optimizing GMP for my code

Blake Huff stangmechanic at gmail.com
Mon Oct 16 09:24:05 CEST 2006

Dr. Roshan:

Did you run 'make check' to see if GMP compiled correctly?  If not,  
that'd probably be a good place to start.   Otherwise, does this  
happen when you run on the cluster, or just a single computer?  Any  
guess on what function/part of the code is giving you this?

Blake Huff
stangmechanic at gmail.com

"When the going gets tough, the tough get aeronautical."-Howling Mad  

On Oct 15, 2006, at 7:21 PM, Usman W Roshan cis faculty/staff wrote:

> Thanks Blake. Your tips are highly useful.
> I tried ABI=32 and GMP compiles on my AMD Opteron 64 cluster.  
> However, I
> get the following error when I run my program:
> GNU MP: Cannot allocate memory (size=16)
> Abort
> This looks like a GMP problem. Does anyone know what this means and  
> how
> can I fix the problem?
> I'm still looking into your last suggestion (see below) about using a
> fixed mantissa. I'm not sure how easy this is going to be.
> Usman
> On Sun, 15 Oct 2006, Blake Huff wrote:
>> Finally, and I'm mostly speculating on this as I haven't dug this
>> deep into the code.  Each variable keeps the size of the mantissa
>> data actually in use regardless of the precision requested.  If
>> you're willing to take the performance hit and use a fixed mantissa
>> size, you might be able to delete this from the mpf_t  variable and
>> alter the functions to assume a fixed mantissa, negating the need to
>> keep track of how many bits are actually in use, again freeing up
>> more memory.
>> Blake Huff
>> stangmechanic at gmail.com

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://gmplib.org/list-archives/gmp-discuss/attachments/20061016/b47d7e92/attachment.html 

More information about the gmp-discuss mailing list