Florian Westphal <fw@xxxxxxxxx> wrote: > remaining problem: > > The ebtables userland makefile has: > ifeq ($(shell uname -m),sparc64) > CFLAGS+=-DEBT_MIN_ALIGN=8 -DKERNEL_64_USERSPACE_32 > endif > > struct ebt_replace, ebt_entry_match etc. then contain userland-side > padding -- this means, if this is actually used in production -- > that even if we are called from a 32 bit userland, the structures > may already be in the right format. > > One possible solution would be to try to handle compat_* calls via > the native handler and then fall back to the compat one if it fails, > but it will cause kernel warnings like: > kernel msg: ebtables bug: please report to author: Wrong size > > Thus this is not done here. > > I am not sure if this "userland-padding" is widely used; it looks > incomplete (eg. I couldn't find any > userland-padding for match/target structures; and some (eg. mark) > do need special handling). Just for the record: Jan Engelhardt was so kind to try "-j mark" with the 32bit sparc ebtables userland and reported that it is not working. So I am not sure if I should bother with supporting the "already padded" userland scenario, as it won't work in some cases anyway. -- To unsubscribe from this list: send the line "unsubscribe netfilter-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
