diff options
| author |   Jason A. Donenfeld <Jason@zx2c4.com> | 2018-02-20 02:23:08 +0100 |
|---|---|---|
| committer | Jason A. Donenfeld <Jason@zx2c4.com> | 2018-02-20 02:24:29 +0100 |
| commit | 740319127f14793a13ad385e8150cd98c715c20c (patch) | |
| tree | 20a5ac1a2a45fd7daf5c842d0b495c16a3bff9f0 | |
| parent | version: bump snapshot (diff) | |
| download | wireguard-monolithic-historical-740319127f14793a13ad385e8150cd98c715c20c.tar.xz wireguard-monolithic-historical-740319127f14793a13ad385e8150cd98c715c20c.zip | |
queueing: skb_reset: mark as xnet
This was avoided for a long time, because I wanted the packet to be
charged to the original socket for as long as possible. However, this
broke net_cls, which looks at skb->sk for additional late-stage routing
decisions. So we had no choice but to ensure that skb->sk is NULL by the
time of xmit, and this means calling the skb destructor.
| -rw-r--r-- | src/queueing.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/queueing.h b/src/queueing.h index de8b7b2..d5948f3 100644 --- a/src/queueing.h +++ b/src/queueing.h @@ -63,7 +63,7 @@ static inline __be16 skb_examine_untrusted_ip_hdr(struct sk_buff *skb) static inline void skb_reset(struct sk_buff *skb) { - skb_scrub_packet(skb, false); + skb_scrub_packet(skb, true); memset(&skb->headers_start, 0, offsetof(struct sk_buff, headers_end) - offsetof(struct sk_buff, headers_start)); skb->queue_mapping = 0; skb->nohdr = 0; |