1
1
Граф коммитов

11826 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
c693d3a5d1 I hadn't honestly considered before that an MPI process might attempt to call functions in the routed framework intended solely for daemons and HNPs. By design, MPI processes are not allowed to route RML/OOB messages, and hence the routed module in an MPI process has no knowledge whatsoever of how a message will reach its destination (except in the direct module). Thus, it has no way to return a valid routing tree, update a routing tree, or get wireup info.
This commit ensures that attempts to access information that is unknowable or undefined returns appropriate invalid or not_supported values to avoid unexpected behavior and/or segfaults.

This commit was SVN r18692.
2008-06-20 03:26:13 +00:00
Jeff Squyres
7905db57bd Slightly decrease the number of buffers for the NetXen adapter
This commit was SVN r18691.
2008-06-20 01:00:22 +00:00
Jeff Squyres
d6e4ea3803 Two tidbits to make autogen.sh a little more robust:
* Ensure that various aliases and color settings don't muck up some
   common shell commands that we use in autogen.sh
 * Per Ralf W's suggestion, properly []-ize the first argument to
   m4_define()

This commit was SVN r18690.
2008-06-20 00:39:43 +00:00
Ralph Castain
5ebe10ebf1 Fix a bad typo - need to look at the node array as the arch array hasn't been built yet
This commit was SVN r18689.
2008-06-19 21:34:39 +00:00
Ralph Castain
174b9f1482 Ensure this module works in heterogeneous environments.
Note: this module is under development, which is why it is not set as the default. Use at your own risk!

This commit was SVN r18688.
2008-06-19 19:40:47 +00:00
Ralph Castain
ccbf194e8f Visibility fix
This commit was SVN r18687.
2008-06-19 19:08:08 +00:00
Pak Lui
119df10349 Fix the debugging messages
This commit was SVN r18686.
2008-06-19 18:54:20 +00:00
Pak Lui
a924b4a7f4 Define the symbols to allow parallel debuggers to dlopen
the shared object when it is compiled with the Sun Studio C compiler.
Depends on where the extern variables that are included in the headers were 
initialized, there can be instances wheter there is no storage allocated 
for the variables and therefore the symbols may or may not be defined 
when the debugger tries to dlopen this message queue dll.

This commit was SVN r18685.
2008-06-19 18:49:25 +00:00
Ralph Castain
26c9ad5799 Clean-up the DSS API to remove two functions that are supposed to be used solely internally to the DSS. These were likely exposed because we need to call them when packing/unpacking declared types, but this means that developers may accidentally use the wrong functions, causing the DSS buffer to get confused. Instead, return the system to the way it used to work and hide those functions.
This commit was SVN r18684.
2008-06-19 18:46:25 +00:00
George Bosilca
bc9b950162 Honor ^ for the PML selection.
This commit was SVN r18683.
2008-06-19 16:50:46 +00:00
Josh Hursey
b78ae13bf3 add back a missing a header taken away in r18664
This commit was SVN r18682.

The following SVN revision numbers were found above:
  r18664 --> open-mpi/ompi@0532d799d6
2008-06-19 16:08:27 +00:00
Ralph Castain
265b4de5de Ensure that the call to orte_routed is properly protected at compile time when RTE support is disabled
This commit was SVN r18681.
2008-06-19 15:20:06 +00:00
Jeff Squyres
e4172a3c44 Shift the AM "if" logic down from orte/tools/Makefile.am down to the
individual orte/tools/*/Makefile.am files.  This causes "make" to
travese into every directory, even if it's not going to build anything
in that directory (which is a good thing).  It also helps cleanup and
dist issues.

This also affects orte-checkpoint and orte-restart, but I couldn't get
--with-ft to compile properly; I'll pass along a heads-up to Josh to
ensure that I didn't break anything.

This commit was SVN r18680.
2008-06-19 14:46:10 +00:00
Jeff Squyres
a884eebdf1 This warning has been bugging me in MTT nightly runs for forever: make
the char string ident in the C++ library be non-static so that other
places can see it.  This makes the C++ library version string
analogous to all the other version strings.

This commit was SVN r18679.
2008-06-19 14:40:37 +00:00
Ralph Castain
571f483c39 Ensure that we don't breakpoint the debugger until -after- all procs have reported their contact info so we can successfully send the release message
This commit was SVN r18678.
2008-06-19 14:37:46 +00:00
Ralph Castain
3b5e80fa61 Shift responsibility for preconnecting the oob to the orte routed framework, which is the only place that knows what needs to be done. Only the direct module will actually do anything - it uses the same algo as the original preconnect function.
This commit was SVN r18677.
2008-06-19 13:48:26 +00:00
Jeff Squyres
7e45b24001 MPIR_being_debugged is an int, not a bool.
This commit was SVN r18676.
2008-06-19 13:31:34 +00:00
Pavel Shamis
4537827973 Making the qp allocation more optimized.
- sq parameter was replaced with max_inline parameter
- inline is allocated only for relevant QPs

This commit was SVN r18675.
2008-06-19 08:40:39 +00:00
Ralph Castain
b56f8ced4f Ensure params are registered prior to parsing global cmd line options in orterun so that debugger options are properly captured and acted upon.
Ensure that routes to remote procs are set on the HNP before completing launch so that the debugger message can be sent. Solves a race condition that can exist in those environments where the HNP does not have local procs.

This commit was SVN r18674.
2008-06-19 02:58:14 +00:00
Ralph Castain
955d117f5e Add a new grpcomm module that mimics the old 1.2 behavior - it -always- does a modex because it always includes the architecture. Hence, we called it "blind-and-dumb" since it doesn't look to see if this is required - moniker of "bad". :-)
Update the ESS API so we can update the stored arch's should the modex include that info. Update ompi/proc to check/set the arch for remote procs, and add that function call to mpi_init right after the modex is done.

Setup to allow other grpcomm modules to decide whether or not to add the arch to the modex, and to detect if other entries have been made. If not, then the modex can just fall through. Begin setting up some logic in the "basic" module to handle different arch situations.

For now, default to the "bad" module so we will work in all situations, even though we may be sending around more info than we really require.

This fixes ticket #1340

This commit was SVN r18673.
2008-06-18 22:17:53 +00:00
Jeff Squyres
da6aa57efb Change up the logic a bit to handle the Red Storm case a bit better
(still need the real yod environment variable name), and add in some
lengthy comments explaining the two different methods of debugger
attach that we're using.

MPI handle debugging doesn't seem to be working at the moment; still
checking into it...

This commit was SVN r18672.
2008-06-18 21:33:08 +00:00
Pak Lui
188c8bce5d Fix the SEGV when module_get finds that no proc is binded. Also make no-intr available for processor binding.
This commit was SVN r18671.
2008-06-18 16:03:08 +00:00
Ralph Castain
282a220e7e Update the debugger interface per email thread with Jeff and Brian. Handoff to them for final test and validation
This commit was SVN r18670.
2008-06-18 15:28:46 +00:00
Brian Barrett
558e68088c Update Cray XT3 (catamount) platform files to use script wrapper compilers,
properly disable building contrib packages, and always build ROMIO

This commit was SVN r18669.
2008-06-18 14:41:25 +00:00
George Bosilca
1dba362a01 Make the ompi_ddt_dump function externally visible.
This commit was SVN r18668.
2008-06-18 08:37:43 +00:00
George Bosilca
f97a728dc6 Dont cast the int32_t pointer into a long pointer. This doesn't work on
64 bits architectures.

This commit was SVN r18667.
2008-06-18 08:33:58 +00:00
George Bosilca
8e7c35e76c These symbols are only available via the module/component structure, so they
don't have to be globally visible.

This commit was SVN r18666.
2008-06-18 08:20:02 +00:00
George Bosilca
0f9b9c0aff Remove a warning and add arequired header (otherwise we cannot compile when
--disable-debug is specified).

This commit was SVN r18665.
2008-06-18 08:10:02 +00:00
Ralph Castain
0532d799d6 Complete implementation of the --without-rte-support configure option. Working with Brian, this has been tested on RedStorm.
Some minor changes to help facilitate debugger support so that both mpirun and yod can operate with it. Still to be completed.

This commit was SVN r18664.
2008-06-18 03:15:56 +00:00
Jeff Squyres
d0cfca5990 Documentation describing how TV (and others like it) attach to MPI
processes.  Originally downloaded from
http://www-unix.mcs.anl.gov/mpi/mpi-debug/mpich-attach.txt -- cached
here in case that file ever disappears someday.

This commit was SVN r18663.
2008-06-17 21:19:34 +00:00
Jeff Squyres
16b2a50543 Slight clarification of help message.
This commit was SVN r18661.
2008-06-17 11:25:32 +00:00
Jeff Squyres
1e00d26c61 Minor word smything
This commit was SVN r18660.
2008-06-16 19:56:13 +00:00
Jeff Squyres
c1d1ffbc56 Fix compile problems on systems with older versions of libnuma (that
don't have MPOL_MF_MOVE).  I know that this is a configure change in
the middle of the US workday, but this compile problem is preventing
work on several kinds of systems (e.g., RHEL4).

This commit was SVN r18659.
2008-06-16 17:26:42 +00:00
Lenny Verkhovsky
f4811d6c4d NUMA Awareness support. Gleb's patch
This commit was SVN r18658.
2008-06-15 13:43:28 +00:00
Lenny Verkhovsky
dee2f1d175 Adding new functionality to Maffinity component to support NUMA awareness
This commit was SVN r18657.
2008-06-15 07:27:29 +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
Brian Barrett
7712b07ac4 Add perl based wrapper compilers for cross-compile environments. The default
is still to use the C based wrapper compilers (which have many more features
and are more well tested).  The Perl compilers are enabled with the option
--enable-script-wrapper-compilers, which also ignores the option
--disable-binaries (ie --enable-script-wrapper-compilers --disable-binaries
will result in perl-based wrapper compilers being installed, but no other
binaries being installed).

This commit was SVN r18655.
2008-06-13 22:52:25 +00:00
Brian Barrett
79ad6d983e - The ptmalloc2 memory manager component is now by default built as
a standalone library named libopenmpi-malloc.  Users wanting to
  use leave_pinned with ptmalloc2 will now need to link the library
  into their application explicitly.  All other users will use the
  libc-provided allocator instead of Open MPI's ptmalloc2.  This change
  may be overriden with the configure option enable-ptmalloc2-internal
- The leave_pinned options will now default to using mallopt on
  Linux in the cases where ptmalloc2 was not linked in.  mallopt
  will also only be available if munmap can be intercepted (the
  default whenever Open MPI is not compiled with --without-memory-
  manager.
- Open MPI will now complain and refuse to use leave_pinned if
  no memory intercept / mallopt option is available.

This commit was SVN r18654.
2008-06-13 22:32:49 +00:00
Galen Shipman
44cd373a87 I also forgot to initialize the convertor max_data, george probably copied
this dumb mistake from me. 

This commit was SVN r18653.
2008-06-13 18:33:43 +00:00
Ralph Castain
378c16b166 Update LANL config files
This commit was SVN r18652.
2008-06-13 12:45:21 +00:00
George Bosilca
170b9c344e Mea culpa. I forget to initialize the max_data before the call to the
convertor.

This commit was SVN r18651.
2008-06-12 17:24:39 +00:00
Pavel Shamis
dc3f14736d Fixing QP initialization stuff.
This commit was SVN r18650.
2008-06-11 16:31:39 +00:00
Ralph Castain
a87aa442e3 Remove last remaining reference to iof_flush - it was #if'd out anyway. The existing flush code appears to have several critical problems. Given the impending rework of the IOF subsystem, there is no point in trying to fix it here.
This commit was SVN r18649.
2008-06-11 16:25:46 +00:00
Ralph Castain
1beb3df065 Let threads exist in these platforms...
This commit was SVN r18648.
2008-06-11 16:24:34 +00:00
Ralph Castain
1f41069ac9 Fix CID 752 - if we can't find the daemon job object, we have to ensure we exit without attempting to dereference it
This commit was SVN r18647.
2008-06-11 14:49:58 +00:00
Ralph Castain
13ea4e4673 Be consistent - since we don't strdup the other values for param, don't strdup this one.
This allows r18645 to fix the memory corruption issue, but also allows us to resolve the memory leaks cited by CID 1039

This commit was SVN r18646.

The following SVN revision numbers were found above:
  r18645 --> open-mpi/ompi@53d83ba1c5
2008-06-11 14:42:47 +00:00
Pak Lui
53d83ba1c5 Take out a couple of free's.
This commit fixes trac:1343

This commit was SVN r18645.

The following Trac tickets were found above:
  Ticket 1343 --> https://svn.open-mpi.org/trac/ompi/ticket/1343
2008-06-11 14:02:49 +00:00
Ralph Castain
d61fe87d04 Use the opal_show_help system if orte_show_help has not been initialized
This fixes ticket #1342

This commit was SVN r18644.
2008-06-11 12:50:40 +00:00
Ralph Castain
f9d809748c Glad someone found that last error - caused me to review the code and find a couple of other cleanups! Nothing major, but just ensure that things flow smoothly since we had a "shadowed" variable.
This commit was SVN r18643.
2008-06-10 19:15:59 +00:00
Camille Coti
67cd1849f7 *map was still NULL in the else statement, inducing a segmentation fault when a field of the structure was accessed to.
This commit was SVN r18642.
2008-06-10 19:00:57 +00:00