KASAN: use-after-free Read in j1939_xtp_rx_dat_one

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit: 0adc313c Merge tag 'gfs2-for-5.10' of git://git.kernel.org..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=134ebaef900000
kernel config: https://syzkaller.appspot.com/x/.config?x=df5c8291513455a2
dashboard link: https://syzkaller.appspot.com/bug?extid=220c1a29987a9a490903
compiler: gcc (GCC) 10.1.0-syz 20200507
userspace arch: i386

Unfortunately, I don't have any reproducer for this issue yet.

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+220c1a...@syzkaller.appspotmail.com

vcan0: j1939_xtp_rx_dat_one: 0x000000009262b4a1: Data of RX-looped back packet (00 ff ff ff ff ff ff) doesn't match TX data (00 00 00 00 00 00 00)!
==================================================================
BUG: KASAN: use-after-free in j1939_xtp_rx_dat_one+0x108d/0x1130 net/can/j1939/transport.c:1825
Read of size 1 at addr ffff8880219f308e by task ksoftirqd/0/10

CPU: 0 PID: 10 Comm: ksoftirqd/0 Not tainted 5.9.0-syzkaller #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.0-59-gc9ba5276e321-prebuilt.qemu.org 04/01/2014
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x107/0x163 lib/dump_stack.c:118
print_address_description.constprop.0.cold+0xae/0x4c8 mm/kasan/report.c:385
__kasan_report mm/kasan/report.c:545 [inline]
kasan_report.cold+0x1f/0x37 mm/kasan/report.c:562
j1939_xtp_rx_dat_one+0x108d/0x1130 net/can/j1939/transport.c:1825
j1939_xtp_rx_dat net/can/j1939/transport.c:1875 [inline]
j1939_tp_recv+0x544/0xb40 net/can/j1939/transport.c:2057
j1939_can_recv+0x5bc/0x7d0 net/can/j1939/main.c:101
deliver net/can/af_can.c:571 [inline]
can_rcv_filter+0x5d4/0x8d0 net/can/af_can.c:605
can_receive+0x2e3/0x520 net/can/af_can.c:662
can_rcv+0x12a/0x1a0 net/can/af_can.c:688
__netif_receive_skb_one_core+0x114/0x180 net/core/dev.c:5311
__netif_receive_skb+0x27/0x1c0 net/core/dev.c:5425
process_backlog+0x232/0x6c0 net/core/dev.c:6315
napi_poll net/core/dev.c:6759 [inline]
net_rx_action+0x4dc/0x1100 net/core/dev.c:6829
__do_softirq+0x2a0/0x9f6 kernel/softirq.c:298
run_ksoftirqd kernel/softirq.c:653 [inline]
run_ksoftirqd+0x2d/0x50 kernel/softirq.c:645
smpboot_thread_fn+0x655/0x9e0 kernel/smpboot.c:165
kthread+0x3af/0x4a0 kernel/kthread.c:292
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:296

Allocated by task 15076:
kasan_save_stack+0x1b/0x40 mm/kasan/common.c:48
kasan_set_track mm/kasan/common.c:56 [inline]
__kasan_kmalloc.constprop.0+0xc2/0xd0 mm/kasan/common.c:461
slab_post_alloc_hook mm/slab.h:526 [inline]
slab_alloc_node mm/slub.c:2891 [inline]
kmem_cache_alloc_node+0x132/0x480 mm/slub.c:2927
__alloc_skb+0x71/0x550 net/core/skbuff.c:198
alloc_skb include/linux/skbuff.h:1094 [inline]
alloc_skb_with_frags+0x92/0x570 net/core/skbuff.c:5832
sock_alloc_send_pskb+0x72a/0x880 net/core/sock.c:2329
j1939_sk_alloc_skb net/can/j1939/socket.c:857 [inline]
j1939_sk_send_loop net/can/j1939/socket.c:1039 [inline]
j1939_sk_sendmsg+0x6bb/0x1380 net/can/j1939/socket.c:1174
sock_sendmsg_nosec net/socket.c:651 [inline]
sock_sendmsg+0xcf/0x120 net/socket.c:671
sock_no_sendpage+0xee/0x130 net/core/sock.c:2833
kernel_sendpage.part.0+0x1ab/0x350 net/socket.c:3646
kernel_sendpage net/socket.c:3643 [inline]
sock_sendpage+0xe5/0x140 net/socket.c:944
pipe_to_sendpage+0x2ad/0x380 fs/splice.c:364
splice_from_pipe_feed fs/splice.c:418 [inline]
__splice_from_pipe+0x3dc/0x830 fs/splice.c:562
splice_from_pipe fs/splice.c:597 [inline]
generic_splice_sendpage+0xd4/0x140 fs/splice.c:743
do_splice_from fs/splice.c:764 [inline]
do_splice+0xbb8/0x1790 fs/splice.c:1061
__do_sys_splice fs/splice.c:1306 [inline]
__se_sys_splice fs/splice.c:1288 [inline]
__ia32_sys_splice+0x195/0x250 fs/splice.c:1288
do_syscall_32_irqs_on arch/x86/entry/common.c:78 [inline]
__do_fast_syscall_32+0x56/0x80 arch/x86/entry/common.c:137
do_fast_syscall_32+0x2f/0x70 arch/x86/entry/common.c:160
entry_SYSENTER_compat_after_hwframe+0x4d/0x5c

Freed by task 22:
kasan_save_stack+0x1b/0x40 mm/kasan/common.c:48
kasan_set_track+0x1c/0x30 mm/kasan/common.c:56
kasan_set_free_info+0x1b/0x30 mm/kasan/generic.c:355
__kasan_slab_free+0x102/0x140 mm/kasan/common.c:422
slab_free_hook mm/slub.c:1544 [inline]
slab_free_freelist_hook+0x5d/0x150 mm/slub.c:1577
slab_free mm/slub.c:3142 [inline]
kmem_cache_free+0x82/0x350 mm/slub.c:3158
kfree_skbmem+0xef/0x1b0 net/core/skbuff.c:622
__kfree_skb net/core/skbuff.c:679 [inline]
kfree_skb net/core/skbuff.c:696 [inline]
kfree_skb+0x140/0x3f0 net/core/skbuff.c:690
j1939_session_skb_drop_old net/can/j1939/transport.c:333 [inline]
j1939_xtp_rx_cts_one net/can/j1939/transport.c:1394 [inline]
j1939_xtp_rx_cts+0xb59/0xec0 net/can/j1939/transport.c:1433
j1939_tp_cmd_recv net/can/j1939/transport.c:2001 [inline]
j1939_tp_recv+0x8be/0xb40 net/can/j1939/transport.c:2067
j1939_can_recv+0x5bc/0x7d0 net/can/j1939/main.c:101
deliver net/can/af_can.c:571 [inline]
can_rcv_filter+0x5d4/0x8d0 net/can/af_can.c:605
can_receive+0x2e3/0x520 net/can/af_can.c:662
can_rcv+0x12a/0x1a0 net/can/af_can.c:688
__netif_receive_skb_one_core+0x114/0x180 net/core/dev.c:5311
__netif_receive_skb+0x27/0x1c0 net/core/dev.c:5425
process_backlog+0x232/0x6c0 net/core/dev.c:6315
napi_poll net/core/dev.c:6759 [inline]
net_rx_action+0x4dc/0x1100 net/core/dev.c:6829
__do_softirq+0x2a0/0x9f6 kernel/softirq.c:298

The buggy address belongs to the object at ffff8880219f3040
which belongs to the cache skbuff_head_cache of size 224
The buggy address is located 78 bytes inside of
224-byte region [ffff8880219f3040, ffff8880219f3120)
The buggy address belongs to the page:
page:000000007e0b9bbc refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x219f2
head:000000007e0b9bbc order:1 compound_mapcount:0
flags: 0xfff00000010200(slab|head)
raw: 00fff00000010200 dead000000000100 dead000000000122 ffff88804030a500
raw: 0000000000000000 0000000000190019 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
ffff8880219f2f80: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
ffff8880219f3000: fc fc fc fc fc fc fc fc fa fb fb fb fb fb fb fb
>ffff8880219f3080: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^
ffff8880219f3100: fb fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc
ffff8880219f3180: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
==================================================================


---
This report is generated by a bot. It may contain errors.
See https://goo.gl/tpsmEJ for more information about syzbot.
syzbot engineers can be reached at syzk...@googlegroups.com.

syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.

[Hillf Danton]

On Mon, 26 Oct 2020 07:41:16 -0700

Find and get skb against the UAF above and free it after use.

--- a/net/can/j1939/transport.c
+++ b/net/can/j1939/transport.c
@@ -371,6 +371,8 @@ sk_buff *j1939_session_skb_find_by_offse
skb = do_skb;
}
}
+ if (skb)
+ skb_get(skb);
spin_unlock_irqrestore(&session->skb_queue.lock, flags);

if (!skb)
@@ -383,9 +385,8 @@ sk_buff *j1939_session_skb_find_by_offse

static struct sk_buff *j1939_session_skb_find(struct j1939_session *session)
{
- unsigned int offset_start;
+ unsigned int offset_start = session->pkt.dpo * 7;

- offset_start = session->pkt.dpo * 7;
return j1939_session_skb_find_by_offset(session, offset_start);
}

@@ -771,7 +772,7 @@ static int j1939_session_tx_dat(struct j
struct j1939_sk_buff_cb *skcb;
int offset, pkt_done, pkt_end;
unsigned int len, pdelay;
- struct sk_buff *se_skb;
+ struct sk_buff *se_skb, *__skb;
const u8 *tpdat;
int ret = 0;
u8 dat[8];
@@ -779,6 +780,7 @@ static int j1939_session_tx_dat(struct j
se_skb = j1939_session_skb_find_by_offset(session, session->pkt.tx * 7);
if (!se_skb)
return -ENOBUFS;
+ __skb = se_skb;

skcb = j1939_skb_to_cb(se_skb);
tpdat = se_skb->data;
@@ -801,6 +803,7 @@ static int j1939_session_tx_dat(struct j
netdev_err_once(priv->ndev,
"%s: 0x%p: requested data outside of queued buffer: offset %i, len %i, pkt.tx: %i\n",
__func__, session, skcb->offset, se_skb->len , session->pkt.tx);
+ kfree_skb(__skb);
return -EOVERFLOW;
}

@@ -835,6 +838,7 @@ static int j1939_session_tx_dat(struct j
if (pkt_done)
j1939_tp_set_rxtimeout(session, 250);

+ kfree_skb(__skb);
return ret;
}

@@ -1019,6 +1023,7 @@ static int j1939_simple_txnext(struct j1
return -ENOMEM;

can_skb_set_owner(skb, se_skb->sk);
+ kfree_skb(se_skb);

j1939_tp_set_rxtimeout(session, J1939_SIMPLE_ECHO_TIMEOUT_MS);

@@ -1173,6 +1178,7 @@ static void j1939_session_completed(stru
skb = j1939_session_skb_find(session);
/* distribute among j1939 receivers */
j1939_sk_recv(session->priv, skb);
+ kfree_skb(skb);
}

j1939_session_deactivate_activate_next(session);
@@ -1744,7 +1750,7 @@ static void j1939_xtp_rx_dat_one(struct
{
struct j1939_priv *priv = session->priv;
struct j1939_sk_buff_cb *skcb;
- struct sk_buff *se_skb;
+ struct sk_buff *se_skb, *__skb;
const u8 *dat;
u8 *tpdat;
int offset;
@@ -1786,7 +1792,7 @@ static void j1939_xtp_rx_dat_one(struct
goto out_session_cancel;
}

- se_skb = j1939_session_skb_find_by_offset(session, packet * 7);
+ __skb = se_skb = j1939_session_skb_find_by_offset(session, packet * 7);
if (!se_skb) {
netdev_warn(priv->ndev, "%s: 0x%p: no skb found\n", __func__,
session);
@@ -1849,6 +1855,7 @@ static void j1939_xtp_rx_dat_one(struct
}
session->last_cmd = 0xff;
j1939_session_put(session);
+ kfree_skb(__skb);

return;

@@ -1856,6 +1863,7 @@ static void j1939_xtp_rx_dat_one(struct
j1939_session_timers_cancel(session);
j1939_session_cancel(session, J1939_XTP_ABORT_FAULT);
j1939_session_put(session);
+ kfree_skb(__skb);
}

static void j1939_xtp_rx_dat(struct j1939_priv *priv, struct sk_buff *skb)

[syzbot]

Auto-closing this bug as obsolete.
Crashes did not happen for a while, no reproducer and no activity.