1
1
Граф коммитов

1148 Коммитов

Автор SHA1 Сообщение Дата
Shiqing Fan
2326f14be5 Remove the unnecessary PROJECT command, I somehow misunderstood how it should be used on Windows....
This commit was SVN r20634.
2009-02-25 16:07:43 +00:00
Eugene Loh
463f11f993 Improve shared-memory allocation:
* compute mmap-file size more wisely and pass requested size to allocator
* change MCA parameters:
  - get rid of mpool_sm_per_peer_size
  - get rid of mpool_sm_max_size
  - set default mpool_sm_min_size to 0
* no longer pad sm allocations to page boundaries
* have sm_btl_first_time_init check return codes on free-list creations

Have mca_btl_sm_prepare_src() check to see if it can allocate an EAGER fragment
rather than a MAX fragment if the smaller size works.

Remove ompi/class/ompi_[circular_buffer_]fifo.h and references thereto.

Remove opal/util/pow2.[c|h] and references thereto.

This commit was SVN r20614.
2009-02-20 19:51:57 +00:00
Jeff Squyres
1a7556d2c9 Refs trac:1805: temporarily disable some assert()s in event_base_free().
This commit was SVN r20609.

The following Trac tickets were found above:
  Ticket 1805 --> https://svn.open-mpi.org/trac/ompi/ticket/1805
2009-02-20 15:03:36 +00:00
Jeff Squyres
ed22f9744e Bring in PLPA v1.3rc3 (add a missing comma, which should fix compiling
issues at Sun).  

Sorry for the middle of the day configure change, but this should fix
a compile break at Sun...

This commit was SVN r20594.
2009-02-19 17:42:43 +00:00
Jeff Squyres
558fc2836d Bump PLPA version to 1.3rc2, which should fix the "make dist" error
from last night's nightly tarball.

This commit was SVN r20576.
2009-02-17 12:54:57 +00:00
Jeff Squyres
4590582807 Bump PLPA to v1.3rc1, which includes a valgrind API fix for a
known-bad memory access pattern.  Specifically, a NULL pointer is
passed in a system call as part of a probe to figure out which
affinity API this system has.  We know it's a NULL and we did it on
purpose, so don't have Valgrind yell about it.

This commit was SVN r20572.
2009-02-17 02:01:30 +00:00
Jeff Squyres
8b9601e35e Put a check in the destructor to ensure that we don't try to free a
NULL pointer.

This commit was SVN r20569.
2009-02-17 01:11:10 +00:00
George Bosilca
918d94f449 Put back the commit r20562 as it had a reason to be there: clean
a memory leak.

This commit was SVN r20566.

The following SVN revision numbers were found above:
  r20562 --> open-mpi/ompi@62c913f851
2009-02-16 20:03:48 +00:00
Josh Hursey
350d9c94ab Backout r20562 since it breaks finalization in the tools (per email to devel).
This commit was SVN r20565.

The following SVN revision numbers were found above:
  r20562 --> open-mpi/ompi@62c913f851
2009-02-16 19:18:43 +00:00
George Bosilca
62c913f851 Release the default base on finalize.
This commit was SVN r20562.
2009-02-14 21:51:09 +00:00
Jeff Squyres
f9043edd39 Ensure to free this string when we're done.
This commit was SVN r20536.
2009-02-12 22:54:43 +00:00
Jeff Squyres
17d9c2c240 Important clarification about the ownership of strings returned by the
current_value parameter to mca_base_param_reg_string() function.

This commit was SVN r20535.
2009-02-12 22:54:29 +00:00
Jeff Squyres
e3ae1468d3 Don't strdup here; there's already a strdup down in
param_set_override().

This commit was SVN r20533.
2009-02-12 22:36:45 +00:00
Ralph Castain
883a0972bc Add missing include file
This commit was SVN r20529.
2009-02-12 16:41:48 +00:00
George Bosilca
84d3ca0c9e Release the memory.
This commit was SVN r20524.
2009-02-11 21:34:27 +00:00
Shiqing Fan
2f1461419c Add a new feature for checking mca subdirectories, i.e. detecting if there is an exclude file list which indicates the files that shouldn't be added to the source list. By default, the CMake build system will simply add all source files in the required sub folders, without knowing which files have to be excluded. The first use of it is in plm/base/.windows.
And clean up the nested variable names, in order to make it readable.

This commit was SVN r20498.
2009-02-10 17:20:13 +00:00
Ralph Castain
4cdf91a8d4 Per the RFC, extend the current use of the ompi_proc_t flags field (without changing the field itself).
The prior ompi_proc_t structure had a uint8_t flag field in it, where only one
bit was used to flag that a proc was "local". In that context, "local" was
constrained to mean "local to this node".

This commit provides a greater degree of granularity on the term "local", to include tests
to see if the proc is on the same socket, PC board, node, switch, CU (computing
unit), and cluster.

Add #define's to designate which bits stand for which local condition. This
was added to the OPAL layer to avoid conflicting with the proposed movement of
the BTLs. To make it easier to use, a set of macros have been defined - e.g.,
OPAL_PROC_ON_LOCAL_SOCKET - that test the specific bit. These can be used in
the code base to clearly indicate which sense of locality is being considered.

All locations in the code base that looked at the current proc_t field have
been changed to use the new macros.

Also modify the orte_ess modules so that each returns a uint8_t (to match the
ompi_proc_t field) that contains a complete description of the locality of this
proc. Obviously, not all environments will be capable of providing such detailed
info. Thus, getting a "false" from a test for "on_local_socket" may simply
indicate a lack of knowledge.

This commit was SVN r20496.
2009-02-10 02:20:16 +00:00
Rainer Keller
64aa384745 - Fix comment to reflect current code.
This commit was SVN r20479.
2009-02-09 00:52:57 +00:00
Shiqing Fan
8086bb1a1b SIGSTOP and SIGTSTP are not supported on Windows. But they have to be defined anyway, although they are not used for Windows.
This commit was SVN r20453.
2009-02-05 17:02:34 +00:00
Shiqing Fan
ff7ca43dd1 Update two configuration files for windows build.
This commit was SVN r20450.
2009-02-05 16:39:40 +00:00
Shiqing Fan
7d2d6b16b1 A fix for windows mainly, adding BEGIN/END_C_DECLS pairs.
This commit was SVN r20448.
2009-02-05 16:35:58 +00:00
Jeff Squyres
67a5374a61 Re CID 1180: Actually, it would be better to also print something in
the case of an error, too...

This commit was SVN r20443.
2009-02-05 15:26:44 +00:00
Jeff Squyres
598e530de9 Fix CID 1180: ensure to check the output from snprintf, since we pass
it to write().

This commit was SVN r20442.
2009-02-05 15:24:48 +00:00
Jeff Squyres
9c2a6da128 Remove errant '>'. How on earth did that work at all?
This commit was SVN r20416.
2009-02-03 23:21:34 +00:00
Jeff Squyres
35c5e28a8e Up to SVN r20383
This commit was SVN r20384.

The following SVN revision numbers were found above:
  r20383 --> open-mpi/ompi@e0638c84c8
2009-01-29 17:59:04 +00:00
Jeff Squyres
bb3d258562 Round up a few places where PATH_MAX was used instead of
OMPI_PATH_MAX.  Thanks to Andrea Iob for the bug report.

This commit was SVN r20360.
2009-01-27 22:57:50 +00:00
Ralph Castain
c92f906d7c Move the daemon collectives out of the ODLS and into the GRPCOMM framework. This removes the inherent assumption that the OOB topology is a tree, thus allowing different grpcomm/routed combinations to implement collectives appropriate to their topology.
This commit was SVN r20357.
2009-01-27 19:13:56 +00:00
Rolf vandeVaart
1872a7b75d This change allows the trunk to be compiled with Sun
Studio compilers again. It has been broken since
1/14/2009 when some changes exposed a bug in autoconf
and how it handles support for the restrict keyword.
Basically, Sun Studio C supports the restrict keyword
but Sun Studio C++ does not.

I am also pursuing a fix with the autoconf folks, but
this change was needed to get things building again.

This commit was SVN r20351.
2009-01-26 20:13:44 +00:00
Jeff Squyres
f3b1432260 Fixes trac:1618: ensure to check to see if the symbol RTLD_NEXT exists
before trying to use it (e.g., it doesn't seem to exist on Cygwin).

This commit was SVN r20343.

The following Trac tickets were found above:
  Ticket 1618 --> https://svn.open-mpi.org/trac/ompi/ticket/1618
2009-01-25 16:38:00 +00:00
Jeff Squyres
6d805eb0dd Ensure to not do the found_files stuff is --disable-dlopen is selected.
This commit was SVN r20320.
2009-01-22 16:46:02 +00:00
Jeff Squyres
58a25cae69 Fixes trac:1271: make the OPAL MCA base read the list of MCA DSO filenames
''once'' and keep the names in an argv-style array.  Each time we go
to open a framework, we just scan that array rather than re-reading
all the filenames from the filesystem.

This commit was SVN r20309.

The following Trac tickets were found above:
  Ticket 1271 --> https://svn.open-mpi.org/trac/ompi/ticket/1271
2009-01-21 22:27:05 +00:00
Josh Hursey
fca3c6e571 Fix the BLCR configuration when explicitly disabling it.
It happened that if we supplied:
 --with-ft=cr --without-blcr
then BLCR would be loaded, due to a logic break in the old m4.

Now this works approprately. This should be moved to v1.3.1

This commit was SVN r20296.
2009-01-19 20:21:58 +00:00
Jeff Squyres
4520b00547 Fixes trac:1587: also check the mca component struct framework and
component name against the filename and ensure that they match.
Ignore the component if they do not.

This commit was SVN r20291.

The following Trac tickets were found above:
  Ticket 1587 --> https://svn.open-mpi.org/trac/ompi/ticket/1587
2009-01-17 12:53:21 +00:00
Ralph Castain
88a0af9726 Revise the way we output resolved hostnames to make life easier for the Eclipse folks. Store aliases for individual nodes (only when requested to show resolved hostnames) and then report them out as part of the display-map option.
This commit was SVN r20284.
2009-01-15 18:11:50 +00:00
Jeff Squyres
d1c6f3f89a * Fix a truckload of Cisco copyrights to be the same as the rest of
the code base.
 * Fix a few misspellings in other copyrights.

This commit was SVN r20241.
2009-01-11 02:30:00 +00:00
Ralph Castain
17e1911afa Remove unneeded include file
This commit was SVN r20204.
2009-01-05 19:20:02 +00:00
Tim Mattox
a5efe3ed77 Refs trac:868, #869
The fix for #868, r14358, introduced an (unneeded?) inconsitency...
For Mac OS X systems, inttypes.h will always be included with opal_config.h,
and NOT included for non-Mac OS X systems.  For developers using Mac OS X,
this masks the need to include inttypes.h or more properly opal_stdint.h.

This changeset corrects one of these oopses.  However, the underlying problem
still exists.  Moving the equivelent of r14358 into opal_stdint.h from
opal_config_bottom.h might be the "right" solution, but AFAIK, we would then
need to replace each direct inclusion of inttypes.h with opal_stdint.h to
properly address tickets #868 and #869.

This commit was SVN r20196.

The following SVN revision numbers were found above:
  r14358 --> open-mpi/ompi@dce72aab70

The following Trac tickets were found above:
  Ticket 868 --> https://svn.open-mpi.org/trac/ompi/ticket/868
2009-01-04 05:09:18 +00:00
Jeff Squyres
ad7cfe63a3 Fix CID 1180: check for negative return from snprintf.
This commit was SVN r20192.
2009-01-03 15:33:54 +00:00
Jeff Squyres
df3a304447 Fix CID 1182: ensure to check return of read() for failure.
This commit was SVN r20191.
2009-01-03 15:30:56 +00:00
Jeff Squyres
ae7dfdd0e0 Fix CID 1136: fix a small memory leak.
This commit was SVN r20188.
2009-01-03 15:12:16 +00:00
Jeff Squyres
c23f8e3981 Fix CIDs 1183-1186 (same as r20186 -- just missed the fact that there
were several more CIDs on the same source line before I committed).

This commit was SVN r20187.

The following SVN revision numbers were found above:
  r20186 --> open-mpi/ompi@423cce4b0a
2009-01-03 14:58:07 +00:00
Jeff Squyres
423cce4b0a Fix CID 1187: use PRIu64 instead of %lu for printing a uint64_t.
This commit was SVN r20186.
2009-01-03 14:55:08 +00:00
Tim Mattox
f911b1a63d Fix a few code comments in the new ompi-top functionality.
This commit was SVN r20166.
2008-12-22 22:36:38 +00:00
Ralph Castain
7787f84540 Per the earlier RFC and some discussion at the Dec ORTE design meeting, add the ompi-top tool and all its supporting infrastructure. This includes a new OPAL pstat framework and data type, currently with rather weak support for Mac OSX and pretty complete support for Linux. The Sun team promised to add Solaris support as well.
Also, per chat with Jeff, modified the Makefile.am's of a few orte tools so that they were consistent in the way we generate the ompi-equivalent cmds.

This commit was SVN r20165.
2008-12-22 20:23:05 +00:00
George Bosilca
80fd24c948 Small cleanups: remove an unused dependency to signal.h and include
output.h.

This commit was SVN r20155.
2008-12-18 22:39:49 +00:00
George Bosilca
24e191a076 Update the MIPS atomics. We can now compile with gcc and Pathscale.
This commit was SVN r20154.
2008-12-18 22:38:31 +00:00
Josh Hursey
ce8d18bfda This commit changes the use of the deprecated cr_request_file() to use the cr_request_checkpoint() interface to BLCR. Additional configure checks are added to use the best available checkpointing interface available for the BLCR installed on the system (default: cr_request_checkpoint()).
This commit fixes trac:1691

Thanks to Matthias Hovestadt for identifying this issue.

This commit was SVN r20114.

The following Trac tickets were found above:
  Ticket 1691 --> https://svn.open-mpi.org/trac/ompi/ticket/1691
2008-12-11 00:08:34 +00:00
Tim Mattox
4fa13a1a4d Fix two typos inside of comments.
This commit was SVN r20112.
2008-12-10 21:18:13 +00:00
Shiqing Fan
5ae5f0e173 - 4/4 commit for Windows Visual Studio and CCP support:
unnecessary clean up to non windows related files (within ifdef __WINDOWS__).

This commit was SVN r20111.
2008-12-10 21:13:27 +00:00
Shiqing Fan
20cea164db - 3/4 commit for Windows Visual Studio and CCP support:
corrections to non-windows files (but within ifdef __WINDOWS__)
  type casts, event library for windows use win32. 
  in orte runtime, add windows sockets handling and object construction.

This commit was SVN r20110.
2008-12-10 21:13:10 +00:00