Re: [PATCH] proc/sysctl: Fix NULL pointer dereference in put_links

From: YueHaibing
Date: Thu Mar 14 2019 - 01:48:15 EST



On 2019/3/14 6:35, Luis Chamberlain wrote:
> On Tue, Mar 12, 2019 at 10:10:53AM +0800, YueHaibing wrote:
>> On 2019/3/9 20:27, YueHaibing wrote:
>>> On 2019/3/9 11:58, Luis Chamberlain wrote:
>>>> I so excited about the report, thanks this is awesome work!! However it seemed possibly like a paper-over a different issue. I'm not convinced yet.
>>>>
>>>> If prefer a bit more investigation until we can be 100% sure. At
>>>> this point my position would be this is more of a risk for more
>>>> regressions and that's an off feeling for a fix.
>>>>
>>>> Luis
>>>
>>> This issue occurs like this:
>>>
>>> __register_sysctl_table
>>> -->get_subdir
>>> -->new_dir //create 'new' dir
>>> -->init_header // &new->header->parent = NULL
>>> -->insert_header
>>> --> insert_links
>>> -->new_links
>>> -->kzalloc //failed
>>> -->&new->header->parent = NULL
>>> -->drop_sysctl_table(&new->header) //In err path of get_subdir
>
> Nope, we drop that earlier actually, note that upon failure on
> insert_links() we also call drop_sysctl_table(). And even in
> the successful case on __register_sysctl_table() we still always
> call drop_sysctl_table().
>
> So it seems that the real issue is we have multiple error paths which
> call drop_sysctl_table(). That is the justification for which we do
> need your patch. Can you adjust your commit log to account for this?

Sure, will resend it with this.

>
> Luis
>
>>> -->put_links //call xlate_dir whose 2nd param is '&new->header->parent'
>>> --> xlate_dir //access dir->header,however 'dir' is NULL which comes from '&new->header->parent'
>>>
>>> Please correct me if this is incorrect.
>>
>> Hello,
>>
>> Any comment will be appreciated.
>>
>>>
>>>>
>>>> On Fri, Mar 8, 2019, 8:04 PM YueHaibing <yuehaibing@xxxxxxxxxx <mailto:yuehaibing@xxxxxxxxxx>> wrote:
>>>>
>>>> +cc Al Viro
>>>>
>>>> On 2019/3/4 21:54, Yue Haibing wrote:
>>>> > From: YueHaibing <yuehaibing@xxxxxxxxxx <mailto:yuehaibing@xxxxxxxxxx>>
>>>> >
>>>> > Syzkaller report this:
>>>> >
>>>> > kasan: GPF could be caused by NULL-ptr deref or user memory access
>>>> > general protection fault: 0000 [#1] SMP KASAN PTI
>>>> > CPU: 1 PID: 5373 Comm: syz-executor.0 Not tainted 5.0.0-rc8+ #3
>>>> > Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1ubuntu1 04/01/2014
>>>> > RIP: 0010:put_links+0x101/0x440 fs/proc/proc_sysctl.c:1599
>>>> > Code: 00 0f 85 3a 03 00 00 48 8b 43 38 48 89 44 24 20 48 83 c0 38 48 89 c2 48 89 44 24 28 48 b8 00 00 00 00 00 fc ff df 48 c1 ea 03 <80> 3c 02 00 0f 85 fe 02 00 00 48 8b 74 24 20 48 c7 c7 60 2a 9d 91
>>>> > RSP: 0018:ffff8881d828f238 EFLAGS: 00010202
>>>> > RAX: dffffc0000000000 RBX: ffff8881e01b1140 RCX: ffffffff8ee98267
>>>> > RDX: 0000000000000007 RSI: ffffc90001479000 RDI: ffff8881e01b1178
>>>> > RBP: dffffc0000000000 R08: ffffed103ee27259 R09: ffffed103ee27259
>>>> > R10: 0000000000000001 R11: ffffed103ee27258 R12: fffffffffffffff4
>>>> > R13: 0000000000000006 R14: ffff8881f59838c0 R15: dffffc0000000000
>>>> > FS: 00007f072254f700(0000) GS:ffff8881f7100000(0000) knlGS:0000000000000000
>>>> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>> > CR2: 00007fff8b286668 CR3: 00000001f0542002 CR4: 00000000007606e0
>>>> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>>>> > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>>>> > PKRU: 55555554
>>>> > Call Trace:
>>>> > drop_sysctl_table+0x152/0x9f0 fs/proc/proc_sysctl.c:1629
>>>> > get_subdir fs/proc/proc_sysctl.c:1022 [inline]
>>>> > __register_sysctl_table+0xd65/0x1090 fs/proc/proc_sysctl.c:1335
>>>> > ? 0xffffffffc1a18000
>>>> > br_netfilter_init+0xbc/0x1000 [br_netfilter]
>>>> > ? 0xffffffffc1a18000
>>>> > ? 0xffffffffc1a18000
>>>> > do_one_initcall+0xfa/0x5ca init/main.c:887
>>>> > do_init_module+0x204/0x5f6 kernel/module.c:3460
>>>> > load_module+0x66b2/0x8570 kernel/module.c:3808
>>>> > __do_sys_finit_module+0x238/0x2a0 kernel/module.c:3902
>>>> > do_syscall_64+0x147/0x600 arch/x86/entry/common.c:290
>>>> > entry_SYSCALL_64_after_hwframe+0x49/0xbe
>>>> > RIP: 0033:0x462e99
>>>> > Code: f7 d8 64 89 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 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 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
>>>> > RSP: 002b:00007f072254ec58 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
>>>> > RAX: ffffffffffffffda RBX: 000000000073bf00 RCX: 0000000000462e99
>>>> > RDX: 0000000000000000 RSI: 0000000020000280 RDI: 0000000000000003
>>>> > RBP: 00007f072254ec70 R08: 0000000000000000 R09: 0000000000000000
>>>> > R10: 0000000000000000 R11: 0000000000000246 R12: 00007f072254f6bc
>>>> > R13: 00000000004bcefa R14: 00000000006f6fb0 R15: 0000000000000004
>>>> > Modules linked in: br_netfilter(+) dvb_usb_dibusb_mc_common dib3000mc dibx000_common dvb_usb_dibusb_common dvb_usb_dw2102 dvb_usb classmate_laptop palmas_regulator cn videobuf2_v4l2 v4l2_common snd_soc_bd28623 mptbase snd_usb_usx2y snd_usbmidi_lib snd_rawmidi wmi libnvdimm lockd sunrpc grace rc_kworld_pc150u rc_core rtc_da9063 sha1_ssse3 i2c_cros_ec_tunnel adxl34x_spi adxl34x nfnetlink lib80211 i5500_temp dvb_as102 dvb_core videobuf2_common videodev media videobuf2_vmalloc videobuf2_memops udc_core lnbp22 leds_lp3952 hid_roccat_ryos s1d13xxxfb mtd vport_geneve openvswitch nf_conncount nf_nat_ipv6 nsh geneve udp_tunnel ip6_udp_tunnel snd_soc_mt6351 sis_agp phylink snd_soc_adau1761_spi snd_soc_adau1761 snd_soc_adau17x1 snd_soc_core snd_pcm_dmaengine ac97_bus snd_compress snd_soc_adau_utils snd_soc_sigmadsp_regmap snd_soc_sigmadsp raid_class hid_roccat_konepure hid_roccat_common hid_roccat c2port_duramar2150 core mdio_bcm_unimac iptable_security iptable_raw iptable_mangle
>>>> > iptable_nat nf_nat_ipv4 nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 iptable_filter bpfilter ip6_vti ip_vti ip_gre ipip sit tunnel4 ip_tunnel hsr veth netdevsim devlink vxcan batman_adv cfg80211 rfkill chnl_net caif nlmon dummy team bonding vcan bridge stp llc ip6_gre gre ip6_tunnel tunnel6 tun crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel joydev mousedev ide_pci_generic piix aesni_intel aes_x86_64 ide_core crypto_simd atkbd cryptd glue_helper serio_raw ata_generic pata_acpi i2c_piix4 floppy sch_fq_codel ip_tables x_tables ipv6 [last unloaded: lm73]
>>>> > Dumping ftrace buffer:
>>>> > (ftrace buffer empty)
>>>> > ---[ end trace 770020de38961fd0 ]---
>>>> >
>>>> > A new dir entry can be created in get_subdir and its header->parent is set
>>>> > to NULL. Only after insert_header success, it will be set to 'dir'.
>>>> > However in err handling path of get_subdir, drop_sysctl_table maybe called
>>>> > on 'new->header' regardless value of header->parent. Then put_links
>>>> > be called, which triggers NULL-ptr deref when access the member of
>>>> > header->parent in xlate_dir.
>>>> >
>>>> > Reported-by: Hulk Robot <hulkci@xxxxxxxxxx <mailto:hulkci@xxxxxxxxxx>>
>>>> > Fixes: 0e47c99d7fe25 ("sysctl: Replace root_list with links between sysctl_table_sets")
>>>> > Signed-off-by: YueHaibing <yuehaibing@xxxxxxxxxx <mailto:yuehaibing@xxxxxxxxxx>>
>>>> > ---
>>>> > fs/proc/proc_sysctl.c | 3 ++-
>>>> > 1 file changed, 2 insertions(+), 1 deletion(-)
>>>> >
>>>> > diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
>>>> > index 4d598a3..4c2a05a 100644
>>>> > --- a/fs/proc/proc_sysctl.c
>>>> > +++ b/fs/proc/proc_sysctl.c
>>>> > @@ -1626,7 +1626,8 @@ static void drop_sysctl_table(struct ctl_table_header *header)
>>>> > if (--header->nreg)
>>>> > return;
>>>> >
>>>> > - put_links(header);
>>>> > + if (parent)
>>>> > + put_links(header);
>>>> > start_unregistering(header);
>>>> > if (!--header->count)
>>>> > kfree_rcu(header, rcu);
>>>> >
>>>>
>
> .
>