[PATCH net] sctp: cache netns in sctp_ep_common

* [PATCH net] sctp: cache netns in sctp_ep_common
@ 2019-11-23  3:56 Xin Long
  2019-11-23 12:56 ` Marcelo Ricardo Leitner
  2019-11-24  2:29 ` Jakub Kicinski
  0 siblings, 2 replies; 3+ messages in thread
From: Xin Long @ 2019-11-23  3:56 UTC (permalink / raw)
  To: network dev, linux-sctp; +Cc: davem, Marcelo Ricardo Leitner, Neil Horman

This patch is to fix a data-race reported by syzbot:

  BUG: KCSAN: data-race in sctp_assoc_migrate / sctp_hash_obj

  write to 0xffff8880b67c0020 of 8 bytes by task 18908 on cpu 1:
    sctp_assoc_migrate+0x1a6/0x290 net/sctp/associola.c:1091
    sctp_sock_migrate+0x8aa/0x9b0 net/sctp/socket.c:9465
    sctp_accept+0x3c8/0x470 net/sctp/socket.c:4916
    inet_accept+0x7f/0x360 net/ipv4/af_inet.c:734
    __sys_accept4+0x224/0x430 net/socket.c:1754
    __do_sys_accept net/socket.c:1795 [inline]
    __se_sys_accept net/socket.c:1792 [inline]
    __x64_sys_accept+0x4e/0x60 net/socket.c:1792
    do_syscall_64+0xcc/0x370 arch/x86/entry/common.c:290
    entry_SYSCALL_64_after_hwframe+0x44/0xa9

  read to 0xffff8880b67c0020 of 8 bytes by task 12003 on cpu 0:
    sctp_hash_obj+0x4f/0x2d0 net/sctp/input.c:894
    rht_key_get_hash include/linux/rhashtable.h:133 [inline]
    rht_key_hashfn include/linux/rhashtable.h:159 [inline]
    rht_head_hashfn include/linux/rhashtable.h:174 [inline]
    head_hashfn lib/rhashtable.c:41 [inline]
    rhashtable_rehash_one lib/rhashtable.c:245 [inline]
    rhashtable_rehash_chain lib/rhashtable.c:276 [inline]
    rhashtable_rehash_table lib/rhashtable.c:316 [inline]
    rht_deferred_worker+0x468/0xab0 lib/rhashtable.c:420
    process_one_work+0x3d4/0x890 kernel/workqueue.c:2269
    worker_thread+0xa0/0x800 kernel/workqueue.c:2415
    kthread+0x1d4/0x200 drivers/block/aoe/aoecmd.c:1253
    ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:352

It was caused by rhashtable access asoc->base.sk when sctp_assoc_migrate
is changing its value. However, what rhashtable wants is netns from asoc
base.sk, and for an asoc, its netns won't change once set. So we can
simply fix it by caching netns since created.

Fixes: d6c0256a60e6 ("sctp: add the rhashtable apis for sctp global transport hashtable")
Reported-by: syzbot+e3b35fe7918ff0ee474e@syzkaller.appspotmail.com
Signed-off-by: Xin Long <lucien.xin@gmail.com>
---
 include/net/sctp/structs.h | 3 +++
 net/sctp/associola.c       | 1 +
 net/sctp/endpointola.c     | 1 +
 net/sctp/input.c           | 4 ++--
 4 files changed, 7 insertions(+), 2 deletions(-)

diff --git a/include/net/sctp/structs.h b/include/net/sctp/structs.h
index 503fbc3..2b6f3f1 100644
--- a/include/net/sctp/structs.h
+++ b/include/net/sctp/structs.h
@@ -1239,6 +1239,9 @@ struct sctp_ep_common {
 	/* What socket does this endpoint belong to?  */
 	struct sock *sk;
 
+	/* Cache netns and it won't change once set */
+	struct net *net;
+
 	/* This is where we receive inbound chunks.  */
 	struct sctp_inq	  inqueue;
 
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index d2ffc9a..41839b8 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -64,6 +64,7 @@ static struct sctp_association *sctp_association_init(
 	/* Discarding const is appropriate here.  */
 	asoc->ep = (struct sctp_endpoint *)ep;
 	asoc->base.sk = (struct sock *)sk;
+	asoc->base.net = sock_net(sk);
 
 	sctp_endpoint_hold(asoc->ep);
 	sock_hold(asoc->base.sk);
diff --git a/net/sctp/endpointola.c b/net/sctp/endpointola.c
index ea53049..3067deb 100644
--- a/net/sctp/endpointola.c
+++ b/net/sctp/endpointola.c
@@ -110,6 +110,7 @@ static struct sctp_endpoint *sctp_endpoint_init(struct sctp_endpoint *ep,
 
 	/* Remember who we are attached to.  */
 	ep->base.sk = sk;
+	ep->base.net = sock_net(sk);
 	sock_hold(ep->base.sk);
 
 	return ep;
diff --git a/net/sctp/input.c b/net/sctp/input.c
index 2277981..4d2bcfc 100644
--- a/net/sctp/input.c
+++ b/net/sctp/input.c
@@ -882,7 +882,7 @@ static inline int sctp_hash_cmp(struct rhashtable_compare_arg *arg,
 	if (!sctp_transport_hold(t))
 		return err;
 
-	if (!net_eq(sock_net(t->asoc->base.sk), x->net))
+	if (!net_eq(t->asoc->base.net, x->net))
 		goto out;
 	if (x->lport != htons(t->asoc->base.bind_addr.port))
 		goto out;
@@ -897,7 +897,7 @@ static inline __u32 sctp_hash_obj(const void *data, u32 len, u32 seed)
 {
 	const struct sctp_transport *t = data;
 
-	return sctp_hashfn(sock_net(t->asoc->base.sk),
+	return sctp_hashfn(t->asoc->base.net,
 			   htons(t->asoc->base.bind_addr.port),
 			   &t->ipaddr, seed);
 }
-- 
2.1.0


^ permalink raw reply related	[flat|nested] 3+ messages in thread