1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
|
#!/usr/bin/env python
#####################
#! \file floating_point.py
## \verbatim
## Top contributors (to current version):
## Eva Darulova, Makai Mann
## This file is part of the CVC4 project.
## Copyright (c) 2009-2018 by the authors listed in the file AUTHORS
## in the top-level source directory) and their institutional affiliations.
## All rights reserved. See the file COPYING in the top-level source
## directory for licensing information.\endverbatim
##
## \brief A simple demonstration of the solving capabilities of the CVC4
## floating point solver through the Python API contributed by Eva
## Darulova. This requires building CVC4 with symfpu.
import pycvc4
from pycvc4 import kinds
if __name__ == "__main__":
slv = pycvc4.Solver()
if not slv.supportsFloatingPoint():
# CVC4 must be built with SymFPU to support the theory of
# floating-point numbers
print("CVC4 was not built with floating-point support.")
exit()
slv.setOption("produce-models", "true")
slv.setLogic("QF_FP")
# single 32-bit precision
fp32 = slv.mkFloatingPointSort(8, 24)
# the standard rounding mode
rm = slv.mkRoundingMode(pycvc4.RoundNearestTiesToEven)
# create a few single-precision variables
x = slv.mkConst(fp32, 'x')
y = slv.mkConst(fp32, 'y')
z = slv.mkConst(fp32, 'z')
# check floating-point arithmetic is commutative, i.e. x + y == y + x
commutative = slv.mkTerm(kinds.FPEq, slv.mkTerm(kinds.FPPlus, rm, x, y), slv.mkTerm(kinds.FPPlus, rm, y, x))
slv.push()
slv.assertFormula(slv.mkTerm(kinds.Not, commutative))
print("Checking floating-point commutativity")
print("Expect SAT (property does not hold for NaN and Infinities).")
print("CVC4:", slv.checkSat())
print("Model for x:", slv.getValue(x))
print("Model for y:", slv.getValue(y))
# disallow NaNs and Infinities
slv.assertFormula(slv.mkTerm(kinds.Not, slv.mkTerm(kinds.FPIsNan, x)))
slv.assertFormula(slv.mkTerm(kinds.Not, slv.mkTerm(kinds.FPIsInf, x)))
slv.assertFormula(slv.mkTerm(kinds.Not, slv.mkTerm(kinds.FPIsNan, y)))
slv.assertFormula(slv.mkTerm(kinds.Not, slv.mkTerm(kinds.FPIsInf, y)))
print("Checking floating-point commutativity assuming x and y are not NaN or Infinity")
print("Expect UNSAT.")
print("CVC4:", slv.checkSat())
# check floating-point arithmetic is not associative
slv.pop()
# constrain x, y, z between -3.14 and 3.14 (also disallows NaN and infinity)
a = slv.mkFloatingPoint(8, 24, slv.mkBitVector("11000000010010001111010111000011", 2)) # -3.14
b = slv.mkFloatingPoint(8, 24, slv.mkBitVector("01000000010010001111010111000011", 2)) # 3.14
bounds_x = slv.mkTerm(kinds.And, slv.mkTerm(kinds.FPLeq, a, x), slv.mkTerm(kinds.FPLeq, x, b))
bounds_y = slv.mkTerm(kinds.And, slv.mkTerm(kinds.FPLeq, a, y), slv.mkTerm(kinds.FPLeq, y, b))
bounds_z = slv.mkTerm(kinds.And, slv.mkTerm(kinds.FPLeq, a, z), slv.mkTerm(kinds.FPLeq, z, b))
slv.assertFormula(slv.mkTerm(kinds.And, slv.mkTerm(kinds.And, bounds_x, bounds_y), bounds_z))
# (x + y) + z == x + (y + z)
lhs = slv.mkTerm(kinds.FPPlus, rm, slv.mkTerm(kinds.FPPlus, rm, x, y), z)
rhs = slv.mkTerm(kinds.FPPlus, rm, x, slv.mkTerm(kinds.FPPlus, rm, y, z))
associative = slv.mkTerm(kinds.Not, slv.mkTerm(kinds.FPEq, lhs, rhs))
slv.assertFormula(associative)
print("Checking floating-point associativity")
print("Expect SAT.")
print("CVC4:", slv.checkSat())
print("Model for x:", slv.getValue(x))
print("Model for y:", slv.getValue(y))
print("Model for z:", slv.getValue(z))
|