Dennis Ruffer wrote:
>
> Now that Tiger has been announced, I can ask about a problem it has
> with autoconf. Here's what I get when running my build script of:
>
> env ac_cv_sizeof_long_long=8 GFORTH="./gforth -i ./kernl32b.fi -
> m4M" ./BUILD-FROM-SCRATCH
>
...
> checking host system type... powerpc-apple-darwin8.0.0
> ./configure: line 3843: syntax error near unexpected token `newline'
> ./configure: line 3843: ` *" ../arch/power/_sync_cache_range.
> $ac_objext "*'
>
> The section in configure that it's complaining about ends up looking
> like this:
>
> ac_cv_func__sync_cache_range=no
> fi
> rm -f conftest.err conftest.$ac_objext \
> conftest$ac_exeext conftest.$ac_ext
> fi
> echo "$as_me:$LINENO: result: $ac_cv_func__sync_cache_range" >&5
> echo "${ECHO_T}$ac_cv_func__sync_cache_range" >&6
> if test $ac_cv_func__sync_cache_range = yes; then
> true
> else
> case $LIBOBJS in
> "../arch/power/_sync_cache_range.$ac_objext" | \
> *" ../arch/power/_sync_cache_range.$ac_objext" | \
> "../arch/power/_sync_cache_range.$ac_objext "* | \
> *" ../arch/power/_sync_cache_range.$ac_objext "*
> fi
> ;;
> *) LIBOBJS="$LIBOBJS ../arch/power/_sync_cache_range.$ac_objext" ;;
> esac
> )
>
> If I don't do the aclocal and autoconf steps, everything works fine.
>
> Any clue as to what's wrong?
The autoconf output just looks wrong (in particular, the "fi ;;"
part), so maybe the particular version of autoconf you are using has a
bug, or it may be that the problem is coming from our usage. What
version of autoconf are you using?
The source for this code is:
AC_CHECK_FUNC(_sync_cache_range,true,AC_LIBOBJ(../arch/power/_sync_cache_range))
Newer versions of autoconf have become stricter about quoting, so you
might try
AC_CHECK_FUNC([_sync_cache_range],true,AC_LIBOBJ([../arch/power/_sync_cache_range]))
- anton
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
