Re: [OMPI users] openmpi-1.10.2rc1: wrong ELF class on Solaris

Thomas, Siegmar and all, there is a bug in configure, it is fixed in both master and v1.10 git repositories (v2.x coming soon) FCFLAGS are incorrectly reset during configure, hence the various and random errors you got. This is a one line fix, but you need recent autotools and re-run autogen

Re: [OMPI users] openmpi-1.10.2rc1: wrong ELF class on Solaris

Hello, On Dec 21, 2015, at 12:46 , Siegmar Gross wrote: LDFLAGS="-m64" CC="gcc" CXX="g++" FC="gfortran" \ CFLAGS="-m64" CXXFLAGS="-m64" FCFLAGS="-m64" \ CPP="cpp" CXXCPP="cpp" \ I might be wrong but are you perhaps missing FFLAGS and F77 settings here? Regards, Thomas -- Thomas Jahns DK

Re: [OMPI users] Help with Binding in 1.8.8: Use only second socket

I tried the following with OpenMPI 1.8.1 and 1.10.1. The both worked. In my case a node has 2 sockets like yours, but each socket has 12 cores and lstopo showed core numbers for the second socket are from 12 to 23. * mpirun --report-bindings --bind-to core --cpu-set 12,13,14,15,16,17,18,19 -np 8 j

Re: [OMPI users] Help with Binding in 1.8.8: Use only second socket

Ralph, Huh. That isn't in the Open MPI 1.8.8 mpirun man page. It is in Open MPI 1.10, so I'm guessing someone noticed it wasn't there. Explains why I didn't try it out. I'm assuming this option is respected on all nodes? Note: a SmarterManThanI™ here at Goddard thought up this: #!/bin/bash rank=

Re: [OMPI users] Help with Binding in 1.8.8: Use only second socket

Try adding —cpu-set a,b,c,… where the a,b,c… are the core id’s of your second socket. I’m working on a cleaner option as this has come up before. > On Dec 21, 2015, at 5:29 AM, Matt Thompson > wrote: > > Dear Open MPI Gurus, > > I'm currently trying to do something

[OMPI users] openmpi-dev-3241-gc895eb7: wrong ELF class on Solaris

Hi, yesterday I tried to build openmpi-dev-3241-gc895eb7 on my machines (Solaris 10 Sparc, Solaris 10 x86_64, and openSUSE Linux 12.1 x86_64) with gcc-5.1.0 and Sun C 5.13. I was successful on my Linux machine, but I got the following error on both Solaris platforms for both compilers for differe

[OMPI users] openmpi-2.0.0rc1: wrong ELF class on Solaris

Hi, yesterday I tried to build openmpi-2.0.0rc1 on my machines (Solaris 10 Sparc, Solaris 10 x86_64, and openSUSE Linux 12.1 x86_64) with gcc-5.1.0 and Sun C 5.13. I was successful on my Linux machine, but I got the following error on both Solaris platforms for both compilers for different files.

[OMPI users] openmpi-1.10.2rc1: wrong ELF class on Solaris

Hi, yesterday I tried to build openmpi-1.10.2rc1 on my machines (Solaris 10 Sparc, Solaris 10 x86_64, and openSUSE Linux 12.1 x86_64) with gcc-5.1.0 and Sun C 5.13. I was successful on my Linux machine, but I got the following error on both Solaris platforms for both compilers for different files

Re: [OMPI users] Odd behavior with subarray datatypes in OpenMPI 1.10.1

Dan, the fix is currently being discussed at https://github.com/open-mpi/ompi/pull/1196 and yes, the plan is to have the issue fixed in 1.10.2 Cheers, Gilles On 12/19/2015 1:42 AM, GARMANN, DANIEL J DR-02 USAF AFMC AFRL/RQVA wrote: Gilles, Thanks for the confirmation and patch. Do you exp

[OMPI users] Help with Binding in 1.8.8: Use only second socket

Dear Open MPI Gurus, I'm currently trying to do something with Open MPI 1.8.8 that I'm pretty sure is possible, but I'm just not smart enough to figure out. Namely, I'm seeing some odd GPU timings and I think it's because I was dumb and assumed the GPU was on the PCI bus next to Socket #0 as some