The if statement already move into sock_{put , hold},
just remove it.
Signed-off-by: Yajun Deng <[email protected]>
---
net/ax25/ax25_in.c | 3 +--
net/bluetooth/mgmt.c | 24 ++++++++----------------
net/can/af_can.c | 6 ++----
net/core/skmsg.c | 6 ++----
net/ipv4/udp_diag.c | 3 +--
net/l2tp/l2tp_ppp.c | 3 +--
net/mptcp/subflow.c | 8 ++------
net/netfilter/nf_queue.c | 8 ++------
net/netlink/af_netlink.c | 3 +--
net/netrom/af_netrom.c | 3 +--
net/phonet/pep.c | 4 +---
net/phonet/socket.c | 3 +--
net/qrtr/mhi.c | 9 +++------
net/unix/af_unix.c | 18 ++++++------------
net/unix/diag.c | 3 +--
net/vmw_vsock/af_vsock.c | 8 ++------
net/vmw_vsock/vmci_transport.c | 3 +--
17 files changed, 36 insertions(+), 79 deletions(-)
diff --git a/net/ax25/ax25_in.c b/net/ax25/ax25_in.c
index cd6afe895db9..713f75e3bbdc 100644
--- a/net/ax25/ax25_in.c
+++ b/net/ax25/ax25_in.c
@@ -380,8 +380,7 @@ static int ax25_rcv(struct sk_buff *skb, struct net_device *dev,
(ax25->digipeat = kmalloc(sizeof(ax25_digi), GFP_ATOMIC)) == NULL) {
kfree_skb(skb);
ax25_destroy_socket(ax25);
- if (sk)
- sock_put(sk);
+ sock_put(sk);
return 0;
}
diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c
index 3663f880df11..48c2299c8096 100644
--- a/net/bluetooth/mgmt.c
+++ b/net/bluetooth/mgmt.c
@@ -1896,8 +1896,7 @@ static void le_enable_complete(struct hci_dev *hdev, u8 status, u16 opcode)
new_settings(hdev, match.sk);
- if (match.sk)
- sock_put(match.sk);
+ sock_put(match.sk);
/* Make sure the controller has a good default for
* advertising data. Restrict the update to when LE
@@ -5335,8 +5334,7 @@ static void set_advertising_complete(struct hci_dev *hdev, u8 status,
new_settings(hdev, match.sk);
- if (match.sk)
- sock_put(match.sk);
+ sock_put(match.sk);
/* Handle suspend notifier */
if (test_and_clear_bit(SUSPEND_PAUSE_ADVERTISING,
@@ -8569,8 +8567,7 @@ void mgmt_power_on(struct hci_dev *hdev, int err)
new_settings(hdev, match.sk);
- if (match.sk)
- sock_put(match.sk);
+ sock_put(match.sk);
hci_dev_unlock(hdev);
}
@@ -8605,8 +8602,7 @@ void __mgmt_power_off(struct hci_dev *hdev)
new_settings(hdev, match.sk);
- if (match.sk)
- sock_put(match.sk);
+ sock_put(match.sk);
}
void mgmt_set_powered_failed(struct hci_dev *hdev, int err)
@@ -8887,8 +8883,7 @@ void mgmt_device_disconnected(struct hci_dev *hdev, bdaddr_t *bdaddr,
mgmt_event(MGMT_EV_DEVICE_DISCONNECTED, hdev, &ev, sizeof(ev), sk);
- if (sk)
- sock_put(sk);
+ sock_put(sk);
mgmt_pending_foreach(MGMT_OP_UNPAIR_DEVICE, hdev, unpair_device_rsp,
hdev);
@@ -9114,8 +9109,7 @@ void mgmt_auth_enable_complete(struct hci_dev *hdev, u8 status)
if (changed)
new_settings(hdev, match.sk);
- if (match.sk)
- sock_put(match.sk);
+ sock_put(match.sk);
}
static void clear_eir(struct hci_request *req)
@@ -9169,8 +9163,7 @@ void mgmt_ssp_enable_complete(struct hci_dev *hdev, u8 enable, u8 status)
if (changed)
new_settings(hdev, match.sk);
- if (match.sk)
- sock_put(match.sk);
+ sock_put(match.sk);
hci_req_init(&req, hdev);
@@ -9211,8 +9204,7 @@ void mgmt_set_class_of_dev_complete(struct hci_dev *hdev, u8 *dev_class,
ext_info_changed(hdev, NULL);
}
- if (match.sk)
- sock_put(match.sk);
+ sock_put(match.sk);
}
void mgmt_set_local_name_complete(struct hci_dev *hdev, u8 *name, u8 status)
diff --git a/net/can/af_can.c b/net/can/af_can.c
index cce2af10eb3e..7f6c6969e203 100644
--- a/net/can/af_can.c
+++ b/net/can/af_can.c
@@ -493,8 +493,7 @@ static void can_rx_delete_receiver(struct rcu_head *rp)
struct sock *sk = rcv->sk;
kmem_cache_free(rcv_cache, rcv);
- if (sk)
- sock_put(sk);
+ sock_put(sk);
}
/**
@@ -562,8 +561,7 @@ void can_rx_unregister(struct net *net, struct net_device *dev, canid_t can_id,
/* schedule the receiver item for deletion */
if (rcv) {
- if (rcv->sk)
- sock_hold(rcv->sk);
+ sock_hold(rcv->sk);
call_rcu(&rcv->rcu, can_rx_delete_receiver);
}
}
diff --git a/net/core/skmsg.c b/net/core/skmsg.c
index 2d6249b28928..187f3f91bbae 100644
--- a/net/core/skmsg.c
+++ b/net/core/skmsg.c
@@ -782,8 +782,7 @@ static void sk_psock_destroy(struct work_struct *work)
sk_psock_link_destroy(psock);
sk_psock_cork_free(psock);
- if (psock->sk_redir)
- sock_put(psock->sk_redir);
+ sock_put(psock->sk_redir);
sock_put(psock->sk);
kfree(psock);
}
@@ -838,8 +837,7 @@ int sk_psock_msg_verdict(struct sock *sk, struct sk_psock *psock,
ret = sk_psock_map_verd(ret, msg->sk_redir);
psock->apply_bytes = msg->apply_bytes;
if (ret == __SK_REDIRECT) {
- if (psock->sk_redir)
- sock_put(psock->sk_redir);
+ sock_put(psock->sk_redir);
psock->sk_redir = msg->sk_redir;
if (!psock->sk_redir) {
ret = __SK_DROP;
diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c
index 1ed8c4d78e5c..629e3f112b08 100644
--- a/net/ipv4/udp_diag.c
+++ b/net/ipv4/udp_diag.c
@@ -80,8 +80,7 @@ static int udp_dump_one(struct udp_table *tbl,
err = nlmsg_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid);
out:
- if (sk)
- sock_put(sk);
+ sock_put(sk);
out_nosk:
return err;
}
diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c
index bf35710127dd..e8f01f1708c8 100644
--- a/net/l2tp/l2tp_ppp.c
+++ b/net/l2tp/l2tp_ppp.c
@@ -141,8 +141,7 @@ static struct sock *pppol2tp_session_get_sock(struct l2tp_session *session)
rcu_read_lock();
sk = rcu_dereference(ps->sk);
- if (sk)
- sock_hold(sk);
+ sock_hold(sk);
rcu_read_unlock();
return sk;
diff --git a/net/mptcp/subflow.c b/net/mptcp/subflow.c
index 966f777d35ce..9fd6737d8a2f 100644
--- a/net/mptcp/subflow.c
+++ b/net/mptcp/subflow.c
@@ -41,9 +41,7 @@ static void subflow_req_destructor(struct request_sock *req)
pr_debug("subflow_req=%p", subflow_req);
- if (subflow_req->msk)
- sock_put((struct sock *)subflow_req->msk);
-
+ sock_put((struct sock *)subflow_req->msk);
mptcp_token_destroy_request(req);
tcp_request_sock_ops.destructor(req);
}
@@ -601,9 +599,7 @@ static void subflow_drop_ctx(struct sock *ssk)
return;
subflow_ulp_fallback(ssk, ctx);
- if (ctx->conn)
- sock_put(ctx->conn);
-
+ sock_put(ctx->conn);
kfree_rcu(ctx, rcu);
}
diff --git a/net/netfilter/nf_queue.c b/net/netfilter/nf_queue.c
index 7f2f69b609d8..a7a199293186 100644
--- a/net/netfilter/nf_queue.c
+++ b/net/netfilter/nf_queue.c
@@ -53,9 +53,7 @@ static void nf_queue_entry_release_refs(struct nf_queue_entry *entry)
/* Release those devices we held, or Alexey will kill me. */
dev_put(state->in);
dev_put(state->out);
- if (state->sk)
- sock_put(state->sk);
-
+ sock_put(state->sk);
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
dev_put(entry->physin);
dev_put(entry->physout);
@@ -93,9 +91,7 @@ void nf_queue_entry_get_refs(struct nf_queue_entry *entry)
dev_hold(state->in);
dev_hold(state->out);
- if (state->sk)
- sock_hold(state->sk);
-
+ sock_hold(state->sk);
#if IS_ENABLED(CONFIG_BRIDGE_NETFILTER)
dev_hold(entry->physin);
dev_hold(entry->physout);
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 24b7cf447bc5..1aad9528416b 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -532,8 +532,7 @@ static struct sock *netlink_lookup(struct net *net, int protocol, u32 portid)
rcu_read_lock();
sk = __netlink_lookup(table, portid, net);
- if (sk)
- sock_hold(sk);
+ sock_hold(sk);
rcu_read_unlock();
return sk;
diff --git a/net/netrom/af_netrom.c b/net/netrom/af_netrom.c
index 6d16e1ab1a8a..dde89b6c2432 100644
--- a/net/netrom/af_netrom.c
+++ b/net/netrom/af_netrom.c
@@ -958,8 +958,7 @@ int nr_rx_frame(struct sk_buff *skb, struct net_device *dev)
if (sk == NULL || sk_acceptq_is_full(sk) ||
(make = nr_make_new(sk)) == NULL) {
nr_transmit_refusal(skb, 0);
- if (sk)
- sock_put(sk);
+ sock_put(sk);
return 0;
}
diff --git a/net/phonet/pep.c b/net/phonet/pep.c
index a1525916885a..2c7ddf6dbb76 100644
--- a/net/phonet/pep.c
+++ b/net/phonet/pep.c
@@ -1314,9 +1314,7 @@ static void pep_sock_unhash(struct sock *sk)
if (hlist_empty(&pn->hlist))
pn_sock_unhash(&pn->pn_sk.sk);
release_sock(sk);
-
- if (skparent)
- sock_put(skparent);
+ sock_put(skparent);
}
static struct proto pep_proto = {
diff --git a/net/phonet/socket.c b/net/phonet/socket.c
index 71e2caf6ab85..0e7de755c1aa 100644
--- a/net/phonet/socket.c
+++ b/net/phonet/socket.c
@@ -620,8 +620,7 @@ struct sock *pn_find_sock_by_res(struct net *net, u8 res)
rcu_read_lock();
sk = rcu_dereference(pnres.sk[res]);
- if (sk)
- sock_hold(sk);
+ sock_hold(sk);
rcu_read_unlock();
return sk;
}
diff --git a/net/qrtr/mhi.c b/net/qrtr/mhi.c
index 1dc955ca57d3..7980d18409af 100644
--- a/net/qrtr/mhi.c
+++ b/net/qrtr/mhi.c
@@ -40,8 +40,7 @@ static void qcom_mhi_qrtr_ul_callback(struct mhi_device *mhi_dev,
{
struct sk_buff *skb = mhi_res->buf_addr;
- if (skb->sk)
- sock_put(skb->sk);
+ sock_put(skb->sk);
consume_skb(skb);
}
@@ -55,8 +54,7 @@ static int qcom_mhi_qrtr_send(struct qrtr_endpoint *ep, struct sk_buff *skb)
if (rc)
goto free_skb;
- if (skb->sk)
- sock_hold(skb->sk);
+ sock_hold(skb->sk);
rc = skb_linearize(skb);
if (rc)
@@ -70,8 +68,7 @@ static int qcom_mhi_qrtr_send(struct qrtr_endpoint *ep, struct sk_buff *skb)
return rc;
free_skb:
- if (skb->sk)
- sock_put(skb->sk);
+ sock_put(skb->sk);
kfree_skb(skb);
return rc;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index ec02e70a549b..f02787e98114 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -206,8 +206,7 @@ struct sock *unix_peer_get(struct sock *s)
unix_state_lock(s);
peer = unix_peer(s);
- if (peer)
- sock_hold(peer);
+ sock_hold(peer);
unix_state_unlock(s);
return peer;
}
@@ -311,8 +310,7 @@ static inline struct sock *unix_find_socket_byname(struct net *net,
spin_lock(&unix_table_lock);
s = __unix_find_socket_byname(net, sunname, len, hash);
- if (s)
- sock_hold(s);
+ sock_hold(s);
spin_unlock(&unix_table_lock);
return s;
}
@@ -1439,8 +1437,7 @@ static int unix_stream_connect(struct socket *sock, struct sockaddr *uaddr,
kfree_skb(skb);
if (newsk)
unix_release_sock(newsk, 0);
- if (other)
- sock_put(other);
+ sock_put(other);
return err;
}
@@ -1884,8 +1881,7 @@ static int unix_dgram_sendmsg(struct socket *sock, struct msghdr *msg,
out_free:
kfree_skb(skb);
out:
- if (other)
- sock_put(other);
+ sock_put(other);
scm_destroy(&scm);
return err;
}
@@ -2765,8 +2761,7 @@ static int unix_shutdown(struct socket *sock, int mode)
unix_state_lock(sk);
sk->sk_shutdown |= mode;
other = unix_peer(sk);
- if (other)
- sock_hold(other);
+ sock_hold(other);
unix_state_unlock(sk);
sk->sk_state_change(sk);
@@ -2788,8 +2783,7 @@ static int unix_shutdown(struct socket *sock, int mode)
else if (peer_mode & RCV_SHUTDOWN)
sk_wake_async(other, SOCK_WAKE_WAITD, POLL_IN);
}
- if (other)
- sock_put(other);
+ sock_put(other);
return 0;
}
diff --git a/net/unix/diag.c b/net/unix/diag.c
index 7e7d7f45685a..e6a03d806b4a 100644
--- a/net/unix/diag.c
+++ b/net/unix/diag.c
@@ -298,8 +298,7 @@ static int unix_diag_get_exact(struct sk_buff *in_skb,
err = nlmsg_unicast(net->diag_nlsk, rep, NETLINK_CB(in_skb).portid);
out:
- if (sk)
- sock_put(sk);
+ sock_put(sk);
out_nosk:
return err;
}
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 3e02cc3b24f8..eccdbdb43808 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -301,9 +301,7 @@ struct sock *vsock_find_bound_socket(struct sockaddr_vm *addr)
spin_lock_bh(&vsock_table_lock);
sk = __vsock_find_bound_socket(addr);
- if (sk)
- sock_hold(sk);
-
+ sock_hold(sk);
spin_unlock_bh(&vsock_table_lock);
return sk;
@@ -317,9 +315,7 @@ struct sock *vsock_find_connected_socket(struct sockaddr_vm *src,
spin_lock_bh(&vsock_table_lock);
sk = __vsock_find_connected_socket(src, dst);
- if (sk)
- sock_hold(sk);
-
+ sock_hold(sk);
spin_unlock_bh(&vsock_table_lock);
return sk;
diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c
index 7aef34e32bdf..944d5302a222 100644
--- a/net/vmw_vsock/vmci_transport.c
+++ b/net/vmw_vsock/vmci_transport.c
@@ -792,8 +792,7 @@ static int vmci_transport_recv_stream_cb(void *data, struct vmci_datagram *dg)
}
out:
- if (sk)
- sock_put(sk);
+ sock_put(sk);
return err;
}
--
2.32.0
August 6, 2021 4:58 PM, "Matthieu Baerts" <[email protected]> wrote:
> Hi Yajun,
>
> Thank you for sharing this patch.
>
> On 06/08/2021 08:38, Yajun Deng wrote:
>
>> The if statement already move into sock_{put , hold},
>> just remove it.
>
> I was wondering in which subtree you had 'sock_put' checking the socket
> pointer but then I realised you sent another patch just before adding
> this check: "net: sock: add the case if sk is NULL"
>
> Please next time send them in the same series to clearly indicate that
> this is the 2nd patch (2/2) and it depends on patch 1/2.
>
OK, Thank you for your advice.
> Related to the modification in MPTCP part: it looks OK but we do a few
> other calls to 'sock_put()' where we don't need to check if the socket
> is NULL or not.
>
> In other words, if your patch "net: sock: add the case if sk is NULL" is
> accepted, then the modification in "net/mptcp/subflow.c" is OK for us.
>
> Cheers,
> Matt
> --
> Tessares | Belgium | Hybrid Access Solutions
> http://www.tessares.net
Hi Yajun,
Thank you for sharing this patch.
On 06/08/2021 08:38, Yajun Deng wrote:
> The if statement already move into sock_{put , hold},
> just remove it.
I was wondering in which subtree you had 'sock_put' checking the socket
pointer but then I realised you sent another patch just before adding
this check: "net: sock: add the case if sk is NULL"
Please next time send them in the same series to clearly indicate that
this is the 2nd patch (2/2) and it depends on patch 1/2.
Related to the modification in MPTCP part: it looks OK but we do a few
other calls to 'sock_put()' where we don't need to check if the socket
is NULL or not.
In other words, if your patch "net: sock: add the case if sk is NULL" is
accepted, then the modification in "net/mptcp/subflow.c" is OK for us.
Cheers,
Matt
--
Tessares | Belgium | Hybrid Access Solutions
http://www.tessares.net