btl sendi functions currently can not handle the descriptor being NULL. The
send inline optimization was assuming (incorrectly) that NULL was ok.
cmr=v1.7.5:ticket=trac:4149
This commit was SVN r30364.
The following Trac tickets were found above:
Ticket 4149 --> https://svn.open-mpi.org/trac/ompi/ticket/4149
allgather.
The new collectives provide a signifigant performance increase over tuned for
small and medium messages. We are initially setting the priority lower than
tuned until this has had some time to soak in the trunk. Please set
coll_ml_priority to 90 for MTT runs.
Credit for this work goes to Manjunath Gorentla Venkata (ORNL), Pavel Shamis (ORNL),
and Nathan Hjelm (LANL).
Commit details (for reference):
Import ORNL's collectives for MPI_Allreduce, MPI_Reduce, and MPI_Allgather.
We need to take the basesmuma header into account when calculating the
ptpcoll small message thresholds. Add a define to bcol.h indicating the
maximum header size so we can take the header into account while not
making ptpcoll dependent on information from basesmuma.
This resolves an issue with allreduce where ptpcoll overwrites the
header of the next buffer in the basesmuma bank.
Fix reduce and make a sequential collective launcher in coll_ml_inlines.h
The root calculation for reduce was wrong for any root != 0. There are
four possibilities for the root:
- The root is not the current process but is in the current hierarchy. In
this case the root is the index of the global root as specified in the
root vector.
- The root is not the current process and is not in the next level of the
hierarchy. In this case 0 must be the local root since this process will
never communicate with the real root.
- The root is not the current process but will be in next level of the
hierarchy. In this case the current process must be the root.
- I am the root. The root is my index.
Tested with IMB which rotates the root on every call to MPI_Reduce. Consider
IMB the reproducer for the issue this commit solves.
Make the bcast algorithm decision an enumerated variable
Resolve various asset failures when destructing coll ml requests.
Two issues:
- Always reset the request to be invalid before returning it to the
free list. This will avoid an asset in ompi_request_t's destructor.
OMPI_REQUEST_FINI does this (and also releases the fortran handle
index).
- Never explicitly construct or destruct the superclass of an opal
object. This screws up the class function tables and will cause
either an assert failure or a segmentation fault when destructing
coll ml requests.
Cleanup allgather.
I removed the duplicate non-blocking and blocking functions and modeled
the cleanup after what I found in allreduce. Also cleaned up the code
somewhat.
Don't bother copying from the send to the recieve buffer in
bcol_basesmuma_allreduce_intra_fanin_fanout if the pointers are the
same.
The eliminates a warning about memcpy and aliasing and avoids an
unnecessary call to memcpy.
Alwasy call CHECK_AND_RELEASE on memsync collectives.
There was a call to OBJ_RELEASE on the collective communicator but
because CHECK_AND_RECYLCE was never called there was not matching call
to OBJ_RELEASE. This caused coll ml to leak communicators.
Make allreduce use the sequential collective launcher in coll_ml_inlines.h
Just launch the next collective in the component progress.
I am a little unsure about this patch. There appears to be some sort
of race between collectives that causes buffer exhaustion in some cases
(IMB Allreduce is a reproducer). Changing progress to only launch the
next bcol seems to resolve the issue but might not be the best fix.
Note that I see little-no performance penalty for this change.
Fix allreduce when there are extra sources.
There was an issue with the buffer offset calculation when there are
extra sources. In the case of extra sources == 1 the offset was set
to buffer_size (just past the header of the next buffer). I adjusted
the buffer size to take into accoun the maximum header size (see the
earlier commit that added this) and simplified the offset calculation.
Make reduce/allreduce non-blocking. This is required for MPI_Comm_idup
to work correctly.
This has been tested with various layouts using the ibm testsuite and
imb and appears to have the same performance as the old blocking version.
Fix allgather for non-contiguous layouts and simplify parsing the
topology.
Some things in this patch:
- There were several comments to the effect that level 0 of the
hierarchy MUST contain all of the ranks. At least one function
made this assumption but it was not true. I changed the sbgp
components and the coll ml initization code to enforce this
requirement.
- Ensure that hierarchy level 0 has the ranks in the correct
scatter gather order. This removes the need for a separate
sort list and fixes the offset calculation for allgather.
- There were several passes over the hierarchy to determine
properties of the hierarchy. I eliminated these extra passes
and the memory allocation associated with them and calculate the
tree properties on the fly. The same DFS recursion also handles
the re-order of level 0.
All these changes have been verified with MPI_Allreduce, MPI_Reduce, and
MPI_Allgather. All functions now pass all IBM/Open MPI, and IMB tests.
coll/ml: correct pointer usage for MPI_BOTTOM
Since contiguous datatypes are copied via memcpy (bypassing the convertor) we
need to adjust for the lb of the datatype. This corrects problems found testing
code that uses MPI_BOTTOM (NULL) as the send pointer.
Add fallback collectives for allreduce and reduce.
cmr=v1.7.5:reviewer=pasha
This commit was SVN r30363.
Per RFC. There are two optimizations in this commit:
- Allocate requests for blocking sends and receives on the stack. This
bypasses the request free list and saves two atomics on the critical path.
This change improves the small message ping-pong by 50-200ns on both AMD
and Intel CPUs.
- For small messages try to use the btl sendi function before intializing a
send request. If the sendi fails or the btl does not have a sendi function
silently fallback on the standard send path.
cmr=v1.7.5:reviewer=brbarret
This commit was SVN r30343.
Gilles Gouaillardet solution attached to ticket #4145.
Closes trac:4145.
cmr=v1.7.4:reviewer=ompi-rm1.7
cmr=v1.6.6:reviewer=ompi-rm1.6
This commit was SVN r30342.
The following Trac tickets were found above:
Ticket 4145 --> https://svn.open-mpi.org/trac/ompi/ticket/4145
Adds coll_hcoll_np mca parameter similar to that of fca component (defaults to 32). Those who use hcoll be aware that from now on the communicators less than 32 procs will run w/o hcoll by default. - Resolves fallback issue in case libhcoll runs out of allowed contexts. The solution is moving hcoll_context_create from comm_enable to comm_query. Shortly, comm_enable should never return OMPI_ERROR in the coll component with highest priority (hcoll). Otherwise the ompi coll_base_select will unselect the coll funtion pointers and module references leaving the communicator w/o coll pointer. This will cause the fail. Same behavior can be reproduced even with tuned if one would hardcore some "return OMPI_ERROR" into it's module_enable funtion. - Additionally, removed all the dead code under #if 0; removed unused variables (path for library, active_modules list) and classes (module list wrapper)
Fixed by Val, Reviewed by Devendar/Josh/Miked
cmr=v1.7.4:reviewer=ompi-rm1.7
This commit was SVN r30341.
This commit fixes an error path that occurs when huge page allocations are
enabled. In this case we allocate a huge page and try to register it but fail.
We then were calling free on the opal object. Fix this by calling the proper free
function.
cmr=v1.7.4:reviewer=rhc
This commit was SVN r30289.
Also add a verbose flag so one can see what devices are selected as well as another flag to override
locality information and use all devices on the node.
This commit was SVN r30287.
NetBSD puts the AIO functions in -lrt, vs. the usual libc. So we
need the fbtl/posix configure.m4 to test for -lrt properly.
Reviewed by Jeff Squyres.
cmr=v1.7.4:reviewer=ompi-rm1.7:subject=Fix NetBSD use of -laio
This commit was SVN r30274.
Add a configure test to see if the Fortran compiler supports the
PROTECTED keyword. If it does, use in mpi-f08-types.F90 (via a macro
defined in configure-fortran-output-bottom.h).
This is needed to support the PGI 9 Fortran compiler, which does not
support the PROTECTED keyword.
Note that regardless of whether we want to support the PGI 9 Fortran
compiler + mpi_f08, we need to correctly detect whether PROTECTED
works or not, and then use that determination as a criteria for
building the mpi_f08 module. Previously, mpi-f08-types.F90 used
PROTECTED unconditionally, and we didn't test for it in configure. So
if a compiler (e.g., PGI 9) supported everything else but didn't
support PROTECTED, it would try to compile the mpi_f08 stuff and choke
on the use of PROTECTED.
Refs trac:4093
This commit was SVN r30273.
The following Trac tickets were found above:
Ticket 4093 --> https://svn.open-mpi.org/trac/ompi/ticket/4093
1. Canary compile-time test: this is compiled whenever you compile
the entire OMPI tree. It's a noinst standalone library comprised
of a single .c file, so no one will notice its addition, and it
doesn't get linked/installed to any real build products. If we
are out of padding space on any predefined MPI object type, it
will fail to compile. This will alert/annoy a human, who will be
able to fix the real problem.
1. Added a "make check" test that will print out the amount of
predefined padding left on all the MPI object types.
This commit was SVN r30268.
NOTE: launch performance will be absolutely awful if you do this with BTLs that aren't configured to modex_recv on first message!
Even with "modex on demand", we still have to do a barrier in place of the modex - we simply don't move any data around, which does reduce the time impact. The barrier is required to ensure that the other proc has in fact registered all its BTL info and therefore is prepared to hand over a complete data package. Otherwise, you may not get the info you need. In addition, the shared memory BTL can fail to properly rendezvous as it expects the barrier to be in place.
This behavior will *only* take effect under the following conditions:
1. launched via mpirun
2. #procs is greater than ompi_hostname_cutoff, which defaults to UINT32_MAX
3. mca param rte_orte_direct_modex is set to 1. At the moment, we are having problems getting this param to register properly, so only the first two conditions are in effect. Still, the bottom line is you have to *want* this behavior to get it.
The planned next evolution of this will be to make the direct modex be non-blocking - this will require two fixes:
1. if the remote proc doesn't have the required info, then let it delay its response until it does. This means we need a way for the MPI layer to tell the RTE "I am done entering modex data".
2. adjust the SM rendezvous logic to loop until the required file has been created
Creating a placeholder to bring this over to 1.7.5 when ready.
cmr=v1.7.5:reviewer=hjelmn:subject=Enable direct modex at scale
This commit was SVN r30259.
This configure option was only relevant when we were generating TKR
"use mpi" interfaces for MPI subroutines with choice buffers. Now
that we aren't, the only interface that needs to accept a choice
buffer is MPI_SIZEOF (which we have to provide).
And since there's now only several dozen interfaces in the "mpi" TKR
module, there's no reason to not generate ''all'' possible array rank
values (when there were thousands of interfaces, generating 4-vs-7
array ranks per interface per type was a big deal). The default used
to be 4; now we can just hard-code it to 7, the max possible value for
Fortran 2003 (I think the max was raised ?to 11? in F2008, but let's
not go there for now).
cmr=v1.7.5:reviewer=dgoodell:subject=Remove even more dead Fortran configury
This commit was SVN r30257.
BIND(C), but not ''all'' of it. So expand our configure checks to
look for multiple different forms of BIND(C):
* ISO_C_BINDING
* SUBROUTINE ... BIND(C)
* TYPE, BIND(C)
* TYPE(foo), BIND(C, name="bar")
If the compiler supports all of these, then declare that we support
BIND(C), and the rest of the mpi_f08 checks can continue. If we miss
any one of those, don't bother continuing -- we won't build the
mpi_f08 module.
Also push the results of all of these tests down to ompi_info so that
they can be reported easily (e.g., "Hey, why doesn't my OMPI
installation have the mpi_f08 module?").
cmr=v1.7.4:reviewer=jsquyres:subject=Expand Fortran BIND(C) configure checks
This commit was SVN r30247.
LIBADD libmpi.la
cmr=v1.7.4:reviewer=brbarret:subject=Add libmpi to libmpi_usempif08_LIBADD
This commit was SVN r30245.
The following SVN revision numbers were found above:
r30244 --> open-mpi/ompi@7015343951
TKR LIBADDs libmpi_mpifh; there is no library for libmpi_usempi ignore
TKR).
Refs trac:4085
This commit was SVN r30244.
The following Trac tickets were found above:
Ticket 4085 --> https://svn.open-mpi.org/trac/ompi/ticket/4085
intended to be used and emits a compile-time warning.
Thanks to Paul Hargrove for identifying the issue.
cmr=v1.7.4:reviewer=hjelmn:subject=remove/replace malloc.h
This commit was SVN r30231.
Avoid compiler warning about (unnecessarily) initializing 2 variables
during instantiation at the top of a switch block (but outside of any
case statements): just declare the variables at the top of the outter
block. They're already safely initialized, so don't worry about
initializing them in the instantiation.
Reviewed by Dave Goodell.
cmr=v1.7.4:reviewer=ompi-rm1.7:subject=Don't instantiate+init variables in a switch block
This commit was SVN r30228.
ob1 dummy registration to actually be used when using udreg. Fix this by
always setting reg to NULL when mpool/udreg's register function fails.
cmr=v1.7.4:reviewer=rhc
This commit was SVN r30214.
It is now possible for orte_proc_applied_binding to be NULL (e.g., if
you mpirun --bind-to none), so we need to ensure we don't pass it down
to opal_hwloc_base_cset2*str().
Also, take the opprotunity to de-duplicate some strings that are used
in multiple places.
Refs trac:4073
This commit was SVN r30204.
The following Trac tickets were found above:
Ticket 4073 --> https://svn.open-mpi.org/trac/ompi/ticket/4073
Set comm attribute with keyval.
Wait for pending hcoll module tasks in comm delete callback where PML
still valid on the communicator. safely destroy hcoll context during
hcoll module destructor.
Author: Devendar Bureddy
reviewed by miked
cmr=v1.7.4:reviewer=ompi-rm1.7
This commit was SVN r30175.
