Skip to contentSkip to conversation

tipc: call start and done ops directly in __tipc_nl_compat_dumpit()

PATCH Superseded Cong Wang 3 participants 6 messages Sep 4, 2018 – Sep 6, 2018

Subsystems (3)

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Cong Wang

author
#0 lore raw

PATCH 1/1 tipc: call start and done ops directly in __tipc_nl_compat_dumpit()

__tipc_nl_compat_dumpit() uses a netlink_callback on stack, so the only way to align it with other ->dumpit() call path is calling tipc_dump_start() and tipc_dump_done() directly inside it. Otherwise ->dumpit() would always get NULL from cb->args[0].

But, tipc_dump_start() uses sock_net(cb->skb->sk) to retrieve net pointer, the cb->skb here doesn't set ->sk, the net pointer is saved in msg->net instead, so introduce a helper function __tipc_dump_start() to pass in msg->net.

Fixes 9a07efa9aea2 ("tipc: switch to rhashtable iterator")
Reported-by [email protected]
Cc Jon Maloy <[email protected]>
Cc Ying Xue <[email protected]>
---
 net/tipc/netlink_compat.c | 2 ++
 net/tipc/socket.c         | 8 ++++++--
 net/tipc/socket.h         | 1 +
 3 files changed, 9 insertions(+), 2 deletions(-)
net/tipc/netlink_compat.c +2 -0
index a2f76743c73a..82f665728382 100644
@@ -185,6 +185,7 @@ static int __tipc_nl_compat_dumpit(struct tipc_nl_compat_cmd_dump *cmd,
 		return -ENOMEM;
 
 	buf->sk = msg->dst_sk;
+	__tipc_dump_start(&cb, msg->net);
 
 	do {
 		int rem;
@@ -216,6 +217,7 @@ static int __tipc_nl_compat_dumpit(struct tipc_nl_compat_cmd_dump *cmd,
 	err = 0;
 
 err_out:
+	tipc_dump_done(&cb);
 	kfree_skb(buf);
 
 	if (err == -EMSGSIZE) {
net/tipc/socket.c +6 -2
index ab7a2a7178f7..a19b2b1c77ed 100644
@@ -3264,9 +3264,14 @@ int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb,
 EXPORT_SYMBOL(tipc_nl_sk_walk);
 
 int tipc_dump_start(struct netlink_callback *cb)
+{
+	return __tipc_dump_start(cb, sock_net(cb->skb->sk));
+}
+EXPORT_SYMBOL(tipc_dump_start);
+
+int __tipc_dump_start(struct netlink_callback *cb, struct net *net)
 {
 	struct rhashtable_iter *iter = (void *)cb->args[0];
-	struct net *net = sock_net(cb->skb->sk);
 	struct tipc_net *tn = tipc_net(net);
 
 	if (!iter) {
@@ -3280,7 +3285,6 @@ int tipc_dump_start(struct netlink_callback *cb)
 	rhashtable_walk_enter(&tn->sk_rht, iter);
 	return 0;
 }
-EXPORT_SYMBOL(tipc_dump_start);
 
 int tipc_dump_done(struct netlink_callback *cb)
 {
net/tipc/socket.h +1 -0
index d43032e26532..5e575f205afe 100644
@@ -69,5 +69,6 @@ int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb,
 				       struct netlink_callback *cb,
 				       struct tipc_sock *tsk));
 int tipc_dump_start(struct netlink_callback *cb);
+int __tipc_dump_start(struct netlink_callback *cb, struct net *net);
 int tipc_dump_done(struct netlink_callback *cb);
 #endif
Signature · 1 line
2.14.4
author
#1 lore raw

PATCH 1/1 tipc: orphan sock in tipc_release()

Before we unlock the sock in tipc_release(), we have to detach sk->sk_socket from sk, otherwise a parallel tipc_sk_fill_sock_diag() could stil read it after we free this socket.

Fixes c30b70deb5f4 ("tipc: implement socket diagnostics for AF_TIPC")
Reported-and-tested-by [email protected]
Cc Jon Maloy <[email protected]>
Cc Ying Xue <[email protected]>
---
 net/tipc/socket.c | 1 +
 1 file changed, 1 insertion(+)
net/tipc/socket.c +1 -0
index a19b2b1c77ed..b5a6635e4dfa 100644
@@ -576,6 +576,7 @@ static int tipc_release(struct socket *sock)
 	sk_stop_timer(sk, &sk->sk_timer);
 	tipc_sk_remove(tsk);
 
+	sock_orphan(sk);
 	/* Reject any messages that accumulated in backlog queue */
 	release_sock(sk);
 	tipc_dest_list_purge(&tsk->cong_links);
Signature · 1 line
2.14.4
Acked-by on v1 1/1
#2 lore raw

Re: [Patch net] tipc: orphan sock in tipc_release()

Quoted 10 lines · Cong Wang

Before we unlock the sock in tipc_release(), we have to detach sk->sk_socket from sk, otherwise a parallel tipc_sk_fill_sock_diag() could stil read it after we free this socket.

Fixes c30b70deb5f4 ("tipc: implement socket diagnostics for AF_TIPC")
Reported-and-tested-by [email protected]
Cc Jon Maloy <[email protected]>
Cc Ying Xue <[email protected]>
Quoted 10 lines
---
 net/tipc/socket.c | 1 +
 1 file changed, 1 insertion(+)
net/tipc/socket.c +0 -0
index a19b2b1c77ed..b5a6635e4dfa 100644

net/tipc/socket.c static int tipc_release(struct socket *sock) · line 582

Quoted 7 lines
net/tipc/socket.c +1 -0
@@ -576,6 +576,7 @@ static int tipc_release(struct socket *sock)
 	sk_stop_timer(sk, &sk->sk_timer);
 	tipc_sk_remove(tsk);
 
+	sock_orphan(sk);
 	/* Reject any messages that accumulated in backlog queue */
 	release_sock(sk);
 	tipc_dest_list_purge(&tsk->cong_links);

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot

maintainer
#3 lore raw

Re: [Patch net] tipc: orphan sock in tipc_release()

In reply to Cong Wang

From: Cong Wang <[email protected]>
Date: Mon, 3 Sep 2018 19:12:41 -0700

Quoted 10 lines

Before we unlock the sock in tipc_release(), we have to detach sk->sk_socket from sk, otherwise a parallel tipc_sk_fill_sock_diag() could stil read it after we free this socket.

Fixes c30b70deb5f4 ("tipc: implement socket diagnostics for AF_TIPC")
Reported-and-tested-by [email protected]
Cc Jon Maloy <[email protected]>
Cc Ying Xue <[email protected]>

Applied and queued up for -stable, thanks.

#4 lore raw
In reply to Cong Wang
Quoted 5 lines · Cong Wang

__tipc_nl_compat_dumpit() uses a netlink_callback on stack, so the only way to align it with other ->dumpit() call path is calling tipc_dump_start() and tipc_dump_done() directly inside it. Otherwise ->dumpit() would always get NULL from cb->args[0].

Thank you for your fix Cong!

Your solution is fine with me.

When we align __tipc_nl_compat_dumpit() with ->dumpit() functions defined in tipc_genl_v2_ops[], cb->args[0] is used to save a rhashtable_iter object allocated in tipc_dump_start(), and the object will be retrieved with cb->args[0] in tipc_dump_done() and will be freed.

But unfortunately cb->args[0] has been used to other purposes in tipc_nl_bearer_dump(), tipc_nl_node_dump_link(), tipc_nl_name_table_dump(), tipc_nl_node_dump() and tipc_nl_net_dump(). It means cb->args[0] saved in __tipc_dump_start() will be overwritten in these ->dumpit() functions. As a consequence, not only the rhashtable_iter object allocated in tipc_dump_start() cannot be properly released in tipc_dump_done(), but also more kernel panics might be triggered in tipc_dump_done().

Quoted 77 lines

But, tipc_dump_start() uses sock_net(cb->skb->sk) to retrieve net pointer, the cb->skb here doesn't set ->sk, the net pointer is saved in msg->net instead, so introduce a helper function __tipc_dump_start() to pass in msg->net.

Fixes 9a07efa9aea2 ("tipc: switch to rhashtable iterator")
Reported-by [email protected]
Cc Jon Maloy <[email protected]>
Cc Ying Xue <[email protected]>
---
 net/tipc/netlink_compat.c | 2 ++
 net/tipc/socket.c         | 8 ++++++--
 net/tipc/socket.h         | 1 +
 3 files changed, 9 insertions(+), 2 deletions(-)
net/tipc/netlink_compat.c +2 -0
index a2f76743c73a..82f665728382 100644
@@ -185,6 +185,7 @@ static int __tipc_nl_compat_dumpit(struct tipc_nl_compat_cmd_dump *cmd,
 		return -ENOMEM;
 
 	buf->sk = msg->dst_sk;
+	__tipc_dump_start(&cb, msg->net);
 
 	do {
 		int rem;
@@ -216,6 +217,7 @@ static int __tipc_nl_compat_dumpit(struct tipc_nl_compat_cmd_dump *cmd,
 	err = 0;
 
 err_out:
+	tipc_dump_done(&cb);
 	kfree_skb(buf);
 
 	if (err == -EMSGSIZE) {
net/tipc/socket.c +6 -2
index ab7a2a7178f7..a19b2b1c77ed 100644
@@ -3264,9 +3264,14 @@ int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb,
 EXPORT_SYMBOL(tipc_nl_sk_walk);
 
 int tipc_dump_start(struct netlink_callback *cb)
+{
+	return __tipc_dump_start(cb, sock_net(cb->skb->sk));
+}
+EXPORT_SYMBOL(tipc_dump_start);
+
+int __tipc_dump_start(struct netlink_callback *cb, struct net *net)
 {
 	struct rhashtable_iter *iter = (void *)cb->args[0];
-	struct net *net = sock_net(cb->skb->sk);
 	struct tipc_net *tn = tipc_net(net);
 
 	if (!iter) {
@@ -3280,7 +3285,6 @@ int tipc_dump_start(struct netlink_callback *cb)
 	rhashtable_walk_enter(&tn->sk_rht, iter);
 	return 0;
 }
-EXPORT_SYMBOL(tipc_dump_start);
 
 int tipc_dump_done(struct netlink_callback *cb)
 {
net/tipc/socket.h +1 -0
index d43032e26532..5e575f205afe 100644
@@ -69,5 +69,6 @@ int tipc_nl_sk_walk(struct sk_buff *skb, struct netlink_callback *cb,
 				       struct netlink_callback *cb,
 				       struct tipc_sock *tsk));
 int tipc_dump_start(struct netlink_callback *cb);
+int __tipc_dump_start(struct netlink_callback *cb, struct net *net);
 int tipc_dump_done(struct netlink_callback *cb);
 #endif
author
#5 lore raw
Quoted 18 lines · Ying Xue
Quoted 5 lines · Cong Wang

__tipc_nl_compat_dumpit() uses a netlink_callback on stack, so the only way to align it with other ->dumpit() call path is calling tipc_dump_start() and tipc_dump_done() directly inside it. Otherwise ->dumpit() would always get NULL from cb->args[0].

Thank you for your fix Cong!

Your solution is fine with me.

When we align __tipc_nl_compat_dumpit() with ->dumpit() functions defined in tipc_genl_v2_ops[], cb->args[0] is used to save a rhashtable_iter object allocated in tipc_dump_start(), and the object will be retrieved with cb->args[0] in tipc_dump_done() and will be freed.

Quoted 8 lines

But unfortunately cb->args[0] has been used to other purposes in tipc_nl_bearer_dump(), tipc_nl_node_dump_link(), tipc_nl_name_table_dump(), tipc_nl_node_dump() and tipc_nl_net_dump(). It means cb->args[0] saved in __tipc_dump_start() will be overwritten in these ->dumpit() functions. As a consequence, not only the rhashtable_iter object allocated in tipc_dump_start() cannot be properly released in tipc_dump_done(), but also more kernel panics might be triggered in tipc_dump_done().

Ah, good catch!

The max utilization of cb->args is tipc_nl_name_table_dump():

net/tipc/name_table.c:  cb->args[0] = last_type;
net/tipc/name_table.c:  cb->args[1] = last_lower;
net/tipc/name_table.c:  cb->args[2] = last_key;
net/tipc/name_table.c:  cb->args[3] = done;

Looks like I should just use cb->args[4] for rhashtable iterator, as we still have some room in cb->args[].