Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Ralph Castain
Okay, 1.8.7rc3 (we already had an rc2) is now out with all these changes - please take one last look. http://www.open-mpi.org/software/ompi/v1.8/ > On Jul 13, 2015, at 7:42 AM, Ralph Castain wrote: > > Yes, I’ll release a new rc once I get it all

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Ralph Castain
Yes, I’ll release a new rc once I get it all merged. Are the linker warnings a change in behavior from 1.8.6? I confess I’ve been seeing them in the master for so long that I’ve been “inoculated” to ignore them. > On Jul 13, 2015, at 7:34 AM, Gilles Gouaillardet > wrote: > > Yes it is. > wi

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Gilles Gouaillardet
Yes it is. will you release rc2 after the PR is merged ? it seems rhel 6.6 ibverbs devel headers are fine, so it would be ideal to confirm the patch works as expected with broken headers... what about the linker warnings with the fortran library ? should we do something before releasing 1.8.7 ? C

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Ralph Castain
Gilles - just to confirm, the patch you provided here is the one in the updated PRs, yes? If so, I’ll consider those PRs as confirmed and commit them > On Jul 13, 2015, at 7:20 AM, Gilles Gouaillardet > wrote: > > Chris, > > thanks for the report. > > recently, we start getting rid of unini

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Gilles Gouaillardet
Chris, thanks for the report. recently, we start getting rid of uninitialized common symbols aka uninitialized global variables. a side effect is the bunch of warnings issued by the linker when linking with fortran (2008 ?) MPI library. I guess I can do something for that, at least with gnu and

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Chris Samuel
On Mon, 13 Jul 2015 05:17:29 PM Gilles Gouaillardet wrote: > Hi Chris, Hi Gilles, > i pushed my tarball into a gist : Thanks for that, I can confirm on our two x86-64 RHEL 6.6 boxes (one circa 2010, one circa 2013) with their included OFED I see: checking if ConnectX XRC support is enabled...

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Gilles Gouaillardet
Hi Chris, i pushed my tarball into a gist : git clone https://gist.github.com/ec20f77ec35533fa575a.git and then the tarball is in ec20f77ec35533fa575a/openmpi-gitclone.tar.bz2 Cheers, Gilles On 7/13/2015 4:59 PM, Chris Samuel wrote: Hi Gilles, On Mon, 13 Jul 2015 03:16:57 PM Gilles Gouaill

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Chris Samuel
Hi Gilles, On Mon, 13 Jul 2015 03:16:57 PM Gilles Gouaillardet wrote: > i made ConnectX XRC (aka XRC) and ConnectIb XRC (aka XRC domains) exclusive, > so yes, you got the desired behavior. Is there a tarball I could test on our x86 systems please? We are tied to the OFED in RHEL due to having

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Gilles Gouaillardet
Paul, thanks for the report, i made ConnectX XRC (aka XRC) and ConnectIb XRC (aka XRC domains) exclusive, so yes, you got the desired behavior. Cheers, Gilles On 7/13/2015 3:11 PM, Paul Hargrove wrote: Giles, With this latest patch on my "new" system I see checking if ConnectX XRC suppo

Re: [OMPI devel] 1.8.7rc1 testing results

2015-07-13 Thread Paul Hargrove
Giles, With this latest patch on my "new" system I see checking if ConnectX XRC support is enabled... no checking if ConnectIB XRC support is enabled... yes On a system which had said "yes" for both probes with all previous patches. I *think* this is the desired behavior (use the non-depre