Re: [Gcl-devel] Bug in mpz_get_d

[Camm Maguire]

Greetings!

Enrique Perez-Terron <address@hidden> writes:

> On Mon, 2009-12-14 at 04:23 +0100, Enrique Perez-Terron wrote:
>> On Fri, 2009-12-11 at 15:12 -0500, Camm Maguire wrote:
> [snip]
>> > >(float a)
>> > 
>> > 1.0715086071862672E301
>> > 
>> > >(float (1+ a))
>> > 
>> > 1.0715086071862673E301
>
> But there is just one bit of difference:
>
> 7e6fffffffffffff
> 7e70000000000000
>

Yes, but all integers input should begin with the last bit pattern,
no?  All are greater than ....73.

> (double, high address byte printed first on x86)
>
> If stability is desired, mpfr is the one to use. GMP does not aim at
> stability, as far as I understand. mpfr is specifically "sold" as GMP
> with correct rounding, even if it has grown beyond that.
>

????

Take care,

> -Enrique
>
>
>
>
>

-- 
Camm Maguire                                        address@hidden
==========================================================================
"The earth is but one country, and mankind its citizens."  --  Baha'u'llah