ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
6c5532072a
openmpi/config
История
Rainer Keller 6c5532072a - Split the datatype engine into two parts: an MPI specific part in 
OMPI
   and a language agnostic part in OPAL. The convertor is completely
   moved into OPAL.  This offers several benefits as described in RFC
   http://www.open-mpi.org/community/lists/devel/2009/07/6387.php
   namely:
    - Fewer basic types (int* and float* types, boolean and wchar
    - Fixing naming scheme to ompi-nomenclature.
    - Usability outside of the ompi-layer.
 - Due to the fixed nature of simple opal types, their information is
   completely
   known at compile time and therefore constified
 - With fewer datatypes (22), the actual sizes of bit-field types may be
   reduced
   from 64 to 32 bits, allowing reorganizing the opal_datatype
   structure, eliminating holes and keeping data required in convertor
   (upon send/recv) in one cacheline...
   This has implications to the convertor-datastructure and other parts
   of the code.
 - Several performance tests have been run, the netpipe latency does not
   change with
   this patch on Linux/x86-64 on the smoky cluster.
 - Extensive tests have been done to verify correctness (no new
   regressions) using:
   1. mpi_test_suite on linux/x86-64 using clean ompi-trunk and
    ompi-ddt:
    a. running both trunk and ompi-ddt resulted in no differences
       (except for MPI_SHORT_INT and MPI_TYPE_MIX_LB_UB do now run
       correctly).
    b. with --enable-memchecker and running under valgrind (one buglet
       when run with static found in test-suite, commited)
   2. ibm testsuite on linux/x86-64 using clean ompi-trunk and ompi-ddt:
      all passed (except for the dynamic/ tests failed!! as trunk/MTT)
   3. compilation and usage of HDF5 tests on Jaguar using PGI and
      PathScale compilers.
   4. compilation and usage on Scicortex.
 - Please note, that for the heterogeneous case, (-m32 compiled
   binaries/ompi), neither
   ompi-trunk, nor ompi-ddt branch would successfully launch.

This commit was SVN r21641.
2009-07-13 04:56:31 +00:00
..
c_get_alignment.m4 Per #1181, make our use of rm be consistent with the rest of AC/AM. 2007-11-06 12:20:58 +00:00
c_weak_symbols.m4 This is a very large change to rename several #define values from 2009-05-06 20:11:28 +00:00
cxx_find_exception_flags.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
cxx_find_template_parameters.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
cxx_find_template_repository.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
cxx_have_exceptions.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
distscript.csh It looks like ftp.gnu.org is disabling the download of repository 2009-01-20 15:53:18 +00:00
f77_check_logical_array.m4 - As discussed revert r21330, Fortran-configure info should 2009-06-01 19:02:34 +00:00
f77_check_real16_c_equiv.m4 - As discussed revert r21330, Fortran-configure info should 2009-06-01 19:02:34 +00:00
f77_check_type.m4 - AC_MSG_RESULT([$OMPI_F90_BUILD_SIZE]), only if F90-bindings enabled. 2009-03-16 02:07:33 +00:00
f77_check.m4 - As discussed revert r21330, Fortran-configure info should 2009-06-01 19:02:34 +00:00
f77_find_ext_symbol_convention.m4 - Update from yesterdays patch -- just check once and for all, getting 2009-06-05 01:00:59 +00:00
f77_get_alignment.m4 Refs trac:1603: 2008-11-07 20:37:21 +00:00
f77_get_fortran_handle_max.m4 - As discussed revert r21330, Fortran-configure info should 2009-06-01 19:02:34 +00:00
f77_get_sizeof.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
f77_get_value_true.m4 - As discussed revert r21330, Fortran-configure info should 2009-06-01 19:02:34 +00:00
f77_purge_unsupported_kind.m4 - As discussed revert r21330, Fortran-configure info should 2009-06-01 19:02:34 +00:00
f90_check_type.m4 - AC_MSG_RESULT([$OMPI_F90_BUILD_SIZE]), only if F90-bindings enabled. 2009-03-16 02:07:33 +00:00
f90_check.m4 - Split the datatype engine into two parts: an MPI specific part in 2009-07-13 04:56:31 +00:00
f90_find_module_include_flag.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
f90_get_int_kind.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
f90_get_precision.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
f90_get_range.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
f90_get_sizeof.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
libltdl_dlopen_global.diff Fix for LT 2.1 libltdl lt_dlopen behavior of putting loaded DSO's in a 2007-03-05 17:21:42 +00:00
lt21a-pathCC.diff Fixes trac:982. 2007-04-21 00:56:47 +00:00
lt224-icc.diff Per advice from Ralf W. (see bug-libtool list post 4:48pm US Eastern 2008-05-27 21:58:09 +00:00
lt1522-pathCC.diff Fixes trac:982. 2007-04-21 00:56:47 +00:00
lt-sun-fortran.diff Add a couple of Libtool tweaks for Sun Studio compatibility: 2008-11-25 15:59:48 +00:00
ltmain_otool.diff Work around a bug involving dependent libraries when building 64 bit on Mac OS X 2007-01-22 20:31:04 +00:00
Makefile.am Rollback the last two commits. 2008-10-28 17:22:29 +00:00
Makefile.options Rollback the last two commits. 2008-10-28 17:22:29 +00:00
mca_acinclude.m4 This is a very large change to rename several #define values from 2009-05-06 20:11:28 +00:00
mca_configure.ac Rollback the last two commits. 2008-10-28 17:22:29 +00:00
mca_make_configure.pl Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_case_sensitive_fs_setup.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_alps.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_bproc.m4 Make better help messages for many configure options that take 2009-04-10 22:32:00 +00:00
ompi_check_broken_qsort.m4 This is a very large change to rename several #define values from 2009-05-06 20:11:28 +00:00
ompi_check_compiler_works.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_func_lib.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_gm.m4 Make better help messages for many configure options that take 2009-04-10 22:32:00 +00:00
ompi_check_icc.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_loadleveler.m4 Make better help messages for many configure options that take 2009-04-10 22:32:00 +00:00
ompi_check_lsf.m4 Up to SVN r20383 2009-01-29 17:59:04 +00:00
ompi_check_mx.m4 Make better help messages for many configure options that take 2009-04-10 22:32:00 +00:00
ompi_check_openib.m4 - Factor out OPAL, ORTE and OMPI specific configure-parameters... 2009-05-27 03:03:18 +00:00
ompi_check_optflags.m4 Fix a few places where we didn't properly escape []; consolidate all debug/optimization flag checking to use AC quadrigraphs properly 2008-12-09 23:42:28 +00:00
ompi_check_package.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_pcie.m4 Once again, into the breach! 2009-06-02 22:26:36 +00:00
ompi_check_portals.m4 Make better help messages for many configure options that take 2009-04-10 22:32:00 +00:00
ompi_check_ps.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_psm.m4 Make better help messages for many configure options that take 2009-04-10 22:32:00 +00:00
ompi_check_sge.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_slurm.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_tm.m4 Make better help messages for many configure options that take 2009-04-10 22:32:00 +00:00
ompi_check_udapl.m4 Make better help messages for many configure options that take 2009-04-10 22:32:00 +00:00
ompi_check_vendor.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_check_withdir.m4 Really fix the quoting this time. Really. 2009-02-04 23:04:21 +00:00
ompi_check_xgrid.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_config_pthreads.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_config_solaris_threads.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_config_subdir_args.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_config_subdir.m4 Disable argument checking in sub-configure scripts to avoid annoying messages about how sub-configures don't support all the same options as the top-level configure script 2009-05-22 21:45:23 +00:00
ompi_configure_options.m4 - Revert OMPI_CSUM_DST, not yet there 2009-05-27 03:24:10 +00:00
ompi_contrib.m4 * Fix a truckload of Cisco copyrights to be the same as the rest of 2009-01-11 02:30:00 +00:00
ompi_ext.m4 Fix one problem with autogen.sh -no-ompi: the ext stuff will now be skipped properly. However, there are now other un-defined AM_CONDITIONALs... Ugh! 2009-06-25 22:37:03 +00:00
ompi_find_type.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_get_libtool_linker_flags.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_get_version.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_get_version.m4sh Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_interix.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_lang_link_with_c.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_load_platform.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_make_stripped_flags.m4 Fix a few places where we didn't properly escape []; consolidate all debug/optimization flag checking to use AC quadrigraphs properly 2008-12-09 23:42:28 +00:00
ompi_mca.m4 Fix a help message. 2009-01-03 01:03:28 +00:00
ompi_microsoft.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_save_version.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_setup_component_package.m4 Make help strings for configure =DIR types of options better/more accurate 2009-04-10 22:10:07 +00:00
ompi_setup_cxx.m4 Fix a problem on some distros that have buggy versions of gcc (e.g., 2008-10-29 12:17:44 +00:00
ompi_setup_f77.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_setup_f90.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_setup_libevent.m4 There is no need to set epevin.data.fd as we have already set epevin.data.ptr to 5 earlier 2009-04-07 12:45:19 +00:00
ompi_setup_wrappers.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
ompi_try_assemble.m4 Rollback the last two commits. 2008-10-28 17:22:29 +00:00
opal_check_attributes.m4 - As proposed in RFC and telcon, warn the user about deprecated 2009-05-22 04:39:43 +00:00
opal_check_compiler_version.m4 - AC_TRY_RUN does not work in case of cross-compilation environments. 2009-06-04 13:00:16 +00:00
opal_check_ident.m4 This change goes along with r21180 to move files into 2009-05-06 20:15:34 +00:00
opal_check_offsetof.m4 Remove the definition of _WIN32_DCOM. It only enables DCOM when _WIN32_WINNT is less then 0x400 (Win 95, 98), and we are supporting 0x502(Win XP) and above in Open MPI. Thanks George for pointing this out. 2009-06-27 23:36:25 +00:00
opal_check_pthread_pids.m4 This change goes along with r21180 to move files into 2009-05-06 20:15:34 +00:00
opal_check_visibility.m4 This change goes along with r21180 to move files into 2009-05-06 20:15:34 +00:00
opal_config_asm.m4 This change goes along with r21180 to move files into 2009-05-06 20:15:34 +00:00
opal_config_threads.m4 This change goes along with r21180 to move files into 2009-05-06 20:15:34 +00:00
opal_configure_options.m4 - Split the datatype engine into two parts: an MPI specific part in 2009-07-13 04:56:31 +00:00
opal_functions.m4 (originally, did not want to do this during business hours, but 2009-05-28 20:28:43 +00:00
opal_setup_cc.m4 This change goes along with r21180 to move files into 2009-05-06 20:15:34 +00:00
orte_configure_options.m4 - Factor out OPAL, ORTE and OMPI specific configure-parameters... 2009-05-27 03:03:18 +00:00