------- Comment #3 from mkl at pengutronix dot de 2009-05-29 08:22 ------- Hello Andrew,
(In reply to comment #2) > http://www.nabble.com/Re:-libFFI-arm-compilation-fails-with-assembly-td20346235.html BTW: I'm not compiling libffi stand alone, here. I'm building the gcc with java enabled. IMHO using "-mfloat-abi=soft" when preprocessing sysv.S can be considered a workaround not a fix. It makes only sense if the compiler passes (by default) floats in the integer registers (-mfloat-abi=soft and -mfloat-abi=softfp), and breaks if passing in the float registers (-mfloat-abi=hard). With the work done in ARM/hardvfp_4_4_branch the -mfloat-abi=hard case is more relevant then ever. This means we cannot use "-mfloat-abi=soft" unconditionally, which boils down (again) to the question: how can we distinguish between the float-passed-in-int-regs and float-passed-in-float-regs case. Ideally this should work if building libffi as part of gcc and if building libffi standalone. cheers, Marc -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=40242
