[systemd-devel] [systemd-commits] configure.ac

[David Herrmann]

Hi

On Tue, Feb 24, 2015 at 4:15 PM, Zbigniew Jędrzejewski-Szmek
<zbyszek at in.waw.pl> wrote:
> On Tue, Feb 24, 2015 at 04:08:17PM +0100, David Herrmann wrote:
>> Hi
>>
>> On Tue, Feb 24, 2015 at 4:07 PM, Zbigniew Jędrzejewski-Szmek
>> <zbyszek at in.waw.pl> wrote:
>> > On Tue, Feb 24, 2015 at 07:02:49AM -0800, David Herrmann wrote:
>> >>  configure.ac |    1 +
>> >>  1 file changed, 1 insertion(+)
>> >>
>> >> New commits:
>> >> commit 0a98d66159e474915afd6597d3aa444a698fdd2d
>> >> Author: David Herrmann <dh.herrmann at gmail.com>
>> >> Date:   Tue Feb 24 15:59:06 2015 +0100
>> >>
>> >>     build: add -Wno-format-signedness
>> >>
>> >>     gcc5 introduced this option (gcc4 silently ignores it, which is fine).
>> >>     Given that gcc5 thinks 'unsigned char'/'unsigned short' is promoted to
>> >>     'int' for var-args, stuff like this spits out warnings:
>> >>         uint8_t x;
>> >>         printf("%" PRIu8", x);
>> >>
>> >>     gcc5 promots 'x' to 'int', instead of 'unsigned int' and thus gets a
>> >>     signedness-warnings as it expects an 'unsigned int'.
>> >>
>> >>     glibc states otherwise: unsigneds are always promoted to 'unsigned int'.
>> >>     Until gcc and glibc figure this out, lets just ignore that warning (which
>> >>     is totally useless in its current form).
>> >>
>> >> diff --git a/configure.ac b/configure.ac
>> >> index 9a2235b..22a6f17 100644
>> >> --- a/configure.ac
>> >> +++ b/configure.ac
>> >> @@ -186,6 +186,7 @@ CC_CHECK_FLAGS_APPEND([with_cflags], [CFLAGS], [\
>> >>          -Wno-unused-parameter \
>> >>          -Wno-missing-field-initializers \
>> >>          -Wno-unused-result \
>> >> +        -Wno-format-signedness \
>> > This doesn't really work, as discussed previously on the list.
>>
>> Why? Works fine with gcc4 and gcc5 here. What's the problem exactly?
> Positive version of the option must be used to detect reliably.
>
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63499

Do we care? I mean I prefer a warning about unknown command-line
options over thousands of warnings regarding totally useless (and imo
wrong) signedness-mismatches. Furthermore, the warning is only printed
if there's another warning.

I can have a look at adding a CC_CHECK_NO_WARNING_APPEND() or
something like that, which properly tests for the non-negated version.

Thanks
David