remove extra division in mpq/aors.c
marc.glisse at inria.fr
Sun Feb 20 19:55:14 CET 2011
On Sun, 20 Feb 2011, Torbjorn Granlund wrote:
> OK, I'll apply that patch instead. Could you please make a change log
> for it?
2011-02-20 Marc Glisse <marc.glisse at inria.fr>
* mpq/aors.c: Remove redundant division
> BTW, did you make sure the temps will be suffient?
I hope I did, but a second look couldn't hurt.
> I wonder if 4 distinct temp areas (gcd, tmp1, tmp2, t) are really
At least I didn't increase that number ;-)
We could certainly merge the first 3 into one area with gcd, tmp1 and tmp2
pointing to different offsets into it. If I remember correctly this is
something that could be done in many places in gmp.
With the first patch, it would be possible to completely remove t by
reusing the numerator of the result, but it may needlessly increase some
> I have attached a patch to modernise a related function. Does it look
> correct to you?
I am sorry but I am not familiar at all with the large number of division
functions. I can try to have a look next week.
More information about the gmp-discuss