---------------------001--------------------------------
        sh-2.05b$ export MRMC="../../../bin/mrmc"
        sh-2.05b$ export TEST_NAME="csrl_discretiz_state_02"
        sh-2.05b$ ${MRMC} csrl ${TEST_NAME}.tra ${TEST_NAME}.lab ${TEST_NAME}.rew
        Running in the CSRL model checking mode.
        States=5, Transitions=14
        Space Occupied:: 712 Bytes.
        Type 'help' to get help.
        >>set d 0.03125
        >>P{>0.1}[ a U[0,25] tt ]
>>>     Overflow
        time to compute: 40 micro sec(s)
        Result: ( 0.0e+00, 0.0e+00, 0.0e+00, 0.0e+00, 0.0e+00 )
        states = {}

        >>P{>0.1}[ a U[0,25][0,110] tt ]
>>>>    Segmentation fault

---------------------002------NOT A BUG-----------------
In pctl_andova_state_01 test case we are getting for
  
        P{>0.1}[ tt U[0,10][0,9] sd ]

the following probabilities

        Result: ( 1.660156250000000e-01, 0.0e+00, 3.320312500000000e-01, 0.0e+00, 0.0e+00,
                   0.0e+00, 6.660156250000000e-01, 0.0e+00, 0.0e+00, 0.0e+00, 0.0e+00, 0.0e+00, 1.0e+00 )
        states = {1, 3, 7, 13, }

This is just wrong because staying in 13 for 10 time units will give you reward equal to
10 which is greater than 9!

The problem is in the "transient.c" file in the method:

double * dtmrm_bounded_until(bitset *phi, bitset *psi, float subi, float supi, float subj, float supj)

There

        //ERROR: This is totally wrong ! what if the node has a self loop !?
        //       or phi is valid here and we can go further to another psi node!?
        if( get1b.b &&
            ( subi <= 0 ) && ( supi >= 0 ) &&
            ( subj <= 0 ) && ( supj >= 0 ) )
        {
                result[i]=1.0;
                continue;
        }

---------------------003--------NOT A BUG------------------------
Some problems with iterative methods, like Gauss-Seidel and Gauss-Jacobi

See test/functional_tests/csl_steady_state_03/csl_steady_state_03.info for
more information.

---------------------004---------FIXED-----------------------
Test cases
    test/functional_tests/csl_steady_state_02
    test/functional_tests/csl_steady_state_03
    
PASS when MRMC is compiled with "-O0 -ggdb" options and FAIL if with "-O3"

---------------------005--------NOT A BUG------------------------
In example test/functional_tests/csl_steady_state_02
steady state probabilities for S{>0.1}[ a5 ] and S{>0.1}[ a6 ] are interchanged comparing to ETMCC 1.0 and PRISM 2.1.
NOTE: It is strange but steady state probabilities computed with Gauss-Seidel and Gauss-Jacobi interchange in
ETMCC 1.0 and PRISM 2.1 !!!! In our implementation we have smth like in PRISM 2.1 (see bug 003)

SOLUTIONS: Here the problem is that Inverted Jacobi does not converge having initial vector & matrix:

(1/3,1/3,1/3),  |-1  1  0|
                | 0 -1  1| 
                | 2  0 -2|

iterations give:
0: (1/3,1/3,1/3)
1: (2/3,1/3,1/6)
2: (1/3,2/3,1/6)
3: (1/3,1/3,1/3)
4:     ....
So depending on what iteration we stop we have alternation for 1 & 2 state probabilities: 1/3 and 2/3

---------------------006--------------------------------
In tests csl_bounded_until_01 - csl_bounded_until_04 values that are close
to 1.0 should be actually 1.0 as I see it. They are within error bound but still ... Check it out ...

---------------------007--------------------------------
In tests csl_bounded_until_01 - csl_bounded_until_0 for P{<0.1}[ tt U[0,0] a4 ] we get
   Result: ( 1.0e+00, 1.0e+00, 1.0e+00, 1.0e+00 )
But it should be 
   Result: ( 0.0e+00, 0.0e+00, 0.0e+00, 1.0e+00 )



