Building gmp library for the iPhone-Update
Pierre-Henri Jondot
mpkju at free.fr
Thu Jul 16 12:21:35 CEST 2009
Le 16 juil. 09 à 11:45, Torbjorn Granlund a écrit :
> Pierre-Henri Jondot <mpkju at free.fr> writes:
>
> Trying to build a gmp library for arm without asm, I replaced arm-
> apple-darwin host with none-apple-darwin (but keeping the arm cross-
> compiler with CC flag) I could build an arm library.
>
> That said, I have been unable to build a simple test case, the link
> stage failing with :
> Undefined symbols:
> "___gmpn_invert_limb", referenced from:
> _mpn_mod_1_unnorm in libgmp.a(mod_1.o)
> _mpn_mod_1_norm in libgmp.a(mod_1.o)
>
> Usually, a host cpu of "none" should pass -DNO_ASM. If you look at
> the
> compiler invokations, didn't that happen? (The logic for -DNO_ASM
> is in
> the configure script, you can locate it easily.)
>
> (Possibly, the cross-compilation motivated overriding of various
> parameters might have inhibited -DNO_ASM.)
>
> --
> Torbjörn
Yes,
For some reason indeed, the -DNO_ASM was not there...
So I've added it to the configure invocation, which is now :
./configure CC="/Developer/Platforms/iPhoneOS.platform/Developer/usr/
bin/gcc-4.2" CPP="/Developer/Platforms/iPhoneOS.platform/Developer/usr/
bin/gcc-4.2 -E" LD="/Developer/Platforms/iPhoneOS.platform/Developer/
usr/bin/ld" CPPFLAGS="-arch armv6 -isysroot /Developer/Platforms/
iPhoneOS.platform/Developer/SDKs/iPhoneOS3.0.sdk -miphoneos-version-
min=3.0 -DNO_ASM" --disable-shared --enable-static --prefix=/Developer/
Platforms/iPhoneOS.platform/Developer/SDKs/iPhoneOS3.0.sdk/usr/local --
host=none-apple-darwin9
The library I obtained this way allowed me this time to build a simple
test case and run it on an iphone.
I'll try to investigate a bit about the asm file that fails to build,
but I know no arm asm (or no asm at all I should say) and so far I
have not been able to find much documentation, or at all, about
apple's compiler, so most probably "porting GMP to the iPhone" will be
out of my league...
Regards,
Pierre-Henri
More information about the gmp-discuss
mailing list