92503eb2cd
- small boxes fft variables placed in a separate datatype, and removed from fft type
...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12671 c92efa57-630b-4861-b058-cf58834340f0
2016-08-06 09:09:42 +00:00
b2f8f3f456
- fixed small inconsistency (not really a bug)
...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12631 c92efa57-630b-4861-b058-cf58834340f0
2016-08-01 10:36:46 +00:00
de2a01c4e6
- Task groups variables moved form FFT type to a new data type
...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12626 c92efa57-630b-4861-b058-cf58834340f0
2016-07-29 14:25:22 +00:00
02ce4df8d6
make.sys => make.inc. All other codes using make.sys should modify their build
...
mechanism accordingly.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12539 c92efa57-630b-4861-b058-cf58834340f0
2016-06-25 07:26:50 +00:00
f2812ffd7d
- remove a dead-lock bug
...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12538 c92efa57-630b-4861-b058-cf58834340f0
2016-06-24 16:02:12 +00:00
02b04a53f9
- adding poor man performance measurments for: Floating point, network bandwidth, network latency
...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12534 c92efa57-630b-4861-b058-cf58834340f0
2016-06-23 16:27:37 +00:00
df30a665aa
Explicit preprocessing produces *_tmp.f90 files instead of *.F90 files, to
...
prevent trouble with OS-X. May or may not work (it won't unless configure
is updated: please somebody with v.2.63 of autoconf do it), may turn out to
be obsolete anyway.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12364 c92efa57-630b-4861-b058-cf58834340f0
2016-04-27 14:53:38 +00:00
9e024fe28c
Obsolete __ESSL preprocessing flag removed (there are no more IBM AIX machines
...
for scientific computing), only __LINUX_ESSL left
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12358 c92efa57-630b-4861-b058-cf58834340f0
2016-04-25 20:05:24 +00:00
0359fd948f
- adding timing for matrix multiplication,
...
and an estimation of the sustained GFlops for the given matrix size
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12132 c92efa57-630b-4861-b058-cf58834340f0
2016-02-14 22:20:11 +00:00
707e58243f
TDDFPT routines, which compute the response density, now use the original PH routines, which allows us to delete the duplicated parts of the code. Some documentation of the PH routines.
...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12120 c92efa57-630b-4861-b058-cf58834340f0
2016-02-14 12:41:19 +00:00
1f2cd6fa4a
- adding command line parameter "-n" to specify the matrix size
...
- adding a short explanation.
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12119 c92efa57-630b-4861-b058-cf58834340f0
2016-02-13 19:30:39 +00:00
023d18237b
- Scalapack context parameter added to the linear algebra descriptor
...
- Linear Algebra Library self testing program, now performs
a matrix diagonalization and print out diagonalization timing.
It can be useful to select the best low level library and/or the
best number of processors to be used for linear algebra.
More validation, auto-testing and performance assesment will follow
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12114 c92efa57-630b-4861-b058-cf58834340f0
2016-02-12 09:41:25 +00:00
b8b48f8df6
- new Linear Algebra library
...
git-svn-id: http://qeforge.qe-forge.org/svn/q-e/trunk/espresso@12053 c92efa57-630b-4861-b058-cf58834340f0
2016-01-23 16:34:07 +00:00
ccavazzoni