1Version 2.0.30 January 23, 2021 2 3 - Added support for lis_esolver_get_specific_*() (thanks to Peter Vermeulen). 4 5Version 2.0.29 December 28, 2020 6 7 - Fixed MPI related bug in configure.ac for macOS (thanks to Carlo de Falco). 8 9Version 2.0.28 December 24, 2020 10 11 - Updated note on win32.mak in win/README. 12 - Added experimental support for Fujitsu A64FX and NEC SX-Aurora TSUBASA. 13 14Version 2.0.27 December 3, 2020 15 16 - Added note on win32.mak to win/README (thanks to Zhengang Liu). 17 18Version 2.0.26 November 25, 2020 19 20 - Added option --disable-test to configure.ac (thanks to Fabian Wein). 21 22Version 2.0.25 November 18, 2020 23 24 - Updated configure.ac for Apple Clang (thanks to Fabian Wein). 25 26Version 2.0.24 November 3, 2020 27 28 - Fixed bugs in lis_error.c and lis_esolver_si.c. 29 30Version 2.0.23 August 26, 2020 31 32 - Fixed bug in lis_input_vector_mm(). 33 34Version 2.0.22 May 22, 2020 35 36 - Updated documents. 37 38Version 2.0.21 March 18, 2020 39 40 - Fixed MPI related bug in test4f.F. 41 42Version 2.0.20 September 4, 2019 43 44 - Disabled unimplemented conv_cond options for some linear solvers 45 (thanks to Kazuhiro Tada). 46 47Version 2.0.19 May 22, 2019 48 49 - Applied Lis patch for K computer (thanks to Shun Ito). 50 51Version 2.0.18 January 30, 2019 52 53 - Updated libtool to 2.4.6 (thanks to Martin Sjoelund). 54 - Updated documents (thanks to Danyang Su). 55 56Version 2.0.17 October 26, 2018 57 58 - Updated documents (thanks to Yasuyo Hatano). 59 60Version 2.0.16 September 26, 2018 61 62 - Disabled option '-f quad' when quad precision is not enabled (thanks to Konstantin). 63 64Version 2.0.15 September 20, 2018 65 66 - Updated copyright notice (thanks to Konstantin). 67 68Version 2.0.14 July 11, 2018 69 70 - Updated .gitignore (thanks to Oliver Evans). 71 72Version 2.0.13 June 28, 2018 73 74 - Added OMPFLAGs to configurations for Intel MIC (thanks to Javier Bahamonde Amor). 75 76Version 2.0.12 May 24, 2018 77 78 - Added test matrix testmat4.mtx in Matrix Market array format. 79 80Version 2.0.11 April 26, 2018 81 82 - Fixed lis.h for compatibility with C++ (thanks to Michael Asam). 83 84Version 2.0.10 April 12, 2018 85 86 - Updated graphics for complex eigenvalues. 87 88Version 2.0.9 March 21, 2018 89 90 - Fixed configuration of win/Makefile.in for SAAMG (thanks to Jian Li). 91 92Version 2.0.8 March 9, 2018 93 94 - Updated descriptions for test matrices. 95 96Version 2.0.7 February 8, 2018 97 98 - Fixed matrix block size related bugs in spmvtest4 and spmvtest5 99 (thanks to Javier Bahamonde Amor). 100 101Version 2.0.6 December 10, 2017 102 103 - Added complex test matrix testmat3.mtx. 104 - Updated print messages for Subspace, Lanczos, and Arnoldi iterations. 105 - Fixed bug in lisf.h.in. 106 107Version 2.0.5 December 7, 2017 108 109 - Updated documents. 110 111Version 2.0.4 November 18, 2017 112 113 - Fixed memory leak bugs related to option -rval. 114 115Version 2.0.3 November 11, 2017 116 117 - Updated lis_matrix_destroy() to avoid conflict with deallocate() in 118 Fortran 90 (thanks to Ste'phane Glockner). 119 120Version 2.0.2 November 6, 2017 121 122 - Changed name of function lis_matrix_shift_general() to lis_matrix_shift_matrix(). 123 - Updated test8f.F90. 124 125Version 2.0.1 November 5, 2017 126 127 - Added support for option -rval. 128 - Added support for graphics/gnuplot3b*. 129 - Fixed shift related bugs. 130 131Version 2.0.0 November 1, 2017 132 133 - Added official support for linear solvers generalized for complex matrices. 134 - Added official support for nonsymmetric eigenproblems. 135 136Version 1.8.15 November 1, 2017 137 138 - Fixed complex related bugs in BiCR. 139 140Version 1.8.14 October 31, 2017 141 142 - Fixed bugs on complex arithmetic. 143 144Version 1.8.13 October 30, 2017 145 146 - Fixed bugs in lis_input_mm.c. 147 148Version 1.8.12 October 25, 2017 149 150 - Generalized linear solvers and preconditioners for complex matrices. 151 152Version 1.8.11 October 18, 2017 153 154 - Fixed bug in lis_esolver_ai.c. 155 - Fixed bug in lis_array.c. 156 - Fixed description in etest7.c. 157 - Added support for Matrix Market array format. 158 159Version 1.8.10 October 17, 2017 160 161 - Updated configurations in win/Makefile.in. 162 163Version 1.8.9 October 11, 2017 164 165 - Updated configurations in win/Makefile.in for Microsoft MPI. 166 - Fixed SAAMG related configurations in win/Makefile.in. 167 168Version 1.8.8 October 9, 2017 169 170 - Updated configurations in win/Makefile.in for Intel Compilers. 171 172Version 1.8.7 October 4, 2017 173 174 - Fixed configure.ac to exclude complex.h by default (thanks to Fuyuki Saito). 175 - Fixed configurations in win/Makefile.in. 176 177Version 1.8.6 September 24, 2017 178 179 - Fixed bug in win/Makefile.in (thanks to Jian Li). 180 - Added support for inputting imaginary shift for eigensolvers. 181 182Version 1.8.5 September 14, 2017 183 184 - Updated documents. 185 186Version 1.8.4 August 29, 2017 187 188 - Fixed documents. 189 190Version 1.8.3 August 10, 2017 191 192 - Updated documents. 193 - Updated eigensolvers. 194 195Version 1.8.2 August 6, 2017 196 197 - Added support for inputting complex vectors in Matrix Market format. 198 199Version 1.8.1 August 4, 2017 200 201 - Fixed bug in lis_solver_sor.c. 202 - Changed names of flags -ssor_w and -hybrid_w to -ssor_omega and -hybrid_omega, 203 respectively. 204 205Version 1.8.0 August 1, 2017 206 207 - Added official support for test/test3c.c for solving multiple-step linear equations. 208 209Version 1.7.37 July 31, 2017 210 211 - Fixed MPI bug in test/test3c.c. 212 - Fixed bug in lis_solver_sor.c. 213 214Version 1.7.36 July 27, 2017 215 216 - Fixed MPI memory leak bug in lis_matrix_assemble() (thanks to Danyang Su). 217 - Fixed lis_input_vector() related bug in test1.c (thanks to Danyang Su). 218 219Version 1.7.35 July 26, 2017 220 221 - Fixed MPI memory leak bug in lis_matrix_assemble() (thanks to Danyang Su). 222 223Version 1.7.34 July 17, 2017 224 225 - Fixed MPI bugs in lis_vector_gather() and lis_vector_scatter() (thanks to Danyang Su). 226 227Version 1.7.33 June 11, 2017 228 229 - Added exception handling for inner linear solvers. 230 231Version 1.7.32 June 10, 2017 232 233 - Fixed exception handling bug. 234 235Version 1.7.31 June 7, 2017 236 237 - Fixed exception handling bug (thanks to Ralf Greve). 238 239Version 1.7.30 May 31, 2017 240 241 - Added exception handlings for wrong parameter settings (thanks to Ralf Greve). 242 - Updated documents. 243 244Version 1.7.29 March 22, 2017 245 246 - Added experimental support for test/test3c.c, which is an update of test/test4b.c 247 for solving multiple-step linear equations (thanks to Fabian Wein). 248 - Updated documents. 249 250Version 1.7.28 March 12, 2017 251 252 - Added support for Intel MIC (thanks to Javier Bahamonde Amor). 253 - Updated documents. 254 255Version 1.7.27 February 17, 2017 256 257 - Fixed matrix file format for lis_esolver_get_evectors() (thanks to Takahiro Okada). 258 - Updated documents. 259 260Version 1.7.26 February 10, 2017 261 262 - Added support for lis_*solver_get_rhistory() in Fortran (thanks to Thomas Goelles). 263 - Updated documents (thanks to Thomas Goelles). 264 265Version 1.7.25 February 3, 2017 266 267 - Updated documents (thanks to Hiroshi Kanayama). 268 269Version 1.7.24 January 25, 2017 270 271 - Updated test/test4b.c. 272 273Version 1.7.23 January 22, 2017 274 275 - Added support for test/test4b.c (thanks to Ganesh Natarajan). 276 - Updated test/spmvtest1.c. 277 278Version 1.7.22 January 13, 2017 279 280 - Updated documents (thanks to Ganesh Natarajan). 281 282Version 1.7.21 January 2, 2017 283 284 - Fixed descriptions. 285 - Updated documents. 286 287Version 1.7.20 December 19, 2016 288 289 - Fixed type of argc in lis_initialize() to avoid conflict with MPI_Init(). 290 291Version 1.7.19 December 19, 2016 292 293 - Fixed critical bug in lis_init.c. 294 - Fixed long long int related bugs. 295 - Fixed long double related bugs. 296 297Version 1.7.18 December 15, 2016 298 299 - Fixed types for print statements. 300 301Version 1.7.17 December 14, 2016 302 303 - Fixed type related MPI bugs in Fortran programs. 304 305Version 1.7.16 December 11, 2016 306 307 - Added support for GCC libquadmath. 308 - Fixed quad precision bugs in Fortran programs. 309 310Version 1.7.15 December 6, 2016 311 312 - Fixed MPI bug in lis_error(). 313 - Fixed print bugs. 314 - Changed specification of lis_print_rhistory(). 315 316Version 1.7.14 December 3, 2016 317 318 - Fixed type related MPI bugs. 319 - Fixed bug in src/system/lis_error.c. 320 - Updated src/system/lis_input.c. 321 - Updated documents. 322 323Version 1.7.13 November 22, 2016 324 325 - Updated documents (thanks to Wei Liu). 326 327Version 1.7.12 November 11, 2016 328 329 - Fixed type related print bugs in lis_error.c. 330 - Fixed bug in spmvtest3b.c. 331 332Version 1.7.11 November 10, 2016 333 334 - Fixed type related print bugs in test/spmvtest*.c. 335 336Version 1.7.10 November 8, 2016 337 338 - Fixed type related print bug in lis_esolver.c. 339 340Version 1.7.9 November 7, 2016 341 342 - Cleaned up print statements using lis_printf(). 343 - Fixed type related print bugs in error messages. 344 345Version 1.7.8 November 4, 2016 346 347 - Fixed installation bugs in win/Makefile.in. 348 - Updated documents. 349 350Version 1.7.7 November 3, 2016 351 352 - Fixed bug in generalized Lanczos. 353 354Version 1.7.6 November 1, 2016 355 356 - Fixed bug in generalized CG. 357 358Version 1.7.5 October 29, 2016 359 360 - Added support for generalized Rayleigh quotient iteration, 361 generalized CG, and generalized Arnoldi iteration. 362 363Version 1.7.4 October 26, 2016 364 365 - Fixed index bug in lis_esolver_ai.c. 366 367Version 1.7.3 October 25, 2016 368 369 - Fixed shift related bugs. 370 - Added support for option -ige for generalized eigenproblems. 371 - Disabled complex operations if complex.h is not available. 372 - Updated documents. 373 374Version 1.7.2 October 24, 2016 375 376 - Added support for generalized Lanczos iteration. 377 378Version 1.7.1 October 23, 2016 379 380 - Inverted sign of shift for eigensolvers in accordance with custom. 381 - Inverted sign of shift in lis_matrix_shift_diagonal(), lis_vector_shift(), 382 and lis_array_shift(), respectively. 383 - Added support for shift for generalized eigenproblems. 384 - Updated man pages. 385 386Version 1.7.0 October 22, 2016 387 388 - Added official support for generalized eigenvalue problems. 389 - Added support for generalized subspace iteration. 390 391Version 1.6.39 October 21, 2016 392 393 - Added experimental support for generalized power iteration. 394 395Version 1.6.38 October 20, 2016 396 397 - Fixed print statements in eigensolvers for option -eprint none. 398 - Added experimental support for generalized inverse iteration. 399 400Version 1.6.37 October 19, 2016 401 402 - Updated test6f.F90. 403 - Added experimental support for generalized CR. 404 405Version 1.6.36 October 17, 2016 406 407 - Updated test6.c (thanks to Hiroshi Kanayama). 408 - Updated lisf_*.c (thanks to Florian Lehner). 409 - Updated lis_esolver.c. 410 411Version 1.6.35 October 16, 2016 412 413 - Updated notations of eigensolvers. 414 - Fixed bug of option -shift. 415 - Deleted approximate inverse iteration from eigensolvers. 416 417Version 1.6.34 October 14, 2016 418 419 - Updated documents (thanks to Hiroshi Kanayama). 420 421Version 1.6.33 October 13, 2016 422 423 - Added support for lis_gesolve(). 424 425Version 1.6.32 October 12, 2016 426 427 - Updated lis_array_*(). 428 - Updated man pages. 429 430Version 1.6.31 October 11, 2016 431 432 - Updated documents (thanks to Hiroshi Kanayama). 433 - Fixed bug in Makefile.am. 434 435Version 1.6.30 October 10, 2016 436 437 - Updated documents (thanks to Hiroshi Kanayama). 438 - Updated man pages. 439 - Updated include/lis.h. 440 441Version 1.6.29 October 7, 2016 442 443 - Fixed quad precision bug in lis_solver.c (thanks to Yonggwi Cho). 444 445Version 1.6.28 October 6, 2016 446 447 - Updated eigensolvers. 448 449Version 1.6.27 October 5, 2016 450 451 - Updated lis_esolver_ai.c. 452 453Version 1.6.26 October 5, 2016 454 455 - Fixed bugs in test/*.F. 456 457Version 1.6.25 October 2, 2016 458 459 - Fixed bugs in test/test*.F90. 460 461Version 1.6.24 September 30, 2016 462 463 - Fixed bugs in test/test1f.F. 464 465Version 1.6.23 September 29, 2016 466 467 - Added support for COCG and COCR. 468 469Version 1.6.22 September 28, 2016 470 471 - Added support for lis_vector_nhdot(). 472 473Version 1.6.21 September 27, 2016 474 475 - Fixed bug in lis_esolver.c. 476 477Version 1.6.20 September 25, 2016 478 479 - Updated documents (thanks to Hiroaki Morimoto). 480 481Version 1.6.19 September 24, 2016 482 483 - Fixed complex related bugs. 484 485Version 1.6.18 September 23, 2016 486 487 - Fixed complex related bugs. 488 489Version 1.6.17 September 23, 2016 490 491 - Fixed complex related bugs. 492 493Version 1.6.16 September 21, 2016 494 495 - Fixed complex related bugs. 496 497Version 1.6.15 September 20, 2016 498 499 - Fixed complex related bugs. 500 501Version 1.6.14 September 19, 2016 502 503 - Fixed complex related bugs. 504 505Version 1.6.13 September 19, 2016 506 507 - Fixed complex related configuration in lis_solver_cg.c. 508 509Version 1.6.12 September 16, 2016 510 511 - Updated win/*. 512 - Fixed memory leak bug in lis_esolver_cg.c. 513 514Version 1.6.11 September 15, 2016 515 516 - Updated win/* (thanks to Hiroaki Morimoto). 517 - Updated lis_input_hb.c. 518 519Version 1.6.10 September 14, 2016 520 521 - Added support for outputting complex matrices. 522 - Added support for reading Hermitian matrices. 523 524Version 1.6.9 September 13, 2016 525 526 - Fixed bugs in lis_input_hb.c. 527 - Updated documents. 528 529Version 1.6.8 September 12, 2016 530 531 - Fixed critical bug in lis_input_mm.c for reading symmetric matrices. 532 533Version 1.6.7 September 11, 2016 534 535 - Added support for lis_vector_conjugate(). 536 537Version 1.6.6 September 10, 2016 538 539 - Fixed bug in win/Makefile.in. 540 - Updated test7.c and test7f.F. 541 542Version 1.6.5 September 9, 2016 543 544 - Fixed complex related bug on fabs(). 545 - Updated test7.c and test7f.F. 546 547Version 1.6.4 September 8, 2016 548 549 - Fixed bug in lisf.h.in. 550 - Updated test7.c and test7f.F. 551 552Version 1.6.3 September 7, 2016 553 554 - Added support for lis_solver_get_status() in Fortran (thanks to 555 Alessandro Brovelli). 556 557Version 1.6.2 September 6, 2016 558 559 - Fixed complex related bug in test4f.F. 560 561Version 1.6.1 September 5, 2016 562 563 - Fixed quad precision bug (thanks to Florian Lehner). 564 565Version 1.6.0 September 4, 2016 566 567 - Added official support for flag --enable-complex. 568 - Fixed complex related bugs in parallel processing. 569 - Added support for inputting complex matrices in Harwell-Boeing format. 570 571Version 1.5.79 September 3, 2016 572 573 - Fixed complex related bugs. 574 575Version 1.5.78 September 2, 2016 576 577 - Added support for outputting complex vectors. 578 - Updated documents. 579 580Version 1.5.77 September 1, 2016 581 582 - Added support for inputting complex matrices in Matrix Market format. 583 584Version 1.5.76 August 31, 2016 585 586 - Fixed complex related bugs. 587 588Version 1.5.75 August 31, 2016 589 590 - Added experimental support for flag --enable-complex for complex scalar. 591 - Updated documents. 592 593Version 1.5.74 August 26, 2016 594 595 - Fixed configurations for complex.h. 596 597Version 1.5.73 August 25, 2016 598 599 - Fixed configurations for complex numbers. 600 601Version 1.5.72 August 25, 2016 602 603 - Fixed bug in test7f.F. 604 605Version 1.5.71 August 25, 2016 606 607 - Fixed bug in Fortran configuration for complex*32. 608 609Version 1.5.70 August 25, 2016 610 611 - Added official support for complex numbers. 612 - Renamed test7f.F90 test8f.F90. 613 - Added test7.c and test7f.F for complex numbers. 614 - Added decralations for PSD functions. 615 616Version 1.5.69 August 21, 2016 617 618 - Fixed bug in test/Makefile.am. 619 620Version 1.5.68 August 21, 2016 621 622 - Added test7.F90 to test_PROGRAMS in test/Makefile.am. 623 624Version 1.5.67 August 20, 2016 625 626 - Added test7.F90 to test/Makefile.am. 627 628Version 1.5.66 June 17, 2016 629 630 - Added note on file format for symmetric matrices to Appendix (thanks to 631 Kodanda Ram Mangipudi). 632 633Version 1.5.65 March 19, 2016 634 635 - Fixed bug in lis_matrix_csr.c. 636 637Version 1.5.64 March 19, 2016 638 639 - Added support for alternative workflow PSD, or Preconditioner and Solver 640 Decoupled (thanks to Neil Hodge). 641 - Updated documents. 642 643Version 1.5.63 January 31, 2016 644 645 - Fixed autoreconf related bugs (thanks to Florian Lehner). 646 647Version 1.5.62 January 7, 2016 648 649 - Changed names of preprocessor macros _LONGLONG and LONGLONG to 650 _LONG__LONG and LONG__LONG, respectively, to avoid conflict with __int64. 651 652Version 1.5.61 January 7, 2016 653 654 - Fixed __int64 related bug in lis.h for Visual C++ 655 (thanks to Takeshi Takeda). 656 657Version 1.5.60 January 2, 2016 658 659 - Deleted "-O3" options associated with Intel compilers 660 so that developers can define optimization themselves 661 (thanks to Neil Hodge). 662 - Moved inclusion of math.h from lis.h to lis_solver.c 663 (thanks to Neil Hodge). 664 - Deleted duplicate defines in lis.h 665 (thanks to Neil Hodge). 666 - Fixed bug in lis_init.c associated with extra whitespace between 667 solver options (thanks to Neil Hodge). 668 - Modified code in lis_vector.c and lis_matrix.c to allow for the case 669 of zero DOFs on a process (thanks to Neil Hodge). 670 - Updated documents. 671 672Version 1.5.59 December 15, 2015 673 674 - Added include statement for math.h to lis.h 675 (thanks to Neil Hodge). 676 677Version 1.5.58 November 31, 2015 678 679 - Added include statements for math.h to lis_solver.c and lis_esolver.c 680 (thanks to James Liu). 681 682Version 1.5.57 July 7, 2015 683 684 - Added support for lis_matrix_unset() in C (thanks to Norihiro Watanabe). 685 686Version 1.5.56 June 25, 2015 687 688 - Updated documents. 689 - Added experimental support for complex numbers. 690 691Version 1.5.55 May 14, 2015 692 693 - Added support for lis_solver_get_precon() and lis_solver_get_preconname() 694 in Fortran (thanks to Domenic D'Ambrosio). 695 - Updated documents (thanks to Domenic D'Ambrosio). 696 697Version 1.5.54 April 30, 2015 698 699 - Fixed configurations for Windows (thanks to Domenic D'Ambrosio). 700 - Added support for combining Intel Fortran Compiler with Visual C 701 (thanks to Domenic D'Ambrosio). 702 703Version 1.5.53 April 5, 2015 704 705 - Updated documents (thanks to Sushilkumar Sreekumar). 706 707Version 1.5.52 March 28, 2015 708 709 - Updated documents (thanks to Sushilkumar Sreekumar). 710 711Version 1.5.51 February 10, 2015 712 713 - Updated documents. 714 715Version 1.5.50 February 9, 2015 716 717 - Updated documents (thanks to Sushilkumar Sreekumar). 718 719Version 1.5.49 February 6, 2015 720 721 - Updated documents (thanks to Carlo de Falco). 722 723Version 1.5.48 February 6, 2015 724 725 - Fixed exception handling bug in lis_matrix_set_size() (thanks to 726 Carlo de Falco). 727 728Version 1.5.47 January 10, 2015 729 730 - Removed unused variables in lis_precision_vec.c. 731 732Version 1.5.46 January 8, 2015 733 734 - Fixed quad precision bug in lis_precision_vec.c. 735 - Updated config/config.*. 736 - Cleaned up programs. 737 738Version 1.5.45 December 30, 2014 739 740 - Added documents for matrix addition. 741 - Updated documents. 742 743Version 1.5.44 December 29, 2014 744 745 - Fixed exception handling in matrix addition. 746 - Updated documents. 747 748Version 1.5.43 December 28, 2014 749 750 - Fixed Fortran functions for matrix operations. 751 752Version 1.5.42 December 27, 2014 753 754 - Added support for matrix addition. 755 756Version 1.5.41 December 23, 2014 757 758 - Fixed bug in win/Makefile.in. 759 760Version 1.5.40 December 20, 2014 761 762 - Fixed variable declaration bug in lis_precision_vec.c. 763 764Version 1.5.39 December 16, 2014 765 766 - Updated documents. 767 768Version 1.5.38 December 15, 2014 769 770 - Updated documents. 771 772Version 1.5.37 December 14, 2014 773 774 - Cleaned up programs (thanks to Florian Lehner). 775 776Version 1.5.36 December 11, 2014 777 778 - Updated lis_esolver_rqi.c. 779 780Version 1.5.35 December 10, 2014 781 782 - Updated documents (thanks to Sushilkumar Sreekumar). 783 784Version 1.5.34 December 8, 2014 785 786 - Updated documents. 787 - Changed default preconditioners for eigensolvers. 788 789Version 1.5.33 December 7, 2014 790 791 - Updated documents. 792 - Added experimental support for Jacobi-Davidson. 793 - Fixed exception handling bug in lis_matrix_set_size() (thanks to Yi Yao). 794 795Version 1.5.32 December 5, 2014 796 797 - Updated documents. 798 799Version 1.5.31 December 4, 2014 800 801 - Updated etest7.c. 802 - Updated lis_esolver_ai.c. 803 804Version 1.5.30 December 3, 2014 805 806 - Updated documents. 807 - Fixed print bug on iteration counts of inner eigensolvers. 808 809Version 1.5.29 December 2, 2014 810 811 - Added experimental support for Arnoldi iteration. 812 - Disabled obsolete option -m for eigensolvers. 813 - Updated documents. 814 815Version 1.5.28 December 1, 2014 816 817 - Changed default linear solver for eigensolvers. 818 - Changed order of eigensolvers: 819 820 flags | eigensolvers flags | eigensolvers 821 ------------+-------------------- ------------+-------------------- 822 -e {si|5} | Subspace ==> -e {cg|5} | CG 823 -e {li|6} | Lanczos -e {cr|6} | CR 824 -e {cg|7} | CG -e {si|7} | Subspace 825 -e {cr|8} | CR -e {li|8} | Lanczos 826 827 - Updated documents. 828 829Version 1.5.27 November 30, 2014 830 831 - Updated documents. 832 833Version 1.5.26 November 29, 2014 834 835 - Changed names of functions lis_array_matvec2() and 836 lis_array_matmat2() to lis_array_matvec_ns() and lis_array_matmat_ns(), 837 respectively. 838 839Version 1.5.25 November 28, 2014 840 841 - Updated lis_array.c. 842 843Version 1.5.24 November 27, 2014 844 845 - Updated lis_array.c (thanks to Florian Lehner). 846 847Version 1.5.23 November 26, 2014 848 849 - Cleaned up lis_esolver_li.c. 850 - Cleaned up lis_array.c. 851 - Fixed convergence condition of lis_array_qr(). 852 853Version 1.5.22 November 25, 2014 854 855 - Updated documents. 856 857Version 1.5.21 November 24, 2014 858 859 - Cleaned up lis_solver_*.c. 860 - Added support for test6f.F90 for array operations. 861 - Updated documents. 862 863Version 1.5.20 November 23, 2014 864 865 - Updated documents. 866 867Version 1.5.19 November 22, 2014 868 869 - Fixed bug in lisf_array.c (thanks to Florian Lehner). 870 871Version 1.5.18 November 21, 2014 872 873 - Updated documents. 874 875Version 1.5.17 November 20, 2014 876 877 - Updated lis_fortran.h for Win64 environment. 878 - Fixed initial value related bug in GMRES to enable option -initx_zeros. 879 880Version 1.5.16 November 19, 2014 881 882 - Updated documents (thanks to Sushilkumar Sreekumar). 883 884Version 1.5.15 November 18, 2014 885 886 - Fixed order of array elements in lis_array_cgs(), lis_array_mgs(), 887 and lis_array_qr(). 888 - Fixed memory initialization bug in lis_array_mgs(). 889 - Updated documents. 890 891Version 1.5.14 November 16, 2014 892 893 - Cleaned up programs. 894 - Updated eigensolvers. 895 - Added sample scripts for graphics. 896 897Version 1.5.13 November 14, 2014 898 899 - Updated documents. 900 901Version 1.5.12 November 13, 2014 902 903 - Updated documents. 904 905Version 1.5.11 November 12, 2014 906 907 - Updated test3b.c. 908 909Version 1.5.10 November 11, 2014 910 911 - Added sample scripts for graphics. 912 913Version 1.5.9 November 9, 2014 914 915 - Cleaned up programs. 916 917Version 1.5.8 November 8, 2014 918 919 - Updated documents. 920 921Version 1.5.7 November 7, 2014 922 923 - Fixed print format. 924 - Fixed elapsed time related bug in preconditioned eigensolvers. 925 - Fixed CR related bug in lis_esolver_cg.c. 926 927Version 1.5.6 November 6, 2014 928 929 - Updated documents. 930 931Version 1.5.5 November 5, 2014 932 933 - Cleaned up programs. 934 - Updated documents. 935 936Version 1.5.4 November 4, 2014 937 938 - Fixed quad precision related bug in lis_esolver_si.c. 939 - Fixed bug in test/Makefile.am. 940 941Version 1.5.3 November 3, 2014 942 943 - Updated eigensolvers. 944 - Fixed array related bug in lis_esolver_cg.c. 945 946Version 1.5.2 November 1, 2014 947 948 - Added support for etest7.c for array operations. 949 - Changed specification of lis_array_qr() to return number of iterations 950 and error of QR algorithm. 951 - Updated documents. 952 953Version 1.5.1 October 31, 2014 954 955 - Added support for test6.c for array operations. 956 - Updated print formats. 957 - Updated documents. 958 959Version 1.5.0 October 30, 2014 960 961 - Fixed print bug in option -conv_cond (thanks to Suguru Miyauchi). 962 - Added official support for array operations. 963 - Updated documents. 964 965Version 1.4.67 October 28, 2014 966 967 - Fixed bug in lis_array_nrmi_f() (thanks to Florian Lehner). 968 - Updated lis_array.c. 969 970Version 1.4.66 October 27, 2014 971 972 - Updated lis_array.c. 973 - Changed names of functions lis_array_dot2() and 974 lis_array_invGauss() to lis_array_dot() and lis_array_ge(), 975 respectively. 976 977Version 1.4.65 October 23, 2014 978 979 - Updated documents (thanks to Suguru Miyauchi). 980 981Version 1.4.64 October 21, 2014 982 983 - Updated lis_m_data_creation_AMGCG.F90. 984 - Updated sample scripts for graphics. 985 986Version 1.4.63 October 19, 2014 987 988 - Added sample scripts for graphics (thanks to Takashi Shiroto 989 and Sushilkumar Sreekumar). 990 991Version 1.4.62 October 17, 2014 992 993 - Fixed exception handling bug in lis_m_data_creation_AMGCG.F90 994 (thanks to Takashi Shiroto). 995 996Version 1.4.61 September 19, 2014 997 998 - Fixed OpenMP related bug in lis_matrix_ell.c (thanks to Gerald Wedel). 999 1000Version 1.4.60 September 13, 2014 1001 1002 - Fixed long double related print bugs. 1003 - Fixed long long int related print bugs. 1004 1005Version 1.4.59 September 9, 2014 1006 1007 - Updated configure.in. 1008 - Updated test programs in Fortran. 1009 - Updated documents. 1010 1011Version 1.4.58 August 27, 2014 1012 1013 - Added support for lis_matrix_unset() in Fortran (thanks to Takumi Yanagawa 1014 and Takashi Shiroto). 1015 - Updated test/test2f.F90. 1016 1017Version 1.4.57 August 15, 2014 1018 1019 - Updated lis_array.c. 1020 1021Version 1.4.56 August 12, 2014 1022 1023 - Updated documents. 1024 - Added experimental support for array operations. 1025 1026Version 1.4.55 August 11, 2014 1027 1028 - Updated lis_matrix_ops.c. 1029 - Updated lisf_matrix.c. 1030 - Updated documents. 1031 1032Version 1.4.54 August 10, 2014 1033 1034 - Updated lisf_matrix.c. 1035 1036Version 1.4.53 August 9, 2014 1037 1038 - Fixed print bug. 1039 - Updated lis_matrix_ops.c (thanks to Sushilkumar Sreekumar). 1040 1041Version 1.4.52 August 7, 2014 1042 1043 - Updated documents. 1044 1045Version 1.4.51 August 6, 2014 1046 1047 - Changed names of lis_matrix_scaling*() to lis_matrix_scale*(), respectively. 1048 - Updated documents. 1049 1050Version 1.4.50 August 5, 2014 1051 1052 - Updated documents (thanks to Sushilkumar Sreekumar). 1053 1054Version 1.4.49 July 25, 2014 1055 1056 - Updated documents. 1057 1058Version 1.4.48 July 23, 2014 1059 1060 - Cleaned up programs. 1061 - Updated documents. 1062 1063Version 1.4.47 July 21, 2014 1064 1065 - Added member intvalue to struct LIS_VECTOR. 1066 - Updated lis_output_vector*() and lis_output_mm_vec(). 1067 - Updated test programs. 1068 - Updated win/Makefile.in. 1069 - Updated documents. 1070 1071Version 1.4.46 July 20, 2014 1072 1073 - Updated win/Makefile.in. 1074 1075Version 1.4.45 July 19, 2014 1076 1077 - Changed names of members *times of structs LIS_SOLVER and LIS_ESOLVER 1078 to *time, respectively. 1079 - Updated documents. 1080 1081Version 1.4.44 July 18, 2014 1082 1083 - Added support for lis_esolver_get_iters(). 1084 - Updated documents. 1085 1086Version 1.4.43 July 17, 2014 1087 1088 - Fixed bug in lis_esolver_li(). 1089 - Changed names of members iters and iters2 of structs LIS_SOLVER and 1090 LIS_ESOLVER to iter and iter2, respectively. 1091 - Changed names of functions lis_solver_get_iters(), 1092 lis_solver_get_itersex(), lis_esolver_get_iters(), and 1093 lis_esolver_get_itersex() to lis_solver_get_iter(), 1094 lis_solver_get_iterex(), lis_esolver_get_iter(), and 1095 lis_esolver_get_iterex(), respectively. 1096 - Updated lis_esolver_si(). 1097 - Updated test programs. 1098 - Updated documents. 1099 1100Version 1.4.42 July 14, 2014 1101 1102 - Changed name of members residual of structs LIS_SOLVER and LIS_ESOLVER 1103 to rhistory. 1104 1105Version 1.4.41 July 11, 2014 1106 1107 - Added support for lis_esolver_get_residualnorms(). 1108 - Updated esolve. 1109 1110Version 1.4.40 July 10, 2014 1111 1112 - Added official support for pointer operations in Fortran. 1113 - Updated documents. 1114 1115Version 1.4.39 June 25, 2014 1116 1117 - Added experimental support for pointer operations in Fortran 1118 (thanks to Svetlana Tkachenko, Alberto Gambaruto, and Takumi Yanagawa). 1119 - Added support for lis_solve_kernel() in Fortran. 1120 - Added support for lis_precon_*() in Fortran. 1121 - Added support for Microsoft MPI. 1122 1123Version 1.4.38 June 20, 2014 1124 1125 - Fixed bug in lis_vector_get_range() in Fortran (thanks to Takumi Yanagawa). 1126 1127Version 1.4.37 May 21, 2014 1128 1129 - Updated documents (thanks to Ryan Sills). 1130 1131Version 1.4.36 April 29, 2014 1132 1133 - Fixed bug in include/lis_vector.h (thanks to Per Oestlund). 1134 1135Version 1.4.35 April 15, 2014 1136 1137 - Updated documents (thanks to Patrick Alken). 1138 1139Version 1.4.34 April 4, 2014 1140 1141 - Updated documents. 1142 1143Version 1.4.33 April 2, 2014 1144 1145 - Updated documents (thanks to Ganesh Natarajan). 1146 1147Version 1.4.32 March 26, 2014 1148 1149 - Added official support for HPCG benchmark kernel. 1150 - Updated test programs. 1151 - Updated man pages. 1152 - Updated documents. 1153 1154Version 1.4.31 March 24, 2014 1155 1156 - Updated test programs. 1157 - Updated man pages. 1158 - Updated documents. 1159 1160Version 1.4.30 March 21, 2014 1161 1162 - Updated win/Makefile.in. 1163 1164Version 1.4.29 March 20, 2014 1165 1166 - Updated man pages. 1167 1168Version 1.4.28 March 17, 2014 1169 1170 - Updated print format. 1171 1172Version 1.4.27 March 16, 2014 1173 1174 - Added experimental support for HPCG benchmark kernel. 1175 1176Version 1.4.26 March 15, 2014 1177 1178 - Fixed print bug. 1179 - Updated test programs. 1180 - Updated documents. 1181 1182Version 1.4.25 February 27, 2014 1183 1184 - Added official support for option -conv_cond (thanks to Ganesh Natarajan). 1185 - Fixed print bugs. 1186 - Updated documents. 1187 1188Version 1.4.24 January 15, 2014 1189 1190 - Updated test programs. 1191 1192Version 1.4.23 January 14, 2014 1193 1194 - Fixed bugs in test programs. 1195 1196Version 1.4.22 January 11, 2014 1197 1198 - Updated test programs. 1199 - Updated documents. 1200 1201Version 1.4.21 January 9, 2014 1202 1203 - Updated test programs. 1204 1205Version 1.4.20 December 15, 2013 1206 1207 - Updated configure.in (thanks to Wei Cai). 1208 1209Version 1.4.19 November 28, 2013 1210 1211 - Updated documents. 1212 1213Version 1.4.18 November 27, 2013 1214 1215 - Fixed specification of configure, which enabled debug option 1216 of Fortran compilers by default (thanks to Svetlana Tkachenko). 1217 1218Version 1.4.17 November 13, 2013 1219 1220 - Updated documents (thanks to Svetlana Tkachenko). 1221 1222Version 1.4.16 November 12, 2013 1223 1224 - Fixed print bug in lis_vector.c (thanks to Svetlana Tkachenko). 1225 1226Version 1.4.15 September 26, 2013 1227 1228 - Updated documents (thanks to Svetlana Tkachenko). 1229 1230Version 1.4.14 September 17, 2013 1231 1232 - Fixed errors in user guide (thanks to Svetlana Tkachenko). 1233 1234Version 1.4.13 September 9, 2013 1235 1236 - Fixed bug in lis_vector_set_size(), which caused error in 1237 Fortran programs (thanks to Ralf Greve and Tatsuro Sato). 1238 1239Version 1.4.12 August 17, 2013 1240 1241 - Fixed errors in user guide (thanks to Sushilkumar Sreekumar). 1242 1243Version 1.4.11 April 19, 2013 1244 1245 - Fixed bug in Makefile.am. 1246 1247Version 1.4.10 April 3, 2013 1248 1249 - Fixed long double related bug, which caused error on Cygwin. 1250 1251Version 1.4.9 March 21, 2013 1252 1253 - Updated documents. 1254 - Changed names of SAAMG related files. 1255 1256Version 1.4.8 March 19, 2013 1257 1258 - Added sample script for gnuplot. 1259 1260Version 1.4.7 March 18, 2013 1261 1262 - Added nonsymmetric test matrix. 1263 1264Version 1.4.6 March 15, 2013 1265 1266 - Updated documents. 1267 1268Version 1.4.5 March 14, 2013 1269 1270 - Added support for gnuplot. 1271 1272Version 1.4.4 March 13, 2013 1273 1274 - Updated documents. 1275 1276Version 1.4.3 March 12, 2013 1277 1278 - Fixed print bug in test*.c. 1279 1280Version 1.4.2 March 11, 2013 1281 1282 - Updated documents. 1283 1284Version 1.4.1 March 10, 2013 1285 1286 - Updated documents. 1287 1288Version 1.4.0 March 8, 2013 1289 1290 - Added official support for long double (quadruple) precision operations. 1291 - Updated documents. 1292 - Changed version number to 1.4. 1293 1294Version 1.3.33 March 6, 2013 1295 1296 - Fixed long double related print bug. 1297 1298Version 1.3.32 March 4, 2013 1299 1300 - Fixed deallocate bug in lis_*_solver_AMGCG.F90 (thanks to Susi Lehtola). 1301 1302Version 1.3.31 March 4, 2013 1303 1304 - Fixed memory leak bug in lis_*_solver_AMGCG.F90 (thanks to Takashi Shiroto). 1305 - Updated documents. 1306 1307Version 1.3.30 March 2, 2013 1308 1309 - Updated documents. 1310 1311Version 1.3.29 March 1, 2013 1312 1313 - Updated documents. 1314 1315Version 1.3.28 February 27, 2013 1316 1317 - Updated documents. 1318 1319Version 1.3.27 February 24, 2013 1320 1321 - Updated documents. 1322 1323Version 1.3.26 February 22, 2013 1324 1325 - Updated documents. 1326 1327Version 1.3.25 February 18, 2013 1328 1329 - Updated documents. 1330 1331Version 1.3.24 February 16, 2013 1332 1333 - Updated documents. 1334 1335Version 1.3.23 February 15, 2013 1336 1337 - Updated documents. 1338 - Updated man pages. 1339 1340Version 1.3.22 February 13, 2013 1341 1342 - Fixed bug in lis_esolver_rqi.c. 1343 1344Version 1.3.21 February 12, 2013 1345 1346 - Updated documents. 1347 - Added support for zip format. 1348 1349Version 1.3.20 February 7, 2013 1350 1351 - Fixed bug in configure.in. 1352 1353Version 1.3.19 February 4, 2013 1354 1355 - Fixed bug in lis_output*.c. 1356 1357Version 1.3.18 February 2, 2013 1358 1359 - Updated documents. 1360 - Updated configure.in. 1361 - Updated win/Makefile.in. 1362 1363Version 1.3.17 February 1, 2013 1364 1365 - Added experimental support for long double (thanks to Kiyoshi Ishikawa). 1366 1367Version 1.3.16 January 24, 2013 1368 1369 - Updated documents. 1370 1371Version 1.3.15 January 21, 2013 1372 1373 - Updated documents. 1374 1375Version 1.3.14 January 18, 2013 1376 1377 - Updated documents. 1378 1379Version 1.3.13 January 17, 2013 1380 1381 - Updated documents. 1382 - Changed name of matrix storage format JDS (Jagged Diagonal Storage) 1383 to JAD (Jagged Diagonal). 1384 - Fixed bug in Makefile.am, which excluded lis_esolver_get_rhistory.3. 1385 1386Version 1.3.12 January 16, 2013 1387 1388 - Updated documents. 1389 - Updated man pages. 1390 1391Version 1.3.11 January 10, 2013 1392 1393 - Updated documents. 1394 - Updated man pages. 1395 1396Version 1.3.10 January 9, 2013 1397 1398 - Updated documents. 1399 1400Version 1.3.9 January 7, 2013 1401 1402 - Updated documents (thanks to Takashi Shiroto). 1403 1404Version 1.3.8 January 6, 2013 1405 1406 - Updated documents (thanks to Jeffrey Kollasch). 1407 1408Version 1.3.7 December 30, 2012 1409 1410 - Updated win/Makefile.in. 1411 - Updated win/configure.bat. 1412 1413Version 1.3.6 December 26, 2012 1414 1415 - Updated win/Makefile.in. 1416 - Changed file name lis_config_win32.h to lis_config_win.h. 1417 1418Version 1.3.5 December 22, 2012 1419 1420 - Updated win/Makefile.in. 1421 - Updated win/configure.bat. 1422 - Updated include/lis_config_win32.h. 1423 1424Version 1.3.4 December 20, 2012 1425 1426 - Updated win/configure.bat. 1427 1428Version 1.3.3 December 18, 2012 1429 1430 - Updated contents of directory win. 1431 - Updated documents. 1432 1433Version 1.3.2 December 17, 2012 1434 1435 - Updated documents. 1436 - Updated win/configure.bat. 1437 1438Version 1.3.1 December 15, 2012 1439 1440 - Added batch script to configure win/Makefile. 1441 - Updated contents of directory win. 1442 1443Version 1.3.0 December 11, 2012 1444 1445 - Added official support for nmake. 1446 - Fixed bug in win/Makefile. 1447 - Changed version number to 1.3. 1448 1449Version 1.2.131 December 11, 2012 1450 1451 - Updated documents. 1452 - Fixed bug in Makefile.mak. 1453 - Changed file name Makefile.mak to Makefile. 1454 1455Version 1.2.130 December 11, 2012 1456 1457 - Updated documents. 1458 - Updated Makefile.mak. 1459 - Changed directory name win32 to win. 1460 1461Version 1.2.129 December 10, 2012 1462 1463 - Updated documents. 1464 - Updated Makefile.mak. 1465 1466Version 1.2.128 December 9, 2012 1467 1468 - Updated documents. 1469 - Updated contents of directory win32. 1470 1471Version 1.2.127 December 8, 2012 1472 1473 - Updated Makefile.mak. 1474 1475Version 1.2.126 December 7, 2012 1476 1477 - Updated Makefile.mak. 1478 1479Version 1.2.125 December 6, 2012 1480 1481 - Updated documents. 1482 - Updated Makefile.mak. 1483 1484Version 1.2.124 December 5, 2012 1485 1486 - Added experimental support for nmake. 1487 1488Version 1.2.123 November 28, 2012 1489 1490 - Updated configure.in. 1491 1492Version 1.2.122 November 14, 2012 1493 1494 - Updated documents (thanks to Yoshikazu Karasawa). 1495 - Updated man pages. 1496 1497Version 1.2.121 November 12, 2012 1498 1499 - Updated test.sh. 1500 - Updated documents. 1501 1502Version 1.2.120 November 7, 2012 1503 1504 - Updated test.sh (thanks to Shoji Itoh). 1505 1506Version 1.2.119 November 5, 2012 1507 1508 - Updated configure.in and documents (thanks to Shoji Itoh). 1509 1510Version 1.2.118 October 27, 2012 1511 1512 - Fixed memory leak bug in eigensolvers (thanks to Christian Schroeppel). 1513 1514Version 1.2.117 October 13, 2012 1515 1516 - Fixed bug in lis_solver_minres.c. 1517 1518Version 1.2.116 September 19, 2012 1519 1520 - Updated man pages. 1521 - Updated Makefile.am. 1522 1523Version 1.2.115 September 12, 2012 1524 1525 - Fixed bug in lis_m_solver_AMGCG.F90. 1526 1527Version 1.2.114 September 11, 2012 1528 1529 - Updated lis_*_solver_AMGCG.F90. 1530 1531Version 1.2.113 September 10, 2012 1532 1533 - Updated documents. 1534 1535Version 1.2.112 September 9, 2012 1536 1537 - Updated man pages. 1538 - Updated documents. 1539 - Changed names of functions lis_get_solvername(), lis_get_preconname(), 1540 and lis_get_esolvername() to lis_solver_get_solvername(), 1541 lis_solver_get_preconname(), and lis_esolver_get_esolvername(), respectively. 1542 1543Version 1.2.111 September 7, 2012 1544 1545 - Updated man pages. 1546 - Updated documents. 1547 1548Version 1.2.110 September 6, 2012 1549 1550 - Updated man pages. 1551 - Updated documents. 1552 1553Version 1.2.109 September 5, 2012 1554 1555 - Fixed OpenMP related bug in lis_matrix_vbr.c. 1556 1557Version 1.2.108 September 3, 2012 1558 1559 - Changed names of matrix storage formats CRS (Compressed Row Storage) 1560 and CCS (Compressed Column Storage) to CSR (Compressed Sparse Row) 1561 and CSC (Compressed Sparse Column), respectively. 1562 1563Version 1.2.107 August 31, 2012 1564 1565 - Fixed long long int related print bug. 1566 - Fixed OpenMP related bug in test.sh. 1567 1568Version 1.2.106 August 29, 2012 1569 1570 - Updated documents. 1571 - Changed specification of Lanczos. 1572 - Restored default configuration of inverse iteration for stability. 1573 1574Version 1.2.105 August 28, 2012 1575 1576 - Updated man pages. 1577 - Updated documents. 1578 - Changed default preconditioner for inverse iteration. 1579 1580Version 1.2.104 August 27, 2012 1581 1582 - Updated documents. 1583 1584Version 1.2.103 August 26, 2012 1585 1586 - Fixed long long int related I/O bug. 1587 1588Version 1.2.102 August 25, 2012 1589 1590 - Added support for man pages. 1591 - Updated documents. 1592 1593Version 1.2.101 August 22, 2012 1594 1595 - Updated documents. 1596 1597Version 1.2.100 August 21, 2012 1598 1599 - Added support for etest6.c. 1600 - Changed specification of lis_eli(). 1601 - Fixed bug in lis_esolver_*.c. 1602 - Updated documents. 1603 1604Version 1.2.99 August 20, 2012 1605 1606 - Changed default configuration of esolve. 1607 - Changed specification of lis_esi(). 1608 - Fixed bug in lis_esolver_*.c. 1609 - Updated documents. 1610 1611Version 1.2.98 August 17, 2012 1612 1613 - Added experimental support for installing programs to ($INSTALLDIR)/bin. 1614 - Updated etest5.c. 1615 - Updated documents. 1616 1617Version 1.2.97 August 15, 2012 1618 1619 - Added support for Fujitsu Technical Computing Suite on PRIMERGY. 1620 1621Version 1.2.96 August 14, 2012 1622 1623 - Added support for lis_solver_get_precon() and lis_get_preconname(). 1624 - Updated print format of eigensolvers. 1625 - Updated documents. 1626 1627Version 1.2.95 August 13, 2012 1628 1629 - Changed specification of lis_solve_kernel() to return residual 1630 computed by lis_solve_execute() (thanks to Jonathan Schleifer). 1631 1632Version 1.2.94 August 12, 2012 1633 1634 - Fixed OpenMP flag bug. 1635 - Updated documents. 1636 1637Version 1.2.93 August 9, 2012 1638 1639 - Fixed OpenMP flag related bug in configure.in for Mac OS X (thanks to Jonathan Schleifer). 1640 - Fixed installation destination of examples. 1641 1642Version 1.2.92 August 7, 2012 1643 1644 - Updated support for out-of-tree build (thanks to Susi Lehtola). 1645 - Updated documents. 1646 1647Version 1.2.91 August 2, 2012 1648 1649 - Fixed Fortran related bug with flag --enable-shared. 1650 1651Version 1.2.90 July 31, 2012 1652 1653 - Changed names of eigensolvers. 1654 - Updated src/esolver/*.c. 1655 - Fixed bug in test2.c and etest2.c. 1656 - Fixed bug in lis_esolver_set_option_esolver(). 1657 - Updated documents. 1658 1659Version 1.2.89 July 29, 2012 1660 1661 - Updated documents. 1662 1663Version 1.2.88 July 27, 2012 1664 1665 - Updated spmvtest*.c. 1666 - Updated documents. 1667 - Fixed long long int related bug in configure.in for Fujitsu Fortran. 1668 - Updated config.guess and config.sub. 1669 1670Version 1.2.87 July 25, 2012 1671 1672 - Added experimental support for 64bit integer with flag --enable-longlong. 1673 - Fixed long long int related bug in configure.in for XL and GNU Fortran. 1674 - Fixed long long int related bug in test/*.c. 1675 - Updated documents. 1676 1677Version 1.2.86 July 24, 2012 1678 1679 - Updated documents. 1680 1681Version 1.2.85 July 21, 2012 1682 1683 - Updated documents. 1684 1685Version 1.2.84 July 20, 2012 1686 1687 - Fixed long long int related bug with flag --enable-mpi. 1688 - Fixed bug in spmvtest*.c. 1689 1690Version 1.2.83 July 19, 2012 1691 1692 - Fixed long long int related bug with flag --enable-quad. 1693 - Fixed bug in test.sh. 1694 - Updated documents. 1695 1696Version 1.2.82 July 17, 2012 1697 1698 - Added experimental support for LIS_INTEGER as integer*8 with flag -DLONGLONG in Fortran. 1699 1700Version 1.2.81 July 13, 2012 1701 1702 - Added experimental support for LIS_INT as long long int with flag -D_LONGLONG in C. 1703 - Fixed long long int related bugs. 1704 1705Version 1.2.80 July 12, 2012 1706 1707 - Replaced integer types (thanks to Alessandro Stagni): 1708 1709 type | description type | default 1710 --------------+-------------------------- ------------------+-------------- 1711 int | integer in C ==> LIS_INT | int 1712 unsigned int | unsigned integer in C LIS_UNSIGNED_INT | unsigned int 1713 integer | integer in Fortran LIS_INTEGER | integer 1714 MPI_INT | MPI integer in C LIS_MPI_INT | MPI_INT 1715 MPI_INTEGER | MPI integer in Fortran LIS_MPI_INTEGER | MPI_INTEGER 1716 MPI_Fint | MPI Fortran integer in C LIS_MPI_Fint | MPI_Fint 1717 1718Version 1.2.79 July 11, 2012 1719 1720 - Updated documents. 1721 1722Version 1.2.78 July 10, 2012 1723 1724 - Updated documents. 1725 1726Version 1.2.77 July 9, 2012 1727 1728 - Updated documents. 1729 1730Version 1.2.76 July 6, 2012 1731 1732 - Updated documents. 1733 1734Version 1.2.75 July 5, 2012 1735 1736 - Updated documents. 1737 - Updated support for Fujitsu FX10. 1738 1739Version 1.2.74 July 4, 2012 1740 1741 - Updated documents. 1742 - Updated support for Fujitsu FX10. 1743 1744Version 1.2.73 July 3, 2012 1745 1746 - Added support for Fujitsu FX10. 1747 1748Version 1.2.72 July 3, 2012 1749 1750 - Updated documents. 1751 1752Version 1.2.71 June 29, 2012 1753 1754 - Updated documents. 1755 1756Version 1.2.70 June 28, 2012 1757 1758 - Updated documents. 1759 - Updated spmvtest*.c. 1760 1761Version 1.2.69 June 28, 2012 1762 1763 - Updated documents. 1764 1765Version 1.2.68 June 26, 2012 1766 1767 - Fixed bug in spmvtest1.c. 1768 1769Version 1.2.67 June 26, 2012 1770 1771 - Updated documents. 1772 - Updated print format. 1773 1774Version 1.2.66 June 19, 2012 1775 1776 - Updated documents. 1777 1778Version 1.2.65 June 6, 2012 1779 1780 - Updated documents. 1781 1782Version 1.2.64 May 1, 2012 1783 1784 - Added support for lis_matrix_malloc_*() in Fortran 1785 (thanks to Hector Espinoza). 1786 1787Version 1.2.63 May 1, 2012 1788 1789 - Fixed bug in lisf_vector.c to support lis_vector_nrmi() in Fortran 1790 (thanks to Hector Espinoza). 1791 1792Version 1.2.62 December 24, 2011 1793 1794 - Added support for preconditioning in MINRES. 1795 - Fixed error handling bug in lis_solve(). 1796 - Updated documents. 1797 1798Version 1.2.61 November 30, 2011 1799 1800 - Added support for lis_matrix_is_assembled() in lis_matrix_setDLU_*(). 1801 1802Version 1.2.60 November 30, 2011 1803 1804 - Fixed bug in lis_matrix_set_ccs() since 1.2.56, which caused segmentation fault. 1805 1806Version 1.2.59 November 25, 2011 1807 1808 - Updated documents. 1809 1810Version 1.2.58 November 22, 2011 1811 1812 - Updated documents. 1813 1814Version 1.2.57 November 9, 2011 1815 1816 - Added support for flag --with-library-suffix to tag suffix to library names 1817 (thanks to Susi Lehtola). 1818 1819Version 1.2.56 November 8, 2011 1820 1821 - Added support for lis_matrix_is_assembled() in lis_matrix_set_*() for calling 1822 solver multiple times without allocating new memory (thanks to Malte Thoma). 1823 - Changed origin of Fortran subroutine lis_vector_set_size() to 1 (thanks to Kengo Tomida). 1824 1825Version 1.2.55 November 7, 2011 1826 1827 - Changed names of configuration header files config.h and config_win32.h 1828 to lis_config.h and lis_config_win32.h to avoid conflict in Fedora packages 1829 (thanks to Susi Lehtola). 1830 1831Version 1.2.54 November 2, 2011 1832 1833 - Fixed bug in lis_solver_set_option() (thanks to Akio Yamamoto and Kenichiro Yoshimi). 1834 1835Version 1.2.53 November 29, 2010 1836 1837 - Added support for Intel Composer XE and Microsoft Visual Studio 2010. 1838 1839Version 1.2.52 October 20, 2010 1840 1841 - Fixed bug in lis_solver_minres() with flag -print (thanks to Kazuhiro Ando). 1842 1843Version 1.2.51 September 16, 2010 1844 1845 - Updated documents. 1846 1847Version 1.2.50 September 15, 2010 1848 1849 - Updated documents. 1850 1851Version 1.2.49 September 14, 2010 1852 1853 - Updated documents. 1854 1855Version 1.2.48 September 12, 2010 1856 1857 - Fixed bug in lis_input_hb.c, which caused input error with symmetric matrices 1858 in Harwell-Boeing Format (thanks to Shoji Itoh). 1859 1860Version 1.2.47 September 10, 2010 1861 1862 - Added instruction for building shared libraries to documents. 1863 1864Version 1.2.46 September 8, 2010 1865 1866 - Fixed bug in OMPFLAG. 1867 1868Version 1.2.45 September 7, 2010 1869 1870 - Updated documents (thanks to Riccardo Rossi). 1871 1872Version 1.2.44 August 26, 2010 1873 1874 - Updated configure.in. 1875 1876Version 1.2.43 August 25, 2010 1877 1878 - Set default configuration of Mac OS X binaries to 64-bit. 1879 - Fixed bug in configure.in for PGI compilers. 1880 - Added experimental support for Pathscale compilers. 1881 1882Version 1.2.42 August 20, 2010 1883 1884 - Enabled AC_CHECK_HEADERS for Mac OS X. 1885 - Fixed parameters of configure.in for Mac OS X. 1886 (Note that default installation of Open MPI for Mac OS does not support Fortran.) 1887 1888Version 1.2.41 August 18, 2010 1889 1890 - Added support for Windows and Mac OS X. 1891 - Added instructions for Visual Studio to documents (thanks to Lizhu Tong). 1892 - Fixed default location of win32 executable files. 1893 1894Version 1.2.40 August 17, 2010 1895 1896 - Fixed bug in lis_with_fortran_mpi.icproj. 1897 - Restored flag USE_SAAMG in project files. 1898 1899Version 1.2.39 August 14, 2010 1900 1901 - Fixed default location of win32 libraries. 1902 1903Version 1.2.38 August 13, 2010 1904 1905 - Fixed character encoding of project files. 1906 - Fixed obsolete library name of MPICH2 in project files. 1907 1908Version 1.2.37 August 12, 2010 1909 1910 - Moved project files to directory win32. 1911 1912Version 1.2.36 August 11, 2010 1913 1914 - Fixed bug in configure.in, which caused configure error with XL Fortran. 1915 - Fixed bug in src/fortran/lisf_system.c, which caused compile error with 1916 Visual Studio. 1917 - Temporarily removed flag USE_SAAMG in project files for Visual Studio. 1918 1919Version 1.2.35 July 29, 2010 1920 1921 - Fixed bug in test/test.sh, which caused make check error with flag --enable-quad. 1922 - Updated documents. 1923 1924Version 1.2.34 July 28, 2010 1925 1926 - Updated documents. 1927 1928Version 1.2.33 July 23, 2010 1929 1930 - Fixed bug in spmvtests. 1931 - Updated documents. 1932 1933Version 1.2.32 July 23, 2010 1934 1935 - Added experimental support for flags -(e)storage and -(e)storage_block. 1936 - Updated spmvtests and documents. 1937 - Changed specification of precision flag: 1938 1939 flags | description flags | description 1940 -------------+-------------------- -------+----------------------------- 1941 -precision | for lis_solve() ==> -f | for lis_solve() 1942 -eprecision | for lis_esolve() -ef | for lis_esolve() 1943 1944Version 1.2.31 July 22, 2010 1945 1946 - Added support for Rayleigh Quotient Iteration. 1947 - Inner solvers for quad precision eigensolvers are switched to quad precision. 1948 - Changed order of eigensolvers: 1949 1950 flags | eigensolvers flags | eigensolvers 1951 ------------+-------------------- ------------+-------------------- 1952 -e {cg|4} | CG ==> -e {rqi|4} | Rayleigh Quotient 1953 -e {li|5} | Lanczos -e {si|5} | Subspace 1954 -e {si|6} | Subspace -e {li|6} | Lanczos 1955 -e {cr|7} | CR -e {cg|7} | CG 1956 -e {cr|8} | CR 1957 1958 - Updated documents. 1959 1960Version 1.2.30 July 21, 2010 1961 1962 - Fixed bug in lis_esolve(), which failed to destroy esolver 1963 with flag --enable-mpi. 1964 1965Version 1.2.29 July 21, 2010 1966 1967 - Fixed bug in lis_esolve(), which failed to return default eigenvector. 1968 - Fixed bug in configure for flag --enable-gprof with f90 compilers. 1969 - Added support for 64bit binary on AIX and Solaris x86 with flag --enable-64bit. 1970 - Updated documents. 1971 1972Version 1.2.28 July 16, 2010 1973 1974 - Updated documents. 1975 1976Version 1.2.27 July 15, 2010 1977 1978 - Added support for quadruple precision operations in eigensolvers. 1979 - Fixed bug in lis.vcproj. 1980 1981Version 1.2.26 July 7, 2010 1982 1983 - Added support for lis_vector_nrmi() for computing infinity norm 1984 (thanks to Osamu Nakamura). 1985 - Updated documents. 1986 1987Version 1.2.25 July 4, 2010 1988 1989 - Fixed bug in configure.in for gcc with --enable-gprof. 1990 1991Version 1.2.24 July 3, 2010 1992 1993 - Added experimental support for performance analysis using GNU profiler 1994 with flag --enable-gprof. 1995 - Added test program for lis_esolver_get_evectors(). 1996 - Updated documents. 1997 1998Version 1.2.23 July 1, 2010 1999 2000 - Added support for lis_output_matrix() (thanks to Jun Sato). 2001 - Changed specification of lis_esolver_get_evectors(). 2002 2003Version 1.2.22 June 30, 2010 2004 2005 - Fixed bug in test.sh. 2006 - Updated spmvtests. 2007 - Updated documents. 2008 2009Version 1.2.21 June 29, 2010 2010 2011 - Updated test programs. 2012 - Updated documents. 2013 2014Version 1.2.20 June 28, 2010 2015 2016 - Added support for parameter irestart for IDR(s). 2017 - Fixed default value of parameter emaxiter for subspace and 2018 Lanczos iteration. 2019 - Updated documents. 2020 2021Version 1.2.19 June 26, 2010 2022 2023 - Fixed default value of parameter etol to avoid breakdown. 2024 2025Version 1.2.18 June 24, 2010 2026 2027 - Fixed minor bug in test.sh for SAAMG with flag --enable-omp. 2028 - Fixed minor bug in lis_solver.c for SAAMG with flag --enable-mpi. 2029 - Added support for three dimensional test programs. 2030 - Updated documents. 2031 2032Version 1.2.17 June 21, 2010 2033 2034 - Fixed print bug in lis_esolver_cg.c. 2035 2036Version 1.2.16 June 18, 2010 2037 2038 - Fixed errors in documents (thanks to Osamu Nakamura). 2039 - Added support for lis_output() in Fortran (thanks to Hiroshi Imamura). 2040 - Updated spmvtests (thanks to Katsunori Matsuzaki). 2041 - Fixed errors in number of nonzero entries printed by test programs 2042 test2, etest2, spmvtest1 and spmvtest2 with flag --enable-mpi. 2043 - Fixed bug in spmvtests, which fail to measure elapsed time 2044 when matrix is very small. 2045 - Fixed bug in spmvtests for single threaded environments. 2046 - Added support for flag -m to specify default mode for Lanczos and 2047 subspace iteration. 2048 - Updated documents. 2049 2050Version 1.2.15 September 8, 2009 2051 2052 - Separated core kernel of lis_solve() to avoid overhead of handling 2053 preconditioner when called multiple times. It is now implemented 2054 as lis_solve_kernel(), to be combined with lis_precon_create() and 2055 lis_precon_destroy() (thanks to Masakazu Akiyama). 2056 - Inverse iteration and subspace iteration were rewritten using 2057 lis_solve_kernel(). Performance of Lanczos iteration, which calls 2058 inverse iteration to compute eigenvectors, is also affected. 2059 - Updated test programs to measure SpMV performance. 2060 2061Version 1.2.14 August 26, 2009 2062 2063 - Fixed bug in configure.in, which omitted OMPFLAG when target architecture 2064 is specified with flag --enable-mpi. 2065 - Updated test programs to measure SpMV performance. 2066 - Fixed description of test program etest3 in documents. 2067 2068Version 1.2.13 August 15, 2009 2069 2070 - Fixed specification of lis_matrix_split() to keep original data of 2071 splitted matrix (thanks to Tim Cramer). 2072 2073Version 1.2.12 July 21, 2009 2074 2075 - Fixed minor bugs related to SpMV test programs. 2076 - Updated documents. 2077 2078Version 1.2.11 July 7, 2009 2079 2080 - Fixed specification of lis_vector_gather() to share vector values 2081 among processes (thanks to Kouji Takatani). 2082 - Updated documents. 2083 2084Version 1.2.10 July 4, 2009 2085 2086 - Added support for lis_vector_scatter() and lis_vector_gather() to handle 2087 vector elements, mainly for MPI configuration. 2088 - Added official support for lis_solver_get_status() and 2089 lis_esolver_get_status() to get correct status of solvers 2090 (thanks to Tim Cramer). 2091 - Fixed bug in link error for Fortran 77 based MPI compilers. 2092 - Updated documents. 2093 2094Version 1.2.9 July 1, 2009 2095 2096 - Added official support for flag -saamg_theta to set drop criteria for AMG. 2097 - Fixed default parameter of theta for unsymmetric AMG, which caused error 2098 with Version 1.2.6 or higher (thanks to Kouji Takatani). 2099 - Added support for test programs to measure SpMV performance. 2100 - Fixed MPILIBS related bug in configure.in. 2101 - Fixed minor bugs. 2102 - Updated documents. 2103 2104Version 1.2.8 June 3, 2009 2105 2106 - Fixed minor bug in configure script. 2107 2108Version 1.2.7 June 1, 2009 2109 2110 - Fixed bug in MPI configuration, which caused error on Hitachi SR16000 2111 and IBM p575. 2112 - Fixed minor bugs. 2113 2114Version 1.2.6 March 23, 2009 2115 2116 - Changed location of config*.h, to be installed in $(includedir). 2117 - Fixed minor bugs. 2118 2119Version 1.2.5 December 14, 2008 2120 2121 - Added full functions for handling elements of vectors and matrices via 2122 Fortran interface. 2123 - Added support for matrix storage format conversion in eigensolvers. 2124 - Added support for PGI and Hitachi compilers. 2125 - Fixed bug in configure, which omitted flag -DUSE_MPI of Fortran based 2126 MPI compilers. 2127 - Fixed bug in test/test3f.F, which caused make check error with -DUSE_MPI. 2128 2129Version 1.2.4 December 5, 2008 2130 2131 - Added experimental support for conjugate residual for nonsymmetric eigenproblems. 2132 - Fixed minor bugs. 2133 2134Version 1.2.3 December 4, 2008 2135 2136 - Fixed origin of lis_vector_get_value() in Fortran (thanks to Kengo Tomida). 2137 - Fixed minor bugs. 2138 2139Version 1.2.2 November 25, 2008 2140 2141 - Fixed bug in test/test2.c, which caused make check error 2142 (thanks to Takahito Miki). 2143 - Fixed bug in configure for combination of SAAMG and gfortran based 2144 MPI compiler. SAAMG can be built with gfortran 4.3 or higher. 2145 - Fixed minor bugs. 2146 2147Version 1.2.1 November 18, 2008 2148 2149 - Fixed origin of lis_vector_set_value() in Fortran (thanks to Yutaka Masuda). 2150 2151Version 1.2.0 November 12, 2008 2152 2153 - Added support for eigensolvers (power iteration, inverse iteration, 2154 approximate inverse iteration, conjugate gradient, Lanczos iteration, 2155 and subspace iteration) for real sparse matrices. 2156 - Added support for MINRES (thanks to Hirofumi Notsu). 2157 - Changed names of lis_output_residual_history() and 2158 lis_get_residual_history() to lis_solver_output_rhistory() and 2159 lis_solver_get_rhistory(), respectively. 2160 - Fixed convergence criteria of TFQMR and BiCGSTAB(l) (thanks to Shoji Itoh). 2161 - Added support for NEC SX-9. 2162 - Updated config.guess and config.sub. 2163 - Added optimization options for Fujitsu Compiler (thanks to Fujitsu 2164 Limited). 2165 - Fixed location of include statement in lisf_init.F to avoid compiling 2166 error with Fujitsu Compiler. 2167 - Fixed configure script to choose mpiexec if mpirun is not available. 2168 - Fixed configure script to choose Fortran 9x based MPI compiler if SAAMG is 2169 enabled. 2170 - Fixed minor bugs. 2171 - Updated documents. 2172 2173Version 1.1.2 December 24, 2007 2174 2175 - Added support for IDR(s). 2176 - Fixed minor bugs. 2177 2178Version 1.1.1 December 6, 2007 2179 2180 - Changed specification of lis_initialize() to avoid MPICH1 related error 2181 (thanks to Takashi Yoshida). 2182 - Fixed minor bugs. 2183 2184Version 1.1.0 October 31, 2007 2185 2186 - Added support for CR, BiCR, CRS, BiCRSTAB, GPBiCR, BiCRSafe, and FGMRES. 2187 - Added support for ILUT preconditioner. 2188 - Added support for binary package for Windows. 2189 - Added support for graph plotting tool for Windows. 2190 - Fixed minor bugs. 2191 2192Version 1.1.0 beta2 June 18, 2007 2193 2194 - Added support for BiCGSafe. 2195 - Fixed bug caused by memory leaks. 2196 2197Version 1.1.0 beta1 November 7, 2006 2198 2199 - Added support for quadruple precision operations for linear solvers. 2200 - Added support for Fortran user interface. 2201 - Added support for unsymmetric SAAMG preconditioner (symmetric structure 2202 only). 2203 - Added support for Crout ILU preconditioner. 2204 - Added support for additive Schwarz preconditioner. 2205 - Added support for user-defined preconditioners. 2206 - Added support for Harwell-Boeing format. 2207 - Integrated Lis-AMG into Lis. 2208 - Introduced Autotools. 2209 - Changed specifications: 2210 - Introduced struct LIS_SOLVER (integrated type parameters). 2211 - Changed declaration of matrices (thanks to David Moloney) and vectors. 2212 - Changed notation of command line options. 2213 2214Version 1.0.2 April 17, 2006 2215 2216 - Fixed bug caused by not correctly setting value of beginning and end 2217 of line when lis_vector_create() is called. 2218 - Fixed bug caused by wrong selection of linear solvers specified 2219 with lis_set_options(): 2220 2221 flags | solvers flags | solvers 2222 ----------------+--------------- ----------------+--------------- 2223 -i {tfqmr|7} | Orthomin ==> -i {tfqmr|7} | TFQMR 2224 -i {orthomin|8}| GMRES -i {orthomin|8}| Orthomin 2225 -i {gmres|9} | TFQMR -i {gmres|9} | GMRES 2226 2227 - Fixed bug in allocation of temporary array of Orthomin. 2228 - Fixed bug in output of residual norm in TFQMR. 2229 - Fixed bug in array assignment of BiCG with ILU preconditioner. 2230 - Fixed minor bugs. 2231 2232Version 1.0.1 November 6, 2005 2233 2234 - Added support for Lis-AMG. 2235 - Added support for English manual. 2236 - Added support for test1b. 2237 - Fixed minor bugs. 2238 2239Version 1.0.0 September 20, 2005 2240 2241 - Released first version. 2242