possibility of configuring the OMTF emulator directly from XMLs #45032

kbunkow · 2024-05-23T18:10:24Z

PR description:

This PR adds the possibility of configuring the OMTF emulator directly from the config XML, without using the L1TMuonOverlapPhase1ParamsESProducer. It should allow to avoid problems with the OMTF emulator that appeared in the #44498 - as discussed in the #44498 (comment).

PR validation:

The PR was validated with the workflow 141.044 (runTheMatrix.py -l 141.044 -i all --ibeos). The OMTF plots look good.

This PR is not a backport.

Details of changes:

OMTFReconstruction.h and OMTFReconstruction.cc
The possibility of configuring the OMTF emulator directly from the config XML is added. When both patternsXMLFile and configXMLFile are given from the python config of the simOmtfDigis or simOmtfPhase2Digis, the L1TMuonOverlapParams are read directly from these files, and the L1TMuonOverlapPhase1ParamsESProducer is not needed. Otherwise, the L1TMuonOverlapParams are obtained from the eventSetup (so from the DB, or from the L1TMuonOverlapPhase1ParamsESProducer - if it is defined).

L1Trigger/L1TMuonOverlapPhase2/python/fakeOmtfParamsPhase2_cff.py is removed, as when it is used, the L1TMuonOverlapPhase1 gets this config as well.

L1Trigger/L1TMuonOverlapPhase2/python/simOmtfPhase2Digis_extrapol_cfi.py is moved to
simOmtfPhase2Digis_cfi.py, as this is the default configuration (besides, there should be only one cfi for a given module).

L1Trigger/L1TMuonOverlapPhase1/src/Omtf/OMTFProcessor.cc line
trackAddr[3] = myCand->getGpResultUnconstr().getPdfSumUnconstr(); is removed, as uGMT expects 3 entries in the TrackAddress (when there were 4, uGMT was generating a warning).

L1Trigger/L1TMuonOverlapPhase2/test - runMuonOverlap*.py are corrected

OMTFReconstruction.h and OMTFReconstruction.cc The possibility of configuring the OMTF emulator directly from the config XML is added. When both patternsXMLFile and configXMLFile are given from the python config of the simOmtfDigis or simOmtfPhase2Digis, the L1TMuonOverlapParams are read directly from these files, and the L1TMuonOverlapPhase1ParamsESProducer is not needed. Otherwise, the L1TMuonOverlapParams are obtained from the eventSetup (so from the DB, or from the L1TMuonOverlapPhase1ParamsESProducer - if it is defined). L1Trigger/L1TMuonOverlapPhase2/python/fakeOmtfParamsPhase2_cff.py is removed, as when it is used, the L1TMuonOverlapPhase1 gets this config as well. L1Trigger/L1TMuonOverlapPhase2/python/simOmtfPhase2Digis_extrapol_cfi.py is moved to simOmtfPhase2Digis_cfi.py, as this is the default configuration (besides, there should be only one cfi for a given module). L1Trigger/L1TMuonOverlapPhase1/src/Omtf/OMTFProcessor.cc line trackAddr[3] = myCand->getGpResultUnconstr().getPdfSumUnconstr(); is removed, as uGMT expects 3 entries in the TrackAddress (when there were 4, uGMT was generating a warning). L1Trigger/L1TMuonOverlapPhase2/test - runMuonOverlap*.py are corrected

cmsbuild · 2024-05-23T18:10:51Z

cms-bot internal usage

cmsbuild · 2024-05-23T18:16:01Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-45032/40324

This PR adds an extra 40KB to repository
There are other open Pull requests which might conflict with changes you have proposed:
- File L1Trigger/L1TMuonOverlapPhase1/src/Omtf/OMTFProcessor.cc modified in PR(s): Set 3 instead of 4 OMTF track sub-address as expected by uGMT #45030

cmsbuild · 2024-05-23T18:16:24Z

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

It involves the following packages:

L1Trigger/L1TMuonOverlapPhase1 (l1)
L1Trigger/L1TMuonOverlapPhase2 (upgrade, l1)

@aloeliger, @epalencia, @srimanob, @subirsarkar, @cmsbuild can you please review it and eventually sign? Thanks.
@JanFSchulte, @thomreis, @Martin-Grunewald, @dinyar, @eyigitba, @missirol this is something you requested to watch as well.
@rappoccio, @sextonkennedy, @antoniovilela you are the release manager for this.

cms-bot commands are listed here

epalencia · 2024-05-24T11:04:52Z

Please test

cmsbuild · 2024-05-24T14:11:33Z

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-6a0d5a/39512/summary.html
COMMIT: 948fd54
CMSSW: CMSSW_14_1_X_2024-05-24-1100/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmssw/45032/39512/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

You potentially removed 59 lines from the logs
Reco comparison results: 6 differences found in the comparisons
DQMHistoTests: Total files compared: 48
DQMHistoTests: Total histograms compared: 3338862
DQMHistoTests: Total failures: 60
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 3338782
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 47 files compared)
Checked 202 log files, 165 edm output root files, 48 DQM output files
TriggerResults: no differences found

artlbv · 2024-05-27T09:23:02Z

Interestingly, this PR now also shows some differences in the Run3 OMTFoutput_vs_uGMTinput DQM, even though when tested alone this PR does not add the new Phase-2 OMTF into the L1 sequence yet.

https://cmssdt.cern.ch/SDT/jenkins-artifacts/baseLineComparisons/CMSSW_14_1_X_2024-05-24-1100+6a0d5a/62881/141.046_RunEGamma2023D/L1T_L1TStage2uGMT_OMTFoutput_vs_uGMTinput.html

Is this expected? @kbunkow @mbluj

kbunkow · 2024-05-27T17:33:46Z

Interestingly, this PR now also shows some differences in the Run3 OMTFoutput_vs_uGMTinput DQM, even though when tested alone this PR does not add the new Phase-2 OMTF into the L1 sequence yet.

https://cmssdt.cern.ch/SDT/jenkins-artifacts/baseLineComparisons/CMSSW_14_1_X_2024-05-24-1100+6a0d5a/62881/141.046_RunEGamma2023D/L1T_L1TStage2uGMT_OMTFoutput_vs_uGMTinput.html

Is this expected? @kbunkow @mbluj

The difference in the L1T_L1TStage2uGMT_OMTFoutput_vs_uGMTinput plots is due to a change in the
L1Trigger/L1TMuonOverlapPhase1/src/Omtf/OMTFProcessor.cc, i.e. removed line
trackAddr[3] = myCand->getGpResultUnconstr().getPdfSumUnconstr();

This line was added by #43687, but it turned out it is not needed, and moreover, uGMT was generating this warning message when the fourth trackAddr was added:

OMTF muon track address map contains 4 instead of the expected 3 subaddresses. Check the data format. Setting track address to 0

The track address map is checked only by the plots L1T_L1TStage2uGMT_OMTFoutput_vs_uGMTinput , therefore the differences are visible only in these plots.

So the fix in this PR in the OMTFProcessor.cc is OK, as well as the changes in the L1T_L1TStage2uGMT_OMTFoutput_vs_uGMTinput plots.

Karol

cmsbuild · 2024-05-28T16:56:42Z

+code-checks

Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-45032/40399

This PR adds an extra 40KB to repository

cmsbuild · 2024-05-28T16:57:04Z

Pull request #45032 was updated. @epalencia, @subirsarkar, @aloeliger, @srimanob, @cmsbuild can you please check and sign again.

aloeliger · 2024-05-29T09:49:14Z

please test

cmsbuild · 2024-05-29T12:51:05Z

+1

Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-6a0d5a/39600/summary.html
COMMIT: 83e721b
CMSSW: CMSSW_14_1_X_2024-05-28-2300/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmssw/45032/39600/install.sh to create a dev area with all the needed externals and cmssw changes.

Comparison Summary

Summary:

You potentially removed 1 lines from the logs
Reco comparison results: 4 differences found in the comparisons
DQMHistoTests: Total files compared: 48
DQMHistoTests: Total histograms compared: 3338862
DQMHistoTests: Total failures: 0
DQMHistoTests: Total nulls: 0
DQMHistoTests: Total successes: 3338842
DQMHistoTests: Total skipped: 20
DQMHistoTests: Total Missing objects: 0
DQMHistoSizes: Histogram memory added: 0.0 KiB( 47 files compared)
Checked 202 log files, 165 edm output root files, 48 DQM output files
TriggerResults: no differences found

aloeliger · 2024-06-04T12:14:43Z

+l1

srimanob · 2024-06-07T14:24:04Z

Note to @cms-sw/l1-l2 for later fix PR

src/L1Trigger/L1TMuonOverlapPhase1/src/Tools/CandidateSimMuonMatcher.cc:444:12: warning: Value stored to 'treshold' during its initialization is never read [deadcode.DeadStores]
444 | double treshold = 6. * sigma;
| ^~~~~~~~ ~~~~~~~~~~
src/L1Trigger/L1TMuonOverlapPhase1/src/Tools/CandidateSimMuonMatcher.cc:446:7: warning: Value stored to 'treshold' is never read [deadcode.DeadStores]
446 | treshold = 7. * sigma;
| ^ ~~~~~~~~~~
src/L1Trigger/L1TMuonOverlapPhase1/src/Tools/CandidateSimMuonMatcher.cc:448:7: warning: Value stored to 'treshold' is never read [deadcode.DeadStores]
448 | treshold = 20. * sigma;
| ^ ~~~~~~~~~~~
3 warnings generated.

srimanob · 2024-06-07T14:24:10Z

+Upgrade

cmsbuild · 2024-06-07T14:24:27Z

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 will now be reviewed by the release team before it's merged. @sextonkennedy, @antoniovilela, @rappoccio (and backports should be raised in the release meeting by the corresponding L2)

rappoccio · 2024-06-12T01:32:59Z

+1

antoniovilela · 2024-06-12T11:43:15Z

ping bot

cmsbuild added this to the CMSSW_14_1_X milestone May 23, 2024

cmsbuild added l1-pending pending-signatures tests-pending orp-pending upgrade-pending code-checks-pending labels May 23, 2024

cmsbuild added code-checks-approved and removed code-checks-pending labels May 23, 2024

kbunkow mentioned this pull request May 23, 2024

Update of Phase-2 GMT Emulator #44498

Merged

cmsbuild added tests-started and removed tests-pending labels May 24, 2024

cmsbuild added tests-approved and removed tests-started labels May 24, 2024

zhenbinwu pushed a commit to zhenbinwu/cmssw that referenced this pull request May 24, 2024

Update OMTF config with PR cms-sw#45032.

d300559

Merge branch 'master' into from-CMSSW_14_1_X_2024-05-22-2300_KBv1

83e721b

cmsbuild added tests-pending code-checks-pending and removed tests-approved code-checks-approved labels May 28, 2024

cmsbuild removed the code-checks-pending label May 28, 2024

cmsbuild added the code-checks-approved label May 28, 2024

cmsbuild added tests-started and removed tests-pending labels May 29, 2024

cmsbuild added tests-approved and removed tests-started labels May 29, 2024

cmsbuild added l1-approved and removed l1-pending labels Jun 4, 2024

cmsbuild added fully-signed upgrade-approved and removed pending-signatures upgrade-pending labels Jun 7, 2024

cmsbuild added orp-approved and removed orp-pending labels Jun 12, 2024

cmsbuild merged commit 51b889c into cms-sw:master Jun 12, 2024
11 checks passed

This was referenced Jun 12, 2024

[AARCH64] jemalloc: explicitly set pagesize cms-sw/cmsdist#9243

Merged

StreamSchedule.h: remove unused variable #45205

Merged

kbunkow mentioned this pull request Sep 12, 2024

possibility of configuring the OMTF emulator directly from XMLs (14_0_X) #45988

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

possibility of configuring the OMTF emulator directly from XMLs #45032

possibility of configuring the OMTF emulator directly from XMLs #45032

kbunkow commented May 23, 2024

cmsbuild commented May 23, 2024 •

edited

Loading

cmsbuild commented May 23, 2024

cmsbuild commented May 23, 2024

epalencia commented May 24, 2024

cmsbuild commented May 24, 2024

artlbv commented May 27, 2024

kbunkow commented May 27, 2024

cmsbuild commented May 28, 2024

cmsbuild commented May 28, 2024

aloeliger commented May 29, 2024

cmsbuild commented May 29, 2024

aloeliger commented Jun 4, 2024

srimanob commented Jun 7, 2024

srimanob commented Jun 7, 2024

cmsbuild commented Jun 7, 2024

rappoccio commented Jun 12, 2024

antoniovilela commented Jun 12, 2024

possibility of configuring the OMTF emulator directly from XMLs #45032

possibility of configuring the OMTF emulator directly from XMLs #45032

Conversation

kbunkow commented May 23, 2024

PR description:

PR validation:

Details of changes:

cmsbuild commented May 23, 2024 • edited Loading

cmsbuild commented May 23, 2024

cmsbuild commented May 23, 2024

epalencia commented May 24, 2024

cmsbuild commented May 24, 2024

Comparison Summary

artlbv commented May 27, 2024

kbunkow commented May 27, 2024

cmsbuild commented May 28, 2024

cmsbuild commented May 28, 2024

aloeliger commented May 29, 2024

cmsbuild commented May 29, 2024

Comparison Summary

aloeliger commented Jun 4, 2024

srimanob commented Jun 7, 2024

srimanob commented Jun 7, 2024

cmsbuild commented Jun 7, 2024

rappoccio commented Jun 12, 2024

antoniovilela commented Jun 12, 2024

cmsbuild commented May 23, 2024 •

edited

Loading