Re: [Chicken-users] bug in configure.in

[felix]

Bruce Hoult wrote:
> The following test in configure.in appears to be reversed, resulting in
> it being pointless to provide a different value for CFLAGS on the
> commandline as it has no effect.
>
>
> *** configure.in        3 Jul 2003 19:25:22 -0000       1.34
> --- configure.in        24 Jul 2003 18:53:30 -0000
> ***************
> *** 139,145 ****
>   AC_SUBST(CPUCONFIGFLAGS)
>   case "$CC" in
>     *gcc*)
> !   if test -n "$CFLAGS"; then
>       CFLAGS="-O3 -fomit-frame-pointer -Wall -Wno-unused
> -Wno-uninitialized"
>       if test -n "$HAVE_ALLOCA_H"; then
>         CFLAGS="$CFLAGS -DHAVE_ALLOCA_H"
> --- 139,145 ----
>   AC_SUBST(CPUCONFIGFLAGS)
>   case "$CC" in
>     *gcc*)
> !   if test -z "$CFLAGS"; then
>       CFLAGS="-O3 -fomit-frame-pointer -Wall -Wno-unused
> -Wno-uninitialized"
>       if test -n "$HAVE_ALLOCA_H"; then
>         CFLAGS="$CFLAGS -DHAVE_ALLOCA_H"

In fact, it's not *really* a bug. CFLAGS is set to "-O2 -g" by default.
(But setting it on the command-line is nevertheless not properly working)


cheers,
felix