[BuildRule] Copy rootmap to lib product store

[smuzaffar]

Looks like edm class checks silently ignore missing rootmap files and suggest a different checksum when rootmap files are not there. New tag makes sure that rootmap files are also moved to lib production store when pcm files are moved.

Looks like if rootmap files (which are generated (along with pcm) when we generate root dict source file) are missing then root suggests different checksum. For normal PR tests it works as rootmap files are read from the release area but for PR tests with full rebuild ( where we remove ref to release area) the rootmap files are missing and root suggests different checksum. A simple script ( copied from edm class check) [a] shows that if we have rootmap files then class checksum is same as we have in classes_def file but if we remove the rootmap files then it shows different checksum [b]

[a]

> cat get_class_checksum.py
#!/usr/bin/env python3
import sys, ROOT
name=sys.argv[1]
lib=sys.argv[2]
ROOT.PyConfig.DisableRootLogon = True
ROOT.gROOT.SetBatch(True)
ROOT.gROOT.ProcessLine(".autodict")
ROOT.gSystem.Load("lib%s.so" % lib)
ROOT.gROOT.ProcessLine("class checkclass {public: int f(char const* name) {TClass* cl = TClass::GetClass(name); bool b = false; cl->GetCheckSum(b); return (int)b;} };")
ROOT.gROOT.ProcessLine("checkclass checkTheClass;")
c = ROOT.TClass.GetClass(name)
temp = "checkTheClass.f(" + '"' + name + '"' + ");"
ROOT.gROOT.ProcessLine(temp)
print(c.GetCheckSum(),c.GetClassVersion())

[b]

• DataFormats/CTPPSDetId/src/classes_def.xml

<class name="CTPPSDetId" ClassVersion="3">
    <version ClassVersion="3" checksum="2579437464"/>
</class>

• content of lib/arch

DataFormatsCTPPSDetId_xr.rootmap    DataFormatsCommon_xr.rootmap
DataFormatsDetId_xr.rootmap    DataFormatsProvenance_xr.rootmap
libDataFormatsCTPPSDetId.so  libDataFormatsDetId.so
libFWCoreMessageLogger.so  libFWCoreReflection.so
DataFormatsCTPPSDetId_xr_rdict.pcm  DataFormatsCommon_xr_rdict.pcm
DataFormatsDetId_xr_rdict.pcm  DataFormatsProvenance_xr_rdict.pcm
libDataFormatsCommon.so      libDataFormatsProvenance.so
libFWCorePluginManager.so  libFWCoreUtilities.so

• running script with rootmap files available

> get_class_checksum.py CTPPSDetId DataFormatsCTPPSDetId
2579437464 3

• running script without rootmap files

> mv lib/el8_amd64_gcc12/*.rootmap backup/
> get_class_checksum.py CTPPSDetId DataFormatsCTPPSDetId
1554945820 3

[smuzaffar]

test parameters:

• full_cmssw = true

[smuzaffar]

please test

[cmsbuild]

A new Pull Request was created by @smuzaffar for branch IB/CMSSW_15_0_X/master.

@aandvalenzuela, @iarspider, @smuzaffar can you please review it and eventually sign? Thanks.
@antoniovilela, @mandrenguyen, @rappoccio, @sextonkennedy you are the release manager for this.
cms-bot commands are listed here

[cmsbuild]

cms-bot internal usage

[makortel]

A simple script ( copied from edm class check) [a] shows that if we have rootmap files then class checksum is same as we have in classes_def file but if we remove the rootmap files then it shows different checksum [b]

Interesting behavior. @pcanal Can you tell if this is an expected behavior from the ROOT side? (I could imagine in absence of the rootmap file ROOT could auto-parse the headers, and that leading to a different checksum)

[pcanal]

This is unexpected. The checksum resulting in parsing the header at run-time should be exactly the same than the checksum seen by rootcling/genreflex. Specifically the 2 environment are meant to see exactly the same shape/layout for the classes involved in I/O.

An alternative explanation may (or may not) that the presence or absence of rootmap file in a specific directory changes which library is auto-loaded for a given class.

[smuzaffar]

please test

[cmsbuild]

Pull request #9568 was updated.

[cmsbuild]

-1

Failed Tests: UnitTests
Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-603fb9/43394/summary.html
COMMIT: d22076e
CMSSW: CMSSW_15_0_X_2024-12-11-0800/el8_amd64_gcc12
User test area: For local testing, you can use /cvmfs/cms-ci.cern.ch/week1/cms-sw/cmsdist/9568/43394/install.sh to create a dev area with all the needed externals and cmssw changes.

The following merge commits were also included on top of IB + this PR after doing git cms-merge-topic:

• @fwyzard Fix PFClusterSoAProducer to read a device collection cmssw#46887
• @stahlleiton Implement eras for Run3 2025 PbPb UPC and oxygen runs cmssw#46878
• @AdrianoDee Restoring 2023 RelVal Inputs cmssw#46909
• @leonardogiannini [mkFit] small fix in propToPlane cmssw#46906
• @fwyzard Let edm::Wrapper<T> use the constructor T{kUninitialized} cmssw#46877
• @mbluj Sort hybrid taus with descending pt [fix] cmssw#46905
• @fwyzard Update NVIDIA gdrcopy to v2.4.3 #9563
• @cms-sw BuildRule: Use correct dev.cc.o for alpaka/serial backend #9566

You can see more details here:
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-603fb9/43394/git-recent-commits.json
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-603fb9/43394/git-merge-result

Unit Tests

I found 1 errors in the following unit tests:

---> test TestDQMOfflineConfigurationGotAll had ERRORS

Comparison Summary

Summary:

• You potentially removed 164 lines from the logs
• Reco comparison results: 17 differences found in the comparisons
• DQMHistoTests: Total files compared: 46
• DQMHistoTests: Total histograms compared: 3510017
• DQMHistoTests: Total failures: 452
• DQMHistoTests: Total nulls: 0
• DQMHistoTests: Total successes: 3509545
• DQMHistoTests: Total skipped: 20
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: 0.0 KiB( 45 files compared)
• Checked 202 log files, 172 edm output root files, 46 DQM output files
• TriggerResults: no differences found

[smuzaffar]

+externals

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next IB/CMSSW_15_0_X/master IBs (but tests are reportedly failing). This pull request will now be reviewed by the release team before it's merged. @sextonkennedy, @antoniovilela, @rappoccio, @mandrenguyen (and backports should be raised in the release meeting by the corresponding L2)