thomreis / cmssw Goto Github PK

Generate ntuples from MP7 buffer dumps for commissioning at P5

translate halo bit in output muon quality

Since it is never written to an EDM file

move remaining developments to 800pre1

BXVector::end() seems to be only called if BXVector::size() was called before. Otherwise a random address value just appears causing the for loop where it is used to not reach the stop condition and ultimately leads to segmentation fault.

Internal muons are currently also put into the uGMT output muon collection.
They should go in a differnet collection

check if L1TBMTFConverter is still needed or if the bmtf sends the correct format now.

Remove hackConditions from DQM configurations

Add a static method to set the correct LorentzVector from hw values for Muon candidates.

Check efficiencies and rates if legacy RPC candidates are fed in the uGMT as additional inputs.

First needs a translator for the RPC muon format to RegionalMuonCand.
Then sort the muons in the MicroGMTProducers last stage.

For studies use RPC muons only if there is no compatible TF muon available.

Remove " without zero suppression" from
https://github.com/cms-l1t-offline/cmssw/blob/l1t-integration-CMSSW_9_0_0/DQM/L1TMonitor/src/L1TMP7ZeroSupp.cc#L95

Physical pT value is currently assigned as hwPt_0.5 but should be (hwPt-1)_0.5 since hwPt == 0 means no muon.

the phi of the RegionalMuonCand is in fact 8 bit signed not unsigned.
Fix the unpacker

Track address based cancel out for BMTF in emulator

Add TwinMux to emulator chain for ntuple production

In the firmware both isolated bits are set to 1 if there is no input from the calorimeters. in the emulator this is set to 0 apparently.

Implement EMTF track addresses in RegionalMuonCand RawDigi translator and change existing BMTF address scheme to the new one with four candidates.

Have to define data format first

Write a module to sort the ugmt input muons in the format we expect them to have correct inputs for the test pattern generation

The BMTF track segment selector bits (bits 23-26 in track address) should be stored in the RegionalMuonCand.

Index bits defining on which input link a muon was received should be set for output muons

only keep necessary modules in muon ntuple producer cfg file

https://github.com/cms-l1t-offline/cmssw/blob/l1t-muon-CMSSW_8_0_0_pre1/L1Trigger/L1TMuon/src/MicroGMTIsolationUnit.cc#L75

Charge is assigned the wrong way around. Should be (-1)^hwCharge

Need to find out if the format is the same as for uGMT output muons.

Add a variable to RegionalMuonCand that specifies the position of the muon in the readout block