ANL/NSE-25/41 Nuclear Science and Engineering Division About Argonne National Laboratory Argonne is a U.S. Department of Energy laboratory managed by UChicago Argonne, LLC under contract DE-AC02-06CH11357. The Laboratory’s main facility is outside Chicago, at 9700 South Cass Avenue, Argonne, Illinois 60439. For information about Argonne and its pioneering science and technology programs, see www.anl.gov. DOCUMENT AVAILABILITY Online Access: U.S. Department of Energy (DOE) reports produced after 1991 and a growing number of pre-1991 documents are available free at OSTI.GOV (http://www.osti.gov/), a service of the US Dept. of Energy’s Office of Scientific and Technical Information. Reports not in digital format may be purchased by the public from the National Technical Information Service (NTIS): U.S. Department of Commerce National Technical Information Service 5301 Shawnee Rd Alexandra, VA 22312 www.ntis.gov Phone: (800) 553-NTIS (6847) or (703) 605-6000 Fax: (703) 605-6900 Email: orders@ntis.gov Reports not in digital format are available to DOE and DOE contractors from the Office of Scientific and Technical Information (OSTI): U.S. Department of Energy Office of Scientific and Technical Information P.O. Box 62 Oak Ridge, TN 37831-0062 www.osti.gov Phone: (865) 576-8401 Fax: (865) 576-5728 ANL/NSE-25/41 prepared by Micheal A. Smith Nuclear Science and Engineering Division, Argonne National Laboratory ABSTRACT The Argonne Reactor Code (ARC) software system supports users in their fast reactor design goals by providing neutronic, thermal-hydraulic, and structural analysis capabilities. REBUS plays a pivotal role in the ARC system as the primary fuel cycle analysis capability for fast reactor problems. Over its 60 year history, ARC software usage with REBUS has been applied to numerous fast and thermal spectrum reactor analysis projects with good to excellent comparison against experiments. The RCT code is a later addition and uses the REBUS restart files to define its input. The RCT code was built to provide pin depletion details on EBR-II models and thus many features of RCT were specifically tailored to the needs of EBR-II models. Additional approximations were invoked which are likely only valid for the EBR-II reactor and the particular fuel management that was done for it. The purpose of the present work is to identify a set of test problems for RCT and assess the code coverage for those test problems. The goal is to document what parts of the existing RCT code are touched by the set of test problems and which are not. Because no detailed verification work has been done on RCT, the existing regression testing suite was chosen for the code coverage assessment. The code coverage analysis of RCT was performed with the Code Coverage Tool of the Intel Fortran compiler which requires modifications to the compilation of RCT. The detailed coverage tables are given for each part of RCT. As will be discussed and shown, some parts of the RCT capability that are known to be used by the EBR-II analysis work are not tested by the regression testing suite. These aspects should be resolved before major source code changes are taken for the RCT software. Because REBUS and DIF3D are not subroutines of RCT, the coverage changes in both of those codes is not altered by RCT. The same is true for all of the modules of DIF3D that are used by RCT such as SYSLIB and SEGLIB. 2 ANL/NSE-25/41 TABLE OF CONTENTS Abstract ........................................................................................................................................... 2 Table of Contents ............................................................................................................................ 3 List of Figures ................................................................................................................................. 4 List of Tables................................................................................................................................... 4 1. Introduction ............................................................................................................................... 5 2. RCT Overview .......................................................................................................................... 7 3. RCT Coverage Assessment ....................................................................................................... 8 3.1 Primary subroutines and driver of the RCT code ............................................................... 8 4. DIF3D and REBUS Coverage Update .................................................................................... 14 4.1 Source code in the directory rebus/source ........................................................................ 14 4.2 Source code in the directory rebus/dif3d/source ............................................................... 14 4.3 Source code in the directory rebus/dif3d/variantbasis/source ........................................... 14 4.4 Source code in the directory rebus/dif3d/arcmodules/source ........................................... 14 4.5 Source code in the directory rebus/dif3d/gnip4c_hmg4c/source ...................................... 15 4.6 Source code in the directory rebus/dif3d/syslib/source .................................................... 18 4.7 Source code in the directory rebus/dif3d/lapack/source and rebus/dif3d/linpack/source . 19 4.8 Source code in the directory rebus/dif3d/seglib/source .................................................... 19 5. Summary and Conclusions...................................................................................................... 21 References ..................................................................................................................................... 22 3 ANL/NSE-25/41 LIST OF FIGURES Figure 1.1 ARC Software Connections for Fast Spectrum Fuel Cycle Analysis .......................... 5 LIST OF TABLES Table 3.1.2 Covered or Partially Covered Files in the Directory source ....................................... 11 Table 4.4.1 Covered or Partially Covered Files in the Directory rebus/dif3d/arcmodules/source 14 Table 4.5.1 Covered or Partially Covered Files in the Directory dif3d/gnip4c_hmg4c/source ..... 15 Table 4.6.1 Covered or Partially Covered Files in the Directory rebus/dif3d/syslib/source ......... 18 Table 4.8.1 Covered or Partially Covered Files in the Directory rebus/dif3d/seglib/source ......... 20 4 ANL/NSE-25/41 1. Introduction The Argonne Reactor Code (ARC) software package consists of 13 primary and many secondary pieces of software that are connected through interface files. The ARC software suite supports users in their fast reactor design goals by providing neutronic, thermal-hydraulic, and structural analysis capabilities. Over the past decades, the ARC software suite has been applied to numerous fast and thermal spectrum reactor analysis projects. For those cases with experimental measurements, ARC performed exceptionally well, yielding results of consistent accuracy to those produced by the Monte Carlo method at a fraction of the cost. At present, the ARC software suite is primarily used for the analysis of advanced reactors. The typical design process workflow for the fuel cycle analysis is shown in Figure 1.1, with the cited codes named in orange boxes, and as can be seen, REBUS [1]-[3] and DIF3D [1][5]-[10] are central to the entire ARC system. Figure 1.1 ARC Software Connections for Fast Spectrum Fuel Cycle Analysis The ARC Software Quality Assurance (SQA) Program [18] is aimed to provide the controls and processes necessary to enable software improvement while meeting user and program sponsor requirements. The Software QA Plan (SQAP) delineates the SQA Program framework for the ARC software by describing the Program activities, organization, and documentation, and by clearly defining the interconnection of all program items. The analysis presented in this report is part of the SQAP for each piece of the ARC software. 5 ANL/NSE-25/41 Much of the ARC software has its origins in the 1960s with many components added or upgraded over the following 60 years of continuous development. RCT [4] was built around REBUS in the 1990s and directly reads the restart files generated by REBUS as part of its input. REBUS is built around the DIF3D code and thus is geometrically limited to the models that can be constructed with DIF3D. RCT is further restricted to the hexagonal 3D geometries of DIF3D-Nodal and DIF3D- VARIANT. Today, the diffusion and transport capabilities of DIF3D-VARIANT are primarily used in the reactor design process with some scattered usage of DIF3D-FD and DIF3D-Nodal and thus this limitation is not really relevant. RCT uses just a few subroutines from REBUS and only links to a few modules that are common to DIF3D and REBUS (e.g. SYSLIB, SEGLIB and GNIP4C). As a consequence, this code coverage assessment will primarily focus on the RCT specific source code coverage but give details on the other connected components as secondary importance. The purpose of the present work is to identify a set of test problems for RCT and document the code coverage of RCT for those test problems. Because RCT is distributed with a regression test suite and there is at present no verification work done on it, the set of test problems for the code coverage assessment will be the existing test suite. In the following sections, the parts of the existing RCT coding which are or are not touched by the set of test problems are documented. The coverage of the connected components is included in a follow up section. The code coverage analysis was performed using the Code Coverage Tool of the Intel Fortran compiler (gprof and codecov). To use this capability, the compilation of RCT must be modified to create a special RCT executable which is different from the production executable. The existing regression test suite will be used to verify that the new executable is correct, in addition to serving as the code coverage assessment test cases. A detailed code coverage report is generated by the Intel Code Coverage Tool, which shows the coverage of subroutines/functions, blocks within each subroutine/function, and the frequency of execution of each subroutine/function. With this code coverage report, it is easy to figure out which subroutines/functions are fully covered, fully uncovered, or partially covered in RCT. A review of the RCT source files that are not fully covered was performed and some indication is reported here for the lack of coverage and its importance for the RCT code usage. One unique behavior of the version of gprof used is that it does not provide call information for all subroutines within a file and thus some details on the call history are not provided in all of the tables here. 6 ANL/NSE-25/41 2. RCT Overview RCT was designed around the REBUS code for the purpose of doing EBR-II pin depletion analysis. RCT is not independent of REBUS in that it requires the user to use REBUS in a non-equilibrium calculation mode and the restart files generated by REBUS to define the RCT input. RCT uses the exact same binary interface files and variable naming scheme as in DIF3D. Over its extensive history, REBUS has been applied to numerous fast and thermal spectrum reactor analysis projects, giving users the confidence needed to select it for use with their reactor analysis project. RCT on the other hand was only constructed for the specific need to do pin level depletion analysis for ongoing EBR-II PIE measurements and thus has only ever been involved in those tasks. Using RCT to carry out pin depletion analysis of all assemblies of an existing core model is not advisable today as it is not an integral part of the search procedure used in REBUS. For more details, the user is directed to the RCT manual [4]. 7 ANL/NSE-25/41 3. RCT Coverage Assessment Because of the limited computational capabilities, the RCT code was developed to be a series of executables that are called in sequence which is termed the modular system. The executables passed information via interface files and had specific roles as part of the RCT execution (e.g., input processing, depletion calculation, and output processing). Today the structure of the RCT modular form is maintained but it is not used, as the driver routines of each non-driver module were converted to subroutines of the driver module. While the approach still uses interface files to transfer information between modules of RCT, the maintenance and use of the software is much easier. All of the modules used in RCT are listed in Table 3.1. Table 3.1 Modules used by RCT Modules Directory locations RCT (driver) source REBUS rebus/source Fortran 90 modules rebus/dif3d/arcmodules/source SYSLIB rebus/dif3d/syslib/source SEGLIB rebus/dif3d/seglib/source G4CH4C rebus/dif3d/gnip4c_hmg4c/source When RCT is compiled, only a few source files in each module (directory) are compiled and linked as part of the executable which is not the case for DIF3D and REBUS. This is consistent with the purpose of RCT as a simple manipulation of the flux solution coming from DIF3D at each time point of a REBUS depletion calculation. The additional complexity is that it also carries out a pin depletion calculation for an identified mesh in the existing REBUS model. Unlike DIF3D and REBUS which have detailed software verification reports [11],[13], RCT does not have one. RCT is distributed with 7 test cases and most have 4 independent RCT calculations. Because RCT was primarily developed for and used on EBR-II, one would expect that one of the test cases would be derived from EBR-II. This is not the case and instead all of the test cases are for a single 60 degree periodic geometry model. The first 5 test cases are focused on DIF3D-Nodal and the last two were added when support was included for DIF3D-VARIANT in RCT. None of these test cases can be considered verification of what the software does and thus the code coverage assessment will highlight those aspects of the RCT capabilities that are not handled because of the limited testing suite. The previous work done on code coverage for DIF3D and REBUS can be found in references [15] and [16]. 3.1 Primary subroutines and driver of the RCT code The primary subroutines of the RCT code are stored in the directory source in addition to the driver routines. The code coverage of all of the files is shown in Table 3.1.1 and Table 3.1.2. Only a single subroutine used for debugging was uncovered as detailed in Table 3.1.2. For the uncovered or partially-covered subroutines, a description of the uncovered parts is also given in each table. For 8 ANL/NSE-25/41 the covered source files, the number of calls by the test package are also listed, showing the usage frequency of each source file for the 7 test cases (22 uses of RCT executable). Starting with the code coverage of the driver and module subroutines in Table 3.1.1, one can see that several of them are partially covered and only the driver routine is fully covered. As detailed in the uncovered portion description, all of the uncovered sections deal with fatal errors and branching for alternative output edits. One item of interest is the older version of REBUS. It is not clear why two versions of REBUS were supported, but this aspect only has to do with the size of the storage arrays for files which has been periodically increased over the years. For the partially-covered subroutines listed in Table 3.1.2, one can see that there are again many parts dealing with fatal errors that are not covered and thus can be ignored. Much of the “blocking” related aspects deal with when the memory limits on the machine were very small and thus these branches are likely to never be reached today (ISOTXS file would have to be generated in that state in order for REBUS/RCT to even deal with it). Similar to this are the various chi treatments. The chi cross section data is not actually used by RCT in its calculations and thus the only consequence of not covering those branches is that it might segfault or obtain incorrect data. A review of the source code does not indicate there would be any such problem and given no anomalous results were generated by recent usage of RCT on EBR-II with an updated ISOTXS file and model that uses the isotopic chi treatment, one can ignore these issues at this time. The cited branch for NRASS=1 in subroutine rdgeo.f pertains to a model being constructed in REBUS that does not use coarse-mesh mapping to regions. It is very difficult for a user to naturally create such a problem through the DIF3D/REBUS interface and while it should be tested, it should not be a great concern as the only difference in source coding is how the MR array in GEODST is handled. The cited LABELS treatment issues in subroutine rdstak.f are interesting but again unlikely to be of concern in RCT. From inspection of the RCT source code, there is no treatment available for the polynomial dependent cross sections. So while the REBUS calculation might use interpolated cross sections (not done for EBR-II), the RCT is incapable of doing so. The various untreated aspects of LABELS are simply not going to occur for the EBR-II models except for maybe the moving control rods and since those aspects are not relevant to the actual RCT calculation not reading them in correctly is of little importance. It is important to note that the GEODST files at each time point that RCT relies upon will have the correct control rod positioning details. One code coverage detail in Table 3.1.2 that is important to resolve is the lack of coverage due to 120 periodic and full core models. Clearly those models need to be added to the testing as all EBR- II problems are full core. Reviewing the source code, there is virtually nothing to worry about as most of the branching just changes the number of sectors that are used when searching for matching region names. A second code coverage aspect that needs to be resolved is the dependence of the testing on the number of active isotopes (NACI). For whatever reason, the developers of RCT choose to unroll 9 ANL/NSE-25/41 the various loops in the code to allow for vectorization. This is an artifact of older compiler methodologies and is considered very unwise today. To explain, instead or programming a loop to go from 1:10 over a set of instructions (e.g. I=1,10 for operation VecA(I)=10.0*VecB(I)), the developers wrote the loop to go 1:10 stepping by 4. This forces the developer to write the loop twice. In the first loop, the instruction would be duplicated four times (e.g. VecA(I)=,VecA(I+1)=,VecA(I+2)=,VecA(I+3)=). In the second loop the parts that the first loop does not cover must be handled (in the example it would be positions 9 and 10). That second loop of course is identical to the one needed for the case when no stepping is used (or stepping by 1). In the uncovered sections of various subroutines, those follow-on loops are not touched because the number of active isotopes in the problem being executed is a perfect multiple of 2 and 4. Further code coverage is wise although it would be preferable to simply eliminate the unrolling of the loops to simplify the coding. A final code coverage aspect in Table 3.1.2 that should be resolved is the fixed format reading branches for the RCT input are not covered. That is because all of the examples use free format inputs. It should be a minor effort to modify some of the existing test cases to use a fixed format input for A.RCT instead of a free format. From the preceding code coverage assessment, one can conclude that most of the critical parts of the RCT code are tested. While some additional coverage would be wise, those aspects should be of no importance with regard to the actual results generated by RCT. 10 ANL/NSE-25/41 Table 3.1.1 Covered or Partially Covered Driver RCT Files in the Directory source Further # of calls by the Coverage File Name Subroutine Uncovered portion description Coverage test package Status needed? 0gotows.F GOTOWS 198 Partially Fatal error and branch for alternative output file No 0rct1.F RCT1 22 Partially Fatal error and branch for older version of REBUS No 0rct2.F RCT2 22 Partially Fatal error and branch for alternative output file No 0rct3.F RCT3 22 Partially Fatal error and branch for alternative output file No 0stp034.F 22 Fully No Table 3.1.2 Covered or Partially Covered Files in the Directory source # of calls Further Coverage File Name Subroutine by the test Uncovered portion description Coverage Status package needed? Fatal errors, fixed format reading, bad input arct06.f ARCT06 22 Partially No corrections arct07.f ARCT07 2 Partially Fatal errors and fixed format reading No arct08.f ARCT08 2 Partially Fatal errors and fixed format reading No arct09.f ARCT09 2 Partially Fatal errors and fixed format reading No arct0.f ARCT0 22 Fully Fatal errors and fixed format reading No assden.f ASSDEN 328 Fully No assflu.f ASSFLU 328 Fully No Branching for inputs that likely cannot be bocden.f BOCDEN 66 Partially No created bunnum.f BUNNUM 38,640 Fully No Fatal errors and card type 7,8,9 setup of the chkbcd.f CHKBCD 22 Partially Yes special pin input option Total scattering data provided on ISOTXS comp4d.f COMP4D 2,134 Partially No instead of elastic, inelastic, n2n. Branch for blocked scattering data in comp7d.f COMP7D 2,134 Partially No COMPXS cpyf.f CPYF 3,952 Partially Branch for blocked data on STACK No dacon.f DACON 1,312 Fully No dcymtx.f DCYMTX 81 Fully No 11 ANL/NSE-25/41 dendcy.f DENDCY 2,316 Fully No eocden.f EOCDEN 66 Fully No getik.f GETIK 132 Partially Branch for 120 periodic and full core models Yes Fatal errors and branch for 120 periodic and hexreg.f HEXREG 44 Partially Yes full core models File chi and isotopic chi data reading of isoiso.f ISOISO 22 Partially No ISOTXS mesh.f MESH 66 Fully No poly13.f POLY13 66 Fully No poly19.f POLY19 66 Fully No pwpeak.f PWPEAK 656 Fully No RCT20 14 Fully No rct20.f RCT20_PRINT 0 Uncovered Debug print routine called from rct20 No rct30.f RCT30 22 Fully No rctip6.f RCTIP6 22 Partially Fatal Errors No rctip7.f RCTIP7 22 Partially Fatal Errors and Output pagination control No rctip8.f RCTIP8 22 Partially Fatal Errors and Special setup of Special Pins Yes rctip9.f RCTIP9 22 Partially Fatal Errors and special output branch No rdcxs.f RDCXS 28 Partially Zone chi matrix reading of COMPXS No rdgeo.f RDGEO 44 Partially Branch for NRASS=1 GEODST setup No Fatal errors, polynomial XS treatments and rdstak.f RDSTAK 5 Partially No other features of LABELS not typically used rdzna.f RDZNA 44 Partially Debug print output branch No Debug print output branch and card type recont.f RECONT 656 Partially Yes 7,8,9 special pin input setups rnhfly.f RNHFLY 28 Partially Debug print output No Number of active isotopes is not an exact spinpw.f SPINPW 656 Partially Yes multiple of 2 spnden.f SPNDEN 984 Fully No spnflu.f SPNFLU 984 Fully No srpeak.f SRPEAK 15,744 Fully No Case where number of active isotopes is not udecay.f UDECAY 38,721 Partially Yes an exact multiple of 4 wrctdn.f WRCTDN 22 Partially Case with record types 5 and 6 in RCTDEN Yes wrctfx.f WRCTFX 22 Fully No 12 ANL/NSE-25/41 wrctpw.f WRCTPW 22 Fully No wrflux.f WRFLUX 656 Fully No xpower.f XPOWER 132 Fully No 13 ANL/NSE-25/41 4. DIF3D and REBUS Coverage Update The preceding RCT subroutine coverage is the primary focus on this work. Because some other parts of the code system were used by RCT, primarily those associated with DIF3D, the code coverage is reiterated here. 4.1 Source code in the directory rebus/source Only the savprg.f subroutine of REBUS is used by RCT and no change in the coverage occurs between it and that seen in the REBUS testing. 4.2 Source code in the directory rebus/dif3d/source No subroutines in rebus/dif3d/source are included in RCT and thus no code coverage data is included here. 4.3 Source code in the directory rebus/dif3d/variantbasis/source No subroutines in rebus/dif3d/variantbasis/source are included in RCT and thus no code coverage data is included here. 4.4 Source code in the directory rebus/dif3d/arcmodules/source The Fortran coding in the directory rebus/dif3d/arcmodules/source of the RCT distribution primarily provides self-contained memory allocation and reading/writing of the binary interface files created in the ARC suite of codes. The RCT code uses the GEODST, NHFLUX, COMPXS, and NE_Kind modules in that directory. Reviewing the source code of RCT, these functions are only used when the DIF3D-VARIANT option of DIF3D is used in the REBUS model. The gprof compilation of the rct code failed to report the number of calls made to these modules, but the code coverage details were provided for all files and are detailed in Table 4.4.1. Table 4.4.1 Covered or Partially Covered Files in the Directory rebus/dif3d/arcmodules/source File Name Subroutine Coverage Details Only uses the define, zero, and void functions and COMPXS.F90 Multiple fully covers except for fatal error branches and delay cross section data inclusion Does not cover delay cross section data inclusion, COMPXS_IMPORT.F90 does not cover zone chi treatment Only uses the define and void functions and fully GEODST.F90 Multiple covers except for fatal errors Does not cover the ascii read subroutine, only the GEODST_IMPORT.F90 binary Does not cover 1D or 2D domain options, fatal error GEODST_IMPORT_BINARY.F90 branches or NRASS=1 branch NE_Kind.F90 Multiple Only uses GetFreeLogicalUnit and UpperCaseString Only touches the define, void, and intel_ideal_open NHFLUX.F90 Multiple functions. Fatal error branches are missed. Only misses the fatal error and debugging output NHFLUX_IMPORT.F90 branches 14 ANL/NSE-25/41 There are no concerning code coverage aspects of these subroutines that are concerning as PERSENT, GAMSRC, and DASSH will more heavily test out all of these modules and subroutines. 4.5 Source code in the directory rebus/dif3d/gnip4c_hmg4c/source The directory rebus/dif3d/gnip4c_hmg4c/source in the RCT distribution contains the GNIP4C and HMG4C submodules of DIF3D. The GNIP4C program imports the A.NIP3 input data and packages it into the GEODST, NDXSRF, ZNATDN, LABELS, ISOTXS, and FIXSRC interface files. The HMG4C program generates the COMPXS file given the presence of GEODST, NDXSRF, ZNATDN, LABELS and ISOTXS files. The coverage summary of those subroutines identified with partial coverage in DIF3D and REBUS is shown in Table 4.5.1. Many of the previously partially or fully covered subroutines in Table 4.5.1 are not covered by the RCT test suite. In fact, most of the covered parts of the module only correspond to the HMG4C usage in RCT. Because DIF3D and REBUS cover these subroutines more completely, the code coverage assessment for RCT is of little importance. Table 4.5.1 Covered or Partially Covered Files in the Directory dif3d/gnip4c_hmg4c/source # of calls by the # of calls by the # of calls by the File Name Subroutine DIF3D tests REBUS tests RCT tests ads.f ADS 4 0 0 ads1.f ADS1 3 0 0 ads2.f ADS2 37 0 0 ads3.f ADS3 1 0 0 anip01.f ANIP01 87 86 0 anip02.f ANIP02 87 86 0 anip03.f ANIP03 75 86 0 anip04.f ANIP04 75 86 0 anip05.f ANIP05 29 16 0 anip06.f ANIP06 34 18 0 anip07.f ANIP07 23 84 0 anip09.f ANIP09 57 82 0 anip10.f ANIP10 11 0 0 anip11.f ANIP11 11 0 0 anip12.f ANIP12 3 10 0 anip13.f ANIP13 4 57 0 anip14.f ANIP14 147 172 0 anip15.f ANIP15 75 86 0 anip19.f ANIP19 12 0 0 anip21.f ANIP21 3 4 0 anip22.f ANIP22 3 4 0 anip23.f ANIP23 2 8 0 anip24.f ANIP24 1 0 0 anip25.f ANIP25 1 0 0 anip34.f ANIP34 8 2 0 anip35.f ANIP35 4 90 0 anip37.f ANIP37 4 90 0 anip39.f ANIP39 2 0 0 anip40.f ANIP40 7 0 0 anip41.f ANIP41 7 0 0 15 ANL/NSE-25/41 anip42.f ANIP42 7 0 0 anip43.f ANIP43 59 85 0 anip44.f ANIP44 1 4 0 aniphx.f ANIPHX 41 68 0 atdn3.f ATDN3 77 1,634 44 bcdxst.f BCDXST 87 86 0 bleep.f BLEEP 233 19,150 0 chek09.f CHEK09 98 104 0 cmesh.f CMESH 106 104 0 copier.f COPIER 4 90 0 cpylbg.f CPYLBG 5 4 0 domods.f DOMODS 4 90 0 edfpws.f EDFPWS 9 0 0 edsorc.f EDSORC 12 0 0 edsrch.f EDSRCH 6 8 0 edtiso.f EDTISO 1 0 0 edtmat.f EDTMAT 4 57 0 edtr5.f EDTR5 2 0 0 edtr6.f EDTR6 2 0 0 edtr8.f EDTR8 2 0 0 edtrod.f EDTROD 3 11 0 edtxs1.f EDTXS1 9 0 0 edtxs2.f EDTXS2 60 0 0 egeods.f EGEODS 59 85 0 fadens.f FADENS 72 86 0 farset.f FARSET 39,149 1,559,136 38,412 fgeods.f FGEODS 75 86 0 fispec.f FISPEC 77 1,634 44 fmesh.f FMESH 106 104 0 getbc.f GETBC 75 86 0 getbuc.f GETBUC 75 86 0 gethgt.f GETHGT 3 10 0 getiso.f GETISO 72 86 0 getmat.f GETMAT 4 57 0 getmr.f GETMR 150 172 0 getmsh.f GETMSH 75 86 0 getreg.f GETREG 75 86 0 getset.f GETSET 72 86 0 getszn.f GETSZN 147 172 0 getzon.f GETZON 147 172 0 gnip4c.F GNIP4C 87 86 0 hexpc1.f HEXPC1 594 1,000 0 hexpc2.f HEXPC2 548 1,026 0 hexpc3.f HEXPC3 1,042 1,907 0 hexpc4.f HEXPC4 614 1,216 0 hexpc6.f HEXPC6 509 915 0 hexpc7.f HEXPC7 8 30 0 hexpic.f HEXPIC 9 45 0 hmg4c.F HMG4C 92 1,634 44 idlr13.f IDLR13 77 1,634 44 indexx.f INDEXX 7 4 0 isor14.f ISOR14 77 1,634 44 isor58.f ISOR58 77 1,634 44 16 ANL/NSE-25/41 lochex.f LOCHEX 11,361 5,264 0 makmsh.f MAKMSH 59 85 0 makrod.f MAKROD 1 4 0 maxbnd.f MAXBND 77 1,634 44 mpgeod.f MPGEOD 16 26 0 mshmap.f MSHMAP 22 26 0 ovl1.f OVL1 77 1,634 44 ovl2.f OVL2 77 1,634 44 ovl3.f OVL3 77 1,634 44 ovl4.f OVL4 9 0 0 ovl5.f OVL5 77 1,634 44 prgeod.f PRGEOD 59 85 0 prntix.f PRNTIX 239 457 0 ranip1.f RANIP1 75 86 0 ranip2.f RANIP2 72 86 0 rcmesh.f RCMESH 106 58 0 rdatdn.f RDATDN 77 1,634 44 rdlabl.f RDLABL 77 1,634 44 rdndx.f RDNDX 77 1,634 44 redmsh.f REDMSH 41 68 0 rwisot.f RWISOT 56 84 0 set2d.f SET2D 14 8 0 sethex.f SETHEX 300,823 204,522 0 sordat.f SORDAT 12 0 0 sordif.f SORDIF 2 0 0 sorlab.f SORLAB 7 0 0 sormsh.f SORMSH 12 0 0 sornzn.f SORNZN 6 0 0 sorrod.f SORROD 1 4 0 sort13.f SORT13 4 57 0 sort14.f SORT14 72 86 0 srch3d.f SRCH3D 1 0 0 srch4d.f SRCH4D 1 4 0 srch6d.f SRCH6D 1 0 0 srflt.f SRFLT 5,436 6,891 0 svscat.f SVSCAT 39,149 1,559,136 38,412 svxs.f SVXS 166 0 0 trigom.f TRIGOM 386 402 0 triplt.f TRIPLT 3 6 0 update.f UPDATE 638 58,049 4,268 valu2d.f VALU2D 333 7,235 0 volreg.f VOLREG 246 230 0 wndxsr.f WNDXSR 72 86 0 wrec1.f WREC1 77 1,634 44 wrec2.f WREC2 77 1,634 44 wrec3.f WREC3 460 37,078 5,016 wrec4.f WREC4 460 37,078 5,016 wrrods.f WRRODS 1 4 0 wrs.f WRS 12 0 0 wrs0.f WRS0 140 0 0 wrs1.f WRS1 68 0 0 wrs2.f WRS2 75 0 0 wrsorc.f WRSORC 87 86 0 17 ANL/NSE-25/41 wrsrch.f WRSRCH 87 86 0 wrtrod.f WRTROD 1 4 0 4.6 Source code in the directory rebus/dif3d/syslib/source The directory rebus/dif3d/syslib/source in the RCT distribution is one of the major libraries of DIF3D which contains BPOINTER memory management and file I/O among other character manipulation functions. The full and partially covered files are listed in Table 4.6.1 with just the updated call information for RCT. As can be seen, a majority of the use of RCT is consistent with usage in DIF3D and there is no reason to believe that RCT covers something that DIF3D or REBUS did not cover. The uncovered files in this source directory were also uncovered in RCT and thus they are not included here. Table 4.6.1 Covered or Partially Covered Files in the Directory rebus/dif3d/syslib/source # of calls by the # of calls by the # of calls by the File Name Subroutines DIF3D tests REBUS tests RCT tests cequat.f CEQUAT 1,104 12,452 0 CHRFLT 286,758 2,146,552 68,526 chrflt.f CFCOPY 286,758 0 0 chrset.f CHRSET 385 2,574 0 cred.f CRED 60 0 0 DOPC 1,394 20,186 110 DOPDEL 83 207 0 dopc.f DOPFIL 83 0 DOPLOK 83 0 DOPSET 449 8,020 88 DRED 30,127 359,591 0 dred.f DREAD 30,127 0 DRED_DOIT 0 0 0 error.F ERROR 1,945 28,873 693 fequat.f FEQUAT 0 2,503,256 2,481 FFORM 3,772 20,477 982 fform.f FFORM1 3,772 0 0 FFORM2 15,013 81,217 6,322 FLTCHR 1,849,042 32,732,597 499,196 fltchr.f FCCOPY 1,849,042 0 fltset.f FLTSET 168,442 1,627,263 10,034 iequat.f IEQUAT 0 2,966,258 8,150 IGTLCM 178 6,834 66 JGT 178 352 FRELCM 122 6,748 66 igtlcm.F LOCFLC 0 0 0 LOCFWD 1,294 25,406 308 LOCF 1,188 28,098 594 intset.f INTSET 6,695 55,922 154 LINES 73,023 768,269 78,268 lines.f PGCNTR 73,023 0 0 LINKMD 112 0 0 linkmd.F FINOUT 0 0 0 LUNREF 58,179 2,821,427 122,789 lunref.f SEKAPP 0 2,088 0 18 ANL/NSE-25/41 makddn.f MAKDDN 4,899 11,645 902 msyncw.F MSYNCW 1,003 7,020 154 POINTR 396 9,366 132 PUTPNT 0 0 0 BULK 396 9,366 132 ECMREF 0 4,992 66 BPFREE 396 9,449 132 WIPOUT 25,078 388,460 6,886 GETPNT 40,576 0 IGET 40,576 711,858 5,808 IPT2 26,117 577,790 5,852 PUTM 22,181 389,733 1,892 IPTERR 7,259 138,353 704 pointr.F ILAST 3,242 29,018 264 REDEF 712 867 0 REDEFM 377 1,900 0 PURGE 4,204 43,711 220 STATIS 0 3,912 132 STATIS_DOIT 0 0 0 PRTI1 0 0 0 PRTI2 0 0 0 PRTR1 0 0 0 PRTR2 0 0 0 PRTECM 0 0 0 REED 20,260 1,375,940 61,947 reed.f RREAD 20,260 0 seek.f SEEK 9,147 202,308 3,326 sekphl.f SEKPHL 3,027 23,923 264 space.f SPACE 6,676 8,063 44 squeze.f SQUEZE 2,204 9,507 164 srlab.f SRLAB 1,596,408 27,247,001 428,269 TIMER 12,278 143,300 726 timer.F DSECOND 12,278 0 0 TLEFT 0 0 0 4.7 Source code in the directory rebus/dif3d/lapack/source and rebus/dif3d/linpack/source The files in directory rebus/dif3d/lapack/source and rebus/dif3d/linpack/source in the RCT distribution were not created as part of the DIF3D, REBUS, or RCT development. While they contain some useful capabilities for linear algebra manipulations, they are not used by RCT and thus no coverage details are given here. 4.8 Source code in the directory rebus/dif3d/seglib/source The files in directory rebus/dif3d/seglib/source of the RCT distribution were developed as a common location separate from SYSLIB for a variety of ARC codes. It primarily contains the SUMMAR and RODPOS subprograms of ARC. SUMMAR produces isotopic reaction rate summary data for a given DIF3D input while RODPOS is used by REBUS to adjust the control rod positions at each time step. The subroutines with full or partial code coverage are listed in Table 4.8.1 and only the updated call information for the RCT tests is provided. As seen, only a single 19 ANL/NSE-25/41 function is used from this library which is associated with the HMG4C usage in RCT. The previously uncovered files are not used by RCT and thus those details are not included here. Table 4.8.1 Covered or Partially Covered Files in the Directory rebus/dif3d/seglib/source # of calls by the # of calls by the # of calls by the File Name Subroutine DIF3D tests REBUS tests RCT tests adarea.f ADAREA 9 766 0 addiso.f ADDISO 11 769 0 bleep.f BLEEP 233 19,150 0 calc.f CALC 11 772 0 codecd.f CODECD 206,916 569,424 2,679 modrec.f MODREC 20 2,233 0 preplt.F PREPLT 11 88 0 prnsum.f PRNSUM 11 772 0 rdcmin.f RDCMIN 11 772 0 rdfrpn.f RDFRPN 11 772 0 rdgeom.f RDGEOM 11 772 0 rdmixr.f RDMIXR 11 772 0 rdmr.f RDMR 23 2,193 0 rdoutm.f RDOUTM 11 772 0 reacto.f REACTO 11 772 0 scan.F SCAN 56 86 22 stmesh.f STMESH 28 2,260 0 STUFF 199 86 22 stuff.F STUFF1 88 86 22 summar.F SUMMAR 11 772 0 sumre1.f SUMRE1 11 772 0 udoit1.F UDOIT1 87 0 0 udoit2.F UDOIT2 97 88 0 udoit3.F UDOIT3 87 0 0 udoit4.F UDOIT4 87 86 0 xsrd.f XSRD 11 772 0 20 ANL/NSE-25/41 5. Summary and Conclusions In the preceding analysis the coverage report for each submodule of RCT was given. The submodule refers to separable libraries (LINPACK, LAPACK, G4CH4C, etc.) that are included either partially or fully as part of RCT (and DIF3D or REBUS) even though the development of the source coding was not specific to RCT. For the RCT specific source code, most of the subroutines were fully or partially covered to the point where additional testing is not necessary. Three primary cases in RCT were identified that need additional testing. The first is that the existing test suite only uses a 60 degree periodic geometry model and thus branching for 120 periodic and full core models are not tested. There are also some branches that are not tested because of loop unrolling scheme taken by the authors of RCT. Finally, the fixed format reading of the RCT input is not tested and new test cases should be added for this specific purpose. No meaningfully important code coverage details were found for all of the connected modules (SYSLIB, SEGLIB, etc…) of RCT beyond that which was identified for DIF3D and REBUS. 21 ANL/NSE-25/41 REFERENCES [1] R. P. Hosteny, “The ARC System Fuel Cycle Analysis Capability, REBUS-2,” ANL-7721, Argonne National Laboratory (1978). [2] B. J. Toppel, “A User’s Guide to the REBUS-3 Fuel Cycle Analysis Capability,” ANL-83-2, Argonne National Laboratory (1983). [3] W. S. Yang, M. A. Smith, “Theory Manual for the Fuel Cycle Analysis Code REBUS,” ANL/NE-19/21, Argonne National Laboratory (2020). [4] W. S. Yang, M. A. Smith, “RCT: REBUS Based Pin Power Reconstruction Using the DIF3D- Nodal and DIF3D-VARIANT Options,” ANL/NE-14/15, Argonne National Laboratory (2014). [5] K. L. Derstine, “DIF3D: A Code to Solve One-, Two-, and Three-Dimensional Finite- Difference Diffusion Theory Problems,” ANL-82-64, Argonne National Laboratory (1984). [6] R. D. Lawrence, "The DIF3D Nodal Neutronics Option for Two- and Three-Dimensional Diffusion Theory Calculations in Hexagonal Geometry," ANL-83-1, Argonne National Laboratory, March (1983). [7] R. D. Lawrence, “Progress in Nodal Methods for the Solution of the Neutron Diffusion and Transport Equations,” Prog. Nucl. Energy, 17, 271 (1986). [8] M. R. Wagner, “Three-Dimensional Nodal Diffusion and Transport Theory for Hexagonal-Z Geometry,” Nucl. Sci. Eng., 103, 377-391 (1989). [9] G. Palmiotti, E. E. Lewis, and C. B. Carrico, “VARIANT: VARIational Anisotropic Nodal Transport for Multidimensional Cartesian and Hexagonal Geometry Calculation,” ANL-95/40, Argonne National Laboratory (1995). [10] M. A. Smith, E. E. Lewis, and E. R. Shemon, “DIF3D-VARIANT 12.0, A Decade of Updates,” ANL/NE-14/1 Rev. 1, Argonne National Laboratory (2025). [11] M. A. Smith and A. Nelson, “Verification of the REBUS Software,” ANL-VTR-50 Rev. 2, Argonne National Laboratory (2025). [12] M. A. Smith and A. Nelson, “Requirements Description of the REBUS Software for the Versatile Test Reactor,” ANL-VTR-38, Argonne National Laboratory (2021). [13] A. Nelson and M. A. Smith, “Verification of the DIF3D Software to Support Fast Reactor Analysis,” ANL/NSE-20/3, Argonne National Laboratory (2020). [14] M. A. Smith, A. Nelson, and F. Heidet, “Requirements Description of the DIF3D Software for the Versatile Test Reactor,” ANL-VTR-19, Argonne National Laboratory (2019). [15] Z. Zhong, M. A. Smith, C. H. Lee, “Code Coverage Status of the ARC Code DIF3D,” ANL/NSE-23/65 Rev. 1, Argonne National Laboratory (2025). [16] Z. Zhong, M. A. Smith, C. H. Lee, “Code Coverage Status of the ARC Code DIF3D,” ANL/NSE-23/65 Rev. 1, Argonne National Laboratory (2025). [17] Micheal. A. Smith and Zhaopeng Zhong, “Code Coverage Status of the ARC Code REBUS,” ANL/NSE-25/36, Argonne National Laboratory (2025). 22 ANL/NSE-25/41 [18] K. Kiesling and M. A. Smith, “Software Quality Assurance Plan,” ANL/NSE-23/13, Argonne National Laboratory (2023). 23 ANL/NSE-25/41 Nuclear Science and Engineering Division Argonne National Laboratory 9700 South Cass Avenue, Bldg. 208 Argonne, IL 60439-4842 www.anl.gov Argonne National Laboratory is a U.S. Department of Energy laboratory managed by UChicago Argonne, LLC