Re: [Syzkaller & bisect] There is KASAN: slab-use-after-free Read in __nf_unregister_net_hook in v6.12-rc1
From: Eric Dumazet
Date: Thu Oct 10 2024 - 05:20:10 EST
On Thu, Oct 10, 2024 at 10:58 AM Eric Dumazet <edumazet@xxxxxxxxxx> wrote:
>
> On Thu, Oct 10, 2024 at 10:19 AM Lai, Yi <yi1.lai@xxxxxxxxxxxxxxx> wrote:
> >
> > Hi Eric,
> >
> > Greetings!
> >
> > I used Syzkaller and found that there is KASAN: slab-use-after-free Read in __nf_unregister_net_hook in v6.12-rc1
> >
> > After bisection and the first bad commit is:
> > "
> > 78c3253f27e5 net: use synchronize_rcu_expedited in cleanup_net()
> > "
> >
> > All detailed into can be found at:
> > https://github.com/laifryiee/syzkaller_logs/tree/main/241001_170248___nf_unregister_net_hook
> > Syzkaller repro code:
> > https://github.com/laifryiee/syzkaller_logs/tree/main/241001_170248___nf_unregister_net_hook/repro.c
> > Syzkaller repro syscall steps:
> > https://github.com/laifryiee/syzkaller_logs/tree/main/241001_170248___nf_unregister_net_hook/repro.prog
> > Syzkaller report:
> > https://github.com/laifryiee/syzkaller_logs/tree/main/241001_170248___nf_unregister_net_hook/repro.report
> > Kconfig(make olddefconfig):
> > https://github.com/laifryiee/syzkaller_logs/tree/main/241001_170248___nf_unregister_net_hook/kconfig_origin
> > Bisect info:
> > https://github.com/laifryiee/syzkaller_logs/tree/main/241001_170248___nf_unregister_net_hook/bisect_info.log
> > bzImage:
> > https://github.com/laifryiee/syzkaller_logs/raw/refs/heads/main/241001_170248___nf_unregister_net_hook/bzImage_9852d85ec9d492ebef56dc5f229416c925758edc
> > Issue dmesg:
> > https://github.com/laifryiee/syzkaller_logs/blob/main/241001_170248___nf_unregister_net_hook/9852d85ec9d492ebef56dc5f229416c925758edc_dmesg.log
> >
> > "
> > [ 16.910304] ==================================================================
> > [ 16.910757] BUG: KASAN: slab-use-after-free in __nf_unregister_net_hook+0x640/0x6b0
> > [ 16.911156] Read of size 8 at addr ffff8880106fe400 by task repro/725
> > [ 16.911529]
> > [ 16.911674] CPU: 1 UID: 0 PID: 725 Comm: repro Not tainted 6.12.0-rc1-9852d85ec9d4+ #1
> > [ 16.912338] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
> > [ 16.913460] Call Trace:
> > [ 16.913859] <TASK>
> > [ 16.913983] dump_stack_lvl+0xea/0x150
> > [ 16.914262] print_report+0xce/0x610
> > [ 16.914585] ? __nf_unregister_net_hook+0x640/0x6b0
> > [ 16.914845] ? kasan_complete_mode_report_info+0x80/0x200
> > [ 16.915133] ? __nf_unregister_net_hook+0x640/0x6b0
> > [ 16.915497] kasan_report+0xcc/0x110
> > [ 16.915687] ? __nf_unregister_net_hook+0x640/0x6b0
> > [ 16.916082] ? __pfx_bpf_link_release+0x10/0x10
> > [ 16.916398] __asan_report_load8_noabort+0x18/0x20
> > [ 16.916655] __nf_unregister_net_hook+0x640/0x6b0
> > [ 16.916895] ? __pfx_bpf_link_release+0x10/0x10
> > [ 16.917177] ? __pfx_bpf_link_release+0x10/0x10
> > [ 16.917467] c+0xea/0x140
> > [ 16.917770] bpf_nf_link_release+0xda/0x1e0
> > [ 16.917983] bpf_link_free+0x139/0x2d0
> > [ 16.918187] bpf_link_release+0x68/0x80
> > [ 16.918397] __fput+0x414/0xb60
> > [ 16.918603] ____fput+0x22/0x30
Florian, Pablo :
It seems that bpf was able to defer the __nf_unregister_net_hook()
after exit()/close() time.
Perhaps a netns reference is missing, because the netns has been
dismantled/freed already.
bpf_nf_link_attach() does :
link->net = net;
But I do not see a reference being taken on net.
Thanks
> > [ 16.918777] task_work_run+0x19c/0x2b0
> > [ 16.919006] ? __pfx_task_work_run+0x10/0x10
> > [ 16.919235] ? free_nsproxy+0x344/0x470
> > [ 16.919448] ? switch_task_namespaces+0xf9/0x110
> > [ 16.919711] do_exit+0xb19/0x2a30
> > [ 16.919913] ? audit_log_end+0x156/0x2c0
> > [ 16.920202] ? __pfx_do_exit+0x10/0x10
> > [ 16.920419] ? audit_seccomp+0x1b2/0x220
> > [ 16.920656] ? audit_seccomp+0x1b9/0x220
> > [ 16.920872] __secure_computing+0x2f5/0x350
> > [ 16.921109] syscall_trace_enter+0x9b/0x230
> > [ 16.921354] do_syscall_64+0x115/0x140
> > [ 16.921551] entry_SYSCALL_64_after_hwframe+0x76/0x7e
> > [ 16.921816] RIP: 0033:0x7f2464518a4d
> > [ 16.922038] Code: Unable to access opcode bytes at 0x7f2464518a23.
> > [ 16.922363] RSP: 002b:00007ffe02122928 EFLAGS: 00000246 ORIG_RAX: 00000000000000e7
> > [ 16.922759] RAX: ffffffffffffffda RBX: 00007f24645f69e0 RCX: 00007f2464518a4d
> > [ 16.923163] RDX: 00000000000000e7 RSI: ffffffffffffff80 RDI: 0000000000000000
> > [ 16.923583] RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000020
> > [ 16.923965] R10: 00007ffe021227d0 R11: 0000000000000246 R12: 00007f24645f69e0
> > [ 16.924353] R13: 00007f24645fbf00 R14: 0000000000000001 R15: 00007f24645fbee8
> > [ 16.924750] </TASK>
> > [ 16.924870]
> > [ 16.924962] Allocated by task 725:
> > [ 16.925153] kasan_save_stack+0x2c/0x60
> > [ 16.925387] kasan_save_track+0x18/0x40
> > [ 16.925607] kasan_save_alloc_info+0x3c/0x50
> > [ 16.925853] __kasan_slab_alloc+0x62/0x80
> > [ 16.926065] kmem_cache_alloc_noprof+0x114/0x370
> > [ 16.926334] copy_net_ns+0xf0/0x6e0
> > [ 16.926520] create_new_namespaces+0x403/0xb40
> > [ 16.926769] unshare_nsproxy_namespaces+0xca/0x200
> > [ 16.927038] ksys_unshare+0x424/0xa10
> > [ 16.927242] __x64_sys_unshare+0x3a/0x50
> > [ 16.927451] x64_sys_call+0xcf1/0x20d0
> > [ 16.927670] do_syscall_64+0x6d/0x140
> > [ 16.927888] entry_SYSCALL_64_after_hwframe+0x76/0x7e
> > [ 16.928148]
> > [ 16.928243] Freed by task 51:
> > [ 16.928414] kasan_save_stack+0x2c/0x60
> > [ 16.928631] kasan_save_track+0x18/0x40
> > [ 16.928841] kasan_save_free_info+0x3f/0x60
> > [ 16.929073] __kasan_slab_free+0x3d/0x60
> > [ 16.929308] kmem_cache_free+0x1aa/0x550
> > [ 16.929529] cleanup_net+0x8af/0xae0
> > [ 16.929745] process_one_work+0x92e/0x1b50
> > [ 16.929981] worker_thread+0x68d/0xe90
> > [ 16.930204] kthread+0x35a/0x470
> > [ 16.930424] ret_from_fork+0x56/0x90
> > [ 16.930626] ret_from_fork_asm+0x1a/0x30
> > [ 16.930843]
> > [ 16.930935] The buggy address belongs to the object at ffff8880106fd400
> > [ 16.930935] which belongs to the cache net_namespace of size 6912
> > [ 16.931597] The buggy address is located 4096 bytes inside of
> > [ 16.931597] freed 6912-byte region [ffff8880106fd400, ffff8880106fef00)
> > [ 16.932280]
> > [ 16.932430] The buggy address belongs to the physical page:
> > [ 16.932757] page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x106f8
> > [ 16.933195] head: order:3 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0
> > [ 16.933592] memcg:ffff88801157d281
> > [ 16.933783] flags: 0xfffffc0000040(head|node=0|zone=1|lastcpupid=0x1fffff)
> > [ 16.934155] page_type: f5(slab)
> > [ 16.934329] raw: 000fffffc0000040 ffff88800d71cdc0 dead000000000122 0000000000000000
> > [ 16.934782] raw: 0000000000000000 0000000080040004 00000001f5000000 ffff88801157d281
> > [ 16.935209] head: 000fffffc0000040 ffff88800d71cdc0 dead000000000122 0000000000000000
> > [ 16.935610] head: 0000000000000000 0000000080040004 00000001f5000000 ffff88801157d281
> > [ 16.936038] head: 000fffffc0000003 ffffea000041be01 ffffffffffffffff 0000000000000000
> > [ 16.936507] head: 0000000000000008 0000000000000000 00000000ffffffff 0000000000000000
> > [ 16.936913] page dumped because: kasan: bad access detected
> > [ 16.937285]
> > [ 16.937398] Memory state around the buggy address:
> > [ 16.937698] ffff8880106fe300: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
> > [ 16.938149] ffff8880106fe380: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
> > [ 16.938634] >ffff8880106fe400: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
> > [ 16.939010] ^
> > "
> >
> > I hope you find it useful.
> >
> > Regards,
> > Yi Lai
> >
> > ---
> >
> > If you don't need the following environment to reproduce the problem or if you
> > already have one reproduced environment, please ignore the following information.
> >
> > How to reproduce:
> > git clone https://gitlab.com/xupengfe/repro_vm_env.git
> > cd repro_vm_env
> > tar -xvf repro_vm_env.tar.gz
> > cd repro_vm_env; ./start3.sh // it needs qemu-system-x86_64 and I used v7.1.0
> > // start3.sh will load bzImage_2241ab53cbb5cdb08a6b2d4688feb13971058f65 v6.2-rc5 kernel
> > // You could change the bzImage_xxx as you want
> > // Maybe you need to remove line "-drive if=pflash,format=raw,readonly=on,file=./OVMF_CODE.fd \" for different qemu version
> > You could use below command to log in, there is no password for root.
> > ssh -p 10023 root@localhost
> >
> > After login vm(virtual machine) successfully, you could transfer reproduced
> > binary to the vm by below way, and reproduce the problem in vm:
> > gcc -pthread -o repro repro.c
> > scp -P 10023 repro root@localhost:/root/
> >
> > Get the bzImage for target kernel:
> > Please use target kconfig and copy it to kernel_src/.config
> > make olddefconfig
> > make -jx bzImage //x should equal or less than cpu num your pc has
> >
> > Fill the bzImage file into above start3.sh to load the target kernel in vm.
> >
> >
> > Tips:
> > If you already have qemu-system-x86_64, please ignore below info.
> > If you want to install qemu v7.1.0 version:
> > git clone https://github.com/qemu/qemu.git
> > cd qemu
> > git checkout -f v7.1.0
> > mkdir build
> > cd build
> > yum install -y ninja-build.x86_64
> > yum -y install libslirp-devel.x86_64
> > ../configure --target-list=x86_64-softmmu --enable-kvm --enable-vnc --enable-gtk --enable-sdl --enable-usb-redir --enable-slirp
> > make
> > make install
>
>
> Probably reported in
> https://lore.kernel.org/all/000000000000635bfa0607ed5cdc@xxxxxxxxxx/T/
>
> You probably should CC netfilter maintainers on this one.