1. Automated Whitebox Fuzz Testing Network and Distributed System Security (NDSS) 2008 by Patrice Godefroid, ‏Michael Y. Levin, and ‏David Molnar Present by Diego Velasquez

2. 2 Acknowledgments  Figures are copy from the paper.  Some slides were taken from the original presentation presented by the authors

3. 3 Outline  Summary  Goals  Motivations  Methods  Experiments  Results  Conclusions  Review  Strengths  Weakness  Extensions  Reference

4. Goals  Propose a novel methodology that performs efficiently fuzz testing.  Introduce a new search algorithm for systematic test generation.  Outcast their system SAGE (Scalable, Automated, Guided Execution) 4

5. Methods  Fuzz testing inserts random data to input of applications in order to find defects of a software system. Heavily used in Security testing.  Pros: Cost effective and can find most of known bugs  Cons: It has some limitations depending on some types of branches, for example on project 2 in order to find bug # 10 we need to execute the if statement below. if(address ==613 && value >= 128 && value<255)//Bug #7 printf("BUG 10 TRIGGERED);  Has (1 in 5000) * (128 in 2^32) in order to be executed if we know that is only 5000 addresses and value is a random 32-bit input 5

6. Methods Cont.  Whitebox Fuzz Testing  Combine fuzz testing with dynamic test generation [2]  Run the code with some initial input  Collect constraints on inputs with symbolic execution  Generate new constraints  Solve constraints with constraint solver  Synthesize new inputs 6

7. Methods Cont.  The Search Algorithm figure 1 from [1]  Black box will do poorly in this case  Dynamic test could do better 7

8. Methods Cont.  Dynamic Approach  Input ‘good’ as example  Collect constrain from trace  Create a new path constraint Figure 2 from [1] 8

9. Methods Cont.  Limitations of Dynamic Testing  Path Explosion  Path doesn’t scale to large in realistic programs.  Can be corrected by modifying the search algorithm.  Imperfect Symbolic Execution  Could be imprecise due to Complex program statements (arithmetic, pointer manipulation)  Calls to OS have to be expensive in order to be precise 9

10. Methods Cont.  New Generation Search Algorithm Figure 3 and figure 4 from [1]  A type of Bread First Search with heuristic to get more input test cases.  Scores return the number of new test cases covered. 10

11. Methods Cont.  Summary of Generation Search Algorithm  Push input to the list  Run&Check(input) check bugs in that input  Traverse the list by selecting from the list base in score  Expanded child paths and adding to the childlist  Traverse childlist Run&Check, assigned score and add to list  Expand Execution  Generates Path constrain  Attempt to expand path constraints and save them  Input.bound is bound is used to limit the backtracking of each sub-search above the branch. 11

12. Experiments  Can test any file-reading program running on Windows by treating bytes read from files as symbolic input.  Another key novelty of SAGE is that it performs symbolic execution of program traces at the x86 binary level 12 FIGURE FROM [2]

13. Experiments Cont.  Sage advantages  Not source-based, SAGE is a machine-code-based, so it can run different languages.  Expensive to build at the beginning, but less expensive over time  Test after shipping,  Since is based in symbolic execution on binary code, SAGE can detects bugs after the production phase  Not source is needed like in another systems  SAGE doesn’t even need specific data types or structures not easy visible in machine code 13

14. Experiments Cont.  MS07-017: Vulnerabilities in Graphics Device Interface (GDI) Could Allow Remote Code Execution.  Test in different Apps such as image processors, media players, file decoders.[2]  Many bugs found rated as “security critical, severity 1, priority 1”[2]  Now used by several teams regularly as part of QA process.[2] 14

15. Experiments Cont.  More in MS07-017, figure below is from [2] left is input right is crashing test case 15 RIFF...ACONLIST B...INFOINAM.... 3D Blue Alternat e v1.1..IART.................... 1996..anih$...$...................rate....................seq....................LIST....framic on........... RIFF...ACONB B...INFOINAM.... 3D Blue Alternat e v1.1..IART.................... 1996..anih$...$...................rate....................seq....................anih....framic on........... Only 1 in 2 32 chance at random!

16. Results  Statistics from 10hour searches on seven test applications, each seeded with a well formed input file. 16

17. Results  Focused on the Media 1 and Media 2 parsers.  Ran a SAGE search for the Media 1 parser with five “well-formed” media files, and five bogus files. Figure 7 from [1] 17

18. Results  Compared with Depth-First Search Method  DFS runs for 10 hours for Media 2 with wff-2 and wff-3, didn’t find anything GS found 15 crashes  Symbolic Execution is slow  Well formed input are better than Bogus files  Non-determinism in Coverage Results.  The heuristic method didn’t have too much impact  Divergences are common 18

19. Results  Most bugs found are “shallow” Figure from [2] 19

20. Conclusions  Blackbox vs. Whitebox Fuzzing  Cost/precision tradeoffs  Blackbox is lightweight, easy and fast, but poor coverage  Whitebox is smarter, but complex and slower  Recent “semi-whitebox” approaches  Less smart but more lightweight: Flayer (taint-flow analysis, may generate false alarms), Bunny-the-fuzzer (taint-flow, source-based, heuristics to fuzz based on input usage), autodafe, etc.  Which is more effective at finding bugs? It depends…  Many apps are so buggy, any form of fuzzing finds bugs!  Once low-hanging bugs are gone, fuzzing must become smarter: use whitebox and/or user-provided guidance (grammars, etc.)‏  Bottom-line: in practice, use both! *Slide From [2] 20

21. 21 Strengths  Novel approach to do fuzz testing  Introduced new search algorithm that use code- coverage maximizing heuristic  Applied as a black box  Not source code was needed  symbolic execution of program at the x86 binary level  Shows results comparing previous results  Test large applications previously tested found more bugs.  Introduced a full system and applied the novel ideas in this paper

22. 22 Weakness  The results were non-determinism  Same input, program and idea different results.  Only focus in specific areas  X86 windows applications  File manipulation applications  Well formed input still some type of regular fuzzing testing  SAGE needs help from different tools  In my opinion the paper extends too much in the implementation of SAGE, and the system could of be too specific to Microsoft

23. 23 Extensions  Make SAGE more general  Easy to implement to another architectures  Use for another types of applications  Linux based applications  Better way to create input files  May be used of grammar  Make the system deterministic  Having different results make me think that it could not be reliable.

24. Reference  [1] P Godefroid, MY Levin, D Molnar, Automated Whitebox Fuzz Testing, NDSS, 2008.  [2] Original presentation slides www.truststc.org/pubs/366/15%20- %20Molnar.ppt www.truststc.org/pubs/366/15%20- %20Molnar.ppt  [3] Wikipedia Fuzz testing http://en.wikipedia.org/wiki/Fuzz_testing. http://en.wikipedia.org/wiki/Fuzz_testing 24

25. 25  Questions, Comments or Suggestions?