[LLVMdev] How to pick default floating point ABI?

[Weitian Leung]

 Hello,
    I'm using clang on x86 cross build for mips, here is my main configure
parameters
    llvm/configure    \
            --prefix=/usr    \
            --target=mipsel-unknown-linux-gnu    \
            --enable-targets=mips    \
            --enable-optimized    \
            --enable-shared    \
            --disable-assertions
    I want to build code for *mips3* with *o32* abi, but fails:

mipsel-unknown-linux-gnu-clang++ test.cpp -mips3
/tmp/test-a51985.s: Assembler messages:
/tmp/test-a51985.s:11: Warning: `fp=64' used with a 32-bit ABI
/usr/local/bin/mipsel-unknown-linux-gnu-ld: Warning: a.out uses
-mdouble-float (set by
/opt/toolchain_clang/host/usr/mipsel-unknown-linux-gnu/sysroot/usr/lib/../lib/crt1.o),
/tmp/test-d48db1.o uses -mgp32 -mfp64

    How can I change the *fp=64* to *fp=32*? Have tried pass -mfp32 to
mipsel-unknown-linux-gnu-clang++ but with no luck.
    Though this is a warning, but it fails to compile Qt.
    Yes I known it works if uses *-mips64* etc, but I need mips3.

    Please let me know if I'm post to the wrong place, I'm new here.
    Thank you.
-- 
*Regards*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20150611/0fcd6366/attachment.html>