[RFC PATCH 0/2] arm64: optional paranoid __{get,put}_user checks

[Kees Cook]

On Wed, Nov 1, 2017 at 4:05 PM, Laura Abbott <labbott at redhat.com> wrote:
> On 11/01/2017 03:28 PM, Kees Cook wrote:
>> On Wed, Nov 1, 2017 at 2:13 PM, Laura Abbott <labbott at redhat.com> wrote:
>>> On 11/01/2017 05:05 AM, Mark Rutland wrote:
>>>> On Tue, Oct 31, 2017 at 04:56:39PM -0700, Laura Abbott wrote:
>>>>> On 10/26/2017 02:09 AM, Mark Rutland wrote:
>>>>>> In Prague, Kees mentioned that it would be nice to have a mechanism to
>>>>>> catch bad __{get,put}_user uses, such as the recent CVE-2017-5123 [1,2]
>>>>>> issue with unsafe_put_user() in waitid().
>>>>>>
>>>>>> These patches allow an optional access_ok() check to be dropped in
>>>>>> arm64's __{get,put}_user() primitives. These will then BUG() if a bad
>>>>>> user pointer is passed (which should only happen in the absence of an
>>>>>> earlier access_ok() check).
>>>>
>>>>> Turning on the option fails as soon as we hit userspace. On my buildroot
>>>>> based environment I get the help text for ld.so (????) and then a message
>>>>> about attempting to kill init.
>>>>
>>>> Ouch. Thanks for the report, and sorry about this.
>>>>
>>>> The problem is that I evaluate the ptr argument twice in
>>>> __{get,put}_user(), and this may have side effects.
>>>>
>>>> e.g. when the ELF loader does things like:
>>>>
>>>>   __put_user((elf_addr_t)p, sp++)
>>>>
>>>> ... we increment sp twice, and write to the wrong user address, leaving
>>>> sp corrupt.
>>>>
>>>> I have an additional patch [1] to fix this, which is in my
>>>> arm64/access-ok branch [2].
>>>>
>>>> Thanks,
>>>> Mark.
>>>>
>>>> [1] https://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git/commit/?h=arm64/access-ok&id=ebb7ff83eb53b8810395d5cf48712a4ae6d678543
>>>> [2] https://git.kernel.org/pub/scm/linux/kernel/git/mark/linux.git/log/?h=arm64/access-ok
>>>>
>>>
>>> Thanks, the updated patch works. I wrote an LKDTM test to verify
>>> the expected behavior (__{get,put}_user panic whereas {get,put}_user
>>> do not). You're welcome to add Tested-by or I can wait for v2.
>>
>> Nice. :) Out of curiosity, can you check if this correctly BUG()s on a
>> waitid() call when the fixes are reverted?
>>
>> 96ca579a1ecc ("waitid(): Avoid unbalanced user_access_end() on
>> access_ok() error")
>> 1c9fec470b81 ("waitid(): Add missing access_ok() checks")
>>
>> -Kees
>>
>
> Yep, we get a nice bug:
>
> [   34.783912] ------------[ cut here ]------------
> [   34.784484] kernel BUG at kernel/exit.c:1614!

Awesome! :)

I wonder how hard it might be to make this happen on x86 too (or
generically). Hmmm

-Kees

> [   34.785016] Internal error: Oops - BUG: 0 [#1] PREEMPT SMP
> [   34.785572] Modules linked in:
> [   34.786177] CPU: 0 PID: 1324 Comm: a.out Not tainted 4.14.0-rc5-00005-ga3bb7b0f72d3 #69
> [   34.786657] Hardware name: linux,dummy-virt (DT)
> [   34.787093] task: ffff80003c4ed400 task.stack: ffff00000ade0000
> [   34.788196] PC is at SyS_waitid+0x1d4/0x210
> [   34.788534] LR is at SyS_waitid+0x20/0x210
> [   34.788839] pc : [<ffff0000080cde1c>] lr : [<ffff0000080cdc68>] pstate: a0000145
> [   34.789310] sp : ffff00000ade3e00
> [   34.789578] x29: ffff00000ade3e00 x28: ffff80003c4ed400
> [   34.790039] x27: ffff0000089e1000 x26: 000000000000005f
> [   34.790397] x25: 0000000000000124 x24: 0000000000000015
> [   34.790649] x23: 0000000080000000 x22: 0000ffffb1eb6b24
> [   34.790897] x21: 00000000ffffffff x20: 000080003600c000
> [   34.791149] x19: ffff800000000000 x18: 0000000000000007
> [   34.791397] x17: 0000000000000001 x16: 0000000000000019
> [   34.791648] x15: 0000000000000033 x14: 000000000000004c
> [   34.791903] x13: 0000000000000068 x12: ffff000008af69d0
> [   34.792156] x11: ffff00000ade3c20 x10: 0000000000000000
> [   34.792451] x9 : 0000000000000000 x8 : 0000000000000000
> [   34.792706] x7 : 0000000000000000 x6 : ffff000008f42f18
> [   34.792965] x5 : dead000000000100 x4 : 0000000000000011
> [   34.793214] x3 : ffff80003c4ed400 x2 : ffff800000000004
> [   34.793462] x1 : 0001000000000000 x0 : 0000000000000000
> [   34.793743] Process a.out (pid: 1324, stack limit = 0xffff00000ade0000)
> [   34.794098] Call trace:
> [   34.794351] Exception stack(0xffff00000ade3cc0 to 0xffff00000ade3e00)
> [   34.794722] 3cc0: 0000000000000000 0001000000000000 ffff800000000004 ffff80003c4ed400
> [   34.795034] 3ce0: 0000000000000011 dead000000000100 ffff000008f42f18 0000000000000000
> [   34.795297] 3d00: 0000000000000000 0000000000000000 0000000000000000 ffff00000ade3c20
> [   34.795549] 3d20: ffff000008af69d0 0000000000000068 000000000000004c 0000000000000033
> [   34.795803] 3d40: 0000000000000019 0000000000000001 0000000000000007 ffff800000000000
> [   34.796066] 3d60: 000080003600c000 00000000ffffffff 0000ffffb1eb6b24 0000000080000000
> [   34.796277] 3d80: 0000000000000015 0000000000000124 000000000000005f ffff0000089e1000
> [   34.796477] 3da0: ffff80003c4ed400 ffff00000ade3e00 ffff0000080cdc68 ffff00000ade3e00
> [   34.796677] 3dc0: ffff0000080cde1c 00000000a0000145 0000000000000000 ffff80003c4ed400
> [   34.796884] 3de0: 0001000000000000 dead000000000100 ffff00000ade3e00 ffff0000080cde1c
> [   34.797153] [<ffff0000080cde1c>] SyS_waitid+0x1d4/0x210
> [   34.797298] Exception stack(0xffff00000ade3ec0 to 0xffff00000ade4000)
> [   34.797470] 3ec0: 0000000000000000 0000000000000000 ffff800000000000 0000000000000004
> [   34.797670] 3ee0: 0000000000000000 0400000055550400 0000ffffb1e0a011 0000000000000cbf
> [   34.797873] 3f00: 000000000000005f 0000000000000a3b 0000000000000000 16170f120a1a1311
> [   34.798073] 3f20: 00000000000001a8 0000ffffb1f8ecb8 0000ffffb1e1c0e0 0000000000000002
> [   34.798272] 3f40: 0000ffffb1eb6af0 0000000000420018 0000000000000000 0000000000400740
> [   34.798471] 3f60: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
> [   34.798668] 3f80: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
> [   34.798870] 3fa0: 0000000000000000 0000fffff7294200 00000000004006fc 0000fffff7294200
> [   34.799068] 3fc0: 0000ffffb1eb6b24 0000000080000000 0000000000000000 000000000000005f
> [   34.799265] 3fe0: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
> [   34.799474] [<ffff0000080837b0>] el0_svc_naked+0x24/0x28
> [   34.799715] Code: f9400fb4 17ffff9a d503201f f9000fb4 (d4210000)
> [   34.800121] ---[ end trace a14ca5cd5d8f9b30 ]---



-- 
Kees Cook
Pixel Security