Hi
Some timings for bobcat in svn size=1000 limbs
./benchmpn
cpu bobcat
add_n2538
sub_n2528
mul_15064this is the 1-way unrolled core2 code
addmul_15308again this
On Friday 04 March 2011 19:28:21 jason wrote:
> On Feb 26, 10:24 am, Jason wrote:
> > Hi
> >
> > An outline of what I hope we can do for MPIR v2.4
> >
> > 1) Release on June 1st
> >
> > 2) More x86_64 assembler code
>
> There is a new dir for the new AMD bobcat cpu , if we cant get hold of
> a
On Feb 26, 10:24 am, Jason wrote:
> Hi
>
> An outline of what I hope we can do for MPIR v2.4
>
> 1) Release on June 1st
>
> 2) More x86_64 assembler code
>
There is a new dir for the new AMD bobcat cpu , if we cant get hold of
a VIA nano , then I will make it default to the K8 code , which will
b
I don't know. It looks suspiciously like an actual bug to me.
On 4 March 2011 10:09, Cactus wrote:
> There is an FFT assertion failure reported here:
>
> http://www.mersenneforum.org/showthread.php?t=15327
>
> on the Windows x64 version of MPIR. It seems to be a very rare issue
> but I wonder i
That's great news. Thanks Jason.
Now let's see if I can get this FFT done for the next release!
Bill.
On 4 March 2011 07:48, Jason wrote:
> Hi
>
> MPIR 2.3.0 was released on 4th March 2011.
>
> * source tarball: http://www.mpir.org/mpir-2.3.0.tar.bz2
> * documentation: http://www.mpir.org/mpir-
On Thu, Mar 3, 2011 at 5:04 AM, Jason wrote:
> OK , but how about tuning and timings?
> Does anyone here have any experience in this ? , or any recommendations
I don't think I have ever actually tested things in virtual machines
such as tuning/timings. Since you still have your 32bit "native"
i
There is an FFT assertion failure reported here:
http://www.mersenneforum.org/showthread.php?t=15327
on the Windows x64 version of MPIR. It seems to be a very rare issue
but I wonder if there is anything that needs to be done about it.
It seems that there might also be some issues that arise w
Hi
The mpir trac hasn't worked in ages , how about removing all references to it
from the web page. I never thought it was useful , I think a simple text file
in svn trunk is good enough?
Jason
--
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
T