1
1
Граф коммитов

969 Коммитов

Автор SHA1 Сообщение Дата
Ralph Castain
abe7dd8274 Cleanup the dss by removing unused functions
This commit was SVN r26644.
2012-06-23 21:20:09 +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
Jeff Squyres
06c4317dd4 Ensure to include external.h in the tarball.
This commit was SVN r26610.
2012-06-15 16:29:21 +00:00
Jeff Squyres
6760840ebb Fix builds with the external hwloc component when we use the
hwloc/openfabrics-verbs.h helper header file.

This commit was SVN r26603.
2012-06-14 19:00:57 +00:00
Terry Dontje
6d7cf4a0e5 corrected picl dependency checking to occur in the hwloc.m4 instead of Makefile.am
This commit was SVN r26595.
2012-06-12 14:47:05 +00:00
Ralph Castain
cee5a75d19 Revert the default configuration to no orte progress thread and no libevent thread support until we can get more of the kinks ironed out.
This commit was SVN r26593.
2012-06-11 20:52:28 +00:00
Nathan Hjelm
ceee4bcb0d libevent2019: libevent_pthreads.la is never built. don't include it
This commit was SVN r26570.
2012-06-07 19:22:45 +00:00
Jeff Squyres
ba040e3a42 Upgrade hwloc from 1.3.2+patches to 1.4.2+patches.
This commit was SVN r26566.
2012-06-07 16:24:46 +00:00
Ralph Castain
5876496f4c Enable orte progress threads and libevent thread support by default
This commit was SVN r26565.
2012-06-07 04:25:00 +00:00
Jeff Squyres
8d161af059 Move hwloc_cpuset_t prettyprint routines down into the hwloc base:
* opal_hwloc_base_cset2str(): Make a human-readable string of a
   hwloc_cpuset_t (e.g., socket 2[core 3[hwt 1]])
 * opal_hwloc_base_cset2mapstr(): Make a map-like string of a
   hwloc_cpuset_t (e.g., [B./..])

This commit was SVN r26532.
2012-06-01 16:02:18 +00:00
Jeff Squyres
05807ef19a Record the upstream SVN commit
This commit was SVN r26525.
2012-05-29 23:44:25 +00:00
Jeff Squyres
b3fbb0a2d5 Ensure to actually exit the non-voice function, even in non-debug
builds (i.e., where assert() is preprocessed away).

This commit was SVN r26524.
2012-05-29 23:41:23 +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
Jeff Squyres
551b53dd89 Keep the help string less than 509 characters so that compilers don't complain.
This commit was SVN r26514.
2012-05-29 18:43:04 +00:00
Jeff Squyres
96901d9503 Slightly change the wording in the help message for the
hwloc_base_mem_alloc_policy MCA parameter to be more explicit.

This commit was SVN r26512.
2012-05-29 18:08:39 +00:00
Jeff Squyres
5391e98d56 Remove some generated files.
This commit was SVN r26495.
2012-05-25 00:45:38 +00:00
Ralph Castain
32337d0d5c Well, we dont need the -levent any more
This commit was SVN r26487.
2012-05-24 01:54:37 +00:00
Ralph Castain
63a23f1f90 Get the event lib built correctly - missing the OMPI-specific pieces from Makefile.am!
Copmlete the symbol renaming to cover new symbols in the updated version.

This commit was SVN r26486.
2012-05-24 00:27:41 +00:00
Ralph Castain
1d41c6bb80 Some more libevent configure fixes. Ensure -levent gets added to the wrapper flags as the LANL machines can't seem to find it otherwise. Remove the duplicate evaluation of --visibility in the libevent area as it was already done by opal for everyone. Set some more ignores.
This commit was SVN r26485.
2012-05-23 21:26:54 +00:00
Ralph Castain
5821e63e27 Remove built files and update ignores
This commit was SVN r26482.
2012-05-23 18:17:29 +00:00
Ralph Castain
8a2ca3d96f Delete built files
This commit was SVN r26481.
2012-05-23 18:14:26 +00:00
Ralph Castain
11d5a31b1e Update ignores, remove build result file from repo
This commit was SVN r26460.
2012-05-21 00:59:04 +00:00
Ralph Castain
40317c0290 Remove the old event lib version - new one seems to be working just fine.
This commit was SVN r26459.
2012-05-21 00:57:47 +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
1ce59d08b5 Continue cleaning up the libevent ignores
This commit was SVN r26455.
2012-05-19 16:11:24 +00:00
Ralph Castain
1826b513ee Add missing windows file
This commit was SVN r26433.
2012-05-12 02:05:25 +00:00
Ralph Castain
09f413025b As per the RFC, upgrade OMPI to libevent 2.0.19. Leave the 2.0.13 release in the system, but inactive, for now in case we discover a need to rollback.
This commit was SVN r26431.
2012-05-11 01:05:36 +00:00
Jeff Squyres
1d7fef001c Record the upstream hwloc commit that we've committed here in the OMPI
tree

This commit was SVN r26422.
2012-05-10 12:15:23 +00:00
Jeff Squyres
9c9d7e77df Commit a fix for hwloc -- still checking with upstream to see if this
will be the final solution.  But I'm committing it now so that
Oracle's Solaris Studio builds can resume.

The issue is that the C++ bindings are now (eventually) including
<hwloc.h>.  We use !__hwloc_inline__ and #define it to an appropriate
value at compile-time.  The issue is that when we're compiling C++
code, we should just set !__hwloc_inline__ to "inline", because that's
a keyword in the C++ language (as opposed to !__inline__, or
somesuch).

This commit was SVN r26418.
2012-05-09 21:03:45 +00:00
Jeff Squyres
de4bbacd13 It turns out that we can't always include the hwloc OpenFabrics verbs
helper file, even if we find that the system has <infiniband/verbs.h>.
The reason is because there are some inline functions in that verbs
helper file that invoke ibv_* functions.  Some linkers (e.g., Solaris
Studio Compilers) will instantiate those static inline functions --
even if we don't use them -- and therefore we need to be able to
resolve the ibv_* symbols at link time.

But since -libverbs is only specified in places where we use other
ibv_* functions (e.g., the OpenFabrics-based BTLs), that means that
linking random executables can/will fail (e.g., orterun).

So instead, introduce a new #define: OPAL_HWLOC_WANT_VERBS_HELPER.  If
this macro is set to 1 before including opal/mca/hwloc/hwloc.h, then
you'll also get the hwloc OpenFabrics verbs helper header file (*if*
hwloc found <infiniband/verbs.h> -- otherwise, it'll #error).

This commit was SVN r26417.
2012-05-09 20:18:31 +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
Jeff Squyres
aba398ce09 Per RFC
(http://www.open-mpi.org/community/lists/devel/2012/04/10905.php), set
opal_cache_line_size via hwloc data, if we have it.
opal_cache_line_size will be set to an hwloc-inspired value by the end
of orte_init(), but will always have a safe value to use (i.e., a
default value 128) -- even before opal_init() has completed.

Default to the same value of 128 that Open MPI has used for several
years if a) we have no hwloc data, or b) we weren't able to find L2
objects in the hwloc data.

This commit was SVN r26322.
2012-04-24 17:31:06 +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
Mike Dubman
ff1c84c53f revert previous commit
This commit was SVN r26206.
2012-03-29 14:07:13 +00:00
Mike Dubman
43a5775e8a performance fix: set alignment for openib internal buffers
This commit was SVN r26205.
2012-03-29 14:00:08 +00:00
Jeff Squyres
028f471a20 Using the right env variable name helps!
This commit was SVN r26204.
2012-03-28 17:59:21 +00:00
Jeff Squyres
8a2df3311d Fixes trac:2812: check for env. markers indicating that we're in a
fakeroot.  If so, exit out of the pre-main hook immediately (without
calling functions such as stat, which will be replaced by fakeroot to
things that are not safe to call in a pre-main environment).

This commit was SVN r26203.

The following Trac tickets were found above:
  Ticket 2812 --> https://svn.open-mpi.org/trac/ompi/ticket/2812
2012-03-28 16:41:29 +00:00
Pavel Shamis
39a55df333 Adding exported libevent globabl variables to the opal_rename file.
Otherwise the varables case name conflicts.

This commit was SVN r26201.
2012-03-27 17:26:21 +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
6f6930eb66 Resolve infinite loop when -cpu-set is specified
This commit was SVN r26184.
2012-03-23 07:18:58 +00:00
Jeff Squyres
95148f3310 Don't force the use of libpci support in hwloc in the default case --
just let hwloc decide for itself.

This commit was SVN r26178.
2012-03-22 15:28:35 +00:00
Jeff Squyres
3bf038bb1c Per RFC from long ago:
http://www.open-mpi.org/community/lists/devel/2011/10/9784.php

Bring support for a DMTCP CRS module into the trunk.  See
http://dmtcp.sourceforge.net/ for a description of DMTCP.  Thanks to
the contribution from Alex Brick at Northeastern University, and all
the others up there who helped shepherd this into being ready to
submit.

This commit was SVN r26176.
2012-03-22 12:01:46 +00:00
Jeff Squyres
d30bbc2ef9 Fix an old issue: enable hwloc PCI detection except on SuSE 10 64 bit.
Worked with Oracle to verify that hwloc PCI detection is correctly
disabled on the Suse 10/64 bit platform and is enabled by default on
all other platforms.  The --[en|dis]able-hwloc-pci switch is also
available for manual override of the configure decision about hwloc
PCI support.

This commit was SVN r26175.
2012-03-22 11:30:57 +00:00
Jeff Squyres
ab543fce58 We have no common components in opal any more, so we can remove this directory.
This commit was SVN r26169.
2012-03-20 21:21:49 +00:00
Jeff Squyres
0322db7cde Bring over r4402 from hwloc trunk.
This commit was SVN r26165.

The following SVN revisions from the original message are invalid or
inconsistent and therefore were not cross-referenced:
  r4402
2012-03-19 16:39:54 +00:00
Jeff Squyres
aeca190744 Refs trac:3046: feedback from Brian -- don't set DYLD_LIBRARY_PATH.
This commit was SVN r26108.

The following Trac tickets were found above:
  Ticket 3046 --> https://svn.open-mpi.org/trac/ompi/ticket/3046
2012-03-07 13:12:22 +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
Jeff Squyres
f84c16bb65 Fixes trac:3043. Looks like some of the improvements to the hwloc132
hwloc component weren't reverse applied to the external hwloc
component.  Additionaly, if we add stuff to LDFLAGS/LIBS, we also may
need to append (DY)LD_LIBRARY_PATH (here in this configure process
only), otherwise future configure tests may fail because they can't
find libhwloc.so (e.g., if you --with-hwloc=/some/path, we need to add
/some/path/lib to (DY)LD_LIBRARY_PATH).

This commit was SVN r26082.

The following Trac tickets were found above:
  Ticket 3043 --> https://svn.open-mpi.org/trac/ompi/ticket/3043
2012-03-02 20:15:07 +00:00
Jeff Squyres
e77653511b Bring in upstream hwloc v1.3 branch SVN commit r4345
This commit was SVN r26048.

The following SVN revision numbers were found above:
  r4345 --> open-mpi/ompi@b6c2a5b602
2012-02-24 13:57:18 +00:00