This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [4.3] making a few real functions work with dfp


On 28/11/06, Ben Elliston <bje@au1.ibm.com> wrote:
> This patch makes exact_real_truncate() and real_isinteger() handle
> dfp values correctly. In order to do so, it adds a
> decimal_normalize() function that normalizes a decimal floating
> point number.

Why are you normalising the decimal float values?  The IEEE 754R draft
standard specifically states that decimal float values are not
normalised by design.  1.0 and 1.00 are both distinctly representable.


By the way, I am not just normalising the dfp values. I am *temporally* normalising them, so I can test for equality. Does the standard says that 1.0 and 1.00 are different values ? Because that would be kind of weird...

Cheers,

Manuel.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]