HcalDQM: Add protection to collection non-exist case for GPUTask, backport of PR #40117 #40119
Conversation
|
A new Pull Request was created by @lwang046 for CMSSW_12_4_X. It involves the following packages:
@emanueleusai, @ahmad3213, @cmsbuild, @jfernan2, @syuvivida, @pmandrik, @micsucmed, @rvenditti can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
|
Pull request #40119 was updated. @emanueleusai, @ahmad3213, @cmsbuild, @jfernan2, @syuvivida, @pmandrik, @micsucmed, @rvenditti can you please check and sign again. |
|
type hcal,bugfix |
|
backport of #40117 |
|
please test |
|
+1 Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-34f298/29164/summary.html Comparison SummarySummary:
|
|
Hi @lwang046 , the PR is now deployed in P5, can you please have a look in Online test gui and see if things looks as expected (assuming that changes can be detected in the GUI) https://cmsweb.cern.ch/dqm/online-playback/start?runnr=0;dataset=/Global/Online/ALL;sampletype=live;filter=all;referencepos=overlay;referenceshow=customise;referencenorm=True;referenceobj1=refobj;referenceobj2=none;referenceobj3=none;referenceobj4=none;search=;striptype=object;stripruns=;stripaxis=run;stripomit=none;workspace=HCAL;size=M;root=Hcal/HcalGPUComparisonTask;focus=;zoom=no; |
|
Hi @rvenditti , looking at the plots, I think there is a potential issue: The axis are in log scale while I try to fit the values of negative into the bins. The idea of filling a constant negative value was trying to differentiating the cases of problematic reconstruction against absent of collections. But now it seems with log scale this can't be done. I think it's more realistic to not fill the event if any collection for comparison is not available, users will have to rely on CLI messages to know the details there. |
…othing if comparison can't be done
|
Pull request #40119 was updated. @emanueleusai, @ahmad3213, @cmsbuild, @jfernan2, @syuvivida, @pmandrik, @micsucmed, @rvenditti can you please check and sign again. |
|
Pull request #40119 was updated. @emanueleusai, @ahmad3213, @cmsbuild, @jfernan2, @syuvivida, @pmandrik, @micsucmed, @rvenditti can you please check and sign again. |
|
please test |
|
+1 Summary: https://cmssdt.cern.ch/SDT/jenkins-artifacts/pull-request-integration/PR-34f298/29265/summary.html Comparison SummarySummary:
|
|
@lwang046 can you confirm this is the final reviewable and deployable version of the PR? we have been testing an intermediate version of this PR already |
Hi @emanueleusai , this is the final version of this PR. |
|
thanks for confirming! |
|
+1 |
cmsbuild
added
dqm-approved
fully-signed
and removed
dqm-pending
pending-signatures
labels
|
This pull request is fully signed and it will be integrated in one of the next CMSSW_12_4_X IBs (tests are also fine) and once validation in the development release cycle CMSSW_13_0_X is complete. This pull request will now be reviewed by the release team before it's merged. @perrotta, @dpiparo, @rappoccio (and backports should be raised in the release meeting by the corresponding L2) |
|
For consistency, backports in older cycles should be propagated to all intermediate ones. You prepared the backports for 12_5_X and 12_4_X: please prepare also the backport for 12_6_X. We'll wait merging these ones until the 12_6_X version of this PR is submitted, at leasr |
Hi @perrotta , I'm a bit confused. I thought #40117 is for 12_6_X, could you help clarify a bit how to create a backport for it, please? |
No, that's for 13_0_X now that we've opened it, please submit directly to the 12_6_X queue in a backport. |
|
+1 |
PR description:
Backport of PR #40117 , corrects the reference&target name mixing in the hcalgpu task, adds some conditions to the collections non-exist case instead of throwing out fetal exception, and correct a minor numeric error in the DigiTask that should mod the LS by 60, instead of 50.