You might be running out of space in /tmp or your builddir, that usually gives that error.
Set TMPDIR. If you're really unlucky it's a compiler bug. On 02/22/2017 05:30 PM, James Birch (IT Research Support) wrote: > Hi all, > > > > I am having some trouble with a build of Molpro 2012.1.0 from source > (using > https://github.com/hpcugent/easybuild-easyconfigs/blob/master/easybuild/easyconfigs/m/Molpro/Molpro-2012.1.27-intel-2015a.eb > for inspiration) with an iomkl 2017.01 toolchain that is very similar to > upstream > (https://github.com/hpcugent/easybuild-easyconfigs/blob/master/easybuild/easyconfigs/i/iomkl/iomkl-2017.01.eb). > The problem is that after about 8 minutes or so (according to ‘time’), > during the building stage, it errors and the log written to /tmp says: > > make[3]: Entering directory > `/dev/shm/build/Molpro/2012.1.0/iomkl-2017.01/Molpro/src/netlib/lapack' > > $(FC) -c $(FFLAGS) -O3 mxvf.F > > ifort: command line remark #10148: option '-vec-report0' not supported > > *ifort: error #10106: Fatal error in > /gpfs/apps/EL7-haswell/software/ifort/2017.1.132/compilers_and_libraries_2017.1.132/linux/bin/intel64/fortcom, > terminated by segmentation violation* > > compilation aborted for rmp2_f12_drv0.F (code 1) > > make[3]: *** [rmp2_f12_drv0.o] Error 1 > > make[3]: Target `lib.a' not remade because of errors. > > > > I’ve stuck quite closely to the upstream example so I’m not setting the > –vec-report0 option manually myself. I’m confused on where to go from > this and Googling didn’t give much back. > > > > I have access to the source and a token for the licence. > > > > Best regards, > > James > > > > --- > > James Birch > > Research Software Assistant > > IT Services > > University of Birmingham > > > -- Ake Sandgren, HPC2N, Umea University, S-90187 Umea, Sweden Internet: [email protected] Phone: +46 90 7866134 Fax: +46 90-580 14 Mobile: +46 70 7716134 WWW: http://www.hpc2n.umu.se
