* [PATCH net] udp: fix race between close() and udp_abort()
@ 2021-06-09 9:49 Paolo Abeni
2021-06-09 21:20 ` patchwork-bot+netdevbpf
0 siblings, 1 reply; 2+ messages in thread
From: Paolo Abeni @ 2021-06-09 9:49 UTC (permalink / raw)
To: netdev; +Cc: Kaustubh Pandey, David Ahern, David S . Miller, Jakub Kicinski
Kaustubh reported and diagnosed a panic in udp_lib_lookup().
The root cause is udp_abort() racing with close(). Both
racing functions acquire the socket lock, but udp{v6}_destroy_sock()
release it before performing destructive actions.
We can't easily extend the socket lock scope to avoid the race,
instead use the SOCK_DEAD flag to prevent udp_abort from doing
any action when the critical race happens.
Diagnosed-and-tested-by: Kaustubh Pandey <kapandey@codeaurora.org>
Fixes: 5d77dca82839 ("net: diag: support SOCK_DESTROY for UDP sockets")
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
net/ipv4/udp.c | 10 ++++++++++
net/ipv6/udp.c | 3 +++
2 files changed, 13 insertions(+)
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 055fceb18bea..bf98a87ac6e6 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -2607,6 +2607,9 @@ void udp_destroy_sock(struct sock *sk)
{
struct udp_sock *up = udp_sk(sk);
bool slow = lock_sock_fast(sk);
+
+ /* protects from races with udp_abort() */
+ sock_set_flag(sk, SOCK_DEAD);
udp_flush_pending_frames(sk);
unlock_sock_fast(sk, slow);
if (static_branch_unlikely(&udp_encap_needed_key)) {
@@ -2857,10 +2860,17 @@ int udp_abort(struct sock *sk, int err)
{
lock_sock(sk);
+ /* udp{v6}_destroy_sock() sets it under the sk lock, avoid racing
+ * with close()
+ */
+ if (sock_flag(sk, SOCK_DEAD))
+ goto out;
+
sk->sk_err = err;
sk->sk_error_report(sk);
__udp_disconnect(sk, 0);
+out:
release_sock(sk);
return 0;
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 199b080d418a..3fcd86f4dfdc 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1598,6 +1598,9 @@ void udpv6_destroy_sock(struct sock *sk)
{
struct udp_sock *up = udp_sk(sk);
lock_sock(sk);
+
+ /* protects from races with udp_abort() */
+ sock_set_flag(sk, SOCK_DEAD);
udp_v6_flush_pending_frames(sk);
release_sock(sk);
--
2.26.3
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH net] udp: fix race between close() and udp_abort()
2021-06-09 9:49 [PATCH net] udp: fix race between close() and udp_abort() Paolo Abeni
@ 2021-06-09 21:20 ` patchwork-bot+netdevbpf
0 siblings, 0 replies; 2+ messages in thread
From: patchwork-bot+netdevbpf @ 2021-06-09 21:20 UTC (permalink / raw)
To: Paolo Abeni; +Cc: netdev, kapandey, dsa, davem, kuba
Hello:
This patch was applied to netdev/net.git (refs/heads/master):
On Wed, 9 Jun 2021 11:49:01 +0200 you wrote:
> Kaustubh reported and diagnosed a panic in udp_lib_lookup().
> The root cause is udp_abort() racing with close(). Both
> racing functions acquire the socket lock, but udp{v6}_destroy_sock()
> release it before performing destructive actions.
>
> We can't easily extend the socket lock scope to avoid the race,
> instead use the SOCK_DEAD flag to prevent udp_abort from doing
> any action when the critical race happens.
>
> [...]
Here is the summary with links:
- [net] udp: fix race between close() and udp_abort()
https://git.kernel.org/netdev/net/c/a8b897c7bcd4
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-06-09 21:20 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-06-09 9:49 [PATCH net] udp: fix race between close() and udp_abort() Paolo Abeni
2021-06-09 21:20 ` patchwork-bot+netdevbpf
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.