On 7/12/19 2:17 PM, Cong Wang wrote: > diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c > index 317339cd7f03..8662a44a28f9 100644 > --- a/net/ipv4/fib_frontend.c > +++ b/net/ipv4/fib_frontend.c > @@ -388,6 +388,12 @@ static int __fib_validate_source(struct sk_buff *skb, > __be32 src, __be32 dst, > fib_combine_itag(itag, &res); > > dev_match = fib_info_nh_uses_dev(res.fi, dev); > + /* This is rare, loopback packets retain skb_dst so normally they > + * would not even hit this slow path. > + */ > + dev_match = dev_match || (res.type == RTN_LOCAL && > + dev == net->loopback_dev &&
The dev should not be needed. res.type == RTN_LOCAL should be enough, no? > + IN_DEV_ACCEPT_LOCAL(idev)); Why is this check needed? Can you give an example use that is fixed - and add one to selftests/net/fib_tests.sh? > if (dev_match) { > ret = FIB_RES_NHC(res)->nhc_scope >= RT_SCOPE_HOST; > return ret; >