lower the current test precision for periodic_systems/17-aluminium.02-td
Description
The tolerance for the periodic_systems/17-aluminium.02-td test seems to be too tight.
Failed Tests
Match Current [step 75] :
Calculated value : 5.215900449153e+00
Reference value : 5.2159004382020004
Difference : 1.09509992185508e-08
Deviation [%] : 2.09954145948495e-07
Tolerance : 1e-08
Tolerance [%] : 1.91721450945623e-07
Current [step 75] : [ FAIL ]
Match Current [step 125] :
Calculated value : -5.882323533101e+00
Reference value : -5.882323521279
Difference : 1.18220002676139e-08
Deviation [%] : 2.00975009702347e-07
Tolerance : 1e-08
Tolerance [%] : 1.70000850239289e-07
Current [step 125] : [ FAIL ]
Test environment
- CPU : Intel Xeon Gold 6148
- Libraries : intel/19.0.4, mkl/2019.4, fftw-mpi/3.3.8, netcdf-serial/4.4.1, impi/2019.4, gsl/2.4, hdf5-serial/1.8.21, libxc/4.3.4
- Configuration options : max-dim=3 openmp mpi sse2 avx libxc4
- Optional libraries : metis mpi2 netcdf scalapack
- Architecture : x86_64
- C compiler : mpiicc -std=gnu99 (icc)
- C compiler flags : -g -traceback -O3 -xCORE-AVX512 -qopt-zmm-usage=high -fma -ip -fzero-initialized-in-bss
- C++ compiler : g++ -std=c++11
- C++ compiler flags : -g -O2
- Fortran compiler : mpiifort (ifort) (Intel(R) Fortran Intel(R) 64 Compiler for applications running on Intel(R) 64, Version 19.0.4.243 Build 20190416)
- Fortran compiler flags : -g -traceback -O3 -xCORE-AVX512 -qopt-zmm-usage=high -fma -ip -fzero-initialized-in-bss -init=arrays -init=zero –fopenmp
Checklist
- [N/A] I have checked that my code follows the Octopus coding standards
- [N/A] I have added tests for all the new features added in this request.