All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Bluetooth: hci_sock: purge socket queues in the destruct() callback
@ 2021-10-07 19:04 ` Nguyen Dinh Phi
  0 siblings, 0 replies; 5+ messages in thread
From: Nguyen Dinh Phi @ 2021-10-07 19:04 UTC (permalink / raw)
  To: marcel, johan.hedberg, luiz.dentz, davem, kuba
  Cc: linux-bluetooth, netdev, linux-kernel, linux-kernel-mentees,
	syzbot+4c4ffd1e1094dae61035

The receive path may take the socket right before hci_sock_release(),
but it may enqueue the packets to the socket queues after the call to
skb_queue_purge(), therefore the socket can be destroyed without clear
its queues completely.

Moving these skb_queue_purge() to the hci_sock_destruct() will fix this
issue, because nothing is referencing the socket at this point.

Signed-off-by: Nguyen Dinh Phi <phind.uet@gmail.com>
Reported-by: syzbot+4c4ffd1e1094dae61035@syzkaller.appspotmail.com
---
 net/bluetooth/hci_sock.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index d0dad1fafe07..446573a12571 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -889,10 +889,6 @@ static int hci_sock_release(struct socket *sock)
 	}
 
 	sock_orphan(sk);
-
-	skb_queue_purge(&sk->sk_receive_queue);
-	skb_queue_purge(&sk->sk_write_queue);
-
 	release_sock(sk);
 	sock_put(sk);
 	return 0;
@@ -2058,6 +2054,12 @@ static int hci_sock_getsockopt(struct socket *sock, int level, int optname,
 	return err;
 }
 
+static void hci_sock_destruct(struct sock *sk)
+{
+	skb_queue_purge(&sk->sk_receive_queue);
+	skb_queue_purge(&sk->sk_write_queue);
+}
+
 static const struct proto_ops hci_sock_ops = {
 	.family		= PF_BLUETOOTH,
 	.owner		= THIS_MODULE,
@@ -2111,6 +2113,7 @@ static int hci_sock_create(struct net *net, struct socket *sock, int protocol,
 
 	sock->state = SS_UNCONNECTED;
 	sk->sk_state = BT_OPEN;
+	sk->sk_destruct = hci_sock_destruct;
 
 	bt_sock_link(&hci_sk_list, sk);
 	return 0;
-- 
2.25.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [PATCH] Bluetooth: hci_sock: purge socket queues in the destruct() callback
@ 2021-10-07 19:04 ` Nguyen Dinh Phi
  0 siblings, 0 replies; 5+ messages in thread
From: Nguyen Dinh Phi @ 2021-10-07 19:04 UTC (permalink / raw)
  To: marcel, johan.hedberg, luiz.dentz, davem, kuba
  Cc: linux-bluetooth, syzbot+4c4ffd1e1094dae61035,
	linux-kernel-mentees, linux-kernel, netdev

The receive path may take the socket right before hci_sock_release(),
but it may enqueue the packets to the socket queues after the call to
skb_queue_purge(), therefore the socket can be destroyed without clear
its queues completely.

Moving these skb_queue_purge() to the hci_sock_destruct() will fix this
issue, because nothing is referencing the socket at this point.

Signed-off-by: Nguyen Dinh Phi <phind.uet@gmail.com>
Reported-by: syzbot+4c4ffd1e1094dae61035@syzkaller.appspotmail.com
---
 net/bluetooth/hci_sock.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/net/bluetooth/hci_sock.c b/net/bluetooth/hci_sock.c
index d0dad1fafe07..446573a12571 100644
--- a/net/bluetooth/hci_sock.c
+++ b/net/bluetooth/hci_sock.c
@@ -889,10 +889,6 @@ static int hci_sock_release(struct socket *sock)
 	}
 
 	sock_orphan(sk);
-
-	skb_queue_purge(&sk->sk_receive_queue);
-	skb_queue_purge(&sk->sk_write_queue);
-
 	release_sock(sk);
 	sock_put(sk);
 	return 0;
@@ -2058,6 +2054,12 @@ static int hci_sock_getsockopt(struct socket *sock, int level, int optname,
 	return err;
 }
 
+static void hci_sock_destruct(struct sock *sk)
+{
+	skb_queue_purge(&sk->sk_receive_queue);
+	skb_queue_purge(&sk->sk_write_queue);
+}
+
 static const struct proto_ops hci_sock_ops = {
 	.family		= PF_BLUETOOTH,
 	.owner		= THIS_MODULE,
@@ -2111,6 +2113,7 @@ static int hci_sock_create(struct net *net, struct socket *sock, int protocol,
 
 	sock->state = SS_UNCONNECTED;
 	sk->sk_state = BT_OPEN;
+	sk->sk_destruct = hci_sock_destruct;
 
 	bt_sock_link(&hci_sk_list, sk);
 	return 0;
-- 
2.25.1

_______________________________________________
Linux-kernel-mentees mailing list
Linux-kernel-mentees@lists.linuxfoundation.org
https://lists.linuxfoundation.org/mailman/listinfo/linux-kernel-mentees

^ permalink raw reply related	[flat|nested] 5+ messages in thread

* RE: Bluetooth: hci_sock: purge socket queues in the destruct() callback
  2021-10-07 19:04 ` Nguyen Dinh Phi
  (?)
@ 2021-10-07 20:02 ` bluez.test.bot
  -1 siblings, 0 replies; 5+ messages in thread
From: bluez.test.bot @ 2021-10-07 20:02 UTC (permalink / raw)
  To: linux-bluetooth, phind.uet

[-- Attachment #1: Type: text/plain, Size: 936 bytes --]

This is automated email and please do not reply to this email!

Dear submitter,

Thank you for submitting the patches to the linux bluetooth mailing list.
This is a CI test results with your patch series:
PW Link:https://patchwork.kernel.org/project/bluetooth/list/?series=559391

---Test result---

Test Summary:
CheckPatch                    PASS      0.75 seconds
GitLint                       PASS      0.32 seconds
BuildKernel                   PASS      709.73 seconds
TestRunner: Setup             PASS      514.21 seconds
TestRunner: l2cap-tester      PASS      11.22 seconds
TestRunner: bnep-tester       PASS      5.71 seconds
TestRunner: mgmt-tester       PASS      95.57 seconds
TestRunner: rfcomm-tester     PASS      7.51 seconds
TestRunner: sco-tester        PASS      7.62 seconds
TestRunner: smp-tester        PASS      7.42 seconds
TestRunner: userchan-tester   PASS      6.19 seconds



---
Regards,
Linux Bluetooth


[-- Attachment #2: l2cap-tester.log --]
[-- Type: application/octet-stream, Size: 44356 bytes --]

[-- Attachment #3: bnep-tester.log --]
[-- Type: application/octet-stream, Size: 3563 bytes --]

[-- Attachment #4: mgmt-tester.log --]
[-- Type: application/octet-stream, Size: 637809 bytes --]

[-- Attachment #5: rfcomm-tester.log --]
[-- Type: application/octet-stream, Size: 11683 bytes --]

[-- Attachment #6: sco-tester.log --]
[-- Type: application/octet-stream, Size: 13923 bytes --]

[-- Attachment #7: smp-tester.log --]
[-- Type: application/octet-stream, Size: 11828 bytes --]

[-- Attachment #8: userchan-tester.log --]
[-- Type: application/octet-stream, Size: 6371 bytes --]

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] Bluetooth: hci_sock: purge socket queues in the destruct() callback
  2021-10-07 19:04 ` Nguyen Dinh Phi
@ 2021-10-12 15:38   ` Marcel Holtmann
  -1 siblings, 0 replies; 5+ messages in thread
From: Marcel Holtmann @ 2021-10-12 15:38 UTC (permalink / raw)
  To: Nguyen Dinh Phi
  Cc: Johan Hedberg, Luiz Augusto von Dentz, David S. Miller,
	Jakub Kicinski, linux-bluetooth, open list:NETWORKING [GENERAL],
	Linux Kernel Mailing List, linux-kernel-mentees,
	syzbot+4c4ffd1e1094dae61035

Hi Nguyen,

> The receive path may take the socket right before hci_sock_release(),
> but it may enqueue the packets to the socket queues after the call to
> skb_queue_purge(), therefore the socket can be destroyed without clear
> its queues completely.
> 
> Moving these skb_queue_purge() to the hci_sock_destruct() will fix this
> issue, because nothing is referencing the socket at this point.
> 
> Signed-off-by: Nguyen Dinh Phi <phind.uet@gmail.com>
> Reported-by: syzbot+4c4ffd1e1094dae61035@syzkaller.appspotmail.com
> ---
> net/bluetooth/hci_sock.c | 11 +++++++----
> 1 file changed, 7 insertions(+), 4 deletions(-)

patch has been applied to bluetooth-next tree.

Regards

Marcel


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] Bluetooth: hci_sock: purge socket queues in the destruct() callback
@ 2021-10-12 15:38   ` Marcel Holtmann
  0 siblings, 0 replies; 5+ messages in thread
From: Marcel Holtmann @ 2021-10-12 15:38 UTC (permalink / raw)
  To: Nguyen Dinh Phi
  Cc: Johan Hedberg, syzbot+4c4ffd1e1094dae61035,
	open list:NETWORKING [GENERAL],
	Linux Kernel Mailing List, linux-bluetooth,
	Luiz Augusto von Dentz, Jakub Kicinski, linux-kernel-mentees,
	David S. Miller

Hi Nguyen,

> The receive path may take the socket right before hci_sock_release(),
> but it may enqueue the packets to the socket queues after the call to
> skb_queue_purge(), therefore the socket can be destroyed without clear
> its queues completely.
> 
> Moving these skb_queue_purge() to the hci_sock_destruct() will fix this
> issue, because nothing is referencing the socket at this point.
> 
> Signed-off-by: Nguyen Dinh Phi <phind.uet@gmail.com>
> Reported-by: syzbot+4c4ffd1e1094dae61035@syzkaller.appspotmail.com
> ---
> net/bluetooth/hci_sock.c | 11 +++++++----
> 1 file changed, 7 insertions(+), 4 deletions(-)

patch has been applied to bluetooth-next tree.

Regards

Marcel

_______________________________________________
Linux-kernel-mentees mailing list
Linux-kernel-mentees@lists.linuxfoundation.org
https://lists.linuxfoundation.org/mailman/listinfo/linux-kernel-mentees

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2021-10-12 15:46 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-07 19:04 [PATCH] Bluetooth: hci_sock: purge socket queues in the destruct() callback Nguyen Dinh Phi
2021-10-07 19:04 ` Nguyen Dinh Phi
2021-10-07 20:02 ` bluez.test.bot
2021-10-12 15:38 ` [PATCH] " Marcel Holtmann
2021-10-12 15:38   ` Marcel Holtmann

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.