* Changes in SLURM 2.4.1
========================
-- Fix bug for job state change from 2.3 -> 2.4 job state can now be preserved
correctly when transitioning. This also applies for 2.4.0 -> 2.4.1, no
state will be lost. (Thanks to Carles Fenoy)
* Changes in SLURM 2.4.0
========================
-- Cray - Improve support for zero compute note resource allocations.
Partition used can now be configured with no nodes nodes.
-- BGQ - make it so srun -i<taskid> works correctly.
-- Fix parse_uint32/16 to complain if a non-digit is given.
-- Add SUBMITHOST to job state passed to Moab vial sched/wiki2. Patch by Jon
Bringhurst (LANL).
-- BGQ - Fix issue when running with AllowSubBlockAllocations=Yes without
compiling with --enable-debug
-- Modify scontrol to require "-dd" option to report batch job's script. Patch
from Don Albert, Bull.
-- Modify SchedulerParamters option to match documentation: "bf_res="
changed to "bf_resolution=". Patch from Rod Schultz, Bull.
-- Fix bug that clears job pending reason field. Patch fron Don Lipari, LLNL.
-- In etc/init.d/slurm move check for scontrol after sourcing
/etc/sysconfig/slurm. Patch from Andy Wettstein, University of Chicago.
-- Fix in scheduling logic that can delay jobs with min/max node counts.
-- BGQ - fix issue where if a step uses the entire allocation and then
the next step in the allocation only uses part of the allocation it gets
the correct cnodes.
-- BGQ - Fix checking for IO on a block with new IBM driver V1R1M1 previous
function didn't always work correctly.
-- BGQ - Fix issue when a nodeboard goes down and you want to combine blocks
to make a larger small block and are running with sub-blocks.
-- BLUEGENE - Better logic for making small blocks around bad nodeboard/card.
-- BGQ - When using an old IBM driver cnodes that go into error because of
a job kill timeout aren't always reported to the system. This is now
handled by the runjob_mux plugin.
-- BGQ - Added information on how to setup the runjob_mux to run as SlurmUser.
-- Improve memory consumption on step layouts with high task count.
-- BGQ - quiter debug when the real time server comes back but there are
still messages we find when we poll but haven't given it back to the real
time yet.
-- BGQ - fix for if a request comes in smaller than the smallest block and
we must use a small block instead of a shared midplane block.
-- Fix issues on large jobs (>64k tasks) to have the correct counter type when
packing the step layout structure.
-- BGQ - fix issue where if a user was asking for tasks and ntasks-per-node
but not node count the node count is correctly figured out.
-- Move logic to always use the 1st alphanumeric node as the batch host for
batch jobs.
-- BLUEGENE - fix race condition where if a nodeboard/card goes down at the
same time a block is destroyed and that block just happens to be the
smallest overlapping block over the bad hardware.
-- Fix bug when querying accounting looking for a job node size.
-- BLUEGENE - fix possible race condition if cleaning up a block and the
removal of the job on the block failed.
-- BLUEGENE - fix issue if a cable was in an error state make it so we can
check if a block is still makable if the cable wasn't in error.
-- Put nodes names in alphabetic order in node table.
-- If preempted job should have a grace time and preempt mode is not cancel
but job is going to be canceled because it is interactive or other reason
it now receives the grace time.
-- BGQ - Modified documents to explain new plugin_flags needed in bg.properties
in order for the runjob_mux to run correctly.
-- BGQ - change linking from libslurm.o to libslurmhelper.la to avoid warning.
* Changes in SLURM 2.4.0.rc1
=============================
-- Improve task binding logic by making fuller use of HWLOC library,
especially with respect to Opteron 6000 series processors. Work contributed
by Komoto Masahiro.
-- Add new configuration parameter PriorityFlags, based upon work by
Carles Fenoy (Barcelona Supercomputer Center).
-- Modify the step completion RPC between slurmd and slurmstepd in order to
eliminate a possible deadlock. Based on work by Matthieu Hautreux, CEA.
-- Change the owner of slurmctld and slurmdbd log files to the appropriate
user. Without this change the files will be created by and owned by the
user starting the daemons (likely user root).
-- Reorganize the slurmstepd logic in order to better support NFS and
Kerberos credentials via the AUKS plugin. Work by Matthieu Hautreux, CEA.
-- Fix bug in allocating GRES that are associated with specific CPUs. In some
cases the code allocated first available GRES to job instead of allocating
GRES accessible to the specific CPUs allocated to the job.
-- spank: Add callbacks in slurmd: slurm_spank_slurmd_{init,exit}
and job epilog/prolog: slurm_spank_job_{prolog,epilog}
-- spank: Add spank_option_getopt() function to api
-- Change resolution of switch wait time from minutes to seconds.
-- Added CrpCPUMins to the output of sshare -l for those using hard limit
accounting. Work contributed by Mark Nelson.
-- Added mpi/pmi2 plugin for complete support of pmi2 including acquiring
additional resources for newly launched tasks. Contributed by Hongjia Cao,
NUDT.
-- BGQ - fixed issue where if a user asked for a specific node count and more
tasks than possible without overcommit the request would be allowed on more
nodes than requested.
-- Add support for new SchedulerParameters of bf_max_job_user, maximum number
of jobs to attempt backfilling per user. Work by BjæËrn-Helge Mevik,
University of Oslo.
-- BLUEGENE - fixed issue where MaxNodes limit on a partition only limited
larger than midplane jobs.
-- Added cpu_run_min to the output of sshare --long. Work contributed by
Mark Nelson.
-- BGQ - allow regular users to resolve Rack-Midplane to AXYZ coords.
-- Add sinfo output format option of "%R" for partition name without "*"
appended for default partition.
-- Cray - Add support for zero compute note resource allocation to run batch
script on front-end node with no ALPS reservation. Useful for pre- or post-
processing.
-- Support for cyclic distribution of cpus in task/cgroup plugin from Martin
Perry, Bull.
-- GrpMEM limit for QOSes and associations added Patch from BjæËrn-Helge Mevik,
University of Oslo.
-- Various performance improvements for up to 500% higher throughput depending
upon configuration. Work supported by the Oak Ridge National Laboratory
Extreme Scale Systems Center.
-- Added jobacct_gather/cgroup plugin. It is not advised to use this in
production as it isn't currently complete and doesn't provide an equivalent
substitution for jobacct_gather/linux yet. Work by Martin Perry, Bull.
Disable automatic detection of SGE and SLURM to avoid PLIST divergence.
Changes in OpenMPI 1.6
v1.6 is a major release; it effectively deprecates both the v1.4.x
and v1.5.x series, and replaces v1.4.x as the current "super stable" series.
A forthcoming v1.7 series will become the new "feature" series.
*All users* -- including production users and downstream providers --
are encouraged to upgrade to v1.6.
Note that v1.6 is ABI compatible with the entire v1.5.x series,
but is not ABI compatible with the v1.4.x series.
See http://www.open-mpi.org/software/ompi/versions/ for a description
of Open MPI's release methodology.
Here is a list of changes in v1.6 as compared to v1.5.5:
- Fix some process affinity issues. When binding a process, Open MPI
will now bind to all available hyperthreads in a core (or socket,
depending on the binding options specified).
--> Note that "mpirun --bind-to-socket ..." does not work on POWER6-
and POWER7-based systems with some Linux kernel versions. See
the FAQ on the Open MPI web site for more information.
- Add support for ARM5 and ARM6 (in addition to the existing ARM7
support). Thanks to Evan Clinton for the patch.
- Minor Mellanox MXM fixes.
- Properly detect FDR10, FDR, and EDR OpenFabrics devices.
- Minor fixes to the mpirun(1) and MPI_Comm_create(3) man pages.
- Prevent segv if COMM_SPAWN_MULTIPLE fails. Thanks to Fujitsu for
the patch.
- Disable interposed memory management in fakeroot environments. This
fixes a problem in some build environments.
- Minor hwloc updates.
- Array versions of MPI_TEST and MPI_WAIT with a count==0 will now
return immediately with MPI_SUCCESS. Thanks to Jeremiah Willcock
for the suggestion.
- Update VampirTrace to v5.12.2.
- Properly handle forwarding stdin to all processes when "mpirun
--stdin all" is used.
- Workaround XLC assembly bug.
- OS X Tiger (10.3) has not been supported for a while, so forcibly
abort configure if we detect it.
- Fix segv in the openib BTL when running on SPARC 64 systems.
- Fix some include file ordering issues on some BSD-based platforms.
Thanks to Paul Hargove for this (and many, many other) fixes.
- Properly handle .FALSE. return parameter value to attribute copy
callback functions.
- Fix a bunch of minor C++ API issues; thanks to Fujitsu for the patch.
- Fixed the default hostfile MCA parameter behavior.
- Per the MPI spec, ensure not to touch the port_name parameter to
MPI_CLOSE_PORT (it's an IN parameter).
instead of "tcsh". This builds ok for me on NetBSD; if it turns out to
cause trouble for anyone, revert it.
The motivation was that the BUILD_DEPENDS accepted either tcsh or
standalone-tcsh, and distbb was latching onto the latter and then
failing trying to install it. If the package turns out to really need
tcsh in some contexts, there are probably other ways to deal with this
issue.
Changes in 1.5.5
----------------
- Many, many portability configure/build fixes courtesy of Paul
Hargrove. Thanks, Paul!
- Fixed shared memory fault tolerance support compiler errors.
- Removed not-production-quality rshd and tmd PLM launchers.
- Minor updates to the Open MPI SRPM spec file.
- Fixed mpirun's --bind-to-socket option.
- A few MPI_THREAD_MULTIPLE fixes in the shared memory BTL.
- Upgrade the GNU Autotools used to bootstrap the 1.5/1.6 series to
all the latest versions at the time of this release.
- Categorically state in the README that if you're having a problem
with Open MPI with the Linux Intel 12.1 compilers, *upgrade your
Intel Compiler Suite to the latest patch version*, and the problems
will go away. :-)
- Fix the --without-memory-manager configure option.
- Fixes for Totalview/DDT MPI-capable debuggers.
- Update rsh/ssh support to properly handle the Mac OS X library path
(i.e., DYLD_LIBRARY_PATH).
- Make warning about shared memory backing files on a networked file
system be optional (i.e., can be disabled via MCA parameter).
- Several fixes to processor and memory affinity.
- Various shared memory infrastructure improvements.
- Various checkpoint/restart fixes.
- Fix MPI_IN_PLACE (and other MPI sentinel values) on OS X. Thanks to
Dave Goodell for providing the magic OS X gcc linker flags necessary.
- Various man page corrections and typo fixes. Thanks to Fujitsu for
the patch.
- Updated wrapper compiler man pages to list the various --showme
options that are available.
- Add PMI direct-launch support (e.g., "srun mpi_application" under
SLURM).
- Correctly compute the aligned address when packing the
datatype description. Thanks to Fujitsu for the patch.
- Fix MPI obscure corner case handling in packing MPI datatypes.
Thanks to Fujitsu for providing the patch.
- Workaround an Intel compiler v12.1.0 2011.6.233 vector optimization
bug.
- Output the MPI API in ompi_info output.
- Major VT update to 5.12.1.4.
- Rankfile 'P'hysical mapping is no longer available.
- Upgrade embedded Hardware Locality (hwloc) v1.3.2, plus some
post-1.3.2-release bug fixes. All processor and memory binding is
now done through hwloc. Woo hoo! Note that this fixes core binding
on AMD Opteron 6200 and 4200 series-based systems (sometimes known
as Interlagos, Valencia, or other Bulldozer-based chips).
- New MCA parameters to control process-wide memory binding policy:
hwloc_base_mem_alloc_policy, hwloc_base_mem_bind_failure_action (see
ompi_info --param hwloc base).
- Removed direct support for libnuma. Libnuma support may now be
picked up through hwloc.
- Added MPI_IN_PLACE support to MPI_EXSCAN.
- Various fixes for building on Windows, including MinGW support.
- Removed support for the OpenFabrics IBCM connection manager.
- Updated Chelsio T4 and Intel NE OpenFabrics default buffer settings.
- Increased the default RDMA CM timeout to 30 seconds.
- Issue a warning if both btl_tcp_if_include and btl_tcp_if_exclude
are specified.
- Many fixes to the Mellanox MXM transport.
Heavily based on Sun Grid Engine package (parallel/sge).
Open Grid Scheduler/Grid Engine is a free and open-source
batch-queuing system for distributed resource management.
OGS/GE is based on Sun Grid Engine, and maintained by the same
group of external (i.e. non-Sun) developers who started
contributing code to Sun Grid Engine since 2001.
The Portable Hardware Locality (hwloc) software package provides
a portable abstraction (across OS, versions, architectures, ...)
of the hierarchical topology of modern architectures, including
NUMA memory nodes, sockets, shared caches, cores and
simultaneous multithreading. It also gathers various system
attributes such as cache and memory information as well as the
locality of I/O devices such as network interfaces, InfiniBand
HCAs or GPUs. It primarily aims at helping applications with
gathering information about modern computing hardware so as to
exploit it accordingly and efficiently.
The Son of Grid Engine is a community project to continue Sun Grid Engine.
Notable changes in Son of Grid Engine
-------------------------------------
Version 8.0.0d
--------------
* Bug fixes
* Man and fixes
* Fix building with older gcc versions
* Provide load average in qstat XML output [#446, #454]
* Partially back out Univa change which broke classic spooling
* Fix -terse in sge_request [#777]
* Other changes (possibly-incompatible)
* Message fixes
Version 8.0.0c
--------------
* Bug fixes
* Man and other documentation fixes
* Build/installation fixes (particularly for Red Hat 6 and Linux 3)
* Fix group ids for submitted jobs [U]
* Fix default value of boolean with JSV [U]
* Windows fixes for helper crashes and Vista GUI jobs [U]
* Ensure parallel jobs are dispatched to the least loaded host [U]
* Correct ownership of qsub -pty output file; was owned by admin user [U]
* Fix format of Windows loadcheck.exe output [U]
* Read from stderr even if stdout is already closed in IJS [U]
* Fix PDC_INTERVAL=NEVER execd parameter [U]
* Fix accounting information for Windows GUI jobs [U]
* Increase default MAX_DYN_EC qmaster param [U]
* Fix qsub -sync y error message and enforce MAX_DYN_EC correctly [U]
* Fix job validation (-w e) behaviour [#716] [U]
* Fix qrsh input redirection [U]
* Avoid warning when submitting a qrsh job [U]
* Print start time in qstat -j -xml output [U]
* Don't raise an error changing resource request on waiting job [#806]
* Don't exit 0 on error with qconf -secl or -sep
* Include string.h in drmaa.h [#712]
* Fix process-scheduler-log with host aliases
* Enhancements
* Base qmake and qtcsh on the current gmake and tcsh source [#289,
#504, #832]
* Support "-binding linear" and "-binding linear:slots"
* Use the hwloc library for all topology information and core
binding, supporting more operating systems (now: AIX, Darwin,
FreeBSD, GNU/Linux, HPUX, MS Windows, OSF/1, Solaris), and more
hardware types (specifically AMD Magny Cours and similar)
* Add task number to execd "exceeds job ... limit"
* Other changes (possibly-incompatible)
* Modify default paths in build files and elsewhere [U]
* Assorted message fixes
* In RPMs, move qsched to qmaster package, and separate drmaa4ruby
* Default to newijs in load_sge_config.sh
* Default to sh, not csh for configured shell
Version 8.0.0b
--------------
* Bug fixes
* Build/installation fixes [including #424, #1349] [(U)]
* Fix execd init script [#1348]
* Man and other documentation fixes [including #614, #764] [(U)]
* Fix contents of admin mail properly [#1307, #1345]
* Fix qalter messages for -tc
* Fix build with -DSGE_PQS_API
* Fix group ids for submitted jobs [U]
* Enhancements
* Update qsched and add man page
* Other changes (possibly-incompatible)
* Avoid the use of /bin/ksh [#1306]
* Change installation defaults to classic spooling, not adding
shadow hosts, and not JMX. [(U)]
Version 8.0.0a
--------------
This is roughly a superset of Univa's 8.0.0 (the V800_TAG from
https://github.com/gridengine/gridengine), with thanks for that.
Changes made there which haven't been included in this version: PLPA
source not removed; some different build/installation defaults
(e.g. for JMX); Univa/UGE "branding" (partly because trademark status
is unknown); authuser not removed (for SDM and testing use),
* Bug fixes
* Many man and other documentation fixes [including #790, #776,
#769, #733, #610, #587, #581, #459, #456, #439, #255, #1288, #797,
#1271, #773] [(U)]
* Some program message fixes [(U)]
* Various build and installation fixes [including #761, #709, #656,
#616, #546, #536, #521, #491, #438, #414, #411, #383, #381, #138,
#455, #344, #438, #1311, #1272, #1273] [(U)]
* Ask for keystore password twice on installation
* Fix qmaster crashes with tightly integrated parallel jobs or
un-discoverable qinstance [#789] [U]
* Report 0 cores and sockets on unsupported Solaris hosts [U]
* Fix malloc hooks which caused crashes, particularly with SuSE 11
[#792, #748, #749] [U]
* Verify the pe task start user in execd in non-CSP mode [U]
* Fix binding parameters parsing [U]
* Fix JSV logging with multiple users submitting jobs on same submit
host [U]
* Fix unresponsive qmaster when modifying the global configuration
in a huge cluster [U]
* Speed up finishing tightly integrated jobs [U]
* Check consistency of JSV binding information properly [U]
* Fix broken project spooling, which caused loss of project when
restarting master when using core binding [U]
* Fix slotwise preemption failure to unsuspend one job per host [#775] [U]
* Fix problems retrieving passwd and group information with large
responses [#1295] [(U)]
* Fix JSV changing default of boolean [U]
* Fix ENABLE_RESCHEDULE_SLAVE=1 [U]
* Allow comma in CMDNAME with Perl JSV scripts [#803]
* Don't put queue into error state when supplementary group id
cannot be set [#185] [U]
* Don't convert LF to CRLF with qrsh -pty [U]
* Fix qconf segfault on bad subordination string [U]
* Fix group ids of submitted jobs [U]
* Disallow -masterq with serial jobs [#155] [U]
* Fix 100% CPU use by shepherd of qsh [U]
* Removed unnecessary binding warning on job starts [U]
* Fix qconf error reports when tmp directory has 755 permissions [U]
* Fix suspending of remote process on qrsh -pty yes <cmd> on Solaris [U]
* Fix starting jobs after global host changed [U]
* Reject invalid load_formula value [U]
* Fix handling of implicitly-requested exclusive resources [U]
* Fix execd vmem reporting on 64-bit Linux [U]
* Fix startup of execd on Windows Vista [U]
* Set xterm's path more appropriately on GNU/Linux [#557]
* Fix generation of admin email from failed jobs [#1307]
* Fix some ill-formed output from qstat -xml [#314]
* Fix handling of multi-line environment variables propagated to
shepherd [#395]
* Fix example MPI PE templates
* Fix bad quoting in JSV sh library
* Fix checking of consumables for parallel jobs across multiple hosts [U]
* Enhancements
* Additional and clarified documentation
* PAM modules for ssh tight integration and access control for
interactive jobs
* Initial core binding support for Solaris/SPARC64 [U]
* Some efficiency improvements and memory leaks fixed [U]
* Ports to S/390 and PARISC GNU/Linux [U]
* New complex m_thread [U]
* Show topology by default in qhost [U]
* qsub -pty switch [#704] [U]
* Improved qmon graphics [#530] [(U)]
* Include bash in default shell list [U]
* A JSV that rejects all jobs [U]
* Files for Scali-MPI
* Ruby DRMAA implementation
* Enable easy building against shared system libraries and use
system openssl and bdb binaries
* New scripts: "qsched" reports resource reservations; "status"
wraps qstat; enable/disable submission; node-selection (idle etc.)
* Restart argument for daemon init scripts
* Improved efficiency of shell JSV if used with bash
* Core dumps from crashing daemons enabled under Linux [U]
* Example host_aliases file [#154]
* Spec file for RPM packaging [#820]
* Other changes (possibly-incompatible)
* Show core binding by default in qstat, qhost (use -ncb for
compatibility) [U]
* Removed Berkeley DB RPC support (recently dropped by BDB) [U]
* Changed position in pending job list for user-rescheduled jobs
(exit99, qmod -rj) and OLD_RESCHEDULE_BEHAVIOR,
OLD_RESCHEDULE_BEHAVIOR_ARRAY_JOB parameters [U]
* Unified GNU/Linux arch strings (lx-*, from lx24-* and lx26-*) [U]
* Default to enabling core binding on GNU/Linux [U]
* Removed Sun service tags support [U]
* Removed obsolete SunHPCT5 files
SLURM is an open-source resource manager designed for Linux
clusters of all sizes. It provides three key functions. First it
allocates exclusive and/or non-exclusive access to resources
(computer nodes) to users for some duration of time so they can
perform work. Second, it provides a framework for starting,
executing, and monitoring work (typically a parallel job) on a
set of allocated nodes. Finally, it arbitrates contention for
resources by managing a queue of pending work.
===============================================================================
Changes in 1.4.1
===============================================================================
# OVERALL: Several improvements to the ARMCI API implementation
within MPICH2.
# Build system: Added beta support for DESTDIR while installing
MPICH2.
# PM/PMI: Upgrade hwloc to 1.2.1rc2.
# PM/PMI: Initial support for the PBS launcher.
# Several other minor bug fixes, memory leak fixes, and code cleanup.
A full list of changes is available using:
svn log -r8675:HEAD https://svn.mcs.anl.gov/repos/mpi/mpich2/tags/release/mpich2-1.4.1
... or at the following link:
https://trac.mcs.anl.gov/projects/mpich2/log/mpich2/tags/release/mpich2-1.4.1?action=follow_copy&rev=HEAD&stop_rev=8675&mode=follow_copy
===============================================================================
Changes in 1.4
===============================================================================
# OVERALL: Improvements to fault tolerance for collective
operations. Thanks to Rui Wang @ ICT for reporting several of these
issues.
# OVERALL: Improvements to the universe size detection. Thanks to
Yauheni Zelenko for reporting this issue.
# OVERALL: Bug fixes for Fortran attributes on some systems. Thanks
to Nicolai Stange for reporting this issue.
# OVERALL: Added new ARMCI API implementation (experimental).
# OVERALL: Added new MPIX_Group_comm_create function to allow
non-collective creation of sub-communicators.
# FORTRAN: Bug fixes in the MPI_DIST_GRAPH_ Fortran bindings.
# PM/PMI: Support for a manual "none" launcher in Hydra to allow for
higher-level tools to be built on top of Hydra. Thanks to Justin
Wozniak for reporting this issue, for providing several patches for
the fix, and testing it.
# PM/PMI: Bug fixes in Hydra to handle non-uniform layouts of hosts
better. Thanks to the MVAPICH group at OSU for reporting this issue
and testing it.
# PM/PMI: Bug fixes in Hydra to handle cases where only a subset of
the available launchers or resource managers are compiled
in. Thanks to Satish Balay @ Argonne for reporting this issue.
# PM/PMI: Support for a different username to be provided for each
host; this only works for launchers that support this (such as
SSH).
# PM/PMI: Bug fixes for using Hydra on AIX machines. Thanks to
Kitrick Sheets @ NCSA for reporting this issue and providing the
first draft of the patch.
# PM/PMI: Bug fixes in memory allocation/management for environment
variables that was showing up on older platforms. Thanks to Steven
Sutphen for reporting the issue and providing detailed analysis to
track down the bug.
# PM/PMI: Added support for providing a configuration file to pick
the default options for Hydra. Thanks to Saurabh T. for reporting
the issues with the current implementation and working with us to
improve this option.
# PM/PMI: Improvements to the error code returned by Hydra.
# PM/PMI: Bug fixes for handling "=" in environment variable values in
hydra.
# PM/PMI: Upgrade the hwloc version to 1.2.
# COLLECTIVES: Performance and memory usage improvements for MPI_Bcast
in certain cases.
# VALGRIND: Fix incorrect Valgrind client request usage when MPICH2 is
built for memory debugging.
# BUILD SYSTEM: "--enable-fast" and "--disable-error-checking" are once
again valid simultaneous options to configure.
# TEST SUITE: Several new tests for MPI RMA operations.
# Several other minor bug fixes, memory leak fixes, and code cleanup.
A full list of changes is available using:
svn log -r7838:HEAD https://svn.mcs.anl.gov/repos/mpi/mpich2/tags/release/mpich2-1.4
... or at the following link:
https://trac.mcs.anl.gov/projects/mpich2/log/mpich2/tags/release/mpich2-1.4?action=follow_copy&rev=HEAD&stop_rev=7838&mode=follow_copy
===============================================================================
Changes in 1.3.2
===============================================================================
# OVERALL: MPICH2 now recognizes the OSX mach_absolute_time as a
native timer type.
# OVERALL: Performance improvements to MPI_Comm_split on large
systems.
# OVERALL: Several improvements to error returns capabilities in the
presence of faults.
# PM/PMI: Several fixes and improvements to Hydra's process binding
capability.
# PM/PMI: Upgrade the hwloc version to 1.1.1.
# PM/PMI: Allow users to sort node lists allocated by resource
managers in Hydra.
# PM/PMI: Improvements to signal handling. Now Hydra respects Ctrl-Z
signals and passes on the signal to the application.
# PM/PMI: Improvements to STDOUT/STDERR handling including improved
support for rank prepending on output. Improvements to STDIN
handling for applications being run in the background.
# PM/PMI: Split the bootstrap servers into "launchers" and "resource
managers", allowing the user to pick a different resource manager
from the launcher. For example, the user can now pick the "SLURM"
resource manager and "SSH" as the launcher.
# PM/PMI: The MPD process manager is deprecated.
# PM/PMI: The PLPA process binding library support is deprecated.
# WINDOWS: Adding support for gfortran and 64-bit gcc libs.
# Several other minor bug fixes, memory leak fixes, and code cleanup.
A full list of changes is available using:
svn log -r7457:HEAD https://svn.mcs.anl.gov/repos/mpi/mpich2/tags/release/mpich2-1.3.2
... or at the following link:
https://trac.mcs.anl.gov/projects/mpich2/log/mpich2/tags/release/mpich2-1.3.2?action=follow_copy&rev=HEAD&stop_rev=7457&mode=follow_copy
Intel(r) Threading Building Blocks (Intel TBB) offers a rich and
complete approach to expressing parallelism in a C++ program.
It is a library that helps you take advantage of multi-core
processor performance without having to be a threading expert.
Intel TBB is not just a threads-replacement library. It represents
a higher-level, task-based parallelism that abstracts platform
details and threading mechanisms for scalability and performance.
Changes in v1.5.4 as compared to v1.5.3:
- Add support for the (as yet unreleased) Mellanox MXM transport.
- Add support for dynamic service levels (SLs) in the openib BTL.
- Fixed C++ bindings cosmetic/warnings issue with
MPI::Comm::NULL_COPY_FN and MPI::Comm::NULL_DELETE_FN. Thanks to
Julio Hoffimann for identifying the issues.
- Also allow the word "slots" in rankfiles (i.e., not just "slot").
(** also to appear in 1.4.4)
- Add Mellanox ConnectX 3 device IDs to the openib BTL defaults.
(** also to appear in 1.4.4)
- Various FCA updates.
- Fix 32 bit SIGBUS errors on Solaris SPARC platforms.
- Add missing ARM assembly code files.
- Update to allow more than 128 entries in an appfile.
(** also to appear in 1.4.4)
- Various VT updates and bug fixes.
- Update description of btl_openib_cq_size to be more accurate.
(** also to appear in 1.4.4)
- Various assembly "clobber" fixes.
- Fix a hang in carto selection in obscure situations.
- Guard the inclusion of execinfo.h since not all platforms have it. Thanks
to Aleksej Saushev for identifying this issue.
(** also to appear in 1.4.4)
- Support Solaris legacy munmap prototype changes.
(** also to appear in 1.4.4)
- Updated to Automake 1.11.1 per
http://www.open-mpi.org/community/lists/devel/2011/07/9492.php.
- Fix compilation of LSF support.
- Update MPI_Comm_spawn_multiple.3 man page to reflect what it
actually does.
- Fix for possible corruption of the environment. Thanks to Peter
Thompson for the suggestion. (** also to appear in 1.4.4)
- Enable use of PSM on direct-launch SLURM jobs.
- Update paffinity hwloc to v1.2, and to fix minor bugs affinity
assignment bugs on PPC64/Linux platforms.
- Let the openib BTL auto-detect its bandwidth.
- Support new MPI-2.2 datatypes.
- Updates to support more datatypes in MPI one-sided communication.
- Fix recursive locking bug when MPI-IO was used with
MPI_THREAD_MULTIPLE. (** also to appear in 1.4.4)
- Fix mpirun handling of prefix conflicts.
- Ensure mpirun's --xterm options leaves sessions attached.
(** also to appear in 1.4.4)
- Fixed type of sendcounts and displs in the "use mpi" F90 module.
ABI is preserved, but applications may well be broken. See the
README for more details. Thanks to Stanislav Sazykin for
identifying the issue. (** also to appear in 1.4.4)
- Fix indexed datatype leaks. Thanks to Pascal Deveze for supplying
the initial patch. (** also to appear in 1.4.4)
- Fix debugger mapping when mpirun's -npernode option is used.
- Fixed support for configure's --disable-dlopen option when used with
"make distclean".
- Fix segv associated with MPI_Comm_create with MPI_GROUP_EMPTY.
Thanks to Dominik Goeddeke for finding this.
(** also to appear in 1.4.4)
- Improved LoadLeveler ORTE support.
- Add new WindVerbs BTL plugin, supporting native OpenFabrics verbs on
Windows (the "wv" BTL).
- Add new btl_openib_gid_index MCA parameter to allow selecting which
GID to use on an OpenFabrics device's GID table.
- Add support for PCI relaxed ordering in the OpenFabrics BTL (when
available).
- Update rsh logic to allow correct SGE operation.
- Ensure that the mca_paffinity_alone MCA parameter only appears once
in the ompi_info output. Thanks to Gus Correa for identifying the
issue.
- Fixed return codes from MPI_PROBE and MPI_IPROBE.
(** also to appear in 1.4.4)
- Remove --enable-progress-thread configure option; it doesn't work on
the v1.5 branch. Rename --enable-mpi-threads to
--enable-mpi-thread-multiple. Add new --enable-opal-multi-threads
option.
- Updates for Intel Fortran compiler version 12.
- Remove bproc support. Farewell bproc!
- If something goes wrong during MPI_INIT, fix the error
message to say that it's illegal to invoke MPI_INIT before
MPI_INIT.
more machines. A job is typically a single command or a small script that
has to be run for each of the lines in the input. The typical input is a
list of files, a list of hosts, a list of users, or a list of tables.
