1 2 SUBROUTINE RSG(NM,N,A,B,W,MATZ,Z,FV1,FV2,IERR) 3C***BEGIN PROLOGUE RSG 4C***DATE WRITTEN 760101 (YYMMDD) 5C***REVISION DATE 830518 (YYMMDD) 6C***CATEGORY NO. D4B1 7C***KEYWORDS EIGENVALUES,EIGENVECTORS,EISPACK 8C***AUTHOR SMITH, B. T., ET AL. 9C***PURPOSE Computes eigenvalues and, optionally, eigenvectors of 10C symmetric generalized eigenproblem: A*X=(LAMBDA)*B*X 11C***DESCRIPTION 12C 13C This subroutine calls the recommended sequence of 14C subroutines from the eigensystem subroutine package (EISPACK) 15C To find the eigenvalues and eigenvectors (if desired) 16C for the REAL SYMMETRIC generalized eigenproblem Ax = (LAMBDA)Bx. 17C 18C On Input 19C 20C NM must be set to the row dimension of the two-dimensional 21C array parameters as declared in the calling program 22C dimension statement. 23C 24C N is the order of the matrices A and B. 25C 26C A contains a real symmetric matrix. 27C 28C B contains a positive definite real symmetric matrix. 29C 30C MATZ is an integer variable set equal to zero if 31C only eigenvalues are desired. Otherwise it is set to 32C any non-zero integer for both eigenvalues and eigenvectors. 33C 34C On Output 35C 36C W contains the eigenvalues in ascending order. 37C 38C Z contains the eigenvectors if MATZ is not ZERO. 39C 40C IERR is an integer output variable set equal to an 41C error completion code described in section 2B of the 42C documentation. The normal completion code is zero. 43C 44C FV1 and FV2 are temporary storage arrays. 45C 46C Questions and comments should be directed to B. S. Garbow, 47C APPLIED MATHEMATICS DIVISION, ARGONNE NATIONAL LABORATORY 48C ------------------------------------------------------------------ 49C***REFERENCES B. T. SMITH, J. M. BOYLE, J. J. DONGARRA, B. S. GARBOW, 50C Y. IKEBE, V. C. KLEMA, C. B. MOLER, *MATRIX EIGEN- 51C SYSTEM ROUTINES - EISPACK GUIDE*, SPRINGER-VERLAG, 52C 1976. 53C***ROUTINES CALLED REBAK,REDUC,TQL2,TQLRAT,TRED1,TRED2 54C***END PROLOGUE RSG 55 56 57