On 26 Dec 2021, at 19:14, Gleb Smirnoff <[email protected]> wrote: > > Jessica, > > On Sun, Dec 26, 2021 at 07:07:06PM +0000, Jessica Clarke wrote: > J> > +struct prison_ip { > J> > + struct epoch_context ctx; > J> > + uint32_t ips; > J> > +#ifdef FUTURE_C > J> > + union { > J> > + struct in_addr pr_ip4[]; > J> > + struct in6_addr pr_ip6[]; > J> > + }; > J> > +#else /* No future C :( */ > J> > +#define PR_IP(pip, i) ((const char *)((pip) + 1) + > pr_families[af].size * (i)) > J> > +#define PR_IPD(pip, i) ((char *)((pip) + 1) + pr_families[af].size * > (i)) > J> > +#endif > J> > +}; > J> > J> You can make this work with a prison_ip base and prison_ipv[46] derived > J> structs. > J> > J> As it stands this is quite gross, you’re assuming things about > J> alignment, and don’t even have a flexible char[] at the end to > J> represent the extra data. > > Will adding char [] to the end be sufficient to guarantee proper alignment?
No. > Using prison_ipv[46] derived structs won't work as most functions are now > made protocol independent. Why not? Just cast from prison_ip * to prison_ipv4 * when known/needed? Jess
