Short:
The gcc -mfloat-abi option should only be passed for ARM builds
unless I am mis-reading the gcc documentation...
Long:
When I build a toolchain for powerpc with buildroot (current SVN)
this results in a compiler that doesn't accept -mfloat-abi option:
# $(STAGING_DIR)/usr/bin/powerpc-linux-uclibc-gcc --version
powerpc-linux-uclibc-gcc (GCC) 4.2.0
Copyright (C) 2007 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
# $(STAGING_DIR)/usr/bin/powerpc-linux-uclibc-gcc -mfloat-abi test.c
cc1: error: unrecognized command line option "-mfloat-abi"
The failure is observed while compiling openssl. Its makefile does:
$(SED) 's,/CFLAG=,/CFLAG= $(TARGET_SOFT_FLOAT) ,g' \
$(OPENSSL_DIR)/Configure
where TARGET_SOFT_FLOAT comes from toolchain/gcc/Makefile.in:
ifeq ($(findstring x4.,x$(GCC_VERSION)),x4.)
SOFT_FLOAT_CONFIG_OPTION:=--with-float=soft
ifeq ($(BR2_SOFT_FLOAT_FP),y)
TARGET_SOFT_FLOAT:=-mfloat-abi=softfp
else # no fp at all
TARGET_SOFT_FLOAT:=-mfloat-abi=soft
#endif
else # not gcc-4.x
TARGET_SOFT_FLOAT:=-msoft-float
endif
So if using gcc4x then -mfloat-abi is passed unconditionally.
(And, because BR2_SOFT_FLOAT_FP is commented out in Config.in,
the value "soft" is always chosen, never "softfp", but that's
not my issue at the moment).
I believe the gcc4 check in toolchain/gcc/Makefile.in should
only set the -mfloat-abi for ARM architecture. I can try cobbling
together a patch for this, but as I am not entirely sure it is
the right answer, I thought I'd get some feedback first.
Thanks,
-R
_______________________________________________
busybox mailing list
[email protected]
http://busybox.net/cgi-bin/mailman/listinfo/busybox
