[PATCH bpf-next v3] bpf: check for BPF_F_ADJ_ROOM_FIXED_GSO when bpf_skb_change_proto

[Dongseok Yi <dseok.yi@samsung.com>]

In the forwarding path GRO -> BPF 6 to 4 -> GSO for TCP traffic, the
coalesced packet payload can be > MSS, but < MSS + 20.
bpf_skb_proto_6_to_4 will upgrade the MSS and it can be > the payload
length. After then tcp_gso_segment checks for the payload length if it
is <= MSS. The condition is causing the packet to be dropped.

tcp_gso_segment():
        [...]
        mss = skb_shinfo(skb)->gso_size;
        if (unlikely(skb->len <= mss))
                goto out;
        [...]

Allow to upgrade/downgrade MSS only when BPF_F_ADJ_ROOM_FIXED_GSO is
not set.

Signed-off-by: Dongseok Yi <dseok.yi@samsung.com>
---
 net/core/filter.c | 22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

v2:
per Willem de Bruijn request,
checked the flag instead of a generic approach.

v3:
per Willem de Bruijn request,
moved to bpf-next, supported for both 6_to_4 and 4_to_6.

diff --git a/net/core/filter.c b/net/core/filter.c
index cae56d0..582ac19 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -3235,7 +3235,7 @@ static int bpf_skb_net_hdr_pop(struct sk_buff *skb, u32 off, u32 len)
 	return ret;
 }
 
-static int bpf_skb_proto_4_to_6(struct sk_buff *skb)
+static int bpf_skb_proto_4_to_6(struct sk_buff *skb, u64 flags)
 {
 	const u32 len_diff = sizeof(struct ipv6hdr) - sizeof(struct iphdr);
 	u32 off = skb_mac_header_len(skb);
@@ -3264,7 +3264,9 @@ static int bpf_skb_proto_4_to_6(struct sk_buff *skb)
 		}
 
 		/* Due to IPv6 header, MSS needs to be downgraded. */
-		skb_decrease_gso_size(shinfo, len_diff);
+		if (!(flags & BPF_F_ADJ_ROOM_FIXED_GSO))
+			skb_decrease_gso_size(shinfo, len_diff);
+
 		/* Header must be checked, and gso_segs recomputed. */
 		shinfo->gso_type |= SKB_GSO_DODGY;
 		shinfo->gso_segs = 0;
@@ -3276,7 +3278,7 @@ static int bpf_skb_proto_4_to_6(struct sk_buff *skb)
 	return 0;
 }
 
-static int bpf_skb_proto_6_to_4(struct sk_buff *skb)
+static int bpf_skb_proto_6_to_4(struct sk_buff *skb, u64 flags)
 {
 	const u32 len_diff = sizeof(struct ipv6hdr) - sizeof(struct iphdr);
 	u32 off = skb_mac_header_len(skb);
@@ -3305,7 +3307,9 @@ static int bpf_skb_proto_6_to_4(struct sk_buff *skb)
 		}
 
 		/* Due to IPv4 header, MSS can be upgraded. */
-		skb_increase_gso_size(shinfo, len_diff);
+		if (!(flags & BPF_F_ADJ_ROOM_FIXED_GSO))
+			skb_increase_gso_size(shinfo, len_diff);
+
 		/* Header must be checked, and gso_segs recomputed. */
 		shinfo->gso_type |= SKB_GSO_DODGY;
 		shinfo->gso_segs = 0;
@@ -3317,17 +3321,17 @@ static int bpf_skb_proto_6_to_4(struct sk_buff *skb)
 	return 0;
 }
 
-static int bpf_skb_proto_xlat(struct sk_buff *skb, __be16 to_proto)
+static int bpf_skb_proto_xlat(struct sk_buff *skb, __be16 to_proto, u64 flags)
 {
 	__be16 from_proto = skb->protocol;
 
 	if (from_proto == htons(ETH_P_IP) &&
 	      to_proto == htons(ETH_P_IPV6))
-		return bpf_skb_proto_4_to_6(skb);
+		return bpf_skb_proto_4_to_6(skb, flags);
 
 	if (from_proto == htons(ETH_P_IPV6) &&
 	      to_proto == htons(ETH_P_IP))
-		return bpf_skb_proto_6_to_4(skb);
+		return bpf_skb_proto_6_to_4(skb, flags);
 
 	return -ENOTSUPP;
 }
@@ -3337,7 +3341,7 @@ BPF_CALL_3(bpf_skb_change_proto, struct sk_buff *, skb, __be16, proto,
 {
 	int ret;
 
-	if (unlikely(flags))
+	if (unlikely(flags & ~(BPF_F_ADJ_ROOM_FIXED_GSO)))
 		return -EINVAL;
 
 	/* General idea is that this helper does the basic groundwork
@@ -3357,7 +3361,7 @@ BPF_CALL_3(bpf_skb_change_proto, struct sk_buff *, skb, __be16, proto,
 	 * that. For offloads, we mark packet as dodgy, so that headers
 	 * need to be verified first.
 	 */
-	ret = bpf_skb_proto_xlat(skb, proto);
+	ret = bpf_skb_proto_xlat(skb, proto, flags);
 	bpf_compute_data_pointers(skb);
 	return ret;
 }
-- 
2.7.4