Code Optimization
Prof. (Dr.) Malay S Bhatt
Department of Computer Engineering
Sankalchand Patel College of Engineering
Sankalchand Patel University
Topics to be covered :
1. Need of Language Processing
2. Analysis – Synthesis Model
3. Language Processing
4. code optimization examples
5. Control Flow Graph
6. Reaching Definitions
7. Register Allocation
An optimization is a transformation that is expected
to improve program in some way.
• Program running time
• Memory utilization
• Power consumption
Basic components of optimization:
• Code Analysis
• Code Rewriting
Optimizers :
• Multiple passes are typical, each does particular
optimization task
• Work on Intermediate representation
Evaluation of Code Optimization :
• Safety
• Opportunity
• Profitability
• Overhead
Scope of Code Optimization :
Code Optimization Examples:
Common Subexpression Elimination:
Common Subexpression Elimination:
t :=d + e
a := t b := t
c := t
Loop level Optimizations:
Loop level Optimizations:
DATA FLOW ANALYSIS
EXAMPLE CALCULATION (ITERATION -1)
EXAMPLE CALCULATION (ITERATION -1)
EXAMPLE CALCULATION (ITERATION -1)
BLOCK REACH_IN REACH_OUT
Entry {} {}
B1 {} {1,2,3}
B2 {1,2,3} {1,2,3}
B3 {1,2,3} {1,2,3,4}
B4 {1,2,3,4} {1,2,3,4}
B5 {1,2,3,4} {1,2,3,4}
B6 {1,2,3,4} {1,5,6,7,8}
EXIT {1,2,3,4} {1,2,3,4}
EXAMPLE CALCULATION (ITERATION -2)
BLOCK REACH_IN REACH_OUT
Entry {} {}
B1 {} {1,2,3}
B2 {1,2,3} {1,2,3}
B3 {1,2,3} {1,2,3,4}
B4 {1,2,3,4,5,6,7,8} {1,2,3,4,5,6,7,8}
B5 {1,2,3,4,5,6,7,8} {1,2,3,4,5,6,7,8}
B6 {1,2,3,4,5,6,7,8} {1,5,6,7,8}
EXIT {1,2,3,4,5,6,7,8} {1,2,3,4,5,6,7,8}
Register Allocation :Interference graph
Instructions Live vars
b=a+2
c=b*b
b=c+1
return b * a
Register Allocation :Interference graph
Instructions Live vars
b=a+2
c=b*b
b=c+1
b,a
return b * a
Register Allocation :Interference graph
Instructions Live vars
b=a+2
c=b*b
a,c
b=c+1
b,a
return b * a
Register Allocation :Interference graph
Instructions Live vars
b=a+2
b,a
c=b*b
a,c
b=c+1
b,a
return b * a
Register Allocation :Interference graph
Instructions Live vars
a
b=a+2
b,a
c=b*b
a,c
b=c+1
b,a
return b * a
Register Allocation :Interference graph
color register
Instructions Live vars eax
a
b=a+2 ebx
a,b
c=b*b
a,c
a
b=c+1
a,b
return b * a
b c
Register Allocation :Interference graph
color register
Instructions Live vars eax
a
b=a+2 ebx
a,b
c=b*b
a,c
a
b=c+1
a,b
return b * a
b c
Register Allocation :Interference graph
• Nodes of the graph = variables
• Edges connect variables that interfere with one another
• Nodes will be assigned a color corresponding to the register assigned
to the variable
• Two colors can’t be next to one another in the graph
Coloring a graph
• Kempe’s algorithm [1879] for finding a K-coloring of a graph
• Assume K=3
• Step 1 (simplify): find a node with at most K-1 edges and cut it out of
the graph. (Remember this node on a stack for later stages.)
Coloring a graph
• Once a coloring is found for the simpler graph, we can always color
the node we saved on the stack
• Step 2 (color): when the simplified subgraph has been colored, add
back the node on the top of the stack and assign it a color not taken
by one of the adjacent nodes
Coloring a graph
color register
eax
ebx
stack:
b c
d e
Coloring a graph
color register
eax
ebx
stack:
b c
c
d e
Coloring a graph
color register
eax
ebx
stack:
b c
e
c
d e
Coloring a graph
color register
eax
ebx
stack:
b c
a
e
c
d e
Coloring a graph
color register
eax
ebx
stack:
b
b c
a
e
c
d e
Coloring a graph
color register
eax
ebx
stack:
d
b c
b
a
e
c
d e
Coloring a graph
color register
eax
ebx
stack:
b c
b
a
e
c
d e
Coloring a graph
color register
eax
ebx
stack:
b c
a
e
c
d e
Coloring a graph
color register
eax
ebx
stack:
b c
e
c
d e
Coloring a graph
color register
eax
ebx
stack:
b c
c
d e
Coloring a graph
color register
eax
ebx
stack:
b c
d e
Final Thought:
