ports/openmpi
ports/openmpi
1
1
Форкнуть 0
Код Релизы Активность
18 043 коммитов 3 Ветки 0 Теги 130 MiB
21ee48de57
Граф коммитов

2938 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
30fd9d7abc MPI procs should definitely not be trapping SIGCHLD - only ORTE tools need to do so 
This commit was SVN r27166.
 2012-08-28 21:39:06 +00:00
Ralph Castain
98580c117b Introduce staged execution. If you don't have adequate resources to run everything without oversubscribing, don't want to oversubscribe, and aren't using MPI, then staged execution lets you (a) run as many procs as there are available resources, and (b) start additional procs as others complete and free up resources. Adds a new mapper as well as a new state machine. 
Remove some stale configure.m4's we no longer need.

Optimize the nidmaps a bit by only sending info that has changed each time, instead of sending a complete copy of everything. Makes no difference for the typical MPI job - only impacts things like staged execution where we are sending multiple (possibly many) launch messages.

This commit was SVN r27165.
 2012-08-28 21:20:17 +00:00
Ralph Castain
aadfe1b61e Fix a missing test that breaks novm operation. 
CMR:v1.7

This commit was SVN r27163.
 2012-08-28 21:13:57 +00:00
Ralph Castain
d310dd8c58 Fix a strange race condition by creating a separate buffer for each send - apparently, just a retain isn't enough protection on some systems 
This commit was SVN r27161.
 2012-08-28 17:17:34 +00:00
Ralph Castain
11c68e2299 Correct the count in the pmi key 
This commit was SVN r27156.
 2012-08-28 15:05:02 +00:00
Ralph Castain
6e8c97c77c Per Sam's eagle-eyed review, free the malloc'd memory if getcwd fails for some strange reason. 
This commit was SVN r27150.
 2012-08-27 19:15:16 +00:00
Ralph Castain
bccc20d13e Deal with one last corner case of positioning a dot-file 
This commit was SVN r27144.
 2012-08-26 03:49:31 +00:00
Ralph Castain
63d41c643d Minor cleanup 
This commit was SVN r27143.
 2012-08-25 14:24:45 +00:00
Ralph Castain
05f0b4c653 Couple of minor cleanups 
This commit was SVN r27135.
 2012-08-24 21:14:40 +00:00
Ralph Castain
d6cbff6d4e Since the preload flags are at the app_context level, we need to link only those files/exe's that pertain to each app_context to the corresponding procs. Also, gain a little optimization by checking to ensure we only send files once - this probably won't work when daemons are created on-the-fly, but that's for some other day 
This commit was SVN r27134.
 2012-08-24 16:16:30 +00:00
Jeff Squyres
20612c4194 Don't close the IOF stdin if we happen to read less than a full 
buffer's worth of data -- interactive stdin will have that behavior
frequently. 

This commit was SVN r27131.
 2012-08-24 14:29:19 +00:00
Ralph Castain
e0c39c94e8 Complete the cleanup of the preload files system. Remove the dest_dir option as moving things to arbitrary locations - especially absolute paths - can prove disastrous. Remove the preload_libs option as these can be treated as just files. Cleanup some of the pack/unpack code as the dss handles NULL strings just fine. Deal a little better with absolute paths, noting that tar now strips the leading '/' for us (showing my age as it didn't used to do so). 
Remove the odls_base_state.c file as that code is now covered by the new broadcast form of preload_files.

This commit was SVN r27127.
 2012-08-24 02:28:29 +00:00
Ralph Castain
b4a544ad2a Per discussion with Josh, use the --preload-xxx cmd line options to broadcast files to all nodes. Add --set-cwd-to-session-dir option to start procs in their session directories. Add OMPI_FILE_LOCATION envar to tell procs where their prepositioned files went. 
This commit was SVN r27125.
 2012-08-23 21:28:05 +00:00
Ralph Castain
855c9ae6cf Support archives .tar, .bz[2,zip], and .gz[ip] 
This commit was SVN r27123.
 2012-08-23 15:38:39 +00:00
Ralph Castain
286c610712 Protect us against the scenario where filem is included in enable-mca-no-build 
This commit was SVN r27122.
 2012-08-23 13:52:06 +00:00
Shiqing Fan
d141d94bd7 Include the new .windows files into the tarball. 
This commit was SVN r27121.
 2012-08-23 12:50:51 +00:00
Ralph Castain
7237a938bf Extend the filem interface to support prepositioning and linking required local files for execution. Create a new "raw" module that uses xcast to send the files to all nodes as this is faster than doing an scp in a linear pattern 
This commit was SVN r27118.
 2012-08-22 21:43:20 +00:00
Ralph Castain
ed4b354846 Ensure we pass along user-specified mca params from the cmd line when doing a tree spawn, but don't extend the cmd line with duplicates or things that shouldn't be there 
This commit was SVN r27117.
 2012-08-22 21:41:50 +00:00
Ralph Castain
5d7872fd68 Cleanup the tag list 
This commit was SVN r27115.
 2012-08-22 21:37:58 +00:00
Ralph Castain
7bcf2f8b5c Stop leaving droppings behind us 
This commit was SVN r27111.
 2012-08-22 17:39:22 +00:00
Shiqing Fan
95b9552546 include several components for Windows build. 
This commit was SVN r27108.
 2012-08-22 14:46:49 +00:00
Jeff Squyres
c8cee23ee7 Priorities really shouldn't be less than 0. 
This commit was SVN r27098.
 2012-08-21 15:47:15 +00:00
Ralph Castain
dacb07000d Turn udcm and ud oob off by default, but allow them to build and be used if someone wants to test them 
cmr:v1.7

This commit was SVN r27097.
 2012-08-21 15:18:34 +00:00
Ralph Castain
64cf75cec5 Add some debug 
This commit was SVN r27087.
 2012-08-17 02:19:26 +00:00
Ralph Castain
35fef87202 Make the "no virtual machine" selection more intuitive by providing a --novm option to mpirun. 
This commit was SVN r27048.
 2012-08-15 14:55:03 +00:00
Nathan Hjelm
8e03f77004 update alps configure scripts 
This commit was SVN r27026.
 2012-08-13 22:57:55 +00:00
Ralph Castain
589acf550c Improve the new MPI_INFO_ENV to better handle Java applications and to correctly report the info for singletons. 
This commit was SVN r27025.
 2012-08-13 22:13:49 +00:00
Ralph Castain
49a757e0bd Silly me - now that all daemons are stripping their prefix on the backend, we no longer need to do it as they report 
This commit was SVN r27023.
 2012-08-13 20:48:13 +00:00
Ralph Castain
b9b41d8662 For cases where the alpha+non-zero prefix must be removed from a node name, be sure to do it everywhere we access node names - otherwise, modex methods such as pmi will fail to correctly identify procs on the same node 
This commit was SVN r27022.
 2012-08-13 20:44:56 +00:00
Ralph Castain
cb48fd52d4 Implement the MPI_Info part of MPI-3 Ticket 313. Add an MPI_info object MPI_INFO_GET_ENV that contains a number of run-time related pieces of info. This includes all the required ones in the ticket, plus a few that specifically address recent user questions: 
"num_app_ctx" - the number of app_contexts in the job
"first_rank" - the MPI rank of the first process in each app_context
"np" - the number of procs in each app_context

Still need clarification on the MPI_Init portion of the ticket. Specifically, does the ticket call for returning an error is someone calls MPI_Init more than once in a program? We set a flag to tell us that we have been initialized, but currently never check it.

This commit was SVN r27005.
 2012-08-12 01:28:23 +00:00
Ralph Castain
c4ee297a60 Cleanup the pmi grpcomm module so it passes non-btl modex data correctly. 
This commit was SVN r26992.
 2012-08-10 20:35:50 +00:00
Ralph Castain
e3e9b7345d First cut at updating the ccp launcher to use the state machine 
This commit was SVN r26986.
 2012-08-10 17:09:33 +00:00
Shiqing Fan
e304c19920 This is also used on Windows. 
This commit was SVN r26975.
 2012-08-08 16:44:00 +00:00
George Bosilca
ba879c2c51 Remove the unused map. 
This commit was SVN r26960.
 2012-08-07 12:06:13 +00:00
Shiqing Fan
2f442799f8 fix several typecasts 
This commit was SVN r26957.
 2012-08-07 10:41:53 +00:00
Ralph Castain
53b1a1c976 Cleanly error out when someone asks to map-to <object> if that object doesn't exist on a node. 
This commit was SVN r26950.
 2012-08-04 21:52:36 +00:00
Ralph Castain
61b09a132b Fix bynode mapping of multiple app-contexts 
This commit was SVN r26949.
 2012-08-03 21:45:40 +00:00
Ralph Castain
96f6f94c24 Ensure we don't get trapped in an infinite loop when ranking bynode if something isn't right 
This commit was SVN r26948.
 2012-08-03 21:45:10 +00:00
Ralph Castain
0d878937fe If a callback is set in the state machine, and the state doesn't yet exist, create it 
This commit was SVN r26947.
 2012-08-03 21:43:36 +00:00
Ralph Castain
431d5361ed For those who really preferred our prior mode of operation that mapped procs and only launched daemons on the nodes that had procs on them, introduce the "novm" state machine component. This recreates the old mode of operation by re-ordering the launch sequence so that we allocate, then map, and then launch daemons only on the reqd nodes (instead of across the entire allocation). 
This commit was SVN r26946.
 2012-08-03 16:30:05 +00:00
Ralph Castain
dc22ea5cde A little cleaner on the message about repeated ctrl-c, and re-enable the event so we can abort if we see multiple ctrl-c's that don't meet the time requirement 
This commit was SVN r26945.
 2012-08-03 01:26:18 +00:00
Ralph Castain
e6c72bfd53 Ensure we can forcibly exit even when we are stuck inside of an event by replacing the libevent signal handler with a POSIX one that (a) attempts to trip a libevent termination event and (b) if anothe ctrl-c hits within 5 seconds, just calls exit. 
This commit was SVN r26943.
 2012-08-02 21:15:35 +00:00
Ralph Castain
d818c9d407 Includes a patch from Jeff and Josh: update the simulator module to allow specification of multiple slot and max_slot counts for each node group (but don't require it). Remove the requirement that each node group provide its own topology. Adjust verbosities to allow showing some light debug output to see what nodes have been added without getting a bunch of other stuff. 
This commit was SVN r26936.
 2012-08-02 04:57:13 +00:00
Jeff Squyres
62c2ff7ee7 It's actually ''not'' an error to exit if all routes and children are 
gone.  So exit with 0, not ORTE_ERROR_DEFAULT_EXIT_CODE (which is 1).

This fixes a race condition in the rsh launcher upon termination,
where ORTE would sometimes think that a daemon failed to launch.

This commit was SVN r26935.
 2012-08-01 19:49:19 +00:00
Nathan Hjelm
4557e15c18 oob/ud fix compile error 
This commit was SVN r26933.
 2012-07-31 21:50:34 +00:00
Ralph Castain
6ee35e4977 Add num_local_peers to orte_process_info so we don't keep re-computing it, ensure it is available for direct launch via pmi as well 
This commit was SVN r26931.
 2012-07-31 21:21:50 +00:00
Jeff Squyres
88cbe9c780 .ompi_ignore this component until it can be fixed. 
This commit was SVN r26930.
 2012-07-31 21:02:06 +00:00
Nathan Hjelm
980692804d oob/ud: don't start listening for ud requests unless we have one usable port 
This commit was SVN r26929.
 2012-07-31 19:00:18 +00:00
Ralph Castain
23c2a315a9 Add missing line to set flag indicating at least one port found 
This commit was SVN r26914.
 2012-07-30 17:54:38 +00:00
Ralph Castain
6285f7d8c0 Per request of Shiqing, restore the ccp components 
This commit was SVN r26904.
 2012-07-29 23:49:59 +00:00
Ralph Castain
94d11e04fd Add an intermediate state when the VM is ready so that third party tools can take action prior to mapping/launching apps 
This commit was SVN r26902.
 2012-07-28 15:33:09 +00:00
Ralph Castain
8bc6694a62 Ensure the daemons don't incorrectly declare a failed launch 
This commit was SVN r26875.
 2012-07-26 19:05:06 +00:00
Ralph Castain
07846f12ae Reconnect the rsh/ssh error reporting code for remote spawns to report failure to launch. Ensure the HNP correctly reports non-zero exit status when ssh encounters a problem. 
Thanks to Terry for spotting it!

This commit was SVN r26868.
 2012-07-25 21:46:45 +00:00
Jeff Squyres
e5cfad0c1a This variable is only used in FT builds. 
This commit was SVN r26854.
 2012-07-24 12:48:47 +00:00
Shiqing Fan
12d99a9ebb Update the hwloc build on Windows and related files. 
This commit was SVN r26818.
 2012-07-20 12:14:28 +00:00
Abhishek Kulkarni
1ce378b5c6 Make C/R work with nodes > 1. This fix makes sure that the app coordinators send 
the "ready-to-checkpoint" signal to the global coordinator only after ORTE has
initialized.

This commit was SVN r26795.
 2012-07-13 23:37:29 +00:00
Abhishek Kulkarni
1878f276cd Replace the pattern while(flag) { opal_progress() }; in the C/R code 
with the ORTE_WAIT_FOR_COMPLETION macro.

This commit was SVN r26794.
 2012-07-13 23:31:56 +00:00
George Bosilca
772ec212eb Fix another compiler warning. 
This commit was SVN r26775.
 2012-07-10 15:57:42 +00:00
Abhishek Kulkarni
eec5a28aa4 More C/R fixes. 
* Fix a typo introduced by the removal of the notifier framework
* Fix to flush the modex cached data correctly using the orte DB API.

This commit was SVN r26773.
 2012-07-10 01:19:46 +00:00
Abhishek Kulkarni
5c58a1c9c1 Fix C/R support in the trunk. 
Among other things, this patch deals with the following issues:
* fix ompi-checkpoint argument parsing
* ompi-restart -showme prints an extraneous "Restarted child with PID" 
  message. Move around the debug statement to avoid this.
* fixes for the state machine changes

This commit was SVN r26770.
 2012-07-09 23:34:13 +00:00
George Bosilca
ec760454a6 Cleaning ... 
This commit was SVN r26747.
 2012-07-04 21:22:13 +00:00
Ralph Castain
6ae5776904 Cleanup IPV6 build 
This commit was SVN r26738.
 2012-07-04 00:03:50 +00:00
Ralph Castain
1a90471374 Drat - missed the other one 
This commit was SVN r26718.
 2012-07-02 22:18:31 +00:00
Ralph Castain
9a6a969f60 Remove debug 
This commit was SVN r26717.
 2012-07-02 22:18:08 +00:00
Ralph Castain
b83fc41d54 Add a state that allows mpirun or other tools to be notified of a job completion prior to terminating so that alternative actions can be performed. 
This commit was SVN r26716.
 2012-07-02 22:16:32 +00:00
Ralph Castain
8bebf2fa47 Ensure we don't build the MR iof components unless hadoop support is enabled 
This commit was SVN r26694.
 2012-06-28 18:20:15 +00:00
Ralph Castain
9aa821d8b4 Add missing file to tarball 
This commit was SVN r26688.
 2012-06-28 02:57:10 +00:00
Ralph Castain
0dfe29b1a6 Roll in the rest of the modex change. Eliminate all non-modex API access of RTE info from the MPI layer - in some cases, the info was already present (either in the ompi_proc_t or in the orte_process_info struct) and no call was necessary. This removes all calls to orte_ess from the MPI layer. Calls to orte_grpcomm remain required. 
Update all the orte ess components to remove their associated APIs for retrieving proc data. Update the grpcomm API to reflect transfer of set/get modex info to the db framework.

Note that this doesn't recreate the old GPR. This is strictly a local db storage that may (at some point) obtain any missing data from the local daemon as part of an async methodology. The framework allows us to experiment with such methods without perturbing the default one.

This commit was SVN r26678.
 2012-06-27 14:53:55 +00:00
Brian Barrett
b22faedd9d Remove the Portals4 SHMEM reference implementation runtime support, as we're 
no longer using the runtime provided by the reference implementation.

Remove the Catamount support from ORTE, since we're no longer supporting
Catamount.  Left the Catamount timer component, because I'm not sure whether
it's used on the XTs running CNL.

This commit was SVN r26677.
 2012-06-27 14:17:43 +00:00
Josh Hursey
28681deffa Backout the ORCA commit. :( 
There is a linking issue on Mac OSX that needs to be addressed before this is able to come back into the trunk.

This commit was SVN r26676.
 2012-06-27 01:28:28 +00:00
Josh Hursey
542330e3a7 Commit of ORCA: Open MPI Runtime Collaborative Abstraction 
This is a runtime interposition project that sits between the OMPI and ORTE layers in Open MPI.

The project is described on the wiki:
  https://svn.open-mpi.org/trac/ompi/wiki/Runtime_Interposition

And on this email thread:
  http://www.open-mpi.org/community/lists/devel/2012/06/11109.php

This commit was SVN r26670.
 2012-06-26 21:42:16 +00:00
Ralph Castain
a34f09e67a Ensure common port is off when not being used 
This commit was SVN r26666.
 2012-06-26 16:09:58 +00:00
Ralph Castain
92527da4e3 Remove unused component 
This commit was SVN r26660.
 2012-06-26 00:49:28 +00:00
Ralph Castain
0103f82918 Turn off the common port for slurm for now 
This commit was SVN r26656.
 2012-06-25 21:55:51 +00:00
Shiqing Fan
6f746cdb33 remove a unused file. 
This commit was SVN r26645.
 2012-06-25 10:17:21 +00:00
Jeff Squyres
148ae6d6e3 This commit unifies the configury of some verbs-lovin' components. 
* Add new configure command line options and deprecate some old ones:
   * --with-verbs replaces --with-openib
   * --with-verbs-libdir replaces --with-openib-libdir
 * If you specify --with-openib[-libdir] without
   --with-verbs[-libdir], you'll get a "these options have been
   deprecated!" warning, but then they'll act just like
   --with-verbs[--libdir]. 

  '''Sidenote:''' Note that we are not renaming any components at this
  time, nor are we renaming the top-level OMPI_CHECK_OPENIB m4 macro
  (which is pretty strongly tied to the openib BTL and is bastaridzed
  by the ofud BTL).  Note that there will likely be more changes in
  this area coming soon (next week?) when some long-standing changes
  move to the SVN trunk: some openib BTL infrastructure will move to
  ompi/mca/common, and its configury gets split up / refactored.

We extend our philosophy of other --with-<foo> configure options of
--with-verbs to ''all'' verbs-lovin components:

 * If you specify --with-verbs, then all verbs-lovin' components must
   configure successfully (or abort).  This currently means: OOB ud,
   BTL ofud, BTL openib.
 * If you specify --with-verbs=DIR, then all verbs-lovin' component
   must configure successfully (or abort), and will use DIR to find
   verbs headers and libraries.
 * If you specify --without-verbs, then all verbs-lovin' components
   will be ignored.

This commit also fixes a problem where the --with-openib=DIR form
would not use DIR for ''all'' verbs-lovin' components (I think only
BTL openib and BTL ofud used that DIR).  Now all of them do, as does
hwloc (because hwloc has some !OpenFabrics helper functions that
require ibv types from verbs.h).

There's a little new m4 infrastructure worth mentioning:

 * If you create a new verbs-lovin' component (i.e., a component that
   need verbs), your configure.m4 should
   AC_REQUIRE([OPAL_CHECK_VERBS_DIR]). 
 * You can then use three global shell variables: $opal_want_verbs,
   $opal_verbs_dir, $opal_verbs_libdir, which will be set as follows:
   * opal_want_verbs will be "yes" and opal_verbs_dir and
     opal_verbs_libdir will both be set to directory values, '''OR'''
   * opal_want_verbs will be "no" and opal_verbs_dir and
     opal_verbs_libdir will both be set empty

This commit was SVN r26640.
 2012-06-22 19:53:56 +00:00
Ralph Castain
e6f3586415 Remove the orte notifier framework, per discussion at the devel meeting and follow-up with Jeff (who took the action item) 
This commit was SVN r26637.
 2012-06-22 18:09:23 +00:00
Ralph Castain
60758faa55 Fix data type 
This commit was SVN r26633.
 2012-06-21 23:48:55 +00:00
Ralph Castain
e9591f2563 Fix tree spawn in the rsh/qrsh environment 
This commit was SVN r26631.
 2012-06-21 21:29:28 +00:00
Ralph Castain
0a713cd27e Add database framework to ORTE and refactor modex code to utilize it. Create the "hash" db component from the prior modex db code. Leave the other components ignored for now - will activate them later. 
Modex is still a blocking operation at this point.

This commit was SVN r26618.
 2012-06-19 13:38:42 +00:00
Ralph Castain
9e0bb6ae28 Revert r26600 and r26601 for a couple of reasons: 
1. they modified the OMPI-ORTE interface, which is something I promised to avoid doing unless absolutely necessary, and

2. the framework ident is already in the component name key provided to the modex db. What is missing is the project ident, but as Jeff and I discussed last week, we really need to add that field to the component struct anyway to avoid multi-project collisions on framework names. That will be done over the next couple of weeks as a separate effort.

This commit was SVN r26613.

The following SVN revision numbers were found above:
  r26600 --> open-mpi/ompi@5ba4deff07
  r26601 --> open-mpi/ompi@0e3094c318
 2012-06-16 09:11:03 +00:00
Ralph Castain
3c2a03b16d Update the other routed components to use common ports. Per conversation with Josh, remove the "cm" component. 
This commit was SVN r26608.
 2012-06-15 15:36:08 +00:00
Ralph Castain
96c778656a Improve launch performance on clusters that use dedicated nodes by instructing the orteds to use the same port as the HNP, thus allowing them to "rollup" their initial callback via the routed network. This substantially reduces the HNP bottleneck and the number of ports opened by the HNP. 
Restore enable-static-ports option by default - the Cray will have to disable it to get around their library issues, but that's just a warning problem as opposed to blocking the build.

This commit was SVN r26606.
 2012-06-15 10:15:07 +00:00
Ralph Castain
0e3094c318 Update the other grpcomm modules to new API 
This commit was SVN r26601.
 2012-06-14 03:28:48 +00:00
Ralph Castain
5ba4deff07 Extend the modex database to support multiple projects and frameworks that might have duplicate component names. No visible API change in the BTL's as it was executed solely in the ompi modex code. 
This commit was SVN r26600.
 2012-06-14 02:55:06 +00:00
Ralph Castain
ecc51d8583 Add missing endif 
This commit was SVN r26596.
 2012-06-12 15:07:09 +00:00
Ralph Castain
078a4667e4 Some more cleanup on direct routed when daemons are involved 
This commit was SVN r26594.
 2012-06-11 23:46:22 +00:00
Ralph Castain
269cb2b8d9 Some cleanup to remove calls to opal_progress when running with orte progress threads, and to ensure that all orte-related events are in the orte event base. 
This commit was SVN r26591.
 2012-06-11 19:59:53 +00:00
Ralph Castain
75e66ad51e Restore the direct routed component 
This commit was SVN r26590.
 2012-06-11 17:16:02 +00:00
Brian Barrett
7406ef1241 Make all the PMI components depend on the common pmi library and properly 
install the common pmi library

This commit was SVN r26588.
 2012-06-11 15:58:09 +00:00
Ralph Castain
2812579246 Just because we find an IB device does not mean we can get a QP on it. Check to see if we can before we select the UD OOB module for use. 
This commit was SVN r26587.
 2012-06-10 01:42:51 +00:00
Ralph Castain
0442a807c0 Default the OOB to the "ud" component IFF the HNP finds itself on a node with a supported Infiniband device. Ensure that the daemons all pick the matching component by dictating the selection via mca param on the orted cmd line. 
This commit was SVN r26582.
 2012-06-08 01:23:08 +00:00
Ralph Castain
05122a2f93 Make debruijn the default routed component. Update the radix component to "short-circuit" the tree when the job size permits 
This commit was SVN r26580.
 2012-06-08 00:35:36 +00:00
Ralph Castain
ffcca0185a Remove no longer needed component 
This commit was SVN r26578.
 2012-06-08 00:18:59 +00:00
Ralph Castain
980768965f Remove unused and unsupported component 
This commit was SVN r26577.
 2012-06-07 23:48:06 +00:00
Ralph Castain
350900f70e Remove unused and unsupported component 
This commit was SVN r26576.
 2012-06-07 23:47:35 +00:00
Nathan Hjelm
625c8078c3 oob/ud: fix typo 
This commit was SVN r26569.
 2012-06-07 19:21:23 +00:00
Ralph Castain
7a94a52420 No reason not to build this 
This commit was SVN r26568.
 2012-06-07 19:11:44 +00:00
Shiqing Fan
2abf783fa0 Remove a unnecessary definition before the real one. 
This commit was SVN r26562.
 2012-06-06 14:15:39 +00:00
Ralph Castain
166d254d4e Add new routed component 
This commit was SVN r26557.
 2012-06-06 11:53:12 +00:00
Ralph Castain
d6279fc971 Fix the debugger daemon launch support to fit the new state machine. Treat debugger daemons just like any other job, except that we map them only to nodes where an app process currently exists (as opposed to every node in the system). Trigger breakpoint and rank0 release only after the debugger daemons are in position. 
This commit was SVN r26556.
 2012-06-06 02:01:23 +00:00
Jeff Squyres
0b8849e2c4 Make "mpirun --report-bindings" have a user-friendly output (i.e., 
readable by normal human beings, vs. having a bitmap of physical
PU's).  Use the new hwloc base prettyprint functions to generate the
output. 

This commit was SVN r26533.
 2012-06-01 16:35:31 +00:00
Jeff Squyres
99c5afb397 Remove clang compiler warnings. 
This commit was SVN r26523.
 2012-05-29 23:36:06 +00:00
Ralph Castain
b0938a254e Dont use mutex where it isn't needed 
This commit was SVN r26521.
 2012-05-29 20:21:11 +00:00
Ralph Castain
32b66c166b Missed one blasted spot 
This commit was SVN r26520.
 2012-05-29 20:20:10 +00:00
Ralph Castain
9bedb25dda Cleanup some compiler warnings, some of which are actual logic errors 
This commit was SVN r26519.
 2012-05-29 20:11:51 +00:00
Ralph Castain
d7ac424d8d Silence optimized build warnings 
This commit was SVN r26518.
 2012-05-29 19:55:47 +00:00
Ralph Castain
bf5ec1ac0c Silence optimized build warnings 
This commit was SVN r26517.
 2012-05-29 19:55:31 +00:00
Ralph Castain
be6ed9c2df Allow partial use of allocations by specifying the max number of daemons (i.e., max VM size) for the job 
This commit was SVN r26499.
 2012-05-27 16:48:19 +00:00
Ralph Castain
7fb49b1559 Silence warning 
This commit was SVN r26480.
 2012-05-23 13:59:41 +00:00
Ralph Castain
da28a4b0e6 Silence warning 
This commit was SVN r26479.
 2012-05-23 13:59:22 +00:00
Nathan Hjelm
b9959a95cd ack! one more 
This commit was SVN r26472.
 2012-05-22 20:52:52 +00:00
Nathan Hjelm
f2d4e95429 doh! add missing include 
This commit was SVN r26471.
 2012-05-22 20:49:13 +00:00
Nathan Hjelm
cdc3c87ba6 move pmi init/finalize into a common component 
This commit was SVN r26470.
 2012-05-22 15:15:39 +00:00
Nathan Hjelm
78b8b3cf76 bug fix: actually close ess components 
This commit was SVN r26469.
 2012-05-22 15:09:18 +00:00
Nathan Hjelm
6eeca66475 add an option to enable static ports. diabled by default 
This commit was SVN r26462.
 2012-05-21 19:56:15 +00:00
Ralph Castain
83d69b6c95 Enable the ORTE progress thread for apps (not needed in the tools as they already continuously loop in the event lib). This appears to be working, at least for MPI apps that only use shared memory (a simple "hello"). More testing is required to identify where problems will occur - this is only intended to allow further development. 
In order to use the progress thread, you must configure with:

--enable-orte-progress-threads --enable-event-thread-support

This commit was SVN r26457.
 2012-05-20 15:14:43 +00:00
Ralph Castain
c4f8043064 Per Nathan, with a little cleanup by me: update the PMI support to aggregate modex info, thus reducing the number of keys required so it fits within Cray default constraints 
This commit was SVN r26456.
 2012-05-19 16:12:52 +00:00
Jeff Squyres
cab31eafce Revert r26413: it was causing too much confusion. When an MPI proc 
exits with status 77, the whole job will be killed, but mpirun will
still return an exit status of 77, so MTT will report it as a skip
anyway. 

This commit was SVN r26445.

The following SVN revision numbers were found above:
  r26413 --> open-mpi/ompi@02aa36f2e5
 2012-05-16 14:45:58 +00:00
Jeff Squyres
02aa36f2e5 ORTE defaults to killing the entire job when any process exits with a 
nonzero status (we polled other MPI implementations since one one in
the OMPI community had a concrete opinion on what behavior to do here
-- all other MPI's seem to adhere to this behavior, too).

This commit adds an MCA parameter that allows us to tell ORTE to
''not'' kill jobs when a process exits with a status of 77, meaning
the GNU testing standard of "this test was skipped".  In all the OMPI
tests, all procs will either return 77 or not.  So if they all return
77, mpirun won't consider it an error, but will still return an exit
status of 77 (so that MTT can know that the test was cleanly skipped).

This commit was SVN r26413.
 2012-05-08 21:49:05 +00:00
Ralph Castain
70a106fa71 Fix binding on remote nodes - need to pass the binding bitmap! 
This commit was SVN r26403.
 2012-05-08 03:52:39 +00:00
Jeff Squyres
2ba10c37fe Per RFC, bring in the following changes: 
* Remove paffinity, maffinity, and carto frameworks -- they've been
   wholly replaced by hwloc.
 * Move ompi_mpi_init() affinity-setting/checking code down to ORTE.
 * Update sm, smcuda, wv, and openib components to no longer use carto.
   Instead, use hwloc data.  There are still optimizations possible in
   the sm/smcuda BTLs (i.e., making multiple mpools).  Also, the old
   carto-based code found out how many NUMA nodes were ''available''
   -- not how many were used ''in this job''.  The new hwloc-using
   code computes the same value -- it was not updated to calculate how
   many NUMA nodes are used ''by this job.''
   * Note that I cannot compile the smcuda and wv BTLs -- I ''think''
     they're right, but they need to be verified by their owners.
 * The openib component now does a bunch of stuff to figure out where
   "near" OpenFabrics devices are.  '''THIS IS A CHANGE IN DEFAULT
   BEHAVIOR!!''' and still needs to be verified by OpenFabrics vendors
   (I do not have a NUMA machine with an OpenFabrics device that is a
   non-uniform distance from multiple different NUMA nodes).
 * Completely rewrite the OMPI_Affinity_str() routine from the
   "affinity" mpiext extension.  This extension now understands
   hyperthreads; the output format of it has changed a bit to reflect
   this new information.
 * Bunches of minor changes around the code base to update names/types
   from maffinity/paffinity-based names to hwloc-based names.
 * Add some helper functions into the hwloc base, mainly having to do
   with the fact that we have the hwloc data reporting ''all''
   topology information, but sometimes you really only want the
   (online | available) data.

This commit was SVN r26391.
 2012-05-07 14:52:54 +00:00
Ralph Castain
44b8608f0a Convert debug to verbose 
This commit was SVN r26384.
 2012-05-05 17:46:10 +00:00
Ralph Castain
96bfeb591c Ensure flag is passed to remote daemons 
This commit was SVN r26383.
 2012-05-03 22:31:25 +00:00
Ralph Castain
45fee2b491 Resolve the case where only the HNP is in the system (i.e., single-node operation) 
This commit was SVN r26382.
 2012-05-03 18:00:01 +00:00
Ralph Castain
c352ca36c2 Minor cleanup 
This commit was SVN r26381.
 2012-05-02 21:23:37 +00:00
Ralph Castain
b2f77bf08f Extend the iof by adding two new components to support map-reduce IO chaining. Add a mapreduce tool for running such applications. 
Fix the state machine to support multiple jobs being simultaneously launched as this is not only required for mapreduce, but can happen under comm-spawn applications as well.

This commit was SVN r26380.
 2012-05-02 21:00:22 +00:00
Ralph Castain
c5da4f24d7 Fix stupid singletons - get the pidmap message correct 
This commit was SVN r26378.
 2012-05-02 17:48:02 +00:00
Ralph Castain
9f724db182 Remove duplicate event assignment 
This commit was SVN r26360.
 2012-04-30 16:06:20 +00:00
Ralph Castain
289f9f41ec From long-term discussions, have the daemons use the node_t and proc_t structs and arrays instead of the pidmap and nidmap arrays. Sets the stage for future work. 
This commit was SVN r26359.
 2012-04-29 00:10:01 +00:00
Ralph Castain
f3e3704c9e Per request from Brian, enable mapping of stddiag output (output from opal_output calls) to stderr of the local process. This allows you to obtain that output in a local window (for example, when using xterm for each process) instead of having it automatically forwarded to mpirun. Turn this on automatically whenever someone uses the -xterm option, and to be set manually using the orte_map_stddiag_to_stderr mca param. 
This commit was SVN r26352.
 2012-04-27 14:39:34 +00:00
Jeff Squyres
46f47e08b6 Remove typo/extra brackets and parens. 
This commit was SVN r26351.
 2012-04-27 13:48:43 +00:00
Jeff Squyres
9d0df5a9a6 Update configury in the new oob ud component: actually check to see if 
it succeeds and run $1 or $2, accordingly.  This allows "make dist" to
run properly on machines that do not have OpenFabrics stuff installed
(e.g., the nightly tarball build machine).

There's still more to be done here -- it doesn't check for non-uniform
directories where the OpenFabrics headers/libraries might be
installed.  We might need to re-tool/combine
ompi/config/ompi_check_openib.m4 (which checks for way more than
oob/ud needs) and move it up to config/ompi_check_ofa.m4, or
something...?

This commit was SVN r26350.
 2012-04-27 11:32:56 +00:00
Jeff Squyres
9829d2279f System-level includes should be at the top of the file, before most 
OPAL/ORTE/OMPI includes.

This commit was SVN r26349.
 2012-04-27 11:29:22 +00:00
Ralph Castain
38af7db183 Ensure the progress message comes out right away. Otherwise, on a large system where proc state messages are arriving frequently, the message doesn't get printed until the launch is done! 
This commit was SVN r26346.
 2012-04-26 23:41:03 +00:00
Nathan Hjelm
e1e0d466e5 Merge ssh://ct-fe1/usr/projects/hpctools/hjelmn/ompi-trunk-git into HEAD 
This commit was SVN r26344.
 2012-04-26 22:06:12 +00:00
Ralph Castain
3461809341 Fix reporting of launch progress so the numbers are correct and appear when they should 
This commit was SVN r26342.
 2012-04-26 00:10:09 +00:00
Ralph Castain
71805bf7e4 Clearout the startup_timeout event if the job did in fact start. Have ORTE_TERMINATE use the job state macro so debug will show where it was called 
This commit was SVN r26334.
 2012-04-25 01:05:17 +00:00
Jeff Squyres
708b497968 Ensure to unset the iof "active" flag after the libevent read callback 
fires (it's already reset once we queue up the read event again).  Failure
to unset the active flag would cause other logic to not queue up the
read event again, because it thought the read event was still active).

This commit was SVN r26311.
 2012-04-23 15:58:12 +00:00
Ralph Castain
7999266f99 Silence warning by removing unused var 
This commit was SVN r26275.
 2012-04-17 22:34:48 +00:00
Ralph Castain
f68487016c Add test code from Terry. Properly terminate if we don't abort on non-zero exit 
This commit was SVN r26271.
 2012-04-16 16:44:23 +00:00
Ralph Castain
ddfbde587f Change the default to "abort" the job when any process exits with a non-zero status. Add the required code to ensure the orted tells the HNP about the problem. 
This commit was SVN r26270.
 2012-04-13 21:19:46 +00:00
Ralph Castain
7741ba47be Fix comm_spawn that spans multiple nodes 
This commit was SVN r26268.
 2012-04-13 01:59:07 +00:00
Ralph Castain
4d16790836 Fix collectives for jobs running across partial allocations 
This commit was SVN r26267.
 2012-04-13 00:38:47 +00:00
Ralph Castain
5d14fa7546 Fix mpi_abort, minimize error output. 
This commit was SVN r26266.
 2012-04-11 14:37:08 +00:00
Ralph Castain
d3dfba3872 Fix the scenario where an MPI error handler causes a proc to exit after finalize, but with non-zero status to indicate an error occurred. 
This commit was SVN r26265.
 2012-04-11 02:23:46 +00:00
Ralph Castain
9cd4c06488 Get things to build and run when --disable-orte is specified 
This commit was SVN r26263.
 2012-04-10 21:50:01 +00:00
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
53bbcf4b5b Plug slot allocation leak 
This commit was SVN r26260.
 2012-04-10 14:56:24 +00:00
Ralph Castain
f5cd996b91 Fix the case where n=1 
This commit was SVN r26258.
 2012-04-09 22:44:56 +00:00
Ralph Castain
a34be856aa Now that we have PMI support, this is no longer needed 
This commit was SVN r26254.
 2012-04-07 13:36:24 +00:00
Ralph Castain
71f9e69c62 Remove stale code 
This commit was SVN r26253.
 2012-04-07 13:34:12 +00:00
Ralph Castain
19630ca28d Remove stale code 
This commit was SVN r26252.
 2012-04-07 13:33:40 +00:00
Ralph Castain
93bbeabc55 Remove stale code 
This commit was SVN r26251.
 2012-04-07 13:33:30 +00:00
Ralph Castain
b6cde9a8d1 Remove stale code 
This commit was SVN r26250.
 2012-04-07 13:33:18 +00:00
George Bosilca
319f76d66a Low hanging fruit. Remove a declared but not defined function. 
This commit was SVN r26245.
 2012-04-06 15:43:28 +00:00
Ralph Castain
ed197acaa2 Eliminate stale code 
This commit was SVN r26244.
 2012-04-06 15:31:13 +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
Ralph Castain
ca3ff58c76 Ensure we get a non-zero exit status when we can't find the specified fork agent. Output a better error message, and ensure we don't multiply report the problem. 
This commit was SVN r26191.
 2012-03-24 00:49:38 +00:00
Ralph Castain
46b040c79f Fix typo 
This commit was SVN r26189.
 2012-03-24 00:31:05 +00:00
Ralph Castain
2bd75ec7e3 Fix Cray XE builds - the priority here needs to equal that of the HNP component so that both build. Otherwise, mpirun tries to use PMI for its basis, and that doesn't work! 
This commit was SVN r26188.
 2012-03-23 20:06:34 +00:00
Ralph Castain
811413e9bc Correctly handle multiple cpu-set ranges. Correctly support optional binding directives combined with cpu-set. 
This commit was SVN r26187.
 2012-03-23 14:50:41 +00:00
Ralph Castain
ce0caf7567 Support -cpu-set by binding to the specified cpus in the absence of any other binding directive. Allows users to subdivide nodes for multiple parallel mpirun invocations. 
This commit was SVN r26186.
 2012-03-23 14:05:52 +00:00
Ralph Castain
33ed3cda07 Update the gridengine allocator to support data from multiple queues by checking for duplicate node entries 
This commit was SVN r26148.
 2012-03-15 17:45:50 +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
e71e871bae Initialize sink location when stdin is forwarded to all ranks 
This commit was SVN r26107.
 2012-03-06 15:47:04 +00:00
Ralph Castain
366f9d1518 Add some missing localities to the hwloc pretty-print, fix pmi modex 
This commit was SVN r26105.
 2012-03-06 06:21:10 +00:00
Ralph Castain
834a86420b Ensure we use the slurm module for slurm environments, and correct init order in pmi module when used by daemons 
This commit was SVN r26089.
 2012-03-02 23:10:48 +00:00
Ralph Castain
ceb34ed0c9 Fix typo 
This commit was SVN r26079.
 2012-03-02 09:58:09 +00:00
Ralph Castain
b2f1bade37 Fix the -H localhost issue 
This commit was SVN r26071.
 2012-02-29 16:56:00 +00:00
Jeff Squyres
81dc6a11ee Fix typo in copyright notice, found by Paul Hargrove 
This commit was SVN r26070.
 2012-02-29 02:02:54 +00:00
Ralph Castain
a83da303c5 When using PMI, we know the ranks that share our node and their relative local/node ranks. Save that info in the pidmap array so that BTLs that require early knowledge of local ranks can access it. 
This commit was SVN r25992.
 2012-02-21 16:43:17 +00:00
Jeff Squyres
b6a90434e4 Fix some include file header ordering issues for some BSDs, suggested 
by Paul Hargrove.

This commit was SVN r25984.
 2012-02-21 13:32:14 +00:00
Jeff Squyres
b295a01d8e Fix another configury error found by Paul Hargrove. Thanks, Paul! 
This commit was SVN r25971.
 2012-02-20 21:38:27 +00:00
Jeff Squyres
cdc783925e (Re-)Add oob_tcp_if_(in|ex)clude functionality to allow CIDR notation, 
just like the btl_tcp_if_(in|ex)clude MCA param.

This commit was SVN r25953.
 2012-02-17 15:38:42 +00:00
Jeff Squyres
3e22450345 Fix the oob_tcp_verbose MCA param; make it actually apply to the OOB 
TCP verbose handle (not the generic/0 handle).

This commit was SVN r25942.
 2012-02-16 22:28:11 +00:00
Ralph Castain
b3aabf1565 Cleanup the --without-hwloc build. Thanks to Paul Hargrove for reporting it broken. 
This commit was SVN r25931.
 2012-02-15 11:08:57 +00:00
Ralph Castain
91977444af Silence warnings 
This commit was SVN r25929.
 2012-02-15 03:42:27 +00:00
Ralph Castain
bba6508b4b Handle the default hostfile case a little better... 
This commit was SVN r25928.
 2012-02-15 03:33:49 +00:00
Ralph Castain
f14c4be580 Correct the ordering logic so the list gets correctly built in daemon vpid order 
This commit was SVN r25818.
 2012-01-30 16:25:07 +00:00
Shiqing Fan
bfbd3c67a5 Add a windows file into the tarball. 
This commit was SVN r25811.
 2012-01-29 10:12:02 +00:00
Ralph Castain
a0edae52f2 Ensure the wrapper flags get entered in the right order, with -lpmi coming before the alps util libs 
This commit was SVN r25809.
 2012-01-27 20:56:21 +00:00
Ralph Castain
3f31feee6f Handle the case where a user's rankfile specifies only cpus, and not socket:cpu pairs. 
This commit was SVN r25803.
 2012-01-27 12:21:45 +00:00
Ralph Castain
07f3a91075 Okay, get srun to play nice. Problem was that everything worked fine so long as the user did "salloc" with an argument requesting a specific number of nodes. However, if the user specified instead a number of processes, then we launched that number of daemons - resulting in multiple daemons/node. Not good. 
So force things to behave correctly either way.

This commit was SVN r25792.
 2012-01-26 19:58:57 +00:00
Ralph Castain
ef94e606c7 Add some debug 
This commit was SVN r25791.
 2012-01-26 19:23:32 +00:00
Ralph Castain
1449b27e9f Ensure that slurm only launches one orted/node, regardless of how the allocation was obtained. 
This commit was SVN r25790.
 2012-01-26 19:23:15 +00:00
Jeff Squyres
64165ce758 r25775 removed the .windows from this directory, but left it in the 
Makefile.am. 

This commit was SVN r25782.

The following SVN revision numbers were found above:
  r25775 --> open-mpi/ompi@2c9a4beffd
 2012-01-26 10:45:06 +00:00
Jeff Squyres
3751495443 Add missing arguments for the new DYLD_LIBRARY_PATH stuff. 
This commit was SVN r25780.
 2012-01-26 00:35:48 +00:00
Ralph Castain
079e4d9156 Per George's comment, just duplicate the lib path envars to provide both Linux and Mac compatible values 
This commit was SVN r25776.
 2012-01-25 14:37:36 +00:00
Shiqing Fan
2c9a4beffd Add and remove a few components for windows build. 
This commit was SVN r25775.
 2012-01-25 09:01:27 +00:00
Ralph Castain
8b115754e6 Fix typo 
This commit was SVN r25763.
 2012-01-21 23:50:39 +00:00
Ralph Castain
469e40ace2 Expand the coverage a little when looking at remote shells for rsh. Prior patch (r25758) works only if both ends of the rsh/ssh connection are Mac. What we really want is to use the Mac version of ld_library_path when the remote end is Mac, regardless of the OS where mpirun is executing. So add a test for system type to the remote_shell test, and set the ld_library_path name to match the remote system type. 
This commit was SVN r25762.

The following SVN revision numbers were found above:
  r25758 --> open-mpi/ompi@1afb77e603
 2012-01-21 23:48:42 +00:00
Ralph Castain
1afb77e603 Mac requires setting DYLD_LIBRARY_PATH instead of the Linux standard LD_LIBRARY_PATH, so ensure we set that when using rsh to launch in Mac environments. 
Thanks to Teng Lin for the patch!

This commit was SVN r25758.
 2012-01-20 19:14:32 +00:00
Ralph Castain
be3dfb6a1a Ensure that we only add -lpmi once to the wrapper compilers, no matter how many components might use it. 
This commit was SVN r25753.
 2012-01-20 04:56:38 +00:00
Ralph Castain
d7fe1615b6 Add missing dollar sign on variable 
This commit was SVN r25745.
 2012-01-19 20:45:22 +00:00
Ralph Castain
0d20f745e2 Remove stale function def 
This commit was SVN r25744.
 2012-01-19 20:40:48 +00:00
Nathan Hjelm
6d0e7a0a0e don't enable ess/alps unless cnos is available 
This commit was SVN r25743.
 2012-01-19 19:36:00 +00:00
Ralph Castain
9d556e2f17 Allow daemons to use PMI to get their name where PMI support is available while using the standard grpcomm and other capabilities. Remove the GNI code from the alps ess component as that component should only be for alps/cnos installations. 
This commit was SVN r25737.
 2012-01-18 20:56:53 +00:00
Ralph Castain
6235a355de Correctly handle co-spawning of daemons when attaching to a running job. We cannot use the general process mappers as we only want debugger daemons spawned on nodes where application procs already exist. So custom build the map for the debugger daemon job, and have the plm just launch that job without doing its usual vm-spawn step. 
This commit was SVN r25736.
 2012-01-18 00:19:49 +00:00
Nathan Hjelm
a2437feba7 removed debug message 
This commit was SVN r25722.
 2012-01-12 20:23:59 +00:00