All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Eric Biggers <ebiggers@google.com>,
	syzbot+6bf095f9becf5efef645@syzkaller.appspotmail.com,
	syzbot+31c16aa4202dace3812e@syzkaller.appspotmail.com,
	Jakub Kicinski <jakub.kicinski@netronome.com>,
	Sasha Levin <sashal@kernel.org>,
	netdev@vger.kernel.org
Subject: [PATCH AUTOSEL 4.4 07/16] llc: fix sk_buff leak in llc_sap_state_process()
Date: Fri, 25 Oct 2019 09:58:31 -0400	[thread overview]
Message-ID: <20191025135842.25977-7-sashal@kernel.org> (raw)
In-Reply-To: <20191025135842.25977-1-sashal@kernel.org>

From: Eric Biggers <ebiggers@google.com>

[ Upstream commit c6ee11c39fcc1fb55130748990a8f199e76263b4 ]

syzbot reported:

    BUG: memory leak
    unreferenced object 0xffff888116270800 (size 224):
       comm "syz-executor641", pid 7047, jiffies 4294947360 (age 13.860s)
       hex dump (first 32 bytes):
         00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
         00 20 e1 2a 81 88 ff ff 00 40 3d 2a 81 88 ff ff  . .*.....@=*....
       backtrace:
         [<000000004d41b4cc>] kmemleak_alloc_recursive  include/linux/kmemleak.h:55 [inline]
         [<000000004d41b4cc>] slab_post_alloc_hook mm/slab.h:439 [inline]
         [<000000004d41b4cc>] slab_alloc_node mm/slab.c:3269 [inline]
         [<000000004d41b4cc>] kmem_cache_alloc_node+0x153/0x2a0 mm/slab.c:3579
         [<00000000506a5965>] __alloc_skb+0x6e/0x210 net/core/skbuff.c:198
         [<000000001ba5a161>] alloc_skb include/linux/skbuff.h:1058 [inline]
         [<000000001ba5a161>] alloc_skb_with_frags+0x5f/0x250  net/core/skbuff.c:5327
         [<0000000047d9c78b>] sock_alloc_send_pskb+0x269/0x2a0  net/core/sock.c:2225
         [<000000003828fe54>] sock_alloc_send_skb+0x32/0x40 net/core/sock.c:2242
         [<00000000e34d94f9>] llc_ui_sendmsg+0x10a/0x540 net/llc/af_llc.c:933
         [<00000000de2de3fb>] sock_sendmsg_nosec net/socket.c:652 [inline]
         [<00000000de2de3fb>] sock_sendmsg+0x54/0x70 net/socket.c:671
         [<000000008fe16e7a>] __sys_sendto+0x148/0x1f0 net/socket.c:1964
	 [...]

The bug is that llc_sap_state_process() always takes an extra reference
to the skb, but sometimes neither llc_sap_next_state() nor
llc_sap_state_process() itself drops this reference.

Fix it by changing llc_sap_next_state() to never consume a reference to
the skb, rather than sometimes do so and sometimes not.  Then remove the
extra skb_get() and kfree_skb() from llc_sap_state_process().

Reported-by: syzbot+6bf095f9becf5efef645@syzkaller.appspotmail.com
Reported-by: syzbot+31c16aa4202dace3812e@syzkaller.appspotmail.com
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 net/llc/llc_s_ac.c | 12 +++++++++---
 net/llc/llc_sap.c  | 23 ++++++++---------------
 2 files changed, 17 insertions(+), 18 deletions(-)

diff --git a/net/llc/llc_s_ac.c b/net/llc/llc_s_ac.c
index a94bd56bcac6f..7ae4cc684d3ab 100644
--- a/net/llc/llc_s_ac.c
+++ b/net/llc/llc_s_ac.c
@@ -58,8 +58,10 @@ int llc_sap_action_send_ui(struct llc_sap *sap, struct sk_buff *skb)
 			    ev->daddr.lsap, LLC_PDU_CMD);
 	llc_pdu_init_as_ui_cmd(skb);
 	rc = llc_mac_hdr_init(skb, ev->saddr.mac, ev->daddr.mac);
-	if (likely(!rc))
+	if (likely(!rc)) {
+		skb_get(skb);
 		rc = dev_queue_xmit(skb);
+	}
 	return rc;
 }
 
@@ -81,8 +83,10 @@ int llc_sap_action_send_xid_c(struct llc_sap *sap, struct sk_buff *skb)
 			    ev->daddr.lsap, LLC_PDU_CMD);
 	llc_pdu_init_as_xid_cmd(skb, LLC_XID_NULL_CLASS_2, 0);
 	rc = llc_mac_hdr_init(skb, ev->saddr.mac, ev->daddr.mac);
-	if (likely(!rc))
+	if (likely(!rc)) {
+		skb_get(skb);
 		rc = dev_queue_xmit(skb);
+	}
 	return rc;
 }
 
@@ -135,8 +139,10 @@ int llc_sap_action_send_test_c(struct llc_sap *sap, struct sk_buff *skb)
 			    ev->daddr.lsap, LLC_PDU_CMD);
 	llc_pdu_init_as_test_cmd(skb);
 	rc = llc_mac_hdr_init(skb, ev->saddr.mac, ev->daddr.mac);
-	if (likely(!rc))
+	if (likely(!rc)) {
+		skb_get(skb);
 		rc = dev_queue_xmit(skb);
+	}
 	return rc;
 }
 
diff --git a/net/llc/llc_sap.c b/net/llc/llc_sap.c
index 5404d0d195cc5..d51ff9df9c959 100644
--- a/net/llc/llc_sap.c
+++ b/net/llc/llc_sap.c
@@ -197,29 +197,22 @@ static int llc_sap_next_state(struct llc_sap *sap, struct sk_buff *skb)
  *	After executing actions of the event, upper layer will be indicated
  *	if needed(on receiving an UI frame). sk can be null for the
  *	datalink_proto case.
+ *
+ *	This function always consumes a reference to the skb.
  */
 static void llc_sap_state_process(struct llc_sap *sap, struct sk_buff *skb)
 {
 	struct llc_sap_state_ev *ev = llc_sap_ev(skb);
 
-	/*
-	 * We have to hold the skb, because llc_sap_next_state
-	 * will kfree it in the sending path and we need to
-	 * look at the skb->cb, where we encode llc_sap_state_ev.
-	 */
-	skb_get(skb);
 	ev->ind_cfm_flag = 0;
 	llc_sap_next_state(sap, skb);
-	if (ev->ind_cfm_flag == LLC_IND) {
-		if (skb->sk->sk_state == TCP_LISTEN)
-			kfree_skb(skb);
-		else {
-			llc_save_primitive(skb->sk, skb, ev->prim);
 
-			/* queue skb to the user. */
-			if (sock_queue_rcv_skb(skb->sk, skb))
-				kfree_skb(skb);
-		}
+	if (ev->ind_cfm_flag == LLC_IND && skb->sk->sk_state != TCP_LISTEN) {
+		llc_save_primitive(skb->sk, skb, ev->prim);
+
+		/* queue skb to the user. */
+		if (sock_queue_rcv_skb(skb->sk, skb) == 0)
+			return;
 	}
 	kfree_skb(skb);
 }
-- 
2.20.1


  parent reply	other threads:[~2019-10-25 14:00 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-25 13:58 [PATCH AUTOSEL 4.4 01/16] PCI/ASPM: Do not initialize link state when aspm_disabled is set Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 02/16] ACPI: video: Use vendor backlight on Sony VPCEH3U1E Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 03/16] driver core: platform: Fix the usage of platform device name(pdev->name) Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 04/16] powerpc/pseries/hvconsole: Fix stack overread via udbg Sasha Levin
2019-10-25 13:58   ` Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 05/16] scsi: RDMA/srp: Fix a sleep-in-invalid-context bug Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 06/16] sch_netem: fix rcu splat in netem_enqueue() Sasha Levin
2019-10-25 13:58 ` Sasha Levin [this message]
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 08/16] llc: fix sk_buff leak in llc_conn_service() Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 09/16] bonding: fix potential NULL deref in bond_update_slave_arr Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 10/16] net: usb: sr9800: fix uninitialized local variable Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 11/16] jbd2: flush_descriptor(): Do not decrease buffer head's ref count Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 12/16] ath9k: dynack: fix possible deadlock in ath_dynack_node_{de}init Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 13/16] Btrfs: fix hang when loading existing inode cache off disk Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 14/16] ubi: ubi_wl_get_peb: Increase the number of attempts while getting PEB Sasha Levin
2019-10-25 13:58   ` Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 15/16] RDMA/iwcm: Fix a lock inversion issue Sasha Levin
2019-10-25 13:58 ` [PATCH AUTOSEL 4.4 16/16] bdi: Do not use freezable workqueue Sasha Levin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20191025135842.25977-7-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=ebiggers@google.com \
    --cc=jakub.kicinski@netronome.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=syzbot+31c16aa4202dace3812e@syzkaller.appspotmail.com \
    --cc=syzbot+6bf095f9becf5efef645@syzkaller.appspotmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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.