Re: [RFC v3 36/45] NFSv4: Fix GETATTR bitmap verification

[Trond Myklebust]

On Thu, May 28, 2015 at 5:40 PM, Trond Myklebust
<trond.myklebust@xxxxxxxxxxxxxxx> wrote:
> On Thu, May 28, 2015 at 5:05 PM, Andreas GrÃnbacher
> <andreas.gruenbacher@xxxxxxxxx> wrote:
>> 2015-05-28 22:50 GMT+02:00 Trond Myklebust <trond.myklebust@xxxxxxxxxxxxxxx>:
>>> On Thu, May 28, 2015 at 4:33 PM, J. Bruce Fields <bfields@xxxxxxxxxxxx> wrote:
>>>> On Fri, Apr 24, 2015 at 01:04:33PM +0200, Andreas Gruenbacher wrote:
>>>>> The NFSv4 client sends the server GETATTR requests with different sets of
>>>>> requested attributes depending on the situation.  The requested set of
>>>>> attributes is encoded in a bitmap; the server replies with the set of
>>>>> attributes it could return.  These bitmaps can be several words wide.  The
>>>>> bitmap returned by the server is a subset of the bitmap sent by the client.
>>>>>
>>>>> While decoding the reply, the client tries to verify the reply bitmap: it
>>>>> checks if any previous, unexpected attributes are left in the same word of the
>>>>> bitmap for each attribute it tries to decode, then it clears the current
>>>>> attribute's bit in the bitmap for the next decode function.
>>>>>
>>>>> The client fails to detect when unexpected attributes are sent after the last
>>>>> expected attribute in each word in the bitmap.
>>>>
>>>> Is it important that the client catch that?
>>>
>>> Right. What is the actual problem or bug that this patch is trying to
>>> fix? Why do we care if a buggy server sends us extra info that we
>>> didn't ask for?
>>
>> I think we do care to correctly decode (and reject) well-formed but illegal
>> server replies. In this case, when switching to the next word of a bitmap, the
>> client doesn't check if the previous word has been completely "consumed" yet.
>> If any attributes are "missed", decoding the attribute values gets out of sync,
>> garbage is decoded, and the error may be missed.
>>
>
> We already do this kind of check with the existing code. What's wrong with it?
>

Actually, you're right, we don't check for the previous word, however
fixing that is a question of adding 2 extra checks in
decode_getfattr_attrs(), one in decode_getfattr_statfs(), and one in
decode_fsinfo().

It shouldn't require a rewrite of the entire nfs4xdr.c.

Trond
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/