Timeline
10/07/09:
- 20:41 Ticket #893 (Hydra: Failures with repeating hostname) created by
- Hydra mpiexec seems to fail when you have two lines in the host file with …
- 19:59 Ticket #892 (Compilation error while compiling mpich2-1.2 with ch3:nemesis:mx device) created by
- Hi MPICH_Developers, The newest mpich2 seems to have problems compiling …
- 16:48 Changeset [5430] by
- Merge r5429 from trunk -- update README to say MPI-2.2 instead of MPI-2.1.
- 15:59 Changeset [5429] by
- The README was still saying MPI-2.1 instead of 2.2. Thanks to Rajeev for …
- 15:00 Milestone mpich2-1.2 completed
- 13:20 Ticket #891 (Can't initialize with MPI_THREAD_MULTIPLE when not run via mpiexec) created by
- MPI_Init_thread() always returns MPI_THREAD_SINGLE when the program is not …
- 12:16 Changeset [5425] by
- Retag mpich2-1.2
- 12:16 Changeset [5424] by
- Remove the older tag for mpich2-1.2.
- 12:15 Changeset [5423] by
- Merging r5422 from the trunk -- update to the CHANGES file to point to 1.2 …
- 12:13 Changeset [5422] by
- Forgot to modify the CHANGES file to point to 1.2 instead of 1.2rc1.
- 12:10 Changeset [5421] by
- Re-tagging mpich2-1.2.
- 12:09 Changeset [5420] by
- Removing the mpich2-1.2 tag to bring in r5419 as well.
- 11:42 Changeset [5419] by
- Merging r5418 from trunk - OPA defns for windows
- 10:37 Changeset [5418] by
- Handle the unhandled OPA defns on windows - added in r5395
10/06/09:
- 19:06 Changeset [5415] by
- Retag mpich2-1.2
- 19:05 Changeset [5414] by
- Remove the older tag for mpich2-1.2.
- 19:04 Changeset [5413] by
- Oops! Wrong location. My bad.
- 19:04 Changeset [5412] by
- Retag mpich2-1.2
- 19:03 Changeset [5411] by
- Merge r5410 from the trunk -- bug fix in the release.pl script to make …
- 19:00 Changeset [5410] by
- Fixed a bug in the release script that was causing the configure to not …
- 16:53 Changeset [5408] by
- Merge r5406 from the mpich2-1.2 release branch, to keep trunk up-to-date …
- 16:44 Changeset [5407] by
- Tagging the 1.2 release.
- 16:43 Changeset [5406] by
- Bumping the version number to 1.2
- 15:55 Changeset [5405] by
- Cutting a branch for the mpich2-1.2 release.
- 15:38 Ticket #885 (Can I submit a bug report?) closed by
- fixed: Ken has added a similar warning message for editing existing tickets as …
- 13:43 Ticket #890 (mpich2-1.1.1p1 lock in MPI_Init_thread/MPI_Comm_spawn_multiple) created by
- I'm calling MPI_Comm_spawn_multiple to create mpi procedures which …
- 05:40 Ticket #889 (Mpich2 not running on 64 bit multple node) created by
- Hi, I cant able to run mpi application on 64 bit windows OS. When i try to …
- 04:16 Changeset [5404] by
- Rmpe2_1_1_1
- 04:15 Changeset [5403] by
- Deleted this branch as some configure.ins do not have AC_PREREQ.
- 04:10 Changeset [5402] by
- Added back AC_PREREQ() to configure.in to work around the Debian/Ubuntu's …
- 02:42 Changeset [5401] by
- Rmpe2_1_1_1
- 02:39 Changeset [5400] by
- updated with clarification of cross-compile step, i.e. no cross-spec file …
10/05/09:
- 17:00 Changeset [5399] by
- Disable testing multiple combinations in a single fulltests.pl run. We …
- 13:45 Ticket #888 (ccreate1 error test fails) closed by
- fixed: Fixed in r5398.
- 13:39 Changeset [5398] by
- Improve error handling in MPI_Comm_create, fixes ccreate1 test (ticket …
- 10:56 Changeset [5397] by
- Updating the MPICH2 version on windows for the upcoming 1.2 rc release
10/03/09:
- 13:28 Ticket #844 (Hydra: rsh - support GNU inetutils rsh) closed by
- fixed: Yes, you are right. It does look like this is only a problem on older …
- 13:28 Changeset [5396] by
- Revert r5289. This was a bug in rsh that was fixed in the newer versions.
- 13:20 Ticket #678 (confdb macros) closed by
- fixed: Most of the cleanup is done. The only parts left are to get rid of the …
10/02/09:
- 17:19 Ticket #888 (ccreate1 error test fails) created by
- See: http://www.mcs.anl.gov/research/projects/mpich2/nightly/old/ […] …
- 17:15 Changeset [5395] by
- Flip the OPA svn:externals to r92 to pick up 64-bit PPC support. No …
- 13:09 Ticket #840 (possible MPD job launch performance regression in 1.1.1p1) closed by
- worksforme: I just retested ch3:nemesis:tcp on the eureka cluster at 796 processes on …
- 08:06 Ticket #598 (MPI_Aint+typecasting issues from IBM patches) closed by
- fixed: I believe that Bill has sorted this out at this point.
- 07:55 Ticket #216 (Incorrect behavior of MPICH2 C++ when Error handler MPI::ERRORS_RETURN is ...) closed by
- wontfix: Now that the C++ bindings have been deprecated in MPI-2.2, this ticket is …
10/01/09:
- 16:49 Changeset [5393] by
- Updating the upgrade code for MPICH2 on windows for upcoming rc release
- 16:46 Changeset [5392] by
- 1) Updating windows proj dependencies for ch3sock channel. This was reqd …
- 16:42 Changeset [5391] by
- Letting gnu compiler run within a cygwin bash prompt
- 16:37 Changeset [5390] by
- Adding decl for thread private vars - was missing in r3177. Without this …
- 16:35 Changeset [5389] by
- Adding channel deprecated msg for users choosing the SHM channel on …
- 13:11 Changeset [5388] by
- Remove the BLCR support information from the CHANGES file (this has been …
- 12:41 Ticket #887 (MPID_nem_init assumes malloc(0) returns a valid pointer) created by
- Assuming that malloc(0) is not null is a Linuxism. See lines 113 in …
- 11:43 Ticket #844 (Hydra: rsh - support GNU inetutils rsh) reopened by
- On what versions of rsh it was tested? There is fast test: rsh localhost …
- 11:40 Changeset [5387] by
- avoid warnings about ifname declaration shadowing another declaration
- 11:36 Ticket #883 (Problems in MPI_Comm_create in mpich2-1.1.1p1 andrecent SVN version) closed by
- fixed: Edric, this is fixed in r5386. Let us know if you have any trouble with …
- 11:31 Changeset [5386] by
- Fix MPI_Comm_create with dynamic procs (ticket #883). MPI_Comm_create was …
09/30/09:
- 13:08 Ticket #886 (clean target in hydra does not remove .o files) closed by
- invalid: I think that this is just detritus left over in your working copy from …
- 12:48 Ticket #886 (clean target in hydra does not remove .o files) created by
- […]
- 07:56 Ticket #771 (Problem with --with-aint-size=8) closed by
- fixed: It appears that I've found all the bugs identified here for …
- 07:51 Ticket #793 (Fwd: [mpich2-dev] fandcattrf90 test on BGP) closed by
- fixed: I believe my recent commits have fixed both the attribute problem when …
- 07:40 Changeset [5385] by
- Removed unnecessary CS_ENTER/EXIT
09/29/09:
- 16:58 Changeset [5384] by
- Require exactly autoconf 2.63 in maint/updatefiles. autoconf-2.64 is not …
- 15:58 Changeset [5383] by
- Fix sharedlib builds broken by the ABI change (r5359). No reviewer.
- 15:15 Changeset [5382] by
- Adding sizes of C99 types and handling other unhandled defns in windows …
- 15:01 Ticket #885 (Can I submit a bug report?) created by
- I'm not sure it this will work - trac tells me to […] I don't know how …
- 12:24 Changeset [5381] by
- Fixes for attributes when sizeof(MPI_Aint) > sizeof(void*), needed when …
- 12:20 Changeset [5380] by
- Quiet compiler warnings about casting pointers to integers of different …
- 12:18 Changeset [5379] by
- Ensure Fortran90 libraries are cleaned
- 12:17 Changeset [5378] by
- Ensure that shared libraries are cleaned as well and static libraries
- 12:09 Changeset [5377] by
- Add MPIU_THREADPRIV_DECLs to the routines needed by non-MPICH2 MPIs (and a …
- 11:06 Ticket #884 (MPI_Get_elements not returning MPI_UNDEFINED when expected) created by
- We have discovered a MPI_Get_elements() bug where the routine does not …
09/28/09:
- 17:22 Changeset [5376] by
- Make mpiu_valgrind.h inclusion unconditional to avoid non-debug link …
- 12:33 Changeset [5375] by
- Enhanced the F90 buildiface so that, like the F77 version, it may be used …
09/26/09:
- 17:20 Ticket #838 (req->dev.iov_offset and MPIDI_CH3U_Request_load_recv_iov()) closed by
- fixed: This should be fixed by r5374. OnDataAvail? functions (all functions …
- 17:04 Changeset [5374] by
- Fix subarray test failures introduced in r4099. The …
- 17:04 Changeset [5373] by
- Minor wibble, put the structure def outside of the macro. No reviewer.
- 17:04 Changeset [5372] by
- Add wrapper for the VALGRIND_CHECK_MEM_IS_DEFINED client request. No …
- 17:03 Changeset [5371] by
- Add missing NULL-pointer check in handle allocation code. No reviewer.
09/25/09:
- 12:03 Changeset [5370] by
- If the RMK already has the number of processes set, just use that instead …
- 11:46 Ticket #883 (Problems in MPI_Comm_create in mpich2-1.1.1p1 andrecent SVN version) created by
- Hi, I’ve been trying some stuff out to see if we can upgrade to …
- 10:16 Changeset [5369] by
- Initial draft of the PBS RMK.
09/24/09:
- 18:16 Ticket #882 (MPICH2 32-bit to 64-bit communication) created by
- I'm starting 3 processes from a batch file using mpiexec on MPICH2 1.0.7. …
- 12:21 Changeset [5368] by
- Use the correct type in the segment calls - it must be an MPI_Aint, not an …
- 11:30 Changeset [5367] by
- Added pragma to tell gcc (if a recent enough version is available) to …
- 11:28 Changeset [5366] by
- Fix misspelled comment
- 11:28 Changeset [5365] by
- Fix the make targets for configure files so that they are correctly built …
09/23/09:
- 20:23 Ticket #40 (Need test for type_create_indexed_block in ROMIO) closed by
- wontfix
- 20:16 Ticket #708 (mpich2-1 on an IBM AIX 5.2 machine) closed by
- wontfix: Don't have access to AIX 5.2
- 20:12 Ticket #779 (Lustre compilation error on FC7 Linux (but not FC4)) closed by
- fixed: Fixed in r5364
- 20:12 Changeset [5364] by
- fixes ticket #779
- 20:09 Ticket #796 (Troubleshooting) closed by
- worksforme: If you don't need to use Fortran, configure with --disable-f77
- 20:06 Ticket #875 (forrtl: severe (24): end-of-file during read, unit 5, file stdin) closed by
- worksforme
- 20:01 Ticket #881 (MPICH2 build fails on Solaris Opteron) created by
- Hi, I'm having issues compiling MPICH2 on Solaris 10 running on a 64-bit …
- 02:27 Changeset [5363] by
- Minor cleanups to the alltoallw_zeros test from r5362. Buffers weren't …
09/22/09:
- 20:35 Ticket #872 (problem in MPI_Alltoallw using empty data types) closed by
- fixed: This is fixed by r5362. Thanks again for the bug report. -Dave
- 20:34 Changeset [5362] by
- Fix alltoall{v,w} zero count/type hangs (ticket #872). This fixes the …
- 19:32 Ticket #880 (mpd generates a buffer overflow on a x86_64 computer (Intel i7)) closed by
- invalid: Hi John, It looks like you are using MPICH1 not MPICH2. MPICH1 is not …
- 19:19 Ticket #880 (mpd generates a buffer overflow on a x86_64 computer (Intel i7)) created by
- I installed mpich using the rpms from OpenSuse?-11.1. The machine is an …
- 15:33 Changeset [5361] by
- Use const for the file name to reduce compiler warning messages
- 13:19 Ticket #879 (weird error message with truncated message) closed by
- fixed: Fixed in mpich2 code base as well. resolving.
- 13:19 Changeset [5360] by
- fixes ticket #879
09/21/09:
- 06:32 Changeset [5359] by
- Updated the code to maintain the ABI versioning in the top-level …
09/20/09:
- 09:22 Changeset [5358] by
- Insure that Fortran ADDRESS_KIND is correctly set if --with-aint-size is …
09/18/09:
09/17/09:
- 15:41 Ticket #877 (Possible incorrect usage of *FUNC_ENTER* macro in MPIC_Sendrecv_replace()) closed by
- fixed: Fixed in r5357 -jayesh
- 15:32 Changeset [5357] by
- Explicitly declaring sendcount and recvcount that are logged by the rlog …
- 14:03 Changeset [5356] by
- Adding Scalable graph topology, MPI_Reduce_scatter_block, MPI_Reduce_local …
- 09:58 Changeset [5355] by
- Added (or moved) MPIU_THREADPRIV_DECL so that the thread-private area is …
- 08:56 Changeset [5354] by
- Ensure that a build failure in the mpi-io directory causes the make to …
09/16/09:
- 15:31 Ticket #856 (MPICH2 on Vista) closed by
- wontfix: resolving until we hear further
- 15:30 Ticket #847 (problem parallelizing on nehalems) closed by
- invalid
- 14:16 Ticket #877 (Possible incorrect usage of *FUNC_ENTER* macro in MPIC_Sendrecv_replace()) created by
- Dave, I believe you added this code in src/mpi/coll/helper_fns.c […] …
- 11:25 Changeset [5352] by
- bugfix for cross compilation.
- 10:30 Changeset [5351] by
- Using the NMPI_Isend() ver instead of directly using MPI_Isend(). Reviewed …
- 10:00 Changeset [5350] by
- Removing old checkpointing code in wintcp - was missed in r5325
09/15/09:
- 21:20 Ticket #634 (mpiexec command issue) reopened by
- […]
- 19:55 Changeset [5349] by
- added missing cflags to .c.lo target. Reported by Brian Smith.
- 19:09 Changeset [5347] by
- Change references to mpich2-maint to point to mpich-discuss in preparation …
- 18:35 Changeset [5346] by
- Fixes for the subarray test, including the infamous " No Errors". Nemesis …
- 18:04 Changeset [5345] by
- Fix for failing rqstatus test. The MPIR_Status_set_empty macro doesn't …
- 16:52 Changeset [5344] by
- Adding support for C99 fixed-width types, C99 complex types, MPI_AINT and …
- 16:20 Ticket #876 (Fwd: Jumpshot question) created by
- […] jumpshot_backwards_arrow.PNG
- 13:32 Changeset [5343] by
- Fix for the timeout test failure. Every process manager seems to have its …
- 11:40 Ticket #875 (forrtl: severe (24): end-of-file during read, unit 5, file stdin) created by
- […] c.txt m.txt mi.txt
- 09:04 Changeset [5342] by
- Forgot to add the copyright notice.
09/14/09:
- 18:37 Changeset [5341] by
- Added a version of the failing test reported by Bryan Putnam to the test …
- 14:57 Changeset [5340] by
- Fix a leak in MPI_Comm_create. Thanks to Bill for catching this. It …
- 14:55 Changeset [5339] by
- Don't require that "weighted" is set for the no-edges case. No reviewer.
- 14:55 Changeset [5338] by
- Copy NULL arrays correctly in topoutil.c. This simplifies the handling of …
- 14:54 Changeset [5337] by
- Don't check memcpy pointer args when len==0. The MPIU_MEM_CHECK_MEMCPY …
- 14:54 Changeset [5336] by
- Correctly handle zero-sized reallocs that return NULL. Realloc can return …
- 14:45 Changeset [5335] by
- Hydra doesn't support spawn yet; disable spawn tests.
- 14:43 Changeset [5334] by
- Updates to the HTML generation script to cleanup older directories.
- 08:40 Ticket #872 (problem in MPI_Alltoallw using empty data types) created by
- […] alltoallvw_bug.c
09/13/09:
- 16:35 Changeset [5333] by
- The MPL's been moved to trunk. This directory is redundant.
09/12/09:
- 10:54 Changeset [5332] by
- Generate the test list in the expected format.
- 10:46 Changeset [5331] by
- Updated nightly test list and added support for PGI compilers in the …
09/11/09:
- 15:50 Changeset [5330] by
- Fixed a bug with respect to vfprintf usage that was causing some arguments …
- 09:37 Ticket #867 (Installing multiple versions of MPICH2 on a windows system) created by
- Currently we prohibit users from installing different versions of MPICH2 …
- 04:40 Ticket #847 (problem parallelizing on nehalems) reopened by
- […]
09/10/09:
- 12:57 Ticket #863 (Add more documentation on debugging MPI apps on windows) created by
- We need to update the existing doc(in windows developers guide) with more …
- 07:11 Changeset [5328] by
- Initial draft of the CHANGES file for the 1.2rc1 release.
- 06:46 Changeset [5327] by
- Upgrade slurm's support level.
- 06:38 Changeset [5326] by
- Fix for the timeout failure from the nightly tests.
- 06:30 Changeset [5325] by
- removing old checkpointing code
- 06:12 Changeset [5324] by
- Got rid of duplicate definition function.
- 03:24 Changeset [5322] by
- warning stomps and eliminated the need for parameters to …
- 03:09 Changeset [5321] by
- added ignores
09/09/09:
- 13:40 Ticket #852 (mpiexec return code 1 on success?) closed by
- worksforme: Resolving until we hear further
- 13:35 Ticket #847 (problem parallelizing on nehalems) closed by
- worksforme: Resolving until we hear further.
- 13:34 Ticket #853 (Problems with mpich) closed by
- worksforme
- 13:34 Ticket #854 (Hello, there are some problem about MPICH2, I did it many days , but I ...) closed by
- worksforme
- 13:27 Ticket #857 (some questions) closed by
- invalid: Yes, it supports all those MPI-2 features.
- 07:36 Changeset [5320] by
- removed unnecessary includes
- 07:23 Ticket #855 (hello ,May I ask you a question? thank you~!) closed by
- worksforme
- 07:03 Changeset [5317] by
- checkpointing interface fixes and warning stomps. Reviewed by pavan@
- 06:10 Changeset [5315] by
- We are planning to release 1.2rc1 first.
- 06:08 Changeset [5314] by
- Update ROMIO's external confdb links to use internal links.
- 06:02 Changeset [5313] by
- Update svn:ignore properties.
- 06:00 Changeset [5312] by
- Moved MPL into the mpich2 trunk and updated the MPL external links to …
- 05:48 Changeset [5311] by
- Cleanup the code to find directories that require autoreconf vs. autoconf.
- 01:24 Changeset [5309] by
- Fix for failing distgraph1 test on 64-bit platforms. No reviewer.
09/08/09:
- 11:40 Ticket #857 (some questions) created by
- […]
- 08:27 Changeset [5308] by
- Update MPI version number to 2.2 (MPI Forum ticket #101). Reviewed by …
- 08:05 Changeset [5307] by
- Scalable graph topology implementation (MPI Forum ticket #33). This …
- 08:05 Changeset [5306] by
- Add MPIU_REALLOC_* macros to simplify proper realloc usage. No reviewer.
- 03:05 Changeset [5305] by
- Update README to encourage users to send general installation/usage …
09/07/09:
- 12:17 Ticket #856 (MPICH2 on Vista) created by
- Hi , I have been trying to make some headaway on this for hours now but I …
- 11:36 Changeset [5304] by
- We need to verify the communicator before freeing it. D'oh.
- 11:34 Changeset [5303] by
- Added query functions to verify that the comm returned by the distributed …
- 08:28 Changeset [5302] by
- Added some more fun cases to the MPI_Dist_graph_create test code.
- 03:09 Changeset [5301] by
- Some more improvements to the error output. Better correlation to the …
Note: See TracTimeline
for information about the timeline view.
