Version 2.1.3 June 14, 2023 - Fixed some bugs in source code (thanks to Thedore Chang). Version 2.1.2 May 21, 2023 - Updated configurations for Intel LLVM C Compiler. - Fixed errors in win/Makefile.in (thanks to K. Yamada). - Fixed errors in user guides (thanks to K. Yamada). Version 2.1.1 April 12, 2023 - Corrected errors on lis_array_ge() in documents (thanks to Ning Li). - Corrected error on lis_array_ge_f() in include/lis_fortran.h (thanks to Ning Li). Version 2.1.0 March 5, 2023 - Updated documents. Version 2.0.35 February 21, 2023 - Fixed win/Makefile.in to include lisf_init.obj in lis.lib without link error for Intel Fortran Compiler (thanks to ChatGPT). Version 2.0.34 April 26, 2022 - Fixed win/Makefile.in to avoid link error with Intel Fortran Compiler (thanks to Yigang Gong). Version 2.0.33 November 14, 2021 - Fixed include/lisf.h.in to avoid Fortran error. - Updated win/Makefile.in to include lisf_init.obj in lis.lib. Version 2.0.32 October 1, 2021 - Fixed type of lis_wtime() in user guides (thanks to Ning Li). Version 2.0.31 September 24, 2021 - Updated include/lisf.h.in experimentally to remove dependency on lisf_init.F (thanks to Ning Li). Version 2.0.30 January 23, 2021 - Added support for lis_esolver_get_specific_*() (thanks to Peter Vermeulen). Version 2.0.29 December 28, 2020 - Fixed MPI related bug in configure.ac for macOS (thanks to Carlo de Falco). Version 2.0.28 December 24, 2020 - Updated note on win32.mak in win/README. - Added experimental support for Fujitsu A64FX and NEC SX-Aurora TSUBASA. Version 2.0.27 December 3, 2020 - Added note on win32.mak to win/README (thanks to Zhengang Liu). Version 2.0.26 November 25, 2020 - Added option --disable-test to configure.ac (thanks to Fabian Wein). Version 2.0.25 November 18, 2020 - Updated configure.ac for Apple Clang (thanks to Fabian Wein). Version 2.0.24 November 3, 2020 - Fixed bugs in lis_error.c and lis_esolver_si.c. Version 2.0.23 August 26, 2020 - Fixed bug in lis_input_vector_mm(). Version 2.0.22 May 22, 2020 - Updated documents. Version 2.0.21 March 18, 2020 - Fixed MPI related bug in test4f.F. Version 2.0.20 September 4, 2019 - Disabled unimplemented conv_cond options for some linear solvers (thanks to Kazuhiro Tada). Version 2.0.19 May 22, 2019 - Applied Lis patch for K computer (thanks to Shun Ito). Version 2.0.18 January 30, 2019 - Updated libtool to 2.4.6 (thanks to Martin Sjoelund). - Updated documents (thanks to Danyang Su). Version 2.0.17 October 26, 2018 - Updated documents (thanks to Yasuyo Hatano). Version 2.0.16 September 26, 2018 - Disabled option '-f quad' when quad precision is not enabled (thanks to Konstantin). Version 2.0.15 September 20, 2018 - Updated copyright notice (thanks to Konstantin). Version 2.0.14 July 11, 2018 - Updated .gitignore (thanks to Oliver Evans). Version 2.0.13 June 28, 2018 - Added OMPFLAGs to configurations for Intel MIC (thanks to Javier Bahamonde Amor). Version 2.0.12 May 24, 2018 - Added test matrix testmat4.mtx in Matrix Market array format. Version 2.0.11 April 26, 2018 - Fixed lis.h for compatibility with C++ (thanks to Michael Asam). Version 2.0.10 April 12, 2018 - Updated graphics for complex eigenvalues. Version 2.0.9 March 21, 2018 - Fixed configuration of win/Makefile.in for SAAMG (thanks to Jian Li). Version 2.0.8 March 9, 2018 - Updated descriptions for test matrices. Version 2.0.7 February 8, 2018 - Fixed matrix block size related bugs in spmvtest4 and spmvtest5 (thanks to Javier Bahamonde Amor). Version 2.0.6 December 10, 2017 - Added complex test matrix testmat3.mtx. - Updated print messages for Subspace, Lanczos, and Arnoldi iterations. - Fixed bug in lisf.h.in. Version 2.0.5 December 7, 2017 - Updated documents. Version 2.0.4 November 18, 2017 - Fixed memory leak bugs related to option -rval. Version 2.0.3 November 11, 2017 - Updated lis_matrix_destroy() to avoid conflict with deallocate() in Fortran 90 (thanks to Ste'phane Glockner). Version 2.0.2 November 6, 2017 - Changed name of function lis_matrix_shift_general() to lis_matrix_shift_matrix(). - Updated test8f.F90. Version 2.0.1 November 5, 2017 - Added support for option -rval. - Added support for graphics/gnuplot3b*. - Fixed shift related bugs. Version 2.0.0 November 1, 2017 - Added official support for linear solvers generalized for complex matrices. - Added official support for nonsymmetric eigenproblems. Version 1.8.15 November 1, 2017 - Fixed complex related bugs in BiCR. Version 1.8.14 October 31, 2017 - Fixed bugs on complex arithmetic. Version 1.8.13 October 30, 2017 - Fixed bugs in lis_input_mm.c. Version 1.8.12 October 25, 2017 - Generalized linear solvers and preconditioners for complex matrices. Version 1.8.11 October 18, 2017 - Fixed bug in lis_esolver_ai.c. - Fixed bug in lis_array.c. - Fixed description in etest7.c. - Added support for Matrix Market array format. Version 1.8.10 October 17, 2017 - Updated configurations in win/Makefile.in. Version 1.8.9 October 11, 2017 - Updated configurations in win/Makefile.in for Microsoft MPI. - Fixed SAAMG related configurations in win/Makefile.in. Version 1.8.8 October 9, 2017 - Updated configurations in win/Makefile.in for Intel Compilers. Version 1.8.7 October 4, 2017 - Fixed configure.ac to exclude complex.h by default (thanks to Fuyuki Saito). - Fixed configurations in win/Makefile.in. Version 1.8.6 September 24, 2017 - Fixed bug in win/Makefile.in (thanks to Jian Li). - Added support for inputting imaginary shift for eigensolvers. Version 1.8.5 September 14, 2017 - Updated documents. Version 1.8.4 August 29, 2017 - Fixed documents. Version 1.8.3 August 10, 2017 - Updated documents. - Updated eigensolvers. Version 1.8.2 August 6, 2017 - Added support for inputting complex vectors in Matrix Market format. Version 1.8.1 August 4, 2017 - Fixed bug in lis_solver_sor.c. - Changed names of flags -ssor_w and -hybrid_w to -ssor_omega and -hybrid_omega, respectively. Version 1.8.0 August 1, 2017 - Added official support for test/test3c.c for solving multiple-step linear equations. Version 1.7.37 July 31, 2017 - Fixed MPI bug in test/test3c.c. - Fixed bug in lis_solver_sor.c. Version 1.7.36 July 27, 2017 - Fixed MPI memory leak bug in lis_matrix_assemble() (thanks to Danyang Su). - Fixed lis_input_vector() related bug in test1.c (thanks to Danyang Su). Version 1.7.35 July 26, 2017 - Fixed MPI memory leak bug in lis_matrix_assemble() (thanks to Danyang Su). Version 1.7.34 July 17, 2017 - Fixed MPI bugs in lis_vector_gather() and lis_vector_scatter() (thanks to Danyang Su). Version 1.7.33 June 11, 2017 - Added exception handling for inner linear solvers. Version 1.7.32 June 10, 2017 - Fixed exception handling bug. Version 1.7.31 June 7, 2017 - Fixed exception handling bug (thanks to Ralf Greve). Version 1.7.30 May 31, 2017 - Added exception handlings for wrong parameter settings (thanks to Ralf Greve). - Updated documents. Version 1.7.29 March 22, 2017 - Added experimental support for test/test3c.c, which is an update of test/test4b.c for solving multiple-step linear equations (thanks to Fabian Wein). - Updated documents. Version 1.7.28 March 12, 2017 - Added support for Intel MIC (thanks to Javier Bahamonde Amor). - Updated documents. Version 1.7.27 February 17, 2017 - Fixed matrix file format for lis_esolver_get_evectors() (thanks to Takahiro Okada). - Updated documents. Version 1.7.26 February 10, 2017 - Added support for lis_*solver_get_rhistory() in Fortran (thanks to Thomas Goelles). - Updated documents (thanks to Thomas Goelles). Version 1.7.25 February 3, 2017 - Updated documents (thanks to Hiroshi Kanayama). Version 1.7.24 January 25, 2017 - Updated test/test4b.c. Version 1.7.23 January 22, 2017 - Added support for test/test4b.c (thanks to Ganesh Natarajan). - Updated test/spmvtest1.c. Version 1.7.22 January 13, 2017 - Updated documents (thanks to Ganesh Natarajan). Version 1.7.21 January 2, 2017 - Fixed descriptions. - Updated documents. Version 1.7.20 December 19, 2016 - Fixed type of argc in lis_initialize() to avoid conflict with MPI_Init(). Version 1.7.19 December 19, 2016 - Fixed critical bug in lis_init.c. - Fixed long long int related bugs. - Fixed long double related bugs. Version 1.7.18 December 15, 2016 - Fixed types for print statements. Version 1.7.17 December 14, 2016 - Fixed type related MPI bugs in Fortran programs. Version 1.7.16 December 11, 2016 - Added support for GCC libquadmath. - Fixed quad precision bugs in Fortran programs. Version 1.7.15 December 6, 2016 - Fixed MPI bug in lis_error(). - Fixed print bugs. - Changed specification of lis_print_rhistory(). Version 1.7.14 December 3, 2016 - Fixed type related MPI bugs. - Fixed bug in src/system/lis_error.c. - Updated src/system/lis_input.c. - Updated documents. Version 1.7.13 November 22, 2016 - Updated documents (thanks to Wei Liu). Version 1.7.12 November 11, 2016 - Fixed type related print bugs in lis_error.c. - Fixed bug in spmvtest3b.c. Version 1.7.11 November 10, 2016 - Fixed type related print bugs in test/spmvtest*.c. Version 1.7.10 November 8, 2016 - Fixed type related print bug in lis_esolver.c. Version 1.7.9 November 7, 2016 - Cleaned up print statements using lis_printf(). - Fixed type related print bugs in error messages. Version 1.7.8 November 4, 2016 - Fixed installation bugs in win/Makefile.in. - Updated documents. Version 1.7.7 November 3, 2016 - Fixed bug in generalized Lanczos. Version 1.7.6 November 1, 2016 - Fixed bug in generalized CG. Version 1.7.5 October 29, 2016 - Added support for generalized Rayleigh quotient iteration, generalized CG, and generalized Arnoldi iteration. Version 1.7.4 October 26, 2016 - Fixed index bug in lis_esolver_ai.c. Version 1.7.3 October 25, 2016 - Fixed shift related bugs. - Added support for option -ige for generalized eigenproblems. - Disabled complex operations if complex.h is not available. - Updated documents. Version 1.7.2 October 24, 2016 - Added support for generalized Lanczos iteration. Version 1.7.1 October 23, 2016 - Inverted sign of shift for eigensolvers in accordance with custom. - Inverted sign of shift in lis_matrix_shift_diagonal(), lis_vector_shift(), and lis_array_shift(), respectively. - Added support for shift for generalized eigenproblems. - Updated man pages. Version 1.7.0 October 22, 2016 - Added official support for generalized eigenvalue problems. - Added support for generalized subspace iteration. Version 1.6.39 October 21, 2016 - Added experimental support for generalized power iteration. Version 1.6.38 October 20, 2016 - Fixed print statements in eigensolvers for option -eprint none. - Added experimental support for generalized inverse iteration. Version 1.6.37 October 19, 2016 - Updated test6f.F90. - Added experimental support for generalized CR. Version 1.6.36 October 17, 2016 - Updated test6.c (thanks to Hiroshi Kanayama). - Updated lisf_*.c (thanks to Florian Lehner). - Updated lis_esolver.c. Version 1.6.35 October 16, 2016 - Updated notations of eigensolvers. - Fixed bug of option -shift. - Deleted approximate inverse iteration from eigensolvers. Version 1.6.34 October 14, 2016 - Updated documents (thanks to Hiroshi Kanayama). Version 1.6.33 October 13, 2016 - Added support for lis_gesolve(). Version 1.6.32 October 12, 2016 - Updated lis_array_*(). - Updated man pages. Version 1.6.31 October 11, 2016 - Updated documents (thanks to Hiroshi Kanayama). - Fixed bug in Makefile.am. Version 1.6.30 October 10, 2016 - Updated documents (thanks to Hiroshi Kanayama). - Updated man pages. - Updated include/lis.h. Version 1.6.29 October 7, 2016 - Fixed quad precision bug in lis_solver.c (thanks to Yonggwi Cho). Version 1.6.28 October 6, 2016 - Updated eigensolvers. Version 1.6.27 October 5, 2016 - Updated lis_esolver_ai.c. Version 1.6.26 October 5, 2016 - Fixed bugs in test/*.F. Version 1.6.25 October 2, 2016 - Fixed bugs in test/test*.F90. Version 1.6.24 September 30, 2016 - Fixed bugs in test/test1f.F. Version 1.6.23 September 29, 2016 - Added support for COCG and COCR. Version 1.6.22 September 28, 2016 - Added support for lis_vector_nhdot(). Version 1.6.21 September 27, 2016 - Fixed bug in lis_esolver.c. Version 1.6.20 September 25, 2016 - Updated documents (thanks to Hiroaki Morimoto). Version 1.6.19 September 24, 2016 - Fixed complex related bugs. Version 1.6.18 September 23, 2016 - Fixed complex related bugs. Version 1.6.17 September 23, 2016 - Fixed complex related bugs. Version 1.6.16 September 21, 2016 - Fixed complex related bugs. Version 1.6.15 September 20, 2016 - Fixed complex related bugs. Version 1.6.14 September 19, 2016 - Fixed complex related bugs. Version 1.6.13 September 19, 2016 - Fixed complex related configuration in lis_solver_cg.c. Version 1.6.12 September 16, 2016 - Updated win/*. - Fixed memory leak bug in lis_esolver_cg.c. Version 1.6.11 September 15, 2016 - Updated win/* (thanks to Hiroaki Morimoto). - Updated lis_input_hb.c. Version 1.6.10 September 14, 2016 - Added support for outputting complex matrices. - Added support for reading Hermitian matrices. Version 1.6.9 September 13, 2016 - Fixed bugs in lis_input_hb.c. - Updated documents. Version 1.6.8 September 12, 2016 - Fixed critical bug in lis_input_mm.c for reading symmetric matrices. Version 1.6.7 September 11, 2016 - Added support for lis_vector_conjugate(). Version 1.6.6 September 10, 2016 - Fixed bug in win/Makefile.in. - Updated test7.c and test7f.F. Version 1.6.5 September 9, 2016 - Fixed complex related bug on fabs(). - Updated test7.c and test7f.F. Version 1.6.4 September 8, 2016 - Fixed bug in lisf.h.in. - Updated test7.c and test7f.F. Version 1.6.3 September 7, 2016 - Added support for lis_solver_get_status() in Fortran (thanks to Alessandro Brovelli). Version 1.6.2 September 6, 2016 - Fixed complex related bug in test4f.F. Version 1.6.1 September 5, 2016 - Fixed quad precision bug (thanks to Florian Lehner). Version 1.6.0 September 4, 2016 - Added official support for flag --enable-complex. - Fixed complex related bugs in parallel processing. - Added support for inputting complex matrices in Harwell-Boeing format. Version 1.5.79 September 3, 2016 - Fixed complex related bugs. Version 1.5.78 September 2, 2016 - Added support for outputting complex vectors. - Updated documents. Version 1.5.77 September 1, 2016 - Added support for inputting complex matrices in Matrix Market format. Version 1.5.76 August 31, 2016 - Fixed complex related bugs. Version 1.5.75 August 31, 2016 - Added experimental support for flag --enable-complex for complex scalar. - Updated documents. Version 1.5.74 August 26, 2016 - Fixed configurations for complex.h. Version 1.5.73 August 25, 2016 - Fixed configurations for complex numbers. Version 1.5.72 August 25, 2016 - Fixed bug in test7f.F. Version 1.5.71 August 25, 2016 - Fixed bug in Fortran configuration for complex*32. Version 1.5.70 August 25, 2016 - Added official support for complex numbers. - Renamed test7f.F90 test8f.F90. - Added test7.c and test7f.F for complex numbers. - Added decralations for PSD functions. Version 1.5.69 August 21, 2016 - Fixed bug in test/Makefile.am. Version 1.5.68 August 21, 2016 - Added test7.F90 to test_PROGRAMS in test/Makefile.am. Version 1.5.67 August 20, 2016 - Added test7.F90 to test/Makefile.am. Version 1.5.66 June 17, 2016 - Added note on file format for symmetric matrices to Appendix (thanks to Kodanda Ram Mangipudi). Version 1.5.65 March 19, 2016 - Fixed bug in lis_matrix_csr.c. Version 1.5.64 March 19, 2016 - Added support for alternative workflow PSD, or Preconditioner and Solver Decoupled (thanks to Neil Hodge). - Updated documents. Version 1.5.63 January 31, 2016 - Fixed autoreconf related bugs (thanks to Florian Lehner). Version 1.5.62 January 7, 2016 - Changed names of preprocessor macros _LONGLONG and LONGLONG to _LONG__LONG and LONG__LONG, respectively, to avoid conflict with __int64. Version 1.5.61 January 7, 2016 - Fixed __int64 related bug in lis.h for Visual C++ (thanks to Takeshi Takeda). Version 1.5.60 January 2, 2016 - Deleted "-O3" options associated with Intel compilers so that developers can define optimization themselves (thanks to Neil Hodge). - Moved inclusion of math.h from lis.h to lis_solver.c (thanks to Neil Hodge). - Deleted duplicate defines in lis.h (thanks to Neil Hodge). - Fixed bug in lis_init.c associated with extra whitespace between solver options (thanks to Neil Hodge). - Modified code in lis_vector.c and lis_matrix.c to allow for the case of zero DOFs on a process (thanks to Neil Hodge). - Updated documents. Version 1.5.59 December 15, 2015 - Added include statement for math.h to lis.h (thanks to Neil Hodge). Version 1.5.58 November 31, 2015 - Added include statements for math.h to lis_solver.c and lis_esolver.c (thanks to James Liu). Version 1.5.57 July 7, 2015 - Added support for lis_matrix_unset() in C (thanks to Norihiro Watanabe). Version 1.5.56 June 25, 2015 - Updated documents. - Added experimental support for complex numbers. Version 1.5.55 May 14, 2015 - Added support for lis_solver_get_precon() and lis_solver_get_preconname() in Fortran (thanks to Domenic D'Ambrosio). - Updated documents (thanks to Domenic D'Ambrosio). Version 1.5.54 April 30, 2015 - Fixed configurations for Windows (thanks to Domenic D'Ambrosio). - Added support for combining Intel Fortran Compiler with Visual C (thanks to Domenic D'Ambrosio). Version 1.5.53 April 5, 2015 - Updated documents (thanks to Sushilkumar Sreekumar). Version 1.5.52 March 28, 2015 - Updated documents (thanks to Sushilkumar Sreekumar). Version 1.5.51 February 10, 2015 - Updated documents. Version 1.5.50 February 9, 2015 - Updated documents (thanks to Sushilkumar Sreekumar). Version 1.5.49 February 6, 2015 - Updated documents (thanks to Carlo de Falco). Version 1.5.48 February 6, 2015 - Fixed exception handling bug in lis_matrix_set_size() (thanks to Carlo de Falco). Version 1.5.47 January 10, 2015 - Removed unused variables in lis_precision_vec.c. Version 1.5.46 January 8, 2015 - Fixed quad precision bug in lis_precision_vec.c. - Updated config/config.*. - Cleaned up programs. Version 1.5.45 December 30, 2014 - Added documents for matrix addition. - Updated documents. Version 1.5.44 December 29, 2014 - Fixed exception handling in matrix addition. - Updated documents. Version 1.5.43 December 28, 2014 - Fixed Fortran functions for matrix operations. Version 1.5.42 December 27, 2014 - Added support for matrix addition. Version 1.5.41 December 23, 2014 - Fixed bug in win/Makefile.in. Version 1.5.40 December 20, 2014 - Fixed variable declaration bug in lis_precision_vec.c. Version 1.5.39 December 16, 2014 - Updated documents. Version 1.5.38 December 15, 2014 - Updated documents. Version 1.5.37 December 14, 2014 - Cleaned up programs (thanks to Florian Lehner). Version 1.5.36 December 11, 2014 - Updated lis_esolver_rqi.c. Version 1.5.35 December 10, 2014 - Updated documents (thanks to Sushilkumar Sreekumar). Version 1.5.34 December 8, 2014 - Updated documents. - Changed default preconditioners for eigensolvers. Version 1.5.33 December 7, 2014 - Updated documents. - Added experimental support for Jacobi-Davidson. - Fixed exception handling bug in lis_matrix_set_size() (thanks to Yi Yao). Version 1.5.32 December 5, 2014 - Updated documents. Version 1.5.31 December 4, 2014 - Updated etest7.c. - Updated lis_esolver_ai.c. Version 1.5.30 December 3, 2014 - Updated documents. - Fixed print bug on iteration counts of inner eigensolvers. Version 1.5.29 December 2, 2014 - Added experimental support for Arnoldi iteration. - Disabled obsolete option -m for eigensolvers. - Updated documents. Version 1.5.28 December 1, 2014 - Changed default linear solver for eigensolvers. - Changed order of eigensolvers: flags | eigensolvers flags | eigensolvers ------------+-------------------- ------------+-------------------- -e {si|5} | Subspace ==> -e {cg|5} | CG -e {li|6} | Lanczos -e {cr|6} | CR -e {cg|7} | CG -e {si|7} | Subspace -e {cr|8} | CR -e {li|8} | Lanczos - Updated documents. Version 1.5.27 November 30, 2014 - Updated documents. Version 1.5.26 November 29, 2014 - Changed names of functions lis_array_matvec2() and lis_array_matmat2() to lis_array_matvec_ns() and lis_array_matmat_ns(), respectively. Version 1.5.25 November 28, 2014 - Updated lis_array.c. Version 1.5.24 November 27, 2014 - Updated lis_array.c (thanks to Florian Lehner). Version 1.5.23 November 26, 2014 - Cleaned up lis_esolver_li.c. - Cleaned up lis_array.c. - Fixed convergence condition of lis_array_qr(). Version 1.5.22 November 25, 2014 - Updated documents. Version 1.5.21 November 24, 2014 - Cleaned up lis_solver_*.c. - Added support for test6f.F90 for array operations. - Updated documents. Version 1.5.20 November 23, 2014 - Updated documents. Version 1.5.19 November 22, 2014 - Fixed bug in lisf_array.c (thanks to Florian Lehner). Version 1.5.18 November 21, 2014 - Updated documents. Version 1.5.17 November 20, 2014 - Updated lis_fortran.h for Win64 environment. - Fixed initial value related bug in GMRES to enable option -initx_zeros. Version 1.5.16 November 19, 2014 - Updated documents (thanks to Sushilkumar Sreekumar). Version 1.5.15 November 18, 2014 - Fixed order of array elements in lis_array_cgs(), lis_array_mgs(), and lis_array_qr(). - Fixed memory initialization bug in lis_array_mgs(). - Updated documents. Version 1.5.14 November 16, 2014 - Cleaned up programs. - Updated eigensolvers. - Added sample scripts for graphics. Version 1.5.13 November 14, 2014 - Updated documents. Version 1.5.12 November 13, 2014 - Updated documents. Version 1.5.11 November 12, 2014 - Updated test3b.c. Version 1.5.10 November 11, 2014 - Added sample scripts for graphics. Version 1.5.9 November 9, 2014 - Cleaned up programs. Version 1.5.8 November 8, 2014 - Updated documents. Version 1.5.7 November 7, 2014 - Fixed print format. - Fixed elapsed time related bug in preconditioned eigensolvers. - Fixed CR related bug in lis_esolver_cg.c. Version 1.5.6 November 6, 2014 - Updated documents. Version 1.5.5 November 5, 2014 - Cleaned up programs. - Updated documents. Version 1.5.4 November 4, 2014 - Fixed quad precision related bug in lis_esolver_si.c. - Fixed bug in test/Makefile.am. Version 1.5.3 November 3, 2014 - Updated eigensolvers. - Fixed array related bug in lis_esolver_cg.c. Version 1.5.2 November 1, 2014 - Added support for etest7.c for array operations. - Changed specification of lis_array_qr() to return number of iterations and error of QR algorithm. - Updated documents. Version 1.5.1 October 31, 2014 - Added support for test6.c for array operations. - Updated print formats. - Updated documents. Version 1.5.0 October 30, 2014 - Fixed print bug in option -conv_cond (thanks to Suguru Miyauchi). - Added official support for array operations. - Updated documents. Version 1.4.67 October 28, 2014 - Fixed bug in lis_array_nrmi_f() (thanks to Florian Lehner). - Updated lis_array.c. Version 1.4.66 October 27, 2014 - Updated lis_array.c. - Changed names of functions lis_array_dot2() and lis_array_invGauss() to lis_array_dot() and lis_array_ge(), respectively. Version 1.4.65 October 23, 2014 - Updated documents (thanks to Suguru Miyauchi). Version 1.4.64 October 21, 2014 - Updated lis_m_data_creation_AMGCG.F90. - Updated sample scripts for graphics. Version 1.4.63 October 19, 2014 - Added sample scripts for graphics (thanks to Takashi Shiroto and Sushilkumar Sreekumar). Version 1.4.62 October 17, 2014 - Fixed exception handling bug in lis_m_data_creation_AMGCG.F90 (thanks to Takashi Shiroto). Version 1.4.61 September 19, 2014 - Fixed OpenMP related bug in lis_matrix_ell.c (thanks to Gerald Wedel). Version 1.4.60 September 13, 2014 - Fixed long double related print bugs. - Fixed long long int related print bugs. Version 1.4.59 September 9, 2014 - Updated configure.in. - Updated test programs in Fortran. - Updated documents. Version 1.4.58 August 27, 2014 - Added support for lis_matrix_unset() in Fortran (thanks to Takumi Yanagawa and Takashi Shiroto). - Updated test/test2f.F90. Version 1.4.57 August 15, 2014 - Updated lis_array.c. Version 1.4.56 August 12, 2014 - Updated documents. - Added experimental support for array operations. Version 1.4.55 August 11, 2014 - Updated lis_matrix_ops.c. - Updated lisf_matrix.c. - Updated documents. Version 1.4.54 August 10, 2014 - Updated lisf_matrix.c. Version 1.4.53 August 9, 2014 - Fixed print bug. - Updated lis_matrix_ops.c (thanks to Sushilkumar Sreekumar). Version 1.4.52 August 7, 2014 - Updated documents. Version 1.4.51 August 6, 2014 - Changed names of lis_matrix_scaling*() to lis_matrix_scale*(), respectively. - Updated documents. Version 1.4.50 August 5, 2014 - Updated documents (thanks to Sushilkumar Sreekumar). Version 1.4.49 July 25, 2014 - Updated documents. Version 1.4.48 July 23, 2014 - Cleaned up programs. - Updated documents. Version 1.4.47 July 21, 2014 - Added member intvalue to struct LIS_VECTOR. - Updated lis_output_vector*() and lis_output_mm_vec(). - Updated test programs. - Updated win/Makefile.in. - Updated documents. Version 1.4.46 July 20, 2014 - Updated win/Makefile.in. Version 1.4.45 July 19, 2014 - Changed names of members *times of structs LIS_SOLVER and LIS_ESOLVER to *time, respectively. - Updated documents. Version 1.4.44 July 18, 2014 - Added support for lis_esolver_get_iters(). - Updated documents. Version 1.4.43 July 17, 2014 - Fixed bug in lis_esolver_li(). - Changed names of members iters and iters2 of structs LIS_SOLVER and LIS_ESOLVER to iter and iter2, respectively. - Changed names of functions lis_solver_get_iters(), lis_solver_get_itersex(), lis_esolver_get_iters(), and lis_esolver_get_itersex() to lis_solver_get_iter(), lis_solver_get_iterex(), lis_esolver_get_iter(), and lis_esolver_get_iterex(), respectively. - Updated lis_esolver_si(). - Updated test programs. - Updated documents. Version 1.4.42 July 14, 2014 - Changed name of members residual of structs LIS_SOLVER and LIS_ESOLVER to rhistory. Version 1.4.41 July 11, 2014 - Added support for lis_esolver_get_residualnorms(). - Updated esolve. Version 1.4.40 July 10, 2014 - Added official support for pointer operations in Fortran. - Updated documents. Version 1.4.39 June 25, 2014 - Added experimental support for pointer operations in Fortran (thanks to Svetlana Tkachenko, Alberto Gambaruto, and Takumi Yanagawa). - Added support for lis_solve_kernel() in Fortran. - Added support for lis_precon_*() in Fortran. - Added support for Microsoft MPI. Version 1.4.38 June 20, 2014 - Fixed bug in lis_vector_get_range() in Fortran (thanks to Takumi Yanagawa). Version 1.4.37 May 21, 2014 - Updated documents (thanks to Ryan Sills). Version 1.4.36 April 29, 2014 - Fixed bug in include/lis_vector.h (thanks to Per Oestlund). Version 1.4.35 April 15, 2014 - Updated documents (thanks to Patrick Alken). Version 1.4.34 April 4, 2014 - Updated documents. Version 1.4.33 April 2, 2014 - Updated documents (thanks to Ganesh Natarajan). Version 1.4.32 March 26, 2014 - Added official support for HPCG benchmark kernel. - Updated test programs. - Updated man pages. - Updated documents. Version 1.4.31 March 24, 2014 - Updated test programs. - Updated man pages. - Updated documents. Version 1.4.30 March 21, 2014 - Updated win/Makefile.in. Version 1.4.29 March 20, 2014 - Updated man pages. Version 1.4.28 March 17, 2014 - Updated print format. Version 1.4.27 March 16, 2014 - Added experimental support for HPCG benchmark kernel. Version 1.4.26 March 15, 2014 - Fixed print bug. - Updated test programs. - Updated documents. Version 1.4.25 February 27, 2014 - Added official support for option -conv_cond (thanks to Ganesh Natarajan). - Fixed print bugs. - Updated documents. Version 1.4.24 January 15, 2014 - Updated test programs. Version 1.4.23 January 14, 2014 - Fixed bugs in test programs. Version 1.4.22 January 11, 2014 - Updated test programs. - Updated documents. Version 1.4.21 January 9, 2014 - Updated test programs. Version 1.4.20 December 15, 2013 - Updated configure.in (thanks to Wei Cai). Version 1.4.19 November 28, 2013 - Updated documents. Version 1.4.18 November 27, 2013 - Fixed specification of configure, which enabled debug option of Fortran compilers by default (thanks to Svetlana Tkachenko). Version 1.4.17 November 13, 2013 - Updated documents (thanks to Svetlana Tkachenko). Version 1.4.16 November 12, 2013 - Fixed print bug in lis_vector.c (thanks to Svetlana Tkachenko). Version 1.4.15 September 26, 2013 - Updated documents (thanks to Svetlana Tkachenko). Version 1.4.14 September 17, 2013 - Fixed errors in user guide (thanks to Svetlana Tkachenko). Version 1.4.13 September 9, 2013 - Fixed bug in lis_vector_set_size(), which caused error in Fortran programs (thanks to Ralf Greve and Tatsuro Sato). Version 1.4.12 August 17, 2013 - Fixed errors in user guide (thanks to Sushilkumar Sreekumar). Version 1.4.11 April 19, 2013 - Fixed bug in Makefile.am. Version 1.4.10 April 3, 2013 - Fixed long double related bug, which caused error on Cygwin. Version 1.4.9 March 21, 2013 - Updated documents. - Changed names of SAAMG related files. Version 1.4.8 March 19, 2013 - Added sample script for gnuplot. Version 1.4.7 March 18, 2013 - Added nonsymmetric test matrix. Version 1.4.6 March 15, 2013 - Updated documents. Version 1.4.5 March 14, 2013 - Added support for gnuplot. Version 1.4.4 March 13, 2013 - Updated documents. Version 1.4.3 March 12, 2013 - Fixed print bug in test*.c. Version 1.4.2 March 11, 2013 - Updated documents. Version 1.4.1 March 10, 2013 - Updated documents. Version 1.4.0 March 8, 2013 - Added official support for long double (quadruple) precision operations. - Updated documents. - Changed version number to 1.4. Version 1.3.33 March 6, 2013 - Fixed long double related print bug. Version 1.3.32 March 4, 2013 - Fixed deallocate bug in lis_*_solver_AMGCG.F90 (thanks to Susi Lehtola). Version 1.3.31 March 4, 2013 - Fixed memory leak bug in lis_*_solver_AMGCG.F90 (thanks to Takashi Shiroto). - Updated documents. Version 1.3.30 March 2, 2013 - Updated documents. Version 1.3.29 March 1, 2013 - Updated documents. Version 1.3.28 February 27, 2013 - Updated documents. Version 1.3.27 February 24, 2013 - Updated documents. Version 1.3.26 February 22, 2013 - Updated documents. Version 1.3.25 February 18, 2013 - Updated documents. Version 1.3.24 February 16, 2013 - Updated documents. Version 1.3.23 February 15, 2013 - Updated documents. - Updated man pages. Version 1.3.22 February 13, 2013 - Fixed bug in lis_esolver_rqi.c. Version 1.3.21 February 12, 2013 - Updated documents. - Added support for zip format. Version 1.3.20 February 7, 2013 - Fixed bug in configure.in. Version 1.3.19 February 4, 2013 - Fixed bug in lis_output*.c. Version 1.3.18 February 2, 2013 - Updated documents. - Updated configure.in. - Updated win/Makefile.in. Version 1.3.17 February 1, 2013 - Added experimental support for long double (thanks to Kiyoshi Ishikawa). Version 1.3.16 January 24, 2013 - Updated documents. Version 1.3.15 January 21, 2013 - Updated documents. Version 1.3.14 January 18, 2013 - Updated documents. Version 1.3.13 January 17, 2013 - Updated documents. - Changed name of matrix storage format JDS (Jagged Diagonal Storage) to JAD (Jagged Diagonal). - Fixed bug in Makefile.am, which excluded lis_esolver_get_rhistory.3. Version 1.3.12 January 16, 2013 - Updated documents. - Updated man pages. Version 1.3.11 January 10, 2013 - Updated documents. - Updated man pages. Version 1.3.10 January 9, 2013 - Updated documents. Version 1.3.9 January 7, 2013 - Updated documents (thanks to Takashi Shiroto). Version 1.3.8 January 6, 2013 - Updated documents (thanks to Jeffrey Kollasch). Version 1.3.7 December 30, 2012 - Updated win/Makefile.in. - Updated win/configure.bat. Version 1.3.6 December 26, 2012 - Updated win/Makefile.in. - Changed file name lis_config_win32.h to lis_config_win.h. Version 1.3.5 December 22, 2012 - Updated win/Makefile.in. - Updated win/configure.bat. - Updated include/lis_config_win32.h. Version 1.3.4 December 20, 2012 - Updated win/configure.bat. Version 1.3.3 December 18, 2012 - Updated contents of directory win. - Updated documents. Version 1.3.2 December 17, 2012 - Updated documents. - Updated win/configure.bat. Version 1.3.1 December 15, 2012 - Added batch script to configure win/Makefile. - Updated contents of directory win. Version 1.3.0 December 11, 2012 - Added official support for nmake. - Fixed bug in win/Makefile. - Changed version number to 1.3. Version 1.2.131 December 11, 2012 - Updated documents. - Fixed bug in Makefile.mak. - Changed file name Makefile.mak to Makefile. Version 1.2.130 December 11, 2012 - Updated documents. - Updated Makefile.mak. - Changed directory name win32 to win. Version 1.2.129 December 10, 2012 - Updated documents. - Updated Makefile.mak. Version 1.2.128 December 9, 2012 - Updated documents. - Updated contents of directory win32. Version 1.2.127 December 8, 2012 - Updated Makefile.mak. Version 1.2.126 December 7, 2012 - Updated Makefile.mak. Version 1.2.125 December 6, 2012 - Updated documents. - Updated Makefile.mak. Version 1.2.124 December 5, 2012 - Added experimental support for nmake. Version 1.2.123 November 28, 2012 - Updated configure.in. Version 1.2.122 November 14, 2012 - Updated documents (thanks to Yoshikazu Karasawa). - Updated man pages. Version 1.2.121 November 12, 2012 - Updated test.sh. - Updated documents. Version 1.2.120 November 7, 2012 - Updated test.sh (thanks to Shoji Itoh). Version 1.2.119 November 5, 2012 - Updated configure.in and documents (thanks to Shoji Itoh). Version 1.2.118 October 27, 2012 - Fixed memory leak bug in eigensolvers (thanks to Christian Schroeppel). Version 1.2.117 October 13, 2012 - Fixed bug in lis_solver_minres.c. Version 1.2.116 September 19, 2012 - Updated man pages. - Updated Makefile.am. Version 1.2.115 September 12, 2012 - Fixed bug in lis_m_solver_AMGCG.F90. Version 1.2.114 September 11, 2012 - Updated lis_*_solver_AMGCG.F90. Version 1.2.113 September 10, 2012 - Updated documents. Version 1.2.112 September 9, 2012 - Updated man pages. - Updated documents. - Changed names of functions lis_get_solvername(), lis_get_preconname(), and lis_get_esolvername() to lis_solver_get_solvername(), lis_solver_get_preconname(), and lis_esolver_get_esolvername(), respectively. Version 1.2.111 September 7, 2012 - Updated man pages. - Updated documents. Version 1.2.110 September 6, 2012 - Updated man pages. - Updated documents. Version 1.2.109 September 5, 2012 - Fixed OpenMP related bug in lis_matrix_vbr.c. Version 1.2.108 September 3, 2012 - Changed names of matrix storage formats CRS (Compressed Row Storage) and CCS (Compressed Column Storage) to CSR (Compressed Sparse Row) and CSC (Compressed Sparse Column), respectively. Version 1.2.107 August 31, 2012 - Fixed long long int related print bug. - Fixed OpenMP related bug in test.sh. Version 1.2.106 August 29, 2012 - Updated documents. - Changed specification of Lanczos. - Restored default configuration of inverse iteration for stability. Version 1.2.105 August 28, 2012 - Updated man pages. - Updated documents. - Changed default preconditioner for inverse iteration. Version 1.2.104 August 27, 2012 - Updated documents. Version 1.2.103 August 26, 2012 - Fixed long long int related I/O bug. Version 1.2.102 August 25, 2012 - Added support for man pages. - Updated documents. Version 1.2.101 August 22, 2012 - Updated documents. Version 1.2.100 August 21, 2012 - Added support for etest6.c. - Changed specification of lis_eli(). - Fixed bug in lis_esolver_*.c. - Updated documents. Version 1.2.99 August 20, 2012 - Changed default configuration of esolve. - Changed specification of lis_esi(). - Fixed bug in lis_esolver_*.c. - Updated documents. Version 1.2.98 August 17, 2012 - Added experimental support for installing programs to ($INSTALLDIR)/bin. - Updated etest5.c. - Updated documents. Version 1.2.97 August 15, 2012 - Added support for Fujitsu Technical Computing Suite on PRIMERGY. Version 1.2.96 August 14, 2012 - Added support for lis_solver_get_precon() and lis_get_preconname(). - Updated print format of eigensolvers. - Updated documents. Version 1.2.95 August 13, 2012 - Changed specification of lis_solve_kernel() to return residual computed by lis_solve_execute() (thanks to Jonathan Schleifer). Version 1.2.94 August 12, 2012 - Fixed OpenMP flag bug. - Updated documents. Version 1.2.93 August 9, 2012 - Fixed OpenMP flag related bug in configure.in for Mac OS X (thanks to Jonathan Schleifer). - Fixed installation destination of examples. Version 1.2.92 August 7, 2012 - Updated support for out-of-tree build (thanks to Susi Lehtola). - Updated documents. Version 1.2.91 August 2, 2012 - Fixed Fortran related bug with flag --enable-shared. Version 1.2.90 July 31, 2012 - Changed names of eigensolvers. - Updated src/esolver/*.c. - Fixed bug in test2.c and etest2.c. - Fixed bug in lis_esolver_set_option_esolver(). - Updated documents. Version 1.2.89 July 29, 2012 - Updated documents. Version 1.2.88 July 27, 2012 - Updated spmvtest*.c. - Updated documents. - Fixed long long int related bug in configure.in for Fujitsu Fortran. - Updated config.guess and config.sub. Version 1.2.87 July 25, 2012 - Added experimental support for 64bit integer with flag --enable-longlong. - Fixed long long int related bug in configure.in for XL and GNU Fortran. - Fixed long long int related bug in test/*.c. - Updated documents. Version 1.2.86 July 24, 2012 - Updated documents. Version 1.2.85 July 21, 2012 - Updated documents. Version 1.2.84 July 20, 2012 - Fixed long long int related bug with flag --enable-mpi. - Fixed bug in spmvtest*.c. Version 1.2.83 July 19, 2012 - Fixed long long int related bug with flag --enable-quad. - Fixed bug in test.sh. - Updated documents. Version 1.2.82 July 17, 2012 - Added experimental support for LIS_INTEGER as integer*8 with flag -DLONGLONG in Fortran. Version 1.2.81 July 13, 2012 - Added experimental support for LIS_INT as long long int with flag -D_LONGLONG in C. - Fixed long long int related bugs. Version 1.2.80 July 12, 2012 - Replaced integer types (thanks to Alessandro Stagni): type | description type | default --------------+-------------------------- ------------------+-------------- int | integer in C ==> LIS_INT | int unsigned int | unsigned integer in C LIS_UNSIGNED_INT | unsigned int integer | integer in Fortran LIS_INTEGER | integer MPI_INT | MPI integer in C LIS_MPI_INT | MPI_INT MPI_INTEGER | MPI integer in Fortran LIS_MPI_INTEGER | MPI_INTEGER MPI_Fint | MPI Fortran integer in C LIS_MPI_Fint | MPI_Fint Version 1.2.79 July 11, 2012 - Updated documents. Version 1.2.78 July 10, 2012 - Updated documents. Version 1.2.77 July 9, 2012 - Updated documents. Version 1.2.76 July 6, 2012 - Updated documents. Version 1.2.75 July 5, 2012 - Updated documents. - Updated support for Fujitsu FX10. Version 1.2.74 July 4, 2012 - Updated documents. - Updated support for Fujitsu FX10. Version 1.2.73 July 3, 2012 - Added support for Fujitsu FX10. Version 1.2.72 July 3, 2012 - Updated documents. Version 1.2.71 June 29, 2012 - Updated documents. Version 1.2.70 June 28, 2012 - Updated documents. - Updated spmvtest*.c. Version 1.2.69 June 28, 2012 - Updated documents. Version 1.2.68 June 26, 2012 - Fixed bug in spmvtest1.c. Version 1.2.67 June 26, 2012 - Updated documents. - Updated print format. Version 1.2.66 June 19, 2012 - Updated documents. Version 1.2.65 June 6, 2012 - Updated documents. Version 1.2.64 May 1, 2012 - Added support for lis_matrix_malloc_*() in Fortran (thanks to Hector Espinoza). Version 1.2.63 May 1, 2012 - Fixed bug in lisf_vector.c to support lis_vector_nrmi() in Fortran (thanks to Hector Espinoza). Version 1.2.62 December 24, 2011 - Added support for preconditioning in MINRES. - Fixed error handling bug in lis_solve(). - Updated documents. Version 1.2.61 November 30, 2011 - Added support for lis_matrix_is_assembled() in lis_matrix_setDLU_*(). Version 1.2.60 November 30, 2011 - Fixed bug in lis_matrix_set_ccs() since 1.2.56, which caused segmentation fault. Version 1.2.59 November 25, 2011 - Updated documents. Version 1.2.58 November 22, 2011 - Updated documents. Version 1.2.57 November 9, 2011 - Added support for flag --with-library-suffix to tag suffix to library names (thanks to Susi Lehtola). Version 1.2.56 November 8, 2011 - Added support for lis_matrix_is_assembled() in lis_matrix_set_*() for calling solver multiple times without allocating new memory (thanks to Malte Thoma). - Changed origin of Fortran subroutine lis_vector_set_size() to 1 (thanks to Kengo Tomida). Version 1.2.55 November 7, 2011 - Changed names of configuration header files config.h and config_win32.h to lis_config.h and lis_config_win32.h to avoid conflict in Fedora packages (thanks to Susi Lehtola). Version 1.2.54 November 2, 2011 - Fixed bug in lis_solver_set_option() (thanks to Akio Yamamoto and Kenichiro Yoshimi). Version 1.2.53 November 29, 2010 - Added support for Intel Composer XE and Microsoft Visual Studio 2010. Version 1.2.52 October 20, 2010 - Fixed bug in lis_solver_minres() with flag -print (thanks to Kazuhiro Ando). Version 1.2.51 September 16, 2010 - Updated documents. Version 1.2.50 September 15, 2010 - Updated documents. Version 1.2.49 September 14, 2010 - Updated documents. Version 1.2.48 September 12, 2010 - Fixed bug in lis_input_hb.c, which caused input error with symmetric matrices in Harwell-Boeing Format (thanks to Shoji Itoh). Version 1.2.47 September 10, 2010 - Added instruction for building shared libraries to documents. Version 1.2.46 September 8, 2010 - Fixed bug in OMPFLAG. Version 1.2.45 September 7, 2010 - Updated documents (thanks to Riccardo Rossi). Version 1.2.44 August 26, 2010 - Updated configure.in. Version 1.2.43 August 25, 2010 - Set default configuration of Mac OS X binaries to 64-bit. - Fixed bug in configure.in for PGI compilers. - Added experimental support for Pathscale compilers. Version 1.2.42 August 20, 2010 - Enabled AC_CHECK_HEADERS for Mac OS X. - Fixed parameters of configure.in for Mac OS X. (Note that default installation of Open MPI for Mac OS does not support Fortran.) Version 1.2.41 August 18, 2010 - Added support for Windows and Mac OS X. - Added instructions for Visual Studio to documents (thanks to Lizhu Tong). - Fixed default location of win32 executable files. Version 1.2.40 August 17, 2010 - Fixed bug in lis_with_fortran_mpi.icproj. - Restored flag USE_SAAMG in project files. Version 1.2.39 August 14, 2010 - Fixed default location of win32 libraries. Version 1.2.38 August 13, 2010 - Fixed character encoding of project files. - Fixed obsolete library name of MPICH2 in project files. Version 1.2.37 August 12, 2010 - Moved project files to directory win32. Version 1.2.36 August 11, 2010 - Fixed bug in configure.in, which caused configure error with XL Fortran. - Fixed bug in src/fortran/lisf_system.c, which caused compile error with Visual Studio. - Temporarily removed flag USE_SAAMG in project files for Visual Studio. Version 1.2.35 July 29, 2010 - Fixed bug in test/test.sh, which caused make check error with flag --enable-quad. - Updated documents. Version 1.2.34 July 28, 2010 - Updated documents. Version 1.2.33 July 23, 2010 - Fixed bug in spmvtests. - Updated documents. Version 1.2.32 July 23, 2010 - Added experimental support for flags -(e)storage and -(e)storage_block. - Updated spmvtests and documents. - Changed specification of precision flag: flags | description flags | description -------------+-------------------- -------+----------------------------- -precision | for lis_solve() ==> -f | for lis_solve() -eprecision | for lis_esolve() -ef | for lis_esolve() Version 1.2.31 July 22, 2010 - Added support for Rayleigh Quotient Iteration. - Inner solvers for quad precision eigensolvers are switched to quad precision. - Changed order of eigensolvers: flags | eigensolvers flags | eigensolvers ------------+-------------------- ------------+-------------------- -e {cg|4} | CG ==> -e {rqi|4} | Rayleigh Quotient -e {li|5} | Lanczos -e {si|5} | Subspace -e {si|6} | Subspace -e {li|6} | Lanczos -e {cr|7} | CR -e {cg|7} | CG -e {cr|8} | CR - Updated documents. Version 1.2.30 July 21, 2010 - Fixed bug in lis_esolve(), which failed to destroy esolver with flag --enable-mpi. Version 1.2.29 July 21, 2010 - Fixed bug in lis_esolve(), which failed to return default eigenvector. - Fixed bug in configure for flag --enable-gprof with f90 compilers. - Added support for 64bit binary on AIX and Solaris x86 with flag --enable-64bit. - Updated documents. Version 1.2.28 July 16, 2010 - Updated documents. Version 1.2.27 July 15, 2010 - Added support for quadruple precision operations in eigensolvers. - Fixed bug in lis.vcproj. Version 1.2.26 July 7, 2010 - Added support for lis_vector_nrmi() for computing infinity norm (thanks to Osamu Nakamura). - Updated documents. Version 1.2.25 July 4, 2010 - Fixed bug in configure.in for gcc with --enable-gprof. Version 1.2.24 July 3, 2010 - Added experimental support for performance analysis using GNU profiler with flag --enable-gprof. - Added test program for lis_esolver_get_evectors(). - Updated documents. Version 1.2.23 July 1, 2010 - Added support for lis_output_matrix() (thanks to Jun Sato). - Changed specification of lis_esolver_get_evectors(). Version 1.2.22 June 30, 2010 - Fixed bug in test.sh. - Updated spmvtests. - Updated documents. Version 1.2.21 June 29, 2010 - Updated test programs. - Updated documents. Version 1.2.20 June 28, 2010 - Added support for parameter irestart for IDR(s). - Fixed default value of parameter emaxiter for subspace and Lanczos iteration. - Updated documents. Version 1.2.19 June 26, 2010 - Fixed default value of parameter etol to avoid breakdown. Version 1.2.18 June 24, 2010 - Fixed minor bug in test.sh for SAAMG with flag --enable-omp. - Fixed minor bug in lis_solver.c for SAAMG with flag --enable-mpi. - Added support for three dimensional test programs. - Updated documents. Version 1.2.17 June 21, 2010 - Fixed print bug in lis_esolver_cg.c. Version 1.2.16 June 18, 2010 - Fixed errors in documents (thanks to Osamu Nakamura). - Added support for lis_output() in Fortran (thanks to Hiroshi Imamura). - Updated spmvtests (thanks to Katsunori Matsuzaki). - Fixed errors in number of nonzero entries printed by test programs test2, etest2, spmvtest1 and spmvtest2 with flag --enable-mpi. - Fixed bug in spmvtests, which fail to measure elapsed time when matrix is very small. - Fixed bug in spmvtests for single threaded environments. - Added support for flag -m to specify default mode for Lanczos and subspace iteration. - Updated documents. Version 1.2.15 September 8, 2009 - Separated core kernel of lis_solve() to avoid overhead of handling preconditioner when called multiple times. It is now implemented as lis_solve_kernel(), to be combined with lis_precon_create() and lis_precon_destroy() (thanks to Masakazu Akiyama). - Inverse iteration and subspace iteration were rewritten using lis_solve_kernel(). Performance of Lanczos iteration, which calls inverse iteration to compute eigenvectors, is also affected. - Updated test programs to measure SpMV performance. Version 1.2.14 August 26, 2009 - Fixed bug in configure.in, which omitted OMPFLAG when target architecture is specified with flag --enable-mpi. - Updated test programs to measure SpMV performance. - Fixed description of test program etest3 in documents. Version 1.2.13 August 15, 2009 - Fixed specification of lis_matrix_split() to keep original data of splitted matrix (thanks to Tim Cramer). Version 1.2.12 July 21, 2009 - Fixed minor bugs related to SpMV test programs. - Updated documents. Version 1.2.11 July 7, 2009 - Fixed specification of lis_vector_gather() to share vector values among processes (thanks to Kouji Takatani). - Updated documents. Version 1.2.10 July 4, 2009 - Added support for lis_vector_scatter() and lis_vector_gather() to handle vector elements, mainly for MPI configuration. - Added official support for lis_solver_get_status() and lis_esolver_get_status() to get correct status of solvers (thanks to Tim Cramer). - Fixed bug in link error for Fortran 77 based MPI compilers. - Updated documents. Version 1.2.9 July 1, 2009 - Added official support for flag -saamg_theta to set drop criteria for AMG. - Fixed default parameter of theta for unsymmetric AMG, which caused error with Version 1.2.6 or higher (thanks to Kouji Takatani). - Added support for test programs to measure SpMV performance. - Fixed MPILIBS related bug in configure.in. - Fixed minor bugs. - Updated documents. Version 1.2.8 June 3, 2009 - Fixed minor bug in configure script. Version 1.2.7 June 1, 2009 - Fixed bug in MPI configuration, which caused error on Hitachi SR16000 and IBM p575. - Fixed minor bugs. Version 1.2.6 March 23, 2009 - Changed location of config*.h, to be installed in $(includedir). - Fixed minor bugs. Version 1.2.5 December 14, 2008 - Added full functions for handling elements of vectors and matrices via Fortran interface. - Added support for matrix storage format conversion in eigensolvers. - Added support for PGI and Hitachi compilers. - Fixed bug in configure, which omitted flag -DUSE_MPI of Fortran based MPI compilers. - Fixed bug in test/test3f.F, which caused make check error with -DUSE_MPI. Version 1.2.4 December 5, 2008 - Added experimental support for conjugate residual for nonsymmetric eigenproblems. - Fixed minor bugs. Version 1.2.3 December 4, 2008 - Fixed origin of lis_vector_get_value() in Fortran (thanks to Kengo Tomida). - Fixed minor bugs. Version 1.2.2 November 25, 2008 - Fixed bug in test/test2.c, which caused make check error (thanks to Takahito Miki). - Fixed bug in configure for combination of SAAMG and gfortran based MPI compiler. SAAMG can be built with gfortran 4.3 or higher. - Fixed minor bugs. Version 1.2.1 November 18, 2008 - Fixed origin of lis_vector_set_value() in Fortran (thanks to Yutaka Masuda). Version 1.2.0 November 12, 2008 - Added support for eigensolvers (power iteration, inverse iteration, approximate inverse iteration, conjugate gradient, Lanczos iteration, and subspace iteration) for real sparse matrices. - Added support for MINRES (thanks to Hirofumi Notsu). - Changed names of lis_output_residual_history() and lis_get_residual_history() to lis_solver_output_rhistory() and lis_solver_get_rhistory(), respectively. - Fixed convergence criteria of TFQMR and BiCGSTAB(l) (thanks to Shoji Itoh). - Added support for NEC SX-9. - Updated config.guess and config.sub. - Added optimization options for Fujitsu Compiler (thanks to Fujitsu Limited). - Fixed location of include statement in lisf_init.F to avoid compiling error with Fujitsu Compiler. - Fixed configure script to choose mpiexec if mpirun is not available. - Fixed configure script to choose Fortran 9x based MPI compiler if SAAMG is enabled. - Fixed minor bugs. - Updated documents. Version 1.1.2 December 24, 2007 - Added support for IDR(s). - Fixed minor bugs. Version 1.1.1 December 6, 2007 - Changed specification of lis_initialize() to avoid MPICH1 related error (thanks to Takashi Yoshida). - Fixed minor bugs. Version 1.1.0 October 31, 2007 - Added support for CR, BiCR, CRS, BiCRSTAB, GPBiCR, BiCRSafe, and FGMRES. - Added support for ILUT preconditioner. - Added support for binary package for Windows. - Added support for graph plotting tool for Windows. - Fixed minor bugs. Version 1.1.0 beta2 June 18, 2007 - Added support for BiCGSafe. - Fixed bug caused by memory leaks. Version 1.1.0 beta1 November 7, 2006 - Added support for quadruple precision operations for linear solvers. - Added support for Fortran user interface. - Added support for unsymmetric SAAMG preconditioner (symmetric structure only). - Added support for Crout ILU preconditioner. - Added support for additive Schwarz preconditioner. - Added support for user-defined preconditioners. - Added support for Harwell-Boeing format. - Integrated Lis-AMG into Lis. - Introduced Autotools. - Changed specifications: - Introduced struct LIS_SOLVER (integrated type parameters). - Changed declaration of matrices (thanks to David Moloney) and vectors. - Changed notation of command line options. Version 1.0.2 April 17, 2006 - Fixed bug caused by not correctly setting value of beginning and end of line when lis_vector_create() is called. - Fixed bug caused by wrong selection of linear solvers specified with lis_set_options(): flags | solvers flags | solvers ----------------+--------------- ----------------+--------------- -i {tfqmr|7} | Orthomin ==> -i {tfqmr|7} | TFQMR -i {orthomin|8}| GMRES -i {orthomin|8}| Orthomin -i {gmres|9} | TFQMR -i {gmres|9} | GMRES - Fixed bug in allocation of temporary array of Orthomin. - Fixed bug in output of residual norm in TFQMR. - Fixed bug in array assignment of BiCG with ILU preconditioner. - Fixed minor bugs. Version 1.0.1 November 6, 2005 - Added support for Lis-AMG. - Added support for English manual. - Added support for test1b. - Fixed minor bugs. Version 1.0.0 September 20, 2005 - Released first version.