1
1
Граф коммитов

474 Коммитов

Автор SHA1 Сообщение Дата
George Bosilca
e0638c84c8 Update the test to check that all data is exposed via the
convertor_raw interface.

This commit was SVN r20383.
2009-01-28 23:07:02 +00:00
George Bosilca
ecdcda9268 Move the datatpye creation functions outside the test itself.
Add a test for the newly added raw functionality.

This commit was SVN r20374.
2009-01-28 15:42:30 +00:00
Shiqing Fan
a5281f0434 - 1/4 commit for Windows Visual Studio and CCP support:
CMakeLists and .windows files.
  In contribs preconfigured and precompiled parts.

This commit was SVN r20108.
2008-12-10 20:59:20 +00:00
Kenneth Matney
94f8189532 Under gcc 4.2.4, make check was failing without the <stdio.h>.
Moreover, I could not figure out why <time.h> would need to be
included twice.  So, I substituted the former for the latter,
in the superfluous instantiation.

This commit was SVN r19859.
2008-10-31 12:18:57 +00:00
Kenneth Matney
68248a32ef Add #include for stdio.h to allow make check to run with gcc 4.2.4 (on
Cray XT platform).

This commit was SVN r19605.
2008-09-22 18:00:30 +00:00
George Bosilca
2bd9ddfc28 The datatype dump function is always visible so we don't need a
fake one.

This commit was SVN r19158.
2008-08-05 14:45:42 +00:00
George Bosilca
e6f700bf04 Reenable the ddt_test as #1242 is now closed.
This commit was SVN r19145.
2008-08-04 15:57:02 +00:00
Brian Barrett
8cff3131d6 Remove memory tests, as they're out of date
This commit was SVN r18656.
2008-06-14 14:01:05 +00:00
Jeff Squyres
1f226b5898 Adjust the comment to be correct, per
http://www.open-mpi.org/community/lists/devel/2008/06/4095.php.

This commit was SVN r18604.
2008-06-06 01:23:58 +00:00
Ralph Castain
7c7b9b0486 Do a little cleanup on the opal graph class and opal carto framework to conform to OMPI naming conventions and avoid potential conflict with user applications - no change in functionality, passes carto test program
This commit was SVN r18407.
2008-05-07 19:33:49 +00:00
Ralph Castain
dc7f45dafd Remove the obsolete and largely unused orte_system_info structure. The only fields that were used in that struct were nodeid and nodename - these have been transferred to the orte_process_info structure.
Only one place used the user name field - session_dir, when formulating the name of the top-level directory. Accordingly, the code for getting the user's id has been moved to the session_dir code.

This commit was SVN r17926.
2008-03-23 23:10:15 +00:00
Jeff Squyres
0fbb399f13 Remove ddt_test from "make check" per #1242.
This commit was SVN r17818.
2008-03-14 14:21:47 +00:00
Jeff Squyres
4133b46ec5 Re-enable "make dist", at least until #1232 is fixed.
This commit was SVN r17796.
2008-03-09 21:36:10 +00:00
Jeff Squyres
498190e326 Add checks to ensure that opal_init() completes successfully so that
we fail gracefully (and don't segv) if opal_init() fails.

This commit was SVN r17760.
2008-03-06 14:55:32 +00:00
Tim Prins
2e1bda6d23 Remove the now-unused arithmatic interface to the dss
This commit was SVN r17654.
2008-02-28 21:36:51 +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
Jeff Squyres
04e026fa98 Fix "make check"; manually include <string.h> since the datatype
header files were re-orged to have fewer dependencies

This commit was SVN r17427.
2008-02-12 13:02:53 +00:00
Shiqing Fan
f5792bbda5 merging the memchecker into trunk.
This commit was SVN r17424.
2008-02-12 08:46:27 +00:00
Sharon Melamed
98e8de264d Wraped the carto API in carto_base_wrapers.c
This commit was SVN r17380.
2008-02-05 19:29:16 +00:00
Sharon Melamed
025b68becf Move the carto framework to the trunk.
This commit was SVN r17177.
2008-01-23 09:20:34 +00:00
George Bosilca
7eca186568 Fix a typo related to the conversion from ompi_pointer_array
to opal_pointer_array.

This commit was SVN r17023.
2007-12-22 05:32:40 +00:00
George Bosilca
906e8bf1d1 Replace the ompi_pointer_array with opal_pointer_array. The next step
(sometimes after the merge with the ORTE branch), the opal_pointer_array
will became the only pointer_array implementation (the orte_pointer_array
will be removed).

This commit was SVN r17007.
2007-12-21 06:02:00 +00:00
Rich Graham
27a748e7eb change all instances of ompi_free_list_init to ompi_free_list_init_new. Header
and payload data are specified separately at this stage.

This commit was SVN r16633.
2007-11-01 23:38:50 +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
Shiqing Fan
0f468f3668 - Remove the solution and project files, will commit them later.
This commit was SVN r15705.
2007-07-31 17:07:02 +00:00
Shiqing Fan
4d7b349cdb - Add VC8 solution and project files.
- If one wants to use this solution, remember to unload the project 'orte-restart' which is currently not working for Windows.

This commit was SVN r15680.
2007-07-30 11:05:34 +00:00
Tim Prins
7445a11f61 Remove duplicate tests. The current version of the dss tests are in orte/test/unit/dss
Remove defunct testing matrix

This commit was SVN r15535.
2007-07-20 13:37:44 +00:00
Ralph Castain
511457feb5 Remove stale test code. At least we were wise enough to have eliminated this code from the "make check" tree, but almost none of it compiles and of what does compile, nothing seems to really work.
This commit was SVN r15446.
2007-07-16 16:34:14 +00:00
Jeff Squyres
f72b52bb1d s/ifdef/if/ fro OMPI_C_HAVE_VISIBILITY to enable static builds.
This commit was SVN r14985.
2007-06-11 13:20:56 +00:00
George Bosilca
29dd535c01 Remove all references to the orte_bitmap as well as the files.
This commit was SVN r14928.
2007-06-06 20:24:07 +00:00
Brian Barrett
42c74b2cf7 fix test case so that condition variables work right, at least on PTHREADS.
I'm pretty sure condition variables are wrong for Solaris threads.

This commit was SVN r14877.
2007-06-05 19:24:17 +00:00
Brian Barrett
60571567a4 Better fix than r14831 -- ddt_pack was removed from TESTS because it
calls MPI_INIT and that causes problems during make distcheck.  Instead
put it in check_PROGRAMS which lets it get built, but doesn't run it.

This commit was SVN r14832.

The following SVN revision numbers were found above:
  r14831 --> open-mpi/ompi@9258c5200a
2007-06-01 14:34:06 +00:00
Rainer Keller
9258c5200a - As we need to reconfigure anyhow, get rid of autogen warning.
This commit was SVN r14831.
2007-06-01 08:20:38 +00:00
Brian Barrett
c7937ec02e until I figure out why MPI_INIT failed during make distcheck
This commit was SVN r14816.
2007-05-31 02:31:12 +00:00
George Bosilca
07f51ae5dc Make the test a little bit more difficult.
This commit was SVN r14814.
2007-05-30 22:40:16 +00:00
Brian Barrett
f02a9525dc add pack / unpack test
This commit was SVN r14801.
2007-05-30 17:41:15 +00:00
Sven Stork
fc932f1fb4 - changes to get the tests running with visibility enabled
This commit was SVN r14730.
2007-05-23 15:02:36 +00:00
Brian Barrett
21e00f6f0c Clean up a couple of configure things:
* Require Autoconf 2.60 or higher and remove some cruft
    required for AC 2.59 or the AC 2.59 / AC 2.60 mix
  * Remove a bunch of now unnecessary AC_SUBST calls
  * Use the libtool-provided variables for the -I and
    library to use when compiling against ltdl

Fixes trac:1000

This commit was SVN r14652.

The following Trac tickets were found above:
  Ticket 1000 --> https://svn.open-mpi.org/trac/ompi/ticket/1000
2007-05-15 04:23:48 +00:00
George Bosilca
f2a6b9394f Deal with the include spree. Protect "environ" on Windows.
Some others minors modifications in order to make it
compile [again] on Windows.

This commit was SVN r14188.
2007-04-01 16:16:54 +00:00
Jeff Squyres
4d8ee3d1e1 Add missing #include; fix the build for some picky compilers.
This commit was SVN r13696.
2007-02-17 11:54:40 +00:00
George Bosilca
beb9be3fe4 Don't import the datatype debug output if we're not in debug mode.
This commit was SVN r13650.
2007-02-14 16:47:12 +00:00
George Bosilca
06044db69a Add another test for the data-type engine. This test pack and unpack
the data in a way similar to the multi-network OB1 PML.

This commit was SVN r13632.
2007-02-13 09:30:19 +00:00
Jeff Squyres
9fb004ab8e remove the legal_numbits tests
This commit was SVN r13575.
2007-02-09 03:18:33 +00:00
Brian Barrett
6f8b366acb Rename liborte to libopen-rte and libopal to libopen-pal per telecon today
and bug #632.

Refs trac:632

This commit was SVN r12762.

The following Trac tickets were found above:
  Ticket 632 --> https://svn.open-mpi.org/trac/ompi/ticket/632
2006-12-05 18:27:24 +00:00
George Bosilca
56748d5f57 Correctly initialize the unpack buffer.
This commit was SVN r12529.
2006-11-10 05:11:02 +00:00
Sven Stork
9cf5b3709c - Add comment for volatile.
This commit was SVN r12436.
2006-11-06 14:00:43 +00:00
Sven Stork
27420fbda3 - Make counter volatile to prohibit compiler to perform optimisations.
Without this a compiler could assume that the counter is not updated
  my the malloc call and remove the test in the assert and always trigger
  the assertion.

This commit was SVN r12419.
2006-11-03 10:46:18 +00:00
George Bosilca
14c49b226a The data-type tests have to be updated too.
This commit was SVN r12334.
2006-10-27 05:34:26 +00:00
George Bosilca
a1c9a374eb Remove all the warnings from the data-type engine testing.
This commit was SVN r12167.
2006-10-18 17:00:43 +00:00
Jeff Squyres
5662122885 Fix "make dist". Temporarily snip some tests from the tarball until
they can be repaired (more changes coming in from the mad branch that
will break them).

This commit was SVN r11560.
2006-09-08 00:09:37 +00:00