*) Fix warnings with new compilers, warnings treated as error
     in maintainer-mode, standard c-89 is enforced
     trunk patch: http://svn.apache.org/r1702948
                  http://svn.apache.org/r1759415
     2.4.x patch:
http://home.apache.org/~ylavic/patches/httpd-2.4.x-r1702948_and_co.patch
     +1: ylavic, jorton,
     jchampion: r1702948 doesn't quite work as advertised. -Werror is never
                added to CFLAGS because when combined with
-Wstrict-prototypes,
                the AC_LANG_PROGRAM won't compile (it uses a bare main()).
     ylavic: Maybe the -Werror case could be handled later, for now this
series
             avoids a lot of "warning: 'aplog_module_index' defined but not
             used [-Wunused-const-variable=]" thanks to AP_MAYBE_UNUSED.
     jchampion: Fine by me. I just think the dead code should be removed
from
                the backport in the meantime, so people don't have a false
sense
                of security.

I usually just read on beyond patches that are 'under discussion'.
STATUS isn't a good bidirectional communications mechanism.

jchampion, did you have a proposed alternative since you felt
strongly about this issue?

Cheers,

Bill

On Mon, Nov 28, 2016 at 1:37 PM, William A Rowe Jr <wr...@rowe-clan.net>
wrote:

> Typical of a long batch of error noise on 2.4 HEAD (with the
> http strict merge patch applied) in --enable-maintainer-mode...
>
> In file included from /home/wrowe/dev/httpd-2.4/modules/mappers/mod_vhost_
> alias.c:45:0:
> /home/wrowe/dev/httpd-2.4/include/http_config.h:427:24: warning:
> 'aplog_module_index' defined but not used [-Wunused-const-variable=]
>      static int * const aplog_module_index = &(foo##_module.module_index)
>                         ^
> /home/wrowe/dev/httpd-2.4/include/http_config.h:437:5: note: in expansion
> of macro 'APLOG_USE_MODULE'
>      APLOG_USE_MODULE(foo);                         \
>      ^~~~~~~~~~~~~~~~
> /home/wrowe/dev/httpd-2.4/modules/mappers/mod_vhost_alias.c:447:1: note:
> in expansion of macro 'AP_DECLARE_MODULE'
>  AP_DECLARE_MODULE(vhost_alias) =
>  ^~~~~~~~~~~~~~~~~
>
> Has anyone investigated far enough for us to pacify gcc?
>

Reply via email to