So, having not had much luck with OpenGL on the mini6410, I've got a basic 3D display running in software. I was wondering if there was any easy way to enable compiling to use the vector FPU in the mini6410 ? Love to hear from anyone who's done it :) Cheers Simon
FriendlyARM has released the new compiler 4.5.1 that support VFP, and use armv6 as the default
I am using 4.5.1, so that's good news :) When you say "uses armv6 as the default", are there any options I can add to the commandline to help FPU performance (or performance in general...) ? When I do: "arm-linux-gcc -v", I get: Using built-in specs. COLLECT_GCC=arm-linux-gcc COLLECT_LTO_WRAPPER=/opt/FriendlyARM/toolschain/4.5.1/libexec/gcc/arm-none-linux -gnueabi/4.5.1/lto-wrapper Target: arm-none-linux-gnueabi Configured with: /work/toolchain/build/src/gcc-4.5.1/configure --build=i686-build_pc-linux-gnu --host=i686-build_pc-linux-gnu --target=arm-none-linux-gnueabi --prefix=/opt/FriendlyARM/toolschain/4.5.1 --with-sysroot=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi/sys-root --enable-languages=c,c++ --disable-multilib --with-cpu=arm1176jzf-s --with-tune=arm1176jzf-s --with-fpu=vfp --with-float=softfp --with-pkgversion=ctng-1.8.1-FA --with-bugurl=http://www.arm9.net/ --disable-sjlj-exceptions --enable-__cxa_atexit --disable-libmudflap --with-host-libstdcxx='-static-libgcc -Wl,-Bstatic,-lstdc++,-Bdynamic -lm' --with-gmp=/work/toolchain/build/arm-none-linux-gnueabi/build/static --with-mpfr=/work/toolchain/build/arm-none-linux-gnueabi/build/static --with-ppl=/work/toolchain/build/arm-none-linux-gnueabi/build/static --with-cloog=/work/toolchain/build/arm-none-linux-gnueabi/build/static --with-mpc=/work/toolchain/build/arm-none-linux-gnueabi/build/static --with-libelf=/work/toolchain/build/arm-none-linux-gnueabi/build/static --enable-threads=posix --with-local-prefix=/opt/FriendlyARM/toolschain/4.5.1/arm-none-linux-gnueabi/sys -root --disable-nls --enable-symvers=gnu --enable-c99 --enable-long-long Thread model: posix gcc version 4.5.1 (ctng-1.8.1-FA) ... looking at that, it seems to say "--with-float=softfp" even though it also says "--with-fpu=vfp". Is this important ? Or does it just mean the floating-point protocol is the same as the software FP, even though it actually uses the VFP ? Thanks :) On the downside, if everything is already set up correctly, that means I'm limited to a 24x24 grid of polygons, getting about 9 frames/second... It's a real pity there's no OpenGL under Linux [sigh] Simon