1
1

30 Коммитов

Автор SHA1 Сообщение Дата
Rolf vandeVaart
8cc6bef090 Refs open-mpi/ompi#627. Fix support for multi-threads with CUDA 7.0 2015-08-10 10:22:45 -04:00
Rolf vandeVaart
1f32fa21ae Fix arguments to error message, remove tabs and trailing spaces 2015-07-23 10:02:45 -04:00
Rolf vandeVaart
7703c96496 Add a workaroud for issue in libcuda.so library 2015-07-22 11:35:27 -04:00
Rolf vandeVaart
ae0f3cfee7 Make explicit call to initalize MCA parameters in common CUDA code. This allows us to view them with ompi_info and possibly modify with tools interface 2015-07-09 12:51:55 -04:00
Ralph Castain
869041f770 Purge whitespace from the repo 2015-06-23 20:59:57 -07:00
Rolf vandeVaart
b3e4ae71d5 Fix finalize code when cuda support is not fully initialized 2015-05-21 13:42:22 -04:00
Rolf vandeVaart
b260dc4228 Cleanup interface that handles events. No functional changes 2015-04-28 15:15:24 -04:00
Rolf vandeVaart
030dddde23 Remove unneeded copy and minor clean up 2015-04-22 13:17:16 -04:00
Rolf vandeVaart
b7913836fc Initialize variables for safety 2015-04-09 12:58:55 -04:00
Rolf vandeVaart
d6d7184703 Enhance verbose message 2015-04-09 12:29:09 -04:00
Rolf vandeVaart
dfb7e00ef5 Make sure context is still around when doing some other cleanup 2015-03-24 16:47:40 -04:00
Rolf vandeVaart
49b5eb6c91 Fix missing initialization of variable 2015-03-10 10:33:27 -04:00
Jeff Squyres
1995f6beba cuda: convert to opal_dl interface 2015-03-09 08:18:13 -07:00
Rolf vandeVaart
237c268a09 Add extra check during cleanup to make sure we really should clean up the CUDA resources. 2015-03-06 13:12:19 -05:00
Howard Pritchard
bf89131f9e add owner files to opa/ompi/orte mca directories
This commit adds an owner file in each of the component directories
for each framework.  This allows for a simple script to parse
the contents of the files and generate, among other things, tables
to be used on the project's wiki page.  Currently there are two
"fields" in the file, an owner and a status.  A tool to parse
the files and generate tables for the wiki page will be added
in a subsequent commit.
2015-02-22 15:10:23 -07:00
Nathan Hjelm
1e518504e4 btl/smcuda: update for BTL 3.0 interface 2015-02-13 11:46:37 -07:00
Rolf vandeVaart
66f6026214 Improve error message to help user figure out what to do 2015-01-16 13:55:27 -05:00
rolfv
f471b09ae9 Add support for CUDA Unified memory. Basically, add a new flag and disable some
optimizations when that flag is detected.  Lightly reviewed by bosilca.
2014-12-10 05:46:00 -08:00
George Bosilca
324e43909d Enable CUDA support on Mac OS X. 2014-11-20 13:51:10 -06:00
Jeff Squyres
9334abc474 Makefile: fix problems with static linking
Avoid a problem with double-derefence of a variable macro name (i.e.,
a macro with part of its name from an AC_SUBST, such as
```$(foo@BAR@baz)```.

In what might be a bug in Automake 1.14.1, if you do a pattern like
this:

```makefile
lib_LTLIBRARIES = lib@A_PREFIX@a_lib.la
noinst_LTLIBRARIES = lib@A_PREFIX@a_noinst.la

lib@A_PREFIX@a_lib_la_SOURCES = a.c

lib@A_PREFIX@a_noinst_la_SOURCES = $(lib@A_PREFIX@a_lib_la_SOURCES)
```

Then in the resulting Makefile, the value of
```$(lib@A_PREFIX@a_lib_la_OBJECTS)``` will be *blank* (when it really
should be ```a.o```).

To workaround this potential bug, I've simply avoided doing
double-derefences like this, and effectively set the second
```_SOURCES``` line equal to ```a.c``` (just like the first
```_SOURCES``` line).

Fixes #250.
2014-10-24 16:27:54 -07:00
Jeff Squyres
c22e1ae33b configury: new OPAL_SET_LIB_PREFIX/ORTE_SET_LIB_PREFIX macros
These two macros set the prefix for the OPAL and ORTE libraries,
respectively.  Specifically, the OPAL library will be named
libPREFIXopen-pal.la and the ORTE library will be named
libPREFIXopen-rte.la.

These macros must be called, even if the prefix argument is empty.

The intent is that Open MPI will call these macros with an empty
prefix, but other projects (such as ORCM) will call these macros with
a non-empty prefix.  For example, ORCM libraries can be named
liborcm-open-pal.la and liborcm-open-rte.la.

This scheme is necessary to allow running Open MPI applications under
systems that use their own versions of ORTE and OPAL.  For example,
when running MPI applications under ORTE, if the ORTE and OPAL
libraries between OMPI and ORCM are not identical (which, because they
are released at different times, are likely to be different), we need
to ensure that the OMPI applications link against their ORTE and OPAL
libraries, but the ORCM executables link against their ORTE and OPAL
libraries.
2014-10-22 10:32:19 -07:00
rolfv
697b18db63 Making async copy the default 2014-10-03 06:42:18 -07:00
Rolf vandeVaart
399dc3db43 Code to check for managed memory. Configure support also.
This commit was SVN r32801.
2014-09-26 16:24:45 +00:00
Rolf vandeVaart
35858f837a Revert r32713. Have different code for this.
This commit was SVN r32800.

The following SVN revision numbers were found above:
  r32713 --> open-mpi/ompi@9a2bab0e27
2014-09-26 14:56:18 +00:00
Rolf vandeVaart
9a2bab0e27 Add support for detecting CUDA managed memory. Disabled for now.
This commit was SVN r32713.
2014-09-11 21:07:17 +00:00
Rolf vandeVaart
c53c981506 Fix initialization and cleanup code for CUDA-aware code. Eliminates all resource leaks.
This commit was SVN r32512.
2014-08-12 19:41:46 +00:00
Rolf vandeVaart
876232e6d0 Fix previous checkin.
This commit was SVN r32468.
2014-08-08 18:58:25 +00:00
Rolf vandeVaart
ac16af0bff Small fix for case where no libcuda.so.1 is found.
This commit was SVN r32461.
2014-08-08 16:05:19 +00:00
Ralph Castain
db89071dc2 Cleanup the moved component's Makefile.am to use the opal instead of ompi directories
This commit was SVN r32370.
2014-07-31 04:41:04 +00:00
Ralph Castain
552c9ca5a0 George did the work and deserves all the credit for it. Ralph did the merge, and deserves whatever blame results from errors in it :-)
WHAT:    Open our low-level communication infrastructure by moving all necessary components (btl/rcache/allocator/mpool) down in OPAL

All the components required for inter-process communications are currently deeply integrated in the OMPI layer. Several groups/institutions have express interest in having a more generic communication infrastructure, without all the OMPI layer dependencies.  This communication layer should be made available at a different software level, available to all layers in the Open MPI software stack. As an example, our ORTE layer could replace the current OOB and instead use the BTL directly, gaining access to more reactive network interfaces than TCP.  Similarly, external software libraries could take advantage of our highly optimized AM (active message) communication layer for their own purpose.  UTK with support from Sandia, developped a version of Open MPI where the entire communication infrastucture has been moved down to OPAL (btl/rcache/allocator/mpool). Most of the moved components have been updated to match the new schema, with few exceptions (mainly BTLs where I have no way of compiling/testing them). Thus, the completion of this RFC is tied to being able to completing this move for all BTLs. For this we need help from the rest of the Open MPI community, especially those supporting some of the BTLs.  A non-exhaustive list of BTLs that qualify here is: mx, portals4, scif, udapl, ugni, usnic.

This commit was SVN r32317.
2014-07-26 00:47:28 +00:00