Message-ID: <cc82200a-6374-0a87-a3e0-471d04a8306a@gmail.com>
Date:   Tue, 2 Apr 2019 13:03:30 -0700
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     syzbot <syzbot+457d3e2ffbcf31aee5c0@...kaller.appspotmail.com>,
        ast@...nel.org, daniel@...earbox.net, linux-kernel@...r.kernel.org,
        netdev@...r.kernel.org, syzkaller-bugs@...glegroups.com
Subject: Re: KASAN: use-after-free Read in __dev_map_entry_free



On 06/20/2018 08:19 AM, syzbot wrote:
> Hello,
> 
> syzbot found the following crash on:
> 
> HEAD commit:    f0dc7f9c6dd9 Merge git://git.kernel.org/pub/scm/linux/kern..
> git tree:       bpf-next
> console output: https://syzkaller.appspot.com/x/log.txt?x=15ad7d10400000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=fa9c20c48788d1c1
> dashboard link: https://syzkaller.appspot.com/bug?extid=457d3e2ffbcf31aee5c0
> compiler:       gcc (GCC) 8.0.1 20180413 (experimental)
> syzkaller repro:https://syzkaller.appspot.com/x/repro.syz?x=1195225f800000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=171a7ce4400000
> 
> IMPORTANT: if you fix the bug, please add the following tag to the commit:
> Reported-by: syzbot+457d3e2ffbcf31aee5c0@...kaller.appspotmail.com
> 
> ==================================================================
> BUG: KASAN: use-after-free in dev_map_flush_old kernel/bpf/devmap.c:365 [inline]
> BUG: KASAN: use-after-free in __dev_map_entry_free+0x2a8/0x300 kernel/bpf/devmap.c:379
> Read of size 8 at addr ffff8801b8da38c8 by task ksoftirqd/1/18
> 
> CPU: 1 PID: 18 Comm: ksoftirqd/1 Not tainted 4.17.0+ #39
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> Call Trace:
>  __dump_stack lib/dump_stack.c:77 [inline]
>  dump_stack+0x1b9/0x294 lib/dump_stack.c:113
>  print_address_description+0x6c/0x20b mm/kasan/report.c:256
>  kasan_report_error mm/kasan/report.c:354 [inline]
>  kasan_report.cold.7+0x242/0x2fe mm/kasan/report.c:412
>  __asan_report_load8_noabort+0x14/0x20 mm/kasan/report.c:433
>  dev_map_flush_old kernel/bpf/devmap.c:365 [inline]
>  __dev_map_entry_free+0x2a8/0x300 kernel/bpf/devmap.c:379
>  __rcu_reclaim kernel/rcu/rcu.h:178 [inline]
>  rcu_do_batch kernel/rcu/tree.c:2558 [inline]
>  invoke_rcu_callbacks kernel/rcu/tree.c:2818 [inline]
>  __rcu_process_callbacks kernel/rcu/tree.c:2785 [inline]
>  rcu_process_callbacks+0xe9d/0x1760 kernel/rcu/tree.c:2802
>  __do_softirq+0x2e0/0xaf5 kernel/softirq.c:284
>  run_ksoftirqd+0x86/0x100 kernel/softirq.c:645
>  smpboot_thread_fn+0x417/0x870 kernel/smpboot.c:164
>  kthread+0x345/0x410 kernel/kthread.c:240
>  ret_from_fork+0x3a/0x50 arch/x86/entry/entry_64.S:412
> 
> Allocated by task 6675:
>  save_stack+0x43/0xd0 mm/kasan/kasan.c:448
>  set_track mm/kasan/kasan.c:460 [inline]
>  kasan_kmalloc+0xc4/0xe0 mm/kasan/kasan.c:553
>  kmem_cache_alloc_trace+0x152/0x780 mm/slab.c:3620
>  kmalloc include/linux/slab.h:513 [inline]
>  kzalloc include/linux/slab.h:706 [inline]
>  dev_map_alloc+0x208/0x7f0 kernel/bpf/devmap.c:102
>  find_and_alloc_map kernel/bpf/syscall.c:129 [inline]
>  map_create+0x393/0x1010 kernel/bpf/syscall.c:453
>  __do_sys_bpf kernel/bpf/syscall.c:2351 [inline]
>  __se_sys_bpf kernel/bpf/syscall.c:2328 [inline]
>  __x64_sys_bpf+0x303/0x510 kernel/bpf/syscall.c:2328
>  do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:290
>  entry_SYSCALL_64_after_hwframe+0x49/0xbe
> 
> Freed by task 26:
>  save_stack+0x43/0xd0 mm/kasan/kasan.c:448
>  set_track mm/kasan/kasan.c:460 [inline]
>  __kasan_slab_free+0x11a/0x170 mm/kasan/kasan.c:521
>  kasan_slab_free+0xe/0x10 mm/kasan/kasan.c:528
>  __cache_free mm/slab.c:3498 [inline]
>  kfree+0xd9/0x260 mm/slab.c:3813
>  dev_map_free+0x4fa/0x670 kernel/bpf/devmap.c:191
>  bpf_map_free_deferred+0xba/0xf0 kernel/bpf/syscall.c:262
>  process_one_work+0xc64/0x1b70 kernel/workqueue.c:2153
>  worker_thread+0x181/0x13a0 kernel/workqueue.c:2296
>  kthread+0x345/0x410 kernel/kthread.c:240
>  ret_from_fork+0x3a/0x50 arch/x86/entry/entry_64.S:412
> 
> The buggy address belongs to the object at ffff8801b8da37c0
>  which belongs to the cache kmalloc-512 of size 512
> The buggy address is located 264 bytes inside of
>  512-byte region [ffff8801b8da37c0, ffff8801b8da39c0)
> The buggy address belongs to the page:
> page:ffffea0006e368c0 count:1 mapcount:0 mapping:ffff8801da800940 index:0xffff8801b8da3540
> flags: 0x2fffc0000000100(slab)
> raw: 02fffc0000000100 ffffea0007217b88 ffffea0006e30cc8 ffff8801da800940
> raw: ffff8801b8da3540 ffff8801b8da3040 0000000100000004 0000000000000000
> page dumped because: kasan: bad access detected
> 
> Memory state around the buggy address:
>  ffff8801b8da3780: fc fc fc fc fc fc fc fc fb fb fb fb fb fb fb fb
>  ffff8801b8da3800: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>> ffff8801b8da3880: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>                                               ^
>  ffff8801b8da3900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>  ffff8801b8da3980: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
> ==================================================================
> 
> 
> ---
> This bug 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 syzkaller@...glegroups.com.
> 
> syzbot will keep track of this bug report. See:
> https://goo.gl/tpsmEJ#bug-status-tracking for how to communicate with syzbot.
> syzbot can test patches for this bug, for details see:
> https://goo.gl/tpsmEJ#testing-patches


What about something like :



diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c
index 191b79948424f4b21b7aa120abc03801264bf0a6..1e525d70f83354e451b738ffb8e42d83b5fa932f 100644
--- a/kernel/bpf/devmap.c
+++ b/kernel/bpf/devmap.c
@@ -164,6 +164,9 @@ static void dev_map_free(struct bpf_map *map)
        bpf_clear_redirect_map(map);
        synchronize_rcu();
 
+       /* Make sure prior __dev_map_entry_free() have completed. */
+       rcu_barrier();
+
        /* To ensure all pending flush operations have completed wait for flush
         * bitmap to indicate all flush_needed bits to be zero on _all_ cpus.
         * Because the above synchronize_rcu() ensures the map is disconnected

Powered by blists - more mailing lists