Update to support passing of HWLOC shmem topology to client procs
Update use of distance API per @bgoglin
Have the openib component lookup its object in the distance matrix
Bring usnic up-to-date
Restore binding for hwloc2
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
All symbols that need to be accessed from a MCA component must be marked
explicitly as visible using PMIX_EXPORT. This patch allows current trunk
to almost work on OsX. More on the devel mailing list.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
passed to make it all flow thru the opal/pmix "put/get" operations. Update the PMIx code to latest master to pickup some required behaviors.
Remove the no-longer-required get_contact_info and set_contact_info from the RML layer.
Add an MCA param to allow the ofi/rml component to route messages if desired. This is mainly for experimentation at this point as we aren't sure if routing wi
ll be beneficial at large scales. Leave it "off" by default.
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
recent changes that broke native launch on cray
using srun or aprun was also broke native launch
using pmi2.
This commit fixes this problem.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
Passed the below set of symbols into a script that added ompi_ to them all.
Note that if processing a symbol named "foo" the script turns
foo into ompi_foo
but doesn't turn
foobar into ompi_foobar
But beyond that the script is blind to C syntax, so it hits strings and
comments etc as well as vars/functions.
coll_base_comm_get_reqs
comm_allgather_pml
comm_allreduce_pml
comm_bcast_pml
fcoll_base_coll_allgather_array
fcoll_base_coll_allgatherv_array
fcoll_base_coll_bcast_array
fcoll_base_coll_gather_array
fcoll_base_coll_gatherv_array
fcoll_base_coll_scatterv_array
fcoll_base_sort_iovec
mpit_big_lock
mpit_init_count
mpit_lock
mpit_unlock
netpatterns_base_err
netpatterns_base_verbose
netpatterns_cleanup_narray_knomial_tree
netpatterns_cleanup_recursive_doubling_tree_node
netpatterns_cleanup_recursive_knomial_allgather_tree_node
netpatterns_cleanup_recursive_knomial_tree_node
netpatterns_init
netpatterns_register_mca_params
netpatterns_setup_multinomial_tree
netpatterns_setup_narray_knomial_tree
netpatterns_setup_narray_tree
netpatterns_setup_narray_tree_contigous_ranks
netpatterns_setup_recursive_doubling_n_tree_node
netpatterns_setup_recursive_doubling_tree_node
netpatterns_setup_recursive_knomial_allgather_tree_node
netpatterns_setup_recursive_knomial_tree_node
pml_v_output_close
pml_v_output_open
intercept_extra_state_t
odls_base_default_wait_local_proc
_event_debug_mode_on
_evthread_cond_fns
_evthread_id_fn
_evthread_lock_debugging_enabled
_evthread_lock_fns
cmd_line_option_t
cmd_line_param_t
crs_base_self_checkpoint_fn
crs_base_self_continue_fn
crs_base_self_restart_fn
event_enable_debug_output
event_global_current_base_
event_module_include
eventops
sync_wait_mt
trigger_user_inc_callback
var_type_names
var_type_sizes
Signed-off-by: Mark Allen <markalle@us.ibm.com>
Remove the opal_ignore from the RML/OFI component, but disable that component unless the user specifically requests it via the "rml_ofi_desired=1" MCA param. This will let us test compile in various environments without interfering with operations while we continue to debug
Fix an error when computing the number of infos during server init
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
This now passes the loop test, and so we believe it resolves the random hangs in finalize.
Changes in PMIx master that are included here:
* Fixed a bug in the PMIx_Get logic
* Fixed self-notification procedure
* Made pmix_output functions thread safe
* Fixed a number of thread safety issues
* Updated configury to use 'uname -n' when hostname is unavailable
Work on cleaning up the event handler thread safety problem
Rarely used functions, but protect them anyway
Fix the last part of the intercomm problem
Ensure we don't cover any PMIx calls with the framework-level lock.
Protect against NULL argv comm_spawn
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Parts of the pmix2x component called the event_* functions directly
instead of the opal_event_* wrappers. This is fine as long as we are
using libevent but becomes a problem with other event libraries.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
Start updating the various mappers to the new procedure. Remove the stale lama component as it is now very out-of-date. Bring round_robin and PPR online, and modify the mindist component (but cannot test/debug it).
Remove unneeded test
Fix memory corruption by re-initializing variable to NULL in loop
Resolve the race condition identified by @ggouaillardet by resetting the
mapped flag within the same event where it was set. There is no need to
retain the flag beyond that point as it isn't used again.
Add a new job attribute ORTE_JOB_FULLY_DESCRIBED to indicate that all the job information (including locations and binding) is included in the launch message. Thus, the backend daemons do not need to do any map computation for the job. Use this for the seq, rankfile, and mindist mappers until someone decides to update them.
Note that this will maintain functionality, but means that users of those three mappers will see large launch messages and less performant scaling than those using the other mappers.
Have the mindist module add procs to the job's proc array as it is a fully described module
Protect the hnp-not-in-allocation case
Per path suggested by Gilles - protect the HNP node when it gets added in the absence of any other allocation or hostfile
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
The direct modex operation is slow, especially at scale for even modestly-connected applications. Likewise, blocking in MPI_Init while we wait for a full modex to complete takes too long. However, as George pointed out, there is a middle ground here. We could kickoff the modex operation in the background, and then trap any modex_recv's until the modex completes and the data is delivered. For most non-benchmark apps, this may prove to be the best of the available options as they are likely to perform other (non-communicating) setup operations after MPI_Init, and so there is a reasonable chance that the modex will actually be done before the first modex_recv gets called.
Once we get instant-on-enabled hardware, this won't be necessary. Clearly, zero time will always out-perform the time spent doing a modex. However, this provides a decent compromise in the interim.
This PR changes the default settings of a few relevant params to make "background modex" the default behavior:
* pmix_base_async_modex -> defaults to true
* pmix_base_collect_data -> continues to default to true (no change)
* async_mpi_init - defaults to true. Note that the prior code attempted to base the default setting of this value on the setting of pmix_base_async_modex. Unfortunately, the pmix value isn't set prior to setting async_mpi_init, and so that attempt failed to accomplish anything.
The logic in MPI_Init is:
* if async_modex AND collect_data are set, AND we have a non-blocking fence available, then we execute the background modex operation
* if async_modex is set, but collect_data is false, then we simply skip the modex entirely - no fence is performed
* if async_modex is not set, then we block until the fence completes (regardless of collecting data or not)
* if we do NOT have a non-blocking fence (e.g., we are not using PMIx), then we always perform the full blocking modex operation.
* if we do perform the background modex, and the user requested the barrier be performed at the end of MPI_Init, then we check to see if the modex has completed when we reach that point. If it has, then we execute the barrier. However, if the modex has NOT completed, then we block until the modex does complete and skip the extra barrier. So we never perform two barriers in that case.
HTH
Ralph
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Set the daemons' state to "running" and mark them as "alive" by default when constructing the nidmap
Get the DVM running again
Fix direct modex by eliminating race condition caused by releasing data while sending it
Up the size limit before compressing
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Cleanup a race condition segfault during finalize by ensuring the PMIx progress thread is stopped prior to starting to tear down the messaging components
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Do not use opal_output_verbose inside O(n) loops. This was causing us
to make O(n) calls to snprintf which was greatly slowing launch at
scale.
Signed-off-by: Nathan Hjelm <hjelmn@lanl.gov>
It's possible that we can have zlib.h but still not have zlib support.
Use the correct macro to protect the usage of calling zlib functions.
This fixes 32-bit MTT builds at Cisco (e.g.,
https://mtt.open-mpi.org/index.php?do_redir=2389).
Submitted upstream to PMIX: https://github.com/pmix/master/pull/290
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
PMIx_server_register_nspace() is an asynchronous operation, so
the pmix glue wait for it completes before returning.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
the argc field from the opal_pmix_app_t struct was removed,
so adjust the pmix/ext11 glue accordingly.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
on Linux, sending the header and then the message data does severely
impact performances of ptl/tcp :
on the receiver, reading the data can often result in an PMIX_ERR_RESOURCE_BUSY
or PMIX_ERR_WOULD_BLOCK, which ends up degrading performances)
this commit send both header and message data at the same time via writev()
and makes ptl/tcp virtually as efficient as ptl/usock.
Short writev generally occur when the kernel buffer is full, so there is no
point for retrying in this case.
fwiw, no such degradation was observed on OSX.
Refs open-mpi/ompi#2657
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Samples are taken after MPI_Init, and then again after MPI_Barrier. This allows the user to see memory consumption caused by add_procs, as well as any modex contribution from forming connections if pmix_base_async_modex is given.
Using the probe simply involves executing it via mpirun, with however many copies you want per node. Example:
$ mpirun -npernode 2 ./mpi_memprobe
Sampling memory usage after MPI_Init
Data for node rhc001
Daemon: 12.483398
Client: 6.514648
Data for node rhc002
Daemon: 11.865234
Client: 4.643555
Sampling memory usage after MPI_Barrier
Data for node rhc001
Daemon: 12.520508
Client: 6.576660
Data for node rhc002
Daemon: 11.879883
Client: 4.703125
Note that the client value on node rhc001 is larger - this is where rank=0 is housed, and apparently it gets a larger footprint for some reason.
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Revamp the event notification integration to rely on the PMIx event chaining and remove the duplicate chaining in OPAL. This ensures we get system-level events that target non-default handlers.
Restore the hostname entries for MPI-level error messages, but provide an MCA param (orte_hostname_cutoff) to remove them for large clusters where the memory footprint is problematic. Set the default at 1000 nodes in the job (not the allocation).
Begin first cut at memory profiler
Some minor cleanups of memprobe
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Plug a minor memory leak. Tell the PMIx server not to create a dstore memory region for the daemon job as there is nobody to share it with.
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Protect users of hwloc membind functions
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Update PMIx to include NULL string protection
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Update to PMIx master to include key overwrite protection
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
There are only five places in the non-daemon code paths where opal_hwloc_topology is currently referenced:
* shared memory BTLs (sm, smcuda). I have added a code path to those components that uses the location string
instead of the topology itself, if available, thus avoiding instantiating the topology
* openib BTL. This uses the distance matrix. At present, I haven't developed a method
for replacing that reference. Thus, this component will instantiate the topology
* usnic BTL. Uses the distance matrix.
* treematch TOPO component. Does some complex tree-based algorithm, so it will instantiate
the topology
* ess base functions. If a process is direct launched and not bound at launch, this
code attempts to bind it. Thus, procs in this scenario will instantiate the
topology
Note that instantiating the topology on complex chips such as KNL can consume
megabytes of memory.
Fix pernode binding policy
Properly handle the unbound case
Correct pointer usage
Do not free static error messages!
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Fine tuning of flux component
Fix a few minor issues with the initial cut:
* Job id could be obtained from the PMI kvsname like SLURM,
but simpler to getenv (FLUX_JOB_ID)
* Flux pmi-1 doesn't define PMI_BOOL, PMI_TRUE, PMI_FALSE
* Flux pmi-1 maps the deprecated PMI_Get_kvs_domain_id() to
PMI_KVS_Get_my_name() internally, so just call that instead.
* Drop residual slurm references.
Add wrappers for PMI functions so that if HAVE_FLUX_PMI_LIBRARY
is not defined, the component can dlopen libpmi.so at location
specified by the FLUX_PMI_LIBRARY_PATH env variable, which adds
flexibility. If HAVE_FLUX_PMI_LIBRARY is defined, link with
libpmi.so at build time in the usual way.
Update configury for flux component
Update m4 so the configure options work as follows:
--with-flux-pmi
Build Flux PMI support (default: yes)
--with-flux-pmi-library
Link Flux PMI support with PMI library at build
time. Otherwise the library is opened at runtime at
location specified by FLUX_PMI_LIBRARY_PATH environment
variable. Use this option to enable Flux support when
building statically or without dlopen support (default: no)
If the latter option is provided, the library/header is located at
build time using the pkg-config module 'flux-pmi'. Otherwise there
is no library/header dependency.
Handle the case where ompi is configured with --disable-dlopen
or --enable-statkc. In those cases, don't build the component
unless --with-flux-pmi-library is provided.
It is fatal if the user explicitly requests --with-flux-pmi but
it cannot be built (e.g. due to --disable-dlopen).
Add a schizo/flux component
Update schizo/flux component
Eliminate slurm-specific usage cases.
Since the module is only loaded if FLUX_JOB_ID is set, there are
only two cases to handle:
1) App was launched indirectly through mpirun. This is not yet
supported with Flux, but hook remains in case this mode is supported
in the future.
2) App was launched directly by Flux, with Flux providing
CPU binding, if any.
Fix up white space in pmix/flux component
Drop non-blocking fence from pmix:flux component
The flux PMI-1 library is not thread safe, therefore
register a regular blocking fence callback instead of the
thread-shifting fencenb().
pmix/flux component avoids extra PMI_KVS_Gets
Keys stored into the base cache under the wildcard
rank are not intended to be part of the global key namespace.
These keys therefore should not trigger a PMI_KVS_Get() if they
are not found in the cache.
Minor pmix/flux component cleanup
pmix/flux: drop code for fetching unused pmix_id
pmix/flux: err_exit must return error
Problem: in flux_init(), although 'ret' (variable holding
err_exit return code) is initialized to OPAL_ERROR, the
variable is reused as a temporary result code, so if there are
some successes followed by a failure that doesn't set 'ret',
flux_init() could return success with PMI not initialized.
Ensure that a "goto err_exit" returns OPAL_ERROR if 'ret'
is not set to some other error code.
pmix/flux: don't mix OPAL_ and PMI_ return codes
Problem: flux_init() can return both PMI_ and OPAL_ return
codes. Although OPAL_SUCCESS and PMI_SUCCESS are both defined
as 0, other codes are not compatible.
Ensure that flux_init() consistently uses 'rc' for PMI_
return codes and 'ret' for OPAL_ return codes.
pmix/flux: factor out repeated code for cache put
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Update ORTE support for dynamic PMIx operations e.g., PMIx_Spawn
Update to track master
Ensure that --disable-pmix-dstore actually disables the dstore. Sync to a few debugger updates
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
It turns that there is an incompatibility between the Cray PMI
library and the default configuration for building Open MPI (master).
To work around this, we now disable use of aprun for direct launch
of Open MPI jobs except under specific conditions.
The problem is that there are now (on master) packages getting
initialized that do not work properly across a fork operation.
As part of a constructor in the Cray PMI library, a fork operation
is done to simplify use of shared memory between the
processes in a job on the same node. This ends up thoroughly
messing up the Open MPI initialization process in the case
that dlopen support is enabled. The initialization process gets
about half-way through when the PMIX framework is opened and
components are loaded, which triggers the Cray PMI constructor
and hence the fork operation.
There are two workarounds for this:
1) configure Open MPI for Cray XE/XC systems using aprun with the
--disable-dlopen option
2) set the PMI_NO_FORK environment variable in the shell in which
the aprun command is run.
Without taking these measures, a Open MPI job will just hang at
job startup in the first attempt to "thread-shift" the PMIx
fence_nb operation. Additional hangs occur at shutdown if this
problem is worked around, again due to the insertion of a fork
operation halfway through the Open MPI initialization procedure.
This commit detects if the conditions that bring out the hang
situation are present, and if so, prints out a message and
aborts the job launch.
Note on systems using slurm, the PMI_NO_FORK environment variable
is set as part of the srun job launch, hence this issue is avoided
on those systems.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
PR open-mpi/ompi#2432 introduced a regression where configure
and build with --disable-dlopn caused build failure owing
to unresolved alps lli symbols in the libopal-pal shared library.
This commit fixes this problem.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
Enhance the cray pmix component to set some OMPI internal
env. variables used to set some key/value pairs
on the MPI_INFO_ENV object. This allows more of the
ompi-tests ibm unit tests to pass when using aprun/srun
direct launch and Cray PMI.
Signed-off-by: Howard Pritchard <howardp@lanl.gov>
Still not completely done as we need a better way of tracking the routed module being used down in the OOB - e.g., when a peer drops connection, we want to remove that route from all conduits that (a) use the OOB and (b) are routed, but we don't want to remove it from an OFI conduit.
- replace MAXHOSTNAMELEN with hardcoded 1024.
unlike Linux, Solaris #define MAXHOSTNAMELEN in <netdb.h>,
so use a hard coded value to keep the test simpl
- stdout cannot be assigned on Solaris, so use freopen instead
(back-ported from upstream commit pmix/master@a63f6e53f4)
this is a convenience macro similar to the PMIX_LIST_FOREACH macro,
that can be used to iterate on all the key/value pairs of a pmix_hash_table_t
(back-ported from upstream commit pmix/master@349971c68c)
theses three pmix components use the same class name,
declare it as static so Open MPI can be built with --disable-dlopen
Thanks Limin Gu for the report
pmix_progress_thread_finalize() invokes libevent event_base_free,
so all libevent stuff cannot be used after.
Hence, pmix_client_globals.myserver must be PMIX_DESTRUCT'ed
before invoking pmix_progress_thread_finalize()
pmix1_value_unload() was added a "key" argument which is unused,
and pmix1_value_unload() was sometimes invoked with two arguments instead of three.
since the "key" argument is unused, simply remove it from the
subroutine prototype and calls.
the LT_* macros do overwrite the enable_dlopen variable,
so it must be tested and saved before invoking LT_INIT.
delay the invokation of the LT_* macros and use the
PMIX_ENABLE_DLOPEN_SUPPORT variable to figure out whether
--disable-dlopen was invoked
