RE: [easybuild] OpenMPI 1.10.3 - mpifort wrong search path for libgfortran.so

2017-11-20 Thread Vanzo, Davide
Kenneth, Yes, the correct path is in both LD_LIBRARY_PATH and in LIBRARY_PATH. Attached you can find the outputs of $ LD_DEBUG=all mpifort multitask_mpi.f90 on two different servers that have the same minimal OS but one works and the other fails. The only difference is the CPU architecture

RE: [easybuild] OpenMPI 1.10.3 - mpifort wrong search path for libgfortran.so

2017-11-20 Thread Vanzo, Davide
John, Thank you for your feedback but your point is exactly what I meant when I said that the /usr/lib64/libgfortran.so does not exist for obvious reasons. I am working on a minimal CentOS 7.2 installation (hence no gfortran installed with the OS) and that is why I cannot understand why it

Re: [easybuild] OpenMPI 1.10.3 - mpifort wrong search path for libgfortran.so

2017-11-20 Thread Kenneth Hoste
Hi Davide, On 20/11/2017 16:06, Vanzo, Davide wrote: Hello EasyBuilders! I have started building the foss-2016b toolchain on a new server and I am stumbling against an error that I have never seen before and for which I cannot pinpoint the cause. When I try to compile a fortran source with

[easybuild] EasyBuild binutils-2.28.eb easyconfig file failed to build with error !

2017-11-20 Thread Sami, Umit D.
Hi All, I am trying to build binutils-2.28 which is a dependency for gcc 7.1 easyconfig. If I run “eb binutils-2.28.eb –r” the build process fails and generates the following errors. I wonder if this has something to do with hardware platform or CC flags. Is this script targeting all

Re: [easybuild] Redhead/CentOS 7.4 and Intel compiler issues

2017-11-20 Thread Strube, Alexandre
Correct. You can test if your system is affected by running the code attached to the issue on Redhat’s ticket system: https://bugzilla.redhat.com/attachment.cgi?id=1334984 dr. Alexandre Strube a.str...@fz-juelich.de Jülich Supercomputing Centre Institute for Advanced Simulation

Re: [easybuild] Redhead/CentOS 7.4 and Intel compiler issues

2017-11-20 Thread Åke Sandgren
The bug affects (as far as i understand it) code built by intel compilers from before 17.0 to 18.0 running on systems with the new glibc. I.e., regardless of where they were built. libc is almost always dynamically loaded. On 11/20/2017 04:14 PM, John-Paul Robinson wrote: > Could someone clarify

Re: [easybuild] Redhead/CentOS 7.4 and Intel compiler issues

2017-11-20 Thread John-Paul Robinson
Could someone clarify if this bug also affects software built with Intel on CentOS/RedHat 7.3 and then run on the 7.4 OS versions? Or is this only affecting new builds on 7.4 with the Intel compilers. In other words, does it break software built with Intel on prior OS releases an run on the 7.4

[easybuild] OpenMPI 1.10.3 - mpifort wrong search path for libgfortran.so

2017-11-20 Thread Vanzo, Davide
Hello EasyBuilders! I have started building the foss-2016b toolchain on a new server and I am stumbling against an error that I have never seen before and for which I cannot pinpoint the cause. When I try to compile a fortran source with mpifort (or mpif90), I get the error you see below. The

Re: [easybuild] Redhead/CentOS 7.4 and Intel compiler issues

2017-11-20 Thread Holanda Rusu Victor
Dear Åke, Thanks. Best wishes, Victor H. Rusu, PhD CSCS Swiss National Supercomputing Centre Via Trevano 131 6900 Lugano Switzerland On 20.11.17, 10:54, "Åke Sandgren" wrote: And it is only the 2018.0.128 version that has shown this specific problem. The

Re: [easybuild] Redhead/CentOS 7.4 and Intel compiler issues

2017-11-20 Thread Åke Sandgren
No it probably isn't. We're running on Ubuntu Xenial with glibc 2.23-0ubuntu9. And it is only the 2018.0.128 version that has shown this specific problem. The older versions that I've tried doesn't show this behaviour on FFTW, nor does 2018.1.163 On 11/20/2017 10:42 AM, Holanda Rusu Victor

Re: [easybuild] Redhead/CentOS 7.4 and Intel compiler issues

2017-11-20 Thread Holanda Rusu Victor
Dear All, I am wondering if this error is associated with the bug reported by intel with glibc. https://software.intel.com/en-us/articles/intel-compiler-not-compatible-with-glibc-224-9-and-newer Code compiled with newer versions of the Intel compilers may result in unexpected numerical values

Re: [easybuild] Redhead/CentOS 7.4 and Intel compiler issues

2017-11-20 Thread Åke Sandgren
On 11/20/2017 09:15 AM, Kenneth Hoste wrote: > Hi Åke, > > Can we somehow integrate these FFTW tests in the sanity check that > EasyBuild runs for FFTW? Those tests are already part of the FFTW build sanity check. That's why it fails to build. > > regards, > > Kenneth > > On 20/11/2017

Re: [easybuild] Redhead/CentOS 7.4 and Intel compiler issues

2017-11-20 Thread Kenneth Hoste
Hi Åke, Can we somehow integrate these FFTW tests in the sanity check that EasyBuild runs for FFTW? regards, Kenneth On 20/11/2017 07:56, Åke Sandgren wrote: 2018.0.128 miscompiles FFTW on KNL (i.e. AVX512) and i suspect it will miscompile FFTW on skylake too. Since that is a fairly well