deactivated by default. It is activated by setting either of the
following two MCA parameters to values greater than 0:
* coll_sync_barrier_before
* coll_sync_barrier_after
If !_before is >0, then the sync coll collective will insert itself
before the underlying collective operations and invoke a barrier
before every Nth barrier (N == coll_sync_barrier_before). Similar for
!_after. Note that N is a _per communicator_ value; not global to the
MPI process.
If both are 0 (which is the default), this component returns NULL for
the comm query, meaning that it is not insertted into the coll module
stack.
The intent of this component is to provide a a workaround for
applications with large numbers of collectives of short messages that
can cause unbounded unexpected messages. Specifically, it is possible
for some iterative collective communication patterns to cause
unbounded unexpected messages. Forcing a barrier before or after
every Nth collective operation would prevent that behavior by forcing
applications to synchronize (and thereby consume any outstanding
unexpected messages caused by collectives on the same communicator).
Open MPI still needs to bound unexpected messages resource consumption
at the receiver, but this is a viable workaround for at least some
symptoms of the problem.
Additionally, there has been anecdotal evidence of some applications
that "perfom better" when they put barriers after other collective
operations. This could be due to many factors -- including shortening
the unexpected message queue. Putting this component in Open MPI
allows people to try this with their own applications and give real
world feedback on this kind of behavior.
This commit was SVN r20584.
* The main thing done here is to convert from multiple FIFOs/queues per
receiver (each receiver has one FIFO for each sender) to a single FIFO/queue
per receiver (all senders sharing the same FIFO for a given receiver).
* This requires rewriting the FIFO support, so that
ompi/class/ompi_[circular_buffer_]fifo.h is no longer used and FIFO
support is instead in btl_sm.h.
* The number of FIFOs per receiver is actually an MCA tunable parameter,
but it appears that 1 or possibly 2 FIFOs (even for 112 local processes)
per receiver is sufficient.
This commit was SVN r20578.
Often, orte/util/show_help.h is included, although no functionality
is required -- instead, most often opal_output.h, or
orte/mca/rml/rml_types.h
Please see orte_show_help_replacement.sh commited next.
- Local compilation (Linux/x86_64) w/ -Wimplicit-function-declaration
actually showed two *missing* #include "orte/util/show_help.h"
in orte/mca/odls/base/odls_base_default_fns.c and
in orte/tools/orte-top/orte-top.c
Manually added these.
Let's have MTT the last word.
This commit was SVN r20557.
by r20496 for the sm BTL, openib BTL on iWarp, and the sm & sm2 coll modules.
This commit was SVN r20515.
The following SVN revision numbers were found above:
r20496 --> open-mpi/ompi@4cdf91a8d4
The prior ompi_proc_t structure had a uint8_t flag field in it, where only one
bit was used to flag that a proc was "local". In that context, "local" was
constrained to mean "local to this node".
This commit provides a greater degree of granularity on the term "local", to include tests
to see if the proc is on the same socket, PC board, node, switch, CU (computing
unit), and cluster.
Add #define's to designate which bits stand for which local condition. This
was added to the OPAL layer to avoid conflicting with the proposed movement of
the BTLs. To make it easier to use, a set of macros have been defined - e.g.,
OPAL_PROC_ON_LOCAL_SOCKET - that test the specific bit. These can be used in
the code base to clearly indicate which sense of locality is being considered.
All locations in the code base that looked at the current proc_t field have
been changed to use the new macros.
Also modify the orte_ess modules so that each returns a uint8_t (to match the
ompi_proc_t field) that contains a complete description of the locality of this
proc. Obviously, not all environments will be capable of providing such detailed
info. Thus, getting a "false" from a test for "on_local_socket" may simply
indicate a lack of knowledge.
This commit was SVN r20496.
we remembered to use strcasecmp() every time I see this entry in the
file... (we did, but I just don't want to have to keep remembering
that ;-) )
This commit was SVN r20461.
btl_openib_connect_rdmacm_reject_causes_connect_error (yes, it's
still long -- on purpose :-) )
* Add INI file parameter rdmacm_reject_causes_connect_error
* Now only treat CONNECT_ERROR events as a REJECT if:
* It's on a connection where we were expecting a REJECT, ''and''
* The MCA parameter is true ''or'' the INI parameter for this
device is true
* Set the INI parameter for true for the NE020
This commit was SVN r20459.
versions of the NE driver will report the OUI while others will report
the PCI ID. We'll put in the Intel values when we get them (may not
be for a few more weeks).
This commit was SVN r20457.
properly at all. NetEffect's current driver (OFED 1.4.0) will return
a CONNECT_ERROR event to the initiator rather than the REJECTED event.
Doh! Additionally -- unfortunately -- NetEffect's vendor_id and
vendor_part_id are reported as 0 in OFED 1.4.0, so we can't
automatically detect these cards and work around the problem. So all
we can do is add a new MCA parameter
(btl_openib_connect_rdmacm_ignore_connect_errors -- yes, it's long on
purpose ;-) ) that says that if we get a CONNECT_ERROR, bascially
treat it exactly as a REJECT for the WRONG_DIRECTION reason (which is
a "good" reject). This allows OMPI to function with NetEffect/Intel
cards on OFED 1.4.0.
Note that NetEffect has been bought by Intel; I'm waiting for
information from them to update the ini file for their new OUI/PCI
ID's and/or new vendor_part_id values.
This commit was SVN r20454.
Fix a mistake in the dpm that hardcoded the update of routes to the HNP. This needs to be done by the individual routing modules so they can take whatever action is required - which will usually include updating the HNP, but might not...and might include additional steps. New routing modules are coming that violated this assumption, so it had to be moved back into init_routes.
All current routed modules know what to do - anyone with routed modules not in the current trunk may need to adjust them (see any of the current routed modules for examples of what to do).
This commit was SVN r20427.
lock -- therefore move the OPAL_THREAD_UNLOCK after
the if-OMPI_ERR_TEMP_OUT_OF_RESOURCE block.
- As mca_rcache_rb_mru_delete is the only setter of rc, move the
error-check right after mca_rcache_rb_mru_delete.
- Removed a few nitty ompi/info/info.h and orte/util/show_help.h
This commit was SVN r20355.
* New "op" MPI layer framework
* Addition of the MPI_REDUCE_LOCAL proposed function (for MPI-2.2)
= Op framework =
Add new "op" framework in the ompi layer. This framework replaces the
hard-coded MPI_Op back-end functions for (MPI_Op, MPI_Datatype) tuples
for pre-defined MPI_Ops, allowing components and modules to provide
the back-end functions. The intent is that components can be written
to take advantage of hardware acceleration (GPU, FPGA, specialized CPU
instructions, etc.). Similar to other frameworks, components are
intended to be able to discover at run-time if they can be used, and
if so, elect themselves to be selected (or disqualify themselves from
selection if they cannot run). If specialized hardware is not
available, there is a default set of functions that will automatically
be used.
This framework is ''not'' used for user-defined MPI_Ops.
The new op framework is similar to the existing coll framework, in
that the final set of function pointers that are used on any given
intrinsic MPI_Op can be a mixed bag of function pointers, potentially
coming from multiple different op modules. This allows for hardware
that only supports some of the operations, not all of them (e.g., a
GPU that only supports single-precision operations).
All the hard-coded back-end MPI_Op functions for (MPI_Op,
MPI_Datatype) tuples still exist, but unlike coll, they're in the
framework base (vs. being in a separate "basic" component) and are
automatically used if no component is found at runtime that provides a
module with the necessary function pointers.
There is an "example" op component that will hopefully be useful to
those writing meaningful op components. It is currently
.ompi_ignore'd so that it doesn't impinge on other developers (it's
somewhat chatty in terms of opal_output() so that you can tell when
its functions have been invoked). See the README file in the example
op component directory. Developers of new op components are
encouraged to look at the following wiki pages:
https://svn.open-mpi.org/trac/ompi/wiki/devel/Autogenhttps://svn.open-mpi.org/trac/ompi/wiki/devel/CreateComponenthttps://svn.open-mpi.org/trac/ompi/wiki/devel/CreateFramework
= MPI_REDUCE_LOCAL =
Part of the MPI-2.2 proposal listed here:
https://svn.mpi-forum.org/trac/mpi-forum-web/ticket/24
is to add a new function named MPI_REDUCE_LOCAL. It is very easy to
implement, so I added it (also because it makes testing the op
framework pretty easy -- you can do it in serial rather than via
parallel reductions). There's even a man page!
This commit was SVN r20280.
* If the accumulate is local, make it short-circuit the request path. Accumulate requires local
ops due to its window rules, so this is likely to help a bunch (on the codes I"m messing
with at least)
* Due a better job at flushing everything that can go out on the wire in a resource constrained problem
* Move some debugging values around to make large problems somewhat easier to deal with
This commit was SVN r20277.
to grow to. Without this change, jobs with np>120 get errors.
This does not change anything for np<16 jobs. It only comes into
play with larger np count on a node. I imagine that this can be
scaled back in the future if the usage of memory in the sm
btl is improved.
This fixes trac:1449.
This commit was SVN r20230.
The following Trac tickets were found above:
Ticket 1449 --> https://svn.open-mpi.org/trac/ompi/ticket/1449
