Building GMP 6.0 for Ivy Bridge

Victor Shoup shoup at
Fri Oct 2 20:04:34 UTC 2015


I tried --build=ivybridge...
On the latest snapshot, that works...the only difference I 
can really discern between the two is that --march and --mtune
resolve to ivybridge, rather than corei7, which is what I get
when I build with coreisbr.

My understanding is that there should be little difference between the
two microarchitectures.  It looks like the same assembly codes are pulled
in --- actually, it looks like GMP lumps ive/sandy bridge together...

Another note: on the official 6.0.0 release.
--build=ivybridge doesn't work at doesn't even
pull x86 assembly codes at all.

For the project I'm working on now, for various reasons, I'd like
to stick with the official GMP release.  So in that case, it seems
--build=coreisbr is the way for me to go.

Also...note that I also went to the official 6.0.0 version,
pulled out the cpuid stuff from config.guess, and ran it.
The output there is "coreisbr". in both cases, that program does output the
right thing, and yet configure itself loses it somewhere,
and everything reverts to a generix x86_64.  So maybe it is
just a bug somewhere else in the configure script??

On Oct 2, 2015, at 9:35 AM, Marc Glisse wrote:

> On Wed, 30 Sep 2015, Victor Shoup wrote:
>> According to lscpu:
>> Intel(R) Xeon(R) CPU E5-2680 v2 @ 2.80GHz
> Actually, could you post the rest of the lscpu output?
> (I should have asked that first...)
> -- 
> Marc Glisse

More information about the gmp-discuss mailing list