update L3 filter, for 2017 online muon reconstruction #19465

sarafiorendi · 2017-06-28T17:56:09Z

Update of the L3 filter in order to:

handle muons reconstructed by the full (L3FromL2 + L3FromL1) iterL3 sequence, and perform the matching with the candidate from the previous step
introduce the possibility of disabling the matching of the L3 to the previous candidate (L2 or L1)

cmsbuild · 2017-06-28T17:56:28Z

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

It involves the following packages:

HLTrigger/Muon

@Martin-Grunewald, @silviodonato, @cmsbuild, @fwyzard, @davidlange6 can you please review it and eventually sign? Thanks.
@battibass, @abbiendi, @jhgoh, @Martin-Grunewald, @calderona, @HuguesBrun, @trocino this is something you requested to watch as well.
@davidlange6 you are the release manager for this.

cms-bot commands are listed here

fwyzard · 2017-06-28T18:04:07Z

HLTrigger/Muon/src/HLTMuonL3PreFilter.cc

 	  }
        } //MTL loop
+
+
+        if ( !( l1CandTag_ == edm::InputTag("")) && check_l1match && matchPreviousCand_){


can you use

if (not l1CandTag_.label().empty() ...

instead, to avoid creating InputTags every time ?

fwyzard · 2017-06-28T19:19:39Z

@cmsbuild , please test

cmsbuild · 2017-06-28T19:19:53Z

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/20973/console Started: 2017/06/28 21:20

cmsbuild · 2017-06-28T20:24:03Z

+1
Tested at: 3f74f56
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-19465/20973/summary.html

The following merge commits were also included on top of IB + this PR after doing git cms-merge-topic:
d598a4a
You can see more details here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-19465/20973/git-log-recent-commits
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-19465/20973/git-merge-result

cmsbuild · 2017-06-28T20:24:07Z

Comparison job queued.

cmsbuild · 2017-06-28T21:34:53Z

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-19465/20973/summary.html

Comparison Summary:

No significant changes to the logs found
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 22
DQMHistoTests: Total histograms compared: 1756063
DQMHistoTests: Total failures: 1231
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 1754666
DQMHistoTests: Total skipped: 166
DQMHistoTests: Total Missing objects: 0
Checked 90 log files, 14 edm output root files, 22 DQM output files

Martin-Grunewald · 2017-06-29T10:51:42Z

Presumably we need a PR for 92X as well - please make one!

Martin-Grunewald · 2017-06-30T11:47:59Z

This change kills off a few triggers (on 100 ttbar events) - new vs old:

HLT-Report ---------- Event  Summary ------------
HLT-Report Summary for Job
HLT-Report Events total = 100 wasrun = 100 passed = 100 errors = 0

HLT-Report ---------- HLTrig Summary ------------
HLT-Report   HLT #  WasRun     L1S     Pre     HLT   %L1sPre    Rate  RateHi  Errors Name

***************
*** 6204 ****
! HLT-Report      24     100     100      49       0   0.00000     0.0     0.0       0 HLT_Mu3_PFJet40_v8
--- 6261 ----
! HLT-Report      24     100     100      49      12  24.48980     0.0     0.0       0 HLT_Mu3_PFJet40_v8
***************
*** 6355 ****
! HLT-Report     175     100     100      49       0   0.00000     0.0     0.0       0 HLT_Mu8_TrkIsoVVL_v7
--- 6412 ----
! HLT-Report     175     100     100      49       8  16.32653     0.0     0.0       0 HLT_Mu8_TrkIsoVVL_v7
***************
*** 6362 ****
! HLT-Report     182     100     100      38       0   0.00000     0.0     0.0       0 HLT_Mu17_TrkIsoVVL_v7
--- 6419 ----
! HLT-Report     182     100     100      38       6  15.78947     0.0     0.0       0 HLT_Mu17_TrkIsoVVL_v7
***************
*** 6440,6441 ****
! HLT-Report     260     100     100      49       0   0.00000     0.0     0.0       0 HLT_Mu8_v7
! HLT-Report     261     100     100      38       0   0.00000     0.0     0.0       0 HLT_Mu17_v7
--- 6497,6498 ----
! HLT-Report     260     100     100      49       8  16.32653     0.0     0.0       0 HLT_Mu8_v7
! HLT-Report     261     100     100      38       6  15.78947     0.0     0.0       0 HLT_Mu17_v7
***************

eg, for HLT_Mu8, the event flow through all modules is

TrigReport     1   64         45         45          0          0 hltL3MuonsIterL3IO
TrigReport     1   65         45         45          0          0 hltIterL3MuonsFromL2LinksCombination
TrigReport     1   66         45         45          0          0 hltIterL3MuonsFromL2
TrigReport     1   67         45         45          0          0 hltIterL3FromL2MuonCandidates
TrigReport     1   68         45          0         45          0 hltL3fL1sMu5L1f0L2f5L3Filtered8
TrigReport     1   69          0          0          0          0 hltBoolEnd

where this is the killing filter:

process.hltL3fL1sMu5L1f0L2f5L3Filtered8 = cms.EDFilter( "HLTMuonL3PreFilter",

Please configure the default working of the new features to reproduce the old behaviour!

sarafiorendi · 2017-07-01T14:45:24Z

Hi @Martin-Grunewald ,

this is expected, and in particular it is due to the change here
https://github.com/cms-sw/cmssw/pull/19465/files#diff-6820f725f6447899fb4b62476d21b33cR214.
In short:

this filter is currently used only for paths where the "L3FromL2" muon sequence is used, which produces L3 candidates and their link collection
to correctly match each L3 candidate with its link, we should compare the globalTrack

The modification in this PR are meant to update the filter to be used after L3 candidates are produced using the full iterL3 sequence.
In this case, due to the way the sequence is built, we need to use the trackerTrack in order to correctly match the L3 candidates with their link.
The usage of the updated filter with the current menu will kill all the paths where the filter is used after the "L3FromL2" sequence.

To conclude, I can introduce another configurable parameter which would make the filter use the globalTrack until we finally switch to the new menu, where L3FromL2 is replaced by full iterL3 everywhere.
Would you agree?

thanks

fwyzard · 2017-07-01T16:33:00Z

If the new parameter is used only during the transition, I would rather not add it.

Instead, can we pick one of these options:

add a python customisation that adjusts this module's configuration until the migration, or
parse this in a private branch, and update the menu and filters in a single PR

.A

Martin-Grunewald · 2017-07-03T07:35:13Z

+1
OK, just that trigger proponents are aware - and have the added incentive to migrate!

cmsbuild · 2017-07-03T07:35:29Z

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 requires discussion in the ORP meeting before it's merged. @davidlange6, @smuzaffar

cmsbuild · 2017-07-04T20:24:13Z

Pull request #19465 was updated. @Martin-Grunewald, @silviodonato, @cmsbuild, @fwyzard, @davidlange6 can you please check and sign again.

fwyzard · 2017-07-04T21:34:52Z

@cmsbuild, please test

fwyzard · 2017-07-04T21:34:54Z

+1

cmsbuild · 2017-07-04T21:35:10Z

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-any-integration/21157/console Started: 2017/07/04 23:36

cmsbuild · 2017-07-04T21:35:11Z

This pull request is fully signed and it will be integrated in one of the next master IBs after it passes the integration tests. This pull request requires discussion in the ORP meeting before it's merged. @davidlange6, @smuzaffar

cmsbuild · 2017-07-04T22:13:30Z

+1
Tested at: ef4cedf
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-19465/21157/summary.html

The following merge commits were also included on top of IB + this PR after doing git cms-merge-topic:
e4d2a49
You can see more details here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-19465/21157/git-log-recent-commits
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-19465/21157/git-merge-result

cmsbuild · 2017-07-04T22:13:33Z

Comparison job queued.

cmsbuild · 2017-07-04T23:47:24Z

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-19465/21157/summary.html

Comparison Summary:

No significant changes to the logs found
Reco comparison results: 0 differences found in the comparisons
DQMHistoTests: Total files compared: 22
DQMHistoTests: Total histograms compared: 1760883
DQMHistoTests: Total failures: 30173
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 1730544
DQMHistoTests: Total skipped: 166
DQMHistoTests: Total Missing objects: 0
Checked 90 log files, 14 edm output root files, 22 DQM output files

davidlange6 · 2017-07-05T11:18:44Z

+1

sarafiorendi added 2 commits June 28, 2017 19:45

modify L3 filter to deal with the full iterL3 reconstruction

f900940

modify L3 filter to deal with the full iterL3 reconstruction

411f95d

cmsbuild added this to the CMSSW_9_3_X milestone Jun 28, 2017

cmsbuild added comparison-pending hlt-pending orp-pending pending-signatures tests-pending labels Jun 28, 2017

fwyzard reviewed Jun 28, 2017

View reviewed changes

BenjaminRS mentioned this pull request Jun 28, 2017

Update dimuon L3 filter, for 2017 online muon reconstruction #19466

Merged

avoid creating inputTag, just checking empty label

3f74f56

cmsbuild added tests-started and removed tests-pending labels Jun 28, 2017

cmsbuild added tests-approved and removed tests-started labels Jun 28, 2017

cmsbuild added comparison-available and removed comparison-pending labels Jun 28, 2017

sarafiorendi mentioned this pull request Jun 29, 2017

[Backport of PR19465] update L3 filter for 2017 online muon reconstruction for 92X #19479

Merged

cmsbuild added comparison-pending hlt-pending pending-signatures tests-pending and removed comparison-available fully-signed hlt-approved tests-approved labels Jul 4, 2017

cmsbuild added fully-signed hlt-approved and removed hlt-pending pending-signatures labels Jul 4, 2017

cmsbuild added tests-started and removed tests-pending labels Jul 4, 2017

cmsbuild added tests-approved and removed tests-started labels Jul 4, 2017

cmsbuild added comparison-available and removed comparison-pending labels Jul 4, 2017

cmsbuild added orp-approved and removed orp-pending labels Jul 5, 2017

cmsbuild merged commit 3ae8d8c into cms-sw:master Jul 5, 2017

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

update L3 filter, for 2017 online muon reconstruction #19465

update L3 filter, for 2017 online muon reconstruction #19465

sarafiorendi commented Jun 28, 2017

cmsbuild commented Jun 28, 2017

fwyzard Jun 28, 2017

fwyzard commented Jun 28, 2017

cmsbuild commented Jun 28, 2017 •

edited

cmsbuild commented Jun 28, 2017

cmsbuild commented Jun 28, 2017

cmsbuild commented Jun 28, 2017

Martin-Grunewald commented Jun 29, 2017

Martin-Grunewald commented Jun 30, 2017 •

edited

sarafiorendi commented Jul 1, 2017

fwyzard commented Jul 1, 2017

Martin-Grunewald commented Jul 3, 2017

cmsbuild commented Jul 3, 2017

cmsbuild commented Jul 4, 2017

fwyzard commented Jul 4, 2017

fwyzard commented Jul 4, 2017

cmsbuild commented Jul 4, 2017 •

edited

cmsbuild commented Jul 4, 2017

cmsbuild commented Jul 4, 2017

cmsbuild commented Jul 4, 2017

cmsbuild commented Jul 4, 2017

davidlange6 commented Jul 5, 2017

update L3 filter, for 2017 online muon reconstruction #19465

update L3 filter, for 2017 online muon reconstruction #19465

Conversation

sarafiorendi commented Jun 28, 2017

cmsbuild commented Jun 28, 2017

fwyzard Jun 28, 2017

Choose a reason for hiding this comment

fwyzard commented Jun 28, 2017

cmsbuild commented Jun 28, 2017 • edited

cmsbuild commented Jun 28, 2017

cmsbuild commented Jun 28, 2017

cmsbuild commented Jun 28, 2017

Martin-Grunewald commented Jun 29, 2017

Martin-Grunewald commented Jun 30, 2017 • edited

sarafiorendi commented Jul 1, 2017

fwyzard commented Jul 1, 2017

Martin-Grunewald commented Jul 3, 2017

cmsbuild commented Jul 3, 2017

cmsbuild commented Jul 4, 2017

fwyzard commented Jul 4, 2017

fwyzard commented Jul 4, 2017

cmsbuild commented Jul 4, 2017 • edited

cmsbuild commented Jul 4, 2017

cmsbuild commented Jul 4, 2017

cmsbuild commented Jul 4, 2017

cmsbuild commented Jul 4, 2017

davidlange6 commented Jul 5, 2017

cmsbuild commented Jun 28, 2017 •

edited

Martin-Grunewald commented Jun 30, 2017 •

edited

cmsbuild commented Jul 4, 2017 •

edited