Date:   Tue, 21 Apr 2020 11:39:33 -0700
From:   Vinicius Costa Gomes <vinicius.gomes@...el.com>
To:     Eric Dumazet <edumazet@...gle.com>,
        "David S . Miller" <davem@...emloft.net>
Cc:     netdev <netdev@...r.kernel.org>,
        Eric Dumazet <edumazet@...gle.com>,
        Eric Dumazet <eric.dumazet@...il.com>,
        syzbot <syzkaller@...glegroups.com>
Subject: Re: [PATCH net] sched: etf: do not assume all sockets are full blown

Hi Eric,

Eric Dumazet <edumazet@...gle.com> writes:

> skb->sk does not always point to a full blown socket,
> we need to use sk_fullsock() before accessing fields which
> only make sense on full socket.
>
> BUG: KASAN: use-after-free in report_sock_error+0x286/0x300 net/sched/sch_etf.c:141
> Read of size 1 at addr ffff88805eb9b245 by task syz-executor.5/9630
>
> CPU: 1 PID: 9630 Comm: syz-executor.5 Not tainted 5.7.0-rc2-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> Call Trace:
>  <IRQ>
>  __dump_stack lib/dump_stack.c:77 [inline]
>  dump_stack+0x188/0x20d lib/dump_stack.c:118
>  print_address_description.constprop.0.cold+0xd3/0x315 mm/kasan/report.c:382
>  __kasan_report.cold+0x35/0x4d mm/kasan/report.c:511
>  kasan_report+0x33/0x50 mm/kasan/common.c:625
>  report_sock_error+0x286/0x300 net/sched/sch_etf.c:141
>  etf_enqueue_timesortedlist+0x389/0x740 net/sched/sch_etf.c:170
>  __dev_xmit_skb net/core/dev.c:3710 [inline]
>  __dev_queue_xmit+0x154a/0x30a0 net/core/dev.c:4021
>  neigh_hh_output include/net/neighbour.h:499 [inline]
>  neigh_output include/net/neighbour.h:508 [inline]
>  ip6_finish_output2+0xfb5/0x25b0 net/ipv6/ip6_output.c:117
>  __ip6_finish_output+0x442/0xab0 net/ipv6/ip6_output.c:143
>  ip6_finish_output+0x34/0x1f0 net/ipv6/ip6_output.c:153
>  NF_HOOK_COND include/linux/netfilter.h:296 [inline]
>  ip6_output+0x239/0x810 net/ipv6/ip6_output.c:176
>  dst_output include/net/dst.h:435 [inline]
>  NF_HOOK include/linux/netfilter.h:307 [inline]
>  NF_HOOK include/linux/netfilter.h:301 [inline]
>  ip6_xmit+0xe1a/0x2090 net/ipv6/ip6_output.c:280
>  tcp_v6_send_synack+0x4e7/0x960 net/ipv6/tcp_ipv6.c:521
>  tcp_rtx_synack+0x10d/0x1a0 net/ipv4/tcp_output.c:3916
>  inet_rtx_syn_ack net/ipv4/inet_connection_sock.c:669 [inline]
>  reqsk_timer_handler+0x4c2/0xb40 net/ipv4/inet_connection_sock.c:763
>  call_timer_fn+0x1ac/0x780 kernel/time/timer.c:1405
>  expire_timers kernel/time/timer.c:1450 [inline]
>  __run_timers kernel/time/timer.c:1774 [inline]
>  __run_timers kernel/time/timer.c:1741 [inline]
>  run_timer_softirq+0x623/0x1600 kernel/time/timer.c:1787
>  __do_softirq+0x26c/0x9f7 kernel/softirq.c:292
>  invoke_softirq kernel/softirq.c:373 [inline]
>  irq_exit+0x192/0x1d0 kernel/softirq.c:413
>  exiting_irq arch/x86/include/asm/apic.h:546 [inline]
>  smp_apic_timer_interrupt+0x19e/0x600 arch/x86/kernel/apic/apic.c:1140
>  apic_timer_interrupt+0xf/0x20 arch/x86/entry/entry_64.S:829
>  </IRQ>
> RIP: 0010:des_encrypt+0x157/0x9c0 lib/crypto/des.c:792
> Code: 85 22 06 00 00 41 31 dc 41 8b 4d 04 44 89 e2 41 83 e4 3f 4a 8d 3c a5 60 72 72 88 81 e2 3f 3f 3f 3f 48 89 f8 48 c1 e8 03 31 d9 <0f> b6 34 28 48 89 f8 c1 c9 04 83 e0 07 83 c0 03 40 38 f0 7c 09 40
> RSP: 0018:ffffc90003b5f6c0 EFLAGS: 00000282 ORIG_RAX: ffffffffffffff13
> RAX: 1ffffffff10e4e55 RBX: 00000000d2f846d0 RCX: 00000000d2f846d0
> RDX: 0000000012380612 RSI: ffffffff839863ca RDI: ffffffff887272a8
> RBP: dffffc0000000000 R08: ffff888091d0a380 R09: 0000000000800081
> R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000012
> R13: ffff8880a8ae8078 R14: 00000000c545c93e R15: 0000000000000006
>  cipher_crypt_one crypto/cipher.c:75 [inline]
>  crypto_cipher_encrypt_one+0x124/0x210 crypto/cipher.c:82
>  crypto_cbcmac_digest_update+0x1b5/0x250 crypto/ccm.c:830
>  crypto_shash_update+0xc4/0x120 crypto/shash.c:119
>  shash_ahash_update+0xa3/0x110 crypto/shash.c:246
>  crypto_ahash_update include/crypto/hash.h:547 [inline]
>  hash_sendmsg+0x518/0xad0 crypto/algif_hash.c:102
>  sock_sendmsg_nosec net/socket.c:652 [inline]
>  sock_sendmsg+0xcf/0x120 net/socket.c:672
>  ____sys_sendmsg+0x308/0x7e0 net/socket.c:2362
>  ___sys_sendmsg+0x100/0x170 net/socket.c:2416
>  __sys_sendmmsg+0x195/0x480 net/socket.c:2506
>  __do_sys_sendmmsg net/socket.c:2535 [inline]
>  __se_sys_sendmmsg net/socket.c:2532 [inline]
>  __x64_sys_sendmmsg+0x99/0x100 net/socket.c:2532
>  do_syscall_64+0xf6/0x7d0 arch/x86/entry/common.c:295
>  entry_SYSCALL_64_after_hwframe+0x49/0xb3
> RIP: 0033:0x45c829
> Code: 0d b7 fb ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 db b6 fb ff c3 66 2e 0f 1f 84 00 00 00 00
> RSP: 002b:00007f6d9528ec78 EFLAGS: 00000246 ORIG_RAX: 0000000000000133
> RAX: ffffffffffffffda RBX: 00000000004fc080 RCX: 000000000045c829
> RDX: 0000000000000001 RSI: 0000000020002640 RDI: 0000000000000004
> RBP: 000000000078bf00 R08: 0000000000000000 R09: 0000000000000000
> R10: 0000000000000000 R11: 0000000000000246 R12: 00000000ffffffff
> R13: 00000000000008d7 R14: 00000000004cb7aa R15: 00007f6d9528f6d4
>
> Fixes: 4b15c7075352 ("net/sched: Make etf report drops on error_queue")
> Fixes: 25db26a91364 ("net/sched: Introduce the ETF Qdisc")
> Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> Reported-by: syzbot <syzkaller@...glegroups.com>
> Cc: Vinicius Costa Gomes <vinicius.gomes@...el.com>
> ---

Reviewed-by: Vinicius Costa Gomes <vinicius.gomes@...el.com>

I am curious about the repro program, do you have it? There might be
other bugs like this in hiding.


Cheers,
-- 
Vinicius

Powered by blists - more mailing lists