[RFC bpf-next v2 5/9] flow_dissector: allow access only to a subset of __sk_buff fields

[Stanislav Fomichev]

Use whitelist instead of a blacklist and allow only a small set of
fields that might be relevant in the context of flow dissector:
* len
* protocol
* vlan_present
* vlan_tci
* vlan_proto
* cb

This is required for the eth_get_headlen case where we construct
temporary skb which might not have full/consistent state to let flow
dissector programs access all the fields (which are irrelevant in for
flow dissector program type).

Signed-off-by: Stanislav Fomichev <s...@google.com>
---
 net/core/filter.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/net/core/filter.c b/net/core/filter.c
index 647c63a7b25b..5f413567ce8a 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -6632,11 +6632,14 @@ static bool flow_dissector_is_valid_access(int off, int 
size,
        case bpf_ctx_range_ptr(struct __sk_buff, flow_keys):
                info->reg_type = PTR_TO_FLOW_KEYS;
                break;
-       case bpf_ctx_range(struct __sk_buff, tc_classid):
-       case bpf_ctx_range(struct __sk_buff, data_meta):
-       case bpf_ctx_range_till(struct __sk_buff, family, local_port):
-       case bpf_ctx_range(struct __sk_buff, tstamp):
-       case bpf_ctx_range(struct __sk_buff, wire_len):
+       case bpf_ctx_range(struct __sk_buff, len):
+       case bpf_ctx_range(struct __sk_buff, protocol):
+       case bpf_ctx_range(struct __sk_buff, vlan_present):
+       case bpf_ctx_range(struct __sk_buff, vlan_tci):
+       case bpf_ctx_range(struct __sk_buff, vlan_proto):
+       case bpf_ctx_range_till(struct __sk_buff, cb[0], cb[4]):
+               break;
+       default:
                return false;
        }
 
-- 
2.21.0.225.g810b269d1ac-goog