Search for the digits to be compressed from the end of the node names.
For example, if the nodelist is c712f6n01,c712f6n02,c712f6n03
the regx/fwd component generates c[3:712]f6n01,c[3:712]f6n02,c[3:712]f6n03@(3)
when the regx/reverse component generates c712f6n[2:1-3]@0(3) which is
a better fit here.
Josh Hursey authored the changes and must be credited.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
typedef int (*orte_regx_base_module_extract_node_names_fn_t)(char *regexp, char ***names);
among other things, that will make testing way easier.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Resolve a race condition between registering for a file to be removed upon termination and actual creation of that file by providing attributes that identify whether the path is a file or directory. This removes the need for PMIx to detect the difference.
Refs #4686
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Handle the need for different regex generator/parsers by moving the
orte/util/nidmap and orte/util/regex code into a new "regx" framework.
Use the original code to complete a "fwd" component, and create a
scaffold for IBM's "reverse" component.
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Now that the daemon calls remote_spawn itself, there is no longer
a need for the "tree_spawn" command nor the associated command
processing code since the HNP is no longer sending a tree-spawn
message to the orted.
Thanks Ralph for the guidance !
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
When the node regex is too long to be sent on the command line,
retrieve it first from the parent, and then spawn the remote orted
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
This parameter can be used to set the node regex max length that can
be passed to the orted command line.
For testing purpose, it can be set to zero in order to force the node regex
being retrieved by orted from its parent.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
since open-mpi/ompi@8f496b01b7
sstore_stage_local_compress_waitpid_cb is invoked with an orte_wait_tracker_t *,
that must be used to reach the orte_sstore_stage_local_app_snapshot_info_t *.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
since open-mpi/ompi@8f496b01b7
rsh_wait_daemon is invoked with an orte_wait_tracker_t *,
that must be used to reach the orte_plm_rsh_caddy_t *.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
If we detect that someone has given us an incorrect node name, provide a helpful message telling them as it is almost certainly a typo.
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
If available, have apps use registration capability to cleanup their session directories. Setup capability for vader to register its shared memory file location - let someone familiar with that code do so.
Final cleanup to track uid/gid, update the opal/pmix API to pass flags for ignore and leave top directory alone
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Somehow, the code for passing a daemon's parent was accidentally removed, thus breaking the tree-spawn callback sequence and causing all daemons to phone directly home. Note that this is noticeably slower than no-tree-spawn for small clusters where directly ssh launch of the child daemons from the HNP doesn't overload the available file descriptors.
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
'-' is not an alpha character nor a digit, but it is a valid hostname
character and should be handled as an alpha character, otherwise, nodes
such as node-001 do not get "compressed" in the regex.
Refs open-mpi/ompi#4621
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Only selectable when specifically requested via "-mca odls pspawn"
Note that there are several concerns:
* we aren't getting SIGCHLD calls when the procs terminate
* we aren't seeing the IO pipes close on termination, though
we are getting output forwarded to mpirun
* I haven't found a way to bind the child process prior to exec.
If we want to use this method, we probably need someone to
implement a cgroup component for the orte/rtc framework
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Change the determination of #spawn threads to be done on basis of #local procs in first job being spawned. Someone can look at an optimization that handles subsequent dynamic spawns that might be larger in size.
Leave the threads running, but blocked, for the life of the daemon, and use them to harvest the local procs as they terminate. This helps short-lived jobs in particular.
Add MCA params to set:
* max number of spawn threads (default: 4)
* set a specific number of spawn threads (default: -1, indicating no set number)
* cutoff - minimum number of local procs before using spawn threads (default: 32)
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
The current error message when the number of slots is insufficient
(e.g. running mpirun -n 4 on a dual core machine) does not mention the
use of `--oversubscribe`.
In earlier version of Open MPI, the over-subscription was automatic
(albeit buggy?); but the important point was no error message was
printed and the application runs. Mentioning the oversubscibe flag in
the message will ease up the transition to the current behaviour where
explicit request is required.
Also make a few other minor tweaks / cleanups to the
orte-rmaps-seq:alloc-error help message.
Signed-off-by: Yu Feng <rainwoodman@gmail.com>
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
if PMIx (version > 1.x) is active since all diagnostic messages will instead flow thru
the PMIx connection. Unfortunately, PMIx v1 does not support this
feature, but we can remove the stddiag support once PMIx v1 slides out
of the support window
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
since some tasks migth end up having /dev/null as their stdin,
simply avoid pipe creation and destruction for these tasks.
From a pragmatic and MPI point of view, and unless explicitly required
otherwise, all MPI tasks but (the first) one end up with /dev/null
as their stdin.
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
The following issues have been fixed for `mindist`:
- computing the job map on the backend nodes
- using slots count (`-host node1:<s1>,nodeN:<sN>`)
- fixed `dist:span` job mapping method
- fixed `oversubcribe` option with `-host`
Signed-off-by: Boris Karasev <karasev.b@gmail.com>
gcc 7.[1,2] (at least) fails to correctly parse the OSX 10.13 sys/syslog.h
header. As a results we need to potect syslog support in OPAL, PMIX and
ORTE.
Signed-off-by: George Bosilca <bosilca@icl.utk.edu>
Signed-off-by: Jeff Squyres <jsquyres@cisco.com>
Debounce "unreachable" notifications for tools when they disconnect
Enable the -x cmd line option for prun
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
(cherry picked from commit 0a5b36180a22959654461ac1303cec35313f8b4a)
Remove some build product. Tell PMIx that we don't need a new nspace generated when OMPI calls connect
Add missing Makefile
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Debugger daemons do not count against available slots. Clean up some leftover errors from the upgrade to HWLOC 2 in the mappers. Properly flag debugger jobs that come in via PMIx.
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Still in the "needs to be done" category:
* mapping/ranking/binding options aren't correctly supported
* if the DVM encounters some errors (e.g., not enough resources for the job), the resulting error is globally set and impacts any subsequent job submission
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
Unlike "orterun", "prun" is a PMIx-only program that discovers the DVM connection instead of requiring that we explicitly provide it. Only build "prun" if PMIx v2.x is available.
This gets the DVM working again, but still is showing problems for multiple executions. I'll detail those in a separate issue. Thus, the DVM should still be considered "broken".
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
to choose the first available non-socket provider.
modified: orte/mca/rml/ofi/rml_ofi_component.c
modified: orte/mca/rml/ofi/rml_ofi_send.c
Signed-off-by: Anandhi Jayakumar <anandhi.s.jayakumar@intel.com>
* Even if we are only launching one app context, we might call spawn
later and the remote groups might want their global rank information.
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
* Resolves#3705
* Components should link against the project level library to better
support `dlopen` with `RTLD_LOCAL`.
* Extend the `mca_FRAMEWORK_COMPONENT_la_LIBADD` in the `Makefile.am`
with the appropriate project level library:
```
MCA components in ompi/
$(top_builddir)/ompi/lib@OMPI_LIBMPI_NAME@.la
MCA components in orte/
$(top_builddir)/orte/lib@ORTE_LIB_PREFIX@open-rte.la
MCA components in opal/
$(top_builddir)/opal/lib@OPAL_LIB_PREFIX@open-pal.la
MCA components in oshmem/
$(top_builddir)/oshmem/liboshmem.la"
```
Note: The changes in this commit were automated by the script in
the commit that proceeds it with the `libadd_mca_comp_update.py`
script. Some components were not included in this change because
they are statically built only.
Signed-off-by: Joshua Hursey <jhursey@us.ibm.com>
There can be multiple [heap] consecutively in proc/<pid>/maps,
and there's no room between them.
Don't use a hole after the first [heap] is there's another [heap]
immediately after it.
This code would fail to find the last [heap] if there were multiple
[heap] interleaved with non-heap VMA, but our kind "after heap"
wouldn't be meaningful anymore anyway.
Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
always re-initialize vmhole *before* mca_base_component_var_register()
otherwise the vmhole gets NULL'ified if orte is initialized a second time.
that typically occurs when Open MPI is configure'd with --disable-dlopen
and the app does MPI_T_init_thread(); MPI_T_finalize(); MPI_T_init_thread();
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
It turns out that the approach of having the HNP do the
fork/exec of MPI ranks on the head node in a SLURM environment
introduces problems when users/sysadmins want to use the SLURM
scancl tool or sbatch --signal option to signal a job.
This commit disables use of the HNP fork/exec procedure when
a job is launched into a SLURM controlled allocation.
update NEWS with a blurb about new ras framework mca parameter.
related to #3998
Signed-off-by: Howard Pritchard <hppritcha@gmail.com>
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>
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>
`send_error()` is only packing status and peer info in the reply.
While remote counterpart in `pmix_server_dmdx_resp()` expects
the "hotel room number" to proceed correctly.
Signed-off-by: Artem Polyakov <artpol84@gmail.com>
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>
(1) rml_ofi_transports mca parameter. This parameter should have the list of transports (currently ethernet,fabric are valid)
fabric is higher priority if provided.
(2) ORTE_RML_TRANSPORT_TYPE key with values "ethernet" or "fabric". "fabric" is higher priority.
If specific provider is required use ORTE_RML_OFI_PROV_NAME key with values "socket" or "OPA" or any other supported in system.
modified: ../orte/mca/rml/ofi/rml_ofi.h
modified: ../orte/mca/rml/ofi/rml_ofi_component.c
modified: ../orte/mca/rml/ofi/rml_ofi_send.c
On send_msg choose the provider on local and peer to follow below rules -
1. if the user specified the transport for this conduit (even giving us a prioritized list of candidates), then the one we selected is the _only_ one we will use. If the remote peer has a matching endpoint, then we use it - otherwise, we error out
2. if the user didn't specify a transport, then we look for matches against _all_ of our available transports, starting with fabric and then going to Ethernet, taking the first one that matches.
3. if we can't find any match, then we error out
modified: ../orte/mca/rml/ofi/rml_ofi.h
modified: ../orte/mca/rml/ofi/rml_ofi_component.c
modified: ../orte/mca/rml/ofi/rml_ofi_send.c
send_msg() -> Fixed case when the local provider chosen at time of opening conduit
is not present in peer (destination) node
modified: ../orte/mca/rml/ofi/rml_ofi.h
modified: ../orte/mca/rml/ofi/rml_ofi_send.c
When opening conduit, checking for the transport preference in below order -
(1) rml_ofi_transports mca parameter. This parameter should have the list of transports (currently ethernet,fabric are valid)
fabric is higher priority if provided.
(2) ORTE_RML_TRANSPORT_TYPE key with values "ethernet" or "fabric". "fabric" is higher priority.
If specific provider is required use ORTE_RML_OFI_PROV_NAME key with values "socket" or "OPA" or any other supported in system.
modified: ../orte/mca/rml/ofi/rml_ofi.h
modified: ../orte/mca/rml/ofi/rml_ofi_component.c
modified: ../orte/mca/rml/ofi/rml_ofi_send.c
On send_msg choose the provider on local and peer to follow below rules -
1. if the user specified the transport for this conduit (even giving us a prioritized list of candidates), then the one we selected is the _only_ one we will use. If the remote peer has a matching endpoint, then we use it - otherwise, we error out
2. if the user didn't specify a transport, then we look for matches against _all_ of our available transports, starting with fabric and then going to Ethernet, taking the first one that matches.
3. if we can't find any match, then we error out
modified: ../orte/mca/rml/ofi/rml_ofi.h
modified: ../orte/mca/rml/ofi/rml_ofi_component.c
modified: ../orte/mca/rml/ofi/rml_ofi_send.c
send_msg() -> Fixed case when the local provider chosen at time of opening conduit
is not present in peer (destination) node
modified: ../orte/mca/rml/ofi/rml_ofi.h
modified: ../orte/mca/rml/ofi/rml_ofi_send.c
Signed-off-by: Anandhi Jayakumar <anandhi.s.jayakumar@intel.com>
Reference: https://bugzilla.kernel.org/show_bug.cgi?id=15272.
Work with both stdin/stdout fds that are known to be always
ready using libevent timers.
Such fds can not be effectively used with non-blocking I/O
functions like epoll, poll, select:
- for poll/select the event will be triggered immediately;
- for epoll `epoll_ctl` will reject an attempt to add this
fd to the working set.
Reference: http://www.wangafu.net/~nickm/libevent-book/Ref4_event.html
Libevent suggests to use timers over event_active for the
reasons provided by the link above.
Signed-off-by: Artem Polyakov <artpol84@gmail.com>
Regular files are always write-ready, so non-blocking I/O does not
give any benefits for them.
More than that - if libevent is using "epoll" to track fd events,
epoll_ctl will refuse attempt to add an fd pointing to a regular
file descriptor with EPERM.
This fix checks the object referenced by fd and avoids event_add
using event_active instead.
In the original configuration that uncovered this issue "epoll"
was used in libevent, it was triggering the following warning
message:
"[warn] Epoll ADD(1) on fd 0 failed. Old events were 0; read
change was 1 (add); write change was 0 (none): Operation not
permitted"
And the side effect was accumulation of all output in mpirun
memory and actually writing it only at mpirun exit.
Signed-off-by: Artem Polyakov <artpol84@gmail.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>
the 'hostname' command might not be available on some platforms
such as Fedora Core 26, so mimick config/libtool.m4 and fallback
to 'uname -n' if needed
Refs. #3680
Signed-off-by: Gilles Gouaillardet <gilles@rist.or.jp>
Cleanup the way we look for matching OFI addresses by using the opal_net_samenetwork helper function. This now works for multi-network environments, but only using the socket provider
Signed-off-by: Ralph Castain <rhc@open-mpi.org>
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>
