[Bug fortran/15441] RRSPACING broken for denormals

["kargl at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>]

------- Comment #8 from kargl at gcc dot gnu dot org  2006-10-04 17:20 -------
gfortran still gets rrspacing of subnormal wrong.

troutmask:sgk[215] ./z
 s               =  4.4081038E-39  <-- Subnormal named constant
 x               =  4.4081038E-39  <-- Subnormal variable
 rrspacing(s)    =  1.2582912E+07  <-- Constant folding
 my_rrspacing(x) =  1.2582912E+07  <-- Fortran code
 rrspacing(x)    =            NaN  <-- gfortran inline

program h
  real, parameter :: s = 0.375 * tiny(1.)
  real x
  x = s
  print *, 's               =', s, ' <-- Subnormal named constant'
  print *, 'x               =', x, ' <-- Subnormal variable'
  print *, 'rrspacing(s)    =', rrspacing(s), ' <-- Constant folding'
  print *, 'my_rrspacing(x) =', my_rrspacing(x), ' <-- Fortran code'
  print *, 'rrspacing(x)    =', rrspacing(x),   ' <-- gfortran inline'
  contains
    function my_rrspacing(x)
    implicit none
    real my_rrspacing
    real, intent(in) :: x
    real y
    y = abs(x)
    if (y == 0.) then
       my_rrspacing = 0.
    else
       my_rrspacing = scale(scale(y, - exponent(y)), digits(y))
    end if
    end function my_rrspacing
end program h

The F95 standard says that rrspacing(x) = | x * b**(-e) | 2**p where b, e,
and p are the radix, exponent of x, and prcision.

trans-intrinsics.c (gfc_conv_intrinsic_rrspacing) appears to mutilate the 
mathematical expression.


-- 

kargl at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|INVALID                     |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15441