static const -> const for thread safety in DeepMETProducer::produce variable (bp of #31738)

[slava77]

backport of #31738

multiple instances of DeepMETProducer (stream module) can end up using the same static const ... input_list initialized by another instance, which is a wrong behavior

I'm not well aware of use cases in 11_1_X, but it may be a useful fix in the context of HLT TDR studies

[cmsbuild]

A new Pull Request was created by @slava77 (Slava Krutelyov) for CMSSW_11_1_X.

It involves the following packages:

RecoMET/METPUSubtraction

@perrotta, @jpata, @cmsbuild, @slava77 can you please review it and eventually sign? Thanks.
@mmarionncern, @gouskos, @ahinzmann, @jdamgov, @jdolen, @nhanvtran, @gkasieczka, @clelange, @schoef, @mariadalfonso, @seemasharmafnal this is something you requested to watch as well.
@silviodonato, @dpiparo, @qliphy you are the release manager for this.

cms-bot commands are listed here

• Backported from static const -> const for thread safety in DeepMETProducer::produce variable #31738

[slava77]

@cmsbuild please test

[cmsbuild]

The tests are being triggered in jenkins.

• CMSSW_11_1_X_2020-10-10-1100/slc7_amd64_gcc820: https://cmssdt.cern.ch/jenkins/job/ib-run-pr-tests/9852/console Started: 2020/10/10 16:55

[cmsbuild]

+1
Tested at: 09f9186
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-a689d5/9852/summary.html
CMSSW: CMSSW_11_1_X_2020-10-10-1100
SCRAM_ARCH: slc7_amd64_gcc820

[cmsbuild]

Comparison job queued.

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-a689d5/9852/summary.html

Comparison Summary:

• No significant changes to the logs found
• Reco comparison results: 2 differences found in the comparisons
• DQMHistoTests: Total files compared: 36
• DQMHistoTests: Total histograms compared: 2781012
• DQMHistoTests: Total failures: 1
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 2780961
• DQMHistoTests: Total skipped: 50
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 35 files compared)
• Checked 152 log files, 16 edm output root files, 36 DQM output files

[slava77]

Reco comparison results: 2 differences found in the comparisons

these are apparently in TTbar14TeV 2026D41 wf 20434.0 in ticlTracksters{HAD,Merge} id_probabilities; these should be unrelated to the change made in this PR (even though both are somewhat related via using TensorFlow, which may give ideas of some longer range cross-talk between TF sessions).

@kpedro88 @rovere @felicepantaleo
do we have some non-reproducibility issue in 11_1_X in TICL?

[slava77]

@cmsbuild please test

trying to see if the diffs in ticlTracksters are repeatable

[cmsbuild]

The tests are being triggered in jenkins.

• CMSSW_11_1_X_2020-10-11-0000/slc7_amd64_gcc820: https://cmssdt.cern.ch/jenkins/job/ib-run-pr-tests/9862/console Started: 2020/10/11 20:07

[cmsbuild]

+1
Tested at: 09f9186
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-a689d5/9862/summary.html
CMSSW: CMSSW_11_1_X_2020-10-11-0000
SCRAM_ARCH: slc7_amd64_gcc820

[cmsbuild]

Comparison job queued.

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-a689d5/9862/summary.html

Comparison Summary:

• No significant changes to the logs found
• Reco comparison results: 4 differences found in the comparisons
• DQMHistoTests: Total files compared: 36
• DQMHistoTests: Total histograms compared: 2781012
• DQMHistoTests: Total failures: 3
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 2780959
• DQMHistoTests: Total skipped: 50
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 35 files compared)
• Checked 152 log files, 16 edm output root files, 36 DQM output files

[slava77]

Reco comparison results: 2 differences found in the comparisons

these are apparently in TTbar14TeV 2026D41 wf 20434.0 in ticlTracksters{HAD,Merge} id_probabilities; these should be unrelated to the change made in this PR (even though both are somewhat related via using TensorFlow, which may give ideas of some longer range cross-talk between TF sessions).

@kpedro88 @rovere @felicepantaleo
do we have some non-reproducibility issue in 11_1_X in TICL?

after rerunning (the tests were using CMSSW_11_1_X_2020-10-11-0000), there are no differences in ticlTracksters this time. So, this doesn't look repeatable.

[slava77]

+1

for #31741 09f9186

I think that this is a "nice to have" update for 11_1_X (not an urgent case for an updated release)

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next CMSSW_11_1_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_11_2_X is complete. This pull request will now be reviewed by the release team before it's merged. @silviodonato, @dpiparo, @qliphy (and backports should be raised in the release meeting by the corresponding L2)

[qliphy]

+1