CP2K in IBM PWR5 machinces

[Axel]

On Mar 17, 6:13 pm, "LIANG Yunfeng" <liangy... at gmail.com> wrote:
> Dear Axel,
>
> Thank you! First could you help me to check whether there is some flag
> problem with the make.inc for lapack-3.1.1? I doesn't go to the end.
> it gave error message when I was compiling lapack:
>
> ....
> LIBRARY: Shared object /usr/lpp/essl.rte.rs1/lib/libessl.a[essl_64.o]: 1632
> symbols imported.
> LIBRARY: Shared object libxlf90.a[io_64.o]: 83 symbols imported.
> LIBRARY: Shared object libxlf90.a[math_64.o]: 328 symbols imported.
> LIBRARY: Shared object libxlf90.a[xlfsys_64.o]: 33 symbols imported.
> LIBRARY: Symbols imported from import file
> /usr/lpp/xlf/lib/libxlf90.a[_xlf_aio.imp]: 10
> LIBRARY: Shared object libxlf90.a[_xlf_posix_64.o]: 88 symbols imported.
> LIBRARY: Shared object libxlomp_ser.a[omp_ser_64.o]: 61 symbols imported.
> LIBRARY: Shared object libc.a[shr_64.o]: 2659 symbols imported.
> LIBRARY: Shared object libc.a[posix_aio_64.o]: 20 symbols imported.
> LIBRARY: Shared object libc.a[aio_64.o]: 14 symbols imported.
> LIBRARY: Shared object libc.a[pse_64.o]: 5 symbols imported.
> LIBRARY: Shared object libc.a[dl_64.o]: 4 symbols imported.
> LIBRARY: Shared object libc.a[pty_64.o]: 1 symbols imported.
> FILELIST: Number of previously inserted files processed: 98
> (ld): resolve
> RESOLVE: 1979 of 15914 symbols were kept.
> (ld): addgl /usr/lib/glink64.o
> ADDGL: Glink code added for 62 symbols.
> (ld): er full
> ER: There are no unresolved symbols.
> (ld): mismatch
> MISMATCH: No type mismatches exist.
> (ld): comprld
> COMPRLD: Kept 15442 of 15442 relocation entries.
> (ld): origin page 0x100000000 0x110000000
> (ld): libpath /usr/lpp/xlf/lib:/usr/lib:/lib
> LIBPATH: Library path set to /usr/lpp/xlf/lib:/usr/lib:/lib
> (ld): save 1L .
> SAVE: Section sizes = 1473056+84312+5729848 (0x167A20+0x14958+0x576E38 hex)
> SAVE: Size of TOC: 6248 (0x1868 hex)
> (ld): rc
> RC: Highest return code was 0.
> gmake[2]: Leaving directory
> `/scratch_ibm/yfliang/library/lapack/TESTING/EIG'
> NEP: Testing Nonsymmetric Eigenvalue Problem routines
> ./xeigtstc < nep.in > cnep.out 2>&1
> gmake[1]: *** [cnep.out] Error 1
> gmake[1]: Leaving directory `/scratch_ibm/yfliang/library/lapack/TESTING'
> gmake: *** [lapack_testing] Error 2
>

this is not failing in the compile itself but in one of the testers.
you have to investigate whether this failure is legit. the testers are
_very_ strict and sometimes little rounding errors can be tolerated.

of course you can also just take the resulting library and hope for
the best.

cheers,
   axel.