From d54072587146dd0db9bb52b513234d944edabda3 Mon Sep 17 00:00:00 2001 From: "Jason A. Donenfeld" Date: Tue, 2 Mar 2021 09:24:45 +0100 Subject: kernel-5.4: backport fd16931a2f51 for chacha neon MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Without this patch, the chacha block counter is not incremented on neon rounds, resulting in incorrect calculations and corrupt packets. This also switches to using `--no-numbered --zero-commit` so that future diffs are smaller. Reported-by: Hans Geiblinger Reviewed-by: Ilya Lipnitskiy Cc: David Bauer Cc: Petr Štetiar Signed-off-by: Jason A. Donenfeld --- ...ocket-mark-skbs-as-not-on-list-when-recei.patch | 34 ++++++++++++++++++++++ 1 file changed, 34 insertions(+) create mode 100644 target/linux/generic/backport-5.4/080-wireguard-0082-wireguard-socket-mark-skbs-as-not-on-list-when-recei.patch (limited to 'target/linux/generic/backport-5.4/080-wireguard-0082-wireguard-socket-mark-skbs-as-not-on-list-when-recei.patch') diff --git a/target/linux/generic/backport-5.4/080-wireguard-0082-wireguard-socket-mark-skbs-as-not-on-list-when-recei.patch b/target/linux/generic/backport-5.4/080-wireguard-0082-wireguard-socket-mark-skbs-as-not-on-list-when-recei.patch new file mode 100644 index 0000000000..779491c8db --- /dev/null +++ b/target/linux/generic/backport-5.4/080-wireguard-0082-wireguard-socket-mark-skbs-as-not-on-list-when-recei.patch @@ -0,0 +1,34 @@ +From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 +From: "Jason A. Donenfeld" +Date: Thu, 2 Jan 2020 17:47:51 +0100 +Subject: [PATCH] wireguard: socket: mark skbs as not on list when receiving + via gro + +commit 736775d06bac60d7a353e405398b48b2bd8b1e54 upstream. + +Certain drivers will pass gro skbs to udp, at which point the udp driver +simply iterates through them and passes them off to encap_rcv, which is +where we pick up. At the moment, we're not attempting to coalesce these +into bundles, but we also don't want to wind up having cascaded lists of +skbs treated separately. The right behavior here, then, is to just mark +each incoming one as not on a list. This can be seen in practice, for +example, with Qualcomm's rmnet_perf driver. + +Signed-off-by: Jason A. Donenfeld +Tested-by: Yaroslav Furman +Signed-off-by: David S. Miller +Signed-off-by: Jason A. Donenfeld +--- + drivers/net/wireguard/socket.c | 1 + + 1 file changed, 1 insertion(+) + +--- a/drivers/net/wireguard/socket.c ++++ b/drivers/net/wireguard/socket.c +@@ -333,6 +333,7 @@ static int wg_receive(struct sock *sk, s + wg = sk->sk_user_data; + if (unlikely(!wg)) + goto err; ++ skb_mark_not_on_list(skb); + wg_packet_receive(wg, skb); + return 0; + -- cgit v1.2.3