MiniAOD warnings in 12_4_0_pre1 using 2021 data #37227

tvami · 2022-03-13T19:29:36Z

Modifying the 2021 data based prompt reco wf 138.4 by adding the MiniAOD step, i.e. doing

voms-proxy-init -voms cms -rfc
echo '{"346512" : [[250, 300]]}' > step1_lumiRanges.log
(dasgoclient --limit 0 --query 'lumi,file dataset=/MinimumBias/Commissioning2021-v1/RAW run=346512' --format json | das-selected-lumis.py 250,300 ) | sort -u > step1_dasquery.log
cmsDriver.py step2 --conditions auto:run3_data_prompt -s RAW2DIGI,L1Reco,RECO,PAT --datatier MINIAOD --eventcontent MINIAOD --data --process reRECO --scenario pp --era Run3 --customise Configuration/DataProcessing/RecoTLR.customisePrompt --triggerResultsProcess RECO -n 100 --filein filelist:step1_dasquery.log --lumiToProcess step1_lumiRanges.log --fileout file:step2.root

A ROOT problem:
I get the following warning

input_line_391:10:7: warning: ignoring return value of function declared with 'nodiscard' attribute [-Wunused-result]
      ((const vector<string>*)obj)->empty();
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

EGM problem
There is also a msg about unused products:

%MSG-w UnusedProductsForCanDeleteEarly:  AfterModDestruction  13-Mar-2022 20:22:23 CET pre-events
The following products in the 'canDeleteEarly' list are not used in this job and will be ignored.
 If possible, remove the producer from the job or add the product to the producer's own 'mightGet' list.
 recoIsoDepositedmValueMap_gedElPFIsoDepositChargedAll__reRECO
 recoIsoDepositedmValueMap_gedElPFIsoDepositCharged__reRECO
 recoIsoDepositedmValueMap_gedElPFIsoDepositGamma__reRECO
 recoIsoDepositedmValueMap_gedElPFIsoDepositNeutral__reRECO
 recoIsoDepositedmValueMap_gedElPFIsoDepositPU__reRECO
%MSG

The text was updated successfully, but these errors were encountered:

cmsbuild · 2022-03-13T19:29:49Z

A new Issue was created by @tvami Tamas Vami.

@Dr15Jones, @perrotta, @dpiparo, @makortel, @smuzaffar, @qliphy can you please review it and eventually sign/assign? Thanks.

cms-bot commands are listed here

tvami · 2022-03-13T19:30:19Z

assign xpog,reconstruction

cmsbuild · 2022-03-13T19:30:39Z

New categories assigned: xpog

@mariadalfonso,@gouskos,@fgolf you have been requested to review this Pull request/Issue and eventually sign? Thanks

cmsbuild · 2022-03-13T19:32:30Z

New categories assigned: reconstruction

@slava77,@jpata,@clacaputo you have been requested to review this Pull request/Issue and eventually sign? Thanks

slava77 · 2022-03-14T03:44:59Z

https://mattermost.web.cern.ch/cms-o-and-c/pl/dy6wygg9mifzpj9nte7eoiz5pa
so, it became an issue now ;)

jpata · 2022-03-14T11:39:04Z

~~I cannot see the mattermost discussion above for some reason, can someone copy it here?~~

I checked that these warnings do not appear in CMSSW_12_3_0_pre1, pre2, pre3, pre4, so one can bisect which release they appeared in.
EDIT: first appears in pre5

At first glance, the first warning (ignoring return value) looks like something gets processed with cling during execution?
EDIT: OK, I see from mattermost that it's the string cut parser...

The second warning (early delete) indicates that gedElPFIsoDeposit are no longer produced, while an early delete is requested.

jpata · 2022-03-14T11:58:19Z

~~Potentially related to this PR? #35953~~
EDIT: no, see #37227 (comment)

jpata · 2022-03-14T12:20:39Z

The early-delete related warning seems to be from #36882.
As far as I can tell, gedElPFIsoValueChargedAll03 or gedElPFIsoDepositChargedAll is not persisted with the config above.

tvami · 2022-03-14T12:24:37Z

assign core

cmsbuild · 2022-03-14T12:24:57Z

New categories assigned: core

@Dr15Jones,@smuzaffar,@makortel you have been requested to review this Pull request/Issue and eventually sign? Thanks

jpata · 2022-03-14T12:26:37Z

As far as I can tell, gedElPFIsoValueChargedAll03 or gedElPFIsoDepositChargedAll is not persisted with the config above.

Note: this does not imply that it's a core issue. Rather, it looks like the cmsDriver config does not keep the gedElPFIsoValueChargedAll03 object, so it's never produced, so the early delete correctly ignores it (if I understood correctly).

tvami · 2022-03-14T12:36:33Z

Note: this does not imply that it's a core issue.

Oh ok, sorry, sure I remove them then?

makortel · 2022-03-14T13:26:14Z

The early deletion warning indeed came with #36882. As far as I can tell, these producers have not been consumed for a long time, and therefore they have not been run (and been silently deleted since 11_3_0_pre1). #36882 made that visible because their products are being asked to be deleted early in

cmssw/CommonTools/ParticleFlow/python/Isolation/customiseEarlyDeleteForCandIsoDeposits.py

Lines 5 to 17 in 689d470

    
           def customiseEarlyDeleteForCandIsoDeposits(process, products): 
        
               # Find the producers 
        
               def _branchName(productType, moduleLabel, instanceLabel=""): 
        
                   return "%s_%s_%s_%s" % (productType, moduleLabel, instanceLabel, process.name_()) 
        
               for name, module in process.producers_().items(): 
        
                   cppType = module._TypedParameterizable__type 
        
                   if cppType == "CandIsoDepositProducer": 
        
                       if module.ExtractorPSet.ComponentName in ["CandViewExtractor", "PFCandWithSuperClusterExtractor"] : 
        
                           products[name].append(_branchName("recoIsoDepositedmValueMap", name)) 
        
               return products

makortel · 2022-03-14T13:34:48Z

@pcanal We're getting the following warning in CMSSW jobs

input_line_391:10:7: warning: ignoring return value of function declared with 'nodiscard' attribute [-Wunused-result]
      ((const vector<string>*)obj)->empty();
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Back in https://mattermost.web.cern.ch/cms-o-and-c/pl/dy6wygg9mifzpj9nte7eoiz5pa I traced this to

#0  0x00007ffff52b3bbd in write () from /lib64/libc.so.6
...
#8  0x00007fffe03af07a in clang::TextDiagnostic::emitDiagnosticLoc(clang::FullSourceLoc, clang::PresumedLoc, clang::DiagnosticsEngine::Level, llvm::ArrayRef<clang::CharSourceRange>) ()
...
#37 0x00007fffdfe7fc40 in cling::IncrementalParser::ParseInternal(llvm::StringRef) () from /cvmfs/cms-ib.cern.ch/week0/slc7_amd64_gcc10/cms/cmssw/CMSSW_12_3_X_2022-02-15-2300/external/slc7_amd64_gcc10/lib/libCling.so
...
#45 0x00007ffff7e613a7 in edm::FunctionWithDict::FunctionWithDict(TMethod*) () from /cvmfs/cms-ib.cern.ch/week0/slc7_amd64_gcc10/cms/cmssw/CMSSW_12_3_X_2022-02-15-2300/lib/slc7_amd64_gcc10/libFWCoreReflection.so
...
#48 0x00007fffc52c6cd6 in reco::parser::MethodSetter::push()
...
#67 0x00007fffc52e54d4 in reco::parser::cutParser()
#68 0x00007fff3a18fc87 in StringCutObjectSelector<pat::TriggerObjectStandAlone, false>::StringCutObjectSelector(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool) ()
#69 0x00007fff3a18ff71 in ObjectSelectorBase<SingleElementCollectionSelector<std::vector<pat::TriggerObjectStandAlone, ...

pointing towards PATTriggerObjectStandAloneSelector. The configuration I looked at the time of that had cut = cms.string('!filterLabels.empty()'). pat::TriggerObjectStandalone::filterLabels() returns const std::vector<std::string> &, matching to the C-style cast on the warning message.

With ROOT in CMSSW_12_3_X_2022-02-15-2300

root [0] std::vector<std::string> foo;
root [1] foo.empty()
ROOT_prompt_1:1:1: warning: ignoring return value of function declared with 'nodiscard' attribute [-Wunused-result]
foo.empty()
^~~~~~~~~
(bool) true

This function call appeared to be the minimal thing causing the warning

cmssw/FWCore/Reflection/src/FunctionWithDict.cc

Line 25 in fec3825

funcptr_ = gInterpreter->CallFunc_IFacePtr(callFunc);

Reproduces also with

root [0] auto cl = TClass::GetClass("std::vector<std::string>");
root [1] auto method = (TMethod*)cl->GetListOfMethods()->FindObject("empty");
root [2] TMethodCall call(method);
root [3] auto ptr = call.GetCallFunc();
root [4] auto ptr2 = gInterpreter->CallFunc_IFacePtr(ptr);
input_line_13:10:7: warning: ignoring return value of function declared with 'nodiscard' attribute [-Wunused-result]
      ((const vector<string>*)obj)->empty();
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
root [5]

Would you have any suggestions how to silence the warning?

jpata · 2022-03-14T13:34:50Z

@cms-sw/egamma-pog-l2 please take note and remove gedElPFIsoDeposits if possible from the configs.

perrotta · 2022-03-14T13:36:50Z

The early-delete related warning seems to be from #36882. As far as I can tell, gedElPFIsoValueChargedAll03 or gedElPFIsoDepositChargedAll is not persisted with the config above.

Event interpretation was removed from RelVal steps with #36465 (in case it may help....)

pcanal · 2022-03-14T19:40:29Z

@Axel-Naumann Any idea for this?

Axel-Naumann · 2022-03-15T07:20:04Z

@jalopezg-r00t FYI!

jpata · 2022-03-15T08:05:23Z

Potentially related to this PR? #35953

FWIW pre4 + 35953 does not raise the "ignoring return value of function" warning.

jpata · 2022-03-15T08:08:55Z

According to the release notes, pre5 introduced root 6.24. As I understand, "ignoring return value of function" is related to that.

Axel-Naumann · 2022-03-15T09:39:49Z

Oh, this is using ROOT v6.24?! Fine, we need to backport root-project/root#9244 - @jalopezg-r00t can you do the necessary?

jalopezg-git · 2022-03-15T10:58:23Z

Oh, this is using ROOT v6.24?! Fine, we need to backport root-project/root#9244 - @jalopezg-r00t can you do the necessary?

Yes, we were missing a backport to v6.24. PR up here: root-project/root#10120!

jalopezg-git · 2022-03-16T14:35:41Z

@makortel @jpata The required patch has been merged into ROOT's v6-24-00-patches branch some hours ago. Let us know if we can do anything else. 🙂

jpata · 2022-03-25T09:20:45Z

@smuzaffar just to check - should v6-24-00-patches be propagated to https://github.com/cms-sw/cmsdist/blob/IB/CMSSW_12_4_X/master/root.spec ?

smuzaffar · 2022-03-25T09:23:53Z

yes

smuzaffar · 2022-03-25T09:24:21Z

I will take care of it

smuzaffar · 2022-03-25T09:37:11Z

cms-sw/cmsdist#7722 updates root in 12.4.X to head of v6-24-00-patches

smuzaffar · 2022-03-28T06:10:53Z

cms-sw/cmsdist#7722 is ready using the tip (commit: 4b08829) of root v6-24-00-patches but I still see the warning

[email protected]>cd /cvmfs/cms-ci.cern.ch/week0/cms-sw/cmsdist/7722/23431/CMSSW_12_4_X_2022-03-27-0000/
[email protected]>cmsenv
**** Following environment variables are going to be unset.
      CMSSW_FULL_RELEASE_BASE

[email protected]>root
   ------------------------------------------------------------------
  | Welcome to ROOT 6.24/07                        https://root.cern |
  | (c) 1995-2021, The ROOT Team; conception: R. Brun, F. Rademakers |
  | Built for linuxx8664gcc on Mar 27 2022, 22:58:00                 |
  | From tag , 2 September 2021                                      |
  | With g++ (GCC) 10.3.0                                            |
  | Try '.help', '.demo', '.license', '.credits', '.quit'/'.q'       |
   ------------------------------------------------------------------

root [0] std::vector<std::string> foo;
root [1] foo.empty()
ROOT_prompt_1:1:1: warning: ignoring return value of function declared with 'nodiscard' attribute [-Wunused-result]
foo.empty()
^~~~~~~~~
(bool) true

smuzaffar · 2022-03-28T06:27:44Z

by the way, this warning is also present in root master and v6.26 based cmssw IBs

jalopezg-git · 2022-03-28T07:23:06Z

I get the following warning

input_line_391:10:7: warning: ignoring return value of function declared with 'nodiscard' attribute [-Wunused-result]
      ((const vector<string>*)obj)->empty();
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The applied patch should fix the original issue reported in this ticket (i.e. invoking a [[nodiscard]] function via CallFunc interface does not emit a warning if the result is not used). However, using such function at the prompt still emits a warning if the result is not directly used.

root [0] auto cl = TClass::GetClass("std::vector<std::string>");
root [1] auto method = (TMethod*)cl->GetListOfMethods()->FindObject("empty");
root [2] TMethodCall call(method);
root [3] auto ptr = call.GetCallFunc();
root [4] auto ptr2 = gInterpreter->CallFunc_IFacePtr(ptr);
input_line_13:10:7: warning: ignoring return value of function declared with 'nodiscard' attribute [-Wunused-result]
      ((const vector<string>*)obj)->empty();
      ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Can you try this and confirm that it does not generate a diagnostic?

smuzaffar · 2022-03-28T07:42:14Z

thanks @jalopezg-r00t , I confirm that with latest root v6.24-patches it does not generate the warning.

smuzaffar · 2022-03-28T07:42:40Z

cms-sw/cmsdist#7722 has been merged and will be available for 11h00 IBs

tvami · 2022-03-28T14:57:23Z

unassign xpog

SohamBhattacharya · 2022-03-28T15:01:09Z

@cms-sw/egamma-pog-l2 please take note and remove gedElPFIsoDeposits if possible from the configs.

Hi @jpata , could you please clarify if this is still required from E/Gamma, or it's already been fixed?

tvami · 2022-03-28T15:50:13Z

Isnt the fix simple to get rid of this line
https://github.com/cms-sw/cmssw/blob/master/RecoEgamma/EgammaElectronProducers/python/electronPFIsolationDeposits_cff.py#L37
?

makortel · 2022-03-28T17:37:32Z

+core

jpata · 2022-03-29T07:32:18Z

could you please clarify if this is still required from E/Gamma, or it's already been fixed?

@SohamBhattacharya they are still in the configs

cmssw/RecoEgamma/EgammaElectronProducers/python/electronPFIsolationValues_cff.py

Line 98 in 6d2f660

gedElPFIsoValueCharged03 = cms.EDProducer("PFCandIsolatorFromDeposits",

, and if Egamma is no longer using them, they should be removed.

SohamBhattacharya · 2022-03-30T09:40:34Z

could you please clarify if this is still required from E/Gamma, or it's already been fixed?

@SohamBhattacharya they are still in the configs

cmssw/RecoEgamma/EgammaElectronProducers/python/electronPFIsolationValues_cff.py

Line 98 in 6d2f660

gedElPFIsoValueCharged03 = cms.EDProducer("PFCandIsolatorFromDeposits",

, and if Egamma is no longer using them, they should be removed.

@jpata Noted, I'll make a PR removing them and also [1] as pointed out by @tvami .

[1] https://github.com/cms-sw/cmssw/blob/master/RecoEgamma/EgammaElectronProducers/python/electronPFIsolationDeposits_cff.py#L24-L37

jpata · 2022-04-27T13:56:11Z

+reconstruction

the ROOT issue (warning about "ignoring return value of function") has been addressed in Updated root to tip of branch v6-24-00-patches cmsdist#7722
the UnusedProductsCanDeleteEarly warning, related to stuff in the configs that was not actually used/run, was addressed in Cleaned GED electron isolation products #37644

Thanks all involved!

cmsbuild · 2022-04-27T13:56:32Z

This issue is fully signed and ready to be closed.

tvami · 2022-04-27T14:04:33Z

Thanks everybody involved, I'm closing the issue now!

cmsbuild added the pending-assignment label Mar 13, 2022

cmsbuild added pending-signatures xpog-pending and removed pending-assignment labels Mar 13, 2022

cmsbuild added the reconstruction-pending label Mar 13, 2022

cmsbuild added the core-pending label Mar 14, 2022

cmsbuild removed the xpog-pending label Mar 28, 2022

cmsbuild added core-approved and removed core-pending labels Mar 28, 2022

smuzaffar mentioned this issue Mar 30, 2022

failure of HLT-Validation tests in reading edm::HepMCProduct #37392

Closed

SohamBhattacharya mentioned this issue Apr 21, 2022

Cleaned GED electron isolation products #37644

Merged

cmsbuild added fully-signed reconstruction-approved and removed reconstruction-pending pending-signatures labels Apr 27, 2022

tvami closed this as completed Apr 27, 2022

MiniAOD warnings in 12_4_0_pre1 using 2021 data #37227

MiniAOD warnings in 12_4_0_pre1 using 2021 data #37227

Comments

tvami commented Mar 13, 2022 • edited Loading

cmsbuild commented Mar 13, 2022

tvami commented Mar 13, 2022 • edited Loading

cmsbuild commented Mar 13, 2022

cmsbuild commented Mar 13, 2022

slava77 commented Mar 14, 2022

jpata commented Mar 14, 2022 • edited Loading

jpata commented Mar 14, 2022 • edited Loading

jpata commented Mar 14, 2022

tvami commented Mar 14, 2022

cmsbuild commented Mar 14, 2022

jpata commented Mar 14, 2022

tvami commented Mar 14, 2022

makortel commented Mar 14, 2022

makortel commented Mar 14, 2022

jpata commented Mar 14, 2022

perrotta commented Mar 14, 2022

pcanal commented Mar 14, 2022

Axel-Naumann commented Mar 15, 2022

jpata commented Mar 15, 2022 • edited Loading

jpata commented Mar 15, 2022

Axel-Naumann commented Mar 15, 2022

jalopezg-git commented Mar 15, 2022

jalopezg-git commented Mar 16, 2022 • edited Loading

jpata commented Mar 25, 2022

smuzaffar commented Mar 25, 2022

smuzaffar commented Mar 25, 2022

smuzaffar commented Mar 25, 2022

smuzaffar commented Mar 28, 2022

smuzaffar commented Mar 28, 2022

jalopezg-git commented Mar 28, 2022

smuzaffar commented Mar 28, 2022

smuzaffar commented Mar 28, 2022

tvami commented Mar 28, 2022

SohamBhattacharya commented Mar 28, 2022

tvami commented Mar 28, 2022

makortel commented Mar 28, 2022

jpata commented Mar 29, 2022

SohamBhattacharya commented Mar 30, 2022

jpata commented Apr 27, 2022

cmsbuild commented Apr 27, 2022

tvami commented Apr 27, 2022

tvami commented Mar 13, 2022 •

edited

Loading

tvami commented Mar 13, 2022 •

edited

Loading

jpata commented Mar 14, 2022 •

edited

Loading

jpata commented Mar 14, 2022 •

edited

Loading

jpata commented Mar 15, 2022 •

edited

Loading

jalopezg-git commented Mar 16, 2022 •

edited

Loading