Commit 795c687e authored by Jakub Kicinski's avatar Jakub Kicinski Committed by Greg Kroah-Hartman
Browse files

net/tls: free the record on encryption error 



[ Upstream commit d10523d0 ]

When tls_do_encryption() fails the SG lists are left with the
SG_END and SG_CHAIN marks in place. One could hope that once
encryption fails we will never see the record again, but that
is in fact not true. Commit d3b18ad3 ("tls: add bpf support
to sk_msg handling") added special handling to ENOMEM and ENOSPC
errors which mean we may see the same record re-submitted.

As suggested by John free the record, the BPF code is already
doing just that.

Reported-by: default avatar <syzbot+df0d4ec12332661dd1f9@syzkaller.appspotmail.com>
Fixes: d3b18ad3 ("tls: add bpf support to sk_msg handling")
Signed-off-by: default avatarJakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: default avatarSimon Horman <simon.horman@netronome.com>
Acked-by: default avatarJohn Fastabend <john.fastabend@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
parent a85ca002

net/tls/tls_sw.c

+8 −2
Original line number Diff line number Diff line
@@ -766,8 +766,14 @@ static int bpf_exec_tx_verdict(struct sk_msg *msg, struct sock *sk, 


	policy = !(flags & MSG_SENDPAGE_NOPOLICY);

	psock = sk_psock_get(sk);

	if (!psock || !policy)

		return tls_push_record(sk, flags, record_type);

	if (!psock || !policy) {

		err = tls_push_record(sk, flags, record_type);

		if (err) {

			*copied -= sk_msg_free(sk, msg);

			tls_free_open_rec(sk);

		}

		return err;

	}

more_data:

	enospc = sk_msg_full(msg);

	if (psock->eval == __SK_NONE) {