Re: [OMPI devel] [1.8.4rc5] preliminary results

Update: I now have 59 of 61 results, with only the QEMU-emulated MIPS platforms outstanding. Those will not likely finish until near to (or after) midnight tonight. Unless something turns up on the MIPS systems my "smoke test" of rc5 is complete. The only issues I found are the vader and fortran

Re: [OMPI devel] [1.8.4rc5] preliminary results

On Thu, Dec 18, 2014 at 8:55 AM, Paul Hargrove wrote: > > I also have unexplained errors on my Solaris-10/SPARC system. > It looks like there may have been a loss of network connectivity during > the tests. > I need to check these deeper, but I expect them to pass when I get a > chance to re-run t

Re: [OMPI devel] [1.8.4rc5] preliminary results

Paul -- Can you send me the config.log and output from configure for these 2 builds? For open64, if you could send the output of "ompi_info --all", that would also be helpful. Thanks. On Dec 18, 2014, at 1:40 PM, Paul Hargrove wrote: > Jeff, > > See below for some failure details. > The lo

Re: [OMPI devel] [1.8.4rc5] preliminary results

On Dec 18, 2014, at 2:10 PM, George Bosilca wrote: > I also noticed a drastic increase in the number of linking warnings. This is > on 64 bits SciLinux Carbon (6.6) with using the Intel compilers 14.0.0 > 20130728. I run some tests and everything seems to work just fine, so this > might not be

Re: [OMPI devel] libfabric, config.h and hwloc

On Dec 18, 2014, at 3:13 AM, Gilles Gouaillardet wrote: > currently, ompi master cannot be built if configured with > --without-hwloc *and without* --without-libfabric. > > the root cause is HAVE_CONFIG_H is defined but no config.h file is found. > > i digged a bit and found that config.h is t

Re: [OMPI devel] [1.8.4rc5] preliminary results

I also noticed a drastic increase in the number of linking warnings. This is on 64 bits SciLinux Carbon (6.6) with using the Intel compilers 14.0.0 20130728. I run some tests and everything seems to work just fine, so this might not be such a deal breaker. George. libtool: install: warning: rel

Re: [OMPI devel] [1.8.4rc5] preliminary results

Jeff, See below for some failure details. The look like different symptoms of the same issue. -Paul Open64 link failure: $ mpifort -g hello_mpifh.f -o hello_mpifh /global/homes/h/hargrove/GSCRATCH/OMPI/openmpi-1.8.4rc5-linux-x86_64-open64/INST/lib/libmpi_mpifh.so: undefined reference to `_Iso_c

Re: [OMPI devel] commit be6d4649

Yeah, I'm an idiot. Typo in the commit. Man, I'm really batting 0 this week. :-( Fix coming shortly (I tested the --without-hwloc before committing; I neglected to test --with-hwloc before committing). On Dec 18, 2014, at 1:03 PM, Howard Pritchard wrote: > Hi Folks, > > commit be6d4649 br

[OMPI devel] commit be6d4649

Hi Folks, commit be6d4649 broke simple ./configure of master. I'd like to revert this commit unless someone can figure out a better solution to Gilles --without-hwloc issue soon. Howard

[OMPI devel] simple ./configure doesn't work on master/HEAD

Hi Folks, I just tried to run ./configure on an updated master and its failing with hwloc header file issues: checking components to build statically... noos xml synthetic custom xml_nolibxml linux linuxpci x86 checking components to build as plugins... checking whether hwloc configure succe

Re: [OMPI devel] ABI compatibility proposal for 1.9/2.0 release stream

+1 On Thu, Dec 18, 2014 at 05:33:36PM +, Jeff Squyres (jsquyres) wrote: > To slightly expand on what Howard said: > > We'd basically fix ABI at the beginning of a stable series (i.e., 2.0.0), and > maintain it for the entire stable series. > > The rationale here is that the adoption we've s

Re: [OMPI devel] ABI compatibility proposal for 1.9/2.0 release stream

To slightly expand on what Howard said: We'd basically fix ABI at the beginning of a stable series (i.e., 2.0.0), and maintain it for the entire stable series. The rationale here is that the adoption we've seen is that production shops are (rightfully) just moving from stable series to stable s

Re: [OMPI devel] [1.8.4rc5] preliminary results

On Dec 18, 2014, at 11:55 AM, Paul Hargrove wrote: > + NEW: I see Fortran bindings failing to compile w/ gfortran > + NEW: I see Fortran bindings fail to link with Open64 Paul -- you make me sad. Sigh. Send some details of these failures when you can. Thanks! -- Jeff Squyres jsquy...@cisco.

[OMPI devel] ABI compatibility proposal for 1.9/2.0 release stream

Hi Folks, Jeff and I have been considering changing the ABI compatibility store fory Open MPI for the 1.9/2.0 release stream. Basically no promises for the odd/feature release series, but keep the current ABI promise for the even release series. I've updated the 1.9/2.0 release page

[OMPI devel] [1.8.4rc5] preliminary results

With results from about 50 out of 61 platforms: + KNOWN: SGI UV is still "broken-by-default" (fails compiling vader unless configured with --without-xpmem) + NEW: I see Fortran bindings failing to compile w/ gfortran + NEW: I see Fortran bindings fail to link with Open64 I also have unexplained e

[OMPI devel] libfabric, config.h and hwloc

Jeff, currently, ompi master cannot be built if configured with --without-hwloc *and without* --without-libfabric. the root cause is HAVE_CONFIG_H is defined but no config.h file is found. i digged a bit and found that config.h is taken from a hwloc directory (if the --without-hwloc option is no

Re: [OMPI devel] 1.8.4rc Status

On Wed, Dec 17, 2014 at 7:17 PM, Paul Hargrove wrote: > > I am going to run the nightly on other configs on both my > Solaris-11/x86-64 and Solaris-10/SPARC systems. > I just want to be sure some other compile/abi/arch combination didn't get > broken by accident. > I will post my results to the li

Re: [OMPI devel] 1.8.4rc5 out

Tests queued on 61 distinct configurations... will share results when I've got them. -Paul On Wed, Dec 17, 2014 at 9:15 PM, Ralph Castain wrote: > > Hi folks > > Trying to bring this to closure, so hopefully this is the last one. Please > give it a smoke test: > > http://www.open-mpi.org/softwar

[OMPI devel] 1.8.4rc5 out

Hi folks Trying to bring this to closure, so hopefully this is the last one. Please give it a smoke test: http://www.open-mpi.org/software/ompi/v1.8/ Ralph