* [Patch net] tipc: check tsk->group in tipc_wait_for_cond()
@ 2018-12-12 5:43 Cong Wang
2018-12-12 11:25 ` Ying Xue
2018-12-14 23:49 ` David Miller
0 siblings, 2 replies; 3+ messages in thread
From: Cong Wang @ 2018-12-12 5:43 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, Ying Xue, Jon Maloy
tipc_wait_for_cond() drops socket lock before going to sleep,
but tsk->group could be freed right after that release_sock().
So we have to re-check and reload tsk->group after it wakes up.
After this patch, tipc_wait_for_cond() returns -ERESTARTSYS when
tsk->group is NULL, instead of continuing with the assumption of
a non-NULL tsk->group.
(It looks like 'dsts' should be re-checked and reloaded too, but
it is a different bug.)
Similar for tipc_send_group_unicast() and tipc_send_group_anycast().
Reported-by: syzbot+10a9db47c3a0e13eb31c@syzkaller.appspotmail.com
Fixes: b7d42635517f ("tipc: introduce flow control for group broadcast messages")
Fixes: ee106d7f942d ("tipc: introduce group anycast messaging")
Fixes: 27bd9ec027f3 ("tipc: introduce group unicast messaging")
Cc: Ying Xue <ying.xue@windriver.com>
Cc: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
---
net/tipc/socket.c | 25 ++++++++++++++-----------
1 file changed, 14 insertions(+), 11 deletions(-)
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index e1396fb87779..656940692a44 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -880,7 +880,6 @@ static int tipc_send_group_unicast(struct socket *sock, struct msghdr *m,
DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
int blks = tsk_blocks(GROUP_H_SIZE + dlen);
struct tipc_sock *tsk = tipc_sk(sk);
- struct tipc_group *grp = tsk->group;
struct net *net = sock_net(sk);
struct tipc_member *mb = NULL;
u32 node, port;
@@ -894,7 +893,9 @@ static int tipc_send_group_unicast(struct socket *sock, struct msghdr *m,
/* Block or return if destination link or member is congested */
rc = tipc_wait_for_cond(sock, &timeout,
!tipc_dest_find(&tsk->cong_links, node, 0) &&
- !tipc_group_cong(grp, node, port, blks, &mb));
+ tsk->group &&
+ !tipc_group_cong(tsk->group, node, port, blks,
+ &mb));
if (unlikely(rc))
return rc;
@@ -924,7 +925,6 @@ static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
struct tipc_sock *tsk = tipc_sk(sk);
struct list_head *cong_links = &tsk->cong_links;
int blks = tsk_blocks(GROUP_H_SIZE + dlen);
- struct tipc_group *grp = tsk->group;
struct tipc_msg *hdr = &tsk->phdr;
struct tipc_member *first = NULL;
struct tipc_member *mbr = NULL;
@@ -941,9 +941,10 @@ static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
type = msg_nametype(hdr);
inst = dest->addr.name.name.instance;
scope = msg_lookup_scope(hdr);
- exclude = tipc_group_exclude(grp);
while (++lookups < 4) {
+ exclude = tipc_group_exclude(tsk->group);
+
first = NULL;
/* Look for a non-congested destination member, if any */
@@ -952,7 +953,8 @@ static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
&dstcnt, exclude, false))
return -EHOSTUNREACH;
tipc_dest_pop(&dsts, &node, &port);
- cong = tipc_group_cong(grp, node, port, blks, &mbr);
+ cong = tipc_group_cong(tsk->group, node, port, blks,
+ &mbr);
if (!cong)
break;
if (mbr == first)
@@ -971,7 +973,8 @@ static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
/* Block or return if destination link or member is congested */
rc = tipc_wait_for_cond(sock, &timeout,
!tipc_dest_find(cong_links, node, 0) &&
- !tipc_group_cong(grp, node, port,
+ tsk->group &&
+ !tipc_group_cong(tsk->group, node, port,
blks, &mbr));
if (unlikely(rc))
return rc;
@@ -1006,8 +1009,7 @@ static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
struct sock *sk = sock->sk;
struct net *net = sock_net(sk);
struct tipc_sock *tsk = tipc_sk(sk);
- struct tipc_group *grp = tsk->group;
- struct tipc_nlist *dsts = tipc_group_dests(grp);
+ struct tipc_nlist *dsts = tipc_group_dests(tsk->group);
struct tipc_mc_method *method = &tsk->mc_method;
bool ack = method->mandatory && method->rcast;
int blks = tsk_blocks(MCAST_H_SIZE + dlen);
@@ -1020,8 +1022,9 @@ static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
return -EHOSTUNREACH;
/* Block or return if any destination link or member is congested */
- rc = tipc_wait_for_cond(sock, &timeout, !tsk->cong_link_cnt &&
- !tipc_group_bc_cong(grp, blks));
+ rc = tipc_wait_for_cond(sock, &timeout,
+ !tsk->cong_link_cnt && tsk->group &&
+ !tipc_group_bc_cong(tsk->group, blks));
if (unlikely(rc))
return rc;
@@ -1036,7 +1039,7 @@ static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
msg_set_hdr_sz(hdr, GROUP_H_SIZE);
msg_set_destport(hdr, 0);
msg_set_destnode(hdr, 0);
- msg_set_grp_bc_seqno(hdr, tipc_group_bc_snd_nxt(grp));
+ msg_set_grp_bc_seqno(hdr, tipc_group_bc_snd_nxt(tsk->group));
/* Avoid getting stuck with repeated forced replicasts */
msg_set_grp_bc_ack_req(hdr, ack);
--
2.19.2
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [Patch net] tipc: check tsk->group in tipc_wait_for_cond()
2018-12-12 5:43 [Patch net] tipc: check tsk->group in tipc_wait_for_cond() Cong Wang
@ 2018-12-12 11:25 ` Ying Xue
2018-12-14 23:49 ` David Miller
1 sibling, 0 replies; 3+ messages in thread
From: Ying Xue @ 2018-12-12 11:25 UTC (permalink / raw)
To: Cong Wang, netdev; +Cc: Jon Maloy
On 12/12/18 1:43 PM, Cong Wang wrote:
> tipc_wait_for_cond() drops socket lock before going to sleep,
> but tsk->group could be freed right after that release_sock().
> So we have to re-check and reload tsk->group after it wakes up.
>
> After this patch, tipc_wait_for_cond() returns -ERESTARTSYS when
> tsk->group is NULL, instead of continuing with the assumption of
> a non-NULL tsk->group.
>
> (It looks like 'dsts' should be re-checked and reloaded too, but
> it is a different bug.)
>
> Similar for tipc_send_group_unicast() and tipc_send_group_anycast().
>
> Reported-by: syzbot+10a9db47c3a0e13eb31c@syzkaller.appspotmail.com
> Fixes: b7d42635517f ("tipc: introduce flow control for group broadcast messages")
> Fixes: ee106d7f942d ("tipc: introduce group anycast messaging")
> Fixes: 27bd9ec027f3 ("tipc: introduce group unicast messaging")
> Cc: Ying Xue <ying.xue@windriver.com>
> Cc: Jon Maloy <jon.maloy@ericsson.com>
> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
> ---
> net/tipc/socket.c | 25 ++++++++++++++-----------
> 1 file changed, 14 insertions(+), 11 deletions(-)
>
> diff --git a/net/tipc/socket.c b/net/tipc/socket.c
> index e1396fb87779..656940692a44 100644
> --- a/net/tipc/socket.c
> +++ b/net/tipc/socket.c
> @@ -880,7 +880,6 @@ static int tipc_send_group_unicast(struct socket *sock, struct msghdr *m,
> DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
> int blks = tsk_blocks(GROUP_H_SIZE + dlen);
> struct tipc_sock *tsk = tipc_sk(sk);
> - struct tipc_group *grp = tsk->group;
> struct net *net = sock_net(sk);
> struct tipc_member *mb = NULL;
> u32 node, port;
> @@ -894,7 +893,9 @@ static int tipc_send_group_unicast(struct socket *sock, struct msghdr *m,
> /* Block or return if destination link or member is congested */
> rc = tipc_wait_for_cond(sock, &timeout,
> !tipc_dest_find(&tsk->cong_links, node, 0) &&
> - !tipc_group_cong(grp, node, port, blks, &mb));
> + tsk->group &&
> + !tipc_group_cong(tsk->group, node, port, blks,
> + &mb));
> if (unlikely(rc))
> return rc;
>
> @@ -924,7 +925,6 @@ static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
> struct tipc_sock *tsk = tipc_sk(sk);
> struct list_head *cong_links = &tsk->cong_links;
> int blks = tsk_blocks(GROUP_H_SIZE + dlen);
> - struct tipc_group *grp = tsk->group;
> struct tipc_msg *hdr = &tsk->phdr;
> struct tipc_member *first = NULL;
> struct tipc_member *mbr = NULL;
> @@ -941,9 +941,10 @@ static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
> type = msg_nametype(hdr);
> inst = dest->addr.name.name.instance;
> scope = msg_lookup_scope(hdr);
> - exclude = tipc_group_exclude(grp);
>
> while (++lookups < 4) {
> + exclude = tipc_group_exclude(tsk->group);
> +
> first = NULL;
>
> /* Look for a non-congested destination member, if any */
> @@ -952,7 +953,8 @@ static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
> &dstcnt, exclude, false))
> return -EHOSTUNREACH;
> tipc_dest_pop(&dsts, &node, &port);
> - cong = tipc_group_cong(grp, node, port, blks, &mbr);
> + cong = tipc_group_cong(tsk->group, node, port, blks,
> + &mbr);
> if (!cong)
> break;
> if (mbr == first)
> @@ -971,7 +973,8 @@ static int tipc_send_group_anycast(struct socket *sock, struct msghdr *m,
> /* Block or return if destination link or member is congested */
> rc = tipc_wait_for_cond(sock, &timeout,
> !tipc_dest_find(cong_links, node, 0) &&
> - !tipc_group_cong(grp, node, port,
> + tsk->group &&
> + !tipc_group_cong(tsk->group, node, port,
> blks, &mbr));
> if (unlikely(rc))
> return rc;
> @@ -1006,8 +1009,7 @@ static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
> struct sock *sk = sock->sk;
> struct net *net = sock_net(sk);
> struct tipc_sock *tsk = tipc_sk(sk);
> - struct tipc_group *grp = tsk->group;
> - struct tipc_nlist *dsts = tipc_group_dests(grp);
> + struct tipc_nlist *dsts = tipc_group_dests(tsk->group);
> struct tipc_mc_method *method = &tsk->mc_method;
> bool ack = method->mandatory && method->rcast;
> int blks = tsk_blocks(MCAST_H_SIZE + dlen);
> @@ -1020,8 +1022,9 @@ static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
> return -EHOSTUNREACH;
>
> /* Block or return if any destination link or member is congested */
> - rc = tipc_wait_for_cond(sock, &timeout, !tsk->cong_link_cnt &&
> - !tipc_group_bc_cong(grp, blks));
> + rc = tipc_wait_for_cond(sock, &timeout,
> + !tsk->cong_link_cnt && tsk->group &&
> + !tipc_group_bc_cong(tsk->group, blks));
> if (unlikely(rc))
> return rc;
>
> @@ -1036,7 +1039,7 @@ static int tipc_send_group_bcast(struct socket *sock, struct msghdr *m,
> msg_set_hdr_sz(hdr, GROUP_H_SIZE);
> msg_set_destport(hdr, 0);
> msg_set_destnode(hdr, 0);
> - msg_set_grp_bc_seqno(hdr, tipc_group_bc_snd_nxt(grp));
> + msg_set_grp_bc_seqno(hdr, tipc_group_bc_snd_nxt(tsk->group));
>
> /* Avoid getting stuck with repeated forced replicasts */
> msg_set_grp_bc_ack_req(hdr, ack);
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [Patch net] tipc: check tsk->group in tipc_wait_for_cond()
2018-12-12 5:43 [Patch net] tipc: check tsk->group in tipc_wait_for_cond() Cong Wang
2018-12-12 11:25 ` Ying Xue
@ 2018-12-14 23:49 ` David Miller
1 sibling, 0 replies; 3+ messages in thread
From: David Miller @ 2018-12-14 23:49 UTC (permalink / raw)
To: xiyou.wangcong; +Cc: netdev, ying.xue, jon.maloy
From: Cong Wang <xiyou.wangcong@gmail.com>
Date: Tue, 11 Dec 2018 21:43:51 -0800
> tipc_wait_for_cond() drops socket lock before going to sleep,
> but tsk->group could be freed right after that release_sock().
> So we have to re-check and reload tsk->group after it wakes up.
>
> After this patch, tipc_wait_for_cond() returns -ERESTARTSYS when
> tsk->group is NULL, instead of continuing with the assumption of
> a non-NULL tsk->group.
>
> (It looks like 'dsts' should be re-checked and reloaded too, but
> it is a different bug.)
>
> Similar for tipc_send_group_unicast() and tipc_send_group_anycast().
>
> Reported-by: syzbot+10a9db47c3a0e13eb31c@syzkaller.appspotmail.com
> Fixes: b7d42635517f ("tipc: introduce flow control for group broadcast messages")
> Fixes: ee106d7f942d ("tipc: introduce group anycast messaging")
> Fixes: 27bd9ec027f3 ("tipc: introduce group unicast messaging")
> Cc: Ying Xue <ying.xue@windriver.com>
> Cc: Jon Maloy <jon.maloy@ericsson.com>
> Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Applied and queued up for -stable, thanks Cong.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-12-14 23:49 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-12-12 5:43 [Patch net] tipc: check tsk->group in tipc_wait_for_cond() Cong Wang
2018-12-12 11:25 ` Ying Xue
2018-12-14 23:49 ` David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).