libquad question

[Steve Kargl]

On Mon, Sep 06, 2010 at 08:48:08AM -0700, Steve Kargl wrote:
> On Mon, Sep 06, 2010 at 02:26:38PM +0200, Dominique Dhumieres wrote:
> > Steve,
> > 
> > I am not sure to understand your problem.
> > 
> 
> The problem is that on FreeBSD (and any other system)
> where precision(1.0_10) and precision(1.0_8) are
> the same, the testsuite may not run tests on REAL(10),
> because the code typcally uses selected_real_kind()
> with the idiom 'precision(1.0_8) + 1'.

I should qualify this to i686-*-freebsd.  On x86_64,
one get the expected result based on precision() 
alone:

troutmask:sgk[202] gfc4x -o z wq.f90 
troutmask:sgk[203] ./z
 4  6   37
 8 15  307
10 18 4931
16 33 4931


-- 
Steve