Fastsim newrechits rebase sep4 #11121

lveldere · 2015-09-04T09:44:33Z

rebase of #11078

fastsim tracker rechit classes are replaced with
- a fastsim base class FastTrackerRecHit
- a fastsim class for single rechits FastSingleTrackerRecHit
  (accomodates the fastsim equivalents of SiPixelRecHit, SiStripRecHit2D and SiStripRecHit1D)
- a fastsim equivalent of SiStripMatchedRecHit2D, FastMatchedTrackerRecHit
- a fastsim equivalent of ProjectedSiStripRecHit2D, FastProjectedTrackerRecHit
trackerHitRTTI::RTTI is extended to accommodate the new rechit containers
a new implementation of FastTrackerRecHit::sharesInput
( will receive further study once merged)
rechit production is further modularised
- actual rechit production by the old SiTrackerGaussianSmearingRecHitConverter
- hit matching by new FastTrackerRecHitMatcher
- predefined combinations of hits by new FastTrackerRecHitCombiner
SiTrackerGaussianSmearingRecHitConverter produces now
OwnVector
vector<\Ref > with one entry per SimHit to find for each SimHit the corresponding RecHit
predefined combinations of hits are now in a new, simple format:
vector<vector<Ref > >
model for hit splitting has been improved and moved to a separate class FastTrackerRecHitSplitter
model for masking of hits in iterative tracking improved
(see HitMaskHelper and FastTrackerRecHitMaskProducer)
GlobalPixelTracking is not fully migrated yet (it produces zero result):
it was not properly configured to begin with,
and we need to study the fullsim equivalent before fixing it
various code files were updated for the new rechit classes and the new rechit combination format
y position of strip hits is set to 0 (before it was the simhit y-position)
unit tests: a few fastsim events are generated on the fly, such to circumvent the backward incompatibility introduced by this pr. Unit tests to be restored when new relvals are available.

lveldere · 2015-09-04T09:44:39Z

please test

cmsbuild · 2015-09-04T09:46:13Z

The tests are being triggered in jenkins.

cmsbuild · 2015-09-04T09:46:14Z

A new Pull Request was created by @lveldere for CMSSW_7_6_X.

Fastsim newrechits rebase sep4

It involves the following packages:

CommonTools/RecoAlgos
DataFormats/TrackerRecHit2D
FastSimulation/Configuration
FastSimulation/Muons
FastSimulation/Tracking
FastSimulation/TrackingRecHitProducer
PhysicsTools/PatAlgos
RecoJets/JetAssociationProducers
RecoMET/METProducers
RecoParticleFlow/PFSimProducer
RecoTracker/TransientTrackingRecHit
SLHCUpgradeSimulations/Geometry
SimGeneral/MixingModule
SimTracker/TrackerHitAssociation

@civanch, @Dr15Jones, @lveldere, @cvuosalo, @ianna, @mdhildreth, @monttj, @cmsbuild, @ssekmen, @slava77, @vadler can you please review it and eventually sign? Thanks.
@ghellwig, @TaiSakuma, @abbiendi, @rappoccio, @Martin-Grunewald, @istaslis, @threus, @mmarionncern, @imarches, @makortel, @acaudron, @jhgoh, @lgray, @jdolen, @ferencek, @cerati, @schoef, @wmtan, @GiacomoSguazzoni, @rovere, @VinInn, @nhanvtran, @wmtford, @mschrode, @dgulhan, @ahinzmann, @gpetruc, @matt-komm, @mariadalfonso, @pvmulder, @bachtis this is something you requested to watch as well.
You can sign-off by replying to this message having '+1' in the first line of your reply.
You can reject by replying to this message having '-1' in the first line of your reply.
If you are a L2 or a release manager you can ask for tests by saying 'please test' in the first line of a comment.
@Degano you are the release manager for this.
You can merge this pull request by typing 'merge' in the first line of your comment.

cmsbuild · 2015-09-04T12:37:04Z

+1
Tested at: 4567eb1
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-11121/7891/summary.html

cmsbuild · 2015-09-04T13:58:14Z

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-11121/7891/summary.html

lveldere · 2015-09-04T15:52:34Z

+1

fullsim comparisons look okay
fastsim changes are expected

@civanch, @Dr15Jones, @cvuosalo, @ianna, @mdhildreth, @monttj, @slava77, @vadler
It would be extremely helpful if you could treat this with some priority.
I only have a few working days left before holidays...

civanch · 2015-09-07T16:40:22Z

+1

slava77 · 2015-09-08T23:31:57Z

FastSimulation/Tracking/plugins/TrackCandidateProducer.cc

+    int32_t icomb = fastTrackingHelper::getRecHitCombinationIndex(seed);
+    if(icomb < 0 || unsigned(icomb) >= recHitCombinations->size()){
+	edm::LogError("TrackCandidateProducer") << " found seed with recHitCombination out or range: " << icomb << std::endl;
+	exit(1);


this should be removed and replaced with an exception

slava77 · 2015-09-09T00:36:47Z

+1

code changes are mostly OK
- *exit(1) system calls should still be removed * (this can be done in a separate PR)
- fullsim/data reco is not affected
jenkins test pass and comparisons with baseline show differences only in the fastsim workflows
I scrolled through a large fraction of comparison plots in 135.4 (ZEE fastsim) and the distributions changed mainly in the tracking algorithm side and less so in the general track or other variables properties
logfiles still suggest there is a problem in the TrajectorySeedProducer:pixelTripletSeeds:
this message changed from something that appears frequently and is filling up the log file

Either region producer, or the seed creator cfg is not available.

to something that shows up less frequently

 RegionFactory is not initialised properly, please check your configuration. Producing empty seed collection

lveldere · 2015-09-09T08:35:48Z

@slava77
Thanks for the comments!
I will get rid of the exits as soon as this pr gets merged.
About the log files: this is a known issue in the HLT paths.
I have a branch with a fix, and I will make a pr when this one gets merged.

Fastsim newrechits rebase sep4

Dr15Jones · 2015-09-11T14:06:33Z

@davidlange6 The removal of the class SiTrackerGSMatchedRecHit2D is causing the IB RelVals to fail with the message

----- Begin Fatal Exception 11-Sep-2015 10:25:21 CEST-----------------------
An exception of category 'FileReadError' occurred while
   [0] Processing run: 1 lumi: 1 event: 1
   [1] Running path 'datamixing_step'
   [2] Calling event method for module DataMixingModule/'mixData'
   [3] Reading branch TrackingRecHitsOwned_mix_generalTracks_DIGI2RAW.
   Additional Info:
      [a] Fatal Root Error: @SUB=TBufferFile::ReadObject
trying to read an emulated class (SiTrackerGSMatchedRecHit2D) to store in a compiled pointer (TrackingRecHit)

----- End Fatal Exception -------------------------------------------------

see
https://cmssdt.cern.ch/SDT/cgi-bin/buildlogs/slc6_amd64_gcc493/CMSSW_7_6_X_2015-09-10-2300/pyRelValMatrixLogs/run/250400.0_ZEE_13+FS_ZEE_13_PRMXUP15_PU25+HARVESTUP15FS+MINIAODMCUP15FS/step1_ZEE_13+FS_ZEE_13_PRMXUP15_PU25+HARVESTUP15FS+MINIAODMCUP15FS.log

Lukas added 19 commits September 4, 2015 09:59

introduce new FastTrackerRecHit classes

c271b36

migrate FastSimulation/TrackingRecHitProducer

b7a2162

add getter for recHitCombinationIndex

e8bddab

migrate FastSimulation/Tracking

2064fb3

migrate CommonTools/RecoAlgos/src/ClusterStorer.cc

217e10b

migrate FastSimulation/Configuration/python/Reconstruction_BefMix_cff.py

50dcba5

migrate FastSimulation/Muons

6d25d6c

migrate RecoJets/JetAssociationProducers

9951486

migrate RecoTracker/TransientTrackingRecHit

c682f43

migrate SimTracker/TrackerHitAssociation

0c106d4

migrate Adjuster

f4da151

migrate PFSimParticleProducer

32c62f6

migrate ntuplizers

be065ef

little fix to FastTrackerRecHit

84e8413

various fixes/improvements

b04be94

several fixes after rebase

b7b7c3d

shrink rechit combinations, replace exit with cms exception

9ba36e4

fix unit tests by producing fastsim on the fly (temporary)

0b9ae0f

fix after rebase

4567eb1

cmsbuild added this to the Next CMSSW_7_6_X milestone Sep 4, 2015

cmsbuild added reconstruction-pending simulation-pending analysis-pending fastsim-pending geometry-pending pending-signatures tests-pending orp-pending labels Sep 4, 2015

cmsbuild added the comparison-pending label Sep 4, 2015

cmsbuild added tests-started and removed tests-pending labels Sep 4, 2015

cmsbuild added tests-approved and removed tests-started labels Sep 4, 2015

cmsbuild added comparison-available and removed comparison-pending labels Sep 4, 2015

slava77 mentioned this pull request Sep 4, 2015

FastSim: new rechits containers #11078

Closed

cmsbuild added fastsim-approved and removed fastsim-pending labels Sep 4, 2015

cmsbuild added geometry-approved simulation-approved and removed geometry-pending simulation-pending labels Sep 7, 2015

slava77 reviewed Sep 8, 2015
View reviewed changes

cmsbuild added reconstruction-approved and removed reconstruction-pending labels Sep 9, 2015

davidlange6 added a commit that referenced this pull request Sep 9, 2015

Merge pull request #11121 from lveldere/fastsim-newrechits-rebaseSep4

070bdb6

Fastsim newrechits rebase sep4

davidlange6 merged commit 070bdb6 into cms-sw:CMSSW_7_6_X Sep 9, 2015

lveldere mentioned this pull request Sep 9, 2015

FastSim, use cms::Exception rather than logError + exit #11201

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fastsim newrechits rebase sep4 #11121

Fastsim newrechits rebase sep4 #11121

lveldere commented Sep 4, 2015

lveldere commented Sep 4, 2015

cmsbuild commented Sep 4, 2015

cmsbuild commented Sep 4, 2015

cmsbuild commented Sep 4, 2015

cmsbuild commented Sep 4, 2015

lveldere commented Sep 4, 2015

civanch commented Sep 7, 2015

slava77 Sep 8, 2015

slava77 commented Sep 9, 2015

lveldere commented Sep 9, 2015

Dr15Jones commented Sep 11, 2015

Fastsim newrechits rebase sep4 #11121

Fastsim newrechits rebase sep4 #11121

Conversation

lveldere commented Sep 4, 2015

lveldere commented Sep 4, 2015

cmsbuild commented Sep 4, 2015

cmsbuild commented Sep 4, 2015

cmsbuild commented Sep 4, 2015

cmsbuild commented Sep 4, 2015

lveldere commented Sep 4, 2015

civanch commented Sep 7, 2015

slava77 Sep 8, 2015

Choose a reason for hiding this comment

slava77 commented Sep 9, 2015

lveldere commented Sep 9, 2015

Dr15Jones commented Sep 11, 2015