llvm dev - Aug 2015 - Fuzzing complex programs

On Sun, Aug 30, 2015 at 9:11 AM, Brian Cain via llvm-dev <
llvm-dev at lists.llvm.org> wrote:
>
>
> On Sun, Aug 30, 2015 at 9:30 AM, Greg Stark via llvm-dev <
> llvm-dev at lists.llvm.org> wrote:
>
>> I have a project I want to do based on Libfuzzer. Is there a separate
>> list for it or should I bring up any ideas for it here?
>>
>> No separate list so far, this one should be good.
> What I have in mind is to fuzz Postgres. Trying to fuzz the SQL
>> interpreter in general
>> is not very productive because traditional fuzzers try to execute the
>> entire program repeatedly and it has a fairly high startup and
>> shutdown cost. Also the instrumentation-guided approach has
>>
>
> One challenge in leaving the daemon up while testing is knowing how well
> isolated the test cases are from one another.  It may be the case that the
> test cases somehow accumulate some global state (test case N triggers heap
> corruption, N + 23 crashes as a result of that earlier corruption).  At
> least that specific failure mode can probably be mitigated by using one or
> more of the sanitizers though.
>
>
This is true, however accumulating global state increases the chances to
find complex bugs (at the cost of increased cost of analyzing bugs).
We have seen a few such cases, e.g.
https://sourceware.org/bugzilla/show_bug.cgi?id=18043#c11

> limitations due to the way lexing and parsing works as well as the
>> large amount of
>> internal state causing non-deterministic internal behaviour (garbage
>> collecting persistent data structures, etc).
>>
>> However there are a number of internal functions that would be very
>> feasible to fuzz. Things like the datatype input/output functions
(I'm
>> particularly thinking of the datetime parser), regular expression
>> library, etc.
>>
>
In my (biased) opinion libFuzzer is particularly well suited for this task
(fuzzing individual libraries, as opposed to fuzzing the whole postgress).
I've played with a dozen of regular expression libs and found bugs in all
of them
(e.g. search for "Fuzzer" in
http://vcs.pcre.org/pcre2/code/trunk/ChangeLog?view=markup&pathrev=360)


>> To do this effectively I think it would be best to invoke the fuzzer
>> from inside Postgres.
>
>
Never tied this.
Can't you just link libFuzzer with a part of the code you want to test?

> Essentially provide bindings for Libfuzzer so
>> you can I can have Libfuzzer provide all the test cases to repeatedly
>> call the internal functions on.
>>
>> Is there any example of doing something like this already? Am I taking
>> a crazy approach?
>>
>>
> I don't have enough experience to say if it's crazy or not.  But if
> your LLVMFuzzerTestOneInput() queues some work for the server and pends on
> a response -- that seems like a sane approach.
>
>
>> There are other approaches possible. It would be nice if I could run
>> afl or libfuzzer on a client program and have the client program tell
>> afl or libfuzzer the pid of the server to watch and then request test
>> cases to feed to the server. That seems like it would be a more
>> flexible approach for a lot of use cases where the server requires
>> setting up a complex environment.
>>
>>
> Great idea, but it seems tricky to get the execution coverage feedback in
> this case.
>
Not very tricky, but less efficient.
The major benefit of libfuzzer is that it get the coverage feedback from
inside the process
avoiding any kind of inter-process communication (no syscalls even).
So for things like simple parsers you can get 50K executions per second
(unless the fuzzer finds  exponential algorithms in the parser).


>
> Let me know if you're interested in collaborating, it sounds
interesting.
> Though at first glance, I'd prefer the "not very productive"
brute force
> option and just toss more resources at it.
>
> --
> -Brian
>
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20150830/bd7af080/attachment.html>

On 30 Aug 2015 20:04, "Kostya Serebryany" <kcc at google.com>
wrote:
>
>
> In my (biased) opinion libFuzzer is particularly well suited for this task
> (fuzzing individual libraries, as opposed to fuzzing the whole postgress).
> I've played with a dozen of regular expression libs and found bugs in
all
of them
> (e.g. search for "Fuzzer" in
http://vcs.pcre.org/pcre2/code/trunk/ChangeLog?view=markup&pathrev=360)
>
>
>>>
>>> To do this effectively I think it would be best to invoke the
fuzzer
>>> from inside Postgres.
>
>
> Never tied this.
> Can't you just link libFuzzer with a part of the code you want to test?
The problem I'm specifically trying to tackle is that the code in question
can use any of the internal postgres APIs and might have dependencies on
anything in the environment.

Even the simplest cases like the date/tone parser will depends on the
timezone library which is initialised at startup, the server session state
which specifies the current timezone and default date format, etc.

The more interesting cases like arrays and other compound objects will
depend on the internal caches of the database schema which is where it
finds things like meta information about the data types stored within.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20150830/87001b01/attachment.html>

On Sun, Aug 30, 2015 at 1:09 PM, Greg Stark <stark at mit.edu> wrote:
>
> On 30 Aug 2015 20:04, "Kostya Serebryany" <kcc at
google.com> wrote:
> >
> >
> > In my (biased) opinion libFuzzer is particularly well suited for this
> task
> > (fuzzing individual libraries, as opposed to fuzzing the whole
> postgress).
> > I've played with a dozen of regular expression libs and found bugs
in
> all of them
> > (e.g. search for "Fuzzer" in
> http://vcs.pcre.org/pcre2/code/trunk/ChangeLog?view=markup&pathrev=360)
> >
> >
> >>>
> >>> To do this effectively I think it would be best to invoke the
fuzzer
> >>> from inside Postgres.
> >
> >
> > Never tied this.
> > Can't you just link libFuzzer with a part of the code you want to
test?
>
> The problem I'm specifically trying to tackle is that the code in
question
> can use any of the internal postgres APIs and might have dependencies on
> anything in the environment.
>
> Even the simplest cases like the date/tone parser will depends on the
> timezone library which is initialised at startup, the server session state
> which specifies the current timezone and default date format, etc.
>
> The more interesting cases like arrays and other compound objects will
> depend on the internal caches of the database schema which is where it
> finds things like meta information about the data types stored within.
>
I see.
At least one of the users of Sanitizer Coverage did something similar with
their server,
but they implemented their own mutator/fuzzer (did not use libFuzzer).

With libFuzzer you can't do that today because it expects to receive the
coverage feedback from the same process.
With somewhat modification you probably can achieve this:
in Fuzzer::RunOneMaximizeTotalCoverage replace the calls to
__sanitizer_get_total_unique_coverage
with something that gets this value for the server process.
https://github.com/llvm-mirror/llvm/blob/master/lib/Fuzzer/FuzzerLoop.cpp#L231

Similarly, for the 8-bit counters (-use_counters=1), in the server process
call
__sanitizer_update_counter_bitset_and_clear_counters and pass the result to
the client app.

If you can may the change general enough the patch would be more than
welcome.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20150830/9558ad7a/attachment.html>