# Invocation command line: # /root/aocc/SPECmpi2007/bin/runspec -c aocc40-mpi2007.cfg --size=mref --tune=all --reportable --rank=96 --iterations=3 --rebuild medium # output_root was not used for this run ############################################################################ # Only modify the binary label extension if you plan to rebuild the binaries. %define ext amd-aocc40-mpi2007 % define build_ncpus 16 # controls number of simultaneous compiles preenv = 1 makeflags = --jobs=%{build_ncpus} --load-average=%{build_ncpus} preENV_OMP_SCHEDULE = static preENV_OMP_DYNAMIC = false preENV_OMP_THREAD_LIMIT = 128 preENV_GOMP_CPU_AFFINITY = 0-128 preENV_OMP_STACKSIZE = 128M preENV_KMP_LIBRARY = turnaround preENV_KMP_BLOCKTIME = 200 preENV_OMP_NESTED = FALSE preENV_OMP_PLACES = threads ################################################################################ # Header settings ################################################################################ backup_config = 1 # set to 0 if you do not want backup files bench_post_setup = sync # command_add_redirect: If set, the generated ${command} will include # redirection operators (stdout, stderr), which are passed along to the shell # that executes the command. If this variable is not set, specinvoke does the # redirection. NOTE: this value must be "yes" for the affinity generation # section of this config file. command_add_redirect = no check_md5 = yes env_vars = no flagsurl000 = http://www.spec.org/mpi2007/flags/amd2021_flags.xml ignore_errors = no size = mref use_version_url = 1 use_submit_for_speed = 1 # label: Arbitrary string, tags your binaries & directories. ext = %{ext} line_width = 1020 log_line_width = 1020 mean_anyway = yes output_format = all reportable = yes size = mtest,mtrain,mref teeout = no teerunout = yes tune = base,peak basepeak = 1 default: CC = mpicc CXX = mpic++ FC = mpif90 CLD = mpicc CXXLD = mpic++ FLD = mpif90 CC_VERSION_OPTION = --version CXX_VERSION_OPTION = --version FC_VERSION_OPTION = --version ranks = 64 submit = mpirun --allow-run-as-root -np $ranks $command default:# data model applies to all benchmarks MATHLIBOPT = -lamdlibm -lm ################################################################################ # Tuning Flags ################################################################################ 121.pop2=default=default=default: CPORTABILITY = -DSPEC_MPI_CASE_FLAG 126.lammps=default=default=default: CXXPORTABILITY = -DMPICH_IGNORE_CXX_SEEK 127.wrf2=default=default=default: CPORTABILITY = -DSPEC_MPI_CASE_FLAG -DSPEC_MPI_LINUX -Wno-return-type #FPORTABILITY = -Mbyteswapio ################################################################# # Optimization flags ################################################################# default=default=default=default: EXTRA_CPORTABILITY = -DSPEC_MPI_LP64 COPTIMIZE = -Ofast -flto -ffast-math -march=znver4 CXXOPTIMIZE = -Ofast -flto -ffast-math -march=znver4 FOPTIMIZE = -Ofast -flto -ffast-math -march=znver4 -funroll-loops EXTRA_LIBS = -lamdlibm -lm -ljemalloc -lflang -flto EXTRA_FLIBS = -lamdlibm -lm -flto # The following section was added automatically, and contains settings that # did not appear in the original configuration file, but were added to the # raw file after the run. default: sw_base_ptrsize = 64-bit sw_peak_ptrsize = 64-bit sw_other = None sw_mpi_other = None sw_preprocessors = No test_sponsor = Supermicro license_num = 6569 tester = Supermicro hw_avail = Nov-2022 sw_avail = Nov-2022 prepared_by = Supermicro system_vendor = Supermicro node_compute_hw_adapter_MISSING_slot_type = PCIeGen4 x16 node_compute_hw_adapter_MISSING_ports_used = 1 node_compute_hw_adapter_MISSING_model = MCX556A-EDAT node_compute_hw_adapter_MISSING_interconnect000= Mellanox Technologies MT28800 Family [ConnectX-5 node_compute_hw_adapter_MISSING_interconnect001 = Ex] node_compute_hw_adapter_MISSING_firmware = 5.7-1.0.2.0 node_compute_hw_adapter_MISSING_driver = Mellanox node_compute_hw_adapter_MISSING_data_rate = 100GbE node_compute_hw_adapter_MISSING_count = 1 node_compute_sw_state = Multi-user, run level 3 node_compute_sw_sharedfile = None node_compute_sw_other = None node_compute_sw_localfile = ext4 node_compute_purpose = compute node_compute_order = 1 node_compute_label = A+ Server 1115CS-TNR node_compute_hw_vendor = Supermicro node_compute_hw_pcache = 32 KB I + 32 KB D on chip per core node_compute_hw_scache = 1 MB I+D on chip per core node_compute_hw_tcache000 = 384 MB I+D on chip per chip node_compute_hw_tcache001 = 32 MB shared / 8 cores node_compute_hw_ocache = None node_compute_hw_other = None node_compute_hw_nthreadspercore = 2 node_compute_hw_ncpuorder = 1 chip node_compute_hw_ncoresperchip = 96 node_compute_hw_ncores = 96 node_compute_hw_nchips = 1 node_compute_hw_model = A+ Server 1115CS-TNR node_compute_hw_memory = 768 GB (12 x 64 GB 2Rx4 PC5-4800B-R) node_compute_hw_disk = 1 x 960 GB NVMe PCIe Gen4.0 node_compute_hw_cpu_name = AMD EPYC 9654 node_compute_hw_cpu_mhz = 2400 node_compute_hw_cpu_char = Max. Boost Clock upto 3.7GHz node_compute_count = 1 node_compute_sw_os000 = SUSE Linux Enterprise Server 15 SP4 node_compute_sw_os001 = Kernel 5.14.21-150400.24.21-default sw_c_compiler000 = AMD Optimizing C/C++ and Fortran Compilers (AOCC) sw_c_compiler001 = Version 4.0.0 Build 389 for Linux sw_cxx_compiler000 = AMD Optimizing C/C++ and Fortran Compilers (AOCC) sw_cxx_compiler001 = Version 4.0.0 Build 389 for Linux sw_f_compiler000 = AMD Optimizing C/C++ and Fortran Compilers (AOCC) sw_f_compiler001 = Version 4.0.0 Build 389 for Linux sw_mpi_library000 = Open MPI Library for Linux sw_mpi_library001 = Version 4.1.1 system_class = Homogeneous system_name = A+ Server 1115CS-TNR (AMD EPYC 9654) notes_055 = MPI startup command: notes_060 = mpiexec command was used to start MPI jobs. notes_065 = RAM configuration: notes_070 = Compute nodes have 1 x 64 GB RDIMM on each memory channel. notes_075 = BIOS settings: notes_080 = NUMA nodes per socket = NPS4 notes_085 = L3 Cache as NUMA Domain = Enabled notes_090 = Determinism Control = Manual notes_095 = Determinism Slider = Power notes_100 = xGMI Link Configuration = 4 xGMI Links notes_105 = 4 Link xGMI max speed = 32Gbps notes_110 = TDP Control = Manual notes_115 = TDP = 400 notes_120 = PPT Control = Manual notes_125 = PPT = 400 notes_130 = Yes: The test sponsor attests, as of date of publication, notes_135 = that CVE-2017-5754 (Meltdown) is mitigated in the system as tested and documented. notes_140 = Yes: The test sponsor attests, as of date of publication, notes_145 = that CVE-2017-5753 (Spectre variant 1) is mitigated in the system as tested and documented. notes_150 = Yes: The test sponsor attests, as of date of publication, notes_155 = that CVE-2017-5715 (Spectre variant 2) is mitigated in the system as tested and documented. # The following section was added automatically, and contains settings that # did not appear in the original configuration file, but were added to the # raw file after the run. default: notes_000 =Environment variables set by runspec before the start of the run: notes_005 =GOMP_CPU_AFFINITY = "0-128" notes_010 =KMP_BLOCKTIME = "200" notes_015 =KMP_LIBRARY = "turnaround" notes_020 =OMP_DYNAMIC = "false" notes_025 =OMP_NESTED = "FALSE" notes_030 =OMP_PLACES = "threads" notes_035 =OMP_SCHEDULE = "static" notes_040 =OMP_STACKSIZE = "128M" notes_045 =OMP_THREAD_LIMIT = "128" notes_050 =