TRS Standard pair #615299805

details

property	value
status	complete
benchmark	Ex4_DLMMU04_Z.xml
ran by	Akihisa Yamada
cpu timeout	1200 seconds
wallclock timeout	300 seconds
memory limit	137438953472 bytes
execution host	n046.star.cs.uiowa.edu
space	Transformed_CSR_04

run statistics

property	value
solver	AutoNon 1.21
configuration	autonon
runtime (wallclock)	104.135364056 seconds
cpu usage	404.590186889
max memory	1.7984E10

stage attributes

key	value
output-size	10872
starexec-result	"NO"

output

/export/starexec/sandbox2/solver/bin/starexec_run_autonon /export/starexec/sandbox2/benchmark/theBenchmark.xml /export/starexec/sandbox2/output/output_files


--------------------------------------------------------------------------------


NO

<html>
<body>
<pre style="font-family: monospace, serif; font-size: 1em; white-space: pre-wrap; margin: 20px; padding: 20px;">

*** AutoNon ***


AutoNon proves looping and non-looping non-termination using regular languages.

AutoNon employs SAT solving to find finite automata as certificates of non-termination.
The automata define (a) the language and (b) which rule may be applied where in a term.
By property (b), the language of the automaton does not need to be closed under rewriting.
It suffices that every term in the language admits some rewrite step to a term in the language.
Finally, we employ a quasi-order on the states of the automaton. This allows rewriting to
change the interpretation of a term (the state of the automaton) and thereby to change what
rules may be applied (the allowed rules depend on the state of the automaton).

The technique is described in the following paper:

    [EZ2015] Proving Non-Termination by Finite Automata
             by J. Endrullis and H. Zantema
             Proc. Conf. on Rewriting Techniques and Applications (RTA 2015)

For further information see <a href="http://joerg.endrullis.de/nonterm/">

       http://joerg.endrullis.de/nonterm/       

</a>

*** Input system ***

    and(tt,T) -> T
    isNatIList(IL) -> isNatList(activate(IL))
    isNat(n__0) -> tt
    isNat(n__s(N)) -> isNat(activate(N))
    isNat(n__length(L)) -> isNatList(activate(L))
    isNatIList(n__zeros) -> tt
    isNatIList(n__cons(N,IL)) -> and(isNat(activate(N)),isNatIList(activate(IL)))
    isNatList(n__nil) -> tt
    isNatList(n__cons(N,L)) -> and(isNat(activate(N)),isNatList(activate(L)))
    isNatList(n__take(N,IL)) -> and(isNat(activate(N)),isNatIList(activate(IL)))
    zeros -> cons(0,n__zeros)
    take(0,IL) -> uTake1(isNatIList(IL))
    uTake1(tt) -> nil
    take(s(M),cons(N,IL)) -> uTake2(and(isNat(M),and(isNat(N),isNatIList(activate(IL)))),M,N,activate(IL))
    uTake2(tt,M,N,IL) -> cons(activate(N),n__take(activate(M),activate(IL)))
    length(cons(N,L)) -> uLength(and(isNat(N),isNatList(activate(L))),activate(L))
    uLength(tt,L) -> s(length(activate(L)))
    0 -> n__0
    s(X) -> n__s(X)
    length(X) -> n__length(X)
    zeros -> n__zeros
    cons(X1,X2) -> n__cons(X1,X2)
    nil -> n__nil
    take(X1,X2) -> n__take(X1,X2)
    activate(n__0) -> 0
    activate(n__s(X)) -> s(X)
    activate(n__length(X)) -> length(X)
    activate(n__zeros) -> zeros
    activate(n__cons(X1,X2)) -> cons(X1,X2)
    activate(n__nil) -> nil
    activate(n__take(X1,X2)) -> take(X1,X2)
    activate(X) -> X



*** Non-looping non-termination using regular languages ***


Input system R is not terminating since R admits a regular language weakly closed under rewriting.


The following finite automaton witnesses non-termination:

  The states are: 0,1,...,3

  The accepting states are: 3

  We use the following order < on the states:

        <  0  1  2  3
        0  1         
        1     1      
        2        1  1
        3           1

  The states determine the rules that may be used:

      state 0:
          zeros -> n__zeros
          nil -> n__nil
      state 1:

popout

output may be truncated. 'popout' for the full output.

job log

popout

actions all output return to TRS Standard