This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: PATCH: PR bootstrap/17369: Bootstrap failure with binutils-2.15.90.0.1.1
> OBJDUMP="$(OBJDUMP)"; export OBJDUMP; \
> + if [ x"@SET_GCC_LIB_PATH@" != x ]; then \
> + SET_GCC_LIB_PATH="@SET_GCC_LIB_PATH@"; \
> + export SET_GCC_LIB_PATH; \
> + else true; fi; \
> + @SET_GCC_LIB_PATH@ \
This looks confusing, using the same name for both an environment
variable and an autoconf macro. Didn't we just agree to use TL_FOO
for global toplevel macros? And why can't we just always set
SET_GCC_LIB_PATH like we do for all the other variables?
> +if test x"$SET_GCC_LIB_PATH" != x; then
> + eval "set_gcc_lib_path=$SET_GCC_LIB_PATH"
> + set_gcc_lib_path="export $set_gcc_lib_path"
> +else
> + fi
> + SET_GCC_LIB_PATH="\$(RPATH_ENVVAR)=$d; export \$(RPATH_ENVVAR);"
> + ;;
> +# Set up library path if needed.
> +@set_gcc_lib_path@
> +
Don't we end up with "export $(RPATH_ENVVAR)=$d; export $(RPATH_ENVVAR)"
here? That seems wrong for a Makefile.