netdev - Re: [PATCH net] bpf: disallow arithmetic operations on context pointer

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <e2c4322a-301f-47f9-d102-725138d1bf67@solarflare.com>
Date:   Mon, 16 Oct 2017 17:47:45 +0100
From:   Edward Cree <ecree@...arflare.com>
To:     Jakub Kicinski <jakub.kicinski@...ronome.com>
CC:     <netdev@...r.kernel.org>, <oss-drivers@...ronome.com>,
        <alexei.starovoitov@...il.com>, <daniel@...earbox.net>
Subject: Re: [PATCH net] bpf: disallow arithmetic operations on context
 pointer

On 16/10/17 17:30, Jakub Kicinski wrote:
> On Mon, 16 Oct 2017 17:16:24 +0100, Edward Cree wrote:
>> On 16/10/17 16:45, Jakub Kicinski wrote:
>>> diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
>>> index 8b8d6ba39e23..8499759d0c7a 100644
>>> --- a/kernel/bpf/verifier.c
>>> +++ b/kernel/bpf/verifier.c
>>> @@ -1116,7 +1116,12 @@ static int check_mem_access(struct bpf_verifier_env *env, int insn_idx, u32 regn
>>>  		/* ctx accesses must be at a fixed offset, so that we can
>>>  		 * determine what type of data were returned.
>>>  		 */
>>> -		if (!tnum_is_const(reg->var_off)) {
>>> +		if (reg->off) {
>>> +			verbose("derefence of modified ctx ptr R%d off=%d+%d, ctx+const is allowed, ctx+const+const is not\n",  
>> This is slightly unclear, it's not that two adds is bad (e.g. r1 += 8;
>>  r0 = *(u32 *)r1 is bad too), it's that the offset must be in the load,
>>  not the register; your message might be accurate for some compilers but
>>  not in full generality (especially for assemblers without compiling).
> I'm happy to hear better suggestions :)  I've spent quite a bit of time
> scratching my head thinking how to phrase this best.  The first
> part of the message is general enough IMHO, the second is targeted
> mostly at C developers.
Hmm, what really bugs me is that if e.g. the compiler turned
   *(ctx + 4 + 4)
 or
   ctx[4 + 4]
 or even
   ctx->arraymemb[4]
 into this kind of arithmetic on ctx, arguably that would be a bug in the
 compiler — if it's doing proper constexpr folding on its IR (or something
 along those lines) it should be able to turn them all into good LDX.  The
 same even goes for if (ctx + 4) got stored in a local, because there's no
 reason that has to map to a register.
So it's not even that "your C source breaks the rules", it's that "your C
 compiler did something silly that we don't handle".
Maybe the message should be "compiler maybe mishandled ctx+const+const"?

-Ed