[13.2.X] Add pixelgpu online DQM client and augment DQM/Integration unit tests

DQM/Integration/python/clients/ecalgpu_dqm_sourceclient-live_cfg.py

@@ -11,8 +11,8 @@
 ### Load cfis ###
 
 if unitTest:
-    process.load("DQM.Integration.config.unittestinputsource_cfi")
-    from DQM.Integration.config.unittestinputsource_cfi import options
+    process.load("DQM.Integration.config.unitteststreamerinputsource_cfi")
+    from DQM.Integration.config.unitteststreamerinputsource_cfi import options
 else:
     process.load("DQM.Integration.config.inputsource_cfi")
     from DQM.Integration.config.inputsource_cfi import options

DQM/Integration/python/clients/hcalgpu_dqm_sourceclient-live_cfg.py

@@ -37,8 +37,8 @@
 else:
 	process.load('DQM.Integration.config.FrontierCondition_GT_cfi')
 if unitTest:
-	process.load("DQM.Integration.config.unittestinputsource_cfi")
-	from DQM.Integration.config.unittestinputsource_cfi import options
+        process.load("DQM.Integration.config.unitteststreamerinputsource_cfi")
+        from DQM.Integration.config.unitteststreamerinputsource_cfi import options
 elif useFileInput:
 	process.load("DQM.Integration.config.fileinputsource_cfi")
 	from DQM.Integration.config.fileinputsource_cfi import options

DQM/Integration/python/clients/pixelgpu_dqm_sourceclient-live_cfg.py

@@ -0,0 +1,127 @@
+#-------------------------------------
+#	Pixel DQM Application using New DQM Sources/Clients
+#-------------------------------------
+
+#-------------------------------------
+#	Standard Python Imports
+#-------------------------------------
+import os, sys, socket, string
+
+#-------------------------------------
+#	Standard CMSSW Imports/Definitions
+#-------------------------------------
+import FWCore.ParameterSet.Config as cms
+from Configuration.Eras.Era_Run3_cff import Run3
+process      = cms.Process('PIXELDQMLIVE', Run3)
+subsystem    = 'Pixel'
+cmssw        = os.getenv("CMSSW_VERSION").split("_")
+debugstr     = "### PixelDQM::cfg::DEBUG: "
+warnstr      = "### PixelDQM::cfg::WARN: "
+errorstr     = "### PixelDQM::cfg::ERROR:"
+useOfflineGT = False
+useFileInput = False
+useMap       = False
+
+unitTest = False
+if 'unitTest=True' in sys.argv:
+	unitTest=True
+	useFileInput=False
+
+#-------------------------------------
+#	Central DQM Stuff imports
+#-------------------------------------
+from DQM.Integration.config.online_customizations_cfi import *
+if useOfflineGT:
+        process.load("Configuration.StandardSequences.FrontierConditions_GlobalTag_cff")
+        process.GlobalTag.globaltag = autoCond['run3_data_prompt'] 
+else:
+	process.load('DQM.Integration.config.FrontierCondition_GT_cfi')
+if unitTest:
+	process.load("DQM.Integration.config.unitteststreamerinputsource_cfi")
+	from DQM.Integration.config.unitteststreamerinputsource_cfi import options
+elif useFileInput:
+	process.load("DQM.Integration.config.fileinputsource_cfi")
+	from DQM.Integration.config.fileinputsource_cfi import options
+else:
+	process.load('DQM.Integration.config.inputsource_cfi')
+	from DQM.Integration.config.inputsource_cfi import options
+process.load('DQM.Integration.config.environment_cfi')
+
+#-------------------------------------
+#	Central DQM Customization
+#-------------------------------------
+process.source.streamLabel = cms.untracked.string("streamDQMGPUvsCPU")
+process.dqmEnv.subSystemFolder = subsystem
+process.dqmSaver.tag = 'PixelGPU'
+process.dqmSaver.runNumber = options.runNumber
+process.dqmSaverPB.tag = 'PixelGPU'
+process.dqmSaverPB.runNumber = options.runNumber
+process = customise(process)
+process.DQMStore.verbose = 0
+if not unitTest and not useFileInput :
+  if not options.BeamSplashRun :
+    process.source.minEventsPerLumi = 100
+
+#-------------------------------------
+#	CMSSW/Pixel non-DQM Related Module import
+#-------------------------------------
+process.load('Configuration.StandardSequences.GeometryRecoDB_cff')
+process.load('FWCore.MessageLogger.MessageLogger_cfi')
+
+#-------------------------------------
+#	CMSSW non-DQM Related Module Settings
+#-------------------------------------
+runType			= process.runType.getRunType()
+runTypeName		= process.runType.getRunTypeName()
+isCosmicRun		= runTypeName=="cosmic_run" or runTypeName=="cosmic_run_stage1"
+isHeavyIon		= runTypeName=="hi_run"
+cmssw			= os.getenv("CMSSW_VERSION").split("_")
+
+#-------------------------------------
+#	Pixel DQM Tasks and Harvesters import
+#-------------------------------------
+process.load('DQM.SiPixelHeterogeneous.SiPixelHeterogenousDQM_FirstStep_cff')
+
+#-------------------------------------
+#	Some Settings before Finishing up
+#-------------------------------------
+process.siPixelPhase1RawDataErrorComparator.pixelErrorSrcGPU = cms.InputTag('hltSiPixelDigisFromSoA')
+process.siPixelPhase1RawDataErrorComparator.pixelErrorSrcCPU = cms.InputTag('hltSiPixelDigisLegacy')
+
+#-------------------------------------
+#       Some Debug
+#-------------------------------------
+process.dump = cms.EDAnalyzer("EventContentAnalyzer")
+process.dumpPath = cms.Path(process.dump)
+
+#-------------------------------------
+#	Hcal DQM Tasks/Clients Sequences Definition
+#-------------------------------------
+process.tasksPath = cms.Path(process.siPixelPhase1RawDataErrorComparator)
+
+#-------------------------------------
+#	Paths/Sequences Definitions
+#-------------------------------------
+process.dqmPath = cms.EndPath(process.dqmEnv)
+process.dqmPath1 = cms.EndPath(process.dqmSaver*process.dqmSaverPB)
+process.schedule = cms.Schedule(process.tasksPath,
+                                #process.dumpPath,  # for debug
+                                process.dqmPath,
+                                process.dqmPath1)
+
+#-------------------------------------
+#	Scheduling and Process Customizations
+#-------------------------------------
+process.options = cms.untracked.PSet(
+		Rethrow = cms.untracked.vstring(
+			"ProductNotFound",
+			"TooManyProducts",
+			"TooFewProducts"
+		)
+)
+process.options.wantSummary = True
+
+# tracer
+#process.Tracer = cms.Service("Tracer")
+print("Final Source settings:", process.source)
+process = customise(process)

DQM/Integration/test/BuildFile.xml

@@ -23,6 +23,9 @@
 <test name="TestDQMOnlineClient-scal_dqm_sourceclient" command="runtest.sh scal_dqm_sourceclient-live_cfg.py"/>
 <test name="TestDQMOnlineClient-sistrip_dqm_sourceclient" command="runtest.sh sistrip_dqm_sourceclient-live_cfg.py"/>
 <test name="TestDQMOnlineClient-onlinebeammonitor_dqm_sourceclient" command="runtest.sh onlinebeammonitor_dqm_sourceclient-live_cfg.py"/>
+<test name="TestDQMOnlineClient-ecalgpu_dqm_sourceclient" command="runtest.sh ecalgpu_dqm_sourceclient-live_cfg.py 369956"/>
+<test name="TestDQMOnlineClient-hcalgpu_dqm_sourceclient" command="runtest.sh hcalgpu_dqm_sourceclient-live_cfg.py 369956"/>
+<test name="TestDQMOnlineClient-pixelgpu_dqm_sourceclient" command="runtest.sh pixelgpu_dqm_sourceclient-live_cfg.py 369956"/>
 <!-- streamDQMCalibration is required -->
 <!-- <test name="TestDQMOnlineClient-ecalcalib_dqm_sourceclient" command="runtest.sh ecalcalib_dqm_sourceclient-live_cfg.py" /> -->
 <!-- streamDQMCalibration is required -->

DQM/Integration/test/runtest.sh

@@ -16,4 +16,10 @@ if [[ -z ${CLIENTS_DIR} ]]; then
 fi
 
 mkdir -p $LOCAL_TEST_DIR/upload
-cmsRun $CLIENTS_DIR/$1 unitTest=True
+
+if [[ $# -eq 1 ]]; then
+    cmsRun $CLIENTS_DIR/$1 unitTest=True
+else
+    echo "Will use streamers files for run $2"
+    cmsRun $CLIENTS_DIR/$1 unitTest=True runNumber=$2
+fi

DQM/SiPixelHeterogeneous/plugins/SiPixelPhase1RawDataErrorComparator.cc

@@ -7,32 +7,26 @@
 //
 // Author: Marco Musich
 //
-#include "DataFormats/Common/interface/Handle.h"
-#include "DataFormats/Math/interface/deltaR.h"
-#include "DataFormats/Math/interface/deltaPhi.h"
-#include "FWCore/Framework/interface/Event.h"
-#include "FWCore/Framework/interface/Frameworkfwd.h"
-#include "FWCore/Framework/interface/MakerMacros.h"
-#include "FWCore/MessageLogger/interface/MessageLogger.h"
-#include "FWCore/ParameterSet/interface/ParameterSet.h"
-#include "FWCore/Utilities/interface/InputTag.h"
-// DQM Histograming
 #include "CondFormats/DataRecord/interface/SiPixelFedCablingMapRcd.h"
 #include "CondFormats/SiPixelObjects/interface/SiPixelFedCablingMap.h"
 #include "CondFormats/SiPixelObjects/interface/SiPixelFedCablingTree.h"
+#include "DQMServices/Core/interface/DQMEDAnalyzer.h"
+#include "DQMServices/Core/interface/DQMStore.h"
+#include "DQMServices/Core/interface/MonitorElement.h"
 #include "DataFormats/Common/interface/DetSetVector.h"
 #include "DataFormats/Common/interface/Handle.h"
 #include "DataFormats/DetId/interface/DetIdCollection.h"
 #include "DataFormats/FEDRawData/interface/FEDNumbering.h"
 #include "DataFormats/SiPixelDetId/interface/PixelFEDChannel.h"
 #include "DataFormats/SiPixelDigi/interface/PixelDigi.h"
-#include "DataFormats/SiPixelRawData/interface/SiPixelErrorsSoA.h"
 #include "EventFilter/SiPixelRawToDigi/interface/PixelDataFormatter.h"
+#include "FWCore/Framework/interface/Event.h"
+#include "FWCore/Framework/interface/Frameworkfwd.h"
+#include "FWCore/Framework/interface/MakerMacros.h"
+#include "FWCore/MessageLogger/interface/MessageLogger.h"
+#include "FWCore/ParameterSet/interface/ParameterSet.h"
+#include "FWCore/Utilities/interface/InputTag.h"
 #include "Geometry/CommonTopologies/interface/SimplePixelTopology.h"
-
-#include "DQMServices/Core/interface/MonitorElement.h"
-#include "DQMServices/Core/interface/DQMEDAnalyzer.h"
-#include "DQMServices/Core/interface/DQMStore.h"
 // for string manipulations
 #include <fmt/printf.h>
 
@@ -84,8 +78,7 @@ namespace {
     k_FED31 = 31,  // 31 indicates an event number error (TBM and FED event number mismatch)
     k_FED32 = 32,  // 32 indicates an incorrectly formatted Slink Header
     k_FED33 = 33,  // 33 indicates an incorrectly formatted Slink Trailer
-    k_FED34 =
-        34,  // 34 indicates the event size encoded in the Slink Trailer is different than size found at raw2digi conversion
+    k_FED34 = 34,  // 34 indicates the evt size encoded in Slink Trailer is different than size found at raw2digi
     k_FED35 = 35,  // 35 indicates an invalid FED channel number
     k_FED36 = 36,  // 36 indicates an invalid ROC value
     k_FED37 = 37,  // 37 indicates an invalid dcol or pixel value
@@ -125,6 +118,8 @@ class SiPixelPhase1RawDataErrorComparator : public DQMEDAnalyzer {
   static void fillDescriptions(edm::ConfigurationDescriptions& descriptions);
 
 private:
+  const edm::InputTag pixelErrorSrcGPU_;
+  const edm::InputTag pixelErrorSrcCPU_;
   const edm::EDGetTokenT<edm::DetSetVector<SiPixelRawDataError>> tokenErrorsGPU_;
   const edm::EDGetTokenT<edm::DetSetVector<SiPixelRawDataError>> tokenErrorsCPU_;
   const std::string topFolderName_;
@@ -133,6 +128,9 @@ class SiPixelPhase1RawDataErrorComparator : public DQMEDAnalyzer {
   MonitorElement* h_FEDerrorVsFEDIdUnbalance_;
   std::unordered_map<SiPixelFEDErrorCodes, MonitorElement*> h_nFEDErrors_;
 
+  // name of the plugins
+  static constexpr const char* kName = "SiPixelPhase1RawDataErrorComparator";
+
   // Define the dimensions of the 2D array
   static constexpr int nFEDs = FEDNumbering::MAXSiPixeluTCAFEDID - FEDNumbering::MINSiPixeluTCAFEDID;
   static constexpr int nErrors = k_FED38 - k_FED25;
@@ -142,10 +140,10 @@ class SiPixelPhase1RawDataErrorComparator : public DQMEDAnalyzer {
 // constructors
 //
 SiPixelPhase1RawDataErrorComparator::SiPixelPhase1RawDataErrorComparator(const edm::ParameterSet& iConfig)
-    : tokenErrorsGPU_(
-          consumes<edm::DetSetVector<SiPixelRawDataError>>(iConfig.getParameter<edm::InputTag>("pixelErrorSrcGPU"))),
-      tokenErrorsCPU_(
-          consumes<edm::DetSetVector<SiPixelRawDataError>>(iConfig.getParameter<edm::InputTag>("pixelErrorSrcCPU"))),
+    : pixelErrorSrcGPU_(iConfig.getParameter<edm::InputTag>("pixelErrorSrcGPU")),
+      pixelErrorSrcCPU_(iConfig.getParameter<edm::InputTag>("pixelErrorSrcCPU")),
+      tokenErrorsGPU_(consumes<edm::DetSetVector<SiPixelRawDataError>>(pixelErrorSrcGPU_)),
+      tokenErrorsCPU_(consumes<edm::DetSetVector<SiPixelRawDataError>>(pixelErrorSrcCPU_)),
       topFolderName_(iConfig.getParameter<std::string>("topFolderName")) {}
 
 //
@@ -172,14 +170,26 @@ void SiPixelPhase1RawDataErrorComparator::analyze(const edm::Event& iEvent, cons
     countsOnGPU[j] = 0.;
   }
 
+  // check upfront if the error collection is present
   edm::Handle<edm::DetSetVector<SiPixelRawDataError>> inputFromCPU;
   iEvent.getByToken(tokenErrorsCPU_, inputFromCPU);
   if (!inputFromCPU.isValid()) {
-    edm::LogWarning("SiPixelCompareTrackSoA") << "reference (cpu) SiPixelRawDataErrors not found; \n"
-                                              << "the comparison will not run.";
+    edm::LogError(kName) << "reference (cpu) SiPixelRawDataErrors collection (" << pixelErrorSrcCPU_.encode()
+                         << ") not found; \n"
+                         << "the comparison will not run.";
+    return;
+  }
+
+  edm::Handle<edm::DetSetVector<SiPixelRawDataError>> inputFromGPU;
+  iEvent.getByToken(tokenErrorsGPU_, inputFromGPU);
+  if (!inputFromGPU.isValid()) {
+    edm::LogError(kName) << "target (gpu) SiPixelRawDataErrors collection (" << pixelErrorSrcGPU_.encode()
+                         << ") not found; \n"
+                         << "the comparison will not run.";
     return;
   }
 
+  // fill the counters on host
   uint errorsOnCPU{0};
   for (auto it = inputFromCPU->begin(); it != inputFromCPU->end(); ++it) {
     for (auto& siPixelRawDataError : *it) {
@@ -191,20 +201,12 @@ void SiPixelPhase1RawDataErrorComparator::analyze(const edm::Event& iEvent, cons
       countsOnCPU[type] += 1;
       countsMatrixOnCPU[fed - FEDNumbering::MINSiPixeluTCAFEDID][type - k_FED25] += 1;
 
-      edm::LogInfo("SiPixelPhase1RawDataErrorComparator")
-          << __PRETTY_FUNCTION__ << " on cpu: FED: " << fed << " detid: " << id.rawId() << " type:" << type;
+      LogDebug(kName) << " (on cpu) FED: " << fed << " detid: " << id.rawId() << " type:" << type;
       errorsOnCPU++;
     }
   }
 
-  edm::Handle<edm::DetSetVector<SiPixelRawDataError>> inputFromGPU;
-  iEvent.getByToken(tokenErrorsGPU_, inputFromGPU);
-  if (!inputFromGPU.isValid()) {
-    edm::LogWarning("SiPixelCompareTrackSoA") << "target (gpu) SiPixelRawDataErrors not found; \n"
-                                              << "the comparison will not run.";
-    return;
-  }
-
+  // fill the counters on device
   uint errorsOnGPU{0};
   for (auto it = inputFromGPU->begin(); it != inputFromGPU->end(); ++it) {
     for (auto& siPixelRawDataError : *it) {
@@ -216,30 +218,28 @@ void SiPixelPhase1RawDataErrorComparator::analyze(const edm::Event& iEvent, cons
       countsOnGPU[type] += 1;
       countsMatrixOnGPU[fed - FEDNumbering::MINSiPixeluTCAFEDID][type - k_FED25] += 1;
 
-      edm::LogInfo("SiPixelPhase1RawDataErrorComparator")
-          << __PRETTY_FUNCTION__ << " on gpu: FED: " << fed << " detid: " << id.rawId() << " type:" << type;
+      LogDebug(kName) << " (on gpu) FED: " << fed << " detid: " << id.rawId() << " type:" << type;
       errorsOnGPU++;
     }
   }
 
-  edm::LogInfo("SiPixelPhase1RawDataErrorComparator")
-      << __PRETTY_FUNCTION__ << " on gpu found: " << errorsOnGPU << " on cpu found: " << errorsOnCPU << std::endl;
+  edm::LogPrint(kName) << " on gpu found: " << errorsOnGPU << " on cpu found: " << errorsOnCPU << std::endl;
 
   h_totFEDErrors_->Fill(errorsOnCPU, errorsOnGPU);
 
   // fill the correlations per error type
   for (unsigned int j = k_FED25; j <= k_FED31; j++) {
     SiPixelFEDErrorCodes code = static_cast<SiPixelFEDErrorCodes>(j);
-    h_nFEDErrors_[code]->Fill(countsOnCPU[j], countsOnGPU[j]);
+    h_nFEDErrors_[code]->Fill(std::min(1000, countsOnCPU[j]), std::min(1000, countsOnGPU[j]));
   }
 
   // fill the error unbalance per FEDid per error type
   for (int i = 0; i < nFEDs; i++) {
     for (int j = 0; j < nErrors; j++) {
       if (countsMatrixOnGPU[i][j] != 0 || countsMatrixOnCPU[i][j] != 0) {
-        edm::LogInfo("SiPixelPhase1RawDataErrorComparator")
-            << "FED: " << i + FEDNumbering::MINSiPixeluTCAFEDID << " error: " << j + k_FED25
-            << " | GPU counts: " << countsMatrixOnGPU[i][j] << " CPU counts:" << countsMatrixOnCPU[i][j] << std::endl;
+        edm::LogVerbatim(kName) << "FED: " << i + FEDNumbering::MINSiPixeluTCAFEDID << " error: " << j + k_FED25
+                                << " | GPU counts: " << countsMatrixOnGPU[i][j]
+                                << " CPU counts:" << countsMatrixOnCPU[i][j] << std::endl;
         h_FEDerrorVsFEDIdUnbalance_->Fill(
             j, i + FEDNumbering::MINSiPixeluTCAFEDID, countsMatrixOnGPU[i][j] - countsMatrixOnCPU[i][j]);
       }
@@ -258,7 +258,7 @@ void SiPixelPhase1RawDataErrorComparator::bookHistograms(DQMStore::IBooker& iBoo
 
   h_FEDerrorVsFEDIdUnbalance_ =
       iBook.book2I("FEErrorVsFEDIdUnbalance",
-                   "difference (GPU-CPE) of FED errors per FEDid per error type;;FED Id number;GPU counts - CPU counts",
+                   "difference (GPU-CPU) of FED errors per FEDid per error type;;FED Id number;GPU counts - CPU counts",
                    nErrors,
                    -0.5,
                    nErrors - 0.5,
@@ -277,26 +277,25 @@ void SiPixelPhase1RawDataErrorComparator::bookHistograms(DQMStore::IBooker& iBoo
                                 "n. of total Pixel FEDError per event; CPU; GPU",
                                 500,
                                 log10(0.5),
-                                log10(500.5),
+                                log10(5000.5),
                                 500,
                                 log10(0.5),
-                                log10(500.5));
+                                log10(5000.5));
 
   for (const auto& element : errorCodeToStringMap) {
     h_nFEDErrors_[element.first] = iBook.book2I(fmt::sprintf("nFED%i_Errors", element.first),
                                                 fmt::sprintf("n. of %ss per event; CPU; GPU", element.second),
-                                                501,
+                                                1000,
                                                 -0.5,
-                                                500.5,
-                                                501,
+                                                1000.5,
+                                                1000,
                                                 -0.5,
-                                                500.5);
+                                                1000.5);
   }
 }
 
 void SiPixelPhase1RawDataErrorComparator::fillDescriptions(edm::ConfigurationDescriptions& descriptions) {
   edm::ParameterSetDescription desc;
-  //desc.add<edm::InputTag>("pixelErrorSrcGPU", edm::InputTag("siPixelDigiErrors"));
   desc.add<edm::InputTag>("pixelErrorSrcGPU", edm::InputTag("siPixelDigis@cuda"))
       ->setComment("input GPU SiPixel FED errors");
   desc.add<edm::InputTag>("pixelErrorSrcCPU", edm::InputTag("siPixelDigis@cpu"))