GMP on PowerPC 970

Brendan Younger bry at mandli.com
Fri Jul 30 23:41:42 CEST 2004


On Jul 30, 2004, at 4:29 PM, Torbjorn Granlund wrote:

> Brendan Younger <bry at mandli.com> writes:
>
>   I've been trying to compile GMP with "longlong" limb support for my 
> PowerPC
>   970 running Mac OS X.  After hacking the configure script to 
> recognize my
>   machine and using the generic powerpc64 umul_ppmm, it compiles nicely
>   (although there are a few warnings related to comparing 32-bit and 
> 64-bit
>   integers) but fails a few tests in MPZ, namely t-mul, t-div, t-bin, 
> etc.
>   Has anyone succeeded in compiling this with 64-bit limb support, or 
> should I
>   just sit around and wait for Mac OS X to become 64-bit?
>
> The mpn/powerpc64 assembly code isn't prepared to deal with a processor
> running in 32-bit mode, which probably will make things rather slow.
> I'd suggest to stick to the 32-bit code.

I didn't use any of the mpn/powerpc64 code aside from umul_ppmm(), 
primarily since it was written for assemblers which support the POWER 
mnemonics and Apple's version of "as" doesn't.  Hence, everything 
compiled was generic C-code.  Also, there is no 64bit/32bit dichotomy 
on the PowerPC 970; any program can call a 64-bit instruction at any 
time so there is no slow-down when using 64-bit code.  I would be 
interested in fixing GMP to work with "longlong" limbs, but I'd like to 
have the support of someone who knows more about the internals of GMP 
than I do.

Brendan Younger



More information about the gmp-discuss mailing list