| #
df4a11de
|
| 18-Jun-2011 |
Barry Smith <bsmith@mcs.anl.gov> |
cleaned up the BasicVersion to only print information once, to not print rms crap and to print sum of rates over all processes
Hg-commit: 5dee010063a8699105ed16fdda2badf05a193b21
|
| #
639d86e3
|
| 13-Jan-2012 |
Sean Farley <sean@mcs.anl.gov> |
Merge
Hg-commit: 0fb617426806c222fb6fb5e47365b6c3688c614c
|
| #
d39137a2
|
| 12-Jan-2012 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: d540875a63b3c979c82b6a7fa7f062571e906d10
|
| #
247e2d92
|
| 06-Jan-2012 |
Jed Brown <jed@59A2.org> |
Mass revert file modes that were made executable in f6e61ee9c329
Hg-commit: 116d90d0b66125cee0db3e7351047057e756baf5
|
| #
832f7384
|
| 01-Jan-2012 |
Matthew G Knepley <knepley@gmail.com> |
Merge
Hg-commit: ee8d91ae0580d1d4fa92be38a18d40c838b9a257
|
| #
14e519b0
|
| 31-Dec-2011 |
Mark F. Adams <mark.adams@columbia.edu> |
changed MatMatMult to MatTranposeMatMult.
Hg-commit: f6e61ee9c329774cd5343eea1ee6b7432be17c3f
|
| #
1c2f5707
|
| 05-Dec-2011 |
Peter Brune <brune@mcs.anl.gov> |
commit after merge
Hg-commit: 439dfdbd682db1b878aeffdee878abfc28af6478
|
| #
0d04baf8
|
| 01-Dec-2011 |
Barry Smith <bsmith@mcs.anl.gov> |
missing float.h include in streams codes
some other stuff I forgot
Hg-commit: 85fd59118ddac7c1b30717712df0ac5319a563f1
|
| #
ce2f17d1
|
| 10-May-2011 |
Dmitry Karpeev <karpeev@mcs.anl.gov> |
Merge.
Hg-commit: 8ad722e18529293bde53db7f4fb83823cee61dc0
|
| #
d6651a8c
|
| 09-May-2011 |
Victor Minden <victorminden@gmail.com> |
commit after merge
Hg-commit: a935ccf20f02d5d32e7e4e31e093474dd9731a2f
|
| #
dcd07f4f
|
| 09-May-2011 |
Barry Smith <bsmith@mcs.anl.gov> |
commit after merge
Hg-commit: cd65c0daecf15ddfa806fd79009fb5a0e5d566bf
|
| #
106b1c52
|
| 09-May-2011 |
Jed Brown <jed@59A2.org> |
Place MPI_Barrier in the necessary places for the STREAM benchmarks.
Note that it is possible for the system to interrupt the process between
the call to MPI_Barrier and getting the current time. Th
Place MPI_Barrier in the necessary places for the STREAM benchmarks.
Note that it is possible for the system to interrupt the process between
the call to MPI_Barrier and getting the current time. This means that
scenarios like
MPI_Barrier(comm);
t1 = time()
work()
MPI_Barrier(comm);
elapsed = time() - t1;
may have each work() running concurrently, or they may run separately,
paired with other active processes on the system. If the other processes
do not have similar memory demands, the aggregate throughput can be
reported as anomalously high. Therefore it is necessary to bracket the
computation with MPI_Barrier.
Hg-commit: def515087559f3ce819b0b673d0ceabaa58d5a03
show more ...
|
| #
c6db04a5
|
| 14-Mar-2011 |
Jed Brown <jed@59A2.org> |
Use #include <header.h> instead of #include "header.h" when there is no intent to search the current directory for header.h
Hg-commit: bd7216e80cc035071a5136364ab9d04bf9f41a07
|
| #
20803bd0
|
| 09-Feb-2011 |
Matthew G Knepley <knepley@gmail.com> |
Merge
Hg-commit: 674bc8187eb8d0423d5e5713bd4ba8a46a2b8020
|
| #
01a79839
|
| 03-Feb-2011 |
Barry Smith <bsmith@mcs.anl.gov> |
added MPI to BasicVersion of streams to get some measure of parallel memory bandwidth
fixed OpenMPI version of streams to actually build with MPI
Hg-commit: 190640f459cfcd17b88515b33cbfb760a32096c2
|
| #
5d28107e
|
| 19-Nov-2009 |
Barry Smith <bsmith@mcs.anl.gov> |
added streams and openmp streams benchmarks and makefile to PETSc-dev
Hg-commit: abdebb2ffacde4fcbdeefe3300138fbdf8c97afa
|