g95 vs. gfortran

[Victor Zhorin]

I have a simple linear programming subroutine that compiles and executes
under both gfortran and g95. However, the results produced are dramatically
different with g95 working well and gfortran not working at all.

For a different test case gfortran sometimes does produce results that are
close to correct ones, but it still is unreliable at some points.

I am very perplexed and I've already tried all obvious options that could
possibly resolve this. It does not make any sense.

Any ideas?

-- 
Victor V. Zhorin