On Oct 24, 10:41 am, Jason Moxham <[email protected]> wrote:
> Hi ,
>
> svn rev trunk 2474 should have the fixes applied .
> I can not test it , so if you would be so kind, thanks.
>
> The new code is : when config.guess or --build is used to pass
> core2-apple-darwin to configure AND we are building with a 32bit ABI , then a
> new mpn code path is selected ie x86/applenopic/core2 which contains no PIC
> code.
Below are the test results:
Yeah, that's what I just tried, but, alas, after configuring with
ABI=32 :
gcc -std=gnu99 -c -DHAVE_CONFIG_H -m32 -O2 -fomit-frame-pointer
-mtune=nocona -march=nocona -D__GMP_WITHIN_GMP -I.. -DOPERATION_dive_1
-I. -I. -I.. tmp-dive_1.s -fno-common -DPIC -o .libs/dive_1.o
tmp-dive_1.s:111:junk `...@got' after expression
make[2]: *** [dive_1.lo] Error 1
make[1]: *** [all-recursive] Error 1
make: *** [all] Error 2
and there it dies (I think the @GOT has something to do with the
-DPIC, etc etc).
So, I'm stalled.
>
> Note: this is ONLY for core2 ,not penryn(45nm core2) , or if someone is
> running darwin on an amd chip then it wont work.
>
> Thanks
> Jason
>
> On Saturday 24 October 2009 16:38:45 Case Vanhorsen wrote:
>
> > On Fri, Oct 23, 2009 at 4:56 PM, Jason Moxham <[email protected]>
> wrote:
> > > Actually being an apple machine can we narrow down the cpu some more , ie
> > > will it just be a core2 in 32bit mode , as apple didn't start using intel
> > > cpus until then ?
>
> > I think it will just be a core2 cpu. Below are the comments I received:
>
> > Can't build MPIR 1.3 in a way that pleases
> > gmpy: unless I use --disable_shared MPIR won't build (there's a note
> > in the .info about problems doing PIC on the Mac), if I do MPIR is
> > fine but gmpy can't link to it (because of the lack of PIC).
>
> > casevh
>
> > > On Saturday 24 October 2009 00:45:03 Jason Moxham wrote:
> > >> On Friday 23 October 2009 19:17:18 casevh wrote:
> > >> > I've been tested the latest version of gmpy with mpir-1.3.0rc1.
> > >> > Everything works fine on Windows and Linux. Another developer tested
> > >> > with OSX and encountered trac issue #94. Python/gmpy wants the PIC
> > >> > option. :(
>
> > >> > Is there a chance this issue will be fixed soon? (I have no idea if
> > >> > the fix is easy or hard.)
>
> > >> > Thanks,
>
> > >> > casevh
>
> > >> This should be easy to do a workaround once we have identitfied which
> > >> systems are affected.Does a config.guess of
> > >> x86*-apple-darwin
> > >> cover all the affected systems and no others?
> > >> And is this only for a shared library ? or is it for static as well?
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"mpir-devel" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/mpir-devel?hl=en
-~----------~----~----~----~------~----~------~--~---
