code optimisation

1. CH10.2
CSE
4100
OverviewOverview
 Code Level OptimizationCode Level Optimization
 Common Sub-expression elimination
 Copy Propagation
 Dead-code elimination
 Peephole optimizationPeephole optimization
 Load/Store elimination
 Unreachable code
 Flow of Control Optimization
 Algebraic simplification
 Strength Reduction

2. CH10.4
CSE
4100
Where can Optimation Occur?Where can Optimation Occur?
 Software Engineer can:Software Engineer can:
 Profile Program
 Change Algorithm Data
 Transform/Improve Loops
Front End
LA, Parse,
Int. Code
Code
Generator
Int. Code Target
Program
Source
Program
 Compiler Can:Compiler Can:
 Improve Loops/Proc
Calls
 Calculate Addresses
 Use Registers
 Selected Instructions
 Perform Peephole Opt.
 All are OptimizationsAll are Optimizations
 1st
is User Controlled and Defined
 At Intermediate Code Level by Compiler
 At Assembly Level for Target Architecture (to take
advantage of different machine features)

3. CH10.6
CSE
4100
First Look at OptimizationFirst Look at Optimization
 Optimization Applied to 3 Address Coding (3AC)Optimization Applied to 3 Address Coding (3AC)
Version of Source Program - Examples:Version of Source Program - Examples:

4. CH10.7
CSE
4100
First Look at OptimizationFirst Look at Optimization
 Identify each Basic Block which
Represents a set of Three Address
Statements where
Execution Enters at Top and Leaves
at Bottom
No Branches within Code
 Represent the Control Flow
Dependencies Among and Between
Basic Blocks
Defines what is Termed a “Flow
Graph”

5. CH10.8
CSE
4100
First Look at OptimizationFirst Look at Optimization
 Steps 1 to 12 from two Slides Back Represented as:Steps 1 to 12 from two Slides Back Represented as:
 Optimization Works with Basic Blocks and FlowOptimization Works with Basic Blocks and Flow
Graph to Perform Transformations that:Graph to Perform Transformations that:

6. CH10.9
CSE
4100
First Look at OptimizationFirst Look at Optimization
 Optimization will PerformOptimization will Perform
Transformations on BasicTransformations on Basic
Blocks/Flow GraphBlocks/Flow Graph
 Resulting Graph(s) PassedResulting Graph(s) Passed
Through to Final Code GenerationThrough to Final Code Generation
to Obtain More Optimal Codeto Obtain More Optimal Code
 Two Fold Goal of OptimizationTwo Fold Goal of Optimization
Reduce Time
Reduce Space

7. CH10.10
CSE
4100
First Look at OptimizationFirst Look at Optimization
 Two Types of TransformationsTwo Types of Transformations
 Structure Preserving
Inherent Structure and Implicit
Functionality of Basic Blocks is
Unchanged
 Algebraic
Elimination of Useless Expressions
x = x + 0 or y = y * 1
Replace Expensive Operators
Change x = y ** 2 to x = y * y

8. CH10.11
CSE
4100
Structure Preserving TransformationsStructure Preserving Transformations
 Common Sub-Expression EliminationCommon Sub-Expression Elimination
 How can Following Code be Improved?
a = b + c
b = a – d
c = b + c
d = a – d
 What Must Make Sure Doesn’t happen?
 Dead-Code EliminationDead-Code Elimination
 If x is not Used in Block, Can it be
Removed?
x = y + z
d = b

9. CH10.12
CSE
4100
Structure Preserving TransformationsStructure Preserving Transformations
 Renaming Temporary VariablesRenaming Temporary Variables
 Consider the code t = b + c
 Can be Changed to u = b + c
 May Reduce the Number of temporaries
 Make Change from all t’s to all u’s
 Interchange of StatementsInterchange of Statements
 Consider and Change to:
t1 = b + c t2 = x + y
t2 = x + y t1 = b + c
 This can Occur as Long as:
 x and y not t1
 b and c not t2

10. CH10.13
CSE
4100
Requirements for OptimizationRequirements for Optimization

11. CH10.14
CSE
4100
The Overall Optimization ProcessThe Overall Optimization Process
 AdvantagesAdvantages
 Intermediate Code has Explicit Operations and Their
Identification Promotes Optimization
 Intermediate Code is Relatively Machine Independent
 Therefore, Optimization Doesn’t Impact Final Code
Generation

12. CH10.15
CSE
4100
Example Source CodeExample Source Code

13. CH10.16
CSE
4100
Generated Three Address CodingGenerated Three Address Coding

14. CH10.17
CSE
4100
Flow Graph of Basic BlocksFlow Graph of Basic Blocks

15. CH10.18
CSE
4100
Indepth Examination of OptimizationIndepth Examination of Optimization
 Code-Transformation Techniques:Code-Transformation Techniques:
 Local – within a “Basic Block”
 Global – between “Basic Blocks”
 Data Flow Dependencies Determined byData Flow Dependencies Determined by
InspectionInspection
what do i, a, and v refer to?what do i, a, and v refer to?
 Dependent in Another Basic BlockDependent in Another Basic Block
 Scoping is Very CriticalScoping is Very Critical

16. CH10.19
CSE
4100
Indepth Examination of OptimizationIndepth Examination of Optimization
 Function Preserving TransformationsFunction Preserving Transformations
 Common Subexpressions
 Copy Propagation
 Deal Code Elimination
 Constant Folding
 Loop OptimizationsLoop Optimizations
 Code Motion
 Induction Variables
 Strength Reduction

17. CH10.20
CSE
4100
Common Sub-ExpressionsCommon Sub-Expressions
 E is a Common Sub-Expression ifE is a Common Sub-Expression if
 E as Previously Computed
 Value of E Unchanged since Previous
Coamputation
 What Can be Saved in B5?What Can be Saved in B5?
 t6 and t7 same computation
 t8 and t10 same computation
 Save:
 Remove 2 temp variables
 Remove 2 multiplications
 Remove 4 variable accesses
 Remove 2 assignments
t6 := 4 * i
x := a[t6]
t7 := 4 * i
t8 := 4 * j
t9 := a[t8]
a[t7] := t9
t10 := 4 * j
a[t10]:= x
Goto B2

18. CH10.21
CSE
4100
Common Sub-ExpressionsCommon Sub-Expressions
 What about B6?What about B6?
 t11 and t12
 t13 and t15
 Similar Savings as in B5Similar Savings as in B5
t11 := 4 * i
x := a[t11]
t12 := 4 * i
t13 := 4 * n
t14 := a[t13]
a[t12]:= t14
t15 := 4 * n
a[t15]:= x
t11 := 4 * i
x := a[t11]
t13 := 4 * n
t14 := a[t13]
a[t11]:= t14
a[t13]:= x

19. CH10.22
CSE
4100
Common Sub-ExpressionsCommon Sub-Expressions
 What else Can be Accomplished?What else Can be Accomplished?
 Where is Variable j Determined?Where is Variable j Determined?
 In B3 – and when drop
through B3 to B4 and into B5,
no change occurs to j!
 What Does B5 Become?What Does B5 Become?
 t9 same as t5!t9 same as t5!
 Again savings in access,Again savings in access,
variables, operations, etc.variables, operations, etc.
t6 := 4 * i
x := a[t6]
t8 := 4 * j
t9 := a[t8]
a[t6] := t9
a[t8]:= x
Goto B2
j := j - 1
t4 := 4 * j
t5 := a[t4]
if t5>4 goto B3
B4
t6 := 4 * i
x := a[t6]
t9 := a[t4]
a[t6] := t9
a[t4]:= x
Goto B2
t6 := 4 * i
x := a[t6]
a[t6] := t5
a[t4]:= x
Goto B2

20. CH10.24
CSE
4100
Common Sub-ExpressionsCommon Sub-Expressions
 B6 is Similarly Changed ….B6 is Similarly Changed ….
t11 := 4 * i
x := a[t11]
t13 := 4 * n
t14 := a[t13]
a[t11]:= t14
a[t13]:= x
x := t3
t14 := a[t1]
a[t2]:= t14
a[t1]:= x

21. CH10.25
CSE
4100
Resulting Flow DiagramResulting Flow Diagram

22. CH10.26
CSE
4100
Copy PropagationCopy Propagation
 Introduce a Common Copy Statement toIntroduce a Common Copy Statement to
Replace an Arithmetic Calculation withReplace an Arithmetic Calculation with
AssignmentAssignment
 Regardless of the Path Chosen, the use ofRegardless of the Path Chosen, the use of
an Assignment Saves Time and Spacean Assignment Saves Time and Space
a:= d + e
a:= t
a:= d + e b:= d + e
c:= d + e
b:= d + e
a:= t
c:= t

23. CH10.27
CSE
4100
Copy PropagationCopy Propagation
 In our Example for B5 and B6 Below:In our Example for B5 and B6 Below:
 Since x is t3, we can replace the use of x on right handSince x is t3, we can replace the use of x on right hand
side as below:side as below:
x := t3
t14 := a[t1]
a[t2]:= t14
a[t1]:= x
x := t3
a[t2] := t5
a[t4]:= x
Goto B2
x := t3
t14 := a[t1]
a[t2] := t14
a[t1] := t3
x := t3
a[t2] := t5
a[t4] := t3
Goto B2

24. CH10.28
CSE
4100
Dead Code EliminationDead Code Elimination
 Variable is “Dead” if its Value will never be UtilizedVariable is “Dead” if its Value will never be Utilized
Again SubsequentlyAgain Subsequently
 Otherwise, Variable is “Live”Otherwise, Variable is “Live”
 What’s True about B5 and B6?What’s True about B5 and B6?
 Can Any Statements be Eliminated? Which Ones?Can Any Statements be Eliminated? Which Ones?
Why?Why?
 B5 and B6 are Now Optimized withB5 and B6 are Now Optimized with
 B5 has 9 Statements Reduced to 3
 B56 has 8 Statements Reduced to 3
x := t3
t14 := a[t1]
a[t2] := t14
a[t1] := t3
x := t3
a[t2] := t5
a[t4] := t3
Goto B2

25. CH10.29
CSE
4100
Loop OptimizationsLoop Optimizations
 Three Types: Code Motion, Induction Variables, andThree Types: Code Motion, Induction Variables, and
Strength ReductionStrength Reduction
 Code MotionCode Motion
 Remove Invariant Operations from Loop
while (limit * 2 > i) do
 Replaced by:
t = limit * 2
while (t > i) do
 Induction VariablesInduction Variables
 Identify Which Variables are Interdependent or in
Step
j = j – 1
t4 = 4 * j
 Replaced by below with an initialization of t4
t4 = t4 - 4

26. CH10.30
CSE
4100
Loop OptimizationsLoop Optimizations
 Strength ReductionStrength Reduction
 Replace an Expensive Operation (Such as
Multiply) with a Cheaper Operation (Such as Add)
 In B4, I and j can be replaced with t2 and t4
 This Eliminates the need for Variables i and j

27. CH10.31
CSE
4100
Final Optimized Flow GraphFinal Optimized Flow Graph

28. CH10.43
CSE
4100
Peephole OptimizationPeephole Optimization
 Simple IdeaSimple Idea
 Slide a window over the code
 Optimize code in the window only.
 Optimizations areOptimizations are
 Local
 Semantic preserving
 Cheap to implement
 UsuallyUsually
 One can repeat the peephole several times!
 Each pass can create new opportunities for
more

29. CH10.44
CSE
4100
Peephole OptimizerPeephole Optimizer
block_3:
mov [esp-4],ebp
mov ebp,esp
mov [ebp-8],esp
sub esp,28
mov eax,[ebp+8]
cmp eax,0
mov eax,0
sete ah
cmp eax,0
jz block_5
block_4:
mov eax,1
jmp block_6
block_5:
mov eax,[ebp+8]
sub eax,1
push eax
mov eax,[ebp+4]
push eax
mov eax,[eax]
mov eax,[eax]
call eax
add esp,8
mov ebx,[ebp+8]
imul ebx,eax
block_3:
mov [esp-4],ebp
mov ebp,esp
mov [ebp-8],esp
sub esp,28
mov eax,[ebp+8]
cmp eax,0
mov eax,0
sete ah
cmp eax,0
jz block_5
block_4:
mov eax,1
jmp block_6
block_5:
mov eax,[ebp+8]
sub eax,1
push eax
mov eax,[ebp+4]
push eax
mov eax,[eax]
mov eax,[eax]
call eax
add esp,8
mov ebx,[ebp+8]
imul ebx,eax

30. CH10.45
CSE
4100
Peephole OptimizationsPeephole Optimizations
 Load/Store elimination
 Get rid of redundant operations
 Unreachable code
 Get rid of code guaranteed to never execute
 Flow of Control Optimization
 Simply jump sequences.
 Algebraic simplification
 Use rules of algebra to rewrite some basic operation
 Strength Reduction
 Replace expensive instructions by equivalent ones (yet
cheaper)
 Machine Idioms
 Replace expensive instructions by equivalent ones (for a
given machine)

31. CH10.46
CSE
4100
Load / Store SequencesLoad / Store Sequences
 Imagine the following sequenceImagine the following sequence
 “a” is a label for a memory location
mov
a,eax
mov
eax,a
mov
a,eax
mov
eax,a

32. CH10.47
CSE
4100
Unreachable CodeUnreachable Code
 ExampleExample
#define debug 0
....
if (debug) {
printf(“This is a trace
messagen”);
}
....
#define debug 0
....
if (debug) {
printf(“This is a trace
messagen”);
}
....

33. CH10.48
CSE
4100
ExampleExample
 The Generated code looks like....The Generated code looks like....
 If we know that...If we know that...
 debug == 0
 Then
....
if (debug == 0) goto L2
printf(“This is a trace
messagen”);
L2: ....
....
if (debug == 0) goto L2
printf(“This is a trace
messagen”);
L2: ....
....
if (0 == 0) goto L2
printf(“This is a trace
messagen”);
L2: ....
....
if (0 == 0) goto L2
printf(“This is a trace
messagen”);
L2: ....
1

34. CH10.49
CSE
4100
ExampleExample
 Final transformationFinal transformation
 Given this codeGiven this code
 There is no way to branch “into” the blue block
 The last instruction (goto L2) jumps over the blue
block
 The blue block is never used. Get rid of it!
....
goto L2
printf(“This is a trace
messagen”);
L2: ....
....
goto L2
printf(“This is a trace
messagen”);
L2: ....

35. CH10.50
CSE
4100
Unreachable Code ExampleUnreachable Code Example
 Bottom LineBottom Line
 Now L2 is instruction after goto...Now L2 is instruction after goto...
 So get rid of goto altogether!
....
goto L2
L2: ....
....
goto L2
L2: ....
....
L2: ....
....
L2: ....

36. CH10.51
CSE
4100
Flow of Control OptimizationFlow of Control Optimization
 SituationSituation
 We can have chains of jumps
 Direct to conditional or vice-versa
 ObjectiveObjective
 Avoid extra jumps.
 ExampleExample
if (x relop y) goto L2
....
L2: goto L4
L3: ....
L4:
L4_BLOCK
if (x relop y) goto L2
....
L2: goto L4
L3: ....
L4:
L4_BLOCK

37. CH10.52
CSE
4100
Flow of ControlFlow of Control
 What can be doneWhat can be done
 Collapse the chain
if (x relop y) goto L4
....
L2: goto L4
L3: ....
L4:
L4_BLOCK
if (x relop y) goto L4
....
L2: goto L4
L3: ....
L4:
L4_BLOCK

38. CH10.53
CSE
4100
Algebraic SimplificationAlgebraic Simplification
 Simple IdeaSimple Idea
 Use algebraic rules to rewrite some code
 ExamplesExamples
x := y + 0x := y + 0
x := yx := y
x := y *
1
x := y *
1
x := yx := y
x := y *
0
x := y *
0
x := 0x := 0

39. CH10.54
CSE
4100
Strength ReductionStrength Reduction
 IdeaIdea
 Replace expensive operation
 By semantically equivalent cheaper ones.
 ExamplesExamples
 Multiplication by 2 is equivalent to a left shift
 Left shift is much faster

40. CH10.55
CSE
4100
Hardware IdiomHardware Idiom
 IdeaIdea
 Replace expensive instructions by...
 Equivalent instruction that are optimized for the
platform
 ExampleExample
add eax,1add eax,1
inc
eax
inc
eax