Ok, looks good now on MacOS 10.8 (at v1.8.4-189-ga6169b4) -Paul
On Thu, Apr 9, 2015 at 10:15 AM, Paul Hargrove <[email protected]> wrote: > Ok, Ralph. > I will test the v1.8 nightly and report. > > -Paul > > On Thu, Apr 9, 2015 at 10:06 AM, Ralph Castain <[email protected]> wrote: > >> I committed a fix to the nightly 1.8.5 tarball that fixed it for me. >> However, Rolf tells me that it broke the CUDA support and we are fixing >> that now. >> >> If you don't enable CUDA, then you should be able to test. >> >> On Thu, Apr 9, 2015 at 10:03 AM, Paul Hargrove <[email protected]> >> wrote: >> >>> Ralph, >>> >>> It appears that Bert and I have iterated to a correct fix for the >>> VT+Open64 issue. >>> >>> Did you make any progress on the MacOS issue? >>> I am using the tarball, and not running autogen, so the system version >>> of libtool *shouldn't* matter. >>> However, at least on my Mac, /usr/bin/libtool is not GNU libtool and >>> /sw/bin/glibtool is version 2.4.2 (which, I understand, is the desired >>> version). >>> So, this problem looks to me to be unrelated to the sort of issue Marco >>> sees (which I have also seen when I did autogen with libtool-2.4.6 on >>> another system. >>> >>> -Paul >>> >>> On Wed, Apr 8, 2015 at 9:41 AM, Paul Hargrove <[email protected]> >>> wrote: >>> >>>> Ralph and Bert, >>>> >>>> I found a couple issues with the VT fix. >>>> I will document them in the PR when I've finished collecting the >>>> outputs.. >>>> >>>> -Paul >>>> >>>> On Wed, Apr 8, 2015 at 8:40 AM, Paul Hargrove <[email protected]> >>>> wrote: >>>> >>>>> Ralph, >>>>> >>>>> According to his comments in the PR, Bert already tested with the same >>>>> compiler that I was using when I originally observed the problem. >>>>> So, I had not put a high priority on retesting. >>>>> However, I'll see what I can do today. >>>>> >>>>> -Paul >>>>> >>>>> On Wed, Apr 8, 2015 at 8:30 AM, Ralph Castain <[email protected]> >>>>> wrote: >>>>> >>>>>> Thanks Paul!! >>>>>> >>>>>> The VT fix is in the queue, if/when you can check it. I'm going to >>>>>> look at the other issue today. >>>>>> >>>>>> >>>>>> On Apr 8, 2015, at 8:10 AM, Paul Hargrove <[email protected]> wrote: >>>>>> >>>>>> My 5 ARM and 3 MIPS testers completed without any problems. >>>>>> >>>>>> The ARM tests include ARMv5, v6, v7 and v8 systems (and include both >>>>>> ARM and THUMB2 ISAs on v7, and AARCH64 on v8) >>>>>> The MIPS tests include the "32" ABI on a 32-bit (MIPS 4Kc) system, >>>>>> and the "n32" and "64" ABIs on a 64-bit (MIPS 5Kc) system. >>>>>> >>>>>> I believe that completely covers the atomics implementations for >>>>>> those architectures. >>>>>> >>>>>> -Paul >>>>>> >>>>>> On Mon, Apr 6, 2015 at 6:38 PM, Paul Hargrove <[email protected]> >>>>>> wrote: >>>>>> >>>>>>> My testing of real hardware is almost complete and I've reported the >>>>>>> only two issues[*] I encountered: >>>>>>> http://www.open-mpi.org/community/lists/devel/2015/04/17183.php >>>>>>> http://www.open-mpi.org/community/lists/devel/2015/04/17184.php >>>>>>> There were roughly 5 failing configurations out of about 70. >>>>>>> >>>>>>> Testing on a Raspberry Pi and QEMU-emulated MIPS32, MIPS64, ARM and >>>>>>> AARCH64 systems will complete overnight. >>>>>>> >>>>>>> -Paul >>>>>>> >>>>>>> [*] Excludes known issues for which bug reports have been filed >>>>>>> against previous releases. >>>>>>> >>>>>>> >>>>>>> On Sun, Apr 5, 2015 at 2:42 PM, Ralph Castain <[email protected]> >>>>>>> wrote: >>>>>>> >>>>>>>> Usual place: >>>>>>>> >>>>>>>> http://www.open-mpi.org/software/ompi/v1.8/ >>>>>>>> >>>>>>>> I'm happy to receive suggestions about the NEWS - here is what I >>>>>>>> have: >>>>>>>> >>>>>>>> 1.8.5 >>>>>>>> ----- >>>>>>>> - Updated the internal HWLOC with several bug fixes >>>>>>>> - Fixed several bugs in OpenSHMEM support >>>>>>>> - Extended vader shared memory support to 32-bit architectures >>>>>>>> - Fixed a bug in handling subarray datatypes >>>>>>>> - Don't use inline functions with Clang compiler >>>>>>>> - Multiple bug fixes for cleanup during finalize >>>>>>>> in unusual situations >>>>>>>> - Added support for TCP keepalive signals to ensure >>>>>>>> timely termination when sockets between daemons >>>>>>>> cannot be created (e.g., due to a firewall) >>>>>>>> - Added MCA parameter to allow full use of a SLURM >>>>>>>> allocation when started from a tool (supports LLNL >>>>>>>> debugger) >>>>>>>> - Fixed several bugs in the configure logic for PMI >>>>>>>> and HWLOC >>>>>>>> - Cleaned up documentation in a few places (e.g., >>>>>>>> MPI_Comm_split_type) >>>>>>>> - Fixed prototypes for MPI_Barrier >>>>>>>> - Fixed a number of one-sided bugs >>>>>>>> - Fixed singleton MPI_Comm_spawn >>>>>>>> - Fixed a number of minor issues with CUDA support, >>>>>>>> including registering of shared memory >>>>>>>> - Improved support for Cray >>>>>>>> - Fixed performance regression introduced by >>>>>>>> inadvertent default enabling of thread support >>>>>>>> >>>>>>>> >>>>>>>> _______________________________________________ >>>>>>>> devel mailing list >>>>>>>> [email protected] >>>>>>>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel >>>>>>>> Link to this post: >>>>>>>> http://www.open-mpi.org/community/lists/devel/2015/04/17180.php >>>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> -- >>>>>>> Paul H. Hargrove [email protected] >>>>>>> Computer Languages & Systems Software (CLaSS) Group >>>>>>> Computer Science Department Tel: +1-510-495-2352 >>>>>>> Lawrence Berkeley National Laboratory Fax: +1-510-486-6900 >>>>>>> >>>>>> >>>>>> >>>>>> >>>>>> -- >>>>>> Paul H. Hargrove [email protected] >>>>>> Computer Languages & Systems Software (CLaSS) Group >>>>>> Computer Science Department Tel: +1-510-495-2352 >>>>>> Lawrence Berkeley National Laboratory Fax: +1-510-486-6900 >>>>>> _______________________________________________ >>>>>> devel mailing list >>>>>> [email protected] >>>>>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel >>>>>> Link to this post: >>>>>> http://www.open-mpi.org/community/lists/devel/2015/04/17197.php >>>>>> >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> devel mailing list >>>>>> [email protected] >>>>>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel >>>>>> Link to this post: >>>>>> http://www.open-mpi.org/community/lists/devel/2015/04/17198.php >>>>>> >>>>> >>>>> >>>>> >>>>> -- >>>>> Paul H. Hargrove [email protected] >>>>> Computer Languages & Systems Software (CLaSS) Group >>>>> Computer Science Department Tel: +1-510-495-2352 >>>>> Lawrence Berkeley National Laboratory Fax: +1-510-486-6900 >>>>> >>>> >>>> >>>> >>>> -- >>>> Paul H. Hargrove [email protected] >>>> Computer Languages & Systems Software (CLaSS) Group >>>> Computer Science Department Tel: +1-510-495-2352 >>>> Lawrence Berkeley National Laboratory Fax: +1-510-486-6900 >>>> >>> >>> >>> >>> -- >>> Paul H. Hargrove [email protected] >>> Computer Languages & Systems Software (CLaSS) Group >>> Computer Science Department Tel: +1-510-495-2352 >>> Lawrence Berkeley National Laboratory Fax: +1-510-486-6900 >>> >>> _______________________________________________ >>> devel mailing list >>> [email protected] >>> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel >>> Link to this post: >>> http://www.open-mpi.org/community/lists/devel/2015/04/17201.php >>> >> >> >> _______________________________________________ >> devel mailing list >> [email protected] >> Subscription: http://www.open-mpi.org/mailman/listinfo.cgi/devel >> Link to this post: >> http://www.open-mpi.org/community/lists/devel/2015/04/17202.php >> > > > > -- > Paul H. Hargrove [email protected] > Computer Languages & Systems Software (CLaSS) Group > Computer Science Department Tel: +1-510-495-2352 > Lawrence Berkeley National Laboratory Fax: +1-510-486-6900 > -- Paul H. Hargrove [email protected] Computer Languages & Systems Software (CLaSS) Group Computer Science Department Tel: +1-510-495-2352 Lawrence Berkeley National Laboratory Fax: +1-510-486-6900
