Re: [OMPI devel] Cuda build break

Fix is here: https://github.com/open-mpi/ompi/pull/4301 > On Oct 4, 2017, at 11:19 AM, Jeff Squyres (jsquyres) > wrote: > > Thanks Ralph. > >> On Oct 4, 2017, at 2:07 PM, r...@open-mpi.org wrote: >> >> I’ll fix >> >>> On Oct 4, 2017, at 10:57 AM,

Re: [OMPI devel] HWLOC / rmaps ppr build failure

Thanks! Fix is here: https://github.com/open-mpi/ompi/pull/4301 > On Oct 4, 2017, at 11:10 AM, Brice Goglin wrote: > > Looks like you're using a hwloc < 1.11. If you want to support this old > API while using the 1.11 names, you can add this to OMPI

Re: [OMPI devel] Cuda build break

Thanks Ralph. > On Oct 4, 2017, at 2:07 PM, r...@open-mpi.org wrote: > > I’ll fix > >> On Oct 4, 2017, at 10:57 AM, Sylvain Jeaugey wrote: >> >> See my last comment on #4257 : >> >> https://github.com/open-mpi/ompi/pull/4257#issuecomment-332900393 >> >> We should completely disable CUDA in h

Re: [OMPI devel] HWLOC / rmaps ppr build failure

Looks like you're using a hwloc < 1.11. If you want to support this old API while using the 1.11 names, you can add this to OMPI after #include #if HWLOC_API_VERSION < 0x00010b00 #define HWLOC_OBJ_NUMANODE HWLOC_OBJ_NODE #define HWLOC_OBJ_PACKAGE HWLOC_OBJ_SOCKET #endif Brice Le 04/10/2017 19

Re: [OMPI devel] Cuda build break

I’ll fix > On Oct 4, 2017, at 10:57 AM, Sylvain Jeaugey wrote: > > See my last comment on #4257 : > > https://github.com/open-mpi/ompi/pull/4257#issuecomment-332900393 > > We should completely disable CUDA in hwloc. It is breaking the build, but > more importantly, it creates an extra depende

Re: [OMPI devel] HWLOC / rmaps ppr build failure

Hmmm...I suspect this is a hwloc v2 vs v1 issue. I’ll fix it > On Oct 4, 2017, at 10:54 AM, Barrett, Brian via devel > wrote: > > It looks like a change in either HWLOC or the rmaps ppr component is causing > Cisco build failures on master for the last couple of days: > > https://mtt.open-mp

Re: [OMPI devel] Cuda build break

See my last comment on #4257 : https://github.com/open-mpi/ompi/pull/4257#issuecomment-332900393 We should completely disable CUDA in hwloc. It is breaking the build, but more importantly, it creates an extra dependency on the CUDA runtime that Open MPI doesn't have, even when compiled with --

[OMPI devel] HWLOC / rmaps ppr build failure

It looks like a change in either HWLOC or the rmaps ppr component is causing Cisco build failures on master for the last couple of days: https://mtt.open-mpi.org/index.php?do_redir=2486 rmaps_ppr.c:665:17: error: ‘HWLOC_OBJ_NUMANODE’ undeclared (first use in this function); did you mean ‘HWLO

[OMPI devel] Cuda build break

All - It looks like nVidia’s MTT started failing on 9/26, due to not finding Cuda. There’s a suspicious commit given the error message in the hwloc cuda changes. Jeff and Brice, it’s your patch, can you dig into the build failures? Brian ___ devel m