1
1
Граф коммитов

96 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
14d5525fb1 Some minor cleanups. Get singletons working. Cleanup abort handling so it gets properly identified.
This commit was SVN r26261.
2012-04-10 19:08:54 +00:00
Ralph Castain
bd8b4f7f1e Sorry for mid-day commit, but I had promised on the call to do this upon my return.
Roll in the ORTE state machine. Remove last traces of opal_sos. Remove UTK epoch code.

Please see the various emails about the state machine change for details. I'll send something out later with more info on the new arch.

This commit was SVN r26242.
2012-04-06 14:23:13 +00:00
Josh Hursey
4dd9f89a99 Create an MCA parameter (ess_base_stream_buffering) that allows the user to override the system default for buffering of stdout/stderr streams. See 'man setvbuf' for more information.
Note: I am working on a system that buffered all output until the application fishished due to a default of 'fully buffered.' This makes debugging painful. This switch fixed the problem by allowing me to adjust the buffering.

This commit was SVN r26119.
2012-03-08 22:02:28 +00:00
Ralph Castain
9b59d8de6f This is actually a much smaller commit than it appears at first glance - it just touches a lot of files. The --without-rte-support configuration option has never really been implemented completely. The option caused various objects not to be defined and conditionally compiled some base functions, but did nothing to prevent build of the component libraries. Unfortunately, since many of those components use objects covered by the option, it caused builds to break if those components were allowed to build.
Brian dealt with this in the past by creating platform files and using "no-build" to block the components. This was clunky, but acceptable when only one organization was using that option. However, that number has now expanded to at least two more locations.

Accordingly, make --without-rte-support actually work by adding appropriate configury to prevent components from building when they shouldn't. While doing so, remove two frameworks (db and rmcast) that are no longer used as ORCM comes to a close (besides, they belonged in ORCM now anyway). Do some minor cleanups along the way.

This commit was SVN r25497.
2011-11-22 21:24:35 +00:00
George Bosilca
61f273b987 Do not tolerate uninitialized variables.
This commit was SVN r25489.
2011-11-18 10:19:24 +00:00
Ralph Castain
b77552c45d Cleanup some include files, return a silent error in open/select as the complaining component already output a message
This commit was SVN r25416.
2011-11-02 17:42:06 +00:00
Ralph Castain
965b04d1a5 Use the new utilities to get a topology that reflects available cpus
This commit was SVN r25394.
2011-10-29 15:07:36 +00:00
Ralph Castain
b44f8d4b28 Complete implementation of the ess.proc_get_locality API. Up to this point, the API was only capable of telling if the specified proc was sharing a node with you. However, the returned value was capable of telling you much more detailed info - e.g., if the proc shares a socket, a cache, or numa node. We just didn't have the data to provide that detail.
Use hwloc to obtain the cpuset for each process during mpi_init, and share that info in the modex. As it arrives, use a new opal_hwloc_base utility function to parse the value against the local proc's cpuset and determine where they overlap. Cache the value in the pmap object as it may be referenced multiple times.

Thus, the return value from orte_ess.proc_get_locality is a 16-bit bitmask that describes the resources being shared with you. This bitmask can be tested using the macros in opal/mca/paffinity/paffinity.h

Locality is available for all procs, whether launched via mpirun or directly with an external launcher such as slurm or aprun.

This commit was SVN r25331.
2011-10-19 20:18:14 +00:00
Ralph Castain
3c4f04f4d9 Ensure opal_hwloc_topology is NULL after being destroyed
This commit was SVN r25138.
2011-09-13 19:21:10 +00:00
Ralph Castain
92c7372e20 Per the RFC from Jeff, move hwloc from opal/mca/common to its own static framework ala libevent. Have ORTE daemons collect the topology info at startup and, if --enable-hwloc-xml is set, send that info back to the HNP for later use. The HNP only retains unique topology "templates" to reduce memory footprint. Have the daemon include the local topology info in the nidmap buffer sent to each app so the apps don't all hammer the local system to discover it for themselves.
Remove the sysinfo framework as hwloc replaces that functionality.

This commit was SVN r25124.
2011-09-11 19:02:24 +00:00
Wesley Bland
4e7ff0bd5e By popular demand the epoch code is now disabled by default.
To enable the epochs and the resilient orte code, use the configure flag:

--enable-resilient-orte

This will define both:

ORTE_ENABLE_EPOCH
ORTE_RESIL_ORTE

This commit was SVN r25093.
2011-08-26 22:16:14 +00:00
Ralph Castain
1c08a4006c Refactor some code to remove a few API handles from errmgr. Reviewed/tested by Wes.
This commit was SVN r25064.
2011-08-18 16:24:45 +00:00
Wesley Bland
5fde3e0e00 Move the resilient orte errmgr code into a seperate errmgr for now while it's
still unstable. Reverted errmgr modules back to the original errmgr (with the
updates since the resilient code was brought into the trunk).

This commit was SVN r24958.
2011-07-28 21:24:34 +00:00
Ralph Castain
6c879f87fb Add a new param "orte_remote_tmpdir_base" for those situations where the compute nodes require a different session directory head than the head node.
This commit was SVN r24956.
2011-07-27 19:37:17 +00:00
Ralph Castain
361bcef253 Close multicast before rml
This commit was SVN r24925.
2011-07-23 20:19:15 +00:00
Ralph Castain
1ee7c39982 Fix some major bit-rot on scalable launch. If static ports are provided, then daemons can connect back to the HNP via the routed connection tree instead of doing so directly. In order to do that at scale, the node list must be passed as a regular expression - otherwise, the orted command line gets too long.
Over the course of time, usage of static ports got corrupted in several places, the "parent" info got incorrectly reset, etc. So correct all that and get the regex-based wireup going again.

Also, don't pass node lists if static ports aren't enabled - they are of no value to the orted and just create the possibility of overly-long cmd lines.

This commit was SVN r24860.
2011-07-07 18:54:30 +00:00
Wesley Bland
0628963506 Fix a return code when a process isn't found.
This commit was SVN r24845.
2011-06-30 15:22:54 +00:00
Brian Barrett
e52fef28ca Do something rational for the disable full support case
This commit was SVN r24844.
2011-06-30 14:48:19 +00:00
Wesley Bland
e1ba09ad51 Add a resilience to ORTE. Allows the runtime to continue after a process (or
ORTED) failure. Note that more work will be necessary to allow the MPI layer to
take advantage of this.

Per RFC:
http://www.open-mpi.org/community/lists/devel/2011/06/9299.php

This commit was SVN r24815.
2011-06-23 20:38:02 +00:00
Josh Hursey
0eb3b3b7b0 Fix missing functionality in MPI_Abort so that the group of peers defined by the communicator that should be aborted with this process are requested from the runtime before the local process exits.
Per RFC:
  http://www.open-mpi.org/community/lists/devel/2011/06/9335.php

This commit was SVN r24775.
2011-06-15 13:10:13 +00:00
Ralph Castain
9b38525d1e Remove unused include files
This commit was SVN r24394.
2011-02-16 00:32:47 +00:00
Ralph Castain
9ea2b196ce Convert the opal_event framework to use direct function calls instead of hiding functions behind function pointers. Eliminate the opal_object_t abstraction of libevent's event struct so it can be directly passed to the libevent functions.
Note: the ompi_check_libfca.m4 file had to be modified to avoid it stomping on global CPPFLAGS and the like. The file was also relocated to the ompi/config directory as it pertains solely to an ompi-layer component.

Forgive the mid-day configure change, but I know Shiqing is working the windows issues and don't want to cause him unnecessary redo work.

This commit was SVN r23966.
2010-10-28 15:22:46 +00:00
Ralph Castain
86c7365e8e Clean up a few initialization issues - don't think these are impacting the shared memory situation as it didn't fix the problem.
Setup the event API to support multiple bases in preparation for splitting the OMPI and ORTE events. Holding here pending shared memory resolution.

This commit was SVN r23943.
2010-10-26 02:41:42 +00:00
Ralph Castain
fceabb2498 Update libevent to the 2.0 series, currently at 2.0.7rc. We will update to their final release when it becomes available. Currently known errors exist in unused portions of the libevent code. This revision passes the IBM test suite on a Linux machine and on a standalone Mac.
This is a fairly intrusive change, but outside of the moving of opal/event to opal/mca/event, the only changes involved (a) changing all calls to opal_event functions to reflect the new framework instead, and (b) ensuring that all opal_event_t objects are properly constructed since they are now true opal_objects.

Note: Shiqing has just returned from vacation and has not yet had a chance to complete the Windows integration. Thus, this commit almost certainly breaks Windows support on the trunk. However, I want this to have a chance to soak for as long as possible before I become less available a week from today (going to be at a class for 5 days, and thus will only be sparingly available) so we can find and fix any problems.

Biggest change is moving the libevent code from opal/event to a new opal/mca/event framework. This was done to make it much easier to update libevent in the future. New versions can be inserted as a new component and tested in parallel with the current version until validated, then we can remove the earlier version if we so choose. This is a statically built framework ala installdirs, so only one component will build at a time. There is no selection logic - the sole compiled component simply loads its function pointers into the opal_event struct.

I have gone thru the code base and converted all the libevent calls I could find. However, I cannot compile nor test every environment. It is therefore quite likely that errors remain in the system. Please keep an eye open for two things:

1. compile-time errors: these will be obvious as calls to the old functions (e.g., opal_evtimer_new) must be replaced by the new framework APIs (e.g., opal_event.evtimer_new)

2. run-time errors: these will likely show up as segfaults due to missing constructors on opal_event_t objects. It appears that it became a typical practice for people to "init" an opal_event_t by simply using memset to zero it out. This will no longer work - you must either OBJ_NEW or OBJ_CONSTRUCT an opal_event_t. I tried to catch these cases, but may have missed some. Believe me, you'll know when you hit it.

There is also the issue of the new libevent "no recursion" behavior. As I described on a recent email, we will have to discuss this and figure out what, if anything, we need to do.

This commit was SVN r23925.
2010-10-24 18:35:54 +00:00
Ralph Castain
12cd07c9a9 Start reducing our dependency on the event library by removing at least one instance where we use it to redirect the program counter. Rolf reported occasional hangs of mpirun in very specific circumstances after all daemons were done. A review of MTT results indicates this may have been happening more generally in a small fraction of cases.
The problem was tracked to use of the grpcomm.onesided_barrier to control daemon/mpirun termination. This relied on messaging -and- required that the program counter jump from the errmgr back to grpcomm. On rare occasions, this jump did not occur, causing mpirun to hang.

This patch looks more invasive than it is - most of the affected files simply had one or two lines removed. The essence of the change is:

* pulled the job_complete and quit routines out of orterun and orted_main and put them in a common place

* modified the errmgr to directly call the new routines when termination is detected

* removed the grpcomm.onesided_barrier and its associated RML tag

* add a new "num_routes" API to the routed framework that reports back the number of dependent routes. When route_lost is called, the daemon's list of "children" is checked and adjusted if that route went to a "leaf" in the routing tree

* use connection termination between daemons to track rollup of the daemon tree. Daemons and HNP now terminate once num_routes returns zero

Also picked up in this commit is the addition of a new bool flag to the app_context struct, and increasing the job_control field from 8 to 16 bits. Both trivial.

This commit was SVN r23429.
2010-07-17 21:03:27 +00:00
Abhishek Kulkarni
afbe3e99c6 * Wrap all the direct error-code checks of the form (OMPI_ERR_* == ret) with
(OMPI_ERR_* = OPAL_SOS_GET_ERR_CODE(ret)), since the return value could be a
 SOS-encoded error. The OPAL_SOS_GET_ERR_CODE() takes in a SOS error and returns
 back the native error code.

* Since OPAL_SUCCESS is preserved by SOS, also change all calls of the form
  (OPAL_ERROR == ret) to (OPAL_SUCCESS != ret). We thus avoid having to
  decode 'ret' to get the native error code.

This commit was SVN r23162.
2010-05-17 23:08:56 +00:00
Ralph Castain
4bd25f587c Begin handling the case of lost connections by having the OOB report it to the errmgr instead of the routed framework. Add an "app" component to t
he errmgr framework so that it can decide how to respond - which for now at least is just to check for lifeline and abort if so.

Add a new error constant to indicate that the error is "unrecoverable" so the oob can know it needs to abort.

This commit was SVN r23112.
2010-05-11 00:34:12 +00:00
Ralph Castain
2ff1ae13e1 Create a new "heartbeat" module in the sensor framework and move the plm_base heartbeat code there. Add new proc and job states for heartbeat_failed. Remove the "heartbeat" cmd line option for orted as this is now done automatically if the --enable-heartbeat configure option is set.
This commit was SVN r23102.
2010-05-05 00:48:43 +00:00
Ralph Castain
9dfb5c7c62 Rename the orte state framework to be "db", which more accurately reflects its overall capabilities since it can store any kind of data (not just state, although that will be its primary purpose). Update tools and tests accordingly. Add a daemon module for storing data on the daemons - requires --enable-multicast, so it won't build unless that is set
This commit was SVN r23082.
2010-05-03 04:11:03 +00:00
Ralph Castain
55889934d8 After hours spent chasing the stupid "abort" file, it became clear that we were always going to be plagued by that idiot contraption when trying to be good citizens and properly cleanup. So get rid of it by instead doing a messaging handshake with the local daemon.
Note that this isn't a problem since MPI_Abort and orte_abort are only called under controlled circumstances - i.e., we are doing an orderly abort and not segfaulting. If we can't get the message out for some reason, then too bad - we'll still see an abnormal process termination and act accordingly.

This commit was SVN r23045.
2010-04-27 03:39:32 +00:00
Ralph Castain
b9893aacc5 Add a sensor framework to ORTE that monitors applications and notifies the errmgr when they exceed specified boundaries. Two modules are included here:
1. file activity - can monitor file size, access and modification times. If these fail to change over a specified number of sampling iterations (rate is an mca param), then the errmgr is notified.

2. memory usage - checks amount of memory used by a process. Limit and sampling rate can be set.

This support must be enabled by configuring --enable-sensors.

ompi_info and orte-info have been updated to include the new framework.

Also includes some initial steps toward restoring the recovery capability. Most notably, the ODLS API has been extended to include a "restart_proc" entry for restarting a local process, and organizes the various ERRMGR framework globals into a single struct as we do in the other ORTE frameworks. Fix an oversight in the ERRMGR framework where a pointer array was constructed, but not initialized.

Implementation continues.

This commit was SVN r23043.
2010-04-26 22:15:57 +00:00
Ralph Castain
efbb5c9b7c Revamp the errmgr framework to provide a greater range of optional behaviors, including different behaviors for daemons, and remove several looping messages across the code base:
* add hnp and orted modules to the errmgr framework. The HNP module contains much of the code that was in the errmgr base since that code could only be executed by the HNP anyway.

* update the odls to report process states directly into the active errmgr module, thus removing the need to send messages looped back into the odls cmd processor. Let the active errmgr module decide what to do at various states.

* remove the code to track application state progress from the plm_base_launch_support.c code. Update the plm modules to call the errmgr directly when a launch fails.

* update the plm_base_receive.c code to call the errmgr with state updates from remote daemons

* update the routed modules to reflect that process state is updated in the errmgr

* ensure that the orted's open the errmgr and select their appropriate module

* add new pretty-print utilities to print process and job state. Move the pretty-print of time info to a globally-accessible place

* define a global orte_comm function to send messages from orted's to the HNP so that others can overlay the standard RML methods, if desired.

* update the orterun help output to reflect that the "term w/o sync" error message can result from three, not two, scenarios

This commit was SVN r23023.
2010-04-23 04:44:41 +00:00
Ralph Castain
4308922f59 Ensure that any application-specific selection of ess module doesn't get overridden by what is given to the orted or orterun
Cleanup tool name determination for CM

This commit was SVN r22980.
2010-04-15 18:10:50 +00:00
Ralph Castain
0b9552cd4e Expand the ESS framework's API to include a new function "query_sys_info" that allows the caller to retrieve key-value pairs of info on the local system capabilities (e.g., cpu type/model). Have each daemon and the HNP "sense" that information and provide it to their local procs to avoid having every proc querying the system directly.
This commit was SVN r22870.
2010-03-23 20:47:41 +00:00
Ralph Castain
ffd5be6aa1 Add a new framework to ORTE for saving and recovering state information. Two components are included that use the db or dbm library for storing the data, with a distributed hash table component coming later.
Note that each of these components will only be selected if specifically requested - otherwise, a "NULL" component will be used.  The framework is only opened by the HNP and orteds, though neither is currently coded to save/restore state

This commit was SVN r22839.
2010-03-16 20:59:48 +00:00
Josh Hursey
e9b5162d79 Fix the configure logic for --with-ft so that it properly takes a comma separated list.
Many of the OPAL_ENABLE_FT should be OPAL_ENABLE_FT_CR, so fix those.

The OPAL Layer INC should call opal_output on restart so that it can refresh the string it prints to reflect the current pid/hostname which may have changed.

This commit was SVN r22824.
2010-03-12 23:57:50 +00:00
Ralph Castain
c88fe1ea54 Create a new mca parameter to control creation of session directories. Defaults to true so that the current behavior of always creating them is preserved. If set to false (0), then don't create session directories. Helps in those environments where session directories are a problem.
Tell the sm btl that it cannot run if no session directories were created.

This commit was SVN r22756.
2010-03-02 15:18:33 +00:00
Ralph Castain
18c7aaff08 Update the grpcomm framework to be more thread-friendly.
Modify the orte configure options to specify --enable-multicast such that it directs components to build or not instead of littering the code base with #if's. Remove those #if's where they used to occur.

Add a new grpcomm "mcast" module to support multicast operations. Still some work required to properly perform daemon collectives for comm_spawn operations. New module only builds when --enable-multicast is provided, and when specifically selected.

This commit was SVN r22709.
2010-02-25 01:11:29 +00:00
Ralph Castain
f66b6cae23 Enable the boot of an orted "virtual machine". Modify the mapper framework to allow mapping of only daemons. Remove the cm ras module as no longer required. Modify the orted code to always send back node arch info. Remove the "--enable-bootstrap" configure option as this feature will now always be available.
This commit was SVN r22480.
2010-01-25 22:25:13 +00:00
Ralph Castain
cec840f6b9 The ability to add procs to a running job was unfortunately borked when we added the detection of a proc exiting before calling init. Re-enable it here, ensuring that procs that are being restarted and/or added to a job do -not- call barrier during orte_init.
This commit was SVN r22404.
2010-01-14 17:59:42 +00:00
Ralph Castain
9f3ccebeaa We need to barrier for orte apps when the job is initially started, but we must not do the barrier when a proc is restarted as the other procs in the job won't know to participate.
This commit was SVN r22388.
2010-01-10 02:21:30 +00:00
Ralph Castain
09763ec711 Since we modified ORTE to declare that any process that terminates after calling "init" while at least one other process has not yet called "init" is an error, we have to ensure that non-MPI ORTE apps (i.e., apps that call orte_init but not mpi_init) include a barrier in orte_init. Otherwise, fast ORTE apps almost always wind up triggering the "abnormal termination" condition.
The barrier is protected with a test to ensure that MPI apps don't execute it and wind up doing two barriers during their init.

This commit was SVN r22378.
2010-01-07 06:58:01 +00:00
Ralph Castain
9acec283af Add a new TCP module to the reliable multicast framework. This module uses ORTE's grpcomm.xcast functionality to "fake" multicasts for environments where regular multicast isn't reliable.
Modify the startup logic to allow for this use-case.

This commit was SVN r22310.
2009-12-15 01:18:27 +00:00
Ralph Castain
a0d5c80ce0 Add a new framework for discovering local resource information such as cpu type/model, #cpus, available physical memory, etc. Two initial components (darwin and linux) are provided. This is needed to support bootstrap operations where daemons are started at node boot, and applications where initial knowledge of cpu identification is needed to guide framework component selection.
Add orte configuration option to control the use of the framework in the system. Although the code will build, it will not be active unless configured with --enable-bootstrap.

If bootstrap is enabled and the new opal_sysinfo framework can successfully determine the cpu model, pass that info to the application as an MCA param to support some work at Sun.

Also, have daemons report back the resources they find to guide process mapping in bootstrap operations (i.e., where the daemon starts at node boot as opposed to being launched at application start).

Adjust some platform files to enable these capabilities.

This commit was SVN r22244.
2009-11-30 23:11:25 +00:00
Ralph Castain
e38a0eab9f Remove the fddp and sensor frameworks - relocated to new cluster mgr project
This commit was SVN r22240.
2009-11-27 22:14:47 +00:00
Ralph Castain
709b36efb4 Cleanup auto-wireup and enable tools to "discover" the HNP via multicast
This commit was SVN r22012.
2009-09-25 01:00:09 +00:00
Josh Hursey
5406fdfb80 Add support for sending SIGSTOP the MPI job after the checkpoint is taken (uses a BLCR feature for the option).
This commit looks larger than it really is since it includes a fair amount of code cleanup.

The SIGSTOP/SIGCONT+checkpointing work uses some of the functionality in r20391. Basic use case below (note that the checkpoint generated is useable as usual if the stopped application is terminated).
{{{
shell 1) mpirun -np 2 -am ft-enable-cr my-app
... running ...

shell 2) ompi-checkpoint --stop -v MPIRUN_PID
[localhost:001300] [  0.00 /   0.20]                 Requested - ...
[localhost:001300] [  0.00 /   0.20]                   Pending - ...
[localhost:001300] [  0.01 /   0.21]                   Running - ...
[localhost:001300] [  1.01 /   1.22]                   Stopped - ompi_global_snapshot_1234.ckpt
Snapshot Ref.: 0 ompi_global_snapshot_1234.ckpt

shell 2) killall -CONT mpirun

... Application Continues execution in shell 1 ...
}}}

Other items in this commit are mostly cleanup that has been sitting off-trunk for too long:
 * Add a new {{{opal_crs_base_ckpt_options_t}}} type that encapsulates the various options that could be passed to the CRS. Currently only TERM and STOP, but this makes adding others ''much'' easier.
 * Eliminate ORTE_SNAPC_CKPT_STATE_PENDING_TERM, since it served a redundant purpose with the new options type.
 * Lay some basic ground work for some future features.

This commit was SVN r21995.

The following SVN revision numbers were found above:
  r20391 --> open-mpi/ompi@0704b98668
2009-09-22 18:26:12 +00:00
Ralph Castain
8da3aa8d5c Some (hopefully final!) adjustments and corrections to the paffinity support:
1. default -npersocket to force -bind-to-socket

2. if we cannot get a value for cores/socket, try using #logical cpus. otherwise, default to 1 core

3. add missing error message for not-enough-processors

4. since we no longer loop through orte_register_params twice, put the auto-detect of
   topology info in the rte_init for hnp and std_orted

5. fix bind-to-core, bysocket combination

This commit was SVN r21992.
2009-09-22 15:41:03 +00:00
Ralph Castain
c3f9096fd9 Add a reliable multicast framework, with an initial basic module. This is configured out unless specifically requested via --enable-multicast.
This commit was SVN r21988.
2009-09-22 00:58:29 +00:00
Ralph Castain
ae31af7dec Enable monitoring if configured to do so. Update the sensor framework
This commit was SVN r21964.
2009-09-09 21:00:27 +00:00