7. Limitations and known issues
8. Copyright and License
-Copyright (c) 2009, Timothy A. Davis. All rights reserved. +Copyright (c) 2009-2010, Timothy A. Davis. All rights reserved.Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: @@ -479,6 +499,8 @@
8. Copyright and License
9. Version and Change-Log
-
+
- Version 1.0.4, October 27, 2010. Update to this help file only. +
- Version 1.0.3, October 9, 2009. Removed the Update button which when
clicked would download a fresh copy of matrices/UFstats.csv. UFgui
updates itself automatically every 30 days. If you want to force a
@@ -506,8 +528,12 @@
9. Version and Change-Log
10. For Further Help
Contact the author of this UFgui program, and the maintainer of the UF Sparse -Matrix Collection: Tim Davis (http://www.cise.ufl.edu/~davis, email -davis@cise.ufl.edu or DrTimothyAldenDavis@gmail.com). To print this document, +Matrix Collection: Tim Davis ( +http://www.cise.ufl.edu/~davis, email +davis@cise.ufl.edu +or +DrTimothyAldenDavis@gmail.com). +To print this document, open the file UFhelp.html in your favorite web browser.11. Acknowledgements
diff --git a/CSparse/MATLAB/UFget/mat/UF_Index.mat b/CSparse/MATLAB/UFget/mat/UF_Index.mat index 858a66c07a..5f549d931c 100644 Binary files a/CSparse/MATLAB/UFget/mat/UF_Index.mat and b/CSparse/MATLAB/UFget/mat/UF_Index.mat differ diff --git a/CSparse/MATLAB/UFget/matrices/UFstats.csv b/CSparse/MATLAB/UFget/matrices/UFstats.csv index e8d55cb937..1c828a6057 100644 --- a/CSparse/MATLAB/UFget/matrices/UFstats.csv +++ b/CSparse/MATLAB/UFget/matrices/UFstats.csv @@ -1,5 +1,5 @@ -2272 -02-Nov-2009 14:30:41 +2379 +14-Jan-2011 15:51:12 HB,1138_bus,1138,1138,4054,1,0,0,1,1,1,power network problem HB,494_bus,494,494,1666,1,0,0,1,1,1,power network problem HB,662_bus,662,662,2474,1,0,0,1,1,1,power network problem @@ -2272,3 +2272,110 @@ Puri,ABACUS_shell_ud,23412,23412,218484,1,0,1,0,1,0.9753731955380578,model reduc Puri,ABACUS_shell_ld,23412,23412,218484,0,0,1,0,1,0.8850885826771654,model reduction problem Puri,ABACUS_shell_md,23412,23412,218484,0,0,1,0,1,0.8850885826771654,model reduction problem Puri,ABACUS_shell_hd,23412,23412,218484,0,0,1,0,1,0.8850885826771654,model reduction problem +Stevenson,LargeRegFile,2111154,801374,4944201,1,0,0,0,0,0,circuit simulation problem +Freescale,memchip,2707524,2707524,13343948,1,0,0,0,0.9058013004223939,0.4015860962358702,circuit simulation problem +Freescale,transient,178866,178866,961368,1,0,0,0,0.9996742983044353,0.243597788123496,circuit simulation problem +Freescale,circuit5M,5558326,5558326,59524291,1,0,0,0,0.9999998517635799,0.4198865531852634,circuit simulation problem +Freescale,circuit5M_dc,3523317,3523317,14865409,1,0,0,0,0.9072118240231114,0.7150571517141635,circuit simulation problem +Grund,poli3,16955,16955,37849,1,0,0,0,0.00507322676366421,0.0005743275581506652,economic problem +Grund,poli4,33833,33833,73249,1,0,0,0,0.003551857113862391,0.0002029632636492795,economic problem +Belcastro,human_gene1,22283,22283,24669643,1,0,0,0,1,1,undirected weighted graph +Belcastro,human_gene2,14340,14340,18068388,1,0,0,0,1,1,undirected weighted graph +Belcastro,mouse_gene,45101,45101,28967291,1,0,0,0,1,1,undirected weighted graph +Um,offshore,259789,259789,4242673,1,0,1,1,1,1,electromagnetics problem +SNAP,soc-Epinions1,75888,75888,508837,1,1,0,0,0.405226035056413,0.405226035056413,directed graph +SNAP,soc-LiveJournal1,4847571,4847571,68993773,1,1,0,0,0.7484193555024753,0.7484193555024753,directed graph +SNAP,soc-Slashdot0811,77360,77360,905468,1,1,0,0,0.8669352939826918,0.8669352939826918,directed graph +SNAP,soc-Slashdot0902,82168,82168,948464,1,1,0,0,0.8410650442849082,0.8410650442849082,directed graph +SNAP,wiki-Vote,8297,8297,103689,1,1,0,0,0.0564572905515532,0.0564572905515532,directed graph +SNAP,email-EuAll,265214,265214,420045,1,1,0,0,0.2600511748250413,0.2600511748250413,directed graph +SNAP,email-Enron,36692,36692,367662,1,1,0,0,1,1,directed graph +SNAP,wiki-Talk,2394385,2394385,5021410,1,1,0,0,0.1441208744157517,0.1441208744157517,directed graph +SNAP,cit-HepPh,34546,34546,421578,1,1,0,0,0.003117186276789061,0.003117186276789061,directed graph +SNAP,cit-HepTh,27770,27770,352807,1,1,0,0,0.002738343613933237,0.002738343613933237,directed graph +SNAP,cit-Patents,3774768,3774768,16518948,1,1,0,0,0,0,directed graph +SNAP,ca-AstroPh,18772,18772,396160,1,1,0,0,1,1,undirected graph +SNAP,ca-CondMat,23133,23133,186936,1,1,0,0,1,1,undirected graph +SNAP,ca-GrQc,5242,5242,28980,1,1,0,0,1,1,undirected graph +SNAP,ca-HepPh,12008,12008,237010,1,1,0,0,1,1,undirected graph +SNAP,ca-HepTh,9877,9877,51971,1,1,0,0,1,1,undirected graph +SNAP,web-BerkStan,685230,685230,7600595,1,1,0,0,0.2502764586193581,0.2502764586193581,directed graph +SNAP,web-Google,916428,916428,5105039,1,1,0,0,0.3067510355944391,0.3067510355944391,directed graph +SNAP,web-NotreDame,325729,325729,1497134,1,1,0,0,0.5165359238309862,0.5165359238309862,directed graph +SNAP,web-Stanford,281903,281903,2312497,1,1,0,0,0.2766368994208425,0.2766368994208425,directed graph +SNAP,amazon0302,262111,262111,1234877,1,1,0,0,0.5427018237443891,0.5427018237443891,directed graph +SNAP,amazon0312,400727,400727,3200440,1,1,0,0,0.5315337891039982,0.5315337891039982,directed graph +SNAP,amazon0505,410236,410236,3356824,1,1,0,0,0.5465803390347542,0.5465803390347542,directed graph +SNAP,amazon0601,403394,403394,3387388,1,1,0,0,0.5573497928197183,0.5573497928197183,directed graph +SNAP,p2p-Gnutella04,10879,10879,39994,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella05,8846,8846,31839,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella06,8717,8717,31525,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella08,6301,6301,20777,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella09,8114,8114,26013,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella24,26518,26518,65369,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella25,22687,22687,54705,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella30,36682,36682,88328,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella31,62586,62586,147892,1,1,0,0,0,0,directed graph +SNAP,roadNet-CA,1971281,1971281,5533214,1,1,0,0,1,1,undirected graph +SNAP,roadNet-PA,1090920,1090920,3083796,1,1,0,0,1,1,undirected graph +SNAP,roadNet-TX,1393383,1393383,3843320,1,1,0,0,1,1,undirected graph +SNAP,as-735,7716,7716,26467,1,1,0,0,1,1,undirected graph sequence +SNAP,as-Skitter,1696415,1696415,22190596,1,1,0,0,1,1,undirected graph +SNAP,as-caida,31379,31379,106762,1,0,0,0,1,0.07869841329311927,directed weighted graph sequence +SNAP,Oregon-1,11492,11492,46818,1,1,0,0,1,1,undirected graph sequence +SNAP,Oregon-2,11806,11806,65460,1,1,0,0,1,1,undirected graph sequence +SNAP,soc-sign-epinions,131828,131828,841372,1,0,0,0,0.3082520316984202,0.3018224331855771,directed weighted graph +SNAP,soc-sign-Slashdot081106,77357,77357,516575,1,0,0,0,0.1859207278710739,0.1786110438948846,directed weighted graph +SNAP,soc-sign-Slashdot090216,81871,81871,545671,1,0,0,0,0.175926519826049,0.168574104176326,directed weighted graph +SNAP,soc-sign-Slashdot090221,82144,82144,549202,1,0,0,0,0.1774246998372184,0.1703271291801559,directed weighted graph +Dziekonski,gsm_106857,589446,589446,21758924,1,0,1,0,1,1,electromagnetics problem +Gleich,flickr,820878,820878,9837214,1,1,0,0,0.653017002578169,0.653017002578169,directed graph +Gleich,minnesota,2642,2642,6606,1,1,0,0,1,1,undirected graph +Gleich,usroads-48,126146,126146,323900,1,1,0,0,1,1,undirected graph +Gleich,usroads,129164,129164,330870,1,1,0,0,1,1,undirected graph +Fluorem,DK01R,903,903,11766,1,0,1,0,0.9614287029365737,0,computational fluid dynamics problem +Fluorem,GT01R,7980,7980,430909,1,0,1,0,0.8811455350661695,9.457852263618717e-06,computational fluid dynamics problem +Fluorem,PR02R,161070,161070,8185136,1,0,1,0,0.9476492341912441,0.0001697393815055858,computational fluid dynamics problem +Fluorem,RM07R,381689,381689,37464962,1,0,1,0,0.9261667922354103,4.260681089287885e-06,computational fluid dynamics problem +Rommes,ww_36_pmec_36,66,66,1194,1,0,1,0,0.3339191564147627,0,eigenvalue/model reduction problem +Rommes,ww_vref_6405,13251,13251,48737,1,0,1,0,0.707540090603374,0.5856980072816208,eigenvalue/model reduction problem +Rommes,xingo_afonso_itaipu,13250,13250,48735,1,0,1,0,0.7075597554196776,0.5857142857142857,eigenvalue/model reduction problem +Rommes,mimo8x8_system,13309,13309,48872,1,0,1,0,0.7060291752177048,0.5844472793832159,eigenvalue/model reduction problem +Rommes,mimo28x28_system,13251,13251,48737,1,0,1,0,0.707540090603374,0.5856980072816208,eigenvalue/model reduction problem +Rommes,mimo46x46_system,13250,13250,48735,1,0,1,0,0.7075597554196776,0.5857142857142857,eigenvalue/model reduction problem +Rommes,juba40k,40337,40337,144945,1,0,1,0,0.6934462083788981,0.5830536596402579,eigenvalue/model reduction problem +Rommes,bauru5727,40366,40366,145019,1,0,1,0,0.6931521565005984,0.5828064190876452,eigenvalue/model reduction problem +Rommes,zeros_nopss_13k,13296,13296,48827,1,0,1,0,0.7083056110002218,0.5842204479929031,eigenvalue/model reduction problem +Rommes,descriptor_xingo6u,20738,20738,73916,1,0,1,0,0.6848202865775669,0.5745268504272713,eigenvalue/model reduction problem +Rommes,nopss_11k,11685,11685,44941,1,0,1,0,0.7502077151335311,0.6233827893175075,eigenvalue/model reduction problem +Rommes,xingo3012,20944,20944,74386,1,0,1,0,0.681002526603102,0.5712336093539642,eigenvalue/model reduction problem +Rommes,bips98_606,7135,7135,34738,1,0,1,0,0.9086859688195991,0.7583878152166104,eigenvalue/model reduction problem +Rommes,bips98_1142,9735,9735,40983,1,0,1,0,0.8020552507215579,0.6694154587839767,eigenvalue/model reduction problem +Rommes,bips98_1450,11305,11305,44678,1,0,1,0,0.7593684272891549,0.6277809047012471,eigenvalue/model reduction problem +Rommes,bips07_1693,13275,13275,49044,1,0,1,0,0.706761938919083,0.5845449278560985,eigenvalue/model reduction problem +Rommes,bips07_1998,15066,15066,62198,1,0,1,0,0.7978660841804587,0.6687809795849786,eigenvalue/model reduction problem +Rommes,bips07_2476,16861,16861,66498,1,0,1,0,0.7574573297176583,0.6349098739830914,eigenvalue/model reduction problem +Rommes,bips07_3078,21128,21128,75729,1,0,1,0,0.6861363964744979,0.5750975292587777,eigenvalue/model reduction problem +Rommes,M10PI_n1,528,528,1317,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,M20PI_n1,1028,1028,2547,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,M40PI_n1,2028,2028,5007,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,M80PI_n1,4028,4028,9927,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,M10PI_n,682,682,1633,1,0,1,0,1,0.1132075471698113,eigenvalue/model reduction problem +Rommes,M20PI_n,1182,1182,2881,1,0,1,0,1,0.04173622704507512,eigenvalue/model reduction problem +Rommes,M40PI_n,2182,2182,5341,1,0,1,0,1,0.02274795268425842,eigenvalue/model reduction problem +Rommes,M80PI_n,4182,4182,10261,1,0,1,0,1,0.01191043353978085,eigenvalue/model reduction problem +Rommes,S10PI_n1,528,528,1317,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,S20PI_n1,1028,1028,2547,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,S40PI_n1,2028,2028,5007,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,S80PI_n1,4028,4028,9927,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,S10PI_n,682,682,1633,1,0,1,0,1,0.1132075471698113,eigenvalue/model reduction problem +Rommes,S20PI_n,1182,1182,2881,1,0,1,0,1,0.04173622704507512,eigenvalue/model reduction problem +Rommes,S40PI_n,2182,2182,5341,1,0,1,0,1,0.02274795268425842,eigenvalue/model reduction problem +Rommes,S80PI_n,4182,4182,10261,1,0,1,0,1,0.01191043353978085,eigenvalue/model reduction problem +Williams,pdb1HYS,36417,36417,4344765,1,0,0,1,1,1,weighted undirected graph +Williams,consph,83334,83334,6010480,1,0,1,1,1,1,2D/3D problem +Williams,cant,62451,62451,4007383,1,0,1,1,1,1,2D/3D problem +Williams,mac_econ_fwd500,206500,206500,1273389,1,0,0,0,0.05995735832675801,0.005599217836274841,economic problem +Williams,mc2depi,525825,525825,2100225,1,0,1,0,0,0,2D/3D problem +Williams,cop20k_A,121192,121192,2624331,1,0,1,0,1,1,2D/3D problem +Williams,webbase-1M,1000005,1000005,3105536,1,0,0,0,0.09552554676231316,0.01588957844838191,weighted directed graph diff --git a/CSparse/Makefile b/CSparse/Makefile index b47121ecc3..ead3ffa775 100644 --- a/CSparse/Makefile +++ b/CSparse/Makefile @@ -1,6 +1,6 @@ # CSparse Makefile -VERSION = 2.2.4 +VERSION = 2.2.5 C: ( cd Lib ; $(MAKE) ) diff --git a/CSparse/README.txt b/CSparse/README.txt index b2396e4f5d..201a6e5684 100644 --- a/CSparse/README.txt +++ b/CSparse/README.txt @@ -1,5 +1,4 @@ -CSparse: a Concise Sparse Matrix package. -Version 2.2.3, Copyright (c) 2006-2009, Timothy A. Davis, Sept 23, 2009. +CSparse: Copyright (c) 2006-2011, Timothy A. Davis. Refer to "Direct Methods for Sparse Linear Systems," Timothy A. Davis, SIAM, Philadelphia, 2006. No detailed user guide is included in this diff --git a/CSparse/Source/cs_util.c b/CSparse/Source/cs_util.c index f714ed8f49..32d9a31e7f 100644 --- a/CSparse/Source/cs_util.c +++ b/CSparse/Source/cs_util.c @@ -35,7 +35,7 @@ cs *cs_spfree (cs *A) cs_free (A->p) ; cs_free (A->i) ; cs_free (A->x) ; - return (cs_free (A)) ; /* free the cs struct and return NULL */ + return ((cs *) cs_free (A)) ; /* free the cs struct and return NULL */ } /* free a numeric factorization */ @@ -46,7 +46,7 @@ csn *cs_nfree (csn *N) cs_spfree (N->U) ; cs_free (N->pinv) ; cs_free (N->B) ; - return (cs_free (N)) ; /* free the csn struct and return NULL */ + return ((csn *) cs_free (N)) ; /* free the csn struct and return NULL */ } /* free a symbolic factorization */ @@ -58,7 +58,7 @@ css *cs_sfree (css *S) cs_free (S->parent) ; cs_free (S->cp) ; cs_free (S->leftmost) ; - return (cs_free (S)) ; /* free the css struct and return NULL */ + return ((css *) cs_free (S)) ; /* free the css struct and return NULL */ } /* allocate a cs_dmperm or cs_scc result */ @@ -82,7 +82,7 @@ csd *cs_dfree (csd *D) cs_free (D->q) ; cs_free (D->r) ; cs_free (D->s) ; - return (cs_free (D)) ; + return ((csd *) cs_free (D)) ; /* free the csd struct and return NULL */ } /* free workspace and return a sparse matrix result */ @@ -98,7 +98,7 @@ int *cs_idone (int *p, cs *C, void *w, int ok) { cs_spfree (C) ; /* free temporary matrix */ cs_free (w) ; /* free workspace */ - return (ok ? p : cs_free (p)) ; /* return result if OK, else free it */ + return (ok ? p : (int *) cs_free (p)) ; /* return result, or free it */ } /* free workspace and return a numeric factorization (Cholesky, LU, or QR) */ diff --git a/CSparse/Tcov/Makefile b/CSparse/Tcov/Makefile index 57ce8442cb..fcdcd37297 100644 --- a/CSparse/Tcov/Makefile +++ b/CSparse/Tcov/Makefile @@ -100,3 +100,4 @@ purge: distclean distclean: clean rm -f cs_demo1 cs_demo2 readhb *.out *.a cs_demo3 cstcov_test cov.sort rm -f cs_*.c + rm -rf *.dSYM diff --git a/CXSparse/Demo/cs_demo.out b/CXSparse/Demo/cs_demo.out index a01cece44b..3cc4a60357 100644 --- a/CXSparse/Demo/cs_demo.out +++ b/CXSparse/Demo/cs_demo.out @@ -1,4 +1,4 @@ -make[1]: Entering directory `/amd/netapp3/vol/research0a/research18/sparse2/SuiteSparse/CXSparse/Demo' +make[1]: Entering directory `/cise/research/sparse2/SuiteSparse/CXSparse/Demo' ./cs_idemo < ../Matrix/t2 --- cs_idemo, size of CS_INT: 4 @@ -394,10 +394,10 @@ tiny entries dropped: 10 blocks: 38 singletons: 37 structural rank: 183 QR natural time: 0.01 resid: 1.42e-27 QR amd(A'*A) time: 0.00 resid: 3.35e-28 -LU natural time: 0.00 resid: 6.20e-28 +LU natural time: 0.01 resid: 6.20e-28 LU amd(A+A') time: 0.00 resid: 1.55e-27 LU amd(S'*S) time: 0.00 resid: 6.98e-28 -LU amd(A'*A) time: 0.01 resid: 6.98e-28 +LU amd(A'*A) time: 0.00 resid: 6.98e-28 ./cs_demo2 < ../Matrix/west0067 --- Matrix: 67-by-67, nnz: 294 (sym: 0: nnz 0), norm: 6.14e+00 @@ -425,7 +425,7 @@ QR amd(A'*A) time: 0.00 resid: 1.61e-02 --- Matrix: 492-by-490, nnz: 49920 (sym: 0: nnz 0), norm: 9.29e-01 blocks: 10 singletons: 8 structural rank: 448 QR natural time: 0.12 resid: nan -QR amd(A'*A) time: 0.15 resid: nan +QR amd(A'*A) time: 0.16 resid: nan ./cs_demo2 < ../Matrix/bcsstk01 --- Matrix: 48-by-48, nnz: 224 (sym: -1: nnz 400), norm: 3.57e+09 @@ -456,24 +456,24 @@ downdate: time: 0.00 (incl solve) resid: 3.74e-17 --- Matrix: 4884-by-4884, nnz: 147631 (sym: -1: nnz 290378), norm: 7.01e+09 blocks: 75 singletons: 74 structural rank: 4884 -QR amd(A'*A) time: 1.98 resid: 2.01e-22 -LU amd(A+A') time: 1.04 resid: 1.10e-22 -LU amd(S'*S) time: 0.99 resid: 1.28e-22 -LU amd(A'*A) time: 1.03 resid: 1.78e-22 -Chol amd(A+A') time: 0.40 resid: 1.19e-22 +QR amd(A'*A) time: 1.66 resid: 2.01e-22 +LU amd(A+A') time: 0.98 resid: 1.10e-22 +LU amd(S'*S) time: 0.95 resid: 1.28e-22 +LU amd(A'*A) time: 0.99 resid: 1.78e-22 +Chol amd(A+A') time: 0.31 resid: 1.19e-22 ./cs_demo3 < ../Matrix/bcsstk16 --- Matrix: 4884-by-4884, nnz: 147631 (sym: -1: nnz 290378), norm: 7.01e+09 chol then update/downdate amd(A+A') symbolic chol time 0.02 -numeric chol time 0.35 +numeric chol time 0.30 solve chol time 0.00 original: resid: 1.19e-22 update: time: 0.00 -update: time: 0.01 (incl solve) resid: 1.12e-23 -rechol: time: 0.37 (incl solve) resid: 1.17e-23 -downdate: time: 0.00 +update: time: 0.00 (incl solve) resid: 1.12e-23 +rechol: time: 0.31 (incl solve) resid: 1.17e-23 +downdate: time: 0.01 downdate: time: 0.01 (incl solve) resid: 4.09e-22 ./cs_di_demo1 < ../Matrix/t1 T: @@ -562,11 +562,11 @@ LU amd(A'*A) time: 0.00 resid: 1.53e-17 zero entries dropped: 71 tiny entries dropped: 10 blocks: 38 singletons: 37 structural rank: 183 -QR natural time: 0.00 resid: 1.42e-27 +QR natural time: 0.01 resid: 1.42e-27 QR amd(A'*A) time: 0.00 resid: 3.35e-28 -LU natural time: 0.01 resid: 6.20e-28 +LU natural time: 0.00 resid: 6.20e-28 LU amd(A+A') time: 0.00 resid: 1.55e-27 -LU amd(S'*S) time: 0.00 resid: 6.98e-28 +LU amd(S'*S) time: 0.01 resid: 6.98e-28 LU amd(A'*A) time: 0.00 resid: 6.98e-28 ./cs_di_demo2 < ../Matrix/west0067 @@ -595,7 +595,7 @@ QR amd(A'*A) time: 0.00 resid: 1.61e-02 --- Matrix: 492-by-490, nnz: 49920 (sym: 0: nnz 0), norm: 9.29e-01 blocks: 10 singletons: 8 structural rank: 448 QR natural time: 0.12 resid: nan -QR amd(A'*A) time: 0.16 resid: nan +QR amd(A'*A) time: 0.15 resid: nan ./cs_di_demo2 < ../Matrix/bcsstk01 --- Matrix: 48-by-48, nnz: 224 (sym: -1: nnz 400), norm: 3.57e+09 @@ -626,23 +626,23 @@ downdate: time: 0.00 (incl solve) resid: 3.74e-17 --- Matrix: 4884-by-4884, nnz: 147631 (sym: -1: nnz 290378), norm: 7.01e+09 blocks: 75 singletons: 74 structural rank: 4884 -QR amd(A'*A) time: 1.73 resid: 2.01e-22 -LU amd(A+A') time: 0.93 resid: 1.10e-22 -LU amd(S'*S) time: 0.98 resid: 1.28e-22 -LU amd(A'*A) time: 1.00 resid: 1.78e-22 -Chol amd(A+A') time: 0.39 resid: 1.19e-22 +QR amd(A'*A) time: 1.67 resid: 2.01e-22 +LU amd(A+A') time: 0.98 resid: 1.10e-22 +LU amd(S'*S) time: 0.95 resid: 1.28e-22 +LU amd(A'*A) time: 0.98 resid: 1.78e-22 +Chol amd(A+A') time: 0.32 resid: 1.19e-22 ./cs_di_demo3 < ../Matrix/bcsstk16 --- Matrix: 4884-by-4884, nnz: 147631 (sym: -1: nnz 290378), norm: 7.01e+09 chol then update/downdate amd(A+A') -symbolic chol time 0.03 -numeric chol time 0.36 -solve chol time 0.02 +symbolic chol time 0.02 +numeric chol time 0.30 +solve chol time 0.00 original: resid: 1.19e-22 update: time: 0.00 update: time: 0.01 (incl solve) resid: 1.12e-23 -rechol: time: 0.38 (incl solve) resid: 1.17e-23 +rechol: time: 0.31 (incl solve) resid: 1.17e-23 downdate: time: 0.00 downdate: time: 0.01 (incl solve) resid: 4.09e-22 ./cs_dl_demo1 < ../Matrix/t1 @@ -732,8 +732,8 @@ LU amd(A'*A) time: 0.00 resid: 1.53e-17 zero entries dropped: 71 tiny entries dropped: 10 blocks: 38 singletons: 37 structural rank: 183 -QR natural time: 0.00 resid: 1.42e-27 -QR amd(A'*A) time: 0.01 resid: 3.35e-28 +QR natural time: 0.01 resid: 1.42e-27 +QR amd(A'*A) time: 0.00 resid: 3.35e-28 LU natural time: 0.00 resid: 6.20e-28 LU amd(A+A') time: 0.00 resid: 1.55e-27 LU amd(S'*S) time: 0.00 resid: 6.98e-28 @@ -765,7 +765,7 @@ QR amd(A'*A) time: 0.00 resid: 1.61e-02 --- Matrix: 492-by-490, nnz: 49920 (sym: 0: nnz 0), norm: 9.29e-01 blocks: 10 singletons: 8 structural rank: 448 QR natural time: 0.13 resid: nan -QR amd(A'*A) time: 0.16 resid: nan +QR amd(A'*A) time: 0.17 resid: nan ./cs_dl_demo2 < ../Matrix/bcsstk01 --- Matrix: 48-by-48, nnz: 224 (sym: -1: nnz 400), norm: 3.57e+09 @@ -796,23 +796,23 @@ downdate: time: 0.00 (incl solve) resid: 3.74e-17 --- Matrix: 4884-by-4884, nnz: 147631 (sym: -1: nnz 290378), norm: 7.01e+09 blocks: 75 singletons: 74 structural rank: 4884 -QR amd(A'*A) time: 1.92 resid: 2.01e-22 -LU amd(A+A') time: 1.01 resid: 1.10e-22 -LU amd(S'*S) time: 1.09 resid: 1.28e-22 +QR amd(A'*A) time: 1.87 resid: 2.01e-22 +LU amd(A+A') time: 1.00 resid: 1.10e-22 +LU amd(S'*S) time: 0.94 resid: 1.28e-22 LU amd(A'*A) time: 0.99 resid: 1.78e-22 -Chol amd(A+A') time: 0.51 resid: 1.19e-22 +Chol amd(A+A') time: 0.39 resid: 1.19e-22 ./cs_dl_demo3 < ../Matrix/bcsstk16 --- Matrix: 4884-by-4884, nnz: 147631 (sym: -1: nnz 290378), norm: 7.01e+09 chol then update/downdate amd(A+A') symbolic chol time 0.03 -numeric chol time 0.49 +numeric chol time 0.34 solve chol time 0.01 original: resid: 1.19e-22 update: time: 0.00 update: time: 0.01 (incl solve) resid: 1.12e-23 -rechol: time: 0.48 (incl solve) resid: 1.17e-23 +rechol: time: 0.36 (incl solve) resid: 1.17e-23 downdate: time: 0.00 downdate: time: 0.01 (incl solve) resid: 4.09e-22 ./cs_ci_demo1 < ../Matrix/t2 @@ -929,7 +929,7 @@ LU amd(A'*A) time: 0.00 resid: 4.76e-17 --- Matrix: 492-by-490, nnz: 49920 (sym: 0: nnz 0), norm: 9.29e-01 blocks: 10 singletons: 8 structural rank: 448 QR natural time: 0.41 resid: nan -QR amd(A'*A) time: 0.48 resid: nan +QR amd(A'*A) time: 0.45 resid: nan ./cs_ci_demo2 < ../Matrix/young1c --- Matrix: 841-by-841, nnz: 4089 (sym: 0: nnz 0), norm: 7.30e+02 @@ -937,16 +937,16 @@ blocks: 1 singletons: 0 structural rank: 841 QR natural time: 0.03 resid: 1.81e-16 QR amd(A'*A) time: 0.02 resid: 1.57e-16 LU natural time: 0.01 resid: 1.39e-16 -LU amd(A+A') time: 0.02 resid: 2.95e-16 -LU amd(S'*S) time: 0.01 resid: 3.37e-16 -LU amd(A'*A) time: 0.00 resid: 3.37e-16 +LU amd(A+A') time: 0.01 resid: 2.95e-16 +LU amd(S'*S) time: 0.00 resid: 3.37e-16 +LU amd(A'*A) time: 0.01 resid: 3.37e-16 ./cs_ci_demo2 < ../Matrix/qc324 --- Matrix: 324-by-324, nnz: 26730 (sym: 0: nnz 0), norm: 1.71e+00 blocks: 1 singletons: 0 structural rank: 324 -QR natural time: 0.07 resid: 9.42e-17 +QR natural time: 0.06 resid: 9.42e-17 QR amd(A'*A) time: 0.07 resid: 8.94e-17 -LU natural time: 0.03 resid: 6.01e-17 +LU natural time: 0.02 resid: 6.01e-17 LU amd(A+A') time: 0.03 resid: 4.05e-17 LU amd(S'*S) time: 0.03 resid: 4.71e-17 LU amd(A'*A) time: 0.03 resid: 4.71e-17 @@ -954,7 +954,7 @@ LU amd(A'*A) time: 0.03 resid: 4.71e-17 --- Matrix: 1600-by-1600, nnz: 7840 (sym: 0: nnz 0), norm: 1.41e+01 blocks: 1 singletons: 0 structural rank: 1600 -QR amd(A'*A) time: 0.05 resid: 1.04e-15 +QR amd(A'*A) time: 0.04 resid: 1.04e-15 LU amd(A+A') time: 0.01 resid: 3.55e-16 LU amd(S'*S) time: 0.02 resid: 4.03e-16 LU amd(A'*A) time: 0.02 resid: 4.03e-16 @@ -993,7 +993,7 @@ QR amd(A'*A) time: 0.01 resid: 6.15e-25 LU amd(A+A') time: 0.01 resid: 2.33e-25 LU amd(S'*S) time: 0.01 resid: 3.96e-25 LU amd(A'*A) time: 0.00 resid: 3.96e-25 -Chol amd(A+A') time: 0.01 resid: 1.58e-25 +Chol amd(A+A') time: 0.00 resid: 1.58e-25 ./cs_ci_demo3 < ../Matrix/mhd1280b --- Matrix: 1280-by-1280, nnz: 12029 (sym: -1: nnz 22778), norm: 8.00e+01 @@ -1005,7 +1005,7 @@ solve chol time 0.00 original: resid: 1.51e-25 update: time: 0.00 update: time: 0.00 (incl solve) resid: 1.75e-25 -rechol: time: 0.00 (incl solve) resid: 1.71e-25 +rechol: time: 0.01 (incl solve) resid: 1.71e-25 downdate: time: 0.00 downdate: time: 0.00 (incl solve) resid: 5.85e-25 ./cs_cl_demo1 < ../Matrix/t2 @@ -1121,8 +1121,8 @@ LU amd(A'*A) time: 0.00 resid: 4.76e-17 --- Matrix: 492-by-490, nnz: 49920 (sym: 0: nnz 0), norm: 9.29e-01 blocks: 10 singletons: 8 structural rank: 448 -QR natural time: 0.51 resid: nan -QR amd(A'*A) time: 0.52 resid: nan +QR natural time: 0.42 resid: nan +QR amd(A'*A) time: 0.45 resid: nan ./cs_cl_demo2 < ../Matrix/young1c --- Matrix: 841-by-841, nnz: 4089 (sym: 0: nnz 0), norm: 7.30e+02 @@ -1138,18 +1138,18 @@ LU amd(A'*A) time: 0.01 resid: 3.37e-16 --- Matrix: 324-by-324, nnz: 26730 (sym: 0: nnz 0), norm: 1.71e+00 blocks: 1 singletons: 0 structural rank: 324 QR natural time: 0.06 resid: 9.42e-17 -QR amd(A'*A) time: 0.08 resid: 8.94e-17 +QR amd(A'*A) time: 0.07 resid: 8.94e-17 LU natural time: 0.03 resid: 6.01e-17 -LU amd(A+A') time: 0.03 resid: 4.05e-17 +LU amd(A+A') time: 0.02 resid: 4.05e-17 LU amd(S'*S) time: 0.03 resid: 4.71e-17 LU amd(A'*A) time: 0.03 resid: 4.71e-17 ./cs_cl_demo2 < ../Matrix/neumann --- Matrix: 1600-by-1600, nnz: 7840 (sym: 0: nnz 0), norm: 1.41e+01 blocks: 1 singletons: 0 structural rank: 1600 -QR amd(A'*A) time: 0.05 resid: 1.04e-15 +QR amd(A'*A) time: 0.04 resid: 1.04e-15 LU amd(A+A') time: 0.01 resid: 3.55e-16 -LU amd(S'*S) time: 0.02 resid: 4.03e-16 +LU amd(S'*S) time: 0.03 resid: 4.03e-16 LU amd(A'*A) time: 0.02 resid: 4.03e-16 ./cs_cl_demo2 < ../Matrix/c4 @@ -1185,8 +1185,8 @@ blocks: 20 singletons: 14 structural rank: 1280 QR amd(A'*A) time: 0.01 resid: 6.15e-25 LU amd(A+A') time: 0.01 resid: 2.33e-25 LU amd(S'*S) time: 0.01 resid: 3.96e-25 -LU amd(A'*A) time: 0.00 resid: 3.96e-25 -Chol amd(A+A') time: 0.01 resid: 1.58e-25 +LU amd(A'*A) time: 0.01 resid: 3.96e-25 +Chol amd(A+A') time: 0.00 resid: 1.58e-25 ./cs_cl_demo3 < ../Matrix/mhd1280b --- Matrix: 1280-by-1280, nnz: 12029 (sym: -1: nnz 22778), norm: 8.00e+01 @@ -1197,8 +1197,8 @@ numeric chol time 0.00 solve chol time 0.00 original: resid: 1.51e-25 update: time: 0.00 -update: time: 0.00 (incl solve) resid: 1.75e-25 -rechol: time: 0.01 (incl solve) resid: 1.71e-25 +update: time: 0.01 (incl solve) resid: 1.75e-25 +rechol: time: 0.00 (incl solve) resid: 1.71e-25 downdate: time: 0.00 downdate: time: 0.00 (incl solve) resid: 5.85e-25 -make[1]: Leaving directory `/amd/netapp3/vol/research0a/research18/sparse2/SuiteSparse/CXSparse/Demo' +make[1]: Leaving directory `/cise/research/sparse2/SuiteSparse/CXSparse/Demo' diff --git a/CXSparse/Doc/ChangeLog b/CXSparse/Doc/ChangeLog index 11bbf5a732..29491885a7 100644 --- a/CXSparse/Doc/ChangeLog +++ b/CXSparse/Doc/ChangeLog @@ -1,3 +1,9 @@ +Jan 25, 2011: version 2.2.5 + + * minor fix to "make install" + * minor change to cs_util.c, typecast return (cs_free (...)) + * minor fixes to UFget: help file, UFsettings.txt + Nov 30, 2009: version 2.2.4 * added "make install" and "make uninstall" to CXSparse diff --git a/CXSparse/MATLAB/UFget/UFget_defaults.m b/CXSparse/MATLAB/UFget/UFget_defaults.m index 00924c1b9a..a907e94bd8 100644 --- a/CXSparse/MATLAB/UFget/UFget_defaults.m +++ b/CXSparse/MATLAB/UFget/UFget_defaults.m @@ -31,7 +31,7 @@ f = -1 ; try - f = fopen (sprintf ('%sUFsettings.txt', s), 'r') ; + f = fopen (sprintf ('%sUFsettings.txt', this), 'r') ; if (f >= 0) % get the location of the mat directory s = fgetl (f) ; diff --git a/CXSparse/MATLAB/UFget/UFgui.jar b/CXSparse/MATLAB/UFget/UFgui.jar index 51c14e28c5..756923a64d 100644 Binary files a/CXSparse/MATLAB/UFget/UFgui.jar and b/CXSparse/MATLAB/UFget/UFgui.jar differ diff --git a/CXSparse/MATLAB/UFget/UFgui.java b/CXSparse/MATLAB/UFget/UFgui.java index 2f4f635e2b..d4f1136bd8 100644 --- a/CXSparse/MATLAB/UFget/UFgui.java +++ b/CXSparse/MATLAB/UFget/UFgui.java @@ -645,7 +645,7 @@ public void actionPerformed (ActionEvent e) real_group.add (real_either_Button) ; real_yes_Button.setToolTipText - ("Select 'real' for real matrices only.") ; + ("Select 'real' for real matrices only (includes integer and binary).") ; real_no_Button.setToolTipText ("Select 'complex' for complex matrices only.") ; real_either_Button.setToolTipText @@ -1172,7 +1172,7 @@ else if (k == 0) "# of rows in the matrix", // 7:nrows "# of columns in the matrix", // 8:ncols "# of nonzeros in the matrix", // 9:nnz - "if the matrix is real", // 10:isReal + "if the matrix is real (not complex)", // 10:isReal "if the matrix is binary", // 11:isBinary "if the matrix arises from a 2D/3D discretization", // 12:isND "if the matrix is symmetric positive definite", // 13:posdef diff --git a/CXSparse/MATLAB/UFget/UFhelp.html b/CXSparse/MATLAB/UFget/UFhelp.html index 01775b2093..935a905acf 100644 --- a/CXSparse/MATLAB/UFget/UFhelp.html +++ b/CXSparse/MATLAB/UFget/UFhelp.html @@ -21,13 +21,17 @@1. Overview
UFgui is a Java GUI interface to the University of Florida Sparse Matrix Collection at http://www.cise.ufl.edu/research/sparse/matrices. It provides an easy way to select matrices from the collection based on their matrix -properties. As of September 2009, there were 2255 matrices ranging in size +properties. As of October 2010, there were 2328 matrices ranging in size from 1-by-2 (with a single nonzero entry) to a square matrix with nearly 28 million rows and 760 million entries. The total size of the collection in all -three formats exceeds 40 GB, and is constantly growing in size. +three formats exceeds 47 GB, and is constantly growing in size.2. Quick Start
+First, download the UFget archive (tar.gz or zip file), and uncompress it. +Open the UFget folder. + +If your web browswer requires an HTTP proxy, or if you wish to change the download location, first see the Customization section below, to configure UFgui before you run it. @@ -40,16 +44,23 @@
2. Quick Start
java -jar UFgui.jar +If the above command fails, then you need to install Java. +See +http://www.java.com/en/download/manual.jsp for details. + +You can skip reading this document by simply navigating the GUI itself. Most of the buttons, check boxes, lists, and table columns have a short "tool tip" which is visible if you hover your mouse over the item.
When the UFgui starts, it checks for any missing matrix icons and -downloads a new table if needed. +downloads a new table if needed. This might take a few minutes, so +just sit back and enjoy the slideshow. It will only happen occassionaly.
3. A Sparse Matrix Problem
-The UF Sparse Matrix Collection a simplified name. It is actually a collection +The UF Sparse Matrix Collection is a simplified name. It is actually a +collection of sparse matrix problems, not just sparse matrices. Each problem includes one primary sparse matrix and meta-data about that matrix. Some problems include additional matrices and vectors (sparse or dense) such as @@ -184,6 +195,11 @@5. Selecting Matrices
the list to add a single item to your highlighted choices. To clear your choices, click the Reset criteria button, described below. +When you select/deselect matrices, the boxes to the left of each matrix in +the Table of Matrices get checked/unchecked. You still see all the matrices +in the list, because you can then modify your selection by checking/unchecking +matrices one at a time in the Table itself. +
At the bottom of the Selection Criteria panel is a row of buttons: @@ -279,9 +295,13 @@
5. Selecting Matrices
- real: this box is checked if the matrix is real. It is complex otherwise. + Note that real matrices include any matrix that is not + complex. In particular, integer and binary matrices are + considered real in this search criterion.
- binary: this box is checked if the matrix is binary. - It is non-binary otherwise (there are no binary complex matrices). + It is non-binary otherwise (there are no binary complex matrices, + so any matrix that is binary is also marked as real).
- 2D/3D: this box is checked if the matrix comes
from a discretization of a 2D or 3D geometry.
@@ -449,7 +469,7 @@
7. Limitations and known issues
8. Copyright and License
-Copyright (c) 2009, Timothy A. Davis. All rights reserved. +Copyright (c) 2009-2010, Timothy A. Davis. All rights reserved.Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: @@ -479,6 +499,8 @@
8. Copyright and License
9. Version and Change-Log
-
+
- Version 1.0.4, October 27, 2010. Update to this help file only. +
- Version 1.0.3, October 9, 2009. Removed the Update button which when
clicked would download a fresh copy of matrices/UFstats.csv. UFgui
updates itself automatically every 30 days. If you want to force a
@@ -506,8 +528,12 @@
9. Version and Change-Log
10. For Further Help
Contact the author of this UFgui program, and the maintainer of the UF Sparse -Matrix Collection: Tim Davis (http://www.cise.ufl.edu/~davis, email -davis@cise.ufl.edu or DrTimothyAldenDavis@gmail.com). To print this document, +Matrix Collection: Tim Davis ( +http://www.cise.ufl.edu/~davis, email +davis@cise.ufl.edu +or +DrTimothyAldenDavis@gmail.com). +To print this document, open the file UFhelp.html in your favorite web browser.11. Acknowledgements
diff --git a/CXSparse/MATLAB/UFget/mat/UF_Index.mat b/CXSparse/MATLAB/UFget/mat/UF_Index.mat index d8678f6e6d..5f549d931c 100644 Binary files a/CXSparse/MATLAB/UFget/mat/UF_Index.mat and b/CXSparse/MATLAB/UFget/mat/UF_Index.mat differ diff --git a/CXSparse/MATLAB/UFget/matrices/UFstats.csv b/CXSparse/MATLAB/UFget/matrices/UFstats.csv index e8d55cb937..1c828a6057 100644 --- a/CXSparse/MATLAB/UFget/matrices/UFstats.csv +++ b/CXSparse/MATLAB/UFget/matrices/UFstats.csv @@ -1,5 +1,5 @@ -2272 -02-Nov-2009 14:30:41 +2379 +14-Jan-2011 15:51:12 HB,1138_bus,1138,1138,4054,1,0,0,1,1,1,power network problem HB,494_bus,494,494,1666,1,0,0,1,1,1,power network problem HB,662_bus,662,662,2474,1,0,0,1,1,1,power network problem @@ -2272,3 +2272,110 @@ Puri,ABACUS_shell_ud,23412,23412,218484,1,0,1,0,1,0.9753731955380578,model reduc Puri,ABACUS_shell_ld,23412,23412,218484,0,0,1,0,1,0.8850885826771654,model reduction problem Puri,ABACUS_shell_md,23412,23412,218484,0,0,1,0,1,0.8850885826771654,model reduction problem Puri,ABACUS_shell_hd,23412,23412,218484,0,0,1,0,1,0.8850885826771654,model reduction problem +Stevenson,LargeRegFile,2111154,801374,4944201,1,0,0,0,0,0,circuit simulation problem +Freescale,memchip,2707524,2707524,13343948,1,0,0,0,0.9058013004223939,0.4015860962358702,circuit simulation problem +Freescale,transient,178866,178866,961368,1,0,0,0,0.9996742983044353,0.243597788123496,circuit simulation problem +Freescale,circuit5M,5558326,5558326,59524291,1,0,0,0,0.9999998517635799,0.4198865531852634,circuit simulation problem +Freescale,circuit5M_dc,3523317,3523317,14865409,1,0,0,0,0.9072118240231114,0.7150571517141635,circuit simulation problem +Grund,poli3,16955,16955,37849,1,0,0,0,0.00507322676366421,0.0005743275581506652,economic problem +Grund,poli4,33833,33833,73249,1,0,0,0,0.003551857113862391,0.0002029632636492795,economic problem +Belcastro,human_gene1,22283,22283,24669643,1,0,0,0,1,1,undirected weighted graph +Belcastro,human_gene2,14340,14340,18068388,1,0,0,0,1,1,undirected weighted graph +Belcastro,mouse_gene,45101,45101,28967291,1,0,0,0,1,1,undirected weighted graph +Um,offshore,259789,259789,4242673,1,0,1,1,1,1,electromagnetics problem +SNAP,soc-Epinions1,75888,75888,508837,1,1,0,0,0.405226035056413,0.405226035056413,directed graph +SNAP,soc-LiveJournal1,4847571,4847571,68993773,1,1,0,0,0.7484193555024753,0.7484193555024753,directed graph +SNAP,soc-Slashdot0811,77360,77360,905468,1,1,0,0,0.8669352939826918,0.8669352939826918,directed graph +SNAP,soc-Slashdot0902,82168,82168,948464,1,1,0,0,0.8410650442849082,0.8410650442849082,directed graph +SNAP,wiki-Vote,8297,8297,103689,1,1,0,0,0.0564572905515532,0.0564572905515532,directed graph +SNAP,email-EuAll,265214,265214,420045,1,1,0,0,0.2600511748250413,0.2600511748250413,directed graph +SNAP,email-Enron,36692,36692,367662,1,1,0,0,1,1,directed graph +SNAP,wiki-Talk,2394385,2394385,5021410,1,1,0,0,0.1441208744157517,0.1441208744157517,directed graph +SNAP,cit-HepPh,34546,34546,421578,1,1,0,0,0.003117186276789061,0.003117186276789061,directed graph +SNAP,cit-HepTh,27770,27770,352807,1,1,0,0,0.002738343613933237,0.002738343613933237,directed graph +SNAP,cit-Patents,3774768,3774768,16518948,1,1,0,0,0,0,directed graph +SNAP,ca-AstroPh,18772,18772,396160,1,1,0,0,1,1,undirected graph +SNAP,ca-CondMat,23133,23133,186936,1,1,0,0,1,1,undirected graph +SNAP,ca-GrQc,5242,5242,28980,1,1,0,0,1,1,undirected graph +SNAP,ca-HepPh,12008,12008,237010,1,1,0,0,1,1,undirected graph +SNAP,ca-HepTh,9877,9877,51971,1,1,0,0,1,1,undirected graph +SNAP,web-BerkStan,685230,685230,7600595,1,1,0,0,0.2502764586193581,0.2502764586193581,directed graph +SNAP,web-Google,916428,916428,5105039,1,1,0,0,0.3067510355944391,0.3067510355944391,directed graph +SNAP,web-NotreDame,325729,325729,1497134,1,1,0,0,0.5165359238309862,0.5165359238309862,directed graph +SNAP,web-Stanford,281903,281903,2312497,1,1,0,0,0.2766368994208425,0.2766368994208425,directed graph +SNAP,amazon0302,262111,262111,1234877,1,1,0,0,0.5427018237443891,0.5427018237443891,directed graph +SNAP,amazon0312,400727,400727,3200440,1,1,0,0,0.5315337891039982,0.5315337891039982,directed graph +SNAP,amazon0505,410236,410236,3356824,1,1,0,0,0.5465803390347542,0.5465803390347542,directed graph +SNAP,amazon0601,403394,403394,3387388,1,1,0,0,0.5573497928197183,0.5573497928197183,directed graph +SNAP,p2p-Gnutella04,10879,10879,39994,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella05,8846,8846,31839,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella06,8717,8717,31525,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella08,6301,6301,20777,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella09,8114,8114,26013,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella24,26518,26518,65369,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella25,22687,22687,54705,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella30,36682,36682,88328,1,1,0,0,0,0,directed graph +SNAP,p2p-Gnutella31,62586,62586,147892,1,1,0,0,0,0,directed graph +SNAP,roadNet-CA,1971281,1971281,5533214,1,1,0,0,1,1,undirected graph +SNAP,roadNet-PA,1090920,1090920,3083796,1,1,0,0,1,1,undirected graph +SNAP,roadNet-TX,1393383,1393383,3843320,1,1,0,0,1,1,undirected graph +SNAP,as-735,7716,7716,26467,1,1,0,0,1,1,undirected graph sequence +SNAP,as-Skitter,1696415,1696415,22190596,1,1,0,0,1,1,undirected graph +SNAP,as-caida,31379,31379,106762,1,0,0,0,1,0.07869841329311927,directed weighted graph sequence +SNAP,Oregon-1,11492,11492,46818,1,1,0,0,1,1,undirected graph sequence +SNAP,Oregon-2,11806,11806,65460,1,1,0,0,1,1,undirected graph sequence +SNAP,soc-sign-epinions,131828,131828,841372,1,0,0,0,0.3082520316984202,0.3018224331855771,directed weighted graph +SNAP,soc-sign-Slashdot081106,77357,77357,516575,1,0,0,0,0.1859207278710739,0.1786110438948846,directed weighted graph +SNAP,soc-sign-Slashdot090216,81871,81871,545671,1,0,0,0,0.175926519826049,0.168574104176326,directed weighted graph +SNAP,soc-sign-Slashdot090221,82144,82144,549202,1,0,0,0,0.1774246998372184,0.1703271291801559,directed weighted graph +Dziekonski,gsm_106857,589446,589446,21758924,1,0,1,0,1,1,electromagnetics problem +Gleich,flickr,820878,820878,9837214,1,1,0,0,0.653017002578169,0.653017002578169,directed graph +Gleich,minnesota,2642,2642,6606,1,1,0,0,1,1,undirected graph +Gleich,usroads-48,126146,126146,323900,1,1,0,0,1,1,undirected graph +Gleich,usroads,129164,129164,330870,1,1,0,0,1,1,undirected graph +Fluorem,DK01R,903,903,11766,1,0,1,0,0.9614287029365737,0,computational fluid dynamics problem +Fluorem,GT01R,7980,7980,430909,1,0,1,0,0.8811455350661695,9.457852263618717e-06,computational fluid dynamics problem +Fluorem,PR02R,161070,161070,8185136,1,0,1,0,0.9476492341912441,0.0001697393815055858,computational fluid dynamics problem +Fluorem,RM07R,381689,381689,37464962,1,0,1,0,0.9261667922354103,4.260681089287885e-06,computational fluid dynamics problem +Rommes,ww_36_pmec_36,66,66,1194,1,0,1,0,0.3339191564147627,0,eigenvalue/model reduction problem +Rommes,ww_vref_6405,13251,13251,48737,1,0,1,0,0.707540090603374,0.5856980072816208,eigenvalue/model reduction problem +Rommes,xingo_afonso_itaipu,13250,13250,48735,1,0,1,0,0.7075597554196776,0.5857142857142857,eigenvalue/model reduction problem +Rommes,mimo8x8_system,13309,13309,48872,1,0,1,0,0.7060291752177048,0.5844472793832159,eigenvalue/model reduction problem +Rommes,mimo28x28_system,13251,13251,48737,1,0,1,0,0.707540090603374,0.5856980072816208,eigenvalue/model reduction problem +Rommes,mimo46x46_system,13250,13250,48735,1,0,1,0,0.7075597554196776,0.5857142857142857,eigenvalue/model reduction problem +Rommes,juba40k,40337,40337,144945,1,0,1,0,0.6934462083788981,0.5830536596402579,eigenvalue/model reduction problem +Rommes,bauru5727,40366,40366,145019,1,0,1,0,0.6931521565005984,0.5828064190876452,eigenvalue/model reduction problem +Rommes,zeros_nopss_13k,13296,13296,48827,1,0,1,0,0.7083056110002218,0.5842204479929031,eigenvalue/model reduction problem +Rommes,descriptor_xingo6u,20738,20738,73916,1,0,1,0,0.6848202865775669,0.5745268504272713,eigenvalue/model reduction problem +Rommes,nopss_11k,11685,11685,44941,1,0,1,0,0.7502077151335311,0.6233827893175075,eigenvalue/model reduction problem +Rommes,xingo3012,20944,20944,74386,1,0,1,0,0.681002526603102,0.5712336093539642,eigenvalue/model reduction problem +Rommes,bips98_606,7135,7135,34738,1,0,1,0,0.9086859688195991,0.7583878152166104,eigenvalue/model reduction problem +Rommes,bips98_1142,9735,9735,40983,1,0,1,0,0.8020552507215579,0.6694154587839767,eigenvalue/model reduction problem +Rommes,bips98_1450,11305,11305,44678,1,0,1,0,0.7593684272891549,0.6277809047012471,eigenvalue/model reduction problem +Rommes,bips07_1693,13275,13275,49044,1,0,1,0,0.706761938919083,0.5845449278560985,eigenvalue/model reduction problem +Rommes,bips07_1998,15066,15066,62198,1,0,1,0,0.7978660841804587,0.6687809795849786,eigenvalue/model reduction problem +Rommes,bips07_2476,16861,16861,66498,1,0,1,0,0.7574573297176583,0.6349098739830914,eigenvalue/model reduction problem +Rommes,bips07_3078,21128,21128,75729,1,0,1,0,0.6861363964744979,0.5750975292587777,eigenvalue/model reduction problem +Rommes,M10PI_n1,528,528,1317,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,M20PI_n1,1028,1028,2547,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,M40PI_n1,2028,2028,5007,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,M80PI_n1,4028,4028,9927,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,M10PI_n,682,682,1633,1,0,1,0,1,0.1132075471698113,eigenvalue/model reduction problem +Rommes,M20PI_n,1182,1182,2881,1,0,1,0,1,0.04173622704507512,eigenvalue/model reduction problem +Rommes,M40PI_n,2182,2182,5341,1,0,1,0,1,0.02274795268425842,eigenvalue/model reduction problem +Rommes,M80PI_n,4182,4182,10261,1,0,1,0,1,0.01191043353978085,eigenvalue/model reduction problem +Rommes,S10PI_n1,528,528,1317,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,S20PI_n1,1028,1028,2547,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,S40PI_n1,2028,2028,5007,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,S80PI_n1,4028,4028,9927,1,0,1,0,1,0,eigenvalue/model reduction problem +Rommes,S10PI_n,682,682,1633,1,0,1,0,1,0.1132075471698113,eigenvalue/model reduction problem +Rommes,S20PI_n,1182,1182,2881,1,0,1,0,1,0.04173622704507512,eigenvalue/model reduction problem +Rommes,S40PI_n,2182,2182,5341,1,0,1,0,1,0.02274795268425842,eigenvalue/model reduction problem +Rommes,S80PI_n,4182,4182,10261,1,0,1,0,1,0.01191043353978085,eigenvalue/model reduction problem +Williams,pdb1HYS,36417,36417,4344765,1,0,0,1,1,1,weighted undirected graph +Williams,consph,83334,83334,6010480,1,0,1,1,1,1,2D/3D problem +Williams,cant,62451,62451,4007383,1,0,1,1,1,1,2D/3D problem +Williams,mac_econ_fwd500,206500,206500,1273389,1,0,0,0,0.05995735832675801,0.005599217836274841,economic problem +Williams,mc2depi,525825,525825,2100225,1,0,1,0,0,0,2D/3D problem +Williams,cop20k_A,121192,121192,2624331,1,0,1,0,1,1,2D/3D problem +Williams,webbase-1M,1000005,1000005,3105536,1,0,0,0,0.09552554676231316,0.01588957844838191,weighted directed graph diff --git a/CXSparse/Makefile b/CXSparse/Makefile index 34794467d0..76ed252772 100644 --- a/CXSparse/Makefile +++ b/CXSparse/Makefile @@ -1,6 +1,6 @@ # CXSparse Makefile -VERSION = 2.2.4 +VERSION = 2.2.5 default: C @@ -38,11 +38,11 @@ distclean: purge # install CSparse install: $(CP) Lib/libcxsparse.a $(INSTALL_LIB)/libcxsparse.$(VERSION).a - ( cd $(INSTALL_LIB) ; ln -s libcxsparse.$(VERSION).a libcxsparse.a ) + ( cd $(INSTALL_LIB) ; ln -sf libcxsparse.$(VERSION).a libcxsparse.a ) $(CP) Include/cs.h $(INSTALL_INCLUDE) # uninstall CSparse uninstall: - $(RM) $(INSTALL_LIB)/libcxsparse.a + $(RM) $(INSTALL_LIB)/libcxsparse*.a $(RM) $(INSTALL_INCLUDE)/cs.h diff --git a/CXSparse/README.txt b/CXSparse/README.txt index 32f59e5ae3..65e2a5898e 100644 --- a/CXSparse/README.txt +++ b/CXSparse/README.txt @@ -1,5 +1,5 @@ -CXSparse: a Concise Sparse Matrix package - Extended. -Version 2.2.3, Copyright (c) 2006-2007, Timothy A. Davis. +CXSparse: Copyright (c) 2006-2011, Timothy A. Davis. + Derived from CSparse. Conversion originally by David Bateman, Motorola, and then modified by Tim Davis. ANSI C99 is required, with support for the _Complex data type. diff --git a/CXSparse/Source/cs_util.c b/CXSparse/Source/cs_util.c index 2a42121174..5a6294200c 100644 --- a/CXSparse/Source/cs_util.c +++ b/CXSparse/Source/cs_util.c @@ -35,7 +35,7 @@ cs *cs_spfree (cs *A) cs_free (A->p) ; cs_free (A->i) ; cs_free (A->x) ; - return (cs_free (A)) ; /* free the cs struct and return NULL */ + return ((cs *) cs_free (A)) ; /* free the cs struct and return NULL */ } /* free a numeric factorization */ @@ -46,7 +46,7 @@ csn *cs_nfree (csn *N) cs_spfree (N->U) ; cs_free (N->pinv) ; cs_free (N->B) ; - return (cs_free (N)) ; /* free the csn struct and return NULL */ + return ((csn *) cs_free (N)) ; /* free the csn struct and return NULL */ } /* free a symbolic factorization */ @@ -58,7 +58,7 @@ css *cs_sfree (css *S) cs_free (S->parent) ; cs_free (S->cp) ; cs_free (S->leftmost) ; - return (cs_free (S)) ; /* free the css struct and return NULL */ + return ((css *) cs_free (S)) ; /* free the css struct and return NULL */ } /* allocate a cs_dmperm or cs_scc result */ @@ -82,7 +82,7 @@ csd *cs_dfree (csd *D) cs_free (D->q) ; cs_free (D->r) ; cs_free (D->s) ; - return (cs_free (D)) ; + return ((csd *) cs_free (D)) ; /* free the csd struct and return NULL */ } /* free workspace and return a sparse matrix result */ @@ -98,7 +98,7 @@ CS_INT *cs_idone (CS_INT *p, cs *C, void *w, CS_INT ok) { cs_spfree (C) ; /* free temporary matrix */ cs_free (w) ; /* free workspace */ - return (ok ? p : cs_free (p)) ; /* return result if OK, else free it */ + return (ok ? p : (CS_INT *) cs_free (p)) ; /* return result, or free it */ } /* free workspace and return a numeric factorization (Cholesky, LU, or QR) */ diff --git a/CXSparse/Tcov/Makefile b/CXSparse/Tcov/Makefile index 04609af611..b4e649fa03 100644 --- a/CXSparse/Tcov/Makefile +++ b/CXSparse/Tcov/Makefile @@ -316,3 +316,4 @@ distclean: clean rm -f cs_idemo cs_ldemo rm -f cs_*.c rm -f cs*_di.c cs*_dl.c cs*_ci.c cs*_cl.c + rm -rf *.dSYM diff --git a/CXSparse_newfiles.tar.gz b/CXSparse_newfiles.tar.gz index 0402627eb9..353ab5c55c 100644 Binary files a/CXSparse_newfiles.tar.gz and b/CXSparse_newfiles.tar.gz differ diff --git a/CXSparse_newfiles/Makefile b/CXSparse_newfiles/Makefile index 34794467d0..76ed252772 100644 --- a/CXSparse_newfiles/Makefile +++ b/CXSparse_newfiles/Makefile @@ -1,6 +1,6 @@ # CXSparse Makefile -VERSION = 2.2.4 +VERSION = 2.2.5 default: C @@ -38,11 +38,11 @@ distclean: purge # install CSparse install: $(CP) Lib/libcxsparse.a $(INSTALL_LIB)/libcxsparse.$(VERSION).a - ( cd $(INSTALL_LIB) ; ln -s libcxsparse.$(VERSION).a libcxsparse.a ) + ( cd $(INSTALL_LIB) ; ln -sf libcxsparse.$(VERSION).a libcxsparse.a ) $(CP) Include/cs.h $(INSTALL_INCLUDE) # uninstall CSparse uninstall: - $(RM) $(INSTALL_LIB)/libcxsparse.a + $(RM) $(INSTALL_LIB)/libcxsparse*.a $(RM) $(INSTALL_INCLUDE)/cs.h diff --git a/CXSparse_newfiles/README.txt b/CXSparse_newfiles/README.txt index 32f59e5ae3..65e2a5898e 100644 --- a/CXSparse_newfiles/README.txt +++ b/CXSparse_newfiles/README.txt @@ -1,5 +1,5 @@ -CXSparse: a Concise Sparse Matrix package - Extended. -Version 2.2.3, Copyright (c) 2006-2007, Timothy A. Davis. +CXSparse: Copyright (c) 2006-2011, Timothy A. Davis. + Derived from CSparse. Conversion originally by David Bateman, Motorola, and then modified by Tim Davis. ANSI C99 is required, with support for the _Complex data type. diff --git a/CXSparse_newfiles/Tcov/Makefile b/CXSparse_newfiles/Tcov/Makefile index 04609af611..b4e649fa03 100644 --- a/CXSparse_newfiles/Tcov/Makefile +++ b/CXSparse_newfiles/Tcov/Makefile @@ -316,3 +316,4 @@ distclean: clean rm -f cs_idemo cs_ldemo rm -f cs_*.c rm -f cs*_di.c cs*_dl.c cs*_ci.c cs*_cl.c + rm -rf *.dSYM diff --git a/Contents.m b/Contents.m index ed867580f1..0e82e531ce 100644 --- a/Contents.m +++ b/Contents.m @@ -129,14 +129,15 @@ % MATLAB_Tools various simple m-files % SSMULT sparse matrix times sparse matrix % LINFACTOR solve Ax=b using LU or CHOL +% UFcollection for managing the UF Sparse Matrix Collection +% RBio for reading/writing Rutherford/Boeing sparse matrices % %------------------------------------------------------------------------------- % % For help on compiling SuiteSparse or the demos, testing functions, etc., -% please see the help for each individual package. UFcollection and RBio -% are two additional toolboxes, for managing the UF Sparse Matrix Collection. +% please see the help for each individual package. % -% Copyright 2008, Timothy A. Davis +% Copyright 2011, Timothy A. Davis % http://www.cise.ufl.edu/research/sparse help SuiteSparse diff --git a/KLU/Demo/kludemo.out b/KLU/Demo/kludemo.out index 8d78ac7a94..e69bd3c4c2 100644 --- a/KLU/Demo/kludemo.out +++ b/KLU/Demo/kludemo.out @@ -5,73 +5,73 @@ x [2] = 3 x [3] = 4 x [4] = 5 ./kludemo < ../Matrix/1c.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 1 nnz(A) 1 nnz(L+U+F) 1 resid 0 recip growth 1 condest 1 rcond 1 flops 0 peak memory usage: 492 bytes ./kludemo < ../Matrix/arrowc.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 100 nnz(A) 298 nnz(L+U+F) 298 resid 1.68007e-14 recip growth 0.019999 condest 295.99 rcond 0.019999 flops 297 peak memory usage: 32244 bytes ./kludemo < ../Matrix/arrow.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 100 nnz(A) 298 nnz(L+U+F) 298 resid 1.77636e-15 recip growth 0.0204082 condest 303 rcond 0.0204082 flops 297 peak memory usage: 20412 bytes ./kludemo < ../Matrix/impcol_a.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 207 nnz(A) 572 nnz(L+U+F) 615 resid 6.98492e-10 recip growth 0.00957447 condest 4.35093e+07 rcond 4.5277e-05 flops 259 peak memory usage: 34276 bytes ./kludemo < ../Matrix/w156.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 156 nnz(A) 362 nnz(L+U+F) 396 resid 6.23816e-08 recip growth 0.00889922 condest 1.79787e+09 rcond 0.000124785 flops 175 peak memory usage: 39516 bytes ./kludemo < ../Matrix/ctina.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 11 nnz(A) 36 nnz(L+U+F) 45 resid 4.44089e-16 recip growth 1 condest 56 rcond 1 flops 73 peak memory usage: 4268 bytes ./kluldemo < ../Matrix/1c.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 1 nnz(A) 1 nnz(L+U+F) 1 resid 0 recip growth 1 condest 1 rcond 1 flops 0 peak memory usage: 600 bytes ./kluldemo < ../Matrix/arrowc.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 100 nnz(A) 298 nnz(L+U+F) 298 resid 1.68007e-14 recip growth 0.019999 condest 295.99 rcond 0.019999 flops 297 peak memory usage: 39000 bytes ./kluldemo < ../Matrix/arrow.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 100 nnz(A) 298 nnz(L+U+F) 298 resid 1.77636e-15 recip growth 0.0204082 condest 303 rcond 0.0204082 flops 297 peak memory usage: 29584 bytes ./kluldemo < ../Matrix/impcol_a.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 207 nnz(A) 572 nnz(L+U+F) 615 resid 6.98492e-10 recip growth 0.00957447 condest 4.35093e+07 rcond 4.5277e-05 flops 259 peak memory usage: 44800 bytes ./kluldemo < ../Matrix/w156.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 156 nnz(A) 362 nnz(L+U+F) 396 resid 6.23816e-08 recip growth 0.00889922 condest 1.79787e+09 rcond 0.000124785 flops 175 peak memory usage: 47480 bytes ./kluldemo < ../Matrix/ctina.mtx -KLU: Nov 30, 2009, version: 1.1.1 +KLU: Jan 25, 2011, version: 1.1.2 n 11 nnz(A) 36 nnz(L+U+F) 45 resid 4.44089e-16 recip growth 1 condest 56 rcond 1 flops 73 peak memory usage: 5144 bytes diff --git a/KLU/Doc/ChangeLog b/KLU/Doc/ChangeLog index db063e7199..e79855befc 100644 --- a/KLU/Doc/ChangeLog +++ b/KLU/Doc/ChangeLog @@ -1,3 +1,8 @@ +Jan 25, 2011: version 1.1.2 + + * minor fix to "make install" + * port of Tcov to Mac OSX 10.6.6 + Nov 30, 2009: version 1.1.1 * added "make install" and "make uninstall" diff --git a/KLU/Doc/KLU_UserGuide.bib b/KLU/Doc/KLU_UserGuide.bib index e69f199f20..123d88e424 100644 --- a/KLU/Doc/KLU_UserGuide.bib +++ b/KLU/Doc/KLU_UserGuide.bib @@ -86,8 +86,8 @@ @article{DavisGilbertLarimoreNg00_algo @article{ChenDavisHagerRajamanickam06, author={Chen, Y. and Davis, T. A. and Hager, W. W. and Rajamanickam, S.}, - title={Algorithm 8xx: {CHOLMOD}, supernodal sparse {Cholesky} factorization and update/downdate}, - journal=TOMS, note={(submitted)}, year={2006}} + title={Algorithm 887: {CHOLMOD}, supernodal sparse {Cholesky} factorization and update/downdate}, + journal=TOMS, year={2009}, volume={35}, number={3}} @article{KarypisKumar98e, author={Karypis, G. and Kumar, V.}, @@ -130,3 +130,23 @@ @techreport{Palamadai05 circuit simulation problems}, note={M.S. Thesis}, address={CISE Department, Univ. of Florida}} + +@article{DavisNatarajan10, + author = {Davis, Timothy A. and Palamadai Natarajan, Ekanathan}, + title = {Algorithm 907: {KLU}, A Direct Sparse Solver for Circuit Simulation Problems}, + journal = {ACM Trans. Math. Softw.}, + volume = {37}, + issue = {3}, + month = {September}, + year = {2010}, + issn = {0098-3500}, + pages = {36:1--36:17}, + articleno = {36}, + numpages = {17}, + url = {http://doi.acm.org/10.1145/1824801.1824814}, + doi = {http://doi.acm.org/10.1145/1824801.1824814}, + acmid = {1824814}, + publisher = {ACM}, + address = {New York, NY, USA}, + keywords = {LU factorization, circuit simulation, sparse matrices}, +} diff --git a/KLU/Doc/KLU_UserGuide.pdf b/KLU/Doc/KLU_UserGuide.pdf index 1d712ec463..303d873c52 100644 Binary files a/KLU/Doc/KLU_UserGuide.pdf and b/KLU/Doc/KLU_UserGuide.pdf differ diff --git a/KLU/Doc/KLU_UserGuide.tex b/KLU/Doc/KLU_UserGuide.tex index 3618a77772..527985be45 100644 --- a/KLU/Doc/KLU_UserGuide.tex +++ b/KLU/Doc/KLU_UserGuide.tex @@ -26,9 +26,9 @@ and Lawrence Berkeley National Laboratory (with funding from Stanford University and the SciDAC program). } -\and Eka Palamadai} +\and Eka Palamadai Natarajan} -\date{March 24, 2009} +\date{VERSION 1.1.2, Jan 25, 2011} \maketitle %------------------------------------------------------------------------------ @@ -40,6 +40,7 @@ for finding a fill-reducing ordering (variants of approximate minimum degree, and nested dissection), and a sparse left-looking LU factorization method to factorize each block. A MATLAB interface is included. +KLU appears as Collected Algorithm 907 of the ACM \cite{DavisNatarajan10}. \end{abstract} %------------------------------------------------------------------------------ @@ -51,7 +52,7 @@ \section{License and Copyright} %------------------------------------------------------------------------------ -KLU Version 1.1.0, Copyright\copyright 2004-2009 University of Florida. +KLU, Copyright\copyright 2004-2011 University of Florida. All Rights Reserved. KLU is available under alternate licenses; contact T. Davis for details. @@ -233,12 +234,17 @@ \section{Overview} For details of the permutation to block triangular form, left-looking sparse LU factorization, and approximate minimum degree, refer to \cite{Davis06book}. Concise versions of these methods can be found in the CSparse package. KLU is -also the topic of a Master's thesis by Palamadai \cite{Palamadai05}; a copy of +also the topic of a Master's thesis by Palamadai Natarajan \cite{Palamadai05}; +a copy of the thesis can be found in the {\tt KLU/Doc} directory. It includes a description of an earlier version of KLU; some of the function names and parameter lists have changed in this version. The descriptions of the methods used still applies to the current version of KLU, however. +KLU appears as {\em Algorithm 907: KLU, a direct sparse solver for circuit +simulation problems}, ACM Transactions on Mathematical Software, vol 37, no 3, +2010. + %------------------------------------------------------------------------------ \section{Availability} %------------------------------------------------------------------------------ diff --git a/KLU/Include/klu.h b/KLU/Include/klu.h index 1f05df595f..92826285da 100644 --- a/KLU/Include/klu.h +++ b/KLU/Include/klu.h @@ -818,11 +818,11 @@ void *klu_l_realloc (size_t, size_t, size_t, void *, klu_l_common *) ; * #endif */ -#define KLU_DATE "Nov 30, 2009" +#define KLU_DATE "Jan 25, 2011" #define KLU_VERSION_CODE(main,sub) ((main) * 1000 + (sub)) #define KLU_MAIN_VERSION 1 #define KLU_SUB_VERSION 1 -#define KLU_SUBSUB_VERSION 1 +#define KLU_SUBSUB_VERSION 2 #define KLU_VERSION KLU_VERSION_CODE(KLU_MAIN_VERSION,KLU_SUB_VERSION) #ifdef __cplusplus diff --git a/KLU/Makefile b/KLU/Makefile index 6d023b1d92..2e94544f16 100644 --- a/KLU/Makefile +++ b/KLU/Makefile @@ -1,4 +1,4 @@ -VERSION = 1.1.1 +VERSION = 1.1.2 default: demos @@ -34,7 +34,7 @@ cov: library # install KLU install: $(CP) Lib/libklu.a $(INSTALL_LIB)/libklu.$(VERSION).a - ( cd $(INSTALL_LIB) ; ln -s libklu.$(VERSION).a libklu.a ) + ( cd $(INSTALL_LIB) ; ln -sf libklu.$(VERSION).a libklu.a ) $(CP) Include/klu.h $(INSTALL_INCLUDE) # uninstall KLU diff --git a/KLU/README.txt b/KLU/README.txt index f34791993b..cad82eb8a3 100644 --- a/KLU/README.txt +++ b/KLU/README.txt @@ -1,5 +1,5 @@ -KLU Version 1.1, March 24, 2009, by Timothy A. Davis and Ekanathan Palamadai. -Copyright (C) 2004-2009, University of Florida +KLU, Copyright (C) 2004-2011, University of Florida +by Timothy A. Davis and Ekanathan Palamadai. KLU is also available under other licenses; contact authors for details. http://www.cise.ufl.edu/research/sparse diff --git a/KLU/Tcov/Makefile b/KLU/Tcov/Makefile index 57fc2c63f7..23ec565d2c 100644 --- a/KLU/Tcov/Makefile +++ b/KLU/Tcov/Makefile @@ -26,7 +26,7 @@ I = -I../../UFconfig -I../../AMD/Include -I../../COLAMD/Include \ -I../../BTF/Include -I../../CHOLMOD/Include -I../../CAMD/Include \ -I../../CCOLAMD/Include -I../../metis-4.0/Lib -I../Include -I../User -PRETTY = grep -v "^\#" | indent -bl -nce -ss -bli0 -i4 -sob +PRETTY = grep -v "^\#" | indent -bl -nce -bli0 -i4 -sob all: purge libs klutest klultest - ./klultests > klultests.out diff --git a/KLU/Tcov/klutest.c b/KLU/Tcov/klutest.c index 723503d1eb..3a9c0fd15d 100644 --- a/KLU/Tcov/klutest.c +++ b/KLU/Tcov/klutest.c @@ -381,7 +381,7 @@ static Int *randperm (Int n, Int seed) static double do_1_solve (cholmod_sparse *A, cholmod_dense *B, cholmod_dense *Xknown, Int *Puser, Int *Quser, - KLU_common *Common, cholmod_common *ch, Int *nan) + KLU_common *Common, cholmod_common *ch, Int *isnan) { Int *Ai, *Ap ; double *Ax, *Bx, *Xknownx, *Xx, *Ax2, *Axx ; @@ -676,7 +676,7 @@ static double do_1_solve (cholmod_sparse *A, cholmod_dense *B, if (SCALAR_IS_NAN (relresid)) { - *nan = TRUE ; + *isnan = TRUE ; } else { @@ -716,7 +716,7 @@ static double do_1_solve (cholmod_sparse *A, cholmod_dense *B, if (SCALAR_IS_NAN (relerr)) { - *nan = TRUE ; + *isnan = TRUE ; } else { @@ -764,17 +764,17 @@ static double do_1_solve (cholmod_sparse *A, cholmod_dense *B, /* test KLU with many options */ static double do_solves (cholmod_sparse *A, cholmod_dense *B, cholmod_dense *X, - Int *Puser, Int *Quser, KLU_common *Common, cholmod_common *ch, Int *nan) + Int *Puser, Int *Quser, KLU_common *Common, cholmod_common *ch, Int *isnan) { double err, maxerr = 0 ; Int n = A->nrow, sflag ; - *nan = FALSE ; + *isnan = FALSE ; /* ---------------------------------------------------------------------- */ /* test KLU with the system A*X=B and default options */ /* ---------------------------------------------------------------------- */ - maxerr = do_1_solve (A, B, X, NULL, NULL, Common, ch, nan) ; + maxerr = do_1_solve (A, B, X, NULL, NULL, Common, ch, isnan) ; /* ---------------------------------------------------------------------- */ /* test with non-default options */ @@ -798,7 +798,7 @@ static double do_solves (cholmod_sparse *A, cholmod_dense *B, cholmod_dense *X, for (Common->ordering = 0 ; Common->ordering <= 3 ; Common->ordering++) { - err = do_1_solve (A, B, X, NULL, NULL, Common, ch, nan) ; + err = do_1_solve (A, B, X, NULL, NULL, Common, ch, isnan) ; maxerr = MAX (maxerr, err) ; } @@ -806,7 +806,7 @@ static double do_solves (cholmod_sparse *A, cholmod_dense *B, cholmod_dense *X, Common->ordering = 3 ; Common->user_data = &sflag ; sflag = 0 ; - err = do_1_solve (A, B, X, NULL, NULL, Common, ch, nan) ; + err = do_1_solve (A, B, X, NULL, NULL, Common, ch, isnan) ; maxerr = MAX (maxerr, err) ; Common->user_data = NULL ; @@ -814,7 +814,7 @@ static double do_solves (cholmod_sparse *A, cholmod_dense *B, cholmod_dense *X, Common->ordering = 2 ; if (n < 200) { - err = do_1_solve (A, B, X, Puser, Quser, Common, ch, nan) ; + err = do_1_solve (A, B, X, Puser, Quser, Common, ch, isnan); maxerr = MAX (maxerr, err) ; } } @@ -847,7 +847,7 @@ int main (void) Int *Ap, *Ai, *Puser, *Quser, *Gunk ; double *Ax, *Bx, *Xx, *A2x ; double one [2], zero [2], xsave, maxerr ; - Int n, i, j, nz, save, isreal, k, nan ; + Int n, i, j, nz, save, isreal, k, isnan ; KLU_symbolic *Symbolic, *Symbolic2 ; KLU_numeric *Numeric ; @@ -936,7 +936,7 @@ int main (void) /* ---------------------------------------------------------------------- */ test_memory_handler (&Common) ; - maxerr = do_solves (A, B, X, Puser, Quser, &Common, &ch, &nan) ; + maxerr = do_solves (A, B, X, Puser, Quser, &Common, &ch, &isnan) ; /* ---------------------------------------------------------------------- */ /* basic error checking */ @@ -1366,7 +1366,7 @@ int main (void) fprintf (stderr, " test FAILED") ; printf (" test FAILED") ; } - if (nan) + if (isnan) { fprintf (stderr, " *") ; printf (" *") ; diff --git a/LDL/Demo/ldlamd.out b/LDL/Demo/ldlamd.out index ac7fa11311..119ad85c93 100644 --- a/LDL/Demo/ldlamd.out +++ b/LDL/Demo/ldlamd.out @@ -5,7 +5,7 @@ Input matrix: name: Dense/0 n: 0 entries: 0 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -13,7 +13,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 0 @@ -35,7 +35,7 @@ Input matrix: name: Dense/0 n: 0 entries: 0 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -43,7 +43,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 0 @@ -65,7 +65,7 @@ Input matrix: name: Dense/1 n: 1 entries: 1 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -73,7 +73,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 1 nz, number of nonzeros in A: 1 @@ -115,7 +115,7 @@ Input matrix: name: Dense/1 n: 1 entries: 2 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -123,7 +123,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 1 nz, number of nonzeros in A: 1 @@ -165,7 +165,7 @@ Input matrix: name: Dense/2 n: 2 entries: 4 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -173,7 +173,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 2 nz, number of nonzeros in A: 4 @@ -215,7 +215,7 @@ Input matrix: name: Dense/2 n: 2 entries: 5 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -223,7 +223,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 2 nz, number of nonzeros in A: 4 @@ -265,7 +265,7 @@ Input matrix: name: Dense/3 n: 3 entries: 9 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -273,7 +273,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 3 nz, number of nonzeros in A: 9 @@ -315,7 +315,7 @@ Input matrix: name: Dense/3 n: 3 entries: 11 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -323,7 +323,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 3 nz, number of nonzeros in A: 9 @@ -365,7 +365,7 @@ Input matrix: name: HB/can_24 n: 24 entries: 160 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -373,7 +373,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 24 nz, number of nonzeros in A: 160 @@ -415,7 +415,7 @@ Input matrix: name: HB/can_24 n: 24 entries: 188 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -423,7 +423,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 24 nz, number of nonzeros in A: 160 @@ -465,7 +465,7 @@ Input matrix: name: FIDAP/ex5 n: 27 entries: 279 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -473,7 +473,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 27 nz, number of nonzeros in A: 279 @@ -515,7 +515,7 @@ Input matrix: name: FIDAP/ex5 n: 27 entries: 325 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -523,7 +523,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 27 nz, number of nonzeros in A: 279 @@ -565,7 +565,7 @@ Input matrix: name: HB/bcsstk01 n: 48 entries: 400 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -573,7 +573,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 48 nz, number of nonzeros in A: 400 @@ -615,7 +615,7 @@ Input matrix: name: HB/bcsstk01 n: 48 entries: 472 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -623,7 +623,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 48 nz, number of nonzeros in A: 400 @@ -665,7 +665,7 @@ Input matrix: name: HB/bcsstm01 n: 48 entries: 24 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -673,7 +673,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 48 nz, number of nonzeros in A: 24 @@ -715,7 +715,7 @@ Input matrix: name: HB/bcsstm01 n: 48 entries: 26 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -723,7 +723,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 48 nz, number of nonzeros in A: 24 @@ -765,7 +765,7 @@ Input matrix: name: Pothen/mesh1e1 n: 48 entries: 306 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -773,7 +773,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 48 nz, number of nonzeros in A: 306 @@ -815,7 +815,7 @@ Input matrix: name: Pothen/mesh1e1 n: 48 entries: 359 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -823,7 +823,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 48 nz, number of nonzeros in A: 306 @@ -865,7 +865,7 @@ Input matrix: name: Bai/bfwb62 n: 62 entries: 342 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -873,7 +873,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 62 nz, number of nonzeros in A: 342 @@ -915,7 +915,7 @@ Input matrix: name: Bai/bfwb62 n: 62 entries: 407 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -923,7 +923,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 62 nz, number of nonzeros in A: 342 @@ -965,7 +965,7 @@ Input matrix: name: HB/bcsstk02 n: 66 entries: 4356 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -973,7 +973,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 66 nz, number of nonzeros in A: 4356 @@ -1015,7 +1015,7 @@ Input matrix: name: HB/bcsstk02 n: 66 entries: 5175 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -1023,7 +1023,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 66 nz, number of nonzeros in A: 4356 @@ -1065,7 +1065,7 @@ Input matrix: name: HB/bcsstm02 n: 66 entries: 66 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -1073,7 +1073,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 66 nz, number of nonzeros in A: 66 @@ -1115,7 +1115,7 @@ Input matrix: name: HB/bcsstm02 n: 66 entries: 72 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -1123,7 +1123,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 4 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 66 nz, number of nonzeros in A: 66 diff --git a/LDL/Demo/ldllamd.out b/LDL/Demo/ldllamd.out index 8c1f63e5a1..d88efab65e 100644 --- a/LDL/Demo/ldllamd.out +++ b/LDL/Demo/ldllamd.out @@ -5,7 +5,7 @@ Input matrix: name: Dense/0 n: 0 entries: 0 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -13,7 +13,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 0 @@ -35,7 +35,7 @@ Input matrix: name: Dense/0 n: 0 entries: 0 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -43,7 +43,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 0 @@ -65,7 +65,7 @@ Input matrix: name: Dense/1 n: 1 entries: 1 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -73,7 +73,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 1 nz, number of nonzeros in A: 1 @@ -115,7 +115,7 @@ Input matrix: name: Dense/1 n: 1 entries: 2 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -123,7 +123,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 1 nz, number of nonzeros in A: 1 @@ -165,7 +165,7 @@ Input matrix: name: Dense/2 n: 2 entries: 4 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -173,7 +173,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 2 nz, number of nonzeros in A: 4 @@ -215,7 +215,7 @@ Input matrix: name: Dense/2 n: 2 entries: 5 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -223,7 +223,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 2 nz, number of nonzeros in A: 4 @@ -265,7 +265,7 @@ Input matrix: name: Dense/3 n: 3 entries: 9 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -273,7 +273,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 3 nz, number of nonzeros in A: 9 @@ -315,7 +315,7 @@ Input matrix: name: Dense/3 n: 3 entries: 11 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -323,7 +323,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 3 nz, number of nonzeros in A: 9 @@ -365,7 +365,7 @@ Input matrix: name: HB/can_24 n: 24 entries: 160 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -373,7 +373,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 24 nz, number of nonzeros in A: 160 @@ -415,7 +415,7 @@ Input matrix: name: HB/can_24 n: 24 entries: 188 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -423,7 +423,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 24 nz, number of nonzeros in A: 160 @@ -465,7 +465,7 @@ Input matrix: name: FIDAP/ex5 n: 27 entries: 279 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -473,7 +473,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 27 nz, number of nonzeros in A: 279 @@ -515,7 +515,7 @@ Input matrix: name: FIDAP/ex5 n: 27 entries: 325 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -523,7 +523,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 27 nz, number of nonzeros in A: 279 @@ -565,7 +565,7 @@ Input matrix: name: HB/bcsstk01 n: 48 entries: 400 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -573,7 +573,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 48 nz, number of nonzeros in A: 400 @@ -615,7 +615,7 @@ Input matrix: name: HB/bcsstk01 n: 48 entries: 472 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -623,7 +623,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 48 nz, number of nonzeros in A: 400 @@ -665,7 +665,7 @@ Input matrix: name: HB/bcsstm01 n: 48 entries: 24 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -673,7 +673,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 48 nz, number of nonzeros in A: 24 @@ -715,7 +715,7 @@ Input matrix: name: HB/bcsstm01 n: 48 entries: 26 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -723,7 +723,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 48 nz, number of nonzeros in A: 24 @@ -765,7 +765,7 @@ Input matrix: name: Pothen/mesh1e1 n: 48 entries: 306 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -773,7 +773,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 48 nz, number of nonzeros in A: 306 @@ -815,7 +815,7 @@ Input matrix: name: Pothen/mesh1e1 n: 48 entries: 359 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -823,7 +823,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 48 nz, number of nonzeros in A: 306 @@ -865,7 +865,7 @@ Input matrix: name: Bai/bfwb62 n: 62 entries: 342 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -873,7 +873,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 62 nz, number of nonzeros in A: 342 @@ -915,7 +915,7 @@ Input matrix: name: Bai/bfwb62 n: 62 entries: 407 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -923,7 +923,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 62 nz, number of nonzeros in A: 342 @@ -965,7 +965,7 @@ Input matrix: name: HB/bcsstk02 n: 66 entries: 4356 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -973,7 +973,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 66 nz, number of nonzeros in A: 4356 @@ -1015,7 +1015,7 @@ Input matrix: name: HB/bcsstk02 n: 66 entries: 5175 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -1023,7 +1023,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 66 nz, number of nonzeros in A: 4356 @@ -1065,7 +1065,7 @@ Input matrix: name: HB/bcsstm02 n: 66 entries: 66 -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -1073,7 +1073,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK n, dimension of A: 66 nz, number of nonzeros in A: 66 @@ -1115,7 +1115,7 @@ Input matrix: name: HB/bcsstm02 n: 66 entries: 72 (jumbled version) -------------------------------------------------------- -AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering +AMD version 2.2.2, Jan 25, 2011: approximate minimum degree ordering dense row parameter: 10 (rows with more than max (10 * sqrt (n), 16) entries are considered "dense", and placed last in output permutation) @@ -1123,7 +1123,7 @@ AMD version 2.2.1, Nov 30, 2009: approximate minimum degree ordering size of AMD integer: 8 -AMD version 2.2.1, Nov 30, 2009, results: +AMD version 2.2.2, Jan 25, 2011, results: status: OK, but jumbled n, dimension of A: 66 nz, number of nonzeros in A: 66 diff --git a/LDL/Demo/ldllmain.out b/LDL/Demo/ldllmain.out new file mode 100644 index 0000000000..f4f8474354 --- /dev/null +++ b/LDL/Demo/ldllmain.out @@ -0,0 +1,334 @@ + + +-------------------------------------------------------- +Input matrix: name: Dense/0 n: 0 entries: 0 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 0 +Factorize A=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 0 + + +-------------------------------------------------------- +Input matrix: name: Dense/0 n: 0 entries: 0 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 0 +Factorize A=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 0 + + +-------------------------------------------------------- +Input matrix: name: Dense/1 n: 1 entries: 1 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 0 +Factorize A=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 0 + + +-------------------------------------------------------- +Input matrix: name: Dense/1 n: 1 entries: 2 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 8.32667e-17 +Factorize A=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 8.32667e-17 + + +-------------------------------------------------------- +Input matrix: name: Dense/2 n: 2 entries: 4 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 1 Flop count: 3 +relative maxnorm of residual: 5.55112e-17 +Factorize A=LDL' and solve Ax=b +Nz in L: 1 Flop count: 3 +relative maxnorm of residual: 5.55112e-17 + + +-------------------------------------------------------- +Input matrix: name: Dense/2 n: 2 entries: 5 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 1 Flop count: 3 +relative maxnorm of residual: 1.11022e-16 +Factorize A=LDL' and solve Ax=b +Nz in L: 1 Flop count: 3 +relative maxnorm of residual: 1.11022e-16 + + +-------------------------------------------------------- +Input matrix: name: Dense/3 n: 3 entries: 9 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 3 Flop count: 11 +relative maxnorm of residual: 1.11022e-16 +Factorize A=LDL' and solve Ax=b +Nz in L: 3 Flop count: 11 +relative maxnorm of residual: 1.11022e-16 + + +-------------------------------------------------------- +Input matrix: name: Dense/3 n: 3 entries: 11 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 3 Flop count: 11 +relative maxnorm of residual: 1.38778e-16 +Factorize A=LDL' and solve Ax=b +Nz in L: 3 Flop count: 11 +relative maxnorm of residual: 1.38778e-16 + + +-------------------------------------------------------- +Input matrix: name: HB/can_24 n: 24 entries: 160 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 96 Flop count: 632 +Ax=b not solved since D(1,1) is zero. +Factorize A=LDL' and solve Ax=b +Nz in L: 146 Flop count: 1360 +Ax=b not solved since D(5,5) is zero. + + +-------------------------------------------------------- +Input matrix: name: HB/can_24 n: 24 entries: 188 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 96 Flop count: 632 +Ax=b not solved since D(1,1) is zero. +Factorize A=LDL' and solve Ax=b +Nz in L: 146 Flop count: 1360 +Ax=b not solved since D(5,5) is zero. + + +-------------------------------------------------------- +Input matrix: name: FIDAP/ex5 n: 27 entries: 279 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 126 Flop count: 954 +relative maxnorm of residual: 2.59625e-10 +Factorize A=LDL' and solve Ax=b +Nz in L: 276 Flop count: 4206 +relative maxnorm of residual: 2.72848e-10 + + +-------------------------------------------------------- +Input matrix: name: FIDAP/ex5 n: 27 entries: 325 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 126 Flop count: 954 +relative maxnorm of residual: 3.27418e-10 +Factorize A=LDL' and solve Ax=b +Nz in L: 276 Flop count: 4206 +relative maxnorm of residual: 2.32376e-10 + + +-------------------------------------------------------- +Input matrix: name: HB/bcsstk01 n: 48 entries: 400 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 441 Flop count: 5961 +relative maxnorm of residual: 2.38712e-13 +Factorize A=LDL' and solve Ax=b +Nz in L: 829 Flop count: 20103 +relative maxnorm of residual: 2.27374e-13 + + +-------------------------------------------------------- +Input matrix: name: HB/bcsstk01 n: 48 entries: 472 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 441 Flop count: 5961 +relative maxnorm of residual: 2.27374e-13 +Factorize A=LDL' and solve Ax=b +Nz in L: 829 Flop count: 20103 +relative maxnorm of residual: 3.83693e-13 + + +-------------------------------------------------------- +Input matrix: name: HB/bcsstm01 n: 48 entries: 24 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +Ax=b not solved since D(3,3) is zero. +Factorize A=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +Ax=b not solved since D(3,3) is zero. + + +-------------------------------------------------------- +Input matrix: name: HB/bcsstm01 n: 48 entries: 26 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +Ax=b not solved since D(3,3) is zero. +Factorize A=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +Ax=b not solved since D(3,3) is zero. + + +-------------------------------------------------------- +Input matrix: name: Pothen/mesh1e1 n: 48 entries: 306 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 288 Flop count: 2630 +relative maxnorm of residual: 5.96745e-16 +Factorize A=LDL' and solve Ax=b +Nz in L: 511 Flop count: 7383 +relative maxnorm of residual: 6.93889e-16 + + +-------------------------------------------------------- +Input matrix: name: Pothen/mesh1e1 n: 48 entries: 359 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 288 Flop count: 2630 +relative maxnorm of residual: 5.55112e-16 +Factorize A=LDL' and solve Ax=b +Nz in L: 511 Flop count: 7383 +relative maxnorm of residual: 5.55112e-16 + + +-------------------------------------------------------- +Input matrix: name: Bai/bfwb62 n: 62 entries: 342 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 226 Flop count: 1472 +relative maxnorm of residual: 5.55112e-16 +Factorize A=LDL' and solve Ax=b +Nz in L: 662 Flop count: 11350 +relative maxnorm of residual: 1.11022e-15 + + +-------------------------------------------------------- +Input matrix: name: Bai/bfwb62 n: 62 entries: 407 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 226 Flop count: 1472 +relative maxnorm of residual: 5.68434e-14 +Factorize A=LDL' and solve Ax=b +Nz in L: 662 Flop count: 11350 +relative maxnorm of residual: 2.15827e-12 + + +-------------------------------------------------------- +Input matrix: name: HB/bcsstk02 n: 66 entries: 4356 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 2145 Flop count: 97955 +relative maxnorm of residual: 7.50219e-13 +Factorize A=LDL' and solve Ax=b +Nz in L: 2145 Flop count: 97955 +relative maxnorm of residual: 7.50219e-13 + + +-------------------------------------------------------- +Input matrix: name: HB/bcsstk02 n: 66 entries: 5175 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 2145 Flop count: 97955 +relative maxnorm of residual: 8.59759e-13 +Factorize A=LDL' and solve Ax=b +Nz in L: 2145 Flop count: 97955 +relative maxnorm of residual: 8.59759e-13 + + +-------------------------------------------------------- +Input matrix: name: HB/bcsstm02 n: 66 entries: 66 +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 2.22045e-16 +Factorize A=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 2.22045e-16 + + +-------------------------------------------------------- +Input matrix: name: HB/bcsstm02 n: 66 entries: 72 (jumbled version) +-------------------------------------------------------- + +Factorize PAP'=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 2.22045e-16 +Factorize A=LDL' and solve Ax=b +Nz in L: 0 Flop count: 0 +relative maxnorm of residual: 2.22045e-16 + + +-------------------------------------------------------- +Input matrix: name: Dense/0 n: 0 entries: 0 (invalid matrix, Ap [0] = 99) +-------------------------------------------------------- + +ldlmain: invalid matrix and/or permutation + + +-------------------------------------------------------- +Input matrix: name: Dense/2 n: 2 entries: 4 (invalid perm, P[1]=99) +-------------------------------------------------------- + +ldlmain: invalid matrix and/or permutation + + +-------------------------------------------------------- +Input matrix: name: Dense/3 n: 3 entries: 9 (invalid perm) +-------------------------------------------------------- + +ldlmain: invalid matrix and/or permutation + + +-------------------------------------------------------- +Input matrix: name: Dense/3 n: 3 entries: 9 (invalid Ap) +-------------------------------------------------------- + +ldlmain: invalid matrix and/or permutation + + +-------------------------------------------------------- +Input matrix: name: Dense/3 n: 3 entries: 9 (invalid Ai) +-------------------------------------------------------- + +ldlmain: invalid matrix and/or permutation + + +-------------------------------------------------------- +Input matrix: name: Dense/3 n: 3 entries: 9 (invalid Ai) +-------------------------------------------------------- + +ldlmain: invalid matrix and/or permutation + +Largest residual during all tests: 3.27418e-10 + +ldlmain: all tests passed diff --git a/LDL/Demo/ldllsimple.out b/LDL/Demo/ldllsimple.out new file mode 100644 index 0000000000..acb48b50a3 --- /dev/null +++ b/LDL/Demo/ldllsimple.out @@ -0,0 +1,11 @@ +Nonzeros in L, excluding diagonal: 13 +x [0] = 0.1 +x [1] = 0.2 +x [2] = 0.3 +x [3] = 0.4 +x [4] = 0.5 +x [5] = 0.6 +x [6] = 0.7 +x [7] = 0.8 +x [8] = 0.9 +x [9] = 1 diff --git a/LDL/Doc/ChangeLog b/LDL/Doc/ChangeLog index efc71d87ec..d0a2e7b776 100644 --- a/LDL/Doc/ChangeLog +++ b/LDL/Doc/ChangeLog @@ -1,3 +1,7 @@ +Jan 25, 2011: version 2.0.3 + + * minor fix to "make install" + Nov 30, 2009: version 2.0.2 * added "make install" and "make uninstall" diff --git a/LDL/Doc/ldl_userguide.pdf b/LDL/Doc/ldl_userguide.pdf index bfe0615e34..b881c8a213 100644 Binary files a/LDL/Doc/ldl_userguide.pdf and b/LDL/Doc/ldl_userguide.pdf differ diff --git a/LDL/Doc/ldl_userguide.tex b/LDL/Doc/ldl_userguide.tex index 6208a5960d..da4400e0d6 100644 --- a/LDL/Doc/ldl_userguide.tex +++ b/LDL/Doc/ldl_userguide.tex @@ -26,7 +26,7 @@ University and the SciDAC program). }} -\date{Nov 1, 2007} +\date{VERSION 2.0.3, Jan 25, 2011} \maketitle diff --git a/LDL/Include/ldl.h b/LDL/Include/ldl.h index 61f45151b4..76cbd0840c 100644 --- a/LDL/Include/ldl.h +++ b/LDL/Include/ldl.h @@ -95,10 +95,10 @@ UF_long ldl_l_valid_matrix ( UF_long n, UF_long Ap [ ], UF_long Ai [ ]) ; /* === LDL version ========================================================== */ /* ========================================================================== */ -#define LDL_DATE "Nov 30, 2009" +#define LDL_DATE "Jan 25, 2011" #define LDL_VERSION_CODE(main,sub) ((main) * 1000 + (sub)) #define LDL_MAIN_VERSION 2 #define LDL_SUB_VERSION 0 -#define LDL_SUBSUB_VERSION 2 +#define LDL_SUBSUB_VERSION 3 #define LDL_VERSION LDL_VERSION_CODE(LDL_MAIN_VERSION,LDL_SUB_VERSION) diff --git a/LDL/Makefile b/LDL/Makefile index 23fc28f864..2ef295bf79 100644 --- a/LDL/Makefile +++ b/LDL/Makefile @@ -2,7 +2,7 @@ # LDL Makefile #------------------------------------------------------------------------------ -VERSION = 2.0.2 +VERSION = 2.0.3 default: demos @@ -59,7 +59,7 @@ mex: # install LDL install: $(CP) Lib/libldl.a $(INSTALL_LIB)/libldl.$(VERSION).a - ( cd $(INSTALL_LIB) ; ln -s libldl.$(VERSION).a libldl.a ) + ( cd $(INSTALL_LIB) ; ln -sf libldl.$(VERSION).a libldl.a ) $(CP) Include/ldl.h $(INSTALL_INCLUDE) # uninstall LDL diff --git a/LDL/README.txt b/LDL/README.txt index 7be8dd1f00..79512be479 100644 --- a/LDL/README.txt +++ b/LDL/README.txt @@ -1,4 +1,6 @@ -LDL Version 2.0: a sparse LDL' factorization and solve package. +LDL Copyright (c) 2005-2011 by Timothy A. Davis. All Rights Reserved. + +LDL: a sparse LDL' factorization and solve package. Written in C, with both a C and MATLAB mexFunction interface. These routines are not terrifically fast (they do not use dense matrix kernels), @@ -28,7 +30,6 @@ Quick start (for MATLAB users); -------------------------------------------------------------------------------- -LDL Copyright (c) 2005 by Timothy A. Davis. All Rights Reserved. LDL License: diff --git a/MATLAB_Tools/spok/Contents.m b/MATLAB_Tools/spok/Contents.m index 35740408eb..6605754841 100644 --- a/MATLAB_Tools/spok/Contents.m +++ b/MATLAB_Tools/spok/Contents.m @@ -22,4 +22,4 @@ % spok (A + 1i*A) % complex sparse % spok (A > .5) % logical sparse -% Copyright 2008, Tim Davis, University of Florida +% Copyright 2008-2011, Tim Davis, University of Florida diff --git a/MATLAB_Tools/spok/spok.m b/MATLAB_Tools/spok/spok.m index 392601abbb..c432eeec58 100644 --- a/MATLAB_Tools/spok/spok.m +++ b/MATLAB_Tools/spok/spok.m @@ -30,6 +30,6 @@ % % See also sparse. -% Copyright 2008, Tim Davis, University of Florida +% Copyright 2008-2011, Tim Davis, University of Florida error ('spok mexFunction not installed') ; diff --git a/MATLAB_Tools/spok/spok_install.m b/MATLAB_Tools/spok/spok_install.m index 651d8f5179..a63ecaa488 100644 --- a/MATLAB_Tools/spok/spok_install.m +++ b/MATLAB_Tools/spok/spok_install.m @@ -8,10 +8,15 @@ % % See also sparse, spok, spok_test -% Copyright 2008, Tim Davis, University of Florida +% Copyright 2008-2011, Tim Davis, University of Florida help spok_install -mex spok.c spok_mex.c +is64 = ~isempty (strfind (computer, '64')) ; +if (is64) + mex -largeArrayDims spok.c spok_mex.c +else + mex spok.c spok_mex.c +end addpath (pwd) fprintf ('Added the following directory to your path:\n') ; diff --git a/MATLAB_Tools/spok/spok_mex.c b/MATLAB_Tools/spok/spok_mex.c index 6c37b160d4..cc29031b8f 100644 --- a/MATLAB_Tools/spok/spok_mex.c +++ b/MATLAB_Tools/spok/spok_mex.c @@ -51,8 +51,8 @@ void mexFunction return ; } - Ap = mxGetJc (pargin [0]) ; - Ai = mxGetIr (pargin [0]) ; + Ap = (SPOK_INT *) mxGetJc (pargin [0]) ; + Ai = (SPOK_INT *) mxGetIr (pargin [0]) ; m = mxGetM (pargin [0]) ; n = mxGetN (pargin [0]) ; nzmax = mxGetNzmax (pargin [0]) ; diff --git a/MATLAB_Tools/spok/spok_test.m b/MATLAB_Tools/spok/spok_test.m index 96ab741239..d7791ce65c 100644 --- a/MATLAB_Tools/spok/spok_test.m +++ b/MATLAB_Tools/spok/spok_test.m @@ -6,7 +6,7 @@ % % See also sparse, spok, spok_install -% Copyright 2008, Tim Davis, University of Florida +% Copyright 2008-2011, Tim Davis, University of Florida % compile and install spok help spok ; diff --git a/RBio/Demo/RBdemo.out b/RBio/Demo/RBdemo.out new file mode 100644 index 0000000000..b69383359f --- /dev/null +++ b/RBio/Demo/RBdemo.out @@ -0,0 +1,19 @@ +./RBdemo < ../RBio/private/west0479.rua + +=========================================================== +title: [1U 8 STAGE COLUMN SECTION, ALL SECTIONS RIGOROUS ( CHEM. ENG. )] +status 0 +key [WEST0479] nrow 479 ncol 479 +mkind 0 skind 0 asize 3820 znz 22 +mtype from file: [RUA] +(A) RBok status 0 0 0 +(Z) RBok status 0 0 0 +mtype from examining matrix: [rua] +mtype after reading from file: [rua] +raw nrow 479 ncol 479 nnz 1910 nelnz 0 xsize 1910 +mtype from raw read from file: [rua] +(P) RBok status 0 0 0 +mtype of pattern: [pua] +mtype of tril(A): [rua] +(L) RBok status 0 0 22 +RBdemo OK diff --git a/RBio/Doc/ChangeLog b/RBio/Doc/ChangeLog index e060ddfe83..4d0b06fdd4 100644 --- a/RBio/Doc/ChangeLog +++ b/RBio/Doc/ChangeLog @@ -1,3 +1,7 @@ +Jan 25, 2011: version 2.0.1 + + * minor fix to "make install" + Nov 30, 2009: version 2.0.0 * rewrote the entire package in C. The advantage is that the code is diff --git a/RBio/Makefile b/RBio/Makefile index ec1f0513a1..185b9a0676 100644 --- a/RBio/Makefile +++ b/RBio/Makefile @@ -2,7 +2,7 @@ # RBio Makefile #------------------------------------------------------------------------------- -VERSION = 2.0.0 +VERSION = 2.0.1 default: all @@ -56,7 +56,7 @@ demos: # install RBio install: $(CP) Lib/librbio.a $(INSTALL_LIB)/librbio.$(VERSION).a - ( cd $(INSTALL_LIB) ; ln -s librbio.$(VERSION).a librbio.a ) + ( cd $(INSTALL_LIB) ; ln -sf librbio.$(VERSION).a librbio.a ) $(CP) Include/RBio.h $(INSTALL_INCLUDE) # uninstall RBio diff --git a/RBio/RBio/mywest b/RBio/RBio/mywest new file mode 100644 index 0000000000..587882cbb9 --- /dev/null +++ b/RBio/RBio/mywest @@ -0,0 +1,601 @@ +WEST0479 chemical eng. problem |west047 + 597 30 95 472 +rua 479 479 1887 0 +(16I5) (20I4) (4E18.10) + 1 4 7 10 12 14 19 24 27 30 33 37 40 43 46 48 + 50 55 58 61 64 67 70 73 76 78 80 82 84 86 88 90 + 92 96 102 104 106 109 113 116 119 122 125 128 130 132 137 142 + 145 148 151 155 158 161 164 166 168 173 176 179 182 185 188 191 + 193 195 197 199 201 203 205 207 209 213 219 221 223 226 230 233 + 236 238 239 246 248 249 256 260 295 303 305 306 320 334 348 352 + 381 386 388 389 400 411 422 430 439 443 446 447 459 471 483 491 + 507 513 516 517 528 539 550 559 568 577 586 595 604 613 622 631 + 640 649 658 667 676 685 694 703 712 721 730 739 748 757 766 774 + 782 790 798 806 814 817 820 823 826 829 832 835 838 841 844 847 + 850 853 856 858 861 864 867 871 875 879 886 896 898 900 907 912 + 918 921 923 925 927 929 931 933 935 937 939 942 945 948 951 954 + 957 960 963 966 969 972 975 979 983 987 992 1000 1002 1004 1010 1013 + 1017 1018 1020 1022 1024 1026 1028 1030 1032 1034 1036 1044 1047 1050 1053 1054 + 1060 1066 1072 1080 1086 1087 1090 1091 1097 1103 1109 1117 1120 1123 1126 1127 + 1133 1139 1145 1147 1149 1151 1158 1164 1165 1168 1169 1175 1181 1187 1195 1198 + 1201 1204 1205 1211 1217 1223 1225 1227 1229 1237 1243 1244 1247 1248 1254 1260 + 1266 1274 1277 1280 1283 1284 1290 1296 1302 1304 1306 1308 1316 1322 1323 1326 + 1327 1333 1339 1345 1353 1356 1359 1362 1363 1369 1375 1381 1383 1385 1387 1395 + 1401 1402 1405 1406 1412 1418 1424 1432 1435 1438 1441 1442 1448 1454 1460 1462 + 1464 1466 1474 1480 1481 1484 1485 1491 1497 1503 1505 1507 1509 1511 1513 1515 + 1517 1519 1521 1523 1525 1527 1529 1531 1533 1535 1537 1539 1541 1543 1545 1547 + 1549 1551 1553 1555 1557 1559 1561 1563 1565 1567 1569 1571 1573 1575 1578 1582 + 1586 1588 1590 1592 1594 1596 1598 1603 1608 1613 1615 1617 1619 1621 1623 1625 + 1628 1631 1634 1636 1638 1643 1648 1651 1654 1657 1661 1664 1667 1670 1672 1674 + 1679 1681 1683 1685 1688 1692 1696 1698 1700 1702 1704 1706 1708 1713 1718 1723 + 1725 1727 1729 1731 1733 1735 1738 1741 1744 1746 1748 1753 1758 1761 1764 1767 + 1771 1774 1777 1780 1782 1784 1789 1792 1795 1798 1801 1804 1807 1809 1811 1813 + 1815 1817 1819 1821 1823 1825 1829 1835 1837 1839 1842 1846 1849 1852 1854 1856 + 1858 1860 1862 1864 1866 1868 1870 1872 1874 1876 1878 1880 1882 1884 1886 1888 + 25 31 87 26 31 88 27 31 89 28 29 29 30 30 31 87 88 89 32 43 + 111 112 113 33 43 111 34 43 112 35 43 113 36 37 38 39 37 43 111 38 + 43 112 39 43 113 40 42 41 42 42 43 111 112 113 2 8 11 3 9 12 + 4 10 13 5 8 11 6 9 12 7 10 13 8 17 27 9 17 10 17 11 + 18 12 18 13 18 14 17 15 17 16 17 17 19 30 42 18 20 23 24 30 + 41 19 21 20 21 21 22 30 22 24 40 41 23 30 42 24 30 42 68 74 + 99 69 74 100 70 74 101 71 72 72 73 73 74 99 100 101 75 86 123 124 + 125 76 86 123 77 86 124 78 86 125 79 80 81 82 80 86 123 81 86 124 + 82 86 125 83 85 84 85 85 86 123 124 125 45 51 54 46 52 55 47 53 + 56 48 51 54 49 52 55 50 53 56 51 60 52 60 53 60 54 61 55 61 + 56 61 57 60 58 60 59 60 60 62 73 85 61 63 66 67 73 84 62 64 + 63 64 64 65 73 65 67 83 84 66 73 85 67 73 85 31 244 43 1 17 + 18 31 35 43 243 74 388 86 44 60 61 74 78 86 387 385 386 387 388 96 + 97 98 120 121 122 141 142 143 185 186 187 389 438 439 440 441 442 443 450 451 + 452 455 456 458 459 461 462 463 464 472 473 474 475 476 182 183 184 391 455 456 + 458 468 388 467 479 203 204 205 209 210 211 382 385 437 453 454 467 469 479 203 + 204 205 209 210 211 383 386 437 453 454 467 470 479 203 204 205 209 210 211 384 + 387 437 453 454 467 471 479 241 242 243 244 108 109 110 132 133 134 245 395 396 + 397 398 399 400 407 408 409 412 413 415 416 418 419 420 421 429 430 431 432 433 + 246 412 413 415 425 244 424 436 160 161 162 238 241 394 410 411 424 426 436 160 + 161 162 239 242 394 410 411 424 427 436 160 161 162 240 243 394 410 411 424 428 + 436 241 242 243 244 253 254 255 256 135 136 137 151 152 153 245 248 249 148 149 + 150 247 244 248 256 249 147 169 170 171 175 176 177 238 241 248 249 253 147 169 + 170 171 175 176 177 239 242 248 249 254 147 169 170 171 175 176 177 240 243 248 + 249 255 363 364 365 366 385 386 387 388 215 216 217 218 219 220 227 228 229 232 + 233 235 236 389 392 393 232 233 235 368 369 390 366 388 392 393 181 194 195 196 + 230 231 363 382 385 392 393 181 194 195 196 230 231 364 383 386 392 393 181 194 + 195 196 230 231 365 384 387 392 393 93 96 248 262 284 306 328 350 372 94 97 + 248 262 284 306 328 350 372 95 98 248 262 284 306 328 350 372 93 96 248 262 + 284 306 328 350 372 94 97 248 262 284 306 328 350 372 95 98 248 262 284 306 + 328 350 372 105 108 264 286 308 330 352 374 392 106 109 264 286 308 330 352 374 + 392 107 110 264 286 308 330 352 374 392 105 108 264 286 308 330 352 374 392 106 + 109 264 286 308 330 352 374 392 107 110 264 286 308 330 352 374 392 117 120 249 + 263 285 307 329 351 373 118 121 249 263 285 307 329 351 373 119 122 249 263 285 + 307 329 351 373 117 120 249 263 285 307 329 351 373 118 121 249 263 285 307 329 + 351 373 119 122 249 263 285 307 329 351 373 129 132 265 287 309 331 353 375 393 + 130 133 265 287 309 331 353 375 393 131 134 265 287 309 331 353 375 393 129 132 + 265 287 309 331 353 375 393 130 133 265 287 309 331 353 375 393 131 134 265 287 + 309 331 353 375 393 135 141 266 288 310 332 354 376 136 142 267 289 311 333 355 + 377 137 143 268 290 312 334 356 378 135 141 266 288 310 332 354 376 136 142 267 + 289 311 333 355 377 137 143 268 290 312 334 356 378 138 148 238 139 149 239 140 + 150 240 144 182 382 145 183 383 146 184 384 395 401 404 396 402 405 397 403 406 + 398 401 404 399 402 405 400 403 406 160 401 410 161 402 410 162 403 163 404 411 + 164 405 411 165 406 411 160 161 407 410 160 162 408 410 161 162 409 410 163 164 + 165 410 412 423 435 157 163 164 165 411 413 416 417 423 434 412 414 413 414 158 + 163 164 165 414 415 423 157 415 417 433 434 163 164 165 416 423 435 417 423 435 + 148 178 149 179 150 180 148 166 149 167 150 168 148 154 149 155 150 156 215 221 + 224 216 222 225 217 223 226 218 221 224 219 222 225 220 223 226 194 221 230 195 + 222 230 196 223 230 197 224 231 198 225 231 199 226 231 194 195 227 230 194 196 + 228 230 195 196 229 230 197 198 199 230 232 191 197 198 199 231 233 236 237 232 + 234 233 234 192 197 198 199 234 235 191 235 237 197 198 199 236 237 182 212 183 + 213 184 214 182 200 183 201 184 202 182 188 183 189 184 190 241 242 243 244 253 + 254 255 256 257 264 265 246 247 258 244 256 264 265 241 250 253 261 264 265 242 + 251 254 261 264 265 243 252 255 261 264 265 253 254 255 256 275 276 277 278 257 + 262 263 266 267 268 259 256 262 278 263 250 253 260 262 263 275 251 254 260 262 + 263 276 252 255 260 262 263 277 253 254 255 256 275 276 277 278 279 286 287 258 + 259 280 256 278 286 287 253 272 275 283 286 287 254 273 276 283 286 287 255 274 + 277 283 286 287 250 269 251 270 252 271 275 276 278 297 298 299 300 279 284 285 + 288 289 290 281 278 284 300 285 272 275 282 284 285 297 273 276 282 284 285 298 + 274 277 282 284 285 299 275 276 277 278 297 298 299 300 301 308 309 280 281 302 + 278 300 308 309 275 294 297 305 308 309 276 295 298 305 308 309 277 296 299 305 + 308 309 272 291 273 292 274 293 297 298 299 300 319 320 321 322 301 306 307 310 + 311 312 303 300 306 322 307 294 297 304 306 307 319 295 298 304 306 307 320 296 + 299 304 306 307 321 297 298 299 300 319 320 321 322 323 330 331 302 303 324 300 + 322 330 331 297 316 319 327 330 331 298 317 320 327 330 331 299 318 321 327 330 + 331 294 313 295 314 296 315 319 320 321 322 341 342 343 344 323 328 329 332 333 + 334 325 322 328 344 329 316 319 326 328 329 341 317 320 326 328 329 342 318 321 + 326 328 329 343 319 320 321 322 341 342 343 344 345 352 353 324 325 346 322 344 + 352 353 319 338 341 349 352 353 320 339 342 349 352 353 321 340 343 349 352 353 + 316 335 317 336 318 337 341 342 343 344 363 364 365 366 345 350 351 354 355 356 + 347 344 350 366 351 338 341 348 350 351 363 339 342 348 350 351 364 340 343 348 + 350 351 365 341 342 343 344 363 364 365 366 367 374 375 346 347 368 344 366 374 + 375 341 360 363 371 374 375 342 361 364 371 374 375 343 362 365 371 374 375 338 + 357 339 358 340 359 363 364 365 366 385 386 387 388 367 372 373 376 377 378 369 + 366 372 388 373 360 363 370 372 373 385 361 364 370 372 373 386 362 365 370 372 + 373 387 360 379 361 380 362 381 87 93 88 94 89 95 90 96 91 97 92 98 + 99 105 100 106 101 107 102 108 103 109 104 110 111 117 112 118 113 119 114 120 + 115 121 116 122 123 129 124 130 125 131 126 132 127 133 128 134 135 138 136 139 + 137 140 141 144 142 145 143 146 151 154 152 155 153 156 157 158 159 158 163 164 + 165 159 163 164 165 160 163 161 164 162 165 163 166 164 167 165 168 169 172 175 + 176 177 170 173 175 176 177 171 174 175 176 177 172 175 173 176 174 177 175 178 + 176 179 177 180 102 418 424 103 419 424 104 420 424 421 422 422 423 102 103 104 + 423 424 126 127 128 425 436 126 426 436 127 427 436 128 428 436 429 430 431 432 + 126 430 436 127 431 436 128 432 436 433 435 434 435 126 127 128 435 436 185 188 + 186 189 187 190 191 192 193 192 197 198 199 193 197 198 199 194 197 195 198 196 + 199 197 200 198 201 199 202 203 206 209 210 211 204 207 209 210 211 205 208 209 + 210 211 206 209 207 210 208 211 209 212 210 213 211 214 90 461 467 91 462 467 + 92 463 467 464 465 465 466 90 91 92 466 467 114 115 116 468 479 114 469 479 + 115 470 479 116 471 479 472 473 474 475 114 473 479 115 474 479 116 475 479 476 + 478 477 478 114 115 116 478 479 438 444 447 439 445 448 440 446 449 441 444 447 + 442 445 448 443 446 449 444 453 445 453 446 453 447 454 448 454 449 454 450 453 + 451 453 452 453 453 455 466 478 454 456 459 460 466 477 455 457 456 457 457 458 + 466 458 460 476 477 459 466 478 460 466 478 266 269 267 270 268 271 288 291 289 + 292 290 293 310 313 311 314 312 315 332 335 333 336 334 337 354 357 355 358 356 + 359 376 379 377 380 378 381 + 1.0000000000E+00 -3.7648000000E-02 -3.4424000000E-01 1.0000000000E+00 + -2.4523000000E-02 -3.7370900000E-01 1.0000000000E+00 -3.6613000000E-02 + -8.3693800000E-01 1.3000000000E+02 -2.4337670000E+00 1.0000000000E+00 + -1.6140910000E+00 1.6140910000E+00 -2.1873200000E-01 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.1383520000E+00 3.6694000000E-02 + 9.9316000000E-02 9.9316000000E-02 9.9316000000E-02 -5.0000000000E-01 + 1.6117000000E-02 8.7246000000E-02 -3.6119200000E-01 1.1643000000E-02 + 1.0504100000E-01 -3.2188800000E-01 1.0376000000E-02 1.4041700000E-01 + -4.3624200000E-01 -7.6804300000E-01 -1.4302800000E-01 -1.5938900000E-01 + 1.0000000000E+00 -4.8561000000E-02 -2.6286800000E-01 1.0000000000E+00 + -3.0926000000E-02 -2.7901300000E-01 1.0000000000E+00 -4.6124000000E-02 + -6.2418800000E-01 5.2824360000E+00 -6.1239200000E-01 2.8868200000E-01 + -2.1638200000E-01 1.3287740000E+00 -3.6946900000E-01 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 4.8176472000E+01 -1.0000000000E+00 + -3.3000000000E-05 8.3500000000E+01 -1.0000000000E+00 -4.1000000000E-05 + 1.7194119300E+02 -1.0000000000E+00 -8.5000000000E-05 9.6651382000E+01 + 2.5000000000E+00 3.3000000000E-05 1.6827059900E+02 2.5000000000E+00 + 4.1000000000E-05 3.4758718900E+02 2.5000000000E+00 8.5000000000E-05 + 1.0000000000E+00 -1.1069700000E-01 1.6052320000E+00 1.0000000000E+00 + -8.9809000000E-02 1.0000000000E+00 -1.5173700000E-01 1.0104550000E+00 + -5.0000000000E-01 1.0059780000E+00 -3.0000000000E-01 1.0028850000E+00 + -2.0000000000E-01 1.0000000000E+00 -1.8118500000E-01 1.0000000000E+00 + -2.4002400000E-01 1.0000000000E+00 -2.2654800000E-01 1.0000000000E+00 + -1.0000000000E+00 1.5000000000E+00 5.0000000000E-01 1.0000000000E+00 + -3.1622000000E+05 -1.2323690430E+04 -1.0000000000E+00 -3.5226800781E+04 + 1.8449019531E+04 5.2983390000E+00 2.4530000000E-03 1.0808590090E+03 + -2.0502200000E-01 1.4419200000E-01 6.3059860000E+01 -1.1593000000E-01 + -1.8449019531E+04 8.4533400000E-01 -1.8449019531E+04 -1.5595580078E+04 + 1.0000000000E+00 2.0204900000E-01 -8.8547100000E-01 1.0000000000E-04 + 2.4483390000E+00 8.1611300000E-01 1.0000000000E+00 -2.2800000000E-04 + -2.6243900000E-01 1.0000000000E+00 -4.1900000000E-04 -3.2162000000E-01 + 1.0000000000E+00 -1.5770000000E-03 -7.2647600000E-01 3.0000000000E+02 + -2.8518910000E+00 1.0000000000E+00 -2.8701600000E-01 2.8701600000E-01 + -4.3410000000E-03 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.1383520000E+00 4.2003000000E-02 9.4148100000E-01 9.4148100000E-01 + 9.4148100000E-01 -3.2188800000E-01 1.1877000000E-02 1.3310950000E+00 + -3.6119200000E-01 1.3327000000E-02 9.9575300000E-01 -5.0000000000E-01 + 1.8449000000E-02 8.2705600000E-01 -1.0000000000E+00 -1.9240000000E+01 + -3.8030000000E+00 -9.4810000000E+00 1.0000000000E+00 -8.8800000000E-04 + -9.9474000000E-02 1.0000000000E+00 -1.3310000000E-03 -9.9474000000E-02 + 1.0000000000E+00 -2.2190000000E-03 -9.9474000000E-02 1.1776130000E+00 + -3.1089630000E+00 9.9194300000E-01 -2.6400560000E+00 3.1921120000E+00 + -4.4614000000E-02 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + 1.0986879700E+02 -1.0000000000E+00 -3.3000000000E-05 1.9138459800E+02 + -1.0000000000E+00 -4.1000000000E-05 3.9547961400E+02 -1.0000000000E+00 + -8.5000000000E-05 2.2173390200E+02 2.5000000000E+00 3.3000000000E-05 + 3.8700918600E+02 2.5000000000E+00 4.1000000000E-05 8.0081652800E+02 + 2.5000000000E+00 8.5000000000E-05 1.0000000000E+00 -9.1700000000E-03 + 1.0000000000E+00 -4.6441000000E-02 1.0000000000E+00 -4.8999200000E-01 + 1.0045300000E+00 -2.0000000000E-01 1.0025910000E+00 -3.0000000000E-01 + 1.0012500000E+00 -5.0000000000E-01 1.0000000000E+00 -3.7501000000E-02 + 1.0000000000E+00 -1.2414300000E-01 1.0000000000E+00 -2.9275300000E-01 + 1.0000000000E+00 -1.0000000000E+00 1.8537300000E-01 6.1791000000E-02 + 1.0000000000E+00 -3.1622000000E+05 -1.6364190430E+04 -1.0000000000E+00 + -4.6967822270E+03 1.3185400400E+02 2.2129129000E+01 9.5375300000E-01 + 2.4942900390E+03 -1.0215870000E+00 8.8782800000E-01 1.0400420000E+00 + -2.6400560000E+00 -1.3185400400E+02 8.0570000000E-03 -1.3185400400E+02 + -1.0624090000E+00 1.0000000000E+00 1.0164300000E-01 -6.1791000000E-02 + 7.6450000000E-03 2.3098949000E+01 7.6996490000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 -3.8502310000E+00 + -8.5000000000E-05 2.0159100000E+00 1.0000000000E+00 1.5961710000E+00 + -7.8975100000E-01 1.0000000000E+00 8.8175600000E-01 1.0000000000E+00 + 1.0000000000E+00 -2.7937600000E+00 -8.4000000000E-05 2.0156650000E+00 + 1.0000000000E+00 1.5939940000E+00 -9.6191500000E-01 2.4250000000E-03 + 3.0363000000E-02 6.7304500000E-01 1.4593600000E+00 1.6897000000E-02 + 3.4848000000E-02 -6.0080000000E-02 -5.5527200000E-01 -4.7704000000E-01 + -1.8582900000E-01 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -5.5188568000E+01 -9.5676857000E+01 -2.1583770800E+02 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 -1.8910000000E-03 -1.5310000000E-03 + -1.0140000000E-03 2.5000000000E+00 2.6063700000E+01 1.0000000000E+00 + -1.5000000000E+00 -9.6031810000E+00 -9.4541850000E+00 -9.4376810000E+00 + -4.8295719000E+01 9.0759200000E-01 -4.3759670000E+00 -8.9979920000E+00 + -8.8819580000E+00 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 -1.0000000000E+00 -2.6063700000E+01 + -1.0000000000E+00 1.0000000000E+00 -1.4593600000E+00 1.0000000000E+00 + 1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 5.6953098000E+01 + 7.0583200000E-01 1.0000000000E+00 -6.4913700000E-01 -3.3000000000E-05 + 5.3807500000E-01 1.0000000000E+00 8.2471000000E-02 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.1583840000E+01 7.0583200000E-01 1.0000000000E+00 + -1.0215420000E+00 -4.1000000000E-05 -3.6302500000E-01 1.0000000000E+00 + 6.9523000000E-02 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 3.2096300000E-01 + 7.0583200000E-01 1.0000000000E+00 -2.0490070000E+00 -8.4000000000E-05 + 9.1353600000E-01 1.0000000000E+00 8.3976400000E-01 5.5083500000E-01 + 4.4892200000E-01 2.4300000000E-04 4.0528300000E-01 -6.7277000000E-02 + -1.5700600000E-01 -9.7476000000E-02 -2.5060100000E-01 -2.8026200000E-01 + 1.0084900000E-01 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + -2.8640000000E-03 -2.3160000000E-03 -1.5310000000E-03 2.5000000000E+00 + 1.1587800000E+01 1.0000000000E+00 -1.1012240000E+00 -1.3644320000E+00 + -1.2188210000E+00 -1.2108590000E+00 -3.4621052000E+01 6.0313600000E-01 + -5.9963800000E-01 -1.4033920000E+00 -1.3703860000E+00 1.0000000000E+00 + 1.0000000000E+00 -1.0000000000E+00 -1.1587800000E+01 -1.0000000000E+00 + 1.0000000000E+00 -4.0528300000E-01 1.0000000000E+00 1.0000000000E+00 + -8.5938400000E-01 -7.7333900000E-01 -7.9517400000E-01 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 -1.6236590000E+00 -3.3000000000E-05 + 3.5720350000E+00 1.0000000000E+00 9.2435300000E-01 -1.1718210000E+00 + -1.2773520000E+00 -1.2505080000E+00 -1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 -2.4602890000E+00 -4.1000000000E-05 -2.1785330000E+00 + 1.0000000000E+00 6.5098400000E-01 -2.3280470000E+00 -2.4161550000E+00 + -2.5121660000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + -4.7536200000E+00 -8.5000000000E-05 6.1671400000E+00 1.0000000000E+00 + 3.3966910000E+00 4.3737000000E-02 5.7396500000E-01 1.5834900000E-01 + 1.8781300000E-01 -4.4344000000E-02 -5.8193800000E-01 -1.6054900000E-01 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -6.2680901000E+01 -1.2388999900E+02 -4.6435549900E+02 1.0000000000E+00 + -5.1647000000E-02 -3.2417600000E-01 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 -1.8781300000E-01 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 9.7738750000E+00 7.7605000000E-01 7.2528310000E+00 + 7.8041000000E-01 -7.8683100000E-01 1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 6.0698200000E-01 7.7605000000E-01 -2.3895080000E+00 + 6.8494000000E-01 -7.8683100000E-01 1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.1890000000E-03 7.7605000000E-01 1.1558830000E+01 + 2.2026440000E+00 -7.8683100000E-01 -1.7051500000E-01 -4.3429600000E-01 + -2.6674200000E-01 -2.6093020000E+00 1.9564500000E-01 4.9830200000E-01 + 3.0605400000E-01 4.2239600000E-01 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + -1.8910000000E-03 -1.5310000000E-03 -1.0140000000E-03 2.5000000000E+00 + 1.6672409000E+01 1.0000000000E+00 -1.5000000000E+00 -1.0000000000E+00 + -4.6212100000E-01 -1.2349600000E-01 -1.0000000000E+00 -1.6672409000E+01 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 + 2.6093020000E+00 -4.2239600000E-01 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 -5.8697600000E-01 -5.0651600000E-01 -5.1399200000E-01 + -1.0366850000E+00 -3.3000000000E-05 -8.7155300000E-01 -1.0000000000E+00 + 1.0000000000E+00 2.6399800000E+00 5.5742100000E-01 1.0000000000E+00 + -8.0016400000E-01 -8.3640900000E-01 -8.0810300000E-01 -1.6376950000E+00 + -4.1000000000E-05 -8.7155300000E-01 -1.0000000000E+00 1.0000000000E+00 + -1.7736780000E+00 4.0754200000E-01 1.0000000000E+00 -1.5893890000E+00 + -1.5818110000E+00 -1.6231180000E+00 -3.2056860000E+00 -8.4000000000E-05 + -8.7155300000E-01 -1.0000000000E+00 1.0000000000E+00 4.4676090000E+00 + 2.2475290000E+00 -1.0000000000E+00 -1.0000000000E+00 -4.0421600000E-01 + -1.8187800000E-01 -1.5634600000E-01 -1.5457000000E-01 -1.5217800000E-01 + -9.1906000000E-02 -7.8710000000E-03 -1.0000000000E+00 -1.0000000000E+00 + -1.8091710000E+00 -3.1896550000E+00 -3.3490190000E+00 -3.3586440000E+00 + -3.3085850000E+00 -1.9678700000E+00 -1.1333600000E-01 -1.0000000000E+00 + -1.0000000000E+00 -2.4566020000E+00 -4.1011280000E+00 -4.2908920000E+00 + -4.3026030000E+00 -4.2541470000E+00 -2.9523890000E+00 -1.1573650000E+00 + -8.1220000000E-03 -1.6897000000E-02 -4.5390000000E-03 -2.1710000000E-03 + -1.8740000000E-03 -1.8530000000E-03 -1.8250000000E-03 -1.1070000000E-03 + -1.0500000000E-04 -1.6750000000E-02 -3.4848000000E-02 -4.1897000000E-02 + -7.8507000000E-02 -8.2794000000E-02 -8.3055000000E-02 -8.1826000000E-02 + -4.8866000000E-02 -3.1320000000E-03 -2.8878000000E-02 -6.0080000000E-02 + -9.8082000000E-02 -1.7402800000E-01 -1.8288500000E-01 -1.8343700000E-01 + -1.8139100000E-01 -1.2639700000E-01 -5.5137000000E-02 -1.0000000000E+00 + -1.0000000000E+00 -7.6362300000E-01 -5.4132100000E-01 -5.2907500000E-01 + -5.2832600000E-01 -5.2965300000E-01 -5.9901100000E-01 -5.7467700000E-01 + -1.0000000000E+00 -1.0000000000E+00 -1.4679790000E+00 -1.2898060000E+00 + -1.2799920000E+00 -1.2793850000E+00 -1.2801740000E+00 -1.3601430000E+00 + -7.1491900000E-01 -1.0000000000E+00 -1.0000000000E+00 -4.3710000000E-03 + -3.9540000000E-03 -3.9320000000E-03 -3.9480000000E-03 -4.7490000000E-03 + -7.2455000000E-02 -1.6023640000E+00 -1.1229300000E-01 -6.7277000000E-02 + -5.4601000000E-02 -3.8877000000E-02 -3.8009000000E-02 -3.7959000000E-02 + -3.8209000000E-02 -4.8086000000E-02 -5.8179000000E-02 -2.6206300000E-01 + -1.5700600000E-01 -2.4496100000E-01 -2.1618100000E-01 -2.1459700000E-01 + -2.1451900000E-01 -2.1552300000E-01 -2.5481000000E-01 -1.6890700000E-01 + -1.6270000000E-01 -9.7476000000E-02 -4.5300000000E-04 -4.1100000000E-04 + -4.0900000000E-04 -4.1100000000E-04 -4.9600000000E-04 -8.4270000000E-03 + -2.3503500000E-01 -1.0000000000E+00 -1.0000000000E+00 -6.9703000000E-02 + -1.9243000000E-02 -1.5838000000E-02 -1.5618000000E-02 -1.5583000000E-02 + -1.4719000000E-02 -5.7600000000E-03 -1.0000000000E+00 -1.0000000000E+00 + -7.4171400000E-01 -8.0234900000E-01 -8.0658500000E-01 -8.0682900000E-01 + -8.0550300000E-01 -7.4927000000E-01 -1.9719700000E-01 -1.0000000000E+00 + -1.0000000000E+00 -5.1276000000E-01 -5.2522500000E-01 -5.2614100000E-01 + -5.2622400000E-01 -5.2730300000E-01 -5.7231900000E-01 -1.0252420000E+00 + -4.6575900000E-01 -9.6900300000E-01 -4.4885000000E-02 -1.3170000000E-02 + -1.0887000000E-02 -1.0739000000E-02 -1.0717000000E-02 -1.0163000000E-02 + -4.4260000000E-03 -5.7440200000E-01 -1.1950330000E+00 -5.8903400000E-01 + -6.7721700000E-01 -6.8380200000E-01 -6.8420500000E-01 -6.8315600000E-01 + -6.3804400000E-01 -1.8686200000E-01 -2.2922800000E-01 -4.7690500000E-01 + -1.6250700000E-01 -1.7691400000E-01 -1.7800600000E-01 -1.7808600000E-01 + -1.7847000000E-01 -1.9449300000E-01 -3.8770300000E-01 -1.0000000000E+00 + -1.0000000000E+00 -3.4272800000E-01 -2.9113800000E-01 -2.8769400000E-01 + -2.8748900000E-01 -2.8823200000E-01 -3.0269900000E-01 -1.7507900000E-01 + -1.0000000000E+00 -1.0000000000E+00 -1.0623340000E+00 -1.1185060000E+00 + -1.1222530000E+00 -1.1225120000E+00 -1.1232850000E+00 -1.1082340000E+00 + -3.5118600000E-01 -1.0000000000E+00 -1.0000000000E+00 -2.4000000000E-03 + -2.6020000000E-03 -2.6160000000E-03 -2.6280000000E-03 -3.1620000000E-03 + -4.4794000000E-02 -5.9723100000E-01 -2.0182530000E+00 -1.2091660000E+00 + -4.4044900000E-01 -3.7580300000E-01 -3.7146400000E-01 -3.7124000000E-01 + -3.7371100000E-01 -4.3672900000E-01 -3.1856300000E-01 -2.5626880000E+00 + -1.5353450000E+00 -1.7335130000E+00 -1.8332450000E+00 -1.8399150000E+00 + -1.8405410000E+00 -1.8492860000E+00 -2.0302660000E+00 -8.1137000000E-01 + -9.2554300000E-01 -5.5450700000E-01 -1.4140000000E-03 -1.5400000000E-03 + -1.5490000000E-03 -1.5560000000E-03 -1.8800000000E-03 -2.9637000000E-02 + -4.9833900000E-01 -1.4338920000E+00 -2.1577040000E+00 -1.5239710000E+00 + -1.5307080000E+00 -1.5311480000E+00 -1.5313160000E+00 -1.5375330000E+00 + -1.7109280000E+00 -9.4320600000E-01 -1.4193260000E+00 -1.0024600000E+00 + -1.0068910000E+00 -1.0071810000E+00 -1.0072910000E+00 -1.0113810000E+00 + -1.1254390000E+00 -5.9645300000E-01 -8.9753500000E-01 -6.3392300000E-01 + -6.3672500000E-01 -6.3690800000E-01 -6.3697800000E-01 -6.3956400000E-01 + -7.1169100000E-01 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 5.5083500000E-01 + -1.0000000000E+00 1.6474950000E+00 7.3959700000E-01 -1.0000000000E+00 + 8.4135162400E+02 2.0404500000E-01 -1.0000000000E+00 1.0000000000E+00 + 1.9564500000E-01 -1.0000000000E+00 1.0000000000E+00 4.9830200000E-01 + -1.0000000000E+00 3.1156220000E+00 9.5354800000E-01 6.6224922000E+01 + -1.0000000000E+00 -3.3000000000E-05 1.1506230200E+02 -1.0000000000E+00 + -4.1000000000E-05 2.3733869900E+02 -1.0000000000E+00 -8.5000000000E-05 + 1.3324499500E+02 2.5000000000E+00 3.3000000000E-05 2.3226390100E+02 + 2.5000000000E+00 4.1000000000E-05 4.8018209800E+02 2.5000000000E+00 + 8.5000000000E-05 -4.7337900000E-01 1.0000000000E+00 -2.1168800000E-01 + -5.7343200000E-01 1.0000000000E+00 -3.1667100000E-01 -6.0900000000E-04 + 1.0000000000E+00 -4.5750039060E+03 1.0075620000E+00 -5.5083500000E-01 + -3.6814160160E+03 1.0043240000E+00 -4.4892200000E-01 -1.7878179930E+03 + 1.0020870000E+00 -2.4300000000E-04 -5.2605600000E-01 -4.2598200000E-01 + 1.0000000000E+00 -4.7048800000E-01 -5.6500000000E-04 -4.3801000000E-01 + 1.0000000000E+00 -5.0500000000E-04 -5.8600000000E-04 -5.6138100000E-01 + 1.0000000000E+00 -6.4700000000E-04 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 1.0000000000E+00 -1.0000000000E+00 6.1444000000E-02 + 2.0481000000E-02 -1.0000000000E+00 4.1240600590E+03 4.1240600590E+03 + 4.1240600590E+03 1.0000000000E+00 -3.1622000000E+05 -2.0034240234E+04 + -1.0000000000E+00 -2.6256569820E+03 2.2212899800E+02 1.8737270000E+01 + 9.4488200000E-01 1.4943669430E+03 -1.0207800000E+00 -9.9186000000E-01 + -1.7922340000E+01 -1.7922340000E+01 -1.7922340000E+01 8.6282900000E-01 + 1.0497190000E+00 -7.3415000000E-01 8.1400000000E-03 -2.2212899800E+02 + 8.1400000000E-03 -2.2212899800E+02 -1.8080990000E+00 -1.0913280000E+00 + -1.6293970000E+00 -1.4448530000E+00 1.0000000000E+00 4.7364000000E-02 + -2.7898000000E-02 4.5390000000E-03 7.5792400000E+00 2.5264130000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.0266460000E+00 -1.0000000000E+00 + 1.0737240000E+00 -1.0000000000E+00 1.2081470000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 9.9149727000E+01 -1.0000000000E+00 + -3.3000000000E-05 1.7263960300E+02 -1.0000000000E+00 -4.1000000000E-05 + 3.5663980100E+02 -1.0000000000E+00 -8.5000000000E-05 2.0000079300E+02 + 2.5000000000E+00 3.3000000000E-05 3.4900329600E+02 2.5000000000E+00 + 4.1000000000E-05 7.2206781000E+02 2.5000000000E+00 8.5000000000E-05 + -1.1483900000E-01 1.0000000000E+00 -1.1646000000E-02 -4.1678400000E-01 + 1.0000000000E+00 -1.7006200000E-01 -4.9676100000E-01 1.0000000000E+00 + -2.4368900000E-01 -5.2768618160E+03 1.0050250000E+00 -1.9564500000E-01 + -4.2415908200E+03 1.0028740000E+00 -4.9830200000E-01 -2.0582939450E+03 + 1.0013870000E+00 -3.0605400000E-01 -3.9872300000E-01 -9.9097000000E-02 + 1.0000000000E+00 -8.0870000000E-02 -4.8643800000E-01 -1.0056000000E-01 + 1.0000000000E+00 -9.8660000000E-02 -4.8411900000E-01 -4.0267900000E-01 + 1.0000000000E+00 -3.9507300000E-01 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + 3.2976230470E+03 3.2976230470E+03 3.2976230470E+03 1.0000000000E+00 + -3.1622000000E+05 -1.8966660156E+04 -1.0000000000E+00 2.2669870000E+01 + 9.5486000000E-01 2.2487060550E+03 -1.0206550000E+00 -9.9229500000E-01 + -2.1898569000E+01 -2.1898569000E+01 -2.1898569000E+01 8.9001000000E-01 + 1.0392050000E+00 7.7050000000E-03 -1.4613620000E+02 7.7050000000E-03 + -6.5890500000E-01 -1.1064970000E+00 -1.0009090000E+00 1.0000000000E+00 + 6.8960000000E-03 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 + -1.0000000000E+00 1.0226760000E+00 -1.0000000000E+00 1.0914180000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 -1.9969600000E-01 + -7.8596200000E-01 -6.4100000000E-04 4.0690400000E-01 2.0247000000E-01 + 7.9688000000E-01 6.5000000000E-04 -2.1665440000E+00 -1.0000000000E+00 + -3.0001500000E-01 -4.0746100000E-01 -1.0000000000E+00 -1.0000000000E+00 + 1.0000000000E+00 4.0690400000E-01 -2.1665440000E+00 1.0000000000E+00 + 1.0000000000E+00 -9.8629900000E-01 -1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 3.5018580000E+00 1.2418840000E+00 -9.8629900000E-01 + -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 -2.1495590000E+00 + 9.3602400000E-01 -9.8629900000E-01 -1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 6.0259860000E+00 4.0868380000E+00 1.1955000000E-02 + 6.1475000000E-01 1.6059600000E-01 5.9918100000E-01 -1.1950000000E-02 + -6.1446100000E-01 -1.6052000000E-01 -1.0000000000E+00 1.0000000000E+00 + -9.3351000000E-02 -3.4681800000E-01 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 1.0000000000E+00 -5.9918100000E-01 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.3333420000E+01 7.8730100000E-01 + 1.0000000000E+00 1.2120260000E+01 7.7025100000E-01 -7.8693100000E-01 + 1.0205510000E+00 7.8730100000E-01 1.0000000000E+00 -3.9843990000E+00 + 6.8134200000E-01 -7.8693100000E-01 3.1870000000E-03 7.8730100000E-01 + 1.0000000000E+00 1.9252159000E+01 2.2369070000E+00 -7.8693100000E-01 + -1.7008100000E-01 -8.2969200000E-01 -6.9700000000E-04 2.5673630000E+00 + 1.7000100000E-01 8.2930200000E-01 6.9700000000E-04 -4.2847870000E+00 + -1.0000000000E+00 -2.5546900000E-01 -4.1224600000E-01 -1.0000000000E+00 + -1.0000000000E+00 1.0000000000E+00 2.5673630000E+00 -4.2847870000E+00 + 1.0000000000E+00 1.0000000000E+00 -1.0004710000E+00 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 2.9555290000E+00 1.2544140000E+00 + -1.0004710000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + -1.8159690000E+00 9.4521200000E-01 -1.0004710000E+00 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 5.0849970000E+00 4.1364790000E+00 + 2.0247000000E-01 -1.0000000000E+00 7.9688000000E-01 -1.0000000000E+00 + 2.0398200000E-01 -1.0000000000E+00 9.8180000000E-03 6.1664200000E-01 + 9.5579400000E-01 -9.8180000000E-03 -6.1661000000E-01 -1.6051500000E-01 + -1.0000000000E+00 1.0000000000E+00 -9.8218000000E-02 -3.4856400000E-01 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 + -9.5579400000E-01 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.3626710000E+01 7.8698300000E-01 1.0000000000E+00 1.2682330000E+01 + 7.6942900000E-01 -7.8694200000E-01 1.0583890000E+00 7.8698300000E-01 + 1.0000000000E+00 -4.1684890000E+00 6.8102800000E-01 -7.8694200000E-01 + 3.4160000000E-03 7.8698300000E-01 1.0000000000E+00 2.0139959000E+01 + 2.2394150000E+00 -7.8694200000E-01 -1.6787000000E-01 -8.3148300000E-01 + -7.0000000000E-04 4.3289930000E+00 1.6786100000E-01 8.3143900000E-01 + 7.0000000000E-04 -4.5292090000E+00 -1.0000000000E+00 -2.5301500000E-01 + -4.1256200000E-01 -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 + 4.3289930000E+00 -4.5292090000E+00 1.0000000000E+00 1.0000000000E+00 + -1.0000520000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 2.9254360000E+00 1.2552490000E+00 -1.0000520000E+00 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 -1.7975930000E+00 9.4582400000E-01 + -1.0000520000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 5.0331660000E+00 4.1397830000E+00 1.7000100000E-01 -1.0000000000E+00 + 8.2930200000E-01 -1.0000000000E+00 2.0397300000E-01 -1.0000000000E+00 + 9.6800000000E-03 6.1671100000E-01 1.6051800000E-01 9.9729800000E-01 + -9.6800000000E-03 -6.1673100000E-01 -1.6052300000E-01 -1.0000000000E+00 + 1.0000000000E+00 -9.8529000000E-02 -3.4867100000E-01 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 -9.9729800000E-01 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 1.3646010000E+01 + 7.8690900000E-01 1.0000000000E+00 1.2716190000E+01 7.6935900000E-01 + -7.8693500000E-01 1.0608970000E+00 7.8690900000E-01 1.0000000000E+00 + -4.1795750000E+00 6.8099400000E-01 -7.8693500000E-01 3.4310000000E-03 + 7.8690900000E-01 1.0000000000E+00 2.0193411000E+01 2.2395320000E+00 + -7.8693500000E-01 -1.6772300000E-01 -8.3154100000E-01 -7.0300000000E-04 + 4.5319110000E+00 1.6772900000E-01 8.3156800000E-01 7.0300000000E-04 + -4.5441880000E+00 -1.0000000000E+00 -2.5288900000E-01 -4.1262900000E-01 + -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 4.5319110000E+00 + -4.5441880000E+00 1.0000000000E+00 1.0000000000E+00 -9.9996700000E-01 + -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 2.9235700000E+00 + 1.2552940000E+00 -9.9996700000E-01 -1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 -1.7964900000E+00 9.4585200000E-01 -9.9996700000E-01 + -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 5.0299350000E+00 + 4.1401400000E+00 1.6786100000E-01 -1.0000000000E+00 8.3143900000E-01 + -1.0000000000E+00 2.0398600000E-01 -1.0000000000E+00 9.6470000000E-03 + 6.1499700000E-01 1.6066400000E-01 1.0122750000E+00 -9.6630000000E-03 + -6.1599900000E-01 -1.6092600000E-01 -1.0000000000E+00 1.0000000000E+00 + -9.7740000000E-02 -3.4838900000E-01 -1.0000000000E+00 -1.0000000000E+00 + -1.0000000000E+00 1.0000000000E+00 -1.0122750000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.3653370000E+01 7.8530800000E-01 + 1.0000000000E+00 1.2536040000E+01 7.6861400000E-01 -7.8658800000E-01 + 1.0618540000E+00 7.8530800000E-01 1.0000000000E+00 -4.1203450000E+00 + 6.8034500000E-01 -7.8658800000E-01 3.4370000000E-03 7.8530800000E-01 + 1.0000000000E+00 1.9907200000E+01 2.2374860000E+00 -7.8658800000E-01 + -1.6769600000E-01 -8.2983400000E-01 -8.4400000000E-04 4.5319130000E+00 + 1.6796900000E-01 8.3118600000E-01 8.4500000000E-04 -4.4769570000E+00 + -1.0000000000E+00 -2.5422800000E-01 -4.1467900000E-01 -1.0000000000E+00 + -1.0000000000E+00 1.0000000000E+00 4.5319120000E+00 -4.4769570000E+00 + 1.0000000000E+00 1.0000000000E+00 -9.9837300000E-01 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 2.9258000000E+00 1.2548710000E+00 + -9.9837300000E-01 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + -1.7994740000E+00 9.4529600000E-01 -9.9837300000E-01 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 5.0329780000E+00 4.1465320000E+00 + 1.6772900000E-01 -1.0000000000E+00 8.3156800000E-01 -1.0000000000E+00 + 2.0458700000E-01 -1.0000000000E+00 8.9580000000E-03 5.6239200000E-01 + 1.7143200000E-01 1.5349870000E+00 -9.3680000000E-03 -5.8815700000E-01 + -1.7928500000E-01 -1.0000000000E+00 1.0000000000E+00 -7.6425000000E-02 + -3.3630700000E-01 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + 1.0000000000E+00 -1.5349870000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.3927700000E+01 7.4278100000E-01 1.0000000000E+00 + 7.7124140000E+00 7.3754000000E-01 -7.7681100000E-01 1.0977920000E+00 + 7.4278100000E-01 1.0000000000E+00 -2.5345340000E+00 6.5320800000E-01 + -7.7681100000E-01 3.6610000000E-03 7.4278100000E-01 1.0000000000E+00 + 1.2244490000E+01 2.1513860000E+00 -7.7681100000E-01 -1.6726400000E-01 + -7.7757800000E-01 -1.1351000000E-02 3.9419710000E+00 1.7492700000E-01 + 8.1320200000E-01 1.1871000000E-02 -2.5680820000E+00 -1.0000000000E+00 + -3.1132300000E-01 -4.5572700000E-01 -1.0000000000E+00 -1.0000000000E+00 + 1.0000000000E+00 3.9419710000E+00 -2.5680820000E+00 1.0000000000E+00 + 1.0000000000E+00 -9.5619400000E-01 -1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 3.1494540000E+00 1.2129980000E+00 -9.5619400000E-01 + -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 -1.9859190000E+00 + 9.0706800000E-01 -9.5619400000E-01 -1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 5.3936870000E+00 4.2274630000E+00 1.6796900000E-01 + -1.0000000000E+00 8.3118600000E-01 -1.0000000000E+00 2.3079700000E-01 + -1.0000000000E+00 4.9560000000E-03 2.0925100000E-01 4.3415700000E-01 + 6.1773840000E+00 -5.6860000000E-03 -2.4009000000E-01 -4.9814100000E-01 + -1.0000000000E+00 1.0000000000E+00 -5.1900000000E-02 -2.2819900000E-01 + -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 + -6.1773840000E+00 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 2.2884661000E+01 6.4836400000E-01 1.0000000000E+00 1.0434500000E+00 + 4.2175900000E-01 -7.4391800000E-01 2.5197030000E+00 6.4836400000E-01 + 1.0000000000E+00 -3.4146600000E-01 3.7989000000E-01 -7.4391800000E-01 + 1.7728000000E-02 6.4836400000E-01 1.0000000000E+00 1.6460520000E+00 + 1.3054760000E+00 -7.4391800000E-01 1.7492700000E-01 -1.0000000000E+00 + 8.1320200000E-01 -1.0000000000E+00 6.6961900000E-01 -1.0000000000E+00 + 6.3180580000E+00 1.0081210000E+00 2.1016520000E+00 9.8325000000E-01 + 1.0216340000E+01 9.7112200000E-01 4.7715330000E+00 1.0168970000E+00 + 1.5445530000E+00 9.6515200000E-01 7.4032580000E+00 9.3992000000E-01 + 2.7676480100E+02 8.8770700000E-01 1.9219039900E+02 1.2620630000E+00 + 4.6529739400E+02 8.3730000000E-01 4.0263531500E+02 9.3272300000E-01 + 2.4395159900E+02 1.1570060000E+00 6.9442572000E+02 9.0252400000E-01 + 2.1471690000E+00 7.3310400000E-01 1.8303540000E+00 7.7070700000E-01 + 5.4194960000E+00 9.1068000000E-01 1.5934600000E+00 4.4472800000E-01 + 1.5193590000E+00 5.2296000000E-01 5.9272710000E+00 8.1417100000E-01 + 8.6013230000E+00 5.8171500000E-01 6.1974850000E+00 5.3220700000E-01 + 3.7670330000E+01 1.1683300000E+00 1.0955350000E+01 7.4939900000E-01 + 8.2864270000E+00 7.1973800000E-01 3.5092930000E+01 1.1008490000E+00 + 4.3389200000E-01 2.2797130000E+00 1.1375700000E-01 6.0698200000E-01 + 4.0354700000E-01 8.0050000000E-03 1.1577040000E+00 4.0422280000E+00 + 4.1932600000E-01 2.4496110000E+00 1.0246500000E-01 3.6475200000E-01 + 1.7257449300E+02 1.4917610000E+01 1.6158450300E+02 1.2093800000E+01 + 1.4531449900E+02 5.7400960000E+00 4.5020000000E-03 9.5263600000E-01 + -1.0000000000E+00 9.4488200000E-01 1.9882059000E+01 1.5998700000E+01 + 7.7694990000E+00 1.0081400000E+00 9.8828979000E+01 1.4755580100E+02 + 1.3084370400E+02 1.0000000000E+00 1.8011980000E+00 1.0000000000E+00 + 2.1962210000E+00 1.0000000000E+00 2.0607380000E+00 1.9882059000E+01 + 9.7404500000E-01 1.5998700000E+01 9.3133800000E-01 7.7694990000E+00 + 8.2771400000E-01 -1.0000000000E+00 -1.0000000000E+00 5.6358000000E-02 + 5.6358000000E-02 5.6358000000E-02 -1.0000000000E+00 -1.0000000000E+00 + 7.3959700000E-01 7.3959700000E-01 7.3959700000E-01 -1.0000000000E+00 + -1.0000000000E+00 2.0404500000E-01 2.0404500000E-01 2.0404500000E-01 + 1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 -1.0000000000E+00 + 1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + -1.9453890000E+01 1.0000000000E+00 -9.5304000000E-02 -2.2090309000E+01 + 1.0000000000E+00 -8.8198000000E-02 -4.9563591000E+01 1.0000000000E+00 + -1.0700000000E-04 1.7973449700E+02 -2.5957400000E+00 1.0000000000E+00 + -9.6217000000E-02 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + 9.6217000000E-02 -8.8940000000E-03 9.3082800000E-01 9.3082800000E-01 + 9.3082800000E-01 -1.1383520000E+00 9.5342000000E-02 8.1769800000E-01 + -5.5083500000E-01 4.6135000000E-02 8.1769800000E-01 -4.4892200000E-01 + 3.7599000000E-02 6.8068650000E+00 -2.0190000000E-03 1.6900000000E-04 + -6.0313600000E-01 -1.1914260000E+00 -2.0901000000E-01 -2.3236600000E-01 + -1.5881990000E+00 1.0000000000E+00 -8.9607000000E-02 -1.7894780000E+00 + 1.0000000000E+00 -8.2283000000E-02 -4.0128040000E+00 1.0000000000E+00 + -1.0000000000E-04 1.1868740000E+00 -8.7134300000E-01 9.9186000000E-01 + -7.3415000000E-01 -1.0000000000E+00 -1.0000000000E+00 -1.0000000000E+00 + 8.9782500000E-01 -1.0242700000E-01 2.6330249000E+02 1.6710230000E+01 + 2.5231250000E+02 1.5091380000E+01 2.3604249600E+02 1.1440290000E+01 + 6.8430000000E-03 9.6227400000E-01 -1.0000000000E+00 9.5486000000E-01 + 3.5042122000E+01 2.8167191000E+01 1.3668540000E+01 1.0077050000E+00 + 8.5846764000E+01 1.4416209400E+02 1.3040539600E+02 1.0000000000E+00 + 1.6589050000E+00 1.0000000000E+00 2.1064970000E+00 1.0000000000E+00 + 2.0009090000E+00 3.5042122000E+01 1.0000000000E+00 2.8167191000E+01 + 9.7782700000E-01 1.3668540000E+01 9.1623900000E-01 -1.0000000000E+00 + -1.0000000000E+00 3.4350000000E-03 3.4350000000E-03 3.4350000000E-03 + -1.0000000000E+00 -1.0000000000E+00 4.3017000000E-02 4.3017000000E-02 + 4.3017000000E-02 -1.0000000000E+00 -1.0000000000E+00 9.5354800000E-01 + 9.5354800000E-01 9.5354800000E-01 1.0000000000E+00 -1.0000000000E+00 + 1.0000000000E+00 -1.0000000000E+00 1.0000000000E+00 -1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 1.0000000000E+00 + 1.0000000000E+00 1.0000000000E+00 -4.7613000000E-02 1.0000000000E+00 + -2.3000000000E-05 -5.7464000000E-02 1.0000000000E+00 -3.5300000000E-04 + -1.2956000000E-01 1.0000000000E+00 -1.7625000000E-02 2.7046249400E+02 + -2.8000670000E+00 1.0000000000E+00 -1.6856930000E+00 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.6856930000E+00 -1.4266700000E-01 + 1.2149700000E-01 1.2149700000E-01 1.2149700000E-01 -1.1383520000E+00 + 2.1231000000E-02 6.0555800000E-01 -1.9490000000E-02 3.6300000000E-04 + 3.3579300000E-01 -1.3533800000E-01 2.5240000000E-03 1.0673100000E-01 + -9.5354800000E-01 1.7784000000E-02 -9.0759200000E-01 -5.7118220000E+00 + -9.3568400000E-01 -1.0346550000E+00 -4.3241000000E-02 1.0000000000E+00 + -2.6000000000E-05 -5.2175000000E-02 1.0000000000E+00 -3.9200000000E-04 + -1.1763100000E-01 1.0000000000E+00 -1.9600000000E-02 5.3140780000E+00 + -1.0021840000E+00 3.4483700000E-01 -2.6478600000E-01 -1.0000000000E+00 + -1.0000000000E+00 -1.0000000000E+00 1.7669710000E+00 -1.7474100000E-01 + 9.9149727000E+01 -1.0000000000E+00 -3.3000000000E-05 1.7263960300E+02 + -1.0000000000E+00 -4.1000000000E-05 3.5663980100E+02 -1.0000000000E+00 + -8.5000000000E-05 2.0000079300E+02 2.5000000000E+00 3.3000000000E-05 + 3.4900329600E+02 2.5000000000E+00 4.1000000000E-05 7.2206781000E+02 + 2.5000000000E+00 8.5000000000E-05 1.0000000000E+00 -1.0000000000E-06 + 1.0000000000E+00 -5.1100000000E-04 1.0000000000E+00 -9.5438900000E-01 + 1.0050250000E+00 -3.4350000000E-03 1.0028740000E+00 -4.3017000000E-02 + 1.0013870000E+00 -9.5354800000E-01 1.0000000000E+00 -4.9000000000E-05 + 1.0000000000E+00 -2.1780000000E-03 1.0000000000E+00 -4.2872000000E-02 + 1.0000000000E+00 -1.0000000000E+00 1.5000000000E+00 5.0000000000E-01 + 1.0000000000E+00 -3.1622000000E+05 -1.2132580078E+04 -1.0000000000E+00 + -2.0451810547E+04 9.1527509770E+03 9.1463570000E+00 3.7730000000E-03 + 2.2487060550E+03 -1.2505300000E-01 6.7588000000E-02 6.5087120000E+01 + -1.8859000000E-01 -9.1527509770E+03 7.5439400000E-01 -9.1527509770E+03 + -6.9047832030E+03 1.0000000000E+00 1.8569300000E-01 -5.0000000000E-01 + 1.9200000000E-04 2.6684520000E+00 8.8948400000E-01 5.2397100000E-01 + 2.5902730000E+00 1.2090400000E-01 1.0000000000E+00 3.6607700000E-01 + 1.8323000000E-02 5.3070800000E-01 2.6120320000E+00 1.2143800000E-01 + 1.0000000000E+00 3.6327500000E-01 1.9399000000E-02 5.3114800000E-01 + 2.6134470000E+00 1.2147300000E-01 1.0000000000E+00 3.6309200000E-01 + 1.9470000000E-02 5.3131600000E-01 2.6139860000E+00 1.2148600000E-01 + 1.0000000000E+00 3.6302200000E-01 1.9498000000E-02 5.3753300000E-01 + 2.6338800000E+00 1.2198000000E-01 1.0000000000E+00 3.6043600000E-01 + 2.0539000000E-02 7.1092800000E-01 3.1304670000E+00 1.3573600000E-01 + 1.0000000000E+00 2.8830900000E-01 7.1490000000E-02 diff --git a/RBio/README.txt b/RBio/README.txt index c143b28120..e2886191ec 100644 --- a/RBio/README.txt +++ b/RBio/README.txt @@ -1,4 +1,5 @@ -RBio: Version 2.0.0, Oct 20, 2009. A MATLAB Toolbox for reading/writing sparse +RBio: Copyright 2009, Timothy A. Davis +A MATLAB Toolbox for reading/writing sparse matrices in Rutherford/Boeing format. To install the MATLAB functions, cd to the RBio directory and type "RBinstall" @@ -32,7 +33,6 @@ MATLAB help for RBio: See also UFget, mread, mwrite. - Copyright 2009, Timothy A. Davis -------------------------------------------------------------------------------- Files and directories: diff --git a/README.txt b/README.txt index 7f3fc5d66d..8f46359ae2 100644 --- a/README.txt +++ b/README.txt @@ -12,7 +12,7 @@ SuiteSparse_install. All packages will be compiled, and several demos will be run. ================================================================================ -Nov 30, 2009. SuiteSparse version 3.5.0 +Jan 25, 2011. SuiteSparse VERSION 3.6.0 AMD approximate minimum degree ordering @@ -37,8 +37,8 @@ Nov 30, 2009. SuiteSparse version 3.5.0 UFconfig/UFconfig.mk is included in the Makefile's of all packages. CSparse and RBio do not use UFconfig. - CSparse a concise sparse matrix package, developed for my upcoming - book, "Direct Methods for Sparse Linear Systems", to be + CSparse a concise sparse matrix package, developed for my + book, "Direct Methods for Sparse Linear Systems", published by SIAM. CXSparse CSparse Extended. Includes support for complex matrices @@ -61,12 +61,12 @@ Nov 30, 2009. SuiteSparse version 3.5.0 SuiteSparseQR sparse QR factorization -CHOLMOD optionally uses METIS 4.0.1 +Some codes optionally use METIS 4.0.1 (http://www-users.cs.umn.edu/~karypis/metis). To use METIS, place a copy of the metis-4.0 directory in the same directory containing this README file. Be sure that you do not have a nested metis-4.0/metis-4.0 directory; SuiteSparse won't find METIS if you do this, which can happen with a zip file of metis-4.0 -on Windows. The use of METIS will improve the ordering quality in CHOLMOD. +on Windows. The use of METIS will improve the ordering quality. Refer to each package for license, copyright, and author information. All codes are authored or co-authored by Timothy A. Davis, CISE Dept., Univ. of @@ -84,13 +84,12 @@ To use "make" in Unix/Linux: (1) Use the right BLAS and LAPACK libraries - See http://www.netlib.org/blas for the Fortran reference BLAS (slow, but - they work). See http://www.tacc.utexas.edu/~kgoto/ or - http://www.cs.utexas.edu/users/flame/goto/ for an optimized BLAS. See - http://www.netlib.org/lapack for LAPACK. The UFconfig/UFconfig.mk file - assumes the vanilla BLAS (-lblas). You should use an optimized BLAS; - otherwise UMFPACK and CHOLMOD will be slow. Change -lblas to -l(your BLAS - library here) in the UFconfig/UFconfig.mk file. + Edit your UFconfig/UFconfig.mk file to point to the right compilers, + and to the correct BLAS and LAPACK libraries. There are many examples + of different computer architectures there. Scroll through to find yours, + and uncomment those lines. + + By default, the UFconfig.mk file assumes -lblas and -llapack. (2) Install Intel's Threading Building Blocks (TBB) diff --git a/SPQR/Demo/qrdemo.cpp b/SPQR/Demo/qrdemo.cpp index c1eb8dbaf4..3e8178b60d 100644 --- a/SPQR/Demo/qrdemo.cpp +++ b/SPQR/Demo/qrdemo.cpp @@ -13,14 +13,57 @@ // platform. For most platforms (except Windows), UF_long is just "long". #define Int UF_long +// ============================================================================= +// check_residual: print the relative residual, norm (A*x-b)/norm(x) +// ============================================================================= + +void check_residual +( + cholmod_sparse *A, + cholmod_dense *X, + cholmod_dense *B, + cholmod_common *cc +) +{ + Int m = A->nrow ; + Int n = A->ncol ; + Int rnk ; + double rnorm, anorm, xnorm ; + double one [2] = {1,0}, minusone [2] = {-1,0} ; + cholmod_dense *Residual ; + + // get the rank(A) estimate + rnk = cc->SPQR_istat [4] ; + + // anorm = norm (A,1) ; + anorm = cholmod_l_norm_sparse (A, 1, cc) ; + + // rnorm = norm (A*X-B) + Residual = cholmod_l_copy_dense (B, cc) ; + cholmod_l_sdmult (A, 0, one, minusone, X, Residual, cc) ; + rnorm = cholmod_l_norm_dense (Residual, 2, cc) ; + + // xnorm = norm (X) + xnorm = cholmod_l_norm_dense (X, 2, cc) ; + + if (m <= n && anorm > 0 && xnorm > 0) + { + // find the relative residual, except for least-squares systems + rnorm /= (anorm * xnorm) ; + } + printf ("residual: %8.1e rank: %6ld\n", rnorm, rnk) ; + cholmod_l_free_dense (&Residual, cc) ; +} + +// ============================================================================= + int main (int argc, char **argv) { cholmod_common Common, *cc ; cholmod_sparse *A ; - cholmod_dense *X, *B, *Residual ; - double anorm, xnorm, rnorm, one [2] = {1,0}, minusone [2] = {-1,0} ; + cholmod_dense *X, *B ; int mtype ; - Int m, n, rnk ; + Int m, n ; // start CHOLMOD cc = &Common ; @@ -38,10 +81,7 @@ int main (int argc, char **argv) m = A->nrow ; n = A->ncol ; - // anorm = norm (A,1) ; - anorm = cholmod_l_norm_sparse (A, 1, cc) ; - - printf ("Matrix %6ld-by-%-6ld nnz: %6ld ", m, n, cholmod_l_nnz (A, cc)) ; + printf ("Matrix %6ld-by-%-6ld nnz: %6ld\n", m, n, cholmod_l_nnz (A, cc)) ; // B = ones (m,1), a dense right-hand-side of the same type as A B = cholmod_l_ones (m, 1, A->xtype, cc) ; @@ -60,28 +100,63 @@ int main (int argc, char **argv) (SPQR_ORDERING_DEFAULT, SPQR_DEFAULT_TOL, A, B, cc) ; } - // get the rank(A) estimate - rnk = cc->SPQR_istat [4] ; - - // rnorm = norm (A*X-B) - Residual = cholmod_l_copy_dense (B, cc) ; - cholmod_l_sdmult (A, 0, one, minusone, X, Residual, cc) ; - rnorm = cholmod_l_norm_dense (Residual, 2, cc) ; + check_residual (A, X, B, cc) ; + cholmod_l_free_dense (&X, cc) ; - // xnorm = norm (X) - xnorm = cholmod_l_norm_dense (X, 2, cc) ; + // ------------------------------------------------------------------------- + // factorizing once then solving twice with different right-hand-sides + // ------------------------------------------------------------------------- - if (m <= n && anorm > 0 && xnorm > 0) + // Just the real case. Complex case is essentially identical + if (A->xtype == CHOLMOD_REAL) { - // find the relative residual, except for least-squares systems - rnorm /= (anorm * xnorm) ; + SuiteSparseQR_factorization*QR ; + cholmod_dense *Y ; + Int i ; + double *Bx ; + + // factorize once + QR = SuiteSparseQR_factorize + (SPQR_ORDERING_DEFAULT, SPQR_DEFAULT_TOL, A, cc) ; + + // solve Ax=b, using the same B as before + + // Y = Q'*B + Y = SuiteSparseQR_qmult (SPQR_QTX, QR, B, cc) ; + // X = R\(E*Y) + X = SuiteSparseQR_solve (SPQR_RETX_EQUALS_B, QR, Y, cc) ; + // check the results + check_residual (A, X, B, cc) ; + // free X and Y + cholmod_l_free_dense (&Y, cc) ; + cholmod_l_free_dense (&X, cc) ; + + // repeat with a different B + Bx = (double *) (B->x) ; + for (i = 0 ; i < m ; i++) + { + Bx [i] = i ; + } + + // Y = Q'*B + Y = SuiteSparseQR_qmult (SPQR_QTX, QR, B, cc) ; + // X = R\(E*Y) + X = SuiteSparseQR_solve (SPQR_RETX_EQUALS_B, QR, Y, cc) ; + // check the results + check_residual (A, X, B, cc) ; + // free X and Y + cholmod_l_free_dense (&Y, cc) ; + cholmod_l_free_dense (&X, cc) ; + + // free QR + SuiteSparseQR_free (&QR, cc) ; } - printf ("residual: %8.1e rank: %6ld\n", rnorm, rnk) ; - // free everything - cholmod_l_free_dense (&Residual, cc) ; + // ------------------------------------------------------------------------- + // free everything that remains + // ------------------------------------------------------------------------- + cholmod_l_free_sparse (&A, cc) ; - cholmod_l_free_dense (&X, cc) ; cholmod_l_free_dense (&B, cc) ; cholmod_l_finish (cc) ; return (0) ; diff --git a/SPQR/Demo/qrdemo_out.txt b/SPQR/Demo/qrdemo_out.txt index eafefcfc62..34f70e99c0 100644 --- a/SPQR/Demo/qrdemo_out.txt +++ b/SPQR/Demo/qrdemo_out.txt @@ -1,66 +1,148 @@ +( cd ../Lib ; make ) +make[1]: Entering directory `/cise/research/sparse2/SuiteSparse/SPQR/Lib' +make[1]: Nothing to be done for `default'. +make[1]: Leaving directory `/cise/research/sparse2/SuiteSparse/SPQR/Lib' +g++ -O3 -fexceptions -I../../CHOLMOD/Include -I../../UFconfig -I../Include qrdemo.cpp -o qrdemo ../Lib/libspqr.a ../../CHOLMOD/Lib/libcholmod.a ../../AMD/Lib/libamd.a ../../COLAMD/Lib/libcolamd.a ../../metis-4.0/libmetis.a ../../CCOLAMD/Lib/libccolamd.a ../../CAMD/Lib/libcamd.a -lm -llapack_plain -lblas_plain -lgfortran -lgfortranbegin +g++ -O3 -fexceptions -I../../CHOLMOD/Include -I../../UFconfig -I../Include qrsimple.cpp -o qrsimple ../Lib/libspqr.a ../../CHOLMOD/Lib/libcholmod.a ../../AMD/Lib/libamd.a ../../COLAMD/Lib/libcolamd.a ../../metis-4.0/libmetis.a ../../CCOLAMD/Lib/libccolamd.a ../../CAMD/Lib/libcamd.a -lm -llapack_plain -lblas_plain -lgfortran -lgfortranbegin +g++ -O3 -fexceptions -I../../CHOLMOD/Include -I../../UFconfig -I../Include -o qrsimplec qrsimplec.o ../Lib/libspqr.a ../../CHOLMOD/Lib/libcholmod.a ../../AMD/Lib/libamd.a ../../COLAMD/Lib/libcolamd.a ../../metis-4.0/libmetis.a ../../CCOLAMD/Lib/libccolamd.a ../../CAMD/Lib/libcamd.a -lm -llapack_plain -lblas_plain -lgfortran -lgfortranbegin ./qrsimple < ../Matrix/ash219.mtx -2-norm of residual: 2.9e-15 +2-norm of residual: 3.8e-15 rank 85 ./qrsimplec < ../Matrix/ash219.mtx -2-norm of residual: 2.9e-15 +2-norm of residual: 3.8e-15 rank 85 ./qrsimple < ../Matrix/west0067.mtx -2-norm of residual: 6.7e-15 +2-norm of residual: 8.3e-15 rank 67 ./qrsimplec < ../Matrix/west0067.mtx -2-norm of residual: 6.7e-15 +2-norm of residual: 8.3e-15 rank 67 ./qrdemo < ../Matrix/a2.mtx -Matrix 2-by-2 nnz: 3 residual: 1.5e-17 rank: 2 +Matrix 2-by-2 nnz: 3 +residual: 0.0e+00 rank: 2 +residual: 0.0e+00 rank: 2 +residual: 0.0e+00 rank: 2 ./qrdemo < ../Matrix/r2.mtx -Matrix 2-by-3 nnz: 4 residual: 1.1e-17 rank: 2 +Matrix 2-by-3 nnz: 4 +residual: 0.0e+00 rank: 2 +residual: 0.0e+00 rank: 2 +residual: 0.0e+00 rank: 2 ./qrdemo < ../Matrix/a04.mtx -Matrix 0-by-4 nnz: 0 residual: 0.0e+00 rank: 0 +Matrix 0-by-4 nnz: 0 +residual: 0.0e+00 rank: 0 +residual: 0.0e+00 rank: 0 +residual: 0.0e+00 rank: 0 ./qrdemo < ../Matrix/a2.mtx -Matrix 2-by-2 nnz: 3 residual: 1.5e-17 rank: 2 +Matrix 2-by-2 nnz: 3 +residual: 0.0e+00 rank: 2 +residual: 0.0e+00 rank: 2 +residual: 0.0e+00 rank: 2 ./qrdemo < ../Matrix/west0067.mtx -Matrix 67-by-67 nnz: 294 residual: 4.1e-17 rank: 67 +Matrix 67-by-67 nnz: 294 +residual: 5.1e-17 rank: 67 +residual: 6.1e-17 rank: 67 +residual: 5.7e-17 rank: 67 ./qrdemo < ../Matrix/c2.mtx -Matrix 2-by-2 nnz: 4 residual: 1.1e-16 rank: 2 +Matrix 2-by-2 nnz: 4 +residual: 1.7e-16 rank: 2 ./qrdemo < ../Matrix/a0.mtx -Matrix 0-by-0 nnz: 0 residual: 0.0e+00 rank: 0 +Matrix 0-by-0 nnz: 0 +residual: 0.0e+00 rank: 0 +residual: 0.0e+00 rank: 0 +residual: 0.0e+00 rank: 0 ./qrdemo < ../Matrix/lfat5b.mtx -Matrix 14-by-14 nnz: 46 residual: 5.1e-17 rank: 14 +Matrix 14-by-14 nnz: 46 +residual: 8.6e-17 rank: 14 +residual: 1.3e-16 rank: 14 +residual: 3.5e-17 rank: 14 ./qrdemo < ../Matrix/bfwa62.mtx -Matrix 62-by-62 nnz: 450 residual: 3.7e-17 rank: 62 +Matrix 62-by-62 nnz: 450 +residual: 6.4e-17 rank: 62 +residual: 5.5e-17 rank: 62 +residual: 4.4e-17 rank: 62 ./qrdemo < ../Matrix/LFAT5.mtx -Matrix 14-by-14 nnz: 46 residual: 2.4e-22 rank: 14 +Matrix 14-by-14 nnz: 46 +residual: 5.7e-22 rank: 14 +residual: 5.7e-22 rank: 14 +residual: 3.7e-21 rank: 14 ./qrdemo < ../Matrix/b1_ss.mtx -Matrix 7-by-7 nnz: 15 residual: 1.0e-17 rank: 7 +Matrix 7-by-7 nnz: 15 +residual: 7.3e-18 rank: 7 +residual: 8.2e-18 rank: 7 +residual: 3.2e-17 rank: 7 ./qrdemo < ../Matrix/bcspwr01.mtx -Matrix 39-by-39 nnz: 131 residual: 5.8e-17 rank: 39 +Matrix 39-by-39 nnz: 131 +residual: 5.7e-17 rank: 39 +residual: 8.0e-17 rank: 39 +residual: 6.0e-17 rank: 39 ./qrdemo < ../Matrix/lpi_galenet.mtx -Matrix 8-by-14 nnz: 22 residual: 0.0e+00 rank: 8 +Matrix 8-by-14 nnz: 22 +residual: 0.0e+00 rank: 8 +residual: 0.0e+00 rank: 8 +residual: 0.0e+00 rank: 8 ./qrdemo < ../Matrix/lpi_itest6.mtx -Matrix 11-by-17 nnz: 29 residual: 0.0e+00 rank: 11 +Matrix 11-by-17 nnz: 29 +residual: 0.0e+00 rank: 11 +residual: 0.0e+00 rank: 11 +residual: 0.0e+00 rank: 11 ./qrdemo < ../Matrix/ash219.mtx -Matrix 219-by-85 nnz: 438 residual: 2.9e-15 rank: 85 +Matrix 219-by-85 nnz: 438 +residual: 3.8e-15 rank: 85 +residual: 4.6e-15 rank: 85 +residual: 1.7e+02 rank: 85 ./qrdemo < ../Matrix/a4.mtx -Matrix 4-by-2 nnz: 4 residual: 1.4e+00 rank: 2 +Matrix 4-by-2 nnz: 4 +residual: 1.4e+00 rank: 2 +residual: 1.4e+00 rank: 2 +residual: 3.6e+00 rank: 2 ./qrdemo < ../Matrix/s32.mtx -Matrix 3-by-2 nnz: 6 residual: 1.1e-02 rank: 2 +Matrix 3-by-2 nnz: 6 +residual: 1.1e-02 rank: 2 +residual: 1.1e-02 rank: 2 +residual: 1.4e-02 rank: 2 ./qrdemo < ../Matrix/c32.mtx -Matrix 3-by-2 nnz: 6 residual: 1.9e-01 rank: 2 +Matrix 3-by-2 nnz: 6 +residual: 1.9e-01 rank: 2 ./qrdemo < ../Matrix/lp_share1b.mtx -Matrix 117-by-253 nnz: 1179 residual: 8.4e-20 rank: 117 +Matrix 117-by-253 nnz: 1179 +residual: 7.2e-20 rank: 117 +residual: 1.2e-19 rank: 117 +residual: 8.9e-20 rank: 117 ./qrdemo < ../Matrix/a1.mtx -Matrix 4-by-4 nnz: 7 residual: 6.1e-01 rank: 2 +Matrix 4-by-4 nnz: 7 +residual: 6.1e-01 rank: 2 +residual: 6.1e-01 rank: 2 +residual: 1.4e+00 rank: 2 ./qrdemo < ../Matrix/GD06_theory.mtx -Matrix 101-by-101 nnz: 380 residual: 7.2e-02 rank: 20 +Matrix 101-by-101 nnz: 380 +residual: 7.2e-02 rank: 20 +residual: 7.2e-02 rank: 20 +residual: 7.3e-02 rank: 20 ./qrdemo < ../Matrix/GD01_b.mtx -Matrix 18-by-18 nnz: 37 residual: 8.6e-17 rank: 17 +Matrix 18-by-18 nnz: 37 +residual: 7.9e-17 rank: 17 +residual: 1.8e-16 rank: 17 +residual: 9.9e-03 rank: 17 ./qrdemo < ../Matrix/Tina_AskCal_perm.mtx -Matrix 11-by-11 nnz: 29 residual: 7.9e-18 rank: 9 +Matrix 11-by-11 nnz: 29 +residual: 3.8e-17 rank: 9 +residual: 5.0e-17 rank: 9 +residual: 1.7e-02 rank: 9 ./qrdemo < ../Matrix/Tina_AskCal.mtx -Matrix 11-by-11 nnz: 29 residual: 5.6e-18 rank: 9 +Matrix 11-by-11 nnz: 29 +residual: 3.7e-17 rank: 9 +residual: 5.0e-17 rank: 9 +residual: 3.1e-02 rank: 9 ./qrdemo < ../Matrix/GD98_a.mtx -Matrix 38-by-38 nnz: 50 residual: 2.4e-01 rank: 14 +Matrix 38-by-38 nnz: 50 +residual: 2.4e-01 rank: 14 +residual: 2.4e-01 rank: 14 +residual: 2.2e-01 rank: 14 ./qrdemo < ../Matrix/Ragusa16.mtx -Matrix 24-by-24 nnz: 81 residual: 2.1e-02 rank: 18 +Matrix 24-by-24 nnz: 81 +residual: 2.1e-02 rank: 18 +residual: 2.1e-02 rank: 18 +residual: 1.5e-02 rank: 18 ./qrdemo < ../Matrix/young1c.mtx -Matrix 841-by-841 nnz: 4089 residual: 1.2e-16 rank: 841 +Matrix 841-by-841 nnz: 4089 +residual: 2.1e-16 rank: 841 diff --git a/SPQR/Demo/qrdemoc_out.txt b/SPQR/Demo/qrdemoc_out.txt index fe7c6c938f..ff2cdf4e87 100644 --- a/SPQR/Demo/qrdemoc_out.txt +++ b/SPQR/Demo/qrdemoc_out.txt @@ -1,33 +1,39 @@ +cc -O3 -fexceptions -ansi -I../../CHOLMOD/Include -I../../UFconfig -I../Include -c qrdemoc.c +( cd ../Lib ; make ) +make[1]: Entering directory `/cise/research/sparse2/SuiteSparse/SPQR/Lib' +make[1]: Nothing to be done for `default'. +make[1]: Leaving directory `/cise/research/sparse2/SuiteSparse/SPQR/Lib' +g++ -O3 -fexceptions -I../../CHOLMOD/Include -I../../UFconfig -I../Include -o qrdemoc qrdemoc.o ../Lib/libspqr.a ../../CHOLMOD/Lib/libcholmod.a ../../AMD/Lib/libamd.a ../../COLAMD/Lib/libcolamd.a ../../metis-4.0/libmetis.a ../../CCOLAMD/Lib/libccolamd.a ../../CAMD/Lib/libcamd.a -lm -llapack_plain -lblas_plain -lgfortran -lgfortranbegin ./qrdemoc < ../Matrix/a2.mtx -Matrix 2-by-2 nnz: 3 residual: 1.5e-17 rank: 2 +Matrix 2-by-2 nnz: 3 residual: 0.0e+00 rank: 2 ./qrdemoc < ../Matrix/r2.mtx -Matrix 2-by-3 nnz: 4 residual: 1.1e-17 rank: 2 +Matrix 2-by-3 nnz: 4 residual: 0.0e+00 rank: 2 ./qrdemoc < ../Matrix/a04.mtx Matrix 0-by-4 nnz: 0 residual: 0.0e+00 rank: 0 ./qrdemoc < ../Matrix/a2.mtx -Matrix 2-by-2 nnz: 3 residual: 1.5e-17 rank: 2 +Matrix 2-by-2 nnz: 3 residual: 0.0e+00 rank: 2 ./qrdemoc < ../Matrix/west0067.mtx -Matrix 67-by-67 nnz: 294 residual: 4.1e-17 rank: 67 +Matrix 67-by-67 nnz: 294 residual: 5.1e-17 rank: 67 ./qrdemoc < ../Matrix/c2.mtx -Matrix 2-by-2 nnz: 4 residual: 1.1e-16 rank: 2 +Matrix 2-by-2 nnz: 4 residual: 1.7e-16 rank: 2 ./qrdemoc < ../Matrix/a0.mtx Matrix 0-by-0 nnz: 0 residual: 0.0e+00 rank: 0 ./qrdemoc < ../Matrix/lfat5b.mtx -Matrix 14-by-14 nnz: 46 residual: 5.1e-17 rank: 14 +Matrix 14-by-14 nnz: 46 residual: 8.6e-17 rank: 14 ./qrdemoc < ../Matrix/bfwa62.mtx -Matrix 62-by-62 nnz: 450 residual: 3.7e-17 rank: 62 +Matrix 62-by-62 nnz: 450 residual: 6.4e-17 rank: 62 ./qrdemoc < ../Matrix/LFAT5.mtx -Matrix 14-by-14 nnz: 46 residual: 2.4e-22 rank: 14 +Matrix 14-by-14 nnz: 46 residual: 5.7e-22 rank: 14 ./qrdemoc < ../Matrix/b1_ss.mtx -Matrix 7-by-7 nnz: 15 residual: 1.0e-17 rank: 7 +Matrix 7-by-7 nnz: 15 residual: 7.3e-18 rank: 7 ./qrdemoc < ../Matrix/bcspwr01.mtx -Matrix 39-by-39 nnz: 131 residual: 5.8e-17 rank: 39 +Matrix 39-by-39 nnz: 131 residual: 5.7e-17 rank: 39 ./qrdemoc < ../Matrix/lpi_galenet.mtx Matrix 8-by-14 nnz: 22 residual: 0.0e+00 rank: 8 ./qrdemoc < ../Matrix/lpi_itest6.mtx Matrix 11-by-17 nnz: 29 residual: 0.0e+00 rank: 11 ./qrdemoc < ../Matrix/ash219.mtx -Matrix 219-by-85 nnz: 438 residual: 2.9e-15 rank: 85 +Matrix 219-by-85 nnz: 438 residual: 3.8e-15 rank: 85 ./qrdemoc < ../Matrix/a4.mtx Matrix 4-by-2 nnz: 4 residual: 1.4e+00 rank: 2 ./qrdemoc < ../Matrix/s32.mtx @@ -35,20 +41,20 @@ Matrix 3-by-2 nnz: 6 residual: 1.1e-02 rank: 2 ./qrdemoc < ../Matrix/c32.mtx Matrix 3-by-2 nnz: 6 residual: 1.9e-01 rank: 2 ./qrdemoc < ../Matrix/lp_share1b.mtx -Matrix 117-by-253 nnz: 1179 residual: 8.4e-20 rank: 117 +Matrix 117-by-253 nnz: 1179 residual: 7.2e-20 rank: 117 ./qrdemoc < ../Matrix/a1.mtx Matrix 4-by-4 nnz: 7 residual: 6.1e-01 rank: 2 ./qrdemoc < ../Matrix/GD06_theory.mtx Matrix 101-by-101 nnz: 380 residual: 7.2e-02 rank: 20 ./qrdemoc < ../Matrix/GD01_b.mtx -Matrix 18-by-18 nnz: 37 residual: 8.6e-17 rank: 17 +Matrix 18-by-18 nnz: 37 residual: 7.9e-17 rank: 17 ./qrdemoc < ../Matrix/Tina_AskCal_perm.mtx -Matrix 11-by-11 nnz: 29 residual: 7.9e-18 rank: 9 +Matrix 11-by-11 nnz: 29 residual: 3.8e-17 rank: 9 ./qrdemoc < ../Matrix/Tina_AskCal.mtx -Matrix 11-by-11 nnz: 29 residual: 5.6e-18 rank: 9 +Matrix 11-by-11 nnz: 29 residual: 3.7e-17 rank: 9 ./qrdemoc < ../Matrix/GD98_a.mtx Matrix 38-by-38 nnz: 50 residual: 2.4e-01 rank: 14 ./qrdemoc < ../Matrix/Ragusa16.mtx Matrix 24-by-24 nnz: 81 residual: 2.1e-02 rank: 18 ./qrdemoc < ../Matrix/young1c.mtx -Matrix 841-by-841 nnz: 4089 residual: 1.2e-16 rank: 841 +Matrix 841-by-841 nnz: 4089 residual: 2.1e-16 rank: 841 diff --git a/SPQR/Doc/ChangeLog b/SPQR/Doc/ChangeLog index a79bcc80c1..d2c7773f2f 100644 --- a/SPQR/Doc/ChangeLog +++ b/SPQR/Doc/ChangeLog @@ -1,3 +1,10 @@ +Jan 25, 2011: version 1.2.1 + + * minor fix to "make install" + * added more examples to SPRQ/Demo/qrdemo.cpp + * fixed error-condition return value for Source/SuiteSparseQR.cpp + * minor fix to SPQR/Tcov/qrtest.cpp + Nov 30, 2009: version 1.2.0 * fixed interface to 64-bit BLAS for MATLAB 7.8 diff --git a/SPQR/Include/SuiteSparseQR_definitions.h b/SPQR/Include/SuiteSparseQR_definitions.h index 0e903cdb45..492dc6100b 100644 --- a/SPQR/Include/SuiteSparseQR_definitions.h +++ b/SPQR/Include/SuiteSparseQR_definitions.h @@ -59,11 +59,11 @@ #endif */ -#define SPQR_DATE "Nov 30, 2009" +#define SPQR_DATE "Jan 25, 2011" #define SPQR_VER_CODE(main,sub) ((main) * 1000 + (sub)) #define SPQR_MAIN_VERSION 1 #define SPQR_SUB_VERSION 2 -#define SPQR_SUBSUB_VERSION 0 +#define SPQR_SUBSUB_VERSION 1 #define SPQR_VERSION SPQR_VER_CODE(SPQR_MAIN_VERSION,SPQR_SUB_VERSION) #endif diff --git a/SPQR/MATLAB/spqr.m b/SPQR/MATLAB/spqr.m index 6df3a52637..56358f8306 100644 --- a/SPQR/MATLAB/spqr.m +++ b/SPQR/MATLAB/spqr.m @@ -1,4 +1,4 @@ -function [out1,out2,P] = spqr (A,arg2,arg3) %#ok +function [Q,R,P,info] = spqr (A,arg2,arg3) %#ok %SPQR multithreaded multifrontal rank-revealing sparse QR. %For a sparse m-by-n matrix A, and sparse or full B of size m-by-k: % @@ -80,10 +80,10 @@ % the default ordering: COLAMD if m<=2*n; otherwise try AMD. If the flops % for chol(SP'SP) / nnz(R) >= 500 and nnz(R)/nnz(S)>=5 with the ordering P % found by AMD, then try METIS, and take the best ordering found. 'amd': -% AMD(S'*S) 'colamd': COLAMD(S) 'metis': METIS(S'*S), only if METIS is -% installed 'best': try all three (AMD, COLAMD, METIS) and take the best -% 'bestamd': try AMD and COLAMD and take the best 'fixed': P=I; this is the -% only option if P is not present in the output 'natural': singleton removal +% AMD(S'*S). 'colamd': COLAMD(S) 'metis': METIS(S'*S), only if METIS is +% installed. 'best': try all three (AMD, COLAMD, METIS) and take the best +% 'bestamd': try AMD and COLAMD and take the best. 'fixed': P=I; this is the +% only option if P is not present in the output. 'natural': singleton removal % only The singleton pre-ordering permutes A prior to factorization into the % form [A11 A12 ; 0 A22] where A11 is upper triangular with % all(abs(diag(A11)) > opts.tol) (see spqr_singletons). @@ -127,5 +127,4 @@ % Copyright 2008-2009, Timothy A. Davis % http://www.cise.ufl.edu/research/sparse -type spqr error ('spqr mexFunction not found') ; diff --git a/SPQR/Makefile b/SPQR/Makefile index 50f5991ac3..3c13aadfa7 100644 --- a/SPQR/Makefile +++ b/SPQR/Makefile @@ -2,7 +2,7 @@ # SuiteSparseQR Makefile #------------------------------------------------------------------------------- -VERSION = 1.2.0 +VERSION = 1.2.1 include ../UFconfig/UFconfig.mk @@ -42,7 +42,7 @@ ccode: demos # install SPQR install: $(CP) Lib/libspqr.a $(INSTALL_LIB)/libspqr.$(VERSION).a - ( cd $(INSTALL_LIB) ; ln -s libspqr.$(VERSION).a libspqr.a ) + ( cd $(INSTALL_LIB) ; ln -sf libspqr.$(VERSION).a libspqr.a ) $(CP) Include/SuiteSparseQR.hpp $(INSTALL_INCLUDE) $(CP) Include/SuiteSparseQR_C.h $(INSTALL_INCLUDE) $(CP) Include/SuiteSparseQR_definitions.h $(INSTALL_INCLUDE) diff --git a/SPQR/README.txt b/SPQR/README.txt index 41b41a0a07..2f79ef1d37 100644 --- a/SPQR/README.txt +++ b/SPQR/README.txt @@ -1,4 +1,4 @@ -SuiteSparseQR version 1.1.0, Sept 20, 2008, Copyright (c) 2008, Timothy A. Davis +SuiteSparseQR Copyright (c) 2008-2011, Timothy A. Davis SuiteSparseQR is a a multithread, multifrontal, rank-revealing sparse QR factorization method. diff --git a/SPQR/Source/SuiteSparseQR.cpp b/SPQR/Source/SuiteSparseQR.cpp index aeec4bcde7..21e4268214 100644 --- a/SPQR/Source/SuiteSparseQR.cpp +++ b/SPQR/Source/SuiteSparseQR.cpp @@ -130,12 +130,12 @@ template Int SuiteSparseQR if (p_HPinv != NULL) *p_HPinv = NULL ; if (p_HTau != NULL) *p_HTau = NULL ; - RETURN_IF_NULL_COMMON (FALSE) ; - RETURN_IF_NULL (A, FALSE) ; + RETURN_IF_NULL_COMMON (EMPTY) ; + RETURN_IF_NULL (A, EMPTY) ; Int xtype = spqr_type ( ) ; - RETURN_IF_XTYPE_INVALID (A, FALSE) ; - if (Bsparse != NULL) RETURN_IF_XTYPE_INVALID (Bsparse, FALSE) ; - if (Bdense != NULL) RETURN_IF_XTYPE_INVALID (Bdense, FALSE) ; + RETURN_IF_XTYPE_INVALID (A, EMPTY) ; + if (Bsparse != NULL) RETURN_IF_XTYPE_INVALID (Bsparse, EMPTY) ; + if (Bdense != NULL) RETURN_IF_XTYPE_INVALID (Bdense, EMPTY) ; cc->status = CHOLMOD_OK ; n = 0 ; diff --git a/SPQR/Source/SuiteSparseQR_expert.cpp b/SPQR/Source/SuiteSparseQR_expert.cpp index 3398efc7cf..ae9afe4431 100644 --- a/SPQR/Source/SuiteSparseQR_expert.cpp +++ b/SPQR/Source/SuiteSparseQR_expert.cpp @@ -174,6 +174,8 @@ SuiteSparseQR_factorization *SuiteSparseQR_symbolic // SuiteSparseQR_symbolic. Alternatively, if SuiteSparseQR_factorize // did not find any singletons (QR->n1cols == 0), a call to // SuiteSparseQR_factorize can be followed by a factorization by this function. +// +// Returns TRUE if successful, FALSE otherwise. template int SuiteSparseQR_numeric ( diff --git a/SPQR/Tcov/qrtest.cpp b/SPQR/Tcov/qrtest.cpp index 6b3af9bab6..d85e5a9295 100644 --- a/SPQR/Tcov/qrtest.cpp +++ b/SPQR/Tcov/qrtest.cpp @@ -2576,7 +2576,15 @@ int do_matrix (int kind, FILE *file, cholmod_common *cc) fprintf (stderr, "Unable to read matrix\n") ; return (1) ; } - fprintf (stderr, "%5ld by %5ld : ", (Int) A->nrow, (Int) A->ncol) ; + Int m = A->nrow ; + Int n = A->ncol ; + fprintf (stderr, "%5ld by %5ld : ", m, n) ; + if (sizeof (Int) > sizeof (int) && (m > 10000 || n > 10000)) + { + fprintf (stderr, "(test skipped on 64-bit systems)\n") ; + cholmod_l_free_sparse (&A, cc) ; + return (0) ; + } // ------------------------------------------------------------------------- // use it to test SuiteSparseQR @@ -2608,9 +2616,6 @@ int do_matrix (int kind, FILE *file, cholmod_common *cc) // free the matrix and report the results // ------------------------------------------------------------------------- - Int m = A->nrow ; - Int n = A->ncol ; - if (kind == 0) { printf ("First Resid and ") ; @@ -2704,7 +2709,7 @@ int main (int argc, char **argv) } fprintf (stderr, "%-30s ", matrix_name) ; FILE *matrix = fopen (matrix_name, "r") ; - if (file == NULL) + if (matrix == NULL) { fprintf (stderr, "Unable to open %s\n", matrix_name) ; nfail++ ; @@ -2712,6 +2717,7 @@ int main (int argc, char **argv) nfail += do_matrix (kind, matrix, cc) ; fclose (matrix) ; } + fclose (file) ; } // ------------------------------------------------------------------------- diff --git a/SSMULT/Makefile b/SSMULT/Makefile index 9b29999c2d..ce847adb59 100644 --- a/SSMULT/Makefile +++ b/SSMULT/Makefile @@ -24,4 +24,4 @@ clean: rm -f *.o purge: clean - rm *.mexglx + rm *.mex* diff --git a/SSMULT/README.txt b/SSMULT/README.txt index c8e93ab036..ef3772e95e 100644 --- a/SSMULT/README.txt +++ b/SSMULT/README.txt @@ -1,5 +1,5 @@ -SSMULT version 2.0, Mar 25, 2009. Distributed under the GNU GPL license (see -below). Copyright (c) 2007-2009, Timothy A. Davis, University of Florida. +SSMULT, Copyright (c) 2007-2009, Timothy A. Davis, University of Florida. +Distributed under the GNU GPL license (see below). SSMULT is also available under other licenses; contact the author for details. http://www.cise.ufl.edu/research/sparse diff --git a/SuiteSparse_install.m b/SuiteSparse_install.m index 3ee09bc0d4..5308e7fb03 100644 --- a/SuiteSparse_install.m +++ b/SuiteSparse_install.m @@ -62,7 +62,7 @@ function SuiteSparse_install (do_demo) % check if METIS 4.0.1 is present where it's supposed to be have_metis = exist ('metis-4.0', 'dir') ; if (~have_metis) - fprintf ('SPQR, CHOLMOD, and KLU optionally use METIS 4.0.1. Download\n') ; + fprintf ('Some codes optionally use METIS 4.0.1. Download\n') ; fprintf ('it from http://glaros.dtc.umn.edu/gkhome/views/metis\n'); fprintf ('and place the metis-4.0 directory in this directory.\n') ; input ('or hit enter to continue without METIS: ', 's') ; diff --git a/UFcollection/Doc/ChangeLog b/UFcollection/Doc/ChangeLog index f6688880ae..754187f269 100644 --- a/UFcollection/Doc/ChangeLog +++ b/UFcollection/Doc/ChangeLog @@ -1,3 +1,8 @@ +Jan 25, 2011: version 1.4.0 + + * added new graphs to UFpage.m (bipartite, A+A') + * METIS no longer used. It causes too many segfaults... + Nov 30, 2009: version 1.3.0 * added code to compute UFstats.csv for UFgui diff --git a/UFcollection/README.txt b/UFcollection/README.txt index 65d7bf25e0..27f57dcf84 100644 --- a/UFcollection/README.txt +++ b/UFcollection/README.txt @@ -1,4 +1,4 @@ -UFcollection, Version 1.3.0, Nov 30, 2009. +UFcollection, Copyright 2007-2011, Timothy A. Davis UFcollection is a MATLAB toolbox for managing the UF Sparse Matrix Collection. If you are a MATLAB user of the collection, you would not normally need to use @@ -54,7 +54,6 @@ MATLAB help for the UFcollection toolbox: Requires UFget, CSparse, CHOLMOD, AMD, COLAMD, RBio, and METIS. - Copyright 2007, Timothy A. Davis -------------------------------------------------------------------------------- Files: diff --git a/UFcollection/UFindex.m b/UFcollection/UFindex.m index b639966dcc..36974c9c8e 100644 --- a/UFcollection/UFindex.m +++ b/UFcollection/UFindex.m @@ -36,16 +36,19 @@ % amd_flops flop count for chol(C(p,p)) where, C=A+A', p=amd(C) % amd_vnz nnz in Householder vectors for qr(A(:,colamd(A))) % amd_rnz nnz in R for qr(A(:,colamd(A))) +% +% These results are for matrices 1:2336 only: % metis_lnz nnz(L) for chol(C(p,p)) where, C=A+A', p=metis(C) % metis_flops flop count for chol(C(p,p)) where, C=A+A', p=metis(C) % metis_vnz nnz in Householder vectors for qr(A(:,metis(A,'col'))) % metis_rnz nnz in R for qr(A(:,metis(A,'col'))) +% % nblocks # of blocks from dmperm % sprank sprank(A) % nzoff # of entries not in diagonal blocks from dmperm % ncc # of strongly connected components -% dmperm_lnz nnz(L), using dmperm plus amd or metis -% dmperm_unz nnz(U), using dmperm plus amd or metis +% dmperm_lnz nnz(L), using dmperm plus amd (or metis for matrices 1:2336) +% dmperm_unz nnz(U), using dmperm plus amd (or metis for matrices 1:2336) % dmperm_flops flop count with dperm plus % dmperm_vnz nnz in Householder vectors for dmperm plus % dmperm_rnz nnz in R for dmperm plus @@ -103,7 +106,14 @@ fprintf ('Loading existing UF_Index.mat file\n') ; %% UF_Index = load ('UF_Index.mat') ; %% UF_Index = UF_Index.UF_Index ; - UF_Index = UFget ; + try + load UF_Index + fprintf ('loaded UF_Index in current directory:\n%s\n', pwd) ; + dir + catch + fprintf ('loading UF_Index=UFget\n') ; + UF_Index = UFget ; + end end % get the current UFkinds @@ -277,7 +287,7 @@ ffile = deblank (files {i}) ; fprintf ('Matrix %d: %s filesize %d\n', i, ffile, filesize (i)) ; if (filesize (i) == 9999999999) - fprintf ('skip this file\n') ; + fprintf ('skip this file (not found)\n') ; continue ; end end @@ -287,8 +297,17 @@ % load the matrices %------------------------------------------------------------------------------- -% metis (A,'col') fails with a seg fault for these matrices: -skip_metis = [850 858 1257 1258] ; +% updated Jan 2011. metis has too many bugs to be used reliably. +% +% % metis (A,'col') fails with a seg fault for these matrices: +% skip_metis = [850 858 1257 1258 2302] ; +% +% % fails in dmperm analysis +% skip_metis = [skip_metis 2337] ; +% +% % metis is exceedingly slow; try again later +% skip_all_metis = [2276 2277] ; +% skip_metis = [skip_metis skip_all_metis] ; % these matrices are known to be positive definite, and indefinite, % respectively, but sparse Cholesky fails (on a 4GB Penitum 4) on some of them: @@ -323,10 +342,12 @@ kinds {id} = Problem.kind ; - nometis = any (id == skip_metis) ; - if (nometis) - fprintf ('skip metis - will fail\n') ; - end +% updated Jan 2010. metis no longer used for matrices 2337 and following +% nometis = any (id == skip_metis) + any (id == skip_all_metis) ; +% if (nometis) +% fprintf ('skip metis on A''A- will fail\n') ; +% end + nometis = 2 ; fprintf ('%s/%s\n', UF_Index.Group {id}, UF_Index.Name {id}) ; @@ -392,6 +413,7 @@ UF_Index.amd_vnz (id) = stats.amd_vnz ; UF_Index.amd_rnz (id) = stats.amd_rnz ; + % Jan 2010: these results are no longer computed UF_Index.metis_lnz (id) = stats.metis_lnz ; UF_Index.metis_flops (id) = stats.metis_flops ; UF_Index.metis_vnz (id) = stats.metis_vnz ; diff --git a/UFcollection/UFpage.m b/UFcollection/UFpage.m index 97f5cb36f0..57b8fa0def 100644 --- a/UFcollection/UFpage.m +++ b/UFcollection/UFpage.m @@ -262,30 +262,88 @@ function UFpage (matrix, index, figures) fprintf (f, 'iso-8859-1"> %s sparse matrix \n', fullname) ; fprintf (f, '\n') ; +% matrix name and description +fprintf (f, 'Matrix: %s\n', fullname) ; +fprintf (f, '
Description: %s
\n', ptitle) ; + % Yifan Hu's medium-sized graph plot, for all matrices yifan_graphs = 'http://www.research.att.com/~yifanhu/GALLERY/GRAPHS/' ; yifan_thumb = [yifan_graphs 'GIF_THUMBNAIL/'] ; yifan_medium = [yifan_graphs 'GIF_SMALL/'] ; yname = strrep (fullname, '/', '@') ; + +% add the primary graph fprintf (f, '\n
') ; +fprintf (f, '
\n
\n') ; + +%------------------------------------------------------------------------------- +% link to matrix images +%------------------------------------------------------------------------------- + +fprintf (f, '\n\n\n') ; fprintf (f, ... - ' \n
\n\n', ...
+ '
\n', fullname, name) ;
-%
-%
-% fprintf (f, ...
- % '\n', ...
- % yifan_medium, yname, fullname, yifan_thumb, yname) ;
+psym = index.pattern_symmetry (id) == 1 ;
-% matrix name and description
-fprintf (f, 'Matrix: %s\n', fullname) ; -fprintf (f, '
Description: %s
\n', ptitle) ; +% add the graph of A+A', for a square matrix with unsymmetric pattern +if (m == n && ~psym) + fprintf (f, ... + '
\n') ; + +% add the caption for the primary graph +if (m == n && psym) + fprintf (f, ' \n(undirected graph drawing) \n') ; +else + fprintf (f, '(bipartite graph drawing) \n') ; +end + +% add the caption for the graph of A+A' +if (m == n && ~psym) + fprintf (f, '(graph drawing of A+A'') \n') ; +end + +% finalize the graph drawing table +fprintf (f, '\n\n
\n
\n\n') ; + +%------------------------------------------------------------------------------- +% download and informational links +%------------------------------------------------------------------------------- % link to UF collection -fprintf (f, '\n') ; + +% primary matrix image +fprintf (f, ' \n
\n\n') ; + +% ccspy, if it exists +if (do_scc) + fprintf (f, ' \n
') ; + fprintf (f, ... +'
- UF Sparse Matrix Collection\n') ; +fprintf (f, '
- Home page of the UF Sparse Matrix Collection\n') ; % link to group fprintf (f, '
- Matrix group: %s\n', grp) ; @@ -323,32 +381,6 @@ function UFpage (matrix, index, figures) fullname) ; fsize (f, [topdir 'RB/' fullname '.tar.gz']) ; -%------------------------------------------------------------------------------- -% link to images -%------------------------------------------------------------------------------- - -fprintf (f, '\n\n
\n', fullname, name) ;
-
-% dmspy, if it exists
-if (do_dmspy)
- fprintf (f, '\n\n', ...
- fullname, name) ;
-end
-
-% ccspy, if it exists
-if (do_scc)
- fprintf (f, '\n', ...
- fullname, name) ;
-end
-
-% gplot, if it exists
-if (do_gplot)
- fprintf (f, '') ; - fprintf (f, ... -'
\n', ...
- name, fullname, name) ;
-end
-
%-------------------------------------------------------------------------------
% table of matrix properties
%-------------------------------------------------------------------------------
diff --git a/UFcollection/UFstats.m b/UFcollection/UFstats.m
index f0f300d16c..9eaf9ab2ce 100644
--- a/UFcollection/UFstats.m
+++ b/UFcollection/UFstats.m
@@ -5,8 +5,10 @@
%
% A: a sparse matrix
% kind: a string with the Problem.kind
-% nometis: if nonzero then metis(A,'col') is not used, nor is metis used in the
-% dmperm+ ordering.
+% nometis:
+% 0: always try METIS
+% 1: use METIS only for A (not for A'A, and not for dmperm)
+% 2: do not use METIS at all
% Z: empty, or a sparse matrix the same size as A. Only used for psym and
% nzero statistics, described below.
%
@@ -60,6 +62,7 @@
if (nargin < 3)
nometis = 0 ;
end
+fprintf ('UFstats. nometis: %d\n', nometis) ;
%-------------------------------------------------------------------------------
% ensure the matrix is sparse
@@ -282,20 +285,22 @@
fprintf ('AMD lnz %d flops %g time: %g\n', ...
stats.amd_lnz, stats.amd_flops, toc) ;
- % order the whole matrix with METIS
- tic ;
- try
- p = metis (C) ;
- c = symbfact (C (p,p)) ;
- stats.metis_lnz = sum (c) ;
- stats.metis_flops = sum (c.^2) ;
- catch
- disp (lasterr) ;
- fprintf ('metis failed\n') ;
+ % order the whole matrix with METIS (A+A')
+ if (nometis <= 1)
+ tic ;
+ try
+ p = metis (C) ;
+ c = symbfact (C (p,p)) ;
+ stats.metis_lnz = sum (c) ;
+ stats.metis_flops = sum (c.^2) ;
+ catch
+ disp (lasterr) ;
+ fprintf ('metis failed\n') ;
+ end
+ clear p c C
+ fprintf ('METIS lnz %d flops %g time: %g\n', ...
+ stats.metis_lnz, stats.metis_flops, toc) ;
end
- clear p c C
- fprintf ('METIS lnz %d flops %g time: %g\n', ...
- stats.metis_lnz, stats.metis_flops, toc) ;
else
@@ -330,7 +335,7 @@
% order entire matrix with METIS, for LU bounds
%-------------------------------------------------------------------------------
-if (~nometis)
+if (nometis == 0)
try
q = metis (A, 'col') ;
[vnz,rnz] = cs_sqr (A (:,q)) ;
@@ -530,9 +535,11 @@
% all blocks are square, analyze a square block
% best of amd and metis
- if (nometis)
+ if (nometis > 0)
+ % use just AMD
[pblock c] = analyze (S|S', 'sym', 1) ;
else
+ % use AMD and METIS
[pblock c] = analyze (S|S', 'sym', 3) ;
end
lnzblock = sum (c) ;
@@ -557,9 +564,11 @@
end
% best of amd and metis
try
- if (nometis)
+ if (nometis > 0)
+ % use just AMD
pblock = analyze (S, 'col', 1) ;
else
+ % use AMD and METIS
pblock = analyze (S, 'col', 3) ;
end
catch
diff --git a/UFconfig/Makefile b/UFconfig/Makefile
index 420a530c07..b8ef789a45 100644
--- a/UFconfig/Makefile
+++ b/UFconfig/Makefile
@@ -1,6 +1,6 @@
# Makefile for UFconfig.c
-VERSION = 3.5.0
+VERSION = 3.6.0
default: ccode
@@ -25,8 +25,8 @@ clean:
# install UFconfig
install:
- $(CP) Lib/libufconfig.a $(INSTALL_LIB)/libufconfig.$(VERSION).a
- ( cd $(INSTALL_LIB) ; ln -s libufconfig.$(VERSION).a libufconfig.a )
+ $(CP) libufconfig.a $(INSTALL_LIB)/libufconfig.$(VERSION).a
+ ( cd $(INSTALL_LIB) ; ln -sf libufconfig.$(VERSION).a libufconfig.a )
$(CP) UFconfig.h $(INSTALL_INCLUDE)
# uninstall UFconfig
diff --git a/UFconfig/UFconfig.h b/UFconfig/UFconfig.h
index 5b39627a38..204e55105e 100644
--- a/UFconfig/UFconfig.h
+++ b/UFconfig/UFconfig.h
@@ -109,23 +109,23 @@ void *UFfree /* always returns NULL */
* version of SuiteSparse, with another package from another version of
* SuiteSparse, may or may not work.
*
- * SuiteSparse Version 3.5.0 contains the following packages:
+ * SuiteSparse Version 3.6.0 contains the following packages:
*
- * AMD version 2.2.1
- * BTF version 1.1.1
- * CAMD version 2.2.1
- * CCOLAMD version 2.7.2
- * CHOLMOD version 1.7.2
- * COLAMD version 2.7.2
- * CSparse version 2.2.4
- * CXSparse version 2.2.4
- * KLU version 1.1.1
- * LDL version 2.0.2
- * RBio version 2.0.0
- * SuiteSparseQR version 1.2.0
- * UFcollection version 1.3.0
+ * AMD version 2.2.2
+ * BTF version 1.1.2
+ * CAMD version 2.2.2
+ * CCOLAMD version 2.7.3
+ * CHOLMOD version 1.7.3
+ * COLAMD version 2.7.3
+ * CSparse version 2.2.5
+ * CXSparse version 2.2.5
+ * KLU version 1.1.2
+ * LDL version 2.0.3
+ * RBio version 2.0.1
+ * SPQR version 1.2.1 (also called SuiteSparseQR)
+ * UFcollection version 1.4.0
* UFconfig version number is the same as SuiteSparse
- * UMFPACK version 5.5.0
+ * UMFPACK version 5.5.1
* LINFACTOR version 1.1.0
* MESHND version 1.1.1
* SSMULT version 2.0.2
@@ -137,11 +137,11 @@ void *UFfree /* always returns NULL */
* METIS 4.0.1 required by CHOLMOD (optional) and KLU (optional)
*/
-#define SUITESPARSE_DATE "Nov 30, 2009"
+#define SUITESPARSE_DATE "Jan 25, 2011"
#define SUITESPARSE_VER_CODE(main,sub) ((main) * 1000 + (sub))
#define SUITESPARSE_MAIN_VERSION 3
-#define SUITESPARSE_SUB_VERSION 5
-#define SUITESPARSE_SUBSUB_VERSION 1
+#define SUITESPARSE_SUB_VERSION 6
+#define SUITESPARSE_SUBSUB_VERSION 0
#define SUITESPARSE_VERSION \
SUITESPARSE_VER_CODE(SUITESPARSE_MAIN_VERSION,SUITESPARSE_SUB_VERSION)
diff --git a/UFconfig/UFconfig.mk b/UFconfig/UFconfig.mk
index c6974d9ac7..60c951b6a9 100644
--- a/UFconfig/UFconfig.mk
+++ b/UFconfig/UFconfig.mk
@@ -90,17 +90,22 @@ INSTALL_INCLUDE = /usr/local/include
# These settings will probably not work, since there is no fixed convention for
# naming the BLAS and LAPACK library (*.a or *.so) files.
-# Using the Goto BLAS:
-# BLAS = -lgoto -lgfortran -lgfortranbegin -lg2c
-
# This is probably slow ... it might connect to the Standard Reference BLAS:
-BLAS = -lblas
+BLAS = -lblas -lgfortran
LAPACK = -llapack
+# NOTE: this next option for the "Goto BLAS" has nothing to do with a "goto"
+# statement. Rather, the Goto BLAS is written by Dr. Kazushige Goto.
+# Using the Goto BLAS:
+# BLAS = -lgoto -lgfortran -lgfortranbegin
+
# Using non-optimized versions:
-# BLAS = -lblas_plain -lgfortran -lgfortranbegin -lg2c
+# BLAS = -lblas_plain -lgfortran -lgfortranbegin
# LAPACK = -llapack_plain
+# BLAS = -lblas_plain -lgfortran -lgfortranbegin
+# LAPACK = -llapack
+
# The BLAS might not contain xerbla, an error-handling routine for LAPACK and
# the BLAS. Also, the standard xerbla requires the Fortran I/O library, and
# stops the application program if an error occurs. A C version of xerbla
diff --git a/UMFPACK/Demo/Makefile b/UMFPACK/Demo/Makefile
index 0626aa19cb..8213aae249 100644
--- a/UMFPACK/Demo/Makefile
+++ b/UMFPACK/Demo/Makefile
@@ -40,12 +40,12 @@ LIBS = $(BLAS) $(XERBLA) $(LIB)
../../CAMD/Lib/libcamd.a:
( cd ../../CAMD ; $(MAKE) library )
-../../metis-4.0/libmetis.a:
- ( cd ../../metis-4.0/Lib ; $(MAKE) )
+$(METIS):
+ ( cd $(METIS_PATH) ; $(MAKE) )
UMFPACK = ../Lib/libumfpack.a ../../AMD/Lib/libamd.a \
../../CHOLMOD/Lib/libcholmod.a ../../COLAMD/Lib/libcolamd.a \
- ../../CAMD/Lib/libcamd.a ../../metis-4.0/libmetis.a \
+ ../../CAMD/Lib/libcamd.a $(METIS) \
../../CCOLAMD/Lib/libccolamd.a
libs: $(UMFPACK)
diff --git a/UMFPACK/Demo/tmp/.gitkeep b/UMFPACK/Demo/tmp/.gitkeep
deleted file mode 100644
index e69de29bb2..0000000000
diff --git a/UMFPACK/Demo/umfpack_di_demo.out b/UMFPACK/Demo/umfpack_di_demo.out
index 585f7c55ed..67001497ef 100644
--- a/UMFPACK/Demo/umfpack_di_demo.out
+++ b/UMFPACK/Demo/umfpack_di_demo.out
@@ -1,5 +1,5 @@
-UMFPACK V5.5 (Nov 30, 2009) demo: _di_ version
+UMFPACK V5.5 (Jan 25, 2011) demo: _di_ version
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
@@ -39,9 +39,9 @@ UMFPACK License:
Availability: http://www.cise.ufl.edu/research/sparse/umfpack
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
-UMFPACK V5.5.0 (Nov 30, 2009), Control:
+UMFPACK V5.5.1 (Jan 25, 2011), Control:
Matrix entry defined as: double
Int (generic integer) defined as: int
@@ -72,7 +72,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Control:
8: BLAS library used: Fortran BLAS. size of BLAS integer: 4
compiled for ANSI C
CPU timer is POSIX times ( ) routine.
- computer/operating system: Mac
+ computer/operating system: Linux
size of int: 4 UF_long: 8 Int: 4 pointer: 8 double: 8 Entry: 8 (in bytes)
@@ -192,7 +192,7 @@ Numeric factorization of A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -280,7 +280,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -354,7 +354,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -371,7 +371,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
x (solution of Ax=b): dense vector, n = 5.
@@ -387,7 +387,7 @@ maxnorm of residual: 1.06581e-14
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
determinant: (1.14) * 10^(2)
@@ -401,7 +401,7 @@ x (solution of Ax=b, solve is split into 3 steps): dense vector, n = 5.
maxnorm of residual: 1.06581e-14
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -475,7 +475,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -545,7 +545,7 @@ Numeric factorization of modified A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -632,7 +632,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -706,7 +706,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 8
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -794,7 +794,7 @@ Numeric factorization of completely modified A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -882,7 +882,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -956,7 +956,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1095,7 +1095,7 @@ Numeric factorization of C: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 3
number of entries stored in L (excl diag): 2
@@ -1276,7 +1276,7 @@ Freeing numeric object:
Loading numeric object:
Done loading numeric object
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -1350,7 +1350,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1377,7 +1377,7 @@ maxnorm of residual: 3.55271e-15
Solving C'x=b again, using umfpack_di_wsolve instead:
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -1451,7 +1451,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
diff --git a/UMFPACK/Demo/umfpack_dl_demo.out b/UMFPACK/Demo/umfpack_dl_demo.out
index 93f788f4ab..59079ea014 100644
--- a/UMFPACK/Demo/umfpack_dl_demo.out
+++ b/UMFPACK/Demo/umfpack_dl_demo.out
@@ -1,5 +1,5 @@
-UMFPACK V5.5 (Nov 30, 2009) demo: _dl_ version
+UMFPACK V5.5 (Jan 25, 2011) demo: _dl_ version
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
@@ -39,9 +39,9 @@ UMFPACK License:
Availability: http://www.cise.ufl.edu/research/sparse/umfpack
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
-UMFPACK V5.5.0 (Nov 30, 2009), Control:
+UMFPACK V5.5.1 (Jan 25, 2011), Control:
Matrix entry defined as: double
Int (generic integer) defined as: UF_long
@@ -72,7 +72,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Control:
8: BLAS library used: Fortran BLAS. size of BLAS integer: 4
compiled for ANSI C
CPU timer is POSIX times ( ) routine.
- computer/operating system: Mac
+ computer/operating system: Linux
size of int: 4 UF_long: 8 Int: 8 pointer: 8 double: 8 Entry: 8 (in bytes)
@@ -192,7 +192,7 @@ Numeric factorization of A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -280,7 +280,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -354,7 +354,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -371,7 +371,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
x (solution of Ax=b): dense vector, n = 5.
@@ -387,7 +387,7 @@ maxnorm of residual: 1.06581e-14
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
determinant: (1.14) * 10^(2)
@@ -401,7 +401,7 @@ x (solution of Ax=b, solve is split into 3 steps): dense vector, n = 5.
maxnorm of residual: 1.06581e-14
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -475,7 +475,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -545,7 +545,7 @@ Numeric factorization of modified A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -632,7 +632,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -706,7 +706,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 8
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -794,7 +794,7 @@ Numeric factorization of completely modified A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -882,7 +882,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -956,7 +956,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1095,7 +1095,7 @@ Numeric factorization of C: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 3
number of entries stored in L (excl diag): 2
@@ -1276,7 +1276,7 @@ Freeing numeric object:
Loading numeric object:
Done loading numeric object
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -1350,7 +1350,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1377,7 +1377,7 @@ maxnorm of residual: 3.55271e-15
Solving C'x=b again, using umfpack_dl_wsolve instead:
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -1451,7 +1451,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1478,4 +1478,4 @@ maxnorm of residual: 3.55271e-15
umfpack_dl_demo complete.
-Total time: 0.00 seconds (CPU time), 0.01 seconds (wallclock time)
+Total time: 0.01 seconds (CPU time), 0.00 seconds (wallclock time)
diff --git a/UMFPACK/Demo/umfpack_zi_demo.out b/UMFPACK/Demo/umfpack_zi_demo.out
index 673e11994e..0b45dc8ae5 100644
--- a/UMFPACK/Demo/umfpack_zi_demo.out
+++ b/UMFPACK/Demo/umfpack_zi_demo.out
@@ -1,5 +1,5 @@
-UMFPACK V5.5 (Nov 30, 2009) demo: _zi_ version
+UMFPACK V5.5 (Jan 25, 2011) demo: _zi_ version
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
@@ -39,9 +39,9 @@ UMFPACK License:
Availability: http://www.cise.ufl.edu/research/sparse/umfpack
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
-UMFPACK V5.5.0 (Nov 30, 2009), Control:
+UMFPACK V5.5.1 (Jan 25, 2011), Control:
Matrix entry defined as: double complex
Int (generic integer) defined as: int
@@ -72,7 +72,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Control:
8: BLAS library used: Fortran BLAS. size of BLAS integer: 4
compiled for ANSI C
CPU timer is POSIX times ( ) routine.
- computer/operating system: Mac
+ computer/operating system: Linux
size of int: 4 UF_long: 8 Int: 4 pointer: 8 double: 8 Entry: 16 (in bytes)
@@ -192,7 +192,7 @@ Numeric factorization of A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -280,7 +280,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -354,7 +354,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -371,7 +371,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
x (solution of Ax=b): dense vector, n = 5.
@@ -387,7 +387,7 @@ maxnorm of residual: 1.77636e-15
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
determinant: (-1.7814+ (2.3784)i) * 10^(2)
@@ -401,7 +401,7 @@ x (solution of Ax=b, solve is split into 3 steps): dense vector, n = 5.
maxnorm of residual: 1.77636e-14
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -475,7 +475,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -545,7 +545,7 @@ Numeric factorization of modified A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 3
number of entries stored in L (excl diag): 1
@@ -632,7 +632,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -706,7 +706,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 8
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -794,7 +794,7 @@ Numeric factorization of completely modified A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -882,7 +882,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -956,7 +956,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1095,7 +1095,7 @@ Numeric factorization of C: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 3
number of entries stored in L (excl diag): 2
@@ -1276,7 +1276,7 @@ Freeing numeric object:
Loading numeric object:
Done loading numeric object
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -1350,7 +1350,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1377,7 +1377,7 @@ maxnorm of residual: 4.88498e-15
Solving C'x=b again, using umfpack_zi_wsolve instead:
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: int
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -1451,7 +1451,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
diff --git a/UMFPACK/Demo/umfpack_zl_demo.out b/UMFPACK/Demo/umfpack_zl_demo.out
index a612739423..81d7f716be 100644
--- a/UMFPACK/Demo/umfpack_zl_demo.out
+++ b/UMFPACK/Demo/umfpack_zl_demo.out
@@ -1,5 +1,5 @@
-UMFPACK V5.5 (Nov 30, 2009) demo: _zl_ version
+UMFPACK V5.5 (Jan 25, 2011) demo: _zl_ version
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
@@ -39,9 +39,9 @@ UMFPACK License:
Availability: http://www.cise.ufl.edu/research/sparse/umfpack
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
-UMFPACK V5.5.0 (Nov 30, 2009), Control:
+UMFPACK V5.5.1 (Jan 25, 2011), Control:
Matrix entry defined as: double complex
Int (generic integer) defined as: UF_long
@@ -72,7 +72,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Control:
8: BLAS library used: Fortran BLAS. size of BLAS integer: 4
compiled for ANSI C
CPU timer is POSIX times ( ) routine.
- computer/operating system: Mac
+ computer/operating system: Linux
size of int: 4 UF_long: 8 Int: 8 pointer: 8 double: 8 Entry: 16 (in bytes)
@@ -192,7 +192,7 @@ Numeric factorization of A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -280,7 +280,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -354,7 +354,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -371,7 +371,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
x (solution of Ax=b): dense vector, n = 5.
@@ -387,7 +387,7 @@ maxnorm of residual: 1.77636e-15
UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.
-UMFPACK V5.5.0 (Nov 30, 2009): OK
+UMFPACK V5.5.1 (Jan 25, 2011): OK
determinant: (-1.7814+ (2.3784)i) * 10^(2)
@@ -401,7 +401,7 @@ x (solution of Ax=b, solve is split into 3 steps): dense vector, n = 5.
maxnorm of residual: 1.77636e-14
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -475,7 +475,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -545,7 +545,7 @@ Numeric factorization of modified A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 3
number of entries stored in L (excl diag): 1
@@ -632,7 +632,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -706,7 +706,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 8
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -794,7 +794,7 @@ Numeric factorization of completely modified A: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 4
number of entries stored in L (excl diag): 2
@@ -882,7 +882,7 @@ diagonal of U: dense vector, n = 5.
Numeric object: OK
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -956,7 +956,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1095,7 +1095,7 @@ Numeric factorization of C: Numeric object:
largest actual frontal matrix size: 4
memory defragmentations: 1
memory reallocations: 1
- costly memory reallocations: 1
+ costly memory reallocations: 0
entries in compressed pattern (L and U): 2
number of nonzeros in L (excl diag): 3
number of entries stored in L (excl diag): 2
@@ -1276,7 +1276,7 @@ Freeing numeric object:
Loading numeric object:
Done loading numeric object
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -1350,7 +1350,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1377,7 +1377,7 @@ maxnorm of residual: 4.88498e-15
Solving C'x=b again, using umfpack_zl_wsolve instead:
-UMFPACK V5.5.0 (Nov 30, 2009), Info:
+UMFPACK V5.5.1 (Jan 25, 2011), Info:
matrix entry defined as: double complex
Int (generic integer) defined as: UF_long
BLAS library used: Fortran BLAS. size of BLAS integer: 4
@@ -1451,7 +1451,7 @@ UMFPACK V5.5.0 (Nov 30, 2009), Info:
numerical values stored in Numeric object: 9
numeric factorization defragmentations: 1
numeric factorization reallocations: 1
- costly numeric factorization reallocations: 1
+ costly numeric factorization reallocations: 0
numeric factorization CPU time (sec): 0.00
numeric factorization wallclock time (sec): 0.00
@@ -1478,4 +1478,4 @@ maxnorm of residual: 4.88498e-15
umfpack_zl_demo complete.
-Total time: 0.00 seconds (CPU time), 0.00 seconds (wallclock time)
+Total time: 0.00 seconds (CPU time), 0.01 seconds (wallclock time)
diff --git a/UMFPACK/Doc/ChangeLog b/UMFPACK/Doc/ChangeLog
index ff25add9f4..e1c5e7ab0c 100644
--- a/UMFPACK/Doc/ChangeLog
+++ b/UMFPACK/Doc/ChangeLog
@@ -1,3 +1,8 @@
+Jan 25, 2011, version 5.5.1
+
+ * minor fix to "make install"
+ * Demo/Makefile updated, to allow for METIS to be in a non-default path
+
Nov 30, 2009, version 5.5.0
* changed and simplified how UMFPACK connects to 64-bit BLAS
diff --git a/UMFPACK/Doc/QuickStart.pdf b/UMFPACK/Doc/QuickStart.pdf
index 1aab56aaf6..43e68ca213 100644
Binary files a/UMFPACK/Doc/QuickStart.pdf and b/UMFPACK/Doc/QuickStart.pdf differ
diff --git a/UMFPACK/Doc/QuickStart.tex b/UMFPACK/Doc/QuickStart.tex
index 972e36074f..6e386a627d 100644
--- a/UMFPACK/Doc/QuickStart.tex
+++ b/UMFPACK/Doc/QuickStart.tex
@@ -18,8 +18,8 @@
\author{Timothy A. Davis \\
Dept. of Computer and Information Science and Engineering \\
Univ. of Florida, Gainesville, FL}
-\title{UMFPACK Version 5.5.0 Quick Start Guide}
-\date{Nov 30, 2009}
+\title{UMFPACK Quick Start Guide}
+\date{VERSION 5.5.1, Jan 25, 2011}
\maketitle
%-------------------------------------------------------------------------------
diff --git a/UMFPACK/Doc/UserGuide.pdf b/UMFPACK/Doc/UserGuide.pdf
index aa920df841..a13ff9c56e 100644
Binary files a/UMFPACK/Doc/UserGuide.pdf and b/UMFPACK/Doc/UserGuide.pdf differ
diff --git a/UMFPACK/Doc/UserGuide.stex b/UMFPACK/Doc/UserGuide.stex
index 5acf844115..9743f3f975 100644
--- a/UMFPACK/Doc/UserGuide.stex
+++ b/UMFPACK/Doc/UserGuide.stex
@@ -20,8 +20,8 @@
\author{Timothy A. Davis \\
Dept. of Computer and Information Science and Engineering \\
Univ. of Florida, Gainesville, FL}
-\title{UMFPACK Version 5.5.0 User Guide}
-\date{Nov 30, 2009}
+\title{UMFPACK User Guide}
+\date{VERSION 5.5.1, Jan 25, 2011}
\maketitle
%-------------------------------------------------------------------------------
diff --git a/UMFPACK/Include/umfpack.h b/UMFPACK/Include/umfpack.h
index 400a48c984..65360ebf24 100644
--- a/UMFPACK/Include/umfpack.h
+++ b/UMFPACK/Include/umfpack.h
@@ -97,7 +97,7 @@ extern "C" {
/* Version, copyright, and license */
/* -------------------------------------------------------------------------- */
-#define UMFPACK_VERSION "UMFPACK V5.5.0 (Nov 30, 2009)"
+#define UMFPACK_VERSION "UMFPACK V5.5.1 (Jan 25, 2011)"
#define UMFPACK_COPYRIGHT \
"UMFPACK: Copyright (c) 2005-2009 by Timothy A. Davis. All Rights Reserved.\n"
@@ -163,11 +163,11 @@ extern "C" {
* above.
*/
-#define UMFPACK_DATE "Nov 30, 2009"
+#define UMFPACK_DATE "Jan 25, 2011"
#define UMFPACK_VER_CODE(main,sub) ((main) * 1000 + (sub))
#define UMFPACK_MAIN_VERSION 5
#define UMFPACK_SUB_VERSION 5
-#define UMFPACK_SUBSUB_VERSION 0
+#define UMFPACK_SUBSUB_VERSION 1
#define UMFPACK_VER UMFPACK_VER_CODE(UMFPACK_MAIN_VERSION,UMFPACK_SUB_VERSION)
/* -------------------------------------------------------------------------- */
diff --git a/UMFPACK/Makefile b/UMFPACK/Makefile
index a5f39980ee..5bc5a1cfa4 100644
--- a/UMFPACK/Makefile
+++ b/UMFPACK/Makefile
@@ -2,7 +2,7 @@
# UMFPACK makefile (for GNU make or original make)
#-------------------------------------------------------------------------------
-VERSION = 5.5.0
+VERSION = 5.5.1
# UMFPACK requires the AMD package to be in ../AMD
@@ -78,7 +78,7 @@ cov: purge
# install UMFPACK
install:
$(CP) Lib/libumfpack.a $(INSTALL_LIB)/libumfpack.$(VERSION).a
- ( cd $(INSTALL_LIB) ; ln -s libumfpack.$(VERSION).a libumfpack.a )
+ ( cd $(INSTALL_LIB) ; ln -sf libumfpack.$(VERSION).a libumfpack.a )
$(CP) Include/*.h $(INSTALL_INCLUDE)
# uninstall UMFPACK
diff --git a/UMFPACK/README.txt b/UMFPACK/README.txt
index 5375c98cd4..ab1e95fa59 100644
--- a/UMFPACK/README.txt
+++ b/UMFPACK/README.txt
@@ -1,3 +1,5 @@
+UMFPACK, Copyright (c) 1995-2011 by Timothy A. Davis. All Rights Reserved.
+
UMFPACK : a set of routines solving sparse linear systems via LU
factorization. Requires three other packages: the BLAS (dense matrix
operations), AMD (sparse matrix minimum degree ordering), and UFconfig.
@@ -34,7 +36,6 @@ Quick start (for MATLAB users):
--------------------------------------------------------------------------------
-UMFPACK, Copyright (c) 1995-2009 by Timothy A. Davis. All Rights Reserved.
UMFPACK is available under alternate licences; contact T. Davis for details.
UMFPACK License:
