options

cp2k.ssmp - 2023-07-10 17:56:28 - MAQAO 2.17.5

Help is available by moving the cursor above any symbol or by checking MAQAO website.

Global Metrics

Total Time (s)924.86
Profiled Time (s)911.27
Time in analyzed loops (%)62.9
Time in analyzed innermost loops (%)36.6
Time in user code (%)66.7
Compilation Options Score (%)75.0
Perfect Flow Complexity1.04
Array Access Efficiency (%)83.1
GFLOPS7.416
Perfect OpenMP + MPI + Pthread1.01
Perfect OpenMP + MPI + Pthread + Perfect Load Distribution1.01
No Scalar IntegerPotential Speedup1.19
Nb Loops to get 80%23
FP VectorisedPotential Speedup1.19
Nb Loops to get 80%31
Fully VectorisedPotential Speedup1.80
Nb Loops to get 80%41
FP Arithmetic OnlyPotential Speedup1.57
Nb Loops to get 80%41

CQA Potential Speedups Summary

Loop Based Profile

Innermost Loop Based Profile

Application Categorization

Compilation Options

Source ObjectIssue
cp2k.ssmp
dbcsr_data_operations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
direct.c-funroll-loops is missing.
dbcsr_index_operations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
ct.c-funroll-loops is missing.
dbcsr_api.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_hash_table.f90-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
ai_contraction.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_block_operations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
grid_ref_task_list.c-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
grid_library.c-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
q1fv_5.c-funroll-loops is missing.
n1fv_3.c-funroll-loops is missing.
qs_neighbor_list_types.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
ai_os_rr.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_mm_common.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
t3fv_5.c-funroll-loops is missing.
qs_ot.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_dist_operations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
qs_kind_types.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
xc.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
qs_kinetic.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
vrank-geq1.c-funroll-loops is missing.
qs_neighbor_lists.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
grid_ref_collint.h-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
timings.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
pw_poisson_methods.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
util.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
core_ppl.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
n1bv_5.c-funroll-loops is missing.
ai_oneelectron.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
grid_ref_prepare_pab.c-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
ai_overlap.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
cp_dbcsr_operations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_block_access.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_vector.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_transformations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
pw_methods.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_mm_dist_operations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
xc_pbe.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
t3bv_5.c-funroll-loops is missing.
task_list_methods.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
routine_map.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
n1bv_3.c-funroll-loops is missing.
callgraph.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
qs_overlap.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_operations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
q1bv_5.c-funroll-loops is missing.
xc_rho_set_types.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
grid_ref_integrate.c-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_toollib.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_ptr_util.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
list_routinestat.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
xc_derivative_set_types.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
ai_kinetic.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_mm_csr.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
ai_overlap_ppl.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_array_sort.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_mm_multrec.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_iterator_operations.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
grid_process_vab.h-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
fast.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
grid_ref_collocate.c-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
qs_ks_utils.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
kahan_sum.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dftw-directsq.c-funroll-loops is missing.
core_ppnl.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dftw-direct.c-funroll-loops is missing.
dbcsr_mpiwrap.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_methods.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
n1fv_5.c-funroll-loops is missing.
cp_array_sort.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
pw_grids.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
ao_util.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
grid_sphere_cache.c-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_mm_hostdrv.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_data_methods_low.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
external_potential_types.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.
dbcsr_mm_cannon.F-march=x86-64 is used but it should be replaced by a more architecture specific option or -march=native.

Loop Path Count Profile

Cumulated Speedup If No Scalar Integer

Cumulated Speedup If FP Vectorized

Cumulated Speedup If Fully Vectorized

Cumulated Speedup If FP Arithmetic Only

Experiment Summary

Application/ccc/work/cont001/ocre/oserete/cp2k/exe/local/cp2k.ssmp
Timestamp2023-07-10 17:56:28 Universal Timestamp1689004588
Number of processes observed1 Number of threads observed1
Experiment TypeOpenMP;
Machineinti6001
Model NameAMD EPYC 7H12 64-Core Processor
Architecturex86_64 Micro ArchitectureZEN_V2
Cache Size512 KB Number of Cores64
OS VersionLinux 4.18.0-305.88.1.el8_4.x86_64 #1 SMP Thu Apr 6 10:22:46 EDT 2023
Architecture used during static analysisx86_64 Micro Architecture used during static analysisZEN_V2
Frequency Driveracpi-cpufreq Frequency Governorperformance
Huge Pagesmadvise Hyperthreadingon
Number of sockets2 Number of cores per socket64
Compilation Options
cp2k.ssmp: GNU Fortran2008 12.2.0 --param=l1-cache-size=32 --param=l1-cache-line-size=64 --param=l2-cache-size=512 -mtune=znver2 -march=x86-64 -g -O3 -std=f2008 -fno-omit-frame-pointer -fopenmp -fcf-protection=none -funroll-loops -fbacktrace -ffree-form -fimplicit-none -fintrinsic-modules-path /ccc/products2/gcc-12.2.0/Rhel_8__x86_64/system/default/bin/../lib/gcc/x86_64-pc-linux-gnu/12.2.0/finclude -fpre-include=/usr/include/finclude/math-vector-fortran.h

Configuration Summary

Dataset
Run Command<executable> H2O-gga.inp
Number Processes1
Number Nodes1
FilterNot Used
Profile StartNot Used
Maximal Path Number4
×