Pages: « Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 Next »
Summary Period: 2003-01-28 to 2002-12-05 (Commits 1817-1866 of 4516)
removed development files
44 lines of code changed in:
check-in test
2 lines of code changed in:
another check-in test
4 lines of code changed in:
*** empty log message ***
2 lines of code changed in:
removed development items
1 lines of code changed in:
revised prompt for input file and/or program options. Program options are
can not be passed at the user prompt for multi-processor runs
72 lines of code changed in:
passing new -decomp_method option overrides previous settings
17 lines of code changed in:
better error handling in setting up free sets
30 lines of code changed in:
corrected FirstConfigure for graph decomposition
4 lines of code changed in:
merging changes from BRANCH_atom_1
116 lines of code changed in:
merging changes and conflicts from BRANCH_atom_1
4060 lines of code changed in:
merging changes and conflicts from BRANCH_atom_1
1729 lines of code changed in:
added enum for ductile fracture model
7 lines of code changed in:
check-in test
2 lines of code changed in:
check-in test
1 lines of code changed in:
Easing more conflicts between namespaces std and Tahoe and removing
SurfacePotential constructors using input streams from the fracture_interface.
23 lines of code changed in:
Reducing the memory fracture_interface uses. Most large arrays'
memory is controlled externally except for the coordinates which
(for the moment at least) get copied in ModelManagerT.
24 lines of code changed in:
Getting rid of casting away const in the fracture_interface parts of
the code since its version of ElementSupport() isn't constant for the
element groups in the fracture_interface.
9 lines of code changed in:
Argument iGroups in function SetElementGroupsNeeded should be passed
by reference instead of value.
4 lines of code changed in:
New Macros to compile with gcc 3.2.
Inspired from GNU.linux.macros
30 lines of code changed in:
corrected bug in accessing beyond last chunk
2 lines of code changed in:
revised exception handling
3 lines of code changed in:
ignore links to header files
2 lines of code changed in:
comment out redefinition of MY_CFLAGS
4 lines of code changed in:
added kModSmithFerrante to TypeT enum
11 lines of code changed in:
added some flags to help convergence
35 lines of code changed in:
set point size to 3.0
5 lines of code changed in:
set default window size
2 lines of code changed in:
fix namespace issue with StringT
12 lines of code changed in:
Add hexagonal lattice and cubic lattice.
Changed data files accordingly.
Correct bug in sorting feature.
Problem:
The hexagonal lattice is not so regular.
Is it a numerical error or a real problem in implementation?
400 lines of code changed in:
resolve geometry to kPoint if number of element nodes = 1 for nsd = 1, 2, 3
6 lines of code changed in:
removed some unneeded MPI code
3 lines of code changed in:
Deriving LJTr2D from CBLatticeT so that it can work with more than
nearest-neighbor interactions. All the energy, stress, and moduli
routines have been replaced to use CBLatticeT's bookkeeping. It still
passes the benchmark.
89 lines of code changed in:
Changing base class of LJTr2DT from NL_E_RotMat2DT
to NL_E_Mat2DT. The angle of rotation of the undeformed lattice will become
an input parameter to LJTr2DT.
8 lines of code changed in:
added access to OutputSetT's
13 lines of code changed in:
group label + 1
2 lines of code changed in:
src/dataio/output/OutputSetT.h
10 lines of code changed in:
write group number of main output if the rank check code isn't silent
9 lines of code changed in:
added "locks" on the residual and the global stiffness matrix, which
control when values can be assembled into either. Previously, both
could be written to at any time. Most solvers now support:
FormRHS: write access to both RHS and LHS
FormLHS: write access to LHS
This means you can assemble values into both the RHS and LHS during the
call to FormRHS
154 lines of code changed in:
corrected output of restart information. Missing a newline
2 lines of code changed in:
Updates to the fracture_interface parts of the code. Most of the
changes are in the various output routines and a few other minor
fixes.
278 lines of code changed in:
minor revisions to output
6 lines of code changed in:
scripts to handle build configuration:
(1) MakeConfigHeaderFile.pl - reads a configuration file and generates a C
header file that #define's pre-processor macros enabling each option.
(2) MakeConfigMakeFile.pl - reads a config file and the header generated by
MakeConfigHeaderFile.pl and write a make-format file which defines a
macro listing the active directories
Neither script overwrites exiting output.
212 lines of code changed in:
macros for the GCC 3.3 release of the GNU compilers for Darwin
107 lines of code changed in:
initial check-in of script to convert a config file to a header file
with user-definable configurations
116 lines of code changed in:
start incorporation of CommManagerT
309 lines of code changed in:
decomposition type enum moved to PartitionT
22 lines of code changed in:
replaced static Min function with conditional expression
4 lines of code changed in:
filling out the CommunicatorT (MPI wrapper) and adding classes to
handle data associated with different types of communications
386 lines of code changed in:
(1) added documentation
(2) added DecompTypeT flag
243 lines of code changed in:
Pages: « Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 Next »
Generated by StatCVS 0.2.2