1
1
Граф коммитов

132 Коммитов

Автор SHA1 Сообщение Дата
Jeff Squyres
dbb932b619 Remove the missing app mpi_after_finalize from the Makefile.
This commit was SVN r19687.
2008-10-06 14:35:15 +00:00
Jeff Squyres
78a25cf116 Commit a few missing header files, etc.
This commit was SVN r19626.
2008-09-24 15:41:42 +00:00
Ralph Castain
20ece3cb86 Add new test that stresses MPI send/recv
This commit was SVN r19530.
2008-09-09 15:47:31 +00:00
Ralph Castain
ba5498cdc6 Repair the MPI-2 dynamic operations. This includes:
1. repair of the linear and direct routed modules

2. repair of the ompi/pubsub/orte module to correctly init routes to the ompi-server, and correctly handle failure to correctly parse the provided ompi-server URI

3. modification of orterun to accept both "file" and "FILE" for designating where the ompi-server URI is to be found - purely a convenience feature

4. resolution of a message ordering problem during the connect/accept handshake that allowed the "send-first" proc to attempt to send to the "recv-first" proc before the HNP had actually updated its routes.

Let this be a further reminder to all - message ordering is NOT guaranteed in the OOB

5. Repair the ompi/dpm/orte module to correctly init routes during connect/accept.

Reminder to all: messages sent to procs in another job family (i.e., started by a different mpirun) are ALWAYS routed through the respective HNPs. As per the comments in orte/routed, this is REQUIRED to maintain connect/accept (where only the root proc on each side is capable of init'ing the routes), allow communication between mpirun's using different routing modules, and to minimize connections on tools such as ompi-server. It is all taken care of "under the covers" by the OOB to ensure that a route back to the sender is maintained, even when the different mpirun's are using different routed modules.

6. corrections in the orte/odls to ensure proper identification of daemons participating in a dynamic launch

7. corrections in build/nidmap to support update of an existing nidmap during dynamic launch

8. corrected implementation of the update_arch function in the ESS, along with consolidation of a number of ESS operations into base functions for easier maintenance. The ability to support info from multiple jobs was added, although we don't currently do so - this will come later to support further fault recovery strategies

9. minor updates to several functions to remove unnecessary and/or no longer used variables and envar's, add some debugging output, etc.

10. addition of a new macro ORTE_PROC_IS_DAEMON that resolves to true if the provided proc is a daemon

There is still more cleanup to be done for efficiency, but this at least works.

Tested on single-node Mac, multi-node SLURM via odin. Tests included connect/accept, publish/lookup/unpublish, comm_spawn, comm_spawn_multiple, and singleton comm_spawn.

Fixes ticket #1256

This commit was SVN r18804.
2008-07-03 17:53:37 +00:00
Ralph Castain
2cc8b2c51f Add yet another test, this one for proper error behavior when someone call an MPI function after calling MPI_Finalize.
Add a minor debug that outputs the orterun exit status to stderr when orte_debug is set.

This commit was SVN r18622.
2008-06-09 19:21:20 +00:00
Ralph Castain
9613b3176c Effectively revert the orte_output system and return to direct use of opal_output at all levels. Retain the orte_show_help subsystem to allow aggregation of show_help messages at the HNP.
After much work by Jeff and myself, and quite a lot of discussion, it has become clear that we simply cannot resolve the infinite loops caused by RML-involved subsystems calling orte_output. The original rationale for the change to orte_output has also been reduced by shifting the output of XML-formatted vs human readable messages to an alternative approach.

I have globally replaced the orte_output/ORTE_OUTPUT calls in the code base, as well as the corresponding .h file name. I have test compiled and run this on the various environments within my reach, so hopefully this will prove minimally disruptive.

This commit was SVN r18619.
2008-06-09 14:53:58 +00:00
Ralph Castain
0da811ce79 Initial work on xml support - allocation and job map outputs completed. More to come.
This commit was SVN r18587.
2008-06-04 20:53:12 +00:00
Ralph Castain
c992e99035 Remove the tags from orte_output_open and the filtering operation from orte_output - this will be handled differently to improve the XML output interface
This commit was SVN r18557.
2008-06-03 14:24:01 +00:00
Jeff Squyres
e7ecd56bd2 This commit represents a bunch of work on a Mercurial side branch. As
such, the commit message back to the master SVN repository is fairly
long.

= ORTE Job-Level Output Messages =

Add two new interfaces that should be used for all new code throughout
the ORTE and OMPI layers (we already make the search-and-replace on
the existing ORTE / OMPI layers):

 * orte_output(): (and corresponding friends ORTE_OUTPUT,
   orte_output_verbose, etc.)  This function sends the output directly
   to the HNP for processing as part of a job-specific output
   channel.  It supports all the same outputs as opal_output()
   (syslog, file, stdout, stderr), but for stdout/stderr, the output
   is sent to the HNP for processing and output.  More on this below.
 * orte_show_help(): This function is a drop-in-replacement for
   opal_show_help(), with two differences in functionality:
   1. the rendered text help message output is sent to the HNP for
      display (rather than outputting directly into the process' stderr
      stream)
   1. the HNP detects duplicate help messages and does not display them
      (so that you don't see the same error message N times, once from
      each of your N MPI processes); instead, it counts "new" instances
      of the help message and displays a message every ~5 seconds when
      there are new ones ("I got X new copies of the help message...")

opal_show_help and opal_output still exist, but they only output in
the current process.  The intent for the new orte_* functions is that
they can apply job-level intelligence to the output.  As such, we
recommend that all new ORTE and OMPI code use the new orte_*
functions, not thei opal_* functions.

=== New code ===

For ORTE and OMPI programmers, here's what you need to do differently
in new code:

 * Do not include opal/util/show_help.h or opal/util/output.h.
   Instead, include orte/util/output.h (this one header file has
   declarations for both the orte_output() series of functions and
   orte_show_help()).
 * Effectively s/opal_output/orte_output/gi throughout your code.
   Note that orte_output_open() takes a slightly different argument
   list (as a way to pass data to the filtering stream -- see below),
   so you if explicitly call opal_output_open(), you'll need to
   slightly adapt to the new signature of orte_output_open().
 * Literally s/opal_show_help/orte_show_help/.  The function signature
   is identical.

=== Notes ===

 * orte_output'ing to stream 0 will do similar to what
   opal_output'ing did, so leaving a hard-coded "0" as the first
   argument is safe.
 * For systems that do not use ORTE's RML or the HNP, the effect of
   orte_output_* and orte_show_help will be identical to their opal
   counterparts (the additional information passed to
   orte_output_open() will be lost!).  Indeed, the orte_* functions
   simply become trivial wrappers to their opal_* counterparts.  Note
   that we have not tested this; the code is simple but it is quite
   possible that we mucked something up.

= Filter Framework =

Messages sent view the new orte_* functions described above and
messages output via the IOF on the HNP will now optionally be passed
through a new "filter" framework before being output to
stdout/stderr.  The "filter" OPAL MCA framework is intended to allow
preprocessing to messages before they are sent to their final
destinations.  The first component that was written in the filter
framework was to create an XML stream, segregating all the messages
into different XML tags, etc.  This will allow 3rd party tools to read
the stdout/stderr from the HNP and be able to know exactly what each
text message is (e.g., a help message, another OMPI infrastructure
message, stdout from the user process, stderr from the user process,
etc.).

Filtering is not active by default.  Filter components must be
specifically requested, such as:

{{{
$ mpirun --mca filter xml ...
}}}

There can only be one filter component active.

= New MCA Parameters =

The new functionality described above introduces two new MCA
parameters:

 * '''orte_base_help_aggregate''': Defaults to 1 (true), meaning that
   help messages will be aggregated, as described above.  If set to 0,
   all help messages will be displayed, even if they are duplicates
   (i.e., the original behavior).
 * '''orte_base_show_output_recursions''': An MCA parameter to help
   debug one of the known issues, described below.  It is likely that
   this MCA parameter will disappear before v1.3 final.

= Known Issues =

 * The XML filter component is not complete.  The current output from
   this component is preliminary and not real XML.  A bit more work
   needs to be done to configure.m4 search for an appropriate XML
   library/link it in/use it at run time.
 * There are possible recursion loops in the orte_output() and
   orte_show_help() functions -- e.g., if RML send calls orte_output()
   or orte_show_help().  We have some ideas how to fix these, but
   figured that it was ok to commit before feature freeze with known
   issues.  The code currently contains sub-optimal workarounds so
   that this will not be a problem, but it would be good to actually
   solve the problem rather than have hackish workarounds before v1.3 final.

This commit was SVN r18434.
2008-05-13 20:00:55 +00:00
Ralph Castain
7b91f8baff Cleanup and fix bugs in the MPI dynamics section. Modify the dpm API so it properly takes ports instead of process names (as correctly identified by Aurelien). Fix race conditions in the use of ompi-server. Fix incompatibilities between the mpi bindings and the dpm implemenation that could cause segfaults due to uninitialized memory.
Fix the ompi-server -h cmd line option so it actually tells you something!

Add two new testing codes to the orte/test/mpi area: accept and connect.

This commit was SVN r18176.
2008-04-16 14:27:42 +00:00
Ralph Castain
5e6dc24e62 Fix ompi-server so it works with unity routed module - still not working with tree routing.
Cleanup debug flag so it activates debugging on the data server code itself

This commit was SVN r18080.
2008-04-04 19:17:28 +00:00
Ralph Castain
6bb139e4f2 One more correction to mpirun exit codes - cleanup the application proc's exit codes in the orted so that non-zero exit codes generated by mpirun itself don't get "munged".
Modify the multi_abort function so they all return different exit codes - allows us to tell which one was being reported.

This commit was SVN r17895.
2008-03-20 13:54:11 +00:00
Ralph Castain
2ed0e60321 Bring some sanity to the exit code returned by mpirun. Ensure that we provide a non-zero code if something goes wrong, including someone exiting after calling mpi_init without calling mpi_finalize.
Jeff is preparing an (undoubtedly lengthy) explanation/matrix of how these codes are determined for the OMPI FAQ.

This commit was SVN r17879.
2008-03-19 19:00:51 +00:00
Ralph Castain
d70e2e8c2b Merge the ORTE devel branch into the main trunk. Details of what this means will be circulated separately.
Remains to be tested to ensure everything came over cleanly, so please continue to withhold commits a little longer

This commit was SVN r17632.
2008-02-28 01:57:57 +00:00
Ralph Castain
6f498c0964 Since we no longer set the APP_NUM attribute in the case of a singleton, protect the code so we don't crash in that case.
This commit was SVN r16461.
2007-10-16 16:17:48 +00:00
Ralph Castain
54b2cf747e These changes were mostly captured in a prior RFC (except for #2 below) and are aimed specifically at improving startup performance and setting up the remaining modifications described in that RFC.
The commit has been tested for C/R and Cray operations, and on Odin (SLURM, rsh) and RoadRunner (TM). I tried to update all environments, but obviously could not test them. I know that Windows needs some work, and have highlighted what is know to be needed in the odls process component.

This represents a lot of work by Brian, Tim P, Josh, and myself, with much advice from Jeff and others. For posterity, I have appended a copy of the email describing the work that was done:

As we have repeatedly noted, the modex operation in MPI_Init is the single greatest consumer of time during startup. To-date, we have executed that operation as an ORTE stage gate that held the process until a startup message containing all required modex (and OOB contact info - see #3 below) info could be sent to it. Each process would send its data to the HNP's registry, which assembled and sent the message when all processes had reported in.

In addition, ORTE had taken responsibility for monitoring process status as it progressed through a series of "stage gates". The process reported its status at each gate, and ORTE would then send a "release" message once all procs had reported in.

The incoming changes revamp these procedures in three ways:

1. eliminating the ORTE stage gate system and cleanly delineating responsibility between the OMPI and ORTE layers for MPI init/finalize. The modex stage gate (STG1) has been replaced by a collective operation in the modex itself that performs an allgather on the required modex info. The allgather is implemented using the orte_grpcomm framework since the BTL's are not active at that point. At the moment, the grpcomm framework only has a "basic" component analogous to OMPI's "basic" coll framework - I would recommend that the MPI team create additional, more advanced components to improve performance of this step.

The other stage gates have been replaced by orte_grpcomm barrier functions. We tried to use MPI barriers instead (since the BTL's are active at that point), but - as we discussed on the telecon - these are not currently true barriers so the job would hang when we fell through while messages were still in process. Note that the grpcomm barrier doesn't actually resolve that problem, but Brian has pointed out that we are unlikely to ever see it violated. Again, you might want to spend a little time on an advanced barrier algorithm as the one in "basic" is very simplistic.

Summarizing this change: ORTE no longer tracks process state nor has direct responsibility for synchronizing jobs. This is now done via collective operations within the MPI layer, albeit using ORTE collective communication services. I -strongly- urge the MPI team to implement advanced collective algorithms to improve the performance of this critical procedure.


2. reducing the volume of data exchanged during modex. Data in the modex consisted of the process name, the name of the node where that process is located (expressed as a string), plus a string representation of all contact info. The nodename was required in order for the modex to determine if the process was local or not - in addition, some people like to have it to print pretty error messages when a connection failed.

The size of this data has been reduced in three ways:

(a) reducing the size of the process name itself. The process name consisted of two 32-bit fields for the jobid and vpid. This is far larger than any current system, or system likely to exist in the near future, can support. Accordingly, the default size of these fields has been reduced to 16-bits, which means you can have 32k procs in each of 32k jobs. Since the daemons must have a vpid, and we require one daemon/node, this also restricts the default configuration to 32k nodes.

To support any future "mega-clusters", a configuration option --enable-jumbo-apps has been added. This option increases the jobid and vpid field sizes to 32-bits. Someday, if necessary, someone can add yet another option to increase them to 64-bits, I suppose.

(b) replacing the string nodename with an integer nodeid. Since we have one daemon/node, the nodeid corresponds to the local daemon's vpid. This replaces an often lengthy string with only 2 (or at most 4) bytes, a substantial reduction.

(c) when the mca param requesting that nodenames be sent to support pretty error messages, a second mca param is now used to request FQDN - otherwise, the domain name is stripped (by default) from the message to save space. If someone wants to combine those into a single param somehow (perhaps with an argument?), they are welcome to do so - I didn't want to alter what people are already using.

While these may seem like small savings, they actually amount to a significant impact when aggregated across the entire modex operation. Since every proc must receive the modex data regardless of the collective used to send it, just reducing the size of the process name removes nearly 400MBytes of communication from a 32k proc job (admittedly, much of this comm may occur in parallel). So it does add up pretty quickly.


3. routing RML messages to reduce connections. The default messaging system remains point-to-point - i.e., each proc opens a socket to every proc it communicates with and sends its messages directly. A new option uses the orteds as routers - i.e., each proc only opens a single socket to its local orted. All messages are sent from the proc to the orted, which forwards the message to the orted on the node where the intended recipient proc is located - that orted then forwards the message to its local proc (the recipient). This greatly reduces the connection storm we have encountered during startup.

It also has the benefit of removing the sharing of every proc's OOB contact with every other proc. The orted routing tables are populated during launch since every orted gets a map of where every proc is being placed. Each proc, therefore, only needs to know the contact info for its local daemon, which is passed in via the environment when the proc is fork/exec'd by the daemon. This alone removes ~50 bytes/process of communication that was in the current STG1 startup message - so for our 32k proc job, this saves us roughly 32k*50 = 1.6MBytes sent to 32k procs = 51GBytes of messaging.

Note that you can use the new routing method by specifying -mca routed tree - if you so desire. This mode will become the default at some point in the future.


There are a few minor additional changes in the commit that I'll just note in passing:

* propagation of command line mca params to the orteds - fixes ticket #1073. See note there for details.

* requiring of "finalize" prior to "exit" for MPI procs - fixes ticket #1144. See note there for details.

* cleanup of some stale header files

This commit was SVN r16364.
2007-10-05 19:48:23 +00:00
Ralph Castain
684aa1bc9f Since universe size now is an orte thing, we may as well give it some direct support. Create rmgr set/get functions so it becomes more obvious where this value is being defined and how to retrieve it. Modify the bproc pls to pass it to the app procs when launched. Modify one of the test programs to verify it has been correctly set.
This commit was SVN r15266.
2007-07-02 16:45:40 +00:00
Ralph Castain
b582d98d4a Fix singleton comm_spawn so it can see available resources
This commit was SVN r14719.
2007-05-22 13:29:07 +00:00
Ralph Castain
677eb5e4bc Ensure the singleton wakes up when comm_spawn fails
This commit was SVN r14714.
2007-05-21 20:13:31 +00:00
Jeff Squyres
97248d6bc6 Add another test to check multiple, concurrent COMM_SPAWN's.
This commit was SVN r14701.
2007-05-19 19:02:24 +00:00
Jeff Squyres
47ba3db3b8 Add a simple MPI_COMM_SPAWN_MULTIPLE test.
This commit was SVN r14700.
2007-05-19 02:30:53 +00:00
Jeff Squyres
2cbcb4abf1 Remove the French and strip the tests down to essentials (no need for
buffer attaching/detaching, for example).

This commit was SVN r14216.
2007-04-04 15:38:23 +00:00
Ralph Castain
d5b5cd2d3c Add test code for multiple comm_spawn calls.
Add ERROR_LOG calls to more clearly document failures in the rsh launcher.

This commit was SVN r14214.
2007-04-04 13:24:39 +00:00
Ralph Castain
26897a626d Add a delayed_abort test code. We seem to handle this case just fine now, but Sun reports still seeing troubles on Solaris.
This commit was SVN r13493.
2007-02-05 15:24:01 +00:00
Jeff Squyres
f6e7016cdd Make this test capable of running more than "-np 1". If you run with
"-np X", it will launch X parents and then MPI_COMM_SPAWN X additional
children.

This commit was SVN r13466.
2007-02-02 14:34:53 +00:00
Ralph Castain
f95e20e2e1 Add another test program - an MPI app that just spins. This supports testing of system response to signal-terminated processes.
Add some debugger output to the ODLS default component.

Modify the orted command communication system so that it is done via non-blocking sends. This removes the linearity of the transmission and improves the response time.

This commit was SVN r12585.
2006-11-13 21:51:34 +00:00
Ralph Castain
17c71f8d2a Fix ticket #545
Setup subscriptions to correctly return the MPI_APPNUM attribute.

Fix an unreported bug that was found. The universe size was incorrectly defined in the attributes code. As coded, it looked for size_t values and based its size computation on those numbers. Unfortunately, the node_slots value had been changed to an orte_std_cntr_t awhile back! So the universe size was never updated.

Update the hello_nodename test to check for MPI_APPNUM.

Add a definition to ns_types for ORTE_PROC_MY_NAME - just a shortcut for orte_process_info.my_name. Brought over from ORTE 2.0 as it will be used extensively there.

This commit was SVN r12377.
2006-10-31 21:29:07 +00:00
Ralph Castain
ec0bb9ffda Fix the bookmark system - we now have children being correctly spawned where they should!
Also, I am no longer seeing any issue with the child job spawning its own daemons - this appears to be fixed. We still don't reuse the existing daemons, however, but that will come.

This commit was SVN r12229.
2006-10-20 18:05:16 +00:00
Ralph Castain
02efd07b60 Fix the MCA param passing issue, at least for rsh at the moment. I will clean this up and move it to the other environments once I shift back to a local computer.
This commit was SVN r12224.
2006-10-20 15:27:29 +00:00
Ralph Castain
99f2986db7 Bring comm_spawn back online. Shift the trigger hosting responsibilities to the HNP.
We still have an issue with the io forwarding going through the spawning process, but that will be dealt with at a future time.

This commit was SVN r11943.
2006-10-03 02:07:58 +00:00
Ralph Castain
37dfdb76eb Here is the major MAD-cure commit. I have written plenty about it, so I refer you here to those messages for a description of everything that was done.
This commit was SVN r11661.
2006-09-14 21:29:51 +00:00
Ralph Castain
5dfd54c778 With the branch to 1.2 made....
Clean up the remainder of the size_t references in the runtime itself. Convert to orte_std_cntr_t wherever it makes sense (only avoid those places where the actual memory size is referenced).

Remove the obsolete oob barrier function (we actually obsoleted it a long time ago - just never bothered to clean it up).

I have done my best to go through all the components and catch everything, even if I couldn't test compile them since I wasn't on that type of system. Still, I cannot guarantee that problems won't show up when you test this on specific systems. Usually, these will just show as "warning: comparison between signed and unsigned" notes which are easily fixed (just change a size_t to orte_std_cntr_t).

In some places, people didn't use size_t, but instead used some other variant (e.g., I found several places with uint32_t). I tried to catch all of them, but...

Once we get all the instances caught and fixed, this should once and for all resolve many of the heterogeneity problems.

This commit was SVN r11204.
2006-08-15 19:54:10 +00:00