Re: [PATCH nf 1/1] netfilter: ebtables: fix OOB read in compat_mtw_from_user
Florian Westphal <[email protected]> Fri, 24 Apr 2026 11:52:11 +0200
| Newsgroups | dev.linux.lists.bridge,org.kernel.vger.netfilter-devel |
|---|---|
| Message-ID | <[email protected]> |
Ren Wei <[email protected]> wrote: > From: Luxiao Xu <[email protected]> > > The function compat_mtw_from_user() converts ebtables extensions from > 32-bit user structures to kernel native structures. However, it lacks > proper validation of the user-supplied match_size/target_size. > > When certain extensions are processed, the kernel-side translation > logic may perform memory accesses based on the extension's expected > size. If the user provides a size smaller than what the extension > requires, it results in an out-of-bounds read as reported by KASAN. > > This fix introduces a check to ensure match_size is at least as large > as the extension's required compatsize. This covers matches, watchers, > and targets, while maintaining compatibility with standard targets. > > Fixes: 81e675c227ec ("netfilter: ebtables: add CONFIG_COMPAT support") > Cc: [email protected] > Reported-by: Yuan Tan <[email protected]> > Reported-by: Yifan Wu <[email protected]> > Reported-by: Juefei Pu <[email protected]> > Reported-by: Xin Liu <[email protected]> > Signed-off-by: Luxiao Xu <[email protected]> > Signed-off-by: Ren Wei <[email protected]> > --- > net/bridge/netfilter/ebtables.c | 11 +++++++++++ > 1 file changed, 11 insertions(+) > > diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c > index aea3e19875c6..80cd0233c088 100644 > --- a/net/bridge/netfilter/ebtables.c > +++ b/net/bridge/netfilter/ebtables.c > @@ -1977,6 +1977,11 @@ static int compat_mtw_from_user(const struct compat_ebt_entry_mwt *mwt, > if (IS_ERR(match)) > return PTR_ERR(match); > > + if (match_size < match->compatsize) { > + module_put(match->me); > + return -EINVAL; > + } > + Are you sure this catches all bad requests? AFAIR compatsize is 0 in most cases, which bypasses this test. should this be: u16 csize = match->compatsize ? : match->matchsize; ... if (match_size < csize) { ... ? @Pablo: I think the 32bit compat layer should be removed in -next, or at least strongly discouraged and slated for removal soon.