Run II Reconstruction Program Status - Version p12

Overview

Major problems reading back the event created by d0reco

• Detector
• SMT - smaller cluster chunk size, apply correct Lorentz drift corrections
  
• CPS, FPS unpacking with better maps
• CAL - Offline zero suppression and non-linear corrections
  
• ID
• Electron - use em_util package for em scale corrections
• Improvements to thumbnail related packages
  

Performance

Upcoming patches, etc.

• bcjet_utilities - get the bcJetInitChunk correctly, otherwise wrong JetChunk will be used
  
• chpart_evt  - unpack x,y and z at DCA
• event_info - source for L2 bit mask changed. don't access the db for MC files, fix L1 initialisation bug
  
• hmreco - remove spurious print out
• muo_hit - update timing and PDT hit equality operator
• muon_raw - update for timing and PDT hit equality operator
• muonid - store track parameter and errors (5x5 matrix, for idnseg=3 get parameters from MuoCentralMatched), fix dynamic range for thumbnails
• silicon_geometry - improve creating map, reduces the memory consumption dramatically
• smt_d0hit - smaller 1D SMT hit chunks
• smt_hitalgs - smaller 1D SMT hit chunks
• smt_hitpackages -smaller 1D SMT hit chunks
• smt_lpos - smaller 1D SMT hit chunks
• smtutil - smaller 1D SMT hit chunks

Known Problems

Major error, Annoying feature, Functionality to be added
Scheduled for next release, Problem fixed.
(common = occurs within 100 events, rare = occurs within 1000 events)

Status	Description	Fixed in version
	d0om_ds - cannot read in the output file created by d0reco. Strongly suspected that the bug is in this package	p12.02.00
	reco_analyze - crashes in gtr_analyze for files reconstructed with p12. It works for files reconstructed with p10	p12.02.00-build3
	d0reco - serious memory leak, memory grows to 700 MB after 200 events
	d0reco - can't pick up correct calorimeter pedestal file needed for offline zero suppression	p12.02.00-build5
	reco_analyze - relocation error related to L3 dependency of cps_unpdata	p12.01.00
	d0reco - relocation error related to L3 dependency of cps_unpdata	p12.01.00

History of Major New Features

• Non-linear correction and off-line zero suppression for CAL
• Small SMT chunk sizes
  

Detailed history of changes in p12

• p12.02.00 build 5
• mc_analyze - check for invalid pointer
  
• muonid - impact parameter was set incorrectly. This bug affected only the reco-analyze output.
• run_config_mgr - Set RCP to use run # 0 for MC files. run_config_mgr_mc.rcp was set incorrectly to use the real run number.
• vertexreco - RCP made unique

• p12.02.00 build 4
• run_d0reco - copy files related to zero suppression and non-linearity corrections to rundata
• calunpdata - change RCP to point to the correct path
• jet_evt - a method was added to return the name of the jet algorithm type. This was needed for the tmb_tree to work correctly.

• p12.02.00 build 3
• bcjet_analyze - store run # and event #
• d0_analyze - add met_analyze and smt_lpos to LIBRARIES. Add smt_lpos_ref to OBJECTS, this is to resolve missing d0om type during running.
• d0reco - add smt_lpos_ref to OBJECTS
• reco_analyze - add smt_lpos to LIBRARIES. Add smt_lpos_ref to OBJECTS, this is to resolve missing d0om type during running.
• run_config_mgr - added dummy variables to RCP files to make them unique, some programs complained.
  
• semreco - include correct headers to fix multiply defined symbols
• vertexreco - use GTrack 401 for real data

• p12.02.00 build 1
• bc_muTagreco - bug fix for null pointer crash
  
• caladress - make Run1-only ICD channels invalid
  
• cal_elec - replace obsolete classes with those in unpack_cal_fe
• caltables - add method to calculate ADC (in float!) to GeV
• cal_unpdata - merge offline zero suppression and non-linearity corrections
• cps_unpdata - new CPS map
• mc_analyze - add information from output kine chunk, fix vertex TMB
• run_config_mgr - TC&C crate for calorimeter added
• unpack_cal_fe - new classes for unpacking TC&C crate

• p12.01.00 build 3
• chpart_reco - put chisq/dof in the chunk, otherwise it would be incompatible with earlier versions of charged particle chunk
• d0reco - don't drop SMT clusters
• em_analyze - bug fix for potential crash
• em_util - fix for compilation errors and now it will be used by em_reco
• em_evt - added component tests
• emreco - use em_util for EM energy scale correction. Fix a serious bug where hmchi2 is affected when a bad em candidate is removed
• hmreco - capability to remove eigen values strongly correlated with Log E and z vertex, protection against negative energy cluster
• muonid - fix for crashes
• reco_analyze - use METAnalyze blocks
• silicon_geometry - correct lorentz drift correction
• smt_d0hit
• smtdigi
• smt_hitalgs - correct lorentz drift correction, events generated with p10 always assumed -2T in d0sim, so if p11 is used to reconstruct the events, wrong drift correction may be applied
• smt_hitpackages
• vertexutil - Read in correct bfield
• vkalman_util - Read in correct bfield
• vertexreco - Magfield initialization in RCP for integrated test

• p12.01.00
• bcjet_evt - accomodation for thumbnail
  
• bcjet_eTagreco - recomputation for thumbnail quantities relegated to this package. Checks the validity of the EMparticle Link index.
• bcjet_muTagreco - same thing as above for muon tag
• caladdress - fix for ICD phi swap for some channels. Implement invalid ADC channels for translation from invalid inputs
• cft_obs - added HEADER.html for d0rte
• cps_unpdata - going back to p11 version. p12 version is dependent on L3 CPS unpacking.
• fps_unpdata - correct map and what not
  
• d0reco - add items to support d0rte (d0 runtime environment protocol), p11 RCPs added
• muo_segmentscintonly - new package added
• muon_raw - initialization for db access
• muonid - fix correlation matrix and thumbnail
  
• reco_analyze - updated RCPs, LIBRARIES and OBJECTS to include features in p11
  
• sftalign - support for copying files to rundata/sftalign as required by d0 run time environment
• tau_hmchisq - added items to work in the d0rte
• thumbnail - fixes to bcjet unpacking
  
• vertex_analyze - turn off V0 search due to timing considerations
  

This page maintained by Suyong Choi