summaryrefslogtreecommitdiffstats
path: root/Documentation/bpf
diff options
context:
space:
mode:
authorStanislav Fomichev <sdf@google.com>2019-07-26 00:52:26 +0200
committerAlexei Starovoitov <ast@kernel.org>2019-07-26 03:00:41 +0200
commit1ac6b126dbe8108c1fe6e29d023a0f2989cd3fea (patch)
tree26829d01b35fb65cb2796d2e4effb3c941a4c77e /Documentation/bpf
parentbpf/flow_dissector: pass input flags to BPF flow dissector program (diff)
downloadlinux-1ac6b126dbe8108c1fe6e29d023a0f2989cd3fea.tar.xz
linux-1ac6b126dbe8108c1fe6e29d023a0f2989cd3fea.zip
bpf/flow_dissector: document flags
Describe what each input flag does and who uses it. Acked-by: Petar Penkov <ppenkov@google.com> Acked-by: Willem de Bruijn <willemb@google.com> Acked-by: Song Liu <songliubraving@fb.com> Cc: Song Liu <songliubraving@fb.com> Cc: Willem de Bruijn <willemb@google.com> Cc: Petar Penkov <ppenkov@google.com> Signed-off-by: Stanislav Fomichev <sdf@google.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
Diffstat (limited to 'Documentation/bpf')
-rw-r--r--Documentation/bpf/prog_flow_dissector.rst18
1 files changed, 18 insertions, 0 deletions
diff --git a/Documentation/bpf/prog_flow_dissector.rst b/Documentation/bpf/prog_flow_dissector.rst
index ed343abe541e..a78bf036cadd 100644
--- a/Documentation/bpf/prog_flow_dissector.rst
+++ b/Documentation/bpf/prog_flow_dissector.rst
@@ -26,6 +26,7 @@ The inputs are:
* ``nhoff`` - initial offset of the networking header
* ``thoff`` - initial offset of the transport header, initialized to nhoff
* ``n_proto`` - L3 protocol type, parsed out of L2 header
+ * ``flags`` - optional flags
Flow dissector BPF program should fill out the rest of the ``struct
bpf_flow_keys`` fields. Input arguments ``nhoff/thoff/n_proto`` should be
@@ -101,6 +102,23 @@ can be called for both cases and would have to be written carefully to
handle both cases.
+Flags
+=====
+
+``flow_keys->flags`` might contain optional input flags that work as follows:
+
+* ``BPF_FLOW_DISSECTOR_F_PARSE_1ST_FRAG`` - tells BPF flow dissector to
+ continue parsing first fragment; the default expected behavior is that
+ flow dissector returns as soon as it finds out that the packet is fragmented;
+ used by ``eth_get_headlen`` to estimate length of all headers for GRO.
+* ``BPF_FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL`` - tells BPF flow dissector to
+ stop parsing as soon as it reaches IPv6 flow label; used by
+ ``___skb_get_hash`` and ``__skb_get_hash_symmetric`` to get flow hash.
+* ``BPF_FLOW_DISSECTOR_F_STOP_AT_ENCAP`` - tells BPF flow dissector to stop
+ parsing as soon as it reaches encapsulated headers; used by routing
+ infrastructure.
+
+
Reference Implementation
========================