installation of ver 08.15.14 version

Hi All

I am trying to install new version of lammps - Aug 15 2014 with Kepler support and I am getting the following error:

mpic++ -O2 -DLAMMPS_MEMALIGN=64 -funroll-loops -fstrict-aliasing -Wall -W -Wno-uninitialized -DLAMMPS_GZIP -DLAMMPS_JPEG -I…/…/lib/qmmm -DLMP_USER_OMP -DLMP_USER_INTEL -I…/…/lib/cuda -DLMP_USER_CUDA -I…/…/lib/colvars -I…/…/lib/awpmd/ivutils/include -I…/…/lib/awpmd/systems/interact -I…/…/lib/atc -I…/…/src -I…/…/lib/reax -I…/…/lib/poems -DLMP_MPIIO -I…/…/lib/meam -I…/…/src -I…/…/src -I…/…/src -I…/…/src -I/share/apps/voro++/local/include/voro++ -I/usr/local/cuda/include -DUNIX -DNO_PREC_TIMING -DFFT_CUFFT -DCUDA_PRECISION=1 -DX_PRECISION=2 -DV_PRECISION=2 -DCUDA_ARCH=20 -c …/fix_intel.cpp
…/fix_intel.h(162): error: expected a “)”
inline void add_results(const ft * restrict const f_in,
^

…/fix_intel.h(165): warning #12: parsing restarts here after previous syntax error
const int offload);
^

…/fix_intel.h(168): error: expected a “)”
inline void add_oresults(const ft * restrict const f_in,
^

…/fix_intel.h(171): warning #12: parsing restarts here after previous syntax error
const int out_offset, const int nall);
^

…/fix_intel.h(230): error: no instance of function template “LAMMPS_NS::FixIntel::add_results” matches the argument list
argument types are: (LAMMPS_NS::IntelBuffers<double, double>::vec3_acc_t *, double *, const int, const int, int)
add_results(f_in, ev_in, eatom, vatom, 0);
^

…/fix_intel.h(253): error: no instance of function template “LAMMPS_NS::FixIntel::add_results” matches the argument list
argument types are: (LAMMPS_NS::IntelBuffers<float, double>::vec3_acc_t *, double *, const int, const int, int)
add_results(f_in, ev_in, eatom, vatom, 0);
^

…/fix_intel.h(276): error: no instance of function template “LAMMPS_NS::FixIntel::add_results” matches the argument list
argument types are: (LAMMPS_NS::IntelBuffers<float, float>::vec3_acc_t *, float *, const int, const int, int)
add_results(f_in, ev_in, eatom, vatom, 0);
^

…/fix_intel.h(287): error: expected a “)”
void FixIntel::add_results(const ft * restrict const f_in,
^

…/fix_intel.h(290): warning #12: parsing restarts here after previous syntax error
const int offload) {
^

…/fix_intel.h(336): error: expected a “)”
void FixIntel::add_oresults(const ft * restrict const f_in,
^

…/fix_intel.h(339): warning #12: parsing restarts here after previous syntax error
const int out_offset, const int nall) {
^

…/fix_intel.cpp(42): warning #1224: #warning directive: “Not building Intel package with Xeon Phi offload support.”
#warning “Not building Intel package with Xeon Phi offload support.”
^

compilation aborted for …/fix_intel.cpp (code 2)
make[1]: *** [fix_intel.o] Error 2
make[1]: Leaving directory `/share/apps/lammps/08.15.14/build/src/Obj_openmpi’
make: *** [openmpi] Error 2

Does someone already report the bug and/or solution?

Alex

Hi All

I am trying to install new version of lammps - Aug 15 2014 with Kepler
support and I am getting the following error:

[...]

compilation aborted for ../fix_intel.cpp (code 2)
make[1]: *** [fix_intel.o] Error 2
make[1]: Leaving directory
`/share/apps/lammps/08.15.14/build/src/Obj_openmpi'
make: *** [openmpi] Error 2

Does someone already report the bug and/or solution?

yes. the solution (for now) is to not install the USER-INTEL package
unless you use the intel compiler (and you may even need a specific
version). work on making this code work with the GNU compilers is
underway. if you feel lucky, you can try the LAMMPS-ICMS branch
http://git.lammps.org/git/?p=lammps-icms.git;a=summary
that one compiles fine with GNU, but is broken with several intel
compiler versions.

axel.

Many thanks Axel

I will try LAMMPS-ICMS branch.

Alex