summaryrefslogtreecommitdiff
path: root/fs/autofs
diff options
context:
space:
mode:
authorFlorian Westphal <fw@strlen.de>2020-03-23 17:34:30 +0100
committerPablo Neira Ayuso <pablo@netfilter.org>2020-03-24 20:00:12 +0100
commita64d558d8cf98424cc5eb9ae6631782cd8bf789c (patch)
tree42a2e03e5bfa5e70bd281e9fedfe9f196786ed00 /fs/autofs
parentbcfabee1afd99484b6ba067361b8678e28bbc065 (diff)
selftests: netfilter: add nfqueue test case
Add a test case to check nf queue infrastructure. Could be extended in the future to also cover serialization of conntrack, uid and secctx attributes in nfqueue. For now, this checks that 'queue bypass' works, that a queue rule with no bypass option blocks traffic and that userspace receives the expected number of packets. For this we add two queues and hook all of prerouting/input/forward/output/postrouting. Packets get queued twice with a dummy base chain in between: This passes with current nf tree, but reverting commit 946c0d8e6ed4 ("netfilter: nf_queue: fix reinject verdict handling") makes this trip (it processes 30 instead of expected 20 packets). v2: update config file with queue and other options missing/needed for other tests. v3: also test with tcp, this reveals problem with commit 28f8bfd1ac94 ("netfilter: Support iif matches in POSTROUTING"), due to skb->dev pointing at another skb in the retransmit rbtree (skb->dev aliases to rbnode child). Signed-off-by: Florian Westphal <fw@strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Diffstat (limited to 'fs/autofs')
0 files changed, 0 insertions, 0 deletions