TCCBlockFormatter Class Reference

#include <TCCBlockFormatter.h>

Inheritance diagram for TCCBlockFormatter:

Public Member Functions
void	DigiToRaw (const EcalTriggerPrimitiveDigi &trigprim, FEDRawData &rawdata, const EcalElectronicsMapping *TheMapping)
	TCCBlockFormatter (BlockFormatter::Config const &, BlockFormatter::Params const &)
Public Member Functions inherited from BlockFormatter
	BlockFormatter (Config const &iC, Params const &iP)
void	CleanUp (FEDRawDataCollection *productRawData, std::map< int, std::map< int, int > > *FEDorder)
void	DigiToRaw (FEDRawDataCollection *productRawData)
void	print (FEDRawData &rawdata)
void	PrintSizes (FEDRawDataCollection *productRawData)

Static Public Attributes
static const int	kCardsPerTower = 5
Static Public Attributes inherited from BlockFormatter
static const int	kCardsPerTower = 5

Private Attributes
bool	AllTPsamples_

Additional Inherited Members
Public Types inherited from BlockFormatter
typedef uint16_t	Word16
typedef uint64_t	Word64
Protected Attributes inherited from BlockFormatter
int	bx_
int	counter_
const bool	debug_
const bool	doBarrel_
const bool	doEndCap_
const bool	doSR_
const bool	doTCC_
const bool	doTower_
int	lv1_
int	orbit_number_
const std::vector< int32_t > *	plistDCCId_
int	runnumber_

Detailed Description

Definition at line 22 of file TCCBlockFormatter.h.

Constructor & Destructor Documentation

TCCBlockFormatter::TCCBlockFormatter	(	BlockFormatter::Config const &	iC,
		BlockFormatter::Params const &	iP
	)

Definition at line 15 of file TCCBlockFormatter.cc.

References AllTPsamples_.

                                                                      : BlockFormatter(iC, iP) {

 AllTPsamples_ = false;
}

Member Function Documentation

void TCCBlockFormatter::DigiToRaw	(	const EcalTriggerPrimitiveDigi &	trigprim,
		FEDRawData &	rawdata,
		const EcalElectronicsMapping *	TheMapping
	)

Definition at line 20 of file TCCBlockFormatter.cc.

References AllTPsamples_, BlockFormatter::bx_, EcalTriggerPrimitiveDigi::compressedEt(), gather_cfg::cout, FEDRawData::data(), BlockFormatter::debug_, TauDecayModes::dec, BlockFormatter::doBarrel_, BlockFormatter::doEndCap_, EcalBarrel, EcalEndcap, stringResolutionProvider_cfi::et, EcalTriggerPrimitiveDigi::fineGrain(), EcalTriggerPrimitiveDigi::id(), EcalTrigTowerDetId::ieta(), EcalTrigTowerDetId::ietaAbs(), ntuplemaker::iline, SurfaceOrientation::inner, createfilelist::int, EcalTrigTowerDetId::iphi(), BlockFormatter::lv1_, EcalElectronicsId::MAX_DCCID_EEM, EcalElectronicsId::MAX_DCCID_EEP, EcalElectronicsMapping::MAX_TCCID, EcalElectronicsId::MIN_DCCID_EEM, EcalElectronicsId::MIN_DCCID_EEP, FEDNumbering::MINECALFEDID, NUMB_SM_EB_PLU_MAX, NUMB_SM_EB_PLU_MIN, BlockFormatter::print(), FEDRawData::resize(), EcalTriggerPrimitiveDigi::size(), FEDRawData::size(), EcalTrigTowerDetId::subDet(), EcalTriggerPrimitiveDigi::ttFlag(), and globals_cff::x1.

Referenced by EcalDigiToRaw::produce().

{

  if (debug_) cout << "enter in TCCBlockFormatter::DigiToRaw " << endl;

  int HEADER_SIZE = 8 * 9;
  int bx = bx_;
  int lv1 = lv1_;


    const EcalTrigTowerDetId& detid = trigprim.id();

           if ( (detid.subDet() == EcalBarrel) && (! doBarrel_) ) return;
           if ( (detid.subDet() == EcalEndcap) && (! doEndCap_) ) return;

    int iDCC = TheMapping -> DCCid(detid);
    int TCCid = TheMapping -> TCCid(detid);


    if (TCCid < EcalElectronicsMapping::MIN_TCCID || TCCid > EcalElectronicsMapping::MAX_TCCID) 
        cout << "Wrong TCCid in TCCBlockFormatter::DigiToRaw " << endl;
    bool IsEndCap = ( (EcalElectronicsId::MIN_DCCID_EEM <= iDCC && iDCC <= EcalElectronicsId::MAX_DCCID_EEM) ||
              (EcalElectronicsId::MIN_DCCID_EEP <= iDCC && iDCC <= EcalElectronicsId::MAX_DCCID_EEP) );

    int FEDid = FEDNumbering::MINECALFEDID + iDCC;

    // note: row is a 64 bit word
    int NTT_max = 68;   // Barrel case
    int Nrows_TCC = 17;     // Barrel case    (without the header row)
    int NTCC = 1;           // Barrel case; number of TCC blocks
    int itcc_block = 1;     // Barrel case

        if (IsEndCap) {
            Nrows_TCC = 8;          // one row is a 64 bit word
            NTCC = 4;       // 4 TTC in EndCap case. Use some custom numbering since
        int pair = TCCid % 2;   // the TCCid is written to the RawData.
                int inner = ( detid.ietaAbs() >= 22) ? 1 : 0;
                itcc_block = 2 * pair + inner + 1;
                if (inner == 1) NTT_max = 28;
                else NTT_max = 16;
        }

    
        int nsamples = trigprim.size();
    if (! AllTPsamples_) nsamples = 1;

    int iTT = TheMapping -> iTT(detid);   // number of tp inside a fed
    if (debug_) cout << "This is a TrigTower  iDCC iTT iTCCBlock TCCid " << dec << 
        iDCC << " " << iTT << " " << itcc_block << " " << TCCid << endl;
        if (debug_) cout << "ieta iphi " << dec << detid.ieta() << " " << detid.iphi() << endl;
        if (iTT <= 0 || iTT > NTT_max)  {
        cout << "invalid iTT " << iTT << endl;
        return;
    }

    int FE_index;
    
    // rawdata points to the block which will be built for TCC data  
    if ((int)rawdata.size() != HEADER_SIZE) {
      FE_index = rawdata.size() / 8 - NTCC*(Nrows_TCC+1);         // as far as raw data have been generated 
      FE_index ++;                                                // infer position in TCC block
      if (debug_) cout << "TCCid already there. FE_index = " << FE_index << endl;
    }
    else {
        if (debug_) cout << "New TTCid added on Raw data, TTCid = " << dec << TCCid << " 0x" << hex << TCCid << endl;
        FE_index = rawdata.size() / 8;                    // size in unites of 64 bits word
            int fe_index = FE_index;
        for (int iblock=0; iblock < NTCC; iblock++) {     // do this once per fed in EB, four times in EE
           rawdata.resize (rawdata.size() + 8);
           unsigned char* ppData = rawdata.data();        // use this to navigate and create the binary
           ppData[8*fe_index] = TCCid & 0xFF;             // fed_index increases in units of bytes
           ppData[8*fe_index+2] = bx & 0xFF;              // bx takes bits 0-11: 0-7+8-11
           ppData[8*fe_index+3] = (bx & 0xF00)>>8;
           ppData[8*fe_index+3] |= 0x60;
           ppData[8*fe_index+4] = lv1 & 0xFF;             // same game done for lv1, which takes bits 0-11: 0-7+8-11 
           ppData[8*fe_index+5] = (lv1 & 0xF00)>>8;       // lv1
           ppData[8*fe_index+6] = NTT_max;
           ppData[8*fe_index+6] |= ((nsamples & 0x1)<<7); // nsamples: number time samples
           ppData[8*fe_index+7] = ((nsamples & 0xE)>>1);
           ppData[8*fe_index+7] |= 0x60;
           if (iblock == 0) FE_index ++;
           fe_index += Nrows_TCC+1;
           rawdata.resize (rawdata.size() + 8*Nrows_TCC);    // 17 lines of TPG data in EB, 8 in EE
        }
        if (debug_) cout << "Added headers and empty lines : " << endl;
        if (debug_) print(rawdata);

        // -- put the B011 already, since for Endcap there can be empty
        // -- lines in the TCC and the SRP blocks
        unsigned char* ppData = rawdata.data();
        for (int iline=FE_index-1; iline < FE_index + (Nrows_TCC+1)*NTCC -1 ; iline++) {
                 ppData[8*iline + 7] |= 0x60;
         ppData[8*iline + 3] |= 0x60;
        }
    }

    unsigned char* pData = rawdata.data();

    // -- Now the TCC Block :

    int jTT = (iTT-1);                                // jTT is the TP number insided a block;
    int irow = jTT/4 + (itcc_block-1)*(Nrows_TCC+1);  // you fit 4 TP's per row; move forward if you're not in the first block;
    int ival = jTT % 4;                               // for each block you have to skip, move of (Nrows_TCC +1) - 1 is for the TCC header

    // RTC required TP's tp follow global phi also in EB+, thus swap them inside the single TCC 
    // here you could swap ival -> 3-ival to swap phi insied EB+ supermodules  
    if(NUMB_SM_EB_PLU_MIN <= iDCC && iDCC <= NUMB_SM_EB_PLU_MAX)
      {ival = 3-ival;}

    FE_index += irow;                                 // ival is location inside a TP row; varies between 0-3

        if (debug_) cout << "Now add tower " << dec << iTT << " irow ival " << dec << irow << " " << dec << ival << endl;
        if (debug_) cout << "new data will be added at line " << dec << FE_index << endl;

    int fg = trigprim.fineGrain();
    int et = trigprim.compressedEt();
    int ttflag = trigprim.ttFlag();

    if (debug_ && (ttflag != 0)) {
        cout << "in TCCBlock : this tower has a non zero flag" << endl;
        cout << "Fedid  iTT  flag " << dec << FEDid << " " << iTT << " " << "0x" << hex << ttflag << endl;
    }
    pData[8*FE_index + ival*2] = et & 0xFF;                   // ival is location inside a TP row; varies between 0-3; tp goes in bits 0-7
    pData[8*FE_index + ival*2+1] = (ttflag<<1) + (fg&0x1);    // fg follows in bit 8; ttfg is in bits 9-11 
    if (IsEndCap) {
        // re-write the TCCid  and N_Tower_Max :
        int ibase = 8*(FE_index - (int)(jTT/4) -1);
        pData[ibase] = TCCid & 0xFF;
                pData[ibase+6] = NTT_max;
                pData[ibase+6] |= ((nsamples & 0x1)<<7);
                pData[ibase+7] |= ((nsamples & 0xE)>>1);
    }
    if (debug_) cout << "pData[8*FE_index + ival*2+1] = " << hex << (int)pData[8*FE_index + ival*2+1] << endl;
    if (debug_) cout << "ttflag ttflag<<1 " << hex << ttflag << " " << hex << (ttflag<<1) << endl;
    if (debug_) cout << "fg&0x1 " << hex << (fg&0x1) << endl;
    if (debug_) cout << "sum " << hex << ( (ttflag<<1) + (fg&0x1) ) << endl;
    if (ival %2 == 1) pData[8*FE_index + ival*2+1] |= 0x60;
    if (debug_) cout << "ttflag et fgbit " << hex << ttflag << " " << hex << et << " " << hex << fg << endl;
    if (debug_) print(rawdata);

    
}

Member Data Documentation

bool TCCBlockFormatter::AllTPsamples_

private

Definition at line 31 of file TCCBlockFormatter.h.

Referenced by DigiToRaw(), and TCCBlockFormatter().

const int TCCBlockFormatter::kCardsPerTower = 5

static

Definition at line 27 of file TCCBlockFormatter.h.