[CP2K:1776] Question regarding regtests

[Teodoro Laino]

Hi Joerg,

I wouldn't really worry for these cases. Reasons:

1) As you said NMR (and all linear response part) is quite floppy to 
numerical errors.. it is enough that you rearrange a bit
the instructions and you get quite different numbers (even in the range 
of 10^-1!!).
2) The graphite example is quite similar to NMR.. that's just a matter 
of having built a quite instable example..
3) The multiple-walkers metadyn regtest is another one for two different 
reasons: (1) it is based on NH3 which handled with FIST turns out to be 
a quite numerically instable example. (2) relies on information written 
on files. If the file system is not fast enough you may easily end up 
with different results because what you would expect to read is not at 
all written in the file.

As usual perfection is out of this world and let's say that if you have 
only troubles with these examples you can safely ignore them and 
consider your executable well built!
It's good you posted this question since one has to develop a certain 
feeling in order to rely on an executable.

Ciao,
Teo

Jörg Saßmannshausen wrote:
> Dear all,
>
> having just spent 1 week at excellent the cp2k tutorial in Zurich, I am 
> feeling less intimidated now in actually using the program.
> However, I have a question regarding the regtests. Whereas I am aware 
> from the list that the regtests are mere for the developers, I think 
> they also could be used to find some problems in the compilation. I am 
> aware that simply comparison between the numbers does not make any sense 
> and that the NMR tests are usually a bit more tricky than the other 
> tests. Whereas I am happy to ignore relative errors in the region of 
> E-08 to E-14 (depending on the actually calculated property), I stumbled 
> over these errors here (for example):
>
> $HOME/cp2k/rt/TEST-Linux-x86-64-intel-sopt-2009-02-16T14:37:17+0000/Fist/regtest-4/nh3-meta-walks_2.inp.out 
> :
>   POTENTIAL ENERGY : old = 0.364412001385E-02 new = 0.328588609073E-02
>   relative error :   1.09022015e-01
>
> $HOME/cp2k/rt/TEST-Linux-x86-64-intel-sopt-2009-02-16T14:37:17+0000/Fist/regtest-4/nh3-meta-walks_2r.inp.out 
> :
>   POTENTIAL ENERGY : old = 0.485712510073E-01 new = 0.533298507726E-02
>   relative error :   8.10770428e+00
>
> $HOME/cp2k/rt/TEST-Linux-x86-64-intel-sopt-2009-02-16T14:37:17+0000/QS/regtest-nonortho/graphite3.inp.out 
> :
>   POTENTIAL ENERGY : old = -0.144703327615E+02 new = -0.180353886104E+02
>   relative error :   1.97670032e-01
>
> [...]
>
> Whereas I quite happily could ignore the FIST results as I right now 
> don't think I need FIST, the QS is a bit more worrying for me.
>
> As a newbie, I would appreciate some feedback here so I don't run into 
> the problem of wasting my time with an obvious missbuild of my binary.
>
> Incidently, some of the errors appear in g95 compiled binaries as well, 
> but I only checked a few cases here.
>
> System information are found at the bottom of the email.
>
> Thanks for your help
>
> Jörg
>
>
> System Informations:
>
> Linux-x86-64-intel.sopt:
> INTEL_INC    = /users/sassy/build/intel/Compiler/10.1/021/include
> ATLAS_PATH   = /users/sassy/build/lib/atlas3.8.2-ifort/lib
> FFTW3_INCL   = /users/sassy/build/fftw3/ifc/include
> FFTW3_PATH   = /users/sassy/build/fftw3/ifc/lib
> LIBINT_WRAP  = /users/sassy/build/src/cp2k/rt/cp2k/libint_tools
> LIBINT_INCL  = /users/sassy/build/fftw3/ifc/include
> LIBINT_PATH  = /users/sassy/build/lib/libint-ifort/lib
> MKL_PATH     = /users/sassy/build/intel/mkl/10.1.0.015/lib/em64t
> MKL_INCL     = /users/sassy/build/intel/mkl/10.1.0.015/include/fftw
>
> CC       = cc
> CPP      =
> FC       = ifort
> LD       = ifort
> AR       = ar -r
> DFLAGS   = -D__INTEL -D__FFTSG -D__FFTW3 -D__LIBINT
> CPPFLAGS = -C -traditional $(DFLAGS) -I$(INTEL_INC) -I$(LIBINT_INCL)
> FCFLAGS  = $(DFLAGS) -I$(INTEL_INC) -I$(FFTW3_INCL) \
>              -O2 -march=pentium3 -pc64 -fpp -free -unroll -mtune=core2
>
> LDFLAGS  = $(FCFLAGS)
> LIBS1     = $(ATLAS_PATH)/liblapack.a \
>             $(ATLAS_PATH)/libcblas.a \
>             $(ATLAS_PATH)/libf77blas.a \
>             $(ATLAS_PATH)/libatlas.a
> LIBS      = $(LIBS1) $(LIBINT_WRAP)/libint_cpp_wrapper.o \
>             $(LIBINT_PATH)/libderiv.a \
>             $(LIBINT_PATH)/libint.a \
>             $(FFTW3_PATH)/libfftw3.a \
>             -lstdc++
>
> OBJECTS_ARCHITECTURE = machine_intel.o
>
>
> fftw-3.2
> ifort (IFORT) 10.1 20081024
> atlas3.8.2
> gcc (GCC) 4.1.0 (SUSE Linux)
>
> CPU:
> Dual-Core AMD Opteron(tm) Processor 2212
>
> OS Linux Suse version?
>
>
>
>