Initial state and check_invariant

From: Jean-Luc Mercadier (jean-luc.mercadier@tni-valiosys.com)
Date: Mon Jul 08 2002 - 08:57:44 MDT

  • Next message: Fabio Somenzi: "Re: Initial state and check_invariant"

    Hello,

    I would like to check the following design (W10.trivial.blif):

    .model w10l.s.pre_blif
    .outputs xx35_out
    .latch xx35 xx35_out 1;
    .names xx35
    1
    .end

    with the following invariant (W10.B.invar):

    xx35_out = 1 ;
    and the following script :
    set vis_stdout ci_trivial_1.vis.output
    set vis_stderr ci_trivial_1.vis.error
    read_blif W10.trivial.blif
    flatten_hierarchy
    static_order
    build_partition_mdds
    check_invariant -d 1 -i -v 2 -F ci_trivial_1.vis.debug W10.B.invar
    time

    I've obtained the following output :

    ********************************
    Reachability analysis results:
    FSM depth = 1
    reachable states = 2
    BDD size = 1
    analysis time = 0
    reachability analysis = complete
    # INV: Summary of invariant pass/fail
    # INV: formula failed --- xx35_out=1
    elapse: 0.0 seconds, total: 0.0 seconds

    and the following debug trace :

    # INV: formula 1 failed --- xx35_out=1
    # INV: calling debugger
    # INV: invariant fails at the following initial state
    --State 0:
    xx35_out:0

    The built FSM accept value 0 for variable xx35_out even with the
    .latch xx35 xx35_out 1;
    initialising xx35_out with value 1.

    How can I force the initial state of the FSM with chosen values ?
    Could this be a bug ?
    Thanks in advance for your response.
    Jean-Luc Mercadier



    This archive was generated by hypermail 2b30 : Mon Jul 08 2002 - 08:10:26 MDT