Rivet3 gcc820

[mseidel42]

@fabiocos I just built these packages using gcc820 without problem:

mseidel@cmsdev23 ~ % ls /build/mseidel/a/BUILD/slc7_amd64_gcc820/external/
herwigpp  herwigpp-toolfile  madgraph5amcatnlo  rivet  rivet-toolfile  thepeg  thepeg-toolfile  yoda

Can you start a test here, please, so that I can see where it goes wrong? Otherwise, I need the logs of the failed builds.

[cmsbuild]

A new Pull Request was created by @intrepid42 (Markus Seidel) for branch IB/CMSSW_11_0_X/gcc820.

@cmsbuild, @smuzaffar, @gudrutis, @mrodozov can you please review it and eventually sign? Thanks.
cms-bot commands are listed here

[mrodozov]

please test

[cmsbuild]

The tests are being triggered in jenkins.
https://cmssdt.cern.ch/jenkins/job/ib-run-pr-tests/2053/console Started: 2019/08/16 11:41

[mrodozov]

I know where it was going wrong - didn't have fastjet contrib . The problem however started becoming more complex when the last cmsdist PR I tried didn't have cmssw IB to test against but the one from aug 12, which is without the rivet interface adjustments. Now the latest IB is with the adjustments so I hope this PR will build

[cmsbuild]

+1
Tested at: 4188c98
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-ae3bf7/2053/summary.html

[cmsbuild]

Comparison job queued.

[fabiocos]

@mrodozov @intrepid42 for my understanding, if the problem is from missing headers why does this appear for gcc8 and gcc9 only, and we haven't broken the main gcc7 IB?

[mseidel42]

Because the gcc700 PR #5107 had the correct Requires and got merged correctly (fe7de33 +66-12) but something went wrong in the gcc820/900 merges, note the different number of added/deleted lines: 61aa0ff e127550 (+65-10)

[mrodozov]

Yes, the forward port merge failed for gcc8 and gcc9 in which case usually the .spec on gcc700 is not the same as on the branches it tries to auto merge.
This is when we have patches for those not same specs in this case we had custom patches which are still there.
The patch for gcc8 is still in use :
https://github.com/cms-sw/cmsdist/blob/IB/CMSSW_11_0_X/gcc820/thepeg-2.1.1-gcc8.patch

What we usually do is to keep the external as it was when it fails to auto merge which is what I tried but in this case also the cmssw interface changes and this doesn't work .

And now I have to fix also the gcc9 external :)

[cmsbuild]

Comparison is ready
https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-ae3bf7/2053/summary.html

Comparison Summary:

• No significant changes to the logs found
• Reco comparison results: 46157 differences found in the comparisons
• DQMHistoTests: Total files compared: 34
• DQMHistoTests: Total histograms compared: 2939508
• DQMHistoTests: Total failures: 120590
• DQMHistoTests: Total nulls: 7
• DQMHistoTests: Total successes: 2818570
• DQMHistoTests: Total skipped: 341
• DQMHistoTests: Total Missing objects: 0
• DQMHistoSizes: Histogram memory added: -1.438 KiB( 33 files compared)
• DQMHistoSizes: changed ( 10224.0 ): -0.352 KiB SiStrip/MechanicalView
• DQMHistoSizes: changed ( 250202.181 ): 0.586 KiB SiStrip/MechanicalView
• DQMHistoSizes: changed ( 136.788 ): -0.004 KiB JetMET/SUSYDQM
• DQMHistoSizes: changed ( 25202.0 ): -0.111 KiB SiStrip/MechanicalView
• DQMHistoSizes: changed ( 7.3 ): -1.557 KiB SiStrip/MechanicalView
• Checked 142 log files, 14 edm output root files, 34 DQM output files

[fabiocos]

@mrodozov I understand that the comparison is made against the gcc7 build, this would explain the large number of differences. Apart for this it looks ok

[mrodozov]

+externals
I agree, lets put it in to fix the IB

[cmsbuild]

This pull request is fully signed and it will be integrated in one of the next IB/CMSSW_11_0_X/gcc820 IBs (tests are also fine). This pull request will now be reviewed by the release team before it's merged. @davidlange6, @slava77, @smuzaffar, @fabiocos (and backports should be raised in the release meeting by the corresponding L2)

[fabiocos]

+1