Software Verification
description
Transcript of Software Verification
![Page 1: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/1.jpg)
1
Software Verification
Computer Science Club, Steklov Math InstituteLecture 1
Natasha SharyginaThe University of Lugano,
Carnegie Mellon University
![Page 2: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/2.jpg)
2
Outline Lecture 1:• Motivation• Model Checking in a Nutshell• Software Model Checking
– SAT-based approach
Lecture 2:• Verification of Evolving Systems (Component
Substitutability Approach)
![Page 3: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/3.jpg)
Bug Catching: Automated Program Analysis
Informatics DepartmentThe University of Lugano
Professor Natasha Sharygina
Guess what this is!
![Page 4: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/4.jpg)
Bug Catching: Automated Program Analysis
Informatics DepartmentThe University of Lugano
Professor Natasha Sharygina
Two trains, one bridge – model transformed with a simulation tool,
Hugo
![Page 5: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/5.jpg)
5
Motivation
• More and more complex computer systemsÞ exploding testing costs
• Increased functionalityÞ dependability concerns
• Increased dependabilityÞ reliability/security concerns
![Page 6: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/6.jpg)
6
System Reliability
• Bugs are unacceptable in safety/security-critical applications: mission control systems, medical devices, banking software, etc.
• Bugs are expensive: earlier we catch them, the better: e.g., Buffer overflows in MS Windows
• Automation is key to improve time-to-market
![Page 7: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/7.jpg)
8
Mars, December 3, 1999Crashed due to uninitialized variable
![Page 8: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/8.jpg)
10
![Page 9: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/9.jpg)
11
![Page 10: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/10.jpg)
12
Traditional Approaches
• Testing: Run the system on select inputs
• Simulation: Simulate a model of the system on select (often symbolic) inputs
• Code review and auditing
![Page 11: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/11.jpg)
13
What are the Problems?
• not exhaustive (missed behaviors)
• not all are automated (manual reviews, manual testing)
• do not scale (large programs are hard to handle)
• no guarantee of results (no mathematical proofs)
• concurrency problems (non-determinism)
![Page 12: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/12.jpg)
14
What is Formal Verification?• Build a mathematical model of the system:
– what are possible behaviors?
• Write correctness requirement in a specification language: – what are desirable behaviors?
• Analysis: (Automatically) check that model satisfies specification
![Page 13: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/13.jpg)
15
What is Formal Verification (2)?• Formal - Correctness claim is a precise mathematical
statement
• Verification - Analysis either proves or disproves the correctness claim
![Page 14: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/14.jpg)
16
Algorithmic Analysis by Model Checking
• Analysis is performed by an algorithm (tool)
• Analysis gives counterexamples for debugging
• Typically requires exhaustive search of state-space
• Limited by high computational complexity
![Page 15: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/15.jpg)
17
Temporal Logic Model Checking[Clarke,Emerson 81][Queille,Sifakis 82]
M |= P
“implementation” (system model)
“specification” (system property)
“satisfies”, “implements”, “refines” (satisfaction relation)
![Page 16: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/16.jpg)
18
M |= P
“implementation” (system model)
“specification” (system property)
“satisfies”, “implements”, “refines”, “confirms”, (satisfaction relation)
more detailed
more abstract
Temporal Logic Model Checking
![Page 17: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/17.jpg)
19
M |= P
system model system property
satisfaction relation
Temporal Logic Model Checking
![Page 18: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/18.jpg)
20
variable-based vs. event-based interleaving vs. true concurrency synchronous vs. asynchronous interaction clocked vs. speed-independent progress etc.
Decisions when choosing a system model:
![Page 19: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/19.jpg)
21
Characteristics of system models
which favor model checking over other verification techniques:
ongoing input/output behavior (not: single input, single result)
concurrency
(not: single control flow) control intensive
(not: lots of data manipulation)
![Page 20: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/20.jpg)
22
While the choice of system model is important for ease of modeling in a given situation,
the only thing that is important for model checking is that the system model can be translated into some form of state-transition graph.
Decisions when choosing a system model:
![Page 21: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/21.jpg)
23
Finite State Machine (FSM)
• Specify state-transition behavior• Transitions depict observable behavior
ERROR
unlock unlock
lock
lock
Acceptable sequences of acquiring and releasing a lock
![Page 22: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/22.jpg)
24
High-level View
LinuxKernel
(C)Spec(FSM)
ConformanceCheck
![Page 23: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/23.jpg)
25
High-level View
LinuxKernel
(C)
Finite StateModel(FSM)
Spec(FSM)
By Construction
Model Checking
![Page 24: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/24.jpg)
26
State-transition graph
Q set of statesI set of initial statesP set of atomic observationT Q Q transition relation [ ]: Q 2P observation function
Low-level View
![Page 25: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/25.jpg)
27
a
a,b b
q1
q3q2
Run: q1 q3 q1 q3 q1 state sequenceTrace: a b a b a observation sequence
![Page 26: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/26.jpg)
28
Model of Computation
Infinite Computation Tree
a b
b c
c
c
a b c
a b
b c c
State Transition Graph
Unwind State Graph to obtain Infinite Tree.
A trace is an infinite sequence of state observations
![Page 27: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/27.jpg)
29
Semantics
Infinite Computation Tree
a b
b c
c
c
a b c
a b
b c c
State Transition Graph
The semantics of a FSM is a set of traces
![Page 28: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/28.jpg)
30
Where is the model?
• Need to extract automatically• Easier to construct from hardware• Fundamental challenge for software
Linux Kernel~1000,000 LOC
Recursion and data structuresPointers and Dynamic memory
Processes and threads
Finite StateModel
![Page 29: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/29.jpg)
31
Mutual-exclusion protocol
loop out: x1 := 1; last := 1 req: await x2 = 0 or last = 2 in: x1 := 0end loop.
loop out: x2 := 1; last := 2 req: await x1 = 0 or last = 1 in: x2 := 0end loop.
||
P1 P2
![Page 30: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/30.jpg)
32
oo001
rr112
ro101 or012
ir112
io101
pc1: {o,r,i} pc2: {o,r,i} x1: {0,1} x2: {0,1} last: {1,2}
33222 = 72 states
![Page 31: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/31.jpg)
33
The translation from a system description to a state-transition graph usually involves an exponential blow-up !!!
e.g., n boolean variables 2n states
This is called the “state-explosion problem.”
State space blow up
![Page 32: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/32.jpg)
34
M |= P
system model system property
satisfaction relation
Temporal Logic Model Checking
![Page 33: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/33.jpg)
35
operational vs. declarative: automata vs. logic
may vs. must: branching vs. linear time
prohibiting bad vs. desiring good behavior: safety vs. liveness
Decisions when choosing system properties:
![Page 34: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/34.jpg)
36
System Properties/Specifications
- Atomic propositions: properties of states- (Linear) Temporal Logic Specifications: properties of
traces.
![Page 35: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/35.jpg)
37
Specification (Property) Examples: Safety (mutual exclusion): no two processes can be at the
critical section at the same time
Liveness (absence of starvation): every request will beeventually granted
Linear Time Logic (LTL) [Pnueli 77]: logic of temporal sequences.
• next (): holds in the next state
• eventually(): holds eventually
• always(): holds from now on
• until : holds until holds
![Page 36: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/36.jpg)
39
Examples of the Robot Control Properties
• Configuration Validity Check:If an instance of EndEffector is in the “FollowingDesiredTrajectory” state, then the instance of the corresponding Arm class is in the ‘Valid” state
Always((ee_reference=1) ->(arm_status=1)
• Control Termination: Eventually the robot control terminates
EventuallyAlways(abort_var=1)
![Page 37: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/37.jpg)
40
What is “satisfy”?M satisfies S if all the reachable states satisfy P
Different Algorithms to check if M |= P.
- Explicit State Space Exploration
For example: Invariant checking Algorithm.
1. Start at the initial states and explore the states of Musing DFS or BFS.
2. In any state, if P is violated then print an “error trace”.
3. If all reachable states have been visited then say “yes”.
![Page 38: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/38.jpg)
41
State Space Explosion
Problem: Size of the state graph can be exponential in size
of the program (both in the number of the program variables and the number of program components)
M = M1 || … || Mn
If each Mi has just 2 local states, potentially 2n global states
Research Directions: State space reduction
![Page 39: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/39.jpg)
42
Abstractions
• They are one of the most useful ways to fight the state explosion problem
• They should preserve properties of interest: properties that hold for the abstract model should hold for the concrete model
• Abstractions should be constructed directly from the program
![Page 40: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/40.jpg)
43
Data AbstractionGiven a program P with variables x1,...xn , each over domain D, the concrete model of P is defined over states (d1,...,dn) D...D Choosing
• Abstract domain A• Abstraction mapping (surjection) h: D A
we get an abstract model over abstract states (a1,...,an) A...A
![Page 41: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/41.jpg)
44
ExampleGiven a program P with variable x over the integersAbstraction 1:A1 = { a–, a0, a+ } a+ if d>0h1(d) = a0 if d=0 a– if d<0
Abstraction 2:A2 = { aeven, aodd }h2(d) = if even( |d| ) then aeven else aodd
![Page 42: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/42.jpg)
45
h h h
Existential Abstraction
M
A
M < A
![Page 43: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/43.jpg)
46
A
Existential Abstraction
1
2 3
4 6
a b
c f
M
[2,3]
[4,5] [6,7]
[1]
5 7
ed
a b
c d fe
![Page 44: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/44.jpg)
47
Existential Abstraction
• Every trace of M is a trace of A
– A over-approximates what M can do (Preserves safety properties!): A satisfies M satisfies
• Some traces of A may not be traces of M
– May yield spurious counterexamples - < a, e >
• Eliminated via abstraction refinement
– Splitting some clusters in smaller ones– Refinement can be automated
![Page 45: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/45.jpg)
48
A
Original Abstraction
1
2 3
4 6
a b
c f
M
[2,3]
[4,5] [6,7]
[1]
5 7
ed
a b
c d fe
![Page 46: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/46.jpg)
49
A
Refined Abstraction
1
2 3
4 6
a b
c f
M
[4,5] [6,7]
[1]
5 7
ed
a b
c d
[2] [3]
e f
![Page 47: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/47.jpg)
50
Predicate Abstraction
[Graf/Saïdi 97]
• Idea: Only keep track of predicates on data
• Abstraction function:
![Page 48: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/48.jpg)
56
Existential AbstractionPredicates
i++;
Basic Block Formula
Current Abstract State Next Abstract State
p1 p2 p3
0 0 0
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1
p’1 p’2 p’3
0 0 0
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1
?Query
![Page 49: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/49.jpg)
57
Existential AbstractionPredicates
i++;
Basic Block Formula
Current Abstract State Next Abstract State
p1 p2 p3
0 0 0
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1
p’1 p’2 p’3
0 0 0
0 0 1
0 1 0
0 1 1
1 0 0
1 0 1
1 1 0
1 1 1
Query
?
… and so on …
![Page 50: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/50.jpg)
59
Predicate Abstraction for Software
• How do we get the predicates?
• Automatic abstraction refinement!
[Kurshan et al. ’93][Clarke et al. ’00]
[Ball, Rajamani ’00]
![Page 51: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/51.jpg)
60
Abstraction Refinement Loop
ActualProgram
ConcurrentBooleanProgram
ModelChecker
Abstraction refinement
VerificationInitial
AbstractionNo error
or bug found
Spurious counterexample
Simulator
Propertyholds
Simulationsuccessful
Bug found
Refinement
Counterexample
![Page 52: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/52.jpg)
61
SLAM• Tool to automatically check device drivers for certain errors
– Takes as input Boolean programs
• Used as a Device Driver Development Kit
• Full detail (and all the slides) available at http://research.microsoft.com/slam/
![Page 53: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/53.jpg)
63
Problems with Existing Tools• Existing tools (BLAST, SLAM, MAGIC) use a Theorem
Prover like Simplify
• Theorem prover works on real or natural numbers, but C uses bit-vectors false positives
• Most theorem provers support only few operators(+, -, <, ≤, …), no bitwise operators
• Idea: Use SAT solver to do bit-vector! - SATABS
![Page 54: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/54.jpg)
64
Abstraction with SAT - SATABS• Successfully used for abstraction of C programs
(Clarke, Kroening, Sharygina, Yorav ’03 – SAT-based predicate abstraction)
• There is now a version of SLAM that has it– Found previously unknown Windows bug
• Create a SAT instance which relates initial value of predicates, basic block, and the values of predicates after the execution of basic block
• SAT also used for simulation and refinement
![Page 55: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/55.jpg)
65
Our Solution
This solves two problems:1. Now can do all ANSI-C
integer operators, including *, /, %, <<, etc.
2. Sound with respect to overflow
No moreunnecessary spurious
counterexamples!
Use SAT solver!1. Generate query equation with
predicates as free variables
2. Transform equation into CNF using Bit Vector Logic
One satisfying assignment matches one abstract transition
3. Obtain all satisfying assignments= most precise abstract transition relation
![Page 56: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/56.jpg)
66
Abstraction Refinement Loop
ActualProgram
ConcurrentBooleanProgram
ModelChecker
Abstraction refinement
VerificationInitial
AbstractionNo error
or bug found
Spurious counterexample
Simulator
Propertyholds
Simulationsuccessful
Bug found
Refinement
![Page 57: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/57.jpg)
67
Model Checkers for Boolean Programs
• Explicit State– Zing– SPIN
• Symbolic– Moped– Bebop– SMV
![Page 58: Software Verification](https://reader035.fdocuments.in/reader035/viewer/2022062520/56815ed3550346895dcd681a/html5/thumbnails/58.jpg)
70
Experimental Results
• Comparison of SLAM with Integer-based theorem prover against SAT-based SLAM
• 308 device drivers
• Timeout: 1200s