Jeremy,
On Sun, Sep 29, 2013 at 08:29:48AM -0700, Jeremy Huddleston Sequoia wrote:
> You had incorrect casting in the IN6_IS_ADDR_* macros. Does Linux (or
> whatever platform you tested on) cast the arguments to those macros for you
> (thus correcting / undoing the casting that you provided)? That seems quite
> dangerous.
>
> I initially reverted this but then noticed it was a trivial error and pushed
> a commit with the fixed cast.
>
Thanks! Good catch!
Unfortunately, with the semi-recent glibc on my Linux build machine
at least, there are casts in both macros which will defeat any type
checking whatsoever :(
Ie:
# define IN6_IS_ADDR_UNSPECIFIED(a) \
(__extension__ \
({ const struct in6_addr *__a = (const struct in6_addr *) (a); \
__a->s6_addr32[0] == 0 \
&& __a->s6_addr32[1] == 0 \
&& __a->s6_addr32[2] == 0 \
&& __a->s6_addr32[3] == 0; }))
Cheers,
Egbert.
_______________________________________________
[email protected]: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel
