# Invocation command line:
# /home/hpc2021/bin/harness/runhpc --config=hpc.cfg --tune=base --reportable --pmodel=MPI --define model=mpi -define ppn=128 --rank=128 --threads=1 --size=ref --iterations=3 small
# output_root was not used for this run
############################################################################
# Invocation command line:
# output_root was not used for this run
############################################################################

# The following setting was inserted automatically as a result of

build_in_build_dir=0        # build in run dir

strict_rundir_verify = 1
%ifndef %{label}         # IF acctype is not set use mpi
%   define label Cisco_UCS
%endif

%ifndef %{model}         # IF acctype is not set use mpi
%   define model mpi
%endif

teeout = yes
makeflags=-j 40


######################################################################
# The header section of the config file.  Must appear
# before any instances of "section markers" (see below)
#
# ext = how the binaries you generated will be identified
# tune = specify "base" or "peak" or "all"

label         = %{label}_%{model}
tune          = all
output_format = text
use_submit_for_speed = 1

default:
AR           = ar
ARFLAGS      = cr
CXX          = mpiicpc -cxx=icpx
CC           = mpiicc -cc=icx
FC           = mpiifort -fc=ifx
system_class = Homogenous

# Compiler Version Flags
CC_VERSION_OPTION  = --version
CXX_VERSION_OPTION = --version
FC_VERSION_OPTION  = --version

default=base=default:
OPTIMIZE      = -Ofast -ipo -mprefer-vector-width=512 -march=common-avx512
COPTIMIZE     = -ansi-alias
CXXOPTIMIZE   = -ansi-alias
FOPTIMIZE     = -nostandard-realloc-lhs -align array64byte
PORTABILITY   = -lstdc++
mpicmd = mpiexec.hydra -bootstrap ssh -genv OMP_NUM_THREADS $threads -np $ranks -ppn $ranks $command
submit = $mpicmd

#default=base=default:
#    ranks   = %{RANKS}
#    threads = %{THREADS}

%if %{model} eq 'mpi'
  pmodel=MPI
%endif

# OpenACC flags
%if %{model} eq 'acc'
  pmodel=ACC
  OPTIMIZE += -fopenacc -foffload=-lm
%endif

# OpenMP (CPU) flags
%if %{model} eq 'omp'
  pmodel=OMP
  OPTIMIZE += -fiopenmp
%endif

# OpenMP Targeting host flags
%if %{model} eq 'tgt'
  pmodel=TGT
  OPTIMIZE += -fopenmp
%endif

# OpenMP Targeting Nvidia GPU flags
%if %{model} eq 'tgtnv'
  pmodel=TGT
  OPTIMIZE += -fopenmp -fopenmp-targets=nvptx64-nvidia-cuda
%endif

# No peak flags set, so make peak use the same flags as base
default=peak=default:
basepeak=1

513.soma_t=base,peak:
    PORTABILITY+=-DSPEC_NO_VAR_ARRAY_REDUCE

528.pot3d_t,628.pot3d_s,728.pot3d_m,828.pot3d_l=base,peak:
#	PORTABILITY+=-DSPEC_NO_REORDER
OPTIMIZE+=-Wno-incompatible-function-pointer-types





# 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:



# 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:
flagsurl000 = http://www.spec.org/hpc2021/flags/Intel_compiler_flags_hpc.2024.xml
sw_other = None
sw_mpi_library = Intel MPI Library for Linux OS, Build 20231005
sw_mpi_other = None
test_sponsor = Cisco Systems
license_num = 9019
showtimer = 0
tester = Cisco Systems
hw_avail = Jun-2024
sw_avail = Feb-2024
prepared_by = Cisco Systems
system_vendor = Cisco Systems
system_name = Cisco UCS C245 M8 (AMD EPYC 9754)
hw_vendor_list = Cisco Systems
hw_total_accel = 0
hw_model_list = Cisco UCS C245 M8
hw_cpu_name_list = AMD EPYC 9754
hw_accel_vendor_list = None
hw_accel_model_list = None
node_compute_syslbl = Cisco UCS C245 M8
node_compute_sw_state = Multi-user, run level 3
node_compute_sw_sharedfile = None
node_compute_sw_other = None
node_compute_sw_localfile = xfs
node_compute_purpose = compute
node_compute_order = 1
node_compute_hw_vendor = Cisco Systems
node_compute_hw_scache = 1 MB I+D on chip per core
node_compute_hw_pcache = 32 KB I + 32 KB D on chip per core
node_compute_hw_other = None
node_compute_hw_ocache = None
node_compute_hw_nthreadspercore = 2
node_compute_hw_ncpuorder = 1,2 chips
node_compute_hw_ncoresperchip = 128
node_compute_hw_ncores = 128
node_compute_hw_nchips = 1
node_compute_hw_model = Cisco UCS C245 M8
node_compute_hw_memory000= 768 GB (12 x 64 GB 2Rx4 PC5-5600B-R,
node_compute_hw_memory001 = running at 4800 MHz)
node_compute_hw_disk = 1 x 960 GB NVMe SSD
node_compute_hw_cpu_name = AMD EPYC 9754
node_compute_hw_cpu_mhz = 2250
node_compute_hw_cpu_char = Max. Boost Clock upto 3.1GHz
node_compute_hw_adapter_fs_slot_type = None
node_compute_hw_adapter_fs_ports_used = 0
node_compute_hw_adapter_fs_model = None
node_compute_hw_adapter_fs_interconnect = None
node_compute_hw_adapter_fs_firmware = None
node_compute_hw_adapter_fs_driver = None
node_compute_hw_adapter_fs_data_rate = None
node_compute_hw_adapter_fs_count = 0
node_compute_hw_accel_vendor = None
node_compute_hw_accel_type = None
node_compute_hw_accel_model = None
node_compute_hw_accel_ecc = None
node_compute_hw_accel_desc = None
node_compute_hw_accel_count = 0
node_compute_hw_accel_connect = None
node_compute_count = 1
node_compute_hw_tcache000 = 256 MB I+D on chip per chip
node_compute_hw_tcache001 = 16 MB shared / 8 cores
node_compute_sw_os000 = SUSE Linux Enterprise Server 15 SP5
node_compute_sw_os001 = Kernel 5.14.21-150500.53-default
sw_compiler000 = Intel oneAPI DPC++/C++ Compiler 2024.0.2
notes_000 = MPI startup command:
notes_005 =   mpirun command was used to start MPI jobs.
notes_submit_000 =mpirun --bind-to core:overload-allowed --oversubscribe --mca topo basic -np $ranks $command
sw_os_list000 = SUSE Linux Enterprise Server 15 SP5Kernel
sw_os_list001 = 5.14.21-150500.53-default