#204 spurious FP contagion warnings

[Sam Steingold]

[20]> (LOG -3/4)

WARNING: Numerical operation combines exact and inexact
numbers (rational
numbers and floating-point numbers), and the
mathematical result is
exact. See ANSI CL 12.1.4.1 and the CLISP
impnotes for details. The
result's actual exactness is controlled by
*FLOATING-POINT-RATIONAL-CONTAGION-ANSI*. To
shut off this warning,
set
*WARN-ON-FLOATING-POINT-RATIONAL-CONTAGION* to NIL.
#C(-0.2876821 3.1415927)

[24]> (cosh #c(1d0 2d0))

WARNING: Floating point operation combines numbers of
different precision. See
ANSI CL 12.1.4.4 and the CLISP impnotes for
details. The result's
actual precision is controlled by
*FLOATING-POINT-CONTAGION-ANSI*. To
shut off this warning, set
*WARN-ON-FLOATING-POINT-CONTAGION* to NIL.
WARNING: Floating point operation combines numbers of
different precision. See
ANSI CL 12.1.4.4 and the CLISP impnotes for
details. The result's
actual precision is controlled by
*FLOATING-POINT-CONTAGION-ANSI*. To
shut off this warning, set
*WARN-ON-FLOATING-POINT-CONTAGION* to NIL.
#C(-0.64214812471552d0 1.0686074213827783d0)

the fix is to insert R_F_float_F calls at the
appropriate places, but one has to be careful because
there are lots of such places.

[Sam Steingold]

Logged In: YES
user_id=5735

this is serios!!!

Break 1 [8]> (expt 0.7877386934673366d0 0.4131787330316742d0)

*** - Numerical operation combines exact and inexact numbers
(rational numbers
and floating-point numbers), and the mathematical
result is exact. See
ANSI CL 12.1.4.1 and the CLISP impnotes for details.
The result's actual
exactness is controlled by
*FLOATING-POINT-RATIONAL-CONTAGION-ANSI*. To
shut off this warning, set
*WARN-ON-FLOATING-POINT-RATIONAL-CONTAGION*
to NIL.

[Bruno Haible]

Logged In: YES
user_id=5923

Fixed in the CVS.