Consolidate the ompi_process_info and opal_process_info structs to
remove duplicate storage and conversion issues. Unwind some interweaving
of include files using opal.h. Silence a couple of warnings.
For now, set the arch to local if PMIX_ARCH is not found.
Signed-off-by: Ralph Castain <rhc@pmix.org>
For direct modex, all procs publish the selected pml module
and then at add_procs pml module for each proc is checked
against every other proc in the add_proc call.
For full modex, there is no change in functionality. Only Rank0
publishes its selected pml, all other procs in the add_proc call
check their selected pml against Rank0.
If pml's do not match, throw error and exit.
Signed-off-by: Dipti Kothari <dkothar@amazon.com>
adding PMIX_NUMA_RANK info to process metadata so that the local NUMA
rank can be accessed through the opal_process_info object.
Signed-off-by: Nikola Dancejic <dancejic@amazon.com>
The locality for remote procs is not provided as it is only a local
concept. Thus, you must _always_ use modex_recv_optional to ensure you
don't hang waiting for a response until dmodex times out.
Signed-off-by: Ralph Castain <rhc@pmix.org>
Do some code cleanup in the connect/accept code. Ensure that the OMPI
layer has access to the PMIx identifier for the process. Add macros for
converting PMIx names to/from strings. Cleanup a few of the simple test
programs. Add a little more info to a btl/tcp error message.
Signed-off-by: Ralph Castain <rhc@pmix.org>
As indicated in the MPI3.2 document 14.3.10 page 599 line 1, the only
MPI error code possible is MPI_SUCCESS. All other errors must be in the
error class MPI_T_ERR*.
Fix the return of few pvar/cvar function that failed to correctly
convert to an MPI error code.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Found a handful of other URLs that weren't https-ized, so I updated
them, too (after verifying that they support https, of course).
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
Pass the correct ompi_proc_t and array length to
mca_pml_base_pml_check_selected() during dynamic modex.
Signed-off-by: Yossi Itigin <yosefe@mellanox.com>
Add a framework to support different types of threading models including
user space thread packages such as Qthreads and argobot:
https://github.com/pmodels/argobotshttps://github.com/Qthreads/qthreads
The default threading model is pthreads. Alternate thread models are
specificed at configure time using the --with-threads=X option.
The framework is static. The theading model to use is selected at
Open MPI configure/build time.
mca/threads: implement Argobots threading layer
config: fix thread configury
- Add double quotations
- Change Argobot to Argobots
config: implement Argobots check
If the poll time is too long, MPI hangs.
This quick fix just sets it to 0, but it is not good for the
Pthreads version. Need to find a good way to abstract it.
Note that even 1 (= 1 millisecond) causes disastrous performance
degradation.
rework threads MCA framework configury
It now works more like the ompi/mca/rte configury,
modulo some edge items that are special for threading package
linking, etc.
qthreads module
some argobots cleanup
Signed-off-by: Noah Evans <noah.evans@gmail.com>
Signed-off-by: Shintaro Iwasaki <siwasaki@anl.gov>
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
Change ompi_mtl_ofi_get_endpoint() to call the active PML's
add_procs() rather than the OFI MTL add_procs() directly when
discovering a new process during operation.
Functionally, this has no impact in correct operation. However,
the current behavior means that the heterogenous and active PML
checks are not being executed in the dynamic discovery case.
Signed-off-by: Brian Barrett <bbarrett@amazon.com>
We currently save the hostname of a proc when we create the ompi_proc_t for it. This was originally done because the only method we had for discovering the host of a proc was to include that info in the modex, and we had to therefore store it somewhere proc-local. Obviously, this ccarried a memory penalty for storing all those strings, and so we added a "cutoff" parameter so that we wouldn't collect hostnames above a certain number of procs.
Unfortunately, this still results in an 8-byte/proc memory cost as we have a char* pointer in the opal_proc_t that is contained in the ompi_proc_t so that we can store the hostname of the other procs if we fall below the cutoff. At scale, this can consume a fair amount of memory.
With the switch to relying on PMIx, there is no longer a need to cache the proc hostnames. Using the "optional" feature of PMIx_Get, we restrict the retrieval to be purely proc-local - i.e., we retrieve the info either via shared memory or from within the proc-internal hash storage (depending upon the active PMIx components). Thus, the retrieval of a hostname is purely a local operation involving no communication.
All RM's are required to provide a complete hostname map of all procs at startup. Thus, we have full access to all hostnames without including them in a modex or having to cache them on each proc. This allows us to remove the char* pointer from the opal_proc_t, saving us 8-bytes/proc.
Unfortunately, PMIx_Get does not currently support the return of a static pointer to memory. Thus, even though PMIx has the hostname in its memory, it can only return a malloc'd version of it. I have therefore ensured that the return from opal_get_proc_hostname is consistently malloc'd and free'd wherever used. This shouldn't be a burden as the hostname is only used in one of two circumstances:
(a) in an error message
(b) in a verbose output for debugging purposes
Thus, there should be no performance penalty associated with the malloc/free requirement. PMIx will eventually be returning static pointers, and so we can eventually simplify this method and return a "const char*" - but as noted, this really isn't an issue even today.
Signed-off-by: Ralph Castain <rhc@pmix.org>
OMPI can only support PMIx v3 and above. PRRTE requires at least PMIx
v4, so protect against the case where OMPI is built against an external
PMIx v3.
Fix check of PMIx_Init return code for singleton operations.
Ensure that the PMIx framework gets properly opened.
Signed-off-by: Ralph Castain <rhc@pmix.org>
Per the devel mailing list, we discussed the need/desirability of this change. Here is the logic behind not including it:
If you call "hwloc_topology_load", then hwloc merrily does its discovery and slams many-core systems. If you call "opal_hwloc_get_topology", then that is fine - it checks if we already have it, tries to get it from PMIx (using shared mem for hwloc 2.x), and only does the discovery if no other method is available.
We previously decided to let those who need the topology call "opal_hwloc_get_topology" to ensure the topo was available so that we don't load it unless someone actually needs it - in the case where it isn't available via PMIx, this avoids paying the startup time and memory footprint penalties for no reason. The function is protected so it will simply return SUCCESS if the topology is already defined.
After discussion, it was decided to stick with that "only setup the topology if someone actually needs it" approach. Hence, we will not blanket init the topology, and the mtl/ofi component will call opal_hwloc_get_topology to ensure the topo has been defined prior to using it.
Signed-off-by: Austen Lauria <awlauria@us.ibm.com>
Use "prte" instead of "prun" for proxy execution of cmds like mpirun.
This avoids the fork/exec-rendezvous complexities and should result in
more reliable operation.
Signed-off-by: Ralph Castain <rhc@pmix.org>
Restore missing call to get_topology - others were doing it in their
components as repeated calls just return success, but let's ensure it is
always present.
Signed-off-by: Ralph Castain <rhc@pmix.org>
Extend the PMIx modex recv macros to cover the full set of
immediate/optional combinations. If PMIx_Init cannot reach a server,
then declare the MPI proc to be a singleton.
Provide full support for info values via PMIx
Catch all the values used in the "info" area of OMPI using data
available from PMIx instead of via envars. Update PMIx and PRRTE to sync
with their capabilities.
PMIx
- ensure cleanup of fork/exec children
- fix bug in gds/hash that left app info off of list
PRRTE
- fix multi-app bugs
- port setup_child logic from orte
- OMPI env changes
- set app->first_rank
- ensure common hostname across prun, prte, and pmix
- Fix "nolocal" support
Silence a warning from btl/vader
Signed-off-by: Ralph Castain <rhc@pmix.org>
The fence logic in MPI_Init got messed up somehow such that we were
always executing a fence, which is not desirable. The logic is supposed
to be:
* if async fence is requested and we are not collecting data, then do
not fence at all
* if async fence is requested and we are collecting data, then execute
the fence in the background - wait for completion at the end of MPI_Init.
* if async fence is not requested, then execute a blocking fence at that
point, collecting data as directed. Note that we cannot actually do a
blocking fence as we need to cycle the event library via opal_progress
as the PMIx progress thread is tied to the OMPI event base.
Signed-off-by: Ralph Castain <rhc@pmix.org>
orte is gone, and we don't want to require other
RM's to either use a prrte specific env, or to
set their own.
OMPI_MCA_orte_ess_num_procs -> OMPI_MCA_num_procs
OMPI_MCA_orte_cpu_type -> OMPI_MCA_cpu_type
See PRRTE PR's:
https://github.com/openpmix/prrte/pull/443https://github.com/openpmix/prrte/pull/440
Signed-off-by: Austen Lauria <awlauria@us.ibm.com>
This commit fixes an issue with the include usage in some
ompi source files. These source files are using the <> form
of include when the "" form is correct (as these are internal,
**not** system headers).
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Signed-off-by: Nathan Hjelm <hjelmn@google.com>
- Port memchecker call from a1d502c.
- Remove unused memcheck macro variables.
- Some code readability improvements.
- Remove some stray +1's in dynamic comm cleanup.
- Re-add OPAL_ENABLE_DEBUG macro to osc header.
- Cleanup some printf's, and includes.
- Refactor cleanup of dpm_disconnect_objs.
Signed-off-by: Austen Lauria <awlauria@us.ibm.com>
- fix handling of contiguous datatypes with a non-zero true lower bound
- fix handling of datatypes using block of non contiguous predefined datatypes
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
OpenPMIx fills in a variety of info when it detects that we are in
singleton mode. Best way of detecting it is to look for the "singleton"
at the beginning of the returned nspace.
Make the modex recvs optional so we don't bounce up to the server and
then to the host trying to retrieve job-level info that must be given to
us at job start.
Signed-off-by: Ralph Castain <rhc@pmix.org>
Temporary solution for the PML inconsistency issue discussed in #7475.
This patch address 2 things: first it make the PMIx key optional so that
if we are not in a full modex mode we don't do a direct modex, and
second it get the PML info from the vpid 0 instead of from the local
rank.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
This commit contains the following changes:
The C++ bindings were removed from the standard in MPI-3.0. This
commit removes the entirety of the C++ bindings as well as the
support configury.
Removes all references to C++ from the man pages. This includes the
bindings themselves, all references to what C++ bindings return,
all not-available comments, and differences between C++ and other
language bindings.
If the user passes --enable-mpi-cxx, --enable-mpi-cxx-seek, or
--enable-cxx-exceptions, print a warning message an abort configure.
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
Signed-off-by: Nathan Hjelm <hjelmn@google.com>
1- Remove the common symbols issue: global variable not initialized. (#7424)
Move the variables to local scope within the set_info function.
2- Remove GPFS hints using datashipping: not used anymore
3- Redirect output stream to corresponding fs framework.
Signed-off-by: raafatfeki <fekiraafat@gmail.com>
The communicator might be not existent yet when mca_fs_gpfs_component_file_query() is called.
Therefore, we need to check it first before calling brodcast function.
Signed-off-by: raafatfeki <fekiraafat@gmail.com>
This commit changes the behavior of the individual sharedfp component. If
the component cannot create either the datafile or the metadatafile during File_open,
no error is being raised going forward. This allows applications that do not use shared
file pointer operations to continue execution without any issue.
If the user however subsequently calls MPI_File_write_shared or similar operations, an error
will be raised.
Fixes issue #7429
Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
If someone gives us a namespace that doesn't easily translate to an
integer, we have to create a mechanism for working around the
disconnect. PRRTE has been updated to give us a flag so we know we were
"natively" launched. If we don't see it, then fall back to generating a
hash of the nspace as our jobid. We then have to translate back/forth
between nspace and jobid using a lookup table.
Probably not the right long-term solution, but hopefully helps get us
thru for a bit.
Includes update of PRRTE pointer
Signed-off-by: Ralph Castain <rhc@pmix.org>
This commit fixes an issue with non-debug builds where adding an
attachment to the attachment list doesn't actually happen. This
causes all MPI_Win_detach calls to fail. The call was within an
assert which is optimized out in optimized builds.
Signed-off-by: Nathan Hjelm <hjelmn@google.com>
This commit increaes the osc_rdma_max_attach variable from 32
to 64. The new default is kept low due to the small number
of registration resources on some systems (Cray Aries). A
larger max attachement value can be set by the user on other
systems.
Signed-off-by: Nathan Hjelm <hjelmn@google.com>
This commit addresses two issues in osc/rdma:
1) It is erroneous to attach regions that overlap. This was being
allowed but the standard does not allow overlapping attachments.
2) Overlapping registration regions (4k alignment of attachments)
appear to be allowed. Add attachment bases to the bookeeping
structure so we can keep better track of what can be detached.
It is possible that the standard did not intend to allow #2. If that
is the case then #2 should fail in the same way as #1. There should
be no technical reason to disallow #2 at this time.
References #7384
Signed-off-by: Nathan Hjelm <hjelmn@google.com>
Fix a couple of spots in OMPI to resolve warnings. The one in comm_cid
in particular may be responsible for some/all of the comm_spawn issues
as it was passing an incorrect pointer to a macro, thus causing memory
corruption.
Update PRRTE and PMIx to deal with v3/v4 differences.
Signed-off-by: Ralph Castain <rhc@pmix.org>
- fix a typo `alloc_shared_contig` to `alloc_shared_noncontig`
- correct the value of `blocking_fence`
Signed-off-by: Tsubasa Yanagibashi <fj2505dt@aa.jp.fujitsu.com>
Will be replaced by PRRTE. Ensure that OMPI and OPAL layers build
without reference to ORTE. Setup opal/pmix framework to be static.
Remove support for all PMI-1 and PMI-2 libraries. Add support for
"external" pmix component as well as internal v4 one.
remove orte: misc fixes
- UCX fixes
- VPATH issue
- oshmem fixes
- remove useless definition
- Add PRRTE submodule
- Get autogen.pl to traverse PRRTE submodule
- Remove stale orcm reference
- Configure embedded PRRTE
- Correctly pass the prefix to PRRTE
- Correctly set the OMPI_WANT_PRRTE am_conditional
- Move prrte configuration to the end of OMPI's configure.ac
- Make mpirun a symlink to prun, when available
- Fix makedist with --no-orte/--no-prrte option
- Add a `--no-prrte` option which is the same as the legacy
`--no-orte` option.
- Remove embedded PMIx tarball. Replace it with new submodule
pointing to OpenPMIx master repo's master branch
- Some cleanup in PRRTE integration and add config summary entry
- Correctly set the hostname
- Fix locality
- Fix singleton operations
- Fix support for "tune" and "am" options
Signed-off-by: Ralph Castain <rhc@pmix.org>
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
Automake's Fortran compilation rules inexplicably use CPPFLAGS and
AM_CPPFLAGS. Unfortunately, this can cause problems in some cases
(e.g., picking up already-installed mpi.mod in a system-default
include search path).
So in relevant module-using Fortran compilation Makefile.am's, zero
out CPPFLAGS and AM_CPPFLAGS.
This has a side-effect of requiring that we compile the one .c file in
the F08 library in a new, separate subdirectory (with its own
Makefile.am that does _not_ have CPPFLAGS/AM_CPPFLAGS zeroed out).
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Fix the C types for the following:
* MPI_UNWEIGHTED
* MPI_WEIGHTS_EMPTY
* MPI_ARGV_NULL
* MPI_ARGVS_NULL
* MPI_ERRCODES_IGNORE
There is lengthy discussion on
https://github.com/open-mpi/ompi/pull/7210 describing the issue; the
gist of it is that the C and Fortran types for several MPI global
sentenial values should agree (specifically: their sizes must(**)
agree). We erroneously had several of these array-like sentinel
values be "array-like" values in C. E.g., MPI_ERRCODES_IGNORE was an
(int *) in C while its corresponding Fortran type was "integer,
dimension(1)". On a 64 bit platform, this resulted in C expecting the
symbol size to be sizeof(int*)==8 while Fortran expected the symbol
size to be sizeof(INTEGER, DIMENSION(1))==4.
That is incorrect -- the corresponding C type needed to be (int).
Then both C and Fortran expect the size of the symbol to be the same.
(**) NOTE: This code has been wrong for years. This mismatch of types
typically worked because, due to Fortran's call-by-reference
semantics, Open MPI was comparing the *addresses* of these instances,
not their *types* (or sizes) -- so even if C expected the size of the
symbol to be X and Fortran expected the size of the symbol to be Y
(where X!=Y), all we really checked at run time was that the addresses
of the symbols were the same. But it caused linker warning messages,
and even caused errors in some cases.
Specifically: due to a GNU ld bug
(https://sourceware.org/bugzilla/show_bug.cgi?id=25236), the 5 common
symbols are incorrectly versioned VER_NDX_LOCAL because their
definitions in Fortran sources have smaller st_size than those in
libmpi.so.
This makes the Fortran library not linkable with lld in distributions
that ship openmpi built with -Wl,--version-script
(https://bugs.llvm.org/show_bug.cgi?id=43748):
% mpifort -fuse-ld=lld /dev/null
ld.lld: error: corrupt input file: version definition index 0 for symbol
mpi_fortran_argv_null_ is out of bounds
>>> defined in /usr/lib/x86_64-linux-gnu/openmpi/lib/libmpi_usempif08.so
...
If we fix the C and Fortran symbols to actually be the same size, the
problem goes away and the GNU ld bug does not come into play.
This commit also fixes a minor issue that MPI_UNWEIGHTED and
MPI_WEIGHTS_EMPTY were not declared as Fortran arrays (not fully fixed
by commit 107c0073dd).
Fixesopen-mpi/ompi#7209
Signed-off-by: Fangrui Song <i@maskray.me>
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
Instead of triggering the fault early in the initialization process, do
a serialized initialization and report the error once all the supporting
infrastructure is up and running.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
OpenMPI doesn't compile anymore with IME because the header
file "ompi/mca/fs/base/base.h" needs to be include in every
file where mca_fs_base_get_mpi_err() is used.
Signed-off-by: Sylvain Didelot <sdidelot@ddn.com>
The opal_gethostname() function provides a more robust mechanism
to retrieve the hostname than gethostname(), which can return
results that are not null-terminated, and which can vary in its
behavior from system to system.
opal_gethostname() just returns the value in opal_process_info.nodename;
this is populated in opal_init_gethostname() inside opal_init.c.
-Changed all gethostname calls in opal subtree to opal_gethostname
-Changed all gethostname calls in orte subtree to opal_gethostname
-Changed all gethostname calls in ompi subdir to opal_gethostname
-Changed all gethostname calls in oshmem subdir to opal_gethostname
-Changed opal_if.c in test subdir to use opal_gethostname
-Changed opal_init.c to include opal_init_gethostname. This function
returns an int and directly sets opal_process_info.nodename per
jsquyres' modifications.
Relates to open-mpi#6801
Signed-off-by: Charles Shereda <cpshereda@lanl.gov>
Some versions of Libfabric contain a bug in EFA where FI_REMOTE_COMM and
FI_LOCAL_COMM are not advertised. In order to workaround this, we need to call
fi_getinfo() without those capability bits to see if EFA is available first.
Also move around some of the provider include/exclude list logic so we can skip
this workaround if applicable.
Signed-off-by: Robert Wespetal <wesper@amazon.com>
Make sure to get an RDM provider that can provide both local and
remote communication. We need this check because some providers could
be selected via RXD or RXM, but can't provide local communication, for
example.
Add OPAL_CHECK_OFI_VERSION_GE() m4 macro to check that the Libfabric
we're building against is >= a target version. Use this check in two
places:
1. MTL/OFI: Make sure it is >= v1.5, because the FI_LOCAL_COMM /
FI_REMOTE_COMM constants were introduced in Libfabric API v1.5.
2. BTL/usnic: It already had similar configury to check for Libfabric
>= v1.1, but the usnic component was checking for >= v1.3. So
update the btl/usnic configury to use the new macro and check for
>= v1.3.
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
Add MCA parameter to allow support for message logging even when the MPI
library is initialized with support for thread multiple.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Add a comment about the stages in the vprotocols initialization and why
we cant use the threading provided during the PML initialization.
Instead, use the OMPI internal threading status to prevent the use of
message logging in multi-threaded applications.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Change the provider include and exclude list name comparison check to
ignore case. The UDP provider's name is uppercase and was being selected
despite being in the exclude list.
Signed-off-by: Robert Wespetal <wesper@amazon.com>
Squash compiler warning due to whitespace/brace problems.
The code block from lines 829-839 was improperly indented, which led to
both the code being confusing and a compiler warning. Comparing this code to
the current version in the MPICH repo made it clear that the code was simply
improperly indented. Fixing the indentation both makes the code readable and
squashes the compiler warning.
Signed-off-by: Maxwell Coil <mcoil@nd.edu>
Squash compiler warning.
ROMIO is third-party software but has an annoying compiler warning;
this is the minimum distance fix.
Signed-off-by: William Bailey <wbailey2@nd.edu>
I removed the implementation and/or prototypes of all unused functions defined for all components.
To reduce recurrent code, I created functions under base for the management of error codes and setting of file permission and amode.
Then, I replaced these recurrent code by those function for all components.
Signed-off-by: raafatfeki <fekiraafat@gmail.com>
add a missing header file
Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
When O_CREAT and O_EXCL are set, open() shall fail if the file exists. Therefore, I assigned the file creation to the root process only.
I also translated the errno codes to their corresponding MPI error codes.
Signed-off-by: raafatfeki <fekiraafat@gmail.com>
I updated the gpfs component to match the latest structures and functions definitions and remove compiler wranings.
Disabled mpi info setting for the gpfs option "Data Shipping" since it is no longer supported in the latest gpfs versions.
Signed-off-by: raafatfeki <fekiraafat@gmail.com>
update the configure logic of the gpfs component
based on what we learned from user feedback over the last
two years for the other components
Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
Delete check for amode which should go to a higler layer, e.g. ompi_file_open.
Only perform Info value check if key is found.
Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
Until now sqrt(n) was missed as a factor for odd square numbers n. This
lead to suboptimal results of MPI_Dims_create for input numbers like 9,
25, 49, ... Fix the results by including sqrt(n) in the search for
factors.
Refs: #7186
Signed-off-by: Michael Lass <bevan@bi-co.net>
In fcoll_two_phase_supprot_fns.c: calculation of the aggregator index
failed for large offsets on 32bit machine, due to improper handling of
64bit offsets.
Fixes Issue #7110
Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
Compiling OMPI on cray systems using latest Cray compilers (clang based)
yielded some compiler warnings from ompio/lustre. Squash these warnings.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
This is closely related to Platform-MPI's old -prot feature.
The long-format of the tables it prints could look like this:
> Host 0 [myhost001] ranks 0 - 1
> Host 1 [myhost002] ranks 2 - 3
> Host 2 [myhost003] ranks 4
> Host 3 [myhost004] ranks 5
> Host 4 [myhost005] ranks 6
> Host 5 [myhost006] ranks 7
> Host 6 [myhost007] ranks 8
> Host 7 [myhost008] ranks 9
> Host 8 [myhost009] ranks 10
>
> host | 0 1 2 3 4 5 6 7 8
> ======|==============================================
> 0 : sm tcp tcp tcp tcp tcp tcp tcp tcp
> 1 : tcp sm tcp tcp tcp tcp tcp tcp tcp
> 2 : tcp tcp self tcp tcp tcp tcp tcp tcp
> 3 : tcp tcp tcp self tcp tcp tcp tcp tcp
> 4 : tcp tcp tcp tcp self tcp tcp tcp tcp
> 5 : tcp tcp tcp tcp tcp self tcp tcp tcp
> 6 : tcp tcp tcp tcp tcp tcp self tcp tcp
> 7 : tcp tcp tcp tcp tcp tcp tcp self tcp
> 8 : tcp tcp tcp tcp tcp tcp tcp tcp self
>
> Connection summary:
> on-host: all connections are sm or self
> off-host: all connections are tcp
In this example hosts 0 and 1 had multiple ranks so "sm" was more
meaningful than "self" to identify how the ranks on the host are
talking to each other. While host 2..8 were one rank per host so
"self" was more meaningful as their btl.
Above a certain number of hosts (12 by default) the above table gets too big
so we shrink to a more abbreviated looking table that has the same data:
> host | 0 1 2 3 4 8
> ======|====================
> 0 : A C C C C C C C C
> 1 : C A C C C C C C C
> 2 : C C B C C C C C C
> 3 : C C C B C C C C C
> 4 : C C C C B C C C C
> 5 : C C C C C B C C C
> 6 : C C C C C C B C C
> 7 : C C C C C C C B C
> 8 : C C C C C C C C B
> key: A == sm
> key: B == self
> key: C == tcp
Then above 36 hosts we stop printing the 2d table entirely and just print the
summary:
> Connection summary:
> on-host: all connections are sm or self
> off-host: all connections are tcp
The options to control it are
-mca comm_method 1 : print the above table at the end of MPI_Init
-mca comm_method 2 : print the above table at the beginning of MPI_Finalize
-mca comm_method_max <n> : number of hosts <n> for which to print a full size 2d
-mca comm_method_brief 1 : only print summary output, no 2d table
-mca comm_method_fakefile <filename> : for debugging only
* printing at init vs finalize:
The most important difference between these two is that when printing the table
during MPI_Init(), we send extra messages to make sure all hosts are connected to
each other. So the table ends up working against the idea of on-demand connections
(although it's only forcing the n^2 connections in the number of hosts, not the
total ranks). If printing at MPI_Finalize() we don't create any connections that
aren't already connected, so the table is more likely to have "n/a" entries if
some hosts never connected to each other.
* how many hosts <n> for which to print a full size 2d table
The option -mca comm_method_max <n> can be used to specify a number of hosts <n>
(default 12) that controls at what host-count the unabbreviated / abbreviated
2d tables get printed:
1 - n : full size 2d table
n+1 - 3n : shortened 2d table
3n+1 - inf : summary only, no 2d table
* brief
The option -mca comm_method_brief 1 can be used to skip the printing of the 2d
table and only show the short summary
* fakefile
This is a debugging option that allows easeir testing of all the printout
routines by letting all the detected communication methods between the hosts
be overridden by fake data from a file.
The source of the information used in the table is the .mca_component_name
In the case of BTLs, the module always had a .btl_component linking back to the
component. The vars mca_pml_base_selected_component and ompi_mtl_base_selected_component
offer similar functionality for pml/mtl.
So with the ability to identify the component, we can then access
the component name with code like this
mca_pml_base_selected_component.pmlm_version.mca_component_name
See the three lookup_{pml,mtl,btl}_name() functions in hook_comm_method_fns.c,
and their use in comm_method() to parse the strings and produce an integer
to represent the connection type being used.
Signed-off-by: Mark Allen <markalle@us.ibm.com>
This is based on a bug reported on the mailing list using a netcdf testcase.
The problem occurs if processes are using a custom file view, but on some
of them it appears as if the default file view is being used. Because of that,
the simple-grouping option lead to different number of aggregators used on different
processes, and ultimately to a deadlock. This patch fixes the problem by not using
the file_view size anymore for the calculation in the simple-grouping option,
but the contiguous chunk size (which is identical on all processes).
Fixes issue #7109
Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
In order to work around an issue with flang based compilers,
avoid declaring bind(C) constants and use plain Fortran parameter
instead.
For example,
type(MPI_Comm), bind(C, name="ompi_f08_mpi_comm_world") OMPI_PROTECTED :: MPI_COMM_WORLD
is changed to
type(MPI_Comm), parameter :: MPI_COMM_WORLD = MPI_Comm(OMPI_MPI_COMM_WORLD)
Note that in order to preserve ABI compatibility, ompi/mpi/fortran/use-mpi-f08/constants.{c,h}
have been kept even if its symbols are no more referenced by Open MPI.
Refs. open-mpi/ompi#7091
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
This commit fixes a segfault in mtl-portals4 finalize(). The segfault
occurs if finalize() is called without any calls to add_procs(). This
commit resolves the segfault by skipping the flow control fini() call if
Portals4 was not initialized.
Signed-off-by: Todd Kordenbrock <thkgcode@gmail.com>
zero-size derived datatypes are now flagged as OPAL_DATATYPE_FLAG_CONTIGUOUS
so update mca_pml_ucx_init_datatype() to correctly handle them.
Since 'size' is a 'size_t', the assertion can simply be removed.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
This commit updates the coll/basic component to correctly order sends
and receives for cartesian communicators with cyclic boundaries. This
addresses an issue identified by mpi-forum/mpi-issues#153. This issue
occurs when the size in any dimension is 1. This gives the same
neighbor in the positive and negative directions. The old code was
sending and receiving in the same order so the -1 buffer contained
the +1 result and vise-versa. The problem is addressed by using
unique tags for each send. This should cover both the case where
overtaking is allowed and is not allowed. The former case will be
possible is a MPI_Cart_create_with_info() call is added to the
standard.
Signed-off-by: Nathan Hjelm <hjelmn@google.com>
Change the ncounts argument to MPI_Count and use
MPI_Status_set_elements_x for enabling read/write operations beyond
the 2GB limit.
Thanks to Richard Warren from the HDF5 group for reporting the issue
and providing the suggested fix for romio.
Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
individual read/write operations exceeding 2GB fail in ompio
due to improper conversions from size_t to int in two different
locations. This commit fixes an issue reported by Richard Warren
from the HDF5 group.
Fixes Issue #397
Signed-off-by: Edgar Gabriel <egabriel@central.uh.edu>
one off patch for v4.0.x. for some reason commit on master
didn't have this problem.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
(cherry picked from commit 5f3dbdb5c8)
Note that this commit is actually a cherry-pick from the v4.0.x
branch. This is the opposite direction than what we nornmally do: we
usually commit to master first and then cherry-pick to the release
branches (vs. the other way around).
As is probably evident from the original commit message above, through
a comedy of errors, this commit was actually applied to the v4.0.x
branch first and then cherry-picked back to master (i.e., the problem
*did* exist in the original master commit
3aca4af548, but it was not recongized at
the time).
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
INTERNAL: STL-59403
The OFI (libfabric) MTL does not respect the maximum message size
parameter that OFI provides in the fi_info data.
This patch adds this missing max_msg_size field to the mca_ofi_module_t
structure and adds a length check to the low-level send routines.
Change-Id: I05aa71d332f2df897133b30c28bf37d98f061996
Signed-off-by: Michael Heinz <michael.william.heinz@intel.com>
Reviewed-by: Adam Goldman <adam.goldman@intel.com>
Reviewed-by: Brendan Cunningham <brendan.cunningham@intel.com>
I'm restoring the info function pointers to the IO module
but allowing the function pointers to be NULL (eg in ompio).
And letting romio321 set its function pointers for those
routines.
This means the info system uses the new OMPI-level info
system for most things, but skips it and uses the pre-existing
romio info system just for the romio module.
It's possible to convert romio, but I went a ways down that
path and found it kind of convoluted. Having pointers from
the lower level ADIO_File back to the higher level ompi_file_t
wasn't too bad, but I got stuck trying to figure out where/how
to register the infosubscribe_subscribe callbacks vs the way
initial k/v values are scattered around the romio code currently.
Signed-off-by: Mark Allen <markalle@us.ibm.com>
open-mpi/ompi@0fe756d416 Introduced
a bug in coll/hcoll component. The ompi_requests allocated by
libhcoll would be treated as coll_base_nbc_request during
ompi_coll_base_retain_<> call. Afterwards this would lead to a
segv in the request cleanup.
Fix: since libhcoll interface does not distinguish between the
blocling/non-blocking requests use coll_base_nbc_request all the
time and initialize it properly in
coll/hcoll/get_coll_handle(). It is still within 2 cache lines.
Signed-off-by: Valentin Petrov <valentinp@mellanox.com>
Within the shuffle iteration, the aggregators have to set a displacement array needed to receive data from other processes. The array had 1 extra element. We adjust the displacement index to match the number of elements.
Signed-off-by: raafatfeki <fekiraafat@gmail.com>
Within the shuffle iteration, the aggregators have to set a displacement array needed to receive data from other processes. The array had 1 extra element. We adjust the displacement index to match the number of elements.
Signed-off-by: raafatfeki <fekiraafat@gmail.com>
a non blocking collective might return ompi_request_null, so we should not
retain anything in that case.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Since ompi_coll_base_nbc_request_t is to be used in an
opal_free_list_t, it must be returned into a "clean" state.
So cleanup some data in the callback completion subroutines.
This fixes a regression introduced in open-mpi/ompi@0fe756d416
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
base ompi_coll_libnbc_request_t on top of ompi_coll_base_nbc_request_t
to correctly support the retention of datatypes/operators
This fixes a regression introduced in open-mpi/ompi@0fe756d416
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
The PSM MTL for Intel's TrueScale Infiniband HCAs is not being actively
maintained and should be removed from the master branch.
Fixes issue: #6877
Signed-off-by: Michael Heinz <michael.william.heinz@intel.com:
Split the sentinel library in ompi/mpi/fortran/use-mpi-f08 into
- the real sentinel that contains no code (only used to build the .mod files)
- an internal library that does contain some code
and have libmpi_usempif08.la slurp the latter.
This fixes a regression introduced in open-mpi/ompi@5de5e751ed
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
These issues were introduced in the recent commit b71af0eca0.
This commit fixes Coverity CID 1451661 and 1451660.
Though `c_info` part was an actual bug, the `c_sendtypes` part was not.
Signed-off-by: KAWASHIMA Takahiro <t-kawashima@jp.fujitsu.com>
Update PMIx to latest master to get supporting updates. For
connect/accept (part of comm_spawn as well), lookup locality for all
participating procs on the node and compute the relative locality so it
can be used for MPI operations.
Signed-off-by: Ralph Castain <rhc@pmix.org>
Use linear with sync alltoall algorithm for certain message/comm size
ranges. Does not affect default fixed decision, unless HPCX (with its
custom parameters) is used or corresponding mca is set.
Signed-off-by: Mikhail Brinskii <mikhailb@mellanox.com>
- ignore sendcounts, sendispls and sendtypes arguments when MPI_IN_PLACE is used
- use the right size when an inter-communicator is used.
Thanks Markus Geimer for reporting this.
Refs. open-mpi/ompi#5459
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
A sentinel is only an internal Fortran module and hence should not
be slurped into libmpi_usempif08.so
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
MPI standard states a user MPI_Op and/or user MPI_Datatype can be free'd
after a call to a non blocking collective and before the non-blocking
collective completes.
Retain user (only) MPI_Op and MPI_Datatype when the non blocking call is
invoked, and set a request callback so they are free'd when the MPI_Request
completes.
Thanks Thomas Ponweiser for reporting this
Fixesopen-mpi/ompi#2151Fixesopen-mpi/ompi#1304
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Move toward a base type of vector (count, type, blocklen, extent, disp)
with disp and extent applying toward the count repertition and blocklen
being a contiguous memory of type type.
Implement 2 optimizations on this description used during type_commit:
- collapse: successive similar datatype descriptions are collapsed
together with an increased count.
- fusion: fuse successive datatype descriptions in order to minimize the
number of resulting memcpy during pack/unpack.
Fixes at the OMPI datatype level including:
- Fix the create_hindexed and vector creation.
- Fix the handling of [get|set]_elements and _count.
- Correctly compute the dispacement for block indexed types.
- Support the MPI_LB and MPI_UB deprecation, aka. OMPI_ENABLE_MPI1_COMPAT.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Since array_of_errcodes is only allocated when MPI_ERRCODES_IGNORE is not used,
it should not be cleaned when MPI_ERRCODES_IGNORE is used.
Correctly allocate array_of_errcodes with the right size (e.g. maxprocs).
Thanks Gyevi-Nagy Laszlo for reporting this issue.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
do not check some input parameters when an {in,out}degree is zero
Thanks Junchao Zhang for analyzing and reporting this issue.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
To avoid fully initializing the osc/ucx component for MPI application
that are not using One-Sided functionality, the initialization happens
at the first MPI window creation.
This commit ensures atomicity of global state modifications.
Signed-off-by: Artem Polyakov <artpol84@gmail.com>
In common_ompi_aggregators calc_cost routine:
do not cast the real division to an int intermediately.
This patch removes the obsolete int variable c and assigns
the result of the P_a/P_x division directly to n_as.
With the intermediate int c variable, n_as gets 0 if P_a < P_x,
resulting in a division by 0 when computing n_s.
Signed-off-by: Harald Klimach <harald.klimach@uni-siegen.de>
The procedure names don't contain "_f08" of Fortran 2008 bindings of
Persistent Collective Operations(mpiext/pcollreq/use-mpi-f08).
This fix adds "_f08" to the procedure names of pcollreq/use-mpi-f08,
same as other Fortran 2008 routines in `ompi/mpi/fortran/use-mpi-f08/mod`.
Signed-off-by: Tsubasa Yanagibashi <fj2505dt@aa.jp.fujitsu.com>
Use the PVAR ctx to save the SPC index, so that no lookup nor
restriction on the SPC vars position is imposed.
Make sure the PVAR are always registered.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
