), new methods, mb1W2Eta(), mb2W2Eta(), etc. Added parameter usePhase2DTPrimitives. AlgoMuon added quality, dxyNN and nnOutputs Added class FinalMuon that stores the candidate parameters in the output format. IOMTFEmulationObserver::observeProcesorEmulation FinalMuons instead of RegionalMuonCands in the argument list OMTFProcessor: run() returns now FinalMuons instead of RegionalMuonCands. Added std::function<void(AlgoMuons::value_type& algoMuon)> assignQuality, which is set to assignQualityPhase1 or assignQualityPhase2 Methods OMTFProcessor::convertToGmtScalesPhase1 and OmtfEmulation::convertToGmtScalesPhase2 set the values in the finalMuon, then the values from the finalMuon are set in l1t::RegionalMuonCand (in getRegionalMuonCands) or SAMuon (in OmtfEmulation::run), respectively. Added method OMTFProcessor<GoldenPatternType>::assignQualityPhase1 assignQuality is called in OMTFProcessor<GoldenPatternType>::run. Adde option useStubQualInExtr - it refers to the ref hit quality. targetStubR is used in extrapolateDtPhiBFixedPoint, instead of targetStubEta CscDigiToStubsConverter::makeStubs Dumping cscDigis to xml. GhostBusterPreferRefDt.cc: added customByRefLayerAndHitQual sorting function, for which the quality of the DT ref hit is the first criterion, not used a.t.m.eicihdebihcfnvlecdgjbekcdnekdlegevvbevejrguf GoldenPatternBase and GoldenPattern::propagateRefPhi adedd iLayer as an argument StubResult.h - added field deltaPhi that stores: deltaPhi = hitPhi - extrapolatedPhi[iStub] - phiRefHit in GoldenPatternBase::process1Layer1RefLayer and is dumped in DataROOTDumper2::observeEventEnd as hit.phiDist . CandidateSimMuonMatcher Removing OMTFConfiguration so that it can be used in the analyzer outside the OMTF package. Added MatchingType enum. reworked CandidateSimMuonMatcher::match. MatchingType::simplePropagation is now propagating the muon to the muon station with the use of the provided histogram of medianDelta, minDelta, and maxDelta. Added method CandidateSimMuonMatcher::propagate which does both propagation withPropagator and simplePropagation. CandidateSimMuonMatcher::propagate: when propagation fails, doSimplePropagation is called for muonRho < 40 added CandidateSimMuonMatcher::collectMuonCands - no matching, just collecting the MuonCands, useful e..g. in minBIas sample, when there is no gen info. match(const l1t::RegionalMuonCand* omtfCand, const AlgoMuonPtr& procMuon, MatchingResult& result, TrajectoryStateOnSurface& tsof) - is used both for the simeTrack and trackingParticle. Added method atStation1. Deleting deltaPhiPropCandMean and deltaPhiPropCandStdDev in destructor. ghostBust - increasing deltaPhi margin from 8 to 10, to reduce the ghost for the displaced muons. match - result.matchingLikelihood = 1. / ( std::abs(result.deltaPhi) + 0.001) as normal_pdf geves nan for low pt muons. mean set to 0 setting result.propagatedPhi and propagatedEta. DataROOTDumper2 Setting omtfEvent.muonEvent = -2 when ResultType::propagationFailed. Fixed setting of omtfEvent.omtfPt. Added rootTree->Write() in DataROOTDumper2::endJob() as it seems it fixes the job crashes. Added vertexEta and vertexPhi. Dumping hit.etaHw instead of hit.z, deltaR insted of eta L1TMuonOverlapPhase2: InputMakerPhase2.h: using L1Phase2MuDTThDigi in the OMTF phase2 algorithm merged from https://github.com/folguera/cmssw/tree/omtf_dtphase2theta_cmssw14_0_0_pre3 DtPhase2DigiToStubsConverterOmtf::addDTphiDigi: R is set for the DT segments. implemented OmtfPhase2AngleConverter::getGlobalEta(). OMTFReconstruction Constructor - using edmParameterSet.copyForModify, as it was suspected that ordinary copying of parameterSet was sometimes causing crashes. FinalMuons are used in reconstruct() PtAssignmentBase Renamed getPts to run. EmulationObserverBase.h changing ParameterSet to reference L1TMuonOverlapPhase2TrackProducer::produce - produces SAMuonCollection OmtfEmulation: Added OmtfEmulation::getQualityFromFiredLayers with firedLayersToQuality tuned for the phase2 OMTF patterns. Added OmtfEmulation::convertToGmtScalesPhase2(). Added OmtfEmulation::getSAMuons(). LutNetworkFixedPointRegressionMultipleOutputs - the new version of the neural network, with multiple outputs. The neurons of the last layer are not fully connected to the last-but-one layer. PtAssignmentNNRegression adapted to the LutNetworkFixedPointRegressionMultipleOutputs with lutNN_omtfRegression_v430_FP.xml omtfHitWithQualAndRToEventInput, OmtfHit now has r instead of eta, no pt calibration, the second pt returned by the NN is not used yet hit.z is not used in OmtfHit deltaR instad of r LutNeuronLayerFixedPoint.h - in floorlog2 and ceillog2 removed throw cms::Exception as it should be const expressions, so runtime throws have no sense. Added template parameter output_F
added valueP2Scale i.e. option ot use the phase-2 GMT eta scale IProcessorEmulator removed setAssignQualityFunction. OMTFProcessor.h assignQualityPhase1 changed to assignQuality. Removed ptAssignment. PatternGenerator.cc removed reCalibratePt(). PtAssignmentBase renamed to MlModelBase and moved to L1Trigger/L1TMuonOverlapPhase2. PtAssignmentNNRegression.h renamed to NNRegression.h. Added class OmtfProcessorPhase2. OmtfEmulation added method beginRun. A few methods and fields moved to OmtfProcessorPhase2. L1Trigger/L1TMuonOverlapPhase2/python/simOmtfPhase2Digis_cfi.py hwToLogicLayer_0x0210.xml - added ref hits from the processors' overlap regions, to improve the efficiency at the processor boundaries. Patterns_ExtraplMB1andMB2RFixedP_ValueP1Scale_DT_2_2_2_t35__classProb17.xml - pattern without pt recalibration, besides the same as the previous version (i.e. Patterns_ExtraplMB1andMB2RFixedP_ValueP1Scale_DT_2_2_2_t35__classProb17_recalib2.xml).
AngleConverterBase was removed, the needed methods were moved to OmtfAngleConverter. A few methods that were not used were dropped. OmtfPhase2AngleConverter method getGlobalEta renamed to getGlobalEtaPhase2, to avoid hiding OmtfAngleConverter methods getGlobalEta.
added << operator to FinalMuon and MatchingResult CandidateSimMuonMatcher.cc Added protections against null simVertex pointer. Added protections against null muonMatcherFile and histograms. Removed cuts on candidates pt and quality. Simplified LogTrace printouts. OmtfProcessorPhase2.cc A small fix in the assignQualityPhase2.
ProcConfigurationBase.h: added dtBxShift. InputMakerPhase2.h: OMTFConfiguration moved from DtPhase2DigiToStubsConverterOmtf to DtPhase2DigiToStubsConverter. InputMakerPhase2.cc: Using config->dtBxShift() instead of the hardcoded value 20.
…of cast from float to int
The bug has had no impact as it was part of the final, redundant condition in an "else if" statement
e.g. better use of const, cout=>MessageLogs, additional comments, etc.
OMTFProcessor.cc and OmtfProcessorPhase2.cc Fixing the condition to assign minimum pT (0 or 1 GeV) for the muons that have only unconstrained pt valid. Checking getFiredLayerCntConstr was missing. updating runMuonOverlap_run3_mc.py and runMuonOverlap.py
RegionalMuonCandBxCollection is no longer produced by L1TMuonOverlapPhase2TrackProducer, RegionalMuonCand are not created by the OmtfEmulation. Only SAMuonCollection (OMTFconstr and OMTFunconstr) are produced now. L1Trigger/Phase2L1GMT/plugins/Phase2L1TGMTFwdMuonTranslator.cc was modified to avoid using RegionalMuonCand and instead use SAMuon from OMTF. L1Trigger/L1TMuonOverlapPhase1/src/Omtf/OMTFConfiguration.cc: fix in procPhiOmtfToGlobalPhiOmtf
Contributor
|
cms-bot internal usage |
Contributor
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51296/49907 |
Contributor
|
A new Pull Request was created by @mbluj for CMSSW_17_0_X. It involves the following packages:
@BenjaminRS, @cmsbuild, @quinnanm can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
Contributor
|
please test |
Contributor
|
-1 Failed Tests: RelVals Failed RelVals |
Contributor
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51296/50025 |
Contributor
|
Pull request #51296 was updated. @BenjaminRS, @cmsbuild, @quinnanm can you please check and sign again. |
Contributor
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-51296/50028 |
Contributor
|
Pull request #51296 was updated. @BenjaminRS, @cmsbuild, @quinnanm can you please check and sign again. |
Contributor
|
please test |
Contributor
|
+1 Size: This PR adds an extra 32KB to repository Comparison SummarySummary:
Max Memory Comparisons exceeding threshold@cms-sw/core-l2 , I found 1 workflow step(s) with memory usage exceeding the error threshold: Expand to see workflows ...
|
Contributor
|
+l1 |
cmsbuild
added
l1-approved
fully-signed
and removed
l1-pending
pending-signatures
labels
Contributor
|
This pull request is fully signed and it will be integrated in one of the next CMSSW_17_0_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_20_1_X is complete. This pull request will now be reviewed by the release team before it's merged. @sextonkennedy, @mandrenguyen, @ftenchini (and backports should be raised in the release meeting by the corresponding L2) |
Contributor
|
+1 |
This was referenced Jul 1, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PR description:
This PR is a backport of #50032 to 17_0_X and introduces recent (as for beginning of 2026) changes in the L1T OMTF emulator for Phase2:
SAMuons(Phase 2 data format):FinalMuonwas introduced that stores the candidate parameters in the output format. It can be either Phase 1 or Phase 2 scales. Then it is converted toRegionalMuonCands(Phase 1) orSAMuons(Phase 2);CandidateSimMuonMatcherandDataROOTDumper2used for developments and validations.Details in this JIRA ticket and K. Bunkowski's presentation here.
This PR requires updated configuration files provided in cms-data/L1Trigger-L1TMuon#29.
PR validation:
Validated with Phase 2 workflows, e.g. 29634.0, 29634.78, 36034.0
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:
It is a backport of #50032 to 17_0_X, which is required for the L1T Annual Review in September. This L1T review will use Spring24 MC, which is incompatible with 20_X_Y.
FYI, @akalinow, @kbunkow