-
Notifications
You must be signed in to change notification settings - Fork 27
/
ChangeLog
5987 lines (4180 loc) · 218 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
2021-12-10 v1.7.0
* new polynomial matrix API
* new FFT code for polynomial matrices
* Dixon solver refactorization
* more examples for rational solutions
* cleanup of BlasMatrixDomain with minor API changes
* update wrt Givaro's new randiter API
* many bug fixes
2019-06-07 v1.6.3
Minor update:
* fix test-suite failures when compiling in builddir!=srcdir
* improve jenkins builder script
* add README.md and other missing files in distribution
2019-05-10 v1.6.0
Major code refactorization:
* solve solution
* echelon form code
* prime generators
* Smith form via Valence
* MPI serialization and communicators
Improvements:
* many bug fixes
* new hadamard bound solutions
* removal of the obsolete sage-interface
* update and new polynomial matrix classes and API with Givaro
* MPI based chinese remaindering applied to solve
* broader testsuite
2017-12-08 v1.5.2
Improvements:
* minor bug fixes (missing config.h file, and SIMD macros)
* documentation build fixes
2017-11-22 v1.5.1
Improvements:
* minor bug fixes in the build system and autoinstall
2017-11-17 v1.5.0
Improvements:
* improved build system (instruction set detection, C++11 and clang compatibility, ...)
* many bug fixes
* increased test-suite coverage
Changes in API:
* new PolynomialRing and DensePolynomial classes, and update interraction with Givaro and FFLAS
2016-07-30 cpernet v1.4.2
* cleanup and new features on polynomial matrices
* many bug fixes ensuring support of gcc-4.8, 5.3, 6.1 clang-3.4 and
icpc on i386, x86_64, ubuntu osx, fedora and ppcle
2016-02-24 cpernet v1.4.1
* update the build system (add pkgconfig file, and a more consistent way
of dealing with dependencies)
* fix all remaining issues for the integration in SageMath
2016-02-24 cpernet
* move almost all finite fields to givaro
* cleanup dead code
* many bug fixes
* change in some parts of the API
* improved order basis computations
2011-03-10 briceboyer
* Implement DublinBox Mother model for dense matrices (BlasM* own, BlasS* share)
* Change BlasSubmatrix private members (and add one to make Iterators faster)
* Change <> includes to "" so I have no surprise when the compiler looks them in a non expected place
* add a const in Field Categories for a weird bug...
* RawIterators are renamed to Iterators (impossible to pronounce, easy to misspell)
* add functions to blas-domain and blas-matrix
2011-24-09 briceboyer
* replace DenseMatrix(Base) by Blas(Matrix/Blackbox)
* put these classes (and DenseSubmatrix) in a Protected namespace (users will be warned...)
* update Blas(Mat/Bb) interface
* clean DenseMatrix(Base) from LinBox
* Delete BlasTag (use LinBoxTag class instead, avoids branching)
* Update to Givaro gcd new clean intefarce
* Update to one/zero in (some) Fflas-Ffpack fields
* Rename Flaflas to more serious FFLAS-FFPACK
* Minor updates in some classes.
* martin albretch patch to update linbox-sage.h
* M4RI checks
* Ekopath (pathscale) checks
2011-05-11 briceboyer
linbox-1.2.0 released:
* Many many bug fixes, many memleaks, much code beautify.
* Licence/copyright added to (hopefully) every file.
* Vi/emacs magic lines for auto-indentation.
* Some documentation added + improved Doxygen framework (dev/user
documentation, more *.doxy items, bug/todo list, bibliography list)
* Upgrade to givaro 3.4.0 series ( new namespace, new API for integer
operations, a few new functions (naturallog...) )
* Upgrade to fflas-ffpack 1.4.0 (new namespace, new includes)
* Some initial work introduced here and there (matrix permutation
representations, symbolic/numeric solver, sparse matrix
representation...)
* CRA algorithms reworked and tested, some added.
* Timer moved to Givaro
* Some debug/util moved to FflasFfpack
* Better configure.ac/Makefiles (more options, icc/gcc
customisation,...)
* More m4 rules
* Possibility to use IML functions as a shared library (and some
examples given)
* 'make fullcheck' checks all possible tests given and reports
* Auto-installer from sources/svn with plenty of options
* Some new field functions like characteristic()
* Work on synchronising field member functions signatures for
better compatibility between all fields (in progress)
2010-11-03 cpernet
linbox-1.1.7 released:
* bug fixes and minor enhancements
* New interface to sage (supporting native floating point
representation)
* new rebind interface
* incorporate the code for computations over the rationals (by A
Ubranska), still waiting to be cleaned-up (it is not used yet by default)
2008-09-22 cpernet
linbox-1.1.6 released:
* compilation pb, and bug fixes, for cygwin support and debianization
2008-06-10 cpernet
* Upgrade fflas-ffpack to v1.3.3 (safer and tighter bounds, bug
fixes,...)
* Design changes for balanced finite field implementations
* Improve efficiency of Chinese Remainder Algorithm
* Add new interface to Sage
* bugs and documentation fixes
* remove gmp++ from LinBox, and thus make Givaro a necessary package
2008_03-02 cpernet
linbox-1.1.5 released:
* drivers are back to the source code,
* shared libraries are enabled by default.
* new code in fflas-ffpack for rank profiles
* new double-determinant algorithm
* changed the rawIndexedIterator interface, fixing the broken previous one.
* many minor fixes, mostly memleaks.
2007-10-27 cpernet
linbox-1.1.4 has been released including the following most important
changes:
* lots of memory leaks fixed
* better compliance with BLAS
* increased system solver functionalities (rectangular, rank deficient
matrices)
* support for g++/gcc v4.2
2007-09-28 cpernet
* Add support for more general dense system solving:
The system can be rectangular (over/underdetermined), singular.
2007-09-07 bds
* This changelog is being widely ignored. Today buildbot test is on.
2007-01-31 bds
* This entry is just to record that version 1.1.0 has been released by
Jean-Guillaume Dumas and the current svn version is just a couple of
steps from the release version.
2006-08-08 bds
* examples/det.C, solutions/det.h, methods.h modified in preparation for
* mpi based paralle cra loop in algorithms/cra-domain.h (by Bryan Youse).
* also a stub of util/mpicpp.h added (will become the mpi interface).
2006-07-27 Hui Wang and bds
* Add blackbox/quad-matrix.h which defines the ZOQuad matrix,
* which is a cache efficient {0,1}-matrix rep.
* Add it's tests, modify also blackbox/zo.h,
* and accumulate-special in field/modular.h to better support sums of field elements efficiently.
2006-06-01 bds
* rank.h adjusted to work around faulty PIVOT_LINEAR code in sparse elim.
* test-rank.C now tests blackbox, elimination, hybrid methods as well as...
2006-04-18 bds and Hui Wang
* examples/smith.C minor changes
* linbox/blackbox/apply.h - don't recall what we did to wreck this.
* linbox/blackbox/archetype.h linbox/blackbox/scalar-matrix.h linbox/blackbox/zero-one.h linbox/randiter/modular.h linbox/vector/stream.h tests/Makefile.am tests/Matio.h - minor touches
* cvs thinks we changed just about everything in tests, but diff shows no change. We're checking in anyway
* Where are our added files blackbox/zo.h zo.inl quad-matrix.h ? - which are a new version of zero-one and a block decomposed sparse matrix, resp.
2006-01-20 bds and Hui Wang
* linbox/algorithms/gauss.inl - we think we made it better but it still isn't working :-)
2005-12-13 Pascal Giorgi
* linbox/algorithms/Makefile.am: add new files
* linbox/algorithms/lifting-container.h: minor change
* linbox/algorithms/matpoly-mult.h: adding FFT and Karatsuba based polynomial matrix multiplication and middle product
* linbox/algorithms/rational-reconstruction.h: use macro to enable timing
* linbox/algorithms/rational-solver.h: minor change
* linbox/algorithms/rational-solver.inl: minor chane
* linbox/algorithms/short-vector.h: work in progress for ternary lattice reduction
* linbox/algorithms/sigma-basis.h: add muti sigma basis computation and special Pade solver.
* linbox/blackbox/block-hankel-inverse.h: enable use of new special Pade solver
* linbox/blackbox/block-hankel.h: use macro to enable timing
* linbox/randiter/{Makefile.am,random-fftprime.h}: add fftprime generator
2005-12-8 bds
* minor cleanups
2005-11-17 Pascal Giorgi:
* macros/blas-check.m4: adding new support for blas libraries
* macros/linbox-opt.m4: adding new support for blas libraries
* macros/linbox-misc.m4: adding new support for blas libraries
* linbox/config-blas.h: linbox headers for BLAS
* linbox/Makefile.am: add new BLAS support
* linbox/fflas/Makefile.am: remove dependencies to ATLAS
* linbox/fflas/fflas.h: add new BLAS support
* linbox/ffpack.Makefile.am: remove dependencies to ATLAS
* configure.in: remove ATLAS checking and replace with BLAS checking
* linbox-config.in: remove ATLAS libs and replace with BLAS libs
* tests/Makefile.am: remove ATLAS dependencies
2005-10-26 Pascal Giorgi:
* linbox/algorithms/Makefile.am: add sigma-basis.h file
* linbox/algorithms/sigma-basis.h: make sigma basis algorithm available
* linbox/algorithms/blackbox-block-container.h: add support for special structured block
* linbox/algorithms/blas-domain.h: fix bug
* linbox/algorithms/lifting-container.h: add support for sparse solver using a Block Hankel inverse
* linbox/algorithms/rational-solver.h(.inl): add new algorithm for solving sparse system
* linbox/blackbox/Makefile.am: add block-hankel.h. block-toeplitz.h and block-hankel-inverse.h
* linbox/blackbox/apply.h: fix optimization compatibility
* linbox/blackbox/block-hankel.h : add a blackbox for block Hankel matrices
* linbox/blackbox/block-hankel-inverse.h: add blackbox to compute the inverse of a block Hankel blackbox
* linbox/blackbox/block-toeplitz.h: add a blackbox for block Toeplitz matrices
* linbox/element/gmp-rational.h: fix this stupid bug using signed long integer to convert rational !!!
* linbox/field/hom.h: add support for PID_integer homomorphism
* linbox/ring/givaro-polynomial.h: fix bug
* linbox/solutions/det.h: add handler for rational domain
* linbox/solutions/charpoly.h: add handler for rational domain
* linbox/solutions/minpoly.h: add handler for rational domain and hybrid method for blas blackbox
* linbox/solutions/rank.h: add handler for rational domain
* linbox/solutions/solve.h: add handler for rational domain and fix indentation
* linbox/solutions/methods.h: fix indentation and add block hankel method
2005-09-01 Dan Roche
* linbox/algorithms/toeplitz-det.h: Forgot to add this before.
2005-08-31 Pascal Giorgi:
* linbox/algorithms/blas-domain.inl: add inverse specialization for multimod double domain
* linbox/algorithms/cia.h: call domain specialized minpoly instead of calling API and make code generic according to integer ring
* linbox/algorithms/lifting-container.h: support for multimod double domain
* linbox/algorithms/matrix-hom.h: support for multimod double domain
* linbox/algorithms/matrix-inverse.h: support for multimod double domain
* linbox/algorithms/rational-solver.inl: support for multimod double domain
* linbox/blackbox/apply.h: support for multimod double domain and new optimization
* linbox/blackbox/blas-blackbox.h: support for multimod double domain, add construction from MatrixReader and finalize apply function
* linbox/blackbox/submatrix.h: fixed indent style problem
* linbox/ffpack/ffpack.h: fix memory leak problem
* linbox/field/modular-double.h: add support for multimod double domain
* linbox/field/ntl-ZZ.h: fix indent style problem
* linbox/field/ntl-ZZ_p.h: set FieldTraits categoryTag
* linbox/matrix/blas-matrix.h: add constructor from MatrixStream
* linbox/matrix/dense-submatrix.h: add constructor from non const MatrixDenseBase
* linbox/matrix/dense-submatrix.inl: add constructor from non const MatrixDenseBase, fix bug in ConstRawIterator operator++(int)
* linbox/matrix/transpose.h: make the reference to the matrix to be non const
* linbox/ring/givaro-polynomial.h: add support for PID-integer and NTL-ZZ domain
* linbox/solutions/charpoly.h: throw an exception if matrix are not square
* linbox/solutions/det.h: fix indentation and throw an exception if matrix are not square
* linbox/solutions/minpoly.h: fix indentation, throw an exception if matrix are not square and make minpoly over the integer generic
* linbox/solutions/rank.h: fix indentation
* linbox/solutions/smith-form.h" fix indentation
* linbox/util/debug.h: allow the PreconditionFailed to be a real LinBox exception
2005-08-19 Z. Wan
* linbox/element/gmp-rational.h Fixed the error which occurs duing conversion
between gmp integer and gmp rational
2005-08-18 Dan Roche
* Added functionality for computing the determinant of a toeplitz matrix
* Added field/ntl-ZZ_pX.h and field/ntl-lzz_pX.h, rings over the
respective NTL polynomial types
* Added functions pow and powin to field/ntl-ZZ_p.h and
field/ntl-lzz_p.h (could be added to all fields)
* Added blackbox/toeplitz.h and blackbox/toeplitz.inl, very similar to
the existing Toeplitz blackbox class, but generalized to use any
polynomial ring. Didn't delete the old blackbox/ntl-toeplitz.h, inl
so old code doesn't break
* Added tests/test-toeplitz-det.C to test all this
* Modified tests/test-ntl-toeplitz.C to use the new Toeplitz class
2005-08-1 bds
* numerous doc tweaks
2005-07-15 bds and zw
* Now modular-balance-int.h and modular-int32.h define different classes.
* Modular-int had been replaced by modular-int32.
* Modular-int is now deprecated.
* Added randiter/modular-balance.h
2005-07-13 bds and zw
* solutions/solve.h call block-lanczos in bbsolve when cardinality small
* algorithms/bbsolve.h call mg-block-lanczos
* algorithms/mg-block-lanczos commentating number of trials.
2005-07-13 Bradford Hovinen <[email protected]>
* linbox/solutions/solve.h: Cast A.coldim () to double before
calling log
(solve): Initialize status
* linbox/algorithms/rational-solver.inl: Use std::max
* linbox/blackbox/apply.h: Move #include ntl-ZZ.h into #ifdef
__LINBOX_HAVE_NTL
* examples/solve.C: Add using namespace std
2005-07-13 bds and zw
* tests/Makefile.am move test-la-block-lanczos to failing tests list
* test-la-block-lanczos use case n = 10 to show problem
* test-mg-block-lanczos restore maxtries to default, comment out nullspace
test which has memory probs.
* test-solve change default blocking factor to 16
2005-07-13 Bradford Hovinen <[email protected]>
* tests/Makefile.am (NTL_TESTS): Moved test-rational-solver,
test-rational-solver-adaptive, test-smith-form-binary,
test-last-invariant-factor here because they depend on NTL
(GIVARO_TESTS): Moved test-ffpack, test-blas-domain,
test-givaro-zpz here because they depend on Givaro
* tests/Makefile.am (NTL_TESTS): Fix definition so that automake
does not complain
2005-07-13 Pascal Giorgi
fix and remove NTL dependencies
- remove NTL dependecies: use of PID_integer
* tests/test-last-invariant-factor.C
* tests/test-rational-solver.C
* tests/test-rational-solver-adaptive.C
* tests/test-smith-form-binary.C
- fix NTL dependencies: by use of __LINBOX_HAVE_NTL preprocessing
* tests/test-smith-form.C
* tests/test-smith-form-iliopoulos.C
* tests/test-smith-form-local.C
* tests/test-smith-form-adaptive.C
* linbox/field/hom.h: specialization for NTL
* linbox/algorithms/smith-form-adaptative.h
- linbox/ring/givaro-polynomial.h: to be done by Jean-Guillaume and Clement
2005-07-12 Pascal Giorgi
* linbox/matrix/factorized.inl: add support for singular system solving (inconsistent system is defined by the zero solution)
fix also bug in nonsingular solving
2005-07-11 Pascal Giorgi (authorized by dictator Dumas))
* linbox/solutions/solve.h: change the interface for integer systems
Make the solver interface working with Wiedemann method, Lanczos method
* linbox/algorithms/solve.h: add support for Lanczos method
* linbox/algorithms/bbsolve.h: fix compilation error
* linbox/algorithms/wiedemann.h: fix compilation error
* linbox/algorithmd/wiedemann.inl: fix compilation error
* linbox/algorithms/lanczos.h: fix compilation error
* linbox/algorithmd/lanczos.inl: fix compilation error
2005-07-07 Dan Roche
* linbox/blackbox/sparse.h
* linbox/blackbox/dense.h
* linbox/matrix/sparse.h: Changed the RawIndexedIterator code so that
dereferencing returns a reference to the value
(same as RawIterator)
* linbox/blackbox/lambda-sparse.h
* tests/test-matrix-stream.C
* linbox/algorithms/matrix-hom.h: Updated the code here to reflect
change above
2005-07-07 Dan Roche
* linbox/matrix/sparse.h: Went through and fixed more iterator problems that could be causing segfaults.
2005-07-06 bds and zw
* linbox/solutions/is-positive-definite.h, is-positive-semidefinite.h, added
functions isPositiveDefinite and isPositiveSemiDefinite.
* linbox/tests/test-isposdef.C, test-ispossemidef.C, Makefile.am, added
tests for them.
* linbox/field/hom.h, linbox/algorithms/signature.h, minor tweaks.
* linbox/algorithms/algorithms.doxy, docs related
2005-07-06 Dan Roche
* examples/rank.C
* examples/det.C: Modified these to use MatrixStream and thereby support multiple file formats.
* linbox/matrix/sparse.h
* linbox/matrix/sparse.inl: Fixed some problems with the iterators: seg faults were caused by poor handling of empty rows.
* linbox/util/matrix-stream.inl: minor clean-ups
2005-07-4 Pascal Giorgi
* linbox/blackbox/Makefile.am: add polynomial.h
* linbox/blackbox/apply.h: add optimization
* linbox/blackbox/blas-blackbox.h: add the use of BLAS asap
* linbox/solutions/methods.h: extend dixon method
* linbox/solutions/solve.h: add rational-solver and diophantine-solver support for BlasElimination method (need to fix denominator output).
also add support for BlasElimination over finite field (not fully implemented)
* linbox/algorithms/rational-reconstruction.h: define early termination according to threshold
(0 means no early termination, x > 0 means early termination with checking each x steps)
2005-07-01 Pascal Giorgi
* configure.in: add linbox/util/formats/Makefile.am support
* add linbox/util/formats/Makefile.am file
* linbox/util/Makefile.am: fixe comment problem
2005-07-01 Dan Roche
* linbox/blackbox/dense.h
* linbox/matrix/dense.inl
* linbox/matrix/dense.h: Added Constructors from a matrix-stream
* linbox/blackbox/sparse.h
* linbox/blackbox/sparse.inl
* linbox/matrix/sparse.h: Added Constructors from a matrix-stream
Changed to use size_t for indexing rather than ints:
* linbox/util/matrix-stream.h
* linbox/util/matrix-stream.inl
* linbox/util/formats/generic-dense.h
* linbox/util/formats/maple-dense1.h
* linbox/util/formats/maple-sparse1.h
* linbox/util/formats/matrix-market.h
* linbox/util/formats/sms.h
* linbox/util/formats/sparse-row.h
* tests/test-matrix-stream.C: Included sparse and dense blackbox matrix constructors in testing
* tests/data/sms.matrix: Fixed a formatting error
2005-07-01 W. J. Turner
* linbox/until/commentator.C: removed using namespace std directive
2005-06-30 Pascal Giorgi
resolve symbol error in libutil.a (commentator.o needed timer.o)
* linbox/util/commetator.C: include timer.C
* linbox/util/Makefile.am: remove timer.c from libutil.a source (included through commentator)
2005-06-29 W. J. Turner
Removed using declarations within several classes and replaced with
this->... to preserve backwards compatability and allow compilation
with gcc 3.4 and newer:
* linbox/algorithms/blackbox-block-container.h
* linbox/algorithms/blackbox-container-symmetric.h
* linbox/algorithms/blackbox-container-symmetrize.h
* linbox/algorithms/blackbox-container.h
* linbox/algorithms/lifting-container.h
* linbox/blackbox/frobenius.h
* linbox/blackbox/lambda-sparse.h
* linbox/blackbox/ntl-hankel.{h,inl}
* linbox/blackbox/submatrix.h
* linbox/field/givaro-zpz.h
* linbox/matrix/blas-matrix.h
* linbox/util/formats/generic-dense.h
* linbox/util/formats/maple-dense1.h
* linbox/util/formats/maple-sparse1.h
* linbox/util/formats/matrix-market.h
* linbox/util/formats/sms.h
* linbox/util/formats/sparse-row.h
2005-06-27 W. J. Turner
* linbox/ffpack/ffpack_charpoly_kgfast.inl: removed using namespace
std directive
* linbox/*/lidia-gfq.h: removed using namespace LiDIA directives
* linbox/vector/sparse.h: removed using namespace LinBox directive
* linbox/util/formats/*.h: removed using declarations
* linbox/util/*.h: removed using declarations
* linbox/util/xml/*.h: removed using declarations
2005-06-25 Bradford Hovinen <[email protected]>
* tests/test-mg-block-lanczos.C (testRandomSolve, testSampleNullspace):
Use preconditioner SYMMETRIZE
(main): Use more sensible defaults
* examples/dot-product.C (main): Declare v1 and v2 to be of length n
(fixes segfault)
2005-06-24 W. J. Turner
* linbox/blackbox/*: removed using declarations and directives
* linbox/field/*.h: removed using declarations and directives
* linbox/field/ntl-lzz_pE.h: added explicit call to namespace in NTL::BuildIrred_zz_pX
* linbox/field/ntl-GF2E.h: added explicit call to namespace in NTL::BuildSparseIrred_GF2X
* linbox/randiter/*.h: removed using declarations
2005-06-17 Pascal Giorgi
* /linbox/field/PID-integer.h: add sqrt function and homomorphism support
* /linbox/field/hom.h: identation fixed
* /example/solver/t-rdisove.C: comment givaro.h to fix compilation errors
2005-06-16 Pascal Giorgi
* /linbox/blackbox/apply.h: add blas-blackbox compatibility
* /linbox/blackbox/blas-blackbox.h: modify constructor and add category support
* /linbox/matrix/blas-matrix.h: new creator and add category support
* /linbox/matrix/matrix-category.h: move MatrixCategory from blas-matrix.h
* /linbox/algorithms/blas-domain.h: fancy indentation
* /linbox/algorithms/blas-domain.inl: add support for BlasBlackbox
* /linbox/algorithms/matrix-hom.h: add support for BlasBlackbox
* /linbox/algorithms/rational-solver.inl: use of MatrixHom and BlasBlackbox instead of MatrixMod and BlasMatrix
2005-06-15 Pascal Giorgi
* /linbox/blackbox/apply.h : compatiblity with 64bits architecture
2005-06-14 Pascal Giorgi
* /doc directory: modification to documentation (linbox.html become central)
* macros/linbox-doc.m4: new documentation checking (use of Doxygen)
* configure.in: take into consideration new doc checking
2005-06-10 bds
methods.h working for det.h in good form - other solutions may be broken:
Can consider det.h a good model for solutions funs now.
2005-06-07 Z. Wan
* linbox/blackbox/matrix-blackbox.h, linbox/matrix/dense.h, sparse.h,
added reind to DenseMatrixBase and SparseMatrixBase.
2005-05-02 Pascal Giorgi
* configure.in , macros/linbox-opt.m4 : add optimization process during linbox configuration
* linbox/fflas/fflas.h : allow the use of optimized Winograd threshold
2005-04-26 -bds
* Name changes *ntl-zz* --> *ntl-lzz* to avoid clashes with ZZ in case
insensitive situations.
* Also change test*.cpp to test*.C for consistency
2005-03-32 Pascal Giorgi
* linbox/fflas/fflas_fgemm.inl: fix compilation error with WinoMain function call
* linbox/algorithms/Makefile.am, rational-solver.h,.inl: set default prime size to 20 bits and make number of prime to try as a user parameter
in function solve
* tests/Makefile.am, test-rational-solver.C : retrieve rational solver test (make commentator progress worrking)
2005-03-30 bds
* removed xml read/write code in numerous places
* misc small stuff (I forgot what I may have changed, to tell the truth!)
2005-03-17 PG and ZW
* linbox/algorithms/cra.h, cra-det-integer.h, minpoly-integer.h, Makefile.am, added
functionalities for integer cases using modular arithmetic and CRA
* linbox/solutions/det.h, minpoly.h, rank.h, unified the interface, added blas elimination
for det and rank
2005-03-15 bds
*tests/test-rank.C linbox/solutions/rank.h methods.h Introduce rank
by blas-elimination, together with check for too small prime in wiedemann rank.
2005-02-17 Zhendong Wan ([email protected])
*linbox/algorithms/rational-reconstruction.h, implemented the fast early termination techniques, which was described
in a preprint paper submitted to ISSAC'05.
2005-02-16 Zhendong Wan ([email protected])
*linbox/algorithms/rational-solver.h, rational-solver.inl, rational-solver-adaptive.h, Merged rational solver using numerical methods with others
2005-02-14 Clement Pernet ([email protected])
* Improvements to FFLAS::fgemm routine: Winograd's fast matrix
multiplication is now working for any operation of the type
C<-alpha.A*B + beta.C
* Dynamic pealing is now generic wrt the field
* The recursive block algorithm is now working over the field if the
BLAS condition is not met
2005-02-07 Clement Pernet ([email protected])
* Change the name fflapack -> ffpack
* bug fixes in blackbox-block-container.h
* compilation failure of test-rational-solver due to "assert" in
rational-solver.h should be fixed
* compilation failuer in test-matrix-stream due to 888888888888888 being to big
should be fixed
2005-02-04 Zhendong Wan ([email protected])
*Add linbox/solutions/cra.h, minpoly-integer.h, add codes to compute the minpoly of an integer matrix.
*Modify linbox/test-minpoly.C, add test for minpoly of an integer matrix.
*Modify linbox/solutions/minpoly.h, it can hanle integer case.
** linbox/solutions/cra.h and linbox/algorithms/cra.h should merge in future
** minpoly-integer.h and minpoly.h should merge, and support blas in future
2005-02-03 Zhendong Wan ([email protected])
*linbox/field/PID-double.h, PID-integer.h, PIR-modular-int.h,
PIR-modular-int32.h, PIR-ntl-ZZ_p.h, double-fmod.h,
field-traits.h gf2.h, givaro-gfq.h, givaro-montg.h,
givaro-rational.h, givaro-zpz.h, gmp-integers.h, gmp-rational.h,
lidia-gfq.h local2_32.h, modular-balance-int.h,
modular-balance-int32.h, modular-byte.h, modular-double.h,
modular-int.h, modular-int32.h, modular-short.h, modular.h,
ntl-GF2E.h ntl-RR.h, ntl-ZZ.h, ntl-ZZ_p.h, ntl-ZZ_pE.h,
ntl-pid-zz_p.h, ntl-zz_p.h, ntl-zz_pE.h, param-fuzzy.h,
unparametric.h,
added information in field traits in order to classify all rings.
2005-03 Zhendong Wan ([email protected])
*linbox/algorithms/matrix-mod.h, change class MatrixMod to namespace,
so that some special mod functions can be addaed
*linbox/algorithms/rational-reconstruction.h, rational-solver.inl, fixed
compiling complaints
2005-01-28 Dan Roche
* examples/linbox-program.C - various fixes
* field/unparametric.h - was casting to long and then to integer - fixed
* field/hom.h - added mapping from UnparametricField to Modular
* util/matrix-stream.h Provides a new, generic way of reading matrices in multiple formats.
* util/formats New directory for matrix-stream format specifications
* util/formats/matrix-stream-readers.h Place to include all the different format specification files:
* util/formats/generic-dense.h
* util/formats/maple-dense1.h
* util/formats/maple-sparse1.h
* util/formats/matrix-market.h
* util/formats/matrix-stream-readers.h
* util/formats/sms.h
* randiter/gmp-random-prime.h Generates random numbers up to a given size
* tests/test-matrix-stream.C Tests to see if matrix-stream is working
* Sample matrices used by test-matrix-stream:
* tests/data/generic-dense.matrix
* tests/data/maple-dense1.matrix
* tests/data/maple-sparse1.matrix
* tests/data/matrix-market-array.matrix
* tests/data/matrix-market-coordinate.matrix
* tests/data/sms.matrix
* tests/data/sparse-row.matrix
2005-01-07 ZW
* linbox/field/gmp-integers.h, fixed init
* linbox/field/modular-double.h, implemented a fast init
2004-12-03 bds
* matrix/sparse.h .inl - add reading of Magma Compact form
* vector/vector-domain.h .inl - can't read into const vectors, so rm 'const
2004-10-08 bds
* examples/blackbox/Makefile.am smith.C - minor
* field/field-traits.h givaro-gfq.h ntl-GF2E.h - put in maxmodulus
* doc/Doxyfile mainpage - minor doc changes
2004-09-30 Pascal Giorgi
* linbox/algorithms/lifting-container.h: take modification of MatrixApplyDomain setup function
* linbox/algorithms/rational-solver.inl: use of BlasMatrix instead of BlasBlackbox
* linbox/algorithms/rational-reconstruction.h: add lifting commentator progress in getRational3 function
* linbox/blackbox/apply.h: redefine MatrixApplyDomain s.t. specialization for dense storage is easier
* linbox/matrix/dense.inl: make reading with sparse format work and fix bug in writing over domain with non positive cardinality
2004-09-21 Pascal Giorgi
* linbox/algorithms/Makefile.am: Add headers for (rationnal-solver.inl, diophantine-solver.inl, vector-fraction.h)
* linbox/algorithms/blackbox-block-container-base.h: Use BlasDomain and BlasMatrix
* linbox/algorithms/blackbox-block-container.h: Make BlackboxBlockContainerRecord class (handling precomputed sequence and row-colum updating)
* linbox/algorithms/block-massey-domain.h: Compute generating matrix polynomial using Sigma Basis through algorithm of GJV-ISSAC'03
* linbox/algorithms/block-wiedemann.h: Add block Wiedemann solver
* linbox/algorithms/lifting-container.h: Add specialisation for Block Wiedemann method and clean optimizations
* linbox/algorithms/rational-solver.h: Add specialisation for Block Wiedemann method
* linbox/algorithms/rational-solver.inl: Add specialisation for Block Wiedemann method
* linbox/blackbox/Makefile.am: Add headers for lambda-sparse.h and blas-blackbox.h
* linbox/blackbox/apply.h: Put optimizations on matrix-vector product - moved from lifting-container
* linbox/fflas/fflas_fgemm.inl: Fix bug on Winomain function
* linbox/fflas/fflas_ftrsm.inl: Fix bug on normalisation before BLAS call
* linbox/field/Makefile.am: Add ntl-GF2E.h header
* linbox/field/ntl-GF2E.h: Add wrapper for GF(2^n) of NTL library
* linbox/field/modular-double.h: Improve DotProductDomain specialization
* linbox/field/ntl.h: Add ntl-GF2E.h header
* linbox/matrix/blas-matrix.h: Take into account modifications of dense-submatrix
* linbox/matrix/dense-submatrix.h: Use pointer instead of reference
* linbox/matrix/dense-submatrix.inl: Use pointer instead of reference
* linbox/matrix/factorized-matrix.inl: Fix the extractions of L, U and S
* linbox/solutions/methods.h: Add BlockWiedemann method
* linbox/util/gmp++/Makefile.am: Remove gmp++_int.c ( redudant for compilation)
2004-09-02 Bradford Hovinen <[email protected]>
* linbox/algorithms/la-block-lanczos.h (class
LABlockLanczosSolver): Add _iter, _total_dim, _rank
* linbox/algorithms/la-block-lanczos.inl (cleanup): Update _rank
(rank): Implement
* linbox/blackbox/diagonal.h (class Diagonal): Compilation fix
* linbox/blackbox/dense.h (class DenseMatrix): Make _AT use
DenseMatrix, not DenseMatrixBase
* tests/test-la-block-lanczos.C (main):
* tests/test-mg-block-lanczos.C (main): Use better defaults
* linbox/algorithms/la-block-lanczos.inl (solve, sampleNullspace):
Clear history when done
* tests/Makefile.am (BASIC_TESTS): Add test-mg-block-lanczos
* linbox/algorithms/la-block-lanczos.h (class
LABlockLanczosSolver::InnerProductArray): Add reset
* linbox/algorithms/la-block-lanczos.inl (iterate): Reset uAv
* linbox/algorithms/Makefile.am (pkgincludesub_HEADERS): Add
eliminator.{h|inl}
* tests/Makefile.am (BASIC_TESTS): Add test-la-block-lanczos
* linbox/algorithms/Makefile.am (pkgincludesub_HEADERS): Add
la-block-lanczos.{h|inl}, mg-block-lanczos.{h|inl}; remove
block-lanczos.{h|inl}
2004-09-01 bds
* solutions/minpoly.h tests/test-minpoly.C: add minpolySymmetrize
2004-09-01 bds wan zw
* linbox/algorithms/rational-solver-adaptive.h, adaptive solve
2004-08-30 bds and zw
* linbox/algorithms/one-invariant-factor.h, smith-form-adaptive.inl, make bonus works
2004-08-30 bds and zw
* linbox/algorithms/smith-form-adaptive.h, inl. Implementation of an adaptive algorithm for
Smith form
2004-08-30 bds
* linbox/randiter/mersenne-twister.h linbox/util/commentator.h
linbox/util/gmp++/gmp++_int_misc.C tests/Makefile.am tests/test-common.C:
LinBoxSrcOnly flag also used in tests/ now (to avoid incompatibility
with use in examples/)
2004-08-30 bds
* examples/ex-integers.C examples/Makefile.am linbox/util/gmp++/Makefile.am: Turn on
LinBoxSrcOnly flag for examples compilations so that user does not confront
load-library problems.
* doc/Makefile.am examples/Matio.h linbox/integer.h linbox/blackbox/*.h etc.:
switch documentation system to doxygen. (it still needs plenty of work)
* add .doxy files (replacing .dxx files) to doc system
* examples/blackbox/smith.C and smithbig.C: smith form examples.
2004-08-28 Pascal Giorgi
* examples/solver/t-rdisolve.C: add sparse matrix format for file and make seed to work
* linbox/algorithms/diophantine-solver{.h,.inl}: adding parameters information of solving
* linbox/algorithms/rational-reconstruction.h: clear bugs in timings
2004-08-26 Z. Wan
* linbox/algoriths/last-invariant-factor.h, implement the bonus idea \ref{....}
2004-08-26 Pascal Giorgi
* examples/fields/ex-fields-archetype.C : put std namespace
* examples/run/run.C : put std namespace
* examples/solver/Makefile.am : change order of libs for compilation
* examples/solver/t-rdisolve.C : change to use the new design of rational solver
* gmp++/gmp++_int.h : Added new function to get the size of integer in base B and base 2
* gmp++/gmp++_int.inl : Added implementation of bitsize() and size_in_base( int base) functions
* linbox/algorithms
* diophantine-solver.h : use new design of rationnal solver
* diophantine-solver.inl : use new design of rationnal solver ( disabling conversion for VectorFraction)
* lifting-container.h : Added fast apply over ring when entries of the matrix are small enough ( say less than 2^16)
* rational-reconstruction.h : Added direct method for reconstruction, added switch between early termination and direct method
* rational-solver.inl : Added blas optimization for preconditionning matrices in random solution solver
* linbox/fflapack/fflapack.h : Maked triangular inverse available
* linbox/field/PID-double.h : Added abs and compare functions, + indentation
* linbox/field/PID-integer.h : Added abs and compare functions, + indentation
* linbox/matrix/dense.inl : problem with resizing in read function, leave code as it was but add the older version as comment
2004-08-24 Z. Wan ([email protected])
* change ith-invariant-factor.h to one-invariant-factor.h
* Change the implementation of the alogrithm for oneinvariantfactor, use the enhanced algorithm in ISSSAC'04 papre by
bds and ZW
2004-08-24 Z. Wan ([email protected])
*linbox/algorithms/rational-solver.h, inl, rational-reconstruction.h, linbox/tests/test-rational-solver.C, change
the format of the rational solution a vector of rationals ====> (a vector of the numerator, the common denominator).
2004-08-19 Z. Wan ([email protected])
* linbox/algorithms/rational-reconstruction.h, implement the V. Pan's idea for final reconstruction step.
* linbox/field/ntl-ZZ.h, add abs and compare member function.
2004-08-19 David Pritchard ([email protected], [email protected])
* gmp++/gmp++_int.inl - Added wrapper for mpz_import (convert word array to LinBox::Integer)
* linbox/algorithms - Added timing to rational solving routines (#define RSTIMING in rational-solver.h)
* diophantine-solver.h - made QSolver field into a reference type
* diophantine-solver.inl - changed comments
* lifting-container.inl - Added blas optimization for muliplication of large integer matrices in Dixon
* rational-reconstruction.h - Changed getRational2 again so that early termination num/denbounds work
* rational-solver.h - Added timing, reportTimes()
* rational-solver.inl - Added timing
* vector-fraction.h - Added comment, about how RandIter may have unrandom low-order bits
* linbox/blackbox/apply.h - Removed bad overload of applyV
* linbox/fflapack/fflapack.h - changed name of function to LQUPtoInverseOfFullRankMinor
* linbox/field/modular-double.h - added constr. from long int, to avoid ambiguity warning when making t-rdisolve
* linbox/matrix/dense.inl - attempted to fix problems with Const{Row/Col}Iterator operator++ (int)
* linbox/matrix/factorized-matrix.h - added comment about how GetQ returns Qt
* linbox/util/timer.C - added a count field to the timer (how many intervals does this timer measure)
* linbox/util/timer.h - added a count, stopped small negative rounding errors from printing
* gmp++/gmp++_int_cstor.C - added mpz_import function wrapper
* tests/test-common.C - fixed reading of large integers
* examples/solver/t-rdisolve.C - added flag to make some columns zero, made better random entries
2004-08-11 bds
* examples/blackbox/smith.C illustrates iliopolous' and local methods
* matrix/dense.inl read() now adjusts to new input size rather
than complaining. Also read() accepts sparse file format.
2004-08-11 David Pritchard ([email protected], [email protected])
* gmp++/gmp++_int.inl - Added nth root function declaration
* linbox/algorithms
* diophantine-solver.inl - minor changes
* lifting-container.h - fixed reconstruction lifting bound
* rational-reconstruction.h - added getRational2; will eventually be las-vegas style reconstructor
- changed early termination so that 2*numbound*denbound is always less than modulus
* rational-solver.inl - more updates, fixes. now only one LQUP is used to check consistency
- added optimization wherein LQUP is used to compute a modular inverse more quickly
* linbox/fflapack/fflapack.h - added InverseFromLQUP function; fixed invL bug
* linbox/matrix/blas-matrix.h - moved indexDomain here (allows application of BlasPermutation to indices, using BlasMatrixDomain<indexDomain>)
* linbox/matrix/dense-submatrix.inl - minor fix
* linbox/util/gmp++/gmp++_int_misc.C - added nth root function
2004-08-09 Dan Roche
* gmp++/gmp++_int.inl: Add arithmetic function support for long long and unsigned long long
* linbox/blackbox/dense.h,inl: Add hadamardBound method to DenseMatrixFactory
* linbox/blackbox/sparse.h: Make hadamardBound return a reference to res, and avoid overflow in computing hadamardBound
* linbox/solutions/det.h: little fixes
2004-08-09 Pascal Giorgi
* /linbox/blackbox/sparse.h : Fix bugs in SparseMatrixFactory. it compiles right now
* /linbox/util/gmp++/gmp++_int_io.C : avoid memory leaks in Integer2string function
* /linbox/field/ntl-zz_pE.h : adding new wrapper for NTL algebraic extension of single precision prime field
* /linbox/field/ntl-ZZ_pE.h : write a class to handle parametric field
* /linbox/field/ntl.h, Makefile.am : adding new field ntl-zz_pE.h
2004-08-08 Dan Roche ([email protected])
* examples/linbox-program.C - now works for determinant also - except that someone recently broke solutions/det.h and blackbox/sparse.h
* gmp++/gmp++_int.h, linbox/util/gmp++/gmp++_int_misc.C - added prevprime so that prime-stream.h works
* linbox/integer.h - added an abs function here so it works for all types from int8 up to integer.
* linbox/algorithms/cra.h - changed invmod to inv so that this compiles
* linbox/blackbox/dense.h,.inl - added DenseBlackboxFactory class so I can use det.h to find determinant of a dense matrix over the integers.
* linbox/solutions/det.h - NUM_BITS is determined from FieldTraits (linbox/field/field-traits.h)
2004-08-06 David Pritchard ([email protected], [email protected])
* examples/solver/t-rdisolve.C - added use of 3 'solver levels' and check of certificate of inconsistency
* linbox/algorithms
* diophantine-solver.h - added use of 3 'solver levels', improved documentation
* diophantine-solver.inl - added use of 3 solver levels
* lifting-container.h - changed normblackbox to properly compute "norm" of composed boxes
* rational-reconstruction.h - added getRational2 for solver use, hopefully will be las-vegas
* rational-solver.h - added use of 3 solver levels, certifying of inconsistency, documentation
* rational-solver.inl - added use of 3 solver levels, certifying inconsistency
- combined 'solve' and 'getRandomSolution' to be both variants of 'solveMonolithic'
* vector-fraction.h - simplified constructor
* linbox/fflapack/fflapack_ludivine.inl - bugfix for LQUP of rectangular matrices (specifically M x 1 case)
* linbox/field/unparametric.h - removed dumb comment
* linbox/matrix/blas-matrix.h - added method extendTrivially to embed permutation in larger row/column size
* linbox/tests/test-common.C - bugfix
2004-07-30 David Pritchard ([email protected], [email protected])
* examples/solver/t-rdisolve.C - added diophantine checking+certification, more options
- moved VectorFraction class to own file in algorithms
- moved new command-line stuff into test-common.C (should be backwards-compatible)
* linbox/algorithms
* vector-fraction.h - added new file. for storing fractions like [n1 n2 n3 n4 ... ]/d
- has various combination functions used by diophantine-solver
* blas-domain.inl - fixed a broken linbox_check for vector-matrix product
* diophantine-solver.h - diophantine checking, min-denom certification works now
* diophantine-solver.inl - diophantine checking, min-denom certification works now
* rational-solver.h - made SolverReturnStatus a global enum, preceded by SS_
* rational-solver.inl - added generation of a partial certificate to randomSolve
- solveDeterministic is now fully Las Vegas
* apply.h - added applyVTrans(vector, const matrix, vector)
* linbox/fflas/fflas_fgemv.inl - fixed errors in transpose matrix product
* linbox/matrix
* blas-matrix.h - fixed flipped row/coldim in constructor
* dense-submatrix.h - fixed typo, removed unimplemented operator[]
* dense.h - removed debug output
* transpose.h - replaced Column with Col
* tests/test-common.h - example string can now be passed as null and intelligently generated by default
* tests/test-common.C - boolean options can now be turned off like "-b false", "-b -", "-b N"
- added a string to write all current switch options to an ostream
* tests/test-rational-solver.C - updated to reflect new return type of rational solver
2004-07-26 David Pritchard ([email protected], [email protected])
* linbox/algorithms
* added diophantine-solver.{h,inl} (no diophantine functionality yet)
* Makefile.am - added diophantine-solver.h
* blas-domain.h - added forms of inv and invin that return the nullity of the matrix
* blas-domain.inl - added forms of inv and invin that return the nullity of the matrix, fixed typo
* last-invariant-factor.h - changed type of b vector from int to Integer
* lifting-container.h - changed blackbox norm to square of 'induced norm', fixed lifting length calculation
- added graceful failure to digit (eg if <double> Ring is used improperly)
* rational-reconstruction.h - added debug switch DEBUG_RR, graceful failure
* matrix-rank.h - changed tabbing
* rational-solver.h - many changes to Dixon; Wiedemann may not work
- added DEFAULT_PRIMESIZE = 14 (since 30 bit default size would screw up some fields)
- added maxPrimes argument to solving routines, will retry that many primes
- added const modifier to solving routines, made _prime mutable
* rational-solver.inl - many changes to Dixon; Wiedemann may not work
- added debug switches, graceful failure for solving routines
- changed BlasMatrixDomain routines to BlasApplyDomain to avoid erroneous calculations
- in singular/random solve, check solution before returning, and multiply by conditioner P correctly
- added consistency check to random solver
* linbox/blackbox/blas-blackbox - added to CVS
* linbox/blackbox/dense.h - matrix reading bugfix
* linbox/fflapack/fflapack.h - pass nullity back in Invert2, and return NULL when matrix is singular
* linbox/fflas/fflas_fgemv.inl - remove assumption that alpha=-1, beta=1
* linbox/fflas/fflas_ftrmm.inl - fixed computation of matrix size bound for embedding into FFLAS
* linbox/field
* PID-double.h - changed 'c=abs(c)' to 'if (c<0) c = -c', no abs found for double
* PID-integer.h - changed documentation of isDivisor(a, b) (returns yes if b|a)
* givaro-montg.h - fixed init from integer
* givaro-zpz.h - fixed init from integer, double and Log16 specializations
* givaro-gfq.h - fixed init from integer
* modular.h - added init from double
- added several specializations of Modular<NTL::ZZ>
* modular-double.h - fixed operator<<
* modular-int.h - fixed init/convert with double
* modular-int32.h - fixed init/convert with double
* ntl-ZZ.h - changed const to static for convert method
- changed documentation of isDivisor(a, b) (returns yes if b|a)
* ntl-ZZ_p.h - added init/convert from/to integer, double
- put NTL_ZZ_p at top and specialization at bottom for compiler
- changed random generator to give fuller range
* ntl-zz_p.h - added init from double, integer
* unparametric.h - added convert to double
* linbox/matrix/dense-submatrix.h - added mapleFormat option to DenseSubmatrix writing
* linbox/matrix/dense-submatrix.inl - added mapleFormat option to DenseSubmatrix writing
* linbox/randiter/random-prime.h - the class now takes a 'bits' argument on construction and makes primes of that size
* linbox/vector/vector-domain.inl - fixed reading bug
* tests/Matio.h - added mapleFormat option to write_field
* tests/test-fflapack.C - matched Invert2 to new 'nullity' argument format
* tests/test-iliopoulos-elimination.C - #define int32 LinBox::int32 to avoid ambiguity with Givaro
* tests/test-last-invariant-factor.C - int32 -> LinBox::int32 to avoid ambiguity with Givaro
* tests/test-last-rational-solver.C - int32 -> LinBox::int32 to avoid ambiguity with Givaro
- now handles solver failures without crashing
- added check to ensure that system is solvable (ie no zero elements on diagonal)
* tests/test-smith-form.C - int32 -> LinBox::int32 to avoid ambiguity with Givaro
* examples/solver/t-rdisolve.C - test driver for rational solver, compile and run t-rdisolve -?
* examples/solver/Makefile.am - make file
2004-07-16 Pascal Giorgi
* linbox/field/lidia-gfq.h : fixing bug and improvement
* tests/Makefile.am : retrieving test-lidia-gfq, fixing modular and gssv redundancy
* tests/test-modular.C : using namespace std
2004-07-16 Pascal Giorgi
* linbox-config.in : switching order of LiDIA in libs
* linbox/randiter/envelope.h : modify random function ( assignement is not needed)
* linbox/matrix/dense-submatrix.h : setting input and output functions public
* linbox/field/givaro-zpz.h : adding linbox/utils/debug.h header
* linbox/field/lidia-gfq.h : setting default degree extension to 1
* linbox/field/ntl-ZZ_p.h : overloading the constructor of UnparamField
2004-07-08 Dan Roche
* Adding examples/linbox-program.C which will eventually do a wide variety of operations on integer matrices
* gmp++/gmp++_int.h,inl : Little fixes
* linbox/algorithms/wiedemann.inl : Little fixes
* linbox/matrix/blas-matrix.h : Little fixes
* linbox/util/gmp++/gmp++_int_compare.C, gmp++_int_mod.C : Little fixes
* linbox/vector/sparse.h : Little fixes
* tests/test-fields.C : Little fixes
* tests/Makefile.am : Moved everything that uses rational-solver to ATLAS_TESTS rational-solver won't work without BLAS.
* fields/field-traits.h : Took implementation out of this file and put it in each field header file.
* fields/givaro-gfq.h : Added init and convert from/to a padic number
* fields/modular.inl : Fixed some problems with MVProductDomain functions over unsigned types
2004-07-06 Bradford Hovinen <[email protected]>
* linbox/blackbox/dense.h (class DenseMatrix): Add _AT, a
TransposeMatrix of this
* linbox/blackbox/dense.inl (DenseMatrix::applyTranspose): Use _AT
* linbox/matrix/matrix-domain.h (class MatrixDomain): Get rid of
template argument MatrixTrait on specialized permute
* linbox/vector/vector-domain.h (class VectorDomain): Get rid of
template argument VectorTrait on specialized permute, swap
* linbox/matrix/sparse.h (struct MatrixTraits):
* linbox/vector/vector-domain.inl (class VectorDomain):
* linbox/vector/vector-domain.h (class VectorDomain):
* linbox/matrix/matrix-domain.h (class MatrixDomain):
* linbox/matrix/transpose.h (struct MatrixTraits):
* linbox/matrix/dense.h (struct MatrixTraits): Remove templates
from matrix tags
2004-07-02 pascal giorgi
* blackbox/apply.h : adding BlasApply domain
* blackbox/compose.h : adding accessor to left an right blackbox
* blackbox/sparse.h : make write function const
* field/ all givaro file : fixe bugs
* field/lidia-gfq.h : make it works
* field/modular-double.h : add reset function in axpy domain
* field/field-traits.h : add specialization for some fields
* field/double-fmod.h : include <math.h>
* field/ntl-ZZ.h : remove all static and replace by const function
* field/PID-double field/PID-integer.h : adding 2 new PID
* matrix/dense.h.inl matrix/sparse.h.inl : make interators work
* randiter/ntl-ZZ.h randiter/lidia-gfq.h : fix bugs and make clearer
* solutions/method.h : add new traits for rational solvers
* algorithms/
* blackbox-container-base.h : adding new constructor allowing size of sequence as parameter
* blackbox-container.h : adding new constructor allowing size of sequence as parameter
* lifting-containe.h : make a new interface for lifting (give specialization for Wiedemann and Dixon methods)
* matrix-inverse.h : put Field as template and require a field as parameter (not restrictive to blackboxes)
* matrix-mod.h : giving a specializatino
* rational-reconstruction.h : make it clear with lifting interface
* rational-solver.h : make a new interface for rational solver