There is a single input file to test all drivers. The input data for each path (testing xGGEV, xGGES, xGGEVX and xGGESX) is preceeded by a single line identifying the path (SGV, SGS, SVX and SGX, respectively, when x=S, and CGV, CGS, CXV and CGX, respectively, when x=C). We discuss each set of input data in turn.

An annotated example of input data for testing SGGEV is shown below (testing CGGEV is identical except CGV replaces SGV):

SGV Data for the Real Nonsymmetric Eigenvalue Problem Driver 6 Number of matrix dimensions 2 6 8 10 15 20 Matrix dimensions 1 1 1 2 1 Parameters NB, NBMIN, NXOVER, NS, NBCOL 10 Threshold value .FALSE. Put .TRUE. to test the error exits 0 Code to interpret the seed SGV 26 Test all 26 matrix types

The first line must contain the characters SGV in columns 1-3. The remaining lines are read using list-directed input and specify the following values:

line 2: | The number of values of N |

line 3: | The values of N, the matrix dimension |

line 4: | The values of the parameters NB, NBMIN, NXOVER, NS and NBCOL |

line 5: | The threshold value THRESH for the test ratios |

line 6: | T to test the error exits |

line 7: | An integer code to interpret the random number seed |

= 0: Set the seed to a default value before each run | |

= 1: Initialize the seed to a default value only before the first run | |

= 2: Like 1, but use the seed values on the next line | |

line 8: | If line 7 was 2, four integer values for the random number seed |

line 9: | Contains `SGV' in columns 1-3, followed by the number of matrix types |

(an integer from 0 to 26) | |

line 10: | (and following) if the number of matrix types is at least one and less than 26, |

a list of integers between 1 and 26 indicating which matrix types are to be tested. |

The input data for testing xGGES has the same format as for xGGEV, except SGS replaces SGV when testing SGGES, and CGS replaces CGV when testing CGGES.

The input data for testing xGGEVX consists of two files. An annotated example of the first type is listed below

SVX Data for the Real Nonsymmetric Eigenvalue Expert Driver 5 Largest matrix dimension 1 1 1 2 1 Parameters NB, NBMIN, NXOVER, NS, NBCOL 10 Threshold for test ratios .FALSE. Put .TRUE. to test the error exits 0 Code to interpret the seed

The first line must contain the characters SXV in columns 1-3. The remaining lines are read using list-directed input and specify the following values:

line 2: | The number of values of N |

line 3: | The values of the parameters NB, NBMIN, NXOVER, NS and NBCOL |

line 4: | The threshold value THRESH for the test ratios |

line 5: | T to test the error exits |

line 6: | An integer code to interpret the random number seed |

= 0: Set the seed to a default value before each run | |

= 1: Initialize the seed to a default value only before the first run | |

= 2: Like 1, but use the seed values on the next line |

The second consists of precomputed data for testing the eigenvalue/eigenvector condition estimation routines. It has the same format as above, except that on line 2, a zero is entered as the matrix size and on line 7 and the following, there is precomputed data. Each example is stored on 2*N+3 lines, where N is its dimension ( 2*N*N+3 lines for complex data). The first line contains the dimension (a single integer). The next N lines contain the matrix 16#16, one row per line. The next N lines contain the matrix 97#97. The next line contains the reciprocals of the eigenvalue condition numbers. The last line contains the reciprocals of the eigenvector condition numbers. The end of data is indicated by dimension N = 0. Even if no data is to be tested, there must be at least one line containing N = 0.

The input data for testing xGGESX consists of two parts. The first part is identical to that for xGGEVX (using SGX instead of SVX and CGX instead of CVX). The second consists of precomputed data for testing the eigenvalue/eigenvector condition estimation routines. Again, it has the same format as the second part for xGGEVX, except that each example is stored on 2*N+3 lines, where N is its dimension ( 2*N*N+3 lines for complex data). The first line contains the dimension (a single integer). The next line contains an integer k such that only the last k eigenvalues will be selected and appear in the leading diagonal blocks of 16#16 and 97#97. The next N lines contain the matrix 16#16, one row per line. The next N lines contain the matrix 97#97. The last line contains the reciprocal of the eigenvalue cluster condition number and the reciprocal of the deflating subspace (associated with selected eigencluster) condition number. The end of data is indicated by dimension N = 0. Even if no data is to be tested, there must be at least one line containing N = 0.