Re: [OMPI devel] 1.10.0rc2

On Jul 21, 2015, at 4:38 PM, Orion Poplawski wrote: > > I see that 1.10.0 can make use of libfabric, but I don't see much > documentation on it and certainly nothing in the README file. Oops; shame on us. We'll update README. > Since libfabric > makes use of libibverbs and librdmacm, does it

[OMPI devel] Open MPI 1.10.0rc2 and "pgcc -m32"

As Pavan Balaji recently reported on the hwloc-devel list, there is a problem with some inline asm in hwloc-1.11.0 that leads to failures with PGI compilers and x86 (32-bit) targets. I have determined that this issue *is* also present in the hwloc version embedded in Open MPI 1.10.0rc2. This is *

Re: [OMPI devel] 1.10.0rc2 failure with pgi-9.0-4 and -m32

Ralph, Thanks for clarifying. -Paul On Tue, Jul 21, 2015 at 5:29 PM, Ralph Castain wrote: > Nathan reminded me today that we started requiring C99 in the 1.8 series, > so yes - 1.10 does require C99 > > I went ahead and removed the designators before he reminded me, but I’ll > leave them out a

Re: [OMPI devel] 1.10.0rc2 failure with pgi-9.0-4 and -m32

Nathan reminded me today that we started requiring C99 in the 1.8 series, so yes - 1.10 does require C99 I went ahead and removed the designators before he reminded me, but I’ll leave them out as they aren’t required > On Jul 21, 2015, at 5:01 PM, Paul Hargrove wrote: > > Ralph, > > I know y

[OMPI devel] 1.10.0rc2 failure with pgi-9.0-4 and -m32

Ralph, I know you recently fixed a C99-related issue I saw in 1.10.0rc1 (removing designators in an initializer). So, I was surprised to see the following with pgi-9.0-4 and -m32: == Compiler and preprocessor tests

Re: [OMPI devel] 1.10.0rc2

On 07/21/2015 01:29 PM, Ralph Castain wrote: > Hey folks > > 1.10.0rc2 is now out for review - excepting the library version numbers, this > should be the final version. Please take a quick gander and let me know of any > problems. > > http://www.open-mpi.org/software/ompi/v1.10/ > > Ralph I se

[OMPI devel] 1.10.0rc2

Hey folks 1.10.0rc2 is now out for review - excepting the library version numbers, this should be the final version. Please take a quick gander and let me know of any problems. http://www.open-mpi.org/software/ompi/v1.10/ Ralph

Re: [OMPI devel] race condition in finalize

I believe I have this fixed - please see if this solves the problem: https://github.com/open-mpi/ompi/pull/730 > On Jul 21, 2015, at 12:22 AM, Gilles Gouaillardet wrote: > > Ralph, > > here is some more detailed information. > > > from orte_ess_b

Re: [OMPI devel] race condition in finalize

Ralph, here is some more detailed information. from orte_ess_base_app_finalize() first orte_rml_base_close() is invoked(via mca_base_framework_close(&orte_rml_base_framework); and it does while (NULL != (item = opal_list_remove_first(&orte_rml_base.posted_recvs))) { OBJ_RELEASE(