[Help-glpk] different results when complied by different version of gcc (fwd)

[Shubhashis Ghosh]

Hi,
   Can anyone tell me what the problem is ? Look upto the end 
of the mail. Please..... say what could be the problem with gcc version 
3.0.4.

-Shu'

   Using glpk-4.7, when I complile it using gcc version 3.0.4 this gives 
the following sequence of results ... I think a cycling in simplex pivot 
step is occuring.

lpx_read_mps: reading problem data from 
`/usr/tees2/cshome/shubhash/integer/glpk-4.7/glpk-4.7/check.mps'...
lpx_read_mps: problem PROBLEM
lpx_read_mps: 1046 rows, 12497 columns, 441030 non-zeros
lpx_read_mps: 12497 integer columns, all of which are binary
lpx_read_mps: 237621 cards were read
lpx_simplex: original LP has 1046 rows, 12497 columns, 441030 non-zeros
lpx_simplex: presolved LP has 1045 rows, 12497 columns, 428533 non-zeros
lpx_adv_basis: size of triangular part = 1045
      0:   objval =   0.000000000e+00   infeas =   1.000000000e+00 (0)
    200:   objval =   4.503364659e+08   infeas =   3.491959209e-02 (0)
    394:   objval =   4.503364659e+08   infeas =   7.183576261e-17 (0)
*   394:   objval =   4.503364659e+08   infeas =   1.697604436e-14 (0)
*   400:   objval =   4.497583912e+08   infeas =   1.543210004e-14 (0)
*   600:   objval =   4.269070296e+08   infeas =   3.223112259e-14 (0)
*   800:   objval =   4.226386563e+08   infeas =   1.071831625e-15 (0)
*  1000:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  1200:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  1400:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  1600:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  1800:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  2000:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  2200:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  2400:   objval =   4.225723474e+08   infeas =   5.561107130e-13 (0)
*  2600:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  2800:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  3000:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  3200:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  3400:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  3600:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  3800:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  4000:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  4200:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  4400:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  4600:   objval =   4.225723474e+08   infeas =   5.733191699e-13 (0)
*  4800:   objval =   4.225723474e+08   infeas =   5.733191699e-13 (0)
*  5000:   objval =   4.225723474e+08   infeas =   5.733191699e-13 (0)
*  5200:   objval =   4.225723474e+08   infeas =   5.733191699e-13 (0)
*  5400:   objval =   4.225723474e+08   infeas =   5.733191699e-13 (0)
*  5600:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  5800:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  6000:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  6200:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  6400:   objval =   4.225723474e+08   infeas =   5.719869023e-13 (0)
*  6600:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  6800:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  7000:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  7200:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  7400:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  7600:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  7800:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  8000:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  8200:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  8400:   objval =   4.225723474e+08   infeas =   4.890532423e-13 (0)
*  8600:   objval =   4.225723474e+08   infeas =   5.733191699e-13 (0)
*  8800:   objval =   4.225723474e+08   infeas =   5.733191699e-13 (0)
*  9000:   objval =   4.225723474e+08   infeas =   5.733191699e-13 (0)
...... and it continues. 

But when I complie it with gcc version 3.2.2, it gives optimal solution as 
follows.

lpx_read_mps: reading problem data from 
`/usr/tees2/cshome/shubhash/integer/glpk-4.7/glpk-4.7/check.mps'...
lpx_read_mps: problem PROBLEM
lpx_read_mps: 1046 rows, 12497 columns, 441030 non-zeros
lpx_read_mps: 12497 integer columns, all of which are binary
lpx_read_mps: 237621 cards were read
lpx_simplex: original LP has 1046 rows, 12497 columns, 441030 non-zeros
lpx_simplex: presolved LP has 1045 rows, 12497 columns, 428533 non-zeros
lpx_adv_basis: size of triangular part = 1045
      0:   objval =   0.000000000e+00   infeas =   1.000000000e+00 (0)
    200:   objval =   4.503364659e+08   infeas =   4.792937564e-02 (0)
    400:   objval =   4.503364659e+08   infeas =   1.148140540e-03 (0)
    425:   objval =   4.503364659e+08   infeas =   9.029227718e-18 (0)
*   425:   objval =   4.503364659e+08   infeas =   2.133764086e-15 (0)
*   600:   objval =   4.275084659e+08   infeas =   9.094905499e-15 (0)
*   800:   objval =   4.225774975e+08   infeas =   1.110223025e-15 (0)
*   814:   objval =   4.225723474e+08   infeas =   2.109423747e-15 (0)
OPTIMAL SOLUTION FOUND
Integer optimization begins...