/
ChangeLog
8640 lines (5364 loc) · 243 KB
/
ChangeLog
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
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
2007-06-26 19:51 sue
* [r7523] rename 0.8.2 manual
2007-06-20 23:46 willic3
* [r7330] Changed default for autoRotateSlipperyNodes to false,
since this adds additional computations that aren't needed.
2007-06-13 20:04 willic3
* [r7213] Made BC symmetric to match test for PyLith 1.0.
2007-06-13 16:40 willic3
* [r7201] Started creating tests for PyLith 0.8 that can be
compared directly against those for 1.0.
2007-06-13 13:48 willic3
* [r7193] Added extra digit to PETSc version number to get
configure working.
2007-05-10 17:32 knepley
* [r6839] Fixed name change
2007-05-10 00:38 leif
* [r6831] Fix for issue108: "change to VecScatterBegin/End".
2007-05-04 21:17 leif
* [r6787] Updated info files.
2007-05-04 20:46 leif
* [r6786] Updated Makefile.
2007-05-04 20:26 willic3
* [r6785] Removed parmetis .cfg file from tractest, since it won't
work, but added one for splittest. Viscosity values for
powertest have been changed.
2007-05-04 03:00 leif
* [r6783] Updated 'setup.py'.
2007-05-03 20:13 leif
* [r6765] ParMETIS partitioning is broken for traction BC; fail
with an error message when partitioner=parmetis and there is a
.traction file.
2007-05-03 15:46 willic3
* [r6764] Moved partitioner parameter to pylith3d list rather than
pylith3d.petsc. Added a Chaco .cfg file for testing, even though
this is already the default.
2007-05-03 15:42 willic3
* [r6763] Removed partitioner from PETSc arguments.
2007-05-03 00:05 leif
* [r6760] Using a Perl script, updated all shell scripts to use
the new property/facility/component layout. Removed shell
scripts that still used 'mpiexec' or had 'lithomop' in the
filename.
2007-05-02 20:25 leif
* [r6757] Removed kludgey support for old 'pl3dscan.xxx' and
'scanner.xxx' options.
2007-05-02 20:09 leif
* [r6756] Updated all *.cfg files to use the new
property/facility/component layout.
2007-05-02 19:29 sue
* [r6754] image prob solved using .ai image version
2007-05-02 19:20 sue
* [r6753] working image problem; pdf pg 62, figure b.1, runs off
page bottom no matter what I do. still troubleshooting
2007-05-02 18:26 sue
* [r6752] working image problem
2007-05-02 17:18 sue
* [r6751] minor edits
2007-05-02 06:55 leif
* [r6750] Updated PyLith for Pythia v0.8.1.3b5.
PetscCommandlineParser and PetscProperty are no longer
supported. Use Pyre command line syntax. Use the "petsc"
facility/component ("petsc.xxx=yyy") to set PETSc options.
2007-05-02 02:45 leif
* [r6748] Removed quotes around quadratureOrder values;
prestressAutoChangeElasticProperties ->
prestressAutoChangeElasticProps.
2007-05-02 02:28 leif
* [r6747] Mechanically converted remaining *.keyval files to *.cfg
files: keyval=`find . -name \*.keyval` for kv in $keyval; do
cfg=`dirname $kv`/`basename $kv .keyval`.cfg; svn cp $kv $cfg ;
echo "[pylith3d]" > $cfg; cat $kv >> $cfg ; svn rm $kv ; done
2007-05-02 02:04 leif
* [r6746] Removed all references to *.keyval files from the
manual; they are no longer read by PyLith.
2007-05-02 01:31 leif
* [r6745] Removed remaining 'runex' scripts, to avoid confusion.
2007-05-01 22:04 leif
* [r6743] Build fixes: install pylith3dapp.py ~ install examples ~
allow tarball to work without Pyrex installed ~ made sure
tarball is complete. Unfortunately, "make distcheck" does not
work, due to an egg-related issue.
2007-05-01 20:27 sue
* [r6741] a few edits; more tomorrow
2007-05-01 18:55 leif
* [r6739] Added one more PETSc config example.
2007-05-01 16:17 willic3
* [r6736] Minor format fix when writing out material model name.
2007-05-01 03:31 leif
* [r6735] Final draft of installation chapter.
2007-04-30 23:19 leif
* [r6732] Wrote PETSc installation section.
2007-04-28 01:15 leif
* [r6725] Temporarily disabled outputFile(). Without ${rank} in
the filenames, there is a race condition revolving around O_EXCL
with more than one processor.
2007-04-27 23:09 leif
* [r6723] Initial rewrite of installation chapter.
2007-04-27 23:03 sue
* [r6722] made journal no. standard format
2007-04-27 22:52 sue
* [r6721] various edits, spacing fixes, image positioning, and
long lines running off the page fixed
2007-04-27 21:47 willic3
* [r6720] Finished draft of materials. It's a little sketchy
toward the end.
2007-04-27 20:49 willic3
* [r6718] Added another reference to userguide. Material models
are complete through generalized Maxwell. Still need Z&T
formulation for power-law Maxwell.
2007-04-27 03:27 willic3
* [r6706] Mostly finished through power-law ESF.
2007-04-27 01:05 willic3
* [r6704] More work on materials. Nearly finished fixing things
through ESF formulation of linear Maxwell. Table and figure
references still need fixing. Still to do: Power-law ESF
Generalized Maxwell Power-law Maxwell using Z&T formulation
2007-04-26 00:11 sue
* [r6688] small changes
2007-04-25 21:31 willic3
* [r6687] Completely reworking materials section: Adding figure of
material models. Changing layout. Changing equations to be
consistent with what we're doing for 1.0. Adding new material
models. Section is still incomplete. I need to finish fixing
after the first subsection in the ESF section.
2007-04-25 02:42 willic3
* [r6681] Multiprocessor .cfg file.
2007-04-24 22:42 sue
* [r6678] finished up standardizing subheader format
2007-04-24 22:36 sue
* [r6677] standardized subsection headers in install file
2007-04-24 22:33 sue
* [r6676] fixed spacing in table and other minor edits
2007-04-24 22:05 sue
* [r6670] edits to tutorials incl standardizing quote marks and
subsection titles, and fixing over long lines
2007-04-24 21:48 willic3
* [r6664] Modified .cfg file to let ParMetis be the partitioner.
2007-04-24 21:44 willic3
* [r6663] Handled case where file could be empty due to
partitioning.
2007-04-24 21:43 willic3
* [r6662] Added 'partitioner' as a PETSc option of interest to
PyLith.
2007-04-24 21:15 willic3
* [r6660] Fixed end of section and created a revised table of
PETSc settings.
2007-04-24 20:38 sue
* [r6659] fixed problem with preface in preview mode in file
userguide.lyx which was causing lyx program to hang when edited,
e.g., turned off preview mode
2007-04-24 20:17 willic3
* [r6658] Updated reversenog tutorial for new utilities, new usage
of .cfg files, etc.
2007-04-24 20:16 sue
* [r6657] various edits, incl. resized table in Materials and
changed quotes to match other manuals
2007-04-24 19:40 willic3
* [r6655] Forgot to change version number.
2007-04-24 19:37 willic3
* [r6654] Updated splitcube tutorial for new utilities, new usage
of .cfg files, etc.
2007-04-24 15:50 willic3
* [r6647] Updated features for current version.
2007-04-23 03:04 willic3
* [r6631] Changed method for determining how many vertices per
face for traction BC.
2007-04-23 03:02 willic3
* [r6630] Changed method for determining how many vertices per
face for traction BC.
2007-04-20 20:33 willic3
* [r6624] Updated .cfg files.
2007-04-20 20:24 willic3
* [r6623] Deleted old shell scripts and added a couple of new .cfg
files.
2007-04-20 19:34 knepley
* [r6622] Small fixes
2007-04-20 17:36 willic3
* [r6620] Added .cfg file to launch multi-processor job. Removed
launcher info from pylith3d.cfg specific to Matt's setup.
2007-04-20 17:19 willic3
* [r6619] Put in changes to ntimdat array (removal of ndtot) that
had been done for fortran code.
2007-04-20 15:37 knepley
* [r6617] Minimally running 0.8 with new PETSc
2007-04-19 20:08 knepley
* [r6609] Fixed compile bug with new PETSc
2007-04-19 17:09 knepley
* [r6607] Trial integration into mainline
2007-04-19 04:11 willic3
* [r6605] Updated vistest examples for current code (no more
scanner), and fixed some timestep stuff.
2007-04-19 02:33 willic3
* [r6604] Removed ndtot, which was no longer being used, along
with corresponding entry in the ntimdat array.
2007-04-19 00:02 willic3
* [r6600] Checked in current version of documentation, which still
needs work.
2007-04-18 18:43 leif
* [r6596] Creating my own forward declarations of "petscmesh.h"
functions turns out to be a bad idea, because such functions can
have different linkage (C or C+) depending upon how PETSC was
configured.
2007-04-18 00:38 leif
* [r6595] Fixed VTK output.
2007-04-17 23:13 leif
* [r6594] Don't open xxx.ascii file when 'asciiOutput' is set to
'none'. Corrected "category" for output files.
2007-04-17 18:47 leif
* [r6591] Re-implemented memory-usage-report feature, but this
time with only a few lines of code. Factored-out
Python-list-to-C-array code. Added __get__() methods so that
public list/array properties are readable as well as writeable.
2007-04-17 17:53 willic3
* [r6587] Fixed incorrect format statement number.
2007-04-17 04:37 leif
* [r6586] Open output files once at the beginning of the run, and
close them once at the end (instead of repeatedly opening and
appending to them). This change elminates the need for the
CIT_FC_OPEN_APPEND Autoconf test and the FORTRAN_POSITION_APPEND
and FORTRAN_ACCESS_APPEND preprocessor macros. It also (quite
incidentally) made the Fortran routine viscos_setup() redundant.
Also, I fixed a bug I introduced. In the Pyrex call to
scan_timdat(), I wrote "len(time_units)" instead of
"sizeof(time_units)". D'oh! Mostly, the Pyrex calls are a
transcription/merging of the old C++ bindings into the Python
code, so hopefully this is the only bug of its kind.
2007-04-14 02:02 leif
* [r6570] Eliminated the bindings by converting the bulk of the
Python code to Pyrex.
2007-04-12 17:30 willic3
* [r6555] More changes to documentation.
2007-04-12 17:29 willic3
* [r6554] Did a little more on Green's functions, but it is
commented out because it still doesn't work.
2007-04-11 20:20 willic3
* [r6546] Started making changes for 0.8.2. There is still a fair
bit to do.
2007-04-10 21:26 willic3
* [r6540] Altered to use pythonTimestep and minor fix to time step
info.
2007-04-10 21:14 willic3
* [r6539] Updated tutorial to use *.cfg files and new methods of
dealing with filenames.
2007-04-10 20:25 willic3
* [r6538] Updated tutorial to use *.cfg files and new ways of
dealing with filenames.
2007-04-09 18:03 willic3
* [r6531] Fixed direction of normal (it was backwards).
2007-04-06 20:06 willic3
* [r6514] Added new 'prescomp' code to source list.
2007-04-06 20:05 willic3
* [r6513] Simple utility code to compute boundary-normal tractions
(pressures) over a specified set of faces.
2007-04-05 21:54 leif
* [r6509] Renamed almost all the variables in the Python code and
in the C++ bindings such that the variable names are consistent
with the Fortran code. E.g., rotationUnits -> rotation_units,
numberRotationEntries -> numrot. This makes the code much easier
to understand and much easier to follow as control flows from
Python through C++ into Fortran, and back again. For instance,
if I grep for 'numrot' (actually I use 'gid' from the 'mkid'
package), I now get matches in the Python and C++ code, and I
can easily see that 'numrot' is defined as the return value of
scan_skew(). Sample matches from the output of "gid numrot": #
---- snip ---- PyLith.py:558: self.numrot = pylith3d.scan_skew(
# [...] ../module/libpylith3d.cc:5319: scan_skew_f(&numrot,
../module/libpylith3d.cc:5336: << "numrot:" << numrot
../module/libpylith3d.cc:5341: return Py_BuildValue((char *)
"i", numrot); # [...] ../libpylith3d/scan_skew.f:31: subroutine
scan_skew(numrot,kr,rotation_units,skfile,ierr, # [...] # ----
snip ---- The light is so much better now!
2007-04-04 20:28 willic3
* [r6508] Minor optimization.
2007-03-30 21:05 leif
* [r6475] Merged all the Fortran bindings into a single file, in
preparation for the next step.
2007-03-30 18:22 leif
* [r6473] Removed 'unused' files.
2007-03-27 22:33 willic3
* [r6444] Made some changes related to generating Green's
functions.
2007-03-27 20:51 sue
* [r6435] standardized typefaces on cover
2007-03-27 02:24 leif
* [r6415] Cleaning, step 10: Removed *all* instance variables used
as temporaries, and *all* unused instance variables. If an
instance variable was only used in a single function (but was
more than a mere temporary), I changed it to an ordinary local
variable. I found all these ill-begotten instance variables with
the help of some introspective scaffolding (__getattribute__,
__settattr__). Refactored/tweaked some code so that temporary
arrays are automatically deallocated (without the goofy "xxx =
None").
2007-03-26 23:19 leif
* [r6414] Cleaning, step 9: Removed 'memorySize' computations.
Maintaining a manual tally such as this one is a tedius and
error-prone way to keep track of memory usage. (I promise to
re-implement this feature later.) Also, I temporarily labeled
all the explicit deallocations, so I can grep for them later.
2007-03-26 22:51 leif
* [r6413] Cleaning, step 8: Eliminated more redundant variable
initializations. Specifically, the (now former) PyLith
convention self.pointerToX = None # ...later... self.pointerToX
= pylith3d.allocateDouble(...) is meaningless in Python.
Variables do not need to be declared or initialized in Python.
Rest assured that any attempt to read 'pointerToX' before it is
defined will result in an attribute error.
2007-03-24 19:07 leif
* [r6389] Fixed Makefile.
2007-03-24 01:21 leif
* [r6387] Cleaning, step 7 {in initialize()}: Eliminated redundant
variable initializations. Moved constants to 'constants.py'.
Eliminated instance variables used as temporaries. Eliminated
unused variables.
2007-03-23 23:39 leif
* [r6383] Cleaning, step 6: Moved all properties out of the nested
Inventory class. They are now properties of the PyLith
application itself. CIG-Pyre automatically makes such properties
accessible as "self.xxx" (which is much less verbose than
"self.inventory.xxx"). This eliminates the especially tiresome
"self.xxx = self.inventory.xxx" mantra.
2007-03-23 22:48 leif
* [r6382] Cleaning, step 5: Merged Application and Pylith3d_scan
into a new class, PyLith. (Removing meaningless Component
structure... done.) There is now a single, top-level component
(the PyLith application class). However, this isn't a
user-visible change (yet), as I added a hack so that old
'pl3dscan.xxx' and 'scanner.xxx' options still work.
2007-03-23 21:26 leif
* [r6379] Cleaning, step 4: Merged GreenFunctionApp with
Application. (Removing meaningless Component structure... 90%
complete...)
2007-03-23 21:00 leif
* [r6377] Cleaning, step 3: Merged Pylith3d_run with
Pylith3d_scan. (Removing meaningless Component structure... 66%
complete...)
2007-03-23 20:40 leif
* [r6375] Cleaning, step 2: Merged Pylith3d_setup with
Pylith3d_scan. (Removing meaningless Component structure...)
2007-03-23 20:21 leif
* [r6373] Cleaning, step 1: removed the leading underscore on
instance variables in Pylith3d_scan, so that they will match the
identifiers used elsewhere in the code. (I'm tired of mold...)
2007-03-23 20:14 willic3
* [r6372] Incomplete and non-functional version of Greeen function
application. Some C code needs to be converted to Python, and
some logic still needs to be worked out.
2007-03-23 17:42 leif
* [r6369] Utilize the new _validate() hook in pythia v0.8.1.3b2,
which PyLith-0.8 now requires. Help and command-completion now
works.
2007-03-22 21:04 willic3
* [r6355] Added sampleLocationFile to inventory for sample
locations of Green's functions. Setup is kludgy right now,
because file is listed as 'optional'. It should actually be
listed as 'required' if Green's functions are being computed,
but I don't have a way of signaling that right now.
2007-03-22 18:13 willic3
* [r6351] Removed debugging statements.
2007-03-21 21:38 willic3
* [r6338] Directory containing simple tests for all of the
viscoelastic models presently available. The default model is a
linear Maxwell model based on the Zienkiewiecz & Taylor
formulation. Other models are obtained simply by putting the
appropriate *.cfg file on the command-line.
2007-03-21 20:23 willic3
* [r6337] Changed call to processMesh so we're now sending it the
base bcInputFile rather than the outputFileRoot.
2007-03-21 20:18 willic3
* [r6336] Made some changes so the per-processor BC filenames are
based on the global BC filename. For everything to be
consistent, this should also be done for coordinates,
connectivities, split nodes, and tractions. That would involve
changes to PETSc, however.
2007-03-21 18:37 leif
* [r6333] Avoid (r)partition, which is new in Python 2.5.
2007-03-21 17:56 willic3
* [r6332] Altered output of BC so that the filenames are based in
the inputFileRoot basename. Now that I think of it, I should add
a new argument so it is based on bcInputFile.
2007-03-21 17:53 willic3
* [r6331] Made some changes to get things working in parallel
again.
2007-03-19 17:01 willic3
* [r6290] Updated makefile for new file names.
2007-03-19 16:59 willic3
* [r6289] Fixed all material models to conform to new arrangement,
and added some documentation strings. Also replaced
commented-out print statements with journal calls.
2007-03-19 16:56 willic3
* [r6288] Completely rearranged material models to correspond with
current preferred methods, although more testing is needed. Also
removed Numerical Recipes linear algebra routines and replaced
them with LAPACK calls.
2007-03-19 16:50 willic3
* [r6287] Parameters for material model 9 (power-law viscoelastic
using Zienkiewicz & Taylor formulation).
2007-03-15 19:39 willic3
* [r6263] Added exception for case where maximum number of
Newton-Raphson iterations is exceeded.
2007-03-15 19:38 willic3
* [r6262] Major fixes to Newton-Raphson solution for power-law.
2007-03-10 21:12 willic3
* [r6220] Updated code to deal with multiple element types.
2007-03-09 21:58 willic3
* [r6215] Traction test for hexahedral elements.
2007-03-08 22:48 leif
* [r6205] Fixed the makefile I broke.
2007-03-08 00:12 leif
* [r6185] Removed support for .keyval files (this is a bit
premature, since I still haven't finished updating the
examples). Moved PETSc/Pyre integration stuff into the
framework. PyLith v0.8 now requires Pythia v0.8.1.3b1. Re-run
'configure' to get it.
2007-03-06 02:32 willic3
* [r6178] Temporarily added Numerical Recipes-derived routines for
LU decomposition and backsubstitution. This should be replaced
with LAPACK calls to routines that take advantage of matrix
symmetry and packed storage.
2007-03-02 19:06 willic3
* [r6161] New and altered routines for new method of dealing with
power-law rheology. Things should compile, but testing/debugging
is needed.
2007-02-28 01:54 leif
* [r6127] Updated about half of the PyLith examples to use *.cfg
files instead of *.keyval files. The other half is linhex/*,
which I will update later.
2007-02-14 00:27 leif
* [r6017] Reverted unintentional check-in of
ComponentHarness-related changes.
2007-02-14 00:23 leif
* [r6016] Ripped-out calls to PetscCommSynchronizeTags, per Matt's
instructions. The function no longer exists in the development
version of PETSc: changeset: 10000:21bdefd4c2d2 user:
bsmith@barry-smiths-computer.local date: Fri Feb 9 19:34:16 2007
-0600 summary: removed PetscCommSynchronizeTags and
PetscCommCheckTags
2007-02-12 20:06 willic3
* [r6010] Fixed 2D Jacobian determinant (needed square root).
Fixed problem where element contributions were replacing
traction vector contents rather than updating them.
2007-02-12 20:05 willic3
* [r6009] Fixed BC to give constant stress throughout cube. Fixed
incorrect node position.
2007-02-12 17:15 willic3
* [r6008] Made nodal spacing more regular so it is easier to
figure out what areas should be.
2007-02-12 17:14 willic3
* [r6007] This file must have been added accidentally, and has
been deleted.
2007-02-05 17:27 willic3
* [r5964] Fixed error in updating viscous strain.
2007-02-03 03:27 leif
* [r5962] Added a "keyval codec" so that *.keyval files can be
processed by the Pyre framework. Keyval parameters are now
ordinary Inventory items, so they can also be set from the
command line or a .cfg file. Another advantage of this is error
reporting: try inserting a bogus key or value into a '.keyval'
file. I still haven't got the implementation exactly right,
however: with this change, errors in .cfg files (for example)
are reported twice. Since the filename of the .keyval file is
itself a Pyre property, this creates order-of-initialization
problems which exceed the limitations of the Pyre framework. I
have one or two ideas about how to fix this, but they are still
in the oven, only half-baked...
2007-02-01 20:26 willic3
* [r5948] Added *.cfg file specifically for debugging to make life
easier.
2007-02-01 17:03 willic3
* [r5947] Fixed mistake in units for viscosityCoefficient.
2007-02-01 16:29 willic3
* [r5946] Fixed stupid typo.
2007-02-01 16:11 willic3
* [r5945] Removed old run scripts and replaced them with *.cfg
files. Fixed time stepping and added option for running
power-law example using older ESF-based solution method. At
present, PyLith is no longer able to parse the material
properties files correctly. I'm not sure why.
2007-01-30 21:28 leif
* [r5936] Moved Python egg stuff into our Autoconf macro library.
2007-01-30 20:50 willic3
* [r5932] Removed python run script, which isn't really necessary
with the new *.cfg files.
2007-01-30 20:45 willic3
* [r5931] Initial version of *.cfg file for tractest example. This
is about the simplest possible file, with nothing related to
MPI, etc.
2007-01-27 20:28 willic3
* [r5920] Fixed parsing of units.
2007-01-27 17:52 willic3
* [r5919] Added new utility to alter boundary conditions along
specified line segments, and added it to the Makefile. This
utility is somewhat specific to the CFEM benchmarks, but may
have some other uses.
2007-01-27 00:19 leif
* [r5918] Removed extraneous quotes around 'traction_units' in the
lintet/tractest example (I'm not sure how this ever worked).
2007-01-26 16:20 willic3
* [r5907] Fixed logic for getting minimum value for a point.
2007-01-26 15:13 willic3
* [r5906] Added faultbm.f to utilities.
2007-01-26 15:12 willic3
* [r5905] Initial version of code to calculate fault BC for CFEM
benchmarks. I need to test it.
2007-01-26 03:02 leif
* [r5900] Removed unused files.
2007-01-24 03:18 leif
* [r5882] Perform I/O file checks earlier, in _configure(), so
that errors are caught before launching. Also, turned the error
message into a 'journal' log message.
2007-01-24 00:44 leif
* [r5879] Cleanly separate input files from output files by giving
their filenames different roots. This allows PyLith to read
input files directly, instead of relying upon symlinks.
2007-01-22 15:47 willic3
* [r5853] Fixed incorrect call to td_matinit.
2007-01-19 21:47 leif
* [r5840] Added 'sys.executable' to the arguments for
PetscInitialize(), so that -start_in_debugger works. Also,
recognize -debugger_pause (the default is too short on my
system).
2007-01-19 00:35 leif
* [r5834] Fix for "Fortran 77 source seen but `F77' is undefined"
error with Automake 1.10.
2007-01-18 16:49 willic3
* [r5826] New example to test ESF version of power-law rheology.
2007-01-18 16:46 willic3
* [r5825] Corrected time info for double precision. Removed gdb
history file that had accidentally been committed.
2007-01-18 16:41 willic3
* [r5824] Starting the process of revising benchmarks. Old Make.mm
and test.sh files have been removed and replaced by runex.py
script to run the benchmark. Time information has been corrected
to yield correct viscoelastic solution. Files originally used
from bm1a have been copied over to make this a standalone
example. The time info needs to be tested to see if it is
overkill.
2007-01-18 16:30 willic3
* [r5823] Starting the process of revising benchmarks. Old Make.mm
and test.sh files have been removed and replaced by runex.py
script to run the benchmark. Time information has been corrected
to yield correct viscoelastic solution.
2007-01-18 16:25 willic3
* [r5822] Starting the process of revising benchmarks. Old Make.mm
and test.sh files have been removed and replaced by runex.py
script to run the benchmark. Time information has been corrected
to yield correct viscoelastic solution.
2007-01-18 15:21 willic3
* [r5821] Added descriptions to stdout info.
2007-01-17 22:24 willic3
* [r5804] Fixed incorrect routine calls.
2007-01-17 21:59 willic3
* [r5803] Fixed incorrect calls.
2007-01-17 21:09 willic3
* [r5801] Fixed incorrect call to td_strs_1.
2007-01-17 20:29 willic3
* [r5800] Fixed time stepping so viscous solution is correct.
2007-01-12 21:34 willic3
* [r5784] Added new material model.
2007-01-12 21:30 willic3
* [r5783] Added new material model to hold original ESF-based
power-law model.
2007-01-12 21:24 willic3
* [r5782] Added exceptions for problems with Cholesky inversion.
2007-01-12 21:23 willic3
* [r5781] Update all material model routines and associated
drivers to make use of iteration counter, which is useful for
some material models. Added option to tangent computation
routines to let them know whether tangent is being computed
using current stresses or stresses from previous time step.
Fixed test makefile to add top-level PETSc directory to include
path.
2007-01-12 21:20 willic3
* [r5780] Replaced original power-law model (model 6) with new
implementation and moved original implementation to material
model 9.
2007-01-12 21:19 willic3
* [r5779] New implementation of Maxwell power-law material model,
using formulation based on Zienkiewicz & Taylor rather than
original ESF-based version. Routines compile but have not yet
been tested.
2007-01-12 21:17 willic3
* [r5778] This material model now corresponds to the original
ESF-version of the Maxwell power-law model.
2007-01-10 03:19 leif
* [r5751] Tied PyLith's configuration even more closely to
PETSc's: use 'petscconf' to determine both CXX and the MPI
flags. I'm not sure why this 'configure' script didn't grab MPI
configuration from PETSc. But, I remember why it didn't grab CXX
from PETSc: Lithomop didn't require PETSc to be configured with
C++, so it couldn't assume that 'petscconf' defined CXX. Now,
with the addition of the Sieve dependency, it can be assumed
that 'petscconf' defines CXX. It is almost certainly an error to
configure PyLith with a C++ compiler different from the one used
to configure PETSc; this change avoids that potential mistake.
This change also avoids confusing 'configure' error messages
about MPIINCLUDES and MPILIBS, which aren't relevant to PyLith
-- PyLith should simply be configured with the MPI library
defined by the PETSc configuration.
2007-01-10 00:45 leif
* [r5746] Use CIG-Pythia's 'launcher' and 'scheduler' facilities.
Users no longer execute 'mpirun'. Instead, they execute
'pylith3dapp.py' directly, and the framework re-executes PyLith
under 'mpirun' automatically. On a cluster with a scheduler
(PBS, LSF, ...) the framework automatically generates a batch
script and submits a job to the queue. One uses
"--launcher.nodes=XXX" to specify the number of processors.
Currently, the framework defaults to MPICH1 with no scheduler;
using PBS, LSF, MPICH2, LAM/MPI, etc. may require additional
configuration, which is beyond the scope of this log comment.
This change breaks the "--disable-embedding" configuration. I
will revisit this later.
2007-01-09 03:02 leif
* [r5698] Transformed "Hello" print statements into 'journal' log
messages. Made MaterialModel subclasses call
MaterialModel.__init__(). Nuked gratuitous use of 'exec'.
2007-01-08 23:27 leif
* [r5696] Issue32: "make install" is now optional. Users can now
simply "configure", "make", and then add the "pylith3d"
subdirectory of the source/build directory to their PATH. In
this mode, changes to C++ simply require a "make" from the
top-level directory. Changes to the Python code do not require
you to run "make" at all -- simply re-run 'pylith3dapp.py'.
Also: Changed the 'configure' option '--with-embedding' to
'--enable-embedding', for consistency with CitcomS. Note: when
configured with '--disable-embedding', "make install" is NOT
optional. This is because the PyLith extension module has the
same name as the top-level package ('pylith3d'). The extension
module must be installed under the package directory in order
for everything to work. There are two possible fixes: 1) tweak
the build so that 'pylith3dmodule.so' lands under 'pylith3d'; or
2) rename the extension module.
2007-01-06 02:11 leif
* [r5688] Implemented strict command-line processing. Previously,
Pyre-based PyLith command-line options and PETSc options were
mixed together. This caused Pyre to issue warnings about the
unrecognized PETSc options, and PETSc to warn about the
unrecognized Pyre options. (Pyre errors were degraded to
warnings by giving "--typos=relaxed" as a matter for course.)
Now, PETSc and Pyre options are cleanly separated. (The use of
"--typos=relaxed" is discouraged.) This is implemented as
follows: PETSc options are designated explicitly using a 'petsc'
dummy facility+component, which allows passing arbitrary options
to PETSc: "--petsc.xxx". In addition,
commonly-used/highly-relevant PETSc options are listed in
PyLith's inventory; so, for example, on can give "-log_summary"
instead of "-petsc.log_summary" (either one will work). Finally,
PyLith overrides the standard Pyre CommandlineParser with
PetscCommandlineParser, which mimics PETSc's command-line
processing logic. This lets one write "-ksp_rtol 1e-09" (as is
customary) instead of "-ksp_rtol=1e-09". ~~~ Also: