SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY COURSE MATERIAL

UNIT V - CODE OPTIMIZATION AND CODE GENERATION

INTRODUCTION

 The code produced by the straight forward compiling algorithms can often be made to run
faster or take less space, or both. This improvement is achieved by program transformations
that are traditionally called optimizations. Compilers that apply code-improving
transformations are called optimizing compilers.

 Optimizations are classified into two categories. They are

Machine independent optimizations:
Machine dependant optimizations:

Machine independent optimizations:

Machine independent optimizations are program transformations that improve the target code
without taking into consideration any properties of the target machine.

Machine dependant optimizations:

Machine dependant optimizations are based on register allocation and utilization of special
machine-instruction sequences.

The criteria for code improvement transformations:

 Simply stated, the best program transformations are those that yield the most benefit for the
least effort.

 The transformation must preserve the meaning of programs. That is, the optimization must
not change the output produced by a program for a given input, or cause an error such as
division by zero, that was not present in the original source program. At all times we take the
“safe” approach of missing an opportunity to apply a transformation rather than risk
changing what the program does.

 A transformation must, on the average, speed up programs by a measurable amount. We are

also interested in reducing the size of the compiled code although the size of the code has
less importance than it once had. Not every transformation succeeds in improving every
program, occasionally an “optimization” may slow down a program slightly.

 The transformation must be worth the effort. It does not make sense for a compiler writer to
expend the intellectual effort to implement a code improving transformation and to have the
compiler expend the additional time compiling source programs if this effort is not repaid
when the target programs are executed. “Peephole” transformations of this kind are simple
enough and beneficial enough to be included in any compiler.
SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY COURSE MATERIAL

Organization for an Optimizing Compiler:

 Flow analysis is a fundamental prerequisite for many important types of code

improvement.
Generally control flow analysis precedes data flow analysis.
Control flow analysis (CFA) represents flow of control usually in form of graphs, CFA
constructs such as
control flow graph
Call graph
Data flow analysis (DFA) is the process of ascerting and collecting information prior to
program execution about the possible modification, preservation, and use of certain
entities (such as values or attributes of variables) in a computer program.
PRINCIPAL SOURCES OF OPTIMISATION

A transformation of a program is called local if it can be performed by looking only at the

statements in a basic block; otherwise, it is called global.
Many transformations can be performed at both the local and global levels. Local
transformations are usually performed first.
Function-Preserving Transformations

There are a number of ways in which a compiler can improve a program without
changing the function it computes.
The transformations

 Common sub expression elimination,

 Copy propagation,
 Dead-code elimination, and
 Constant folding

are common examples of such function-preserving transformations. The other

transformations come up primarily when global optimizations are performed.
SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY COURSE MATERIAL

Frequently, a program will include several calculations of the same value, such as an
offset in an array. Some of the duplicate calculations cannot be avoided by the
programmer because they lie below the level of detail accessible within the source
language.

 Common Sub expressions elimination:

An occurrence of an expression E is called a common sub-expression if E was previously

computed, and the values of variables in E have not changed since the previous
computation. We can avoid recomputing the expression if we can use the previously
computed value.
For example
t1: = 4*i
t2: = a [t1]
t3: = 4*j
t4: = 4*i
t5 : = n
t6: = b [t4] +t5

The above code can be optimized using the common sub-expression elimination as
t1: = 4*i
t2: = a [t1]
t3: = 4*j
t5 : = n
t6: = b [t1] +t5

The common sub expression t4: =4*i is eliminated as its computation is already in t 1. And
value of i is not been changed from definition to use.

 Copy Propagation:

Assignments of the form f : = g called copy statements, or copies for short. The idea
behind the copy-propagation transformation is to use g for f, whenever possible after the
copy statement f: = g. Copy propagation means use of one variable instead of another.
This may not appear to be an improvement, but as we shall see it gives us an opportunity
to eliminate x.
For example:

x=Pi;
……
A=x*r*r;
The optimization using copy propagation can be done as follows:

A=Pi*r*r;

Here the variable x is eliminated

 Dead-Code Eliminations:

A variable is live at a point in a program if its value can be used subsequently; otherwise,
it is dead at that point. A related idea is dead or useless code, statements that compute
SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY COURSE MATERIAL

values that never get used. While the programmer is unlikely to introduce any dead code
intentionally, it may appear as the result of previous transformations. An optimization can
be done by eliminating dead code.
Example:

i=0;
if(i=1)
{
a=b+5;
}

Here, „if‟ statement is dead code because this condition will never get satisfied.

 Constant folding:

We can eliminate both the test and printing from the object code. More generally,
deducing at compile time that the value of an expression is a constant and using the
constant instead is known as constant folding.

One advantage of copy propagation is that it often turns the copy statement into dead
code.
 For example,
a=3.14157/2 can be replaced by
a=1.570 there by eliminating a division operation.

 Loop Optimizations:
We now give a brief introduction to a very important place for optimizations, namely
loops, especially the inner loops where programs tend to spend the bulk of their time. The
running time of a program may be improved if we decrease the number of instructions in
an inner loop, even if we increase the amount of code outside that loop.
Three techniques are important for loop optimization:

 code motion, which moves code outside a loop;

 Induction-variable elimination, which we apply to replace variables from inner loop.
 Reduction in strength, which replaces and expensive operation by a cheaper one, such as
a multiplication by an addition.

 Code Motion:
An important modification that decreases the amount of code in a loop is code motion.
This transformation takes an expression that yields the same result independent of the
number of times a loop is executed ( a loop-invariant computation) and places the
expression before the loop. Note that the notion “before the loop” assumes the existence
of an entry for the loop. For example, evaluation of limit-2 is a loop-invariant
computation in the following while-statement:

while (i <= limit-2) /* statement does not change limit*/

Code motion will result in the equivalent of

SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY COURSE MATERIAL

t= limit-2;
while (i<=t) /* statement does not change limit or t */

 Induction Variables :
Loops are usually processed inside out. For example consider the loop around B3.
Note that the values of j and t 4 remain in lock-step; every time the value of j decreases by
1, that of t4 decreases by 4 because 4*j is assigned to t 4. Such identifiers are called
induction variables.
When there are two or more induction variables in a loop, it may be possible to get rid of
all but one, by the process of induction-variable elimination. For the inner loop around
B3 in Fig. we cannot get rid of either j or t4 completely; t4 is used in B3 and j in B4.
However, we can illustrate reduction in strength and illustrate a part of the process of
induction-variable elimination. Eventually j will be eliminated when the outer loop of B2
- B5 is considered.
Example:
As the relationship t4:=4*j surely holds after such an assignment to t4 in Fig. and t4 is not
changed elsewhere in the inner loop around B3, it follows that just after the statement
j:=j-1 the relationship t4:= 4*j-4 must hold. We may therefore replace the assignment t 4:=
4*j by t4:= t4-4. The only problem is that t4 does not have a value when we enter block B3
for the first time. Since we must maintain the relationship t 4=4*j on entry to the block B3,
we place an initializations of t4 at the end of the block where j itself is

before after

initialized, shown by the dashed addition to block B1 in second Fig.

SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY COURSE MATERIAL

The replacement of a multiplication by a subtraction will speed up the object code if

multiplication takes more time than addition or subtraction, as is the case on many
machines.

 Reduction In Strength:

Reduction in strength replaces expensive operations by equivalent cheaper ones on the

target machine. Certain machine instructions are considerably cheaper than others and
can often be used as special cases of more expensive operators.
For example, x² is invariably cheaper to implement as x*x than as a call to an
exponentiation routine. Fixed-point multiplication or division by a power of two is
cheaper to implement as a shift. Floating-point division by a constant can be implemented
as multiplication by a constant, which may be cheaper.

OPTIMIZATION OF BASIC BLOCKS

There are two types of basic block optimizations. They are :

 Structure-Preserving Transformations
 Algebraic Transformations

Structure-Preserving Transformations:
The primary Structure-Preserving Transformation on basic blocks are:

 Common sub-expression elimination

 Dead code elimination
 Renaming of temporary variables
 Interchange of two independent adjacent statements.

 Common sub-expression elimination:

Common sub expressions need not be computed over and over again. Instead they can be
computed once and kept in store from where it‟s referenced when encountered again – of course
providing the variable values in the expression still remain constant.
DAG ( Directed Acyclic Graph ) is used to eliminate common subexpressions.
Example:
t1:=4*i after
a: =b+c t2:=a[t1] t1: = 4*i
t3:=4*j t2: = a [t1]
b: =a-d t4=4*i t3: = 4*j
c: =b+c t5:=n t4: = t1
d: =a-d t6:=b[t4]+n t5: = n
t6: = b [t4] +t5
The 2nd and 4th statements compute the same expression: b+c and a-d
Basic block can be transformed to
a: = b+c
b: = a-d
c: = a
d: = b
 int main()
{
SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY x=2 COURSE MATERIAL
if (x > 2)
cout << "code"; // Dead code
else
 Dead code elimination: cout << "Optimization";
return 0;
}
It‟s possible that a large amount of dead (useless) code may exist in the program. This
might be especially caused when introducing variables and procedures as part of constructio n or
error-correction of a program – once declared and defined, one forgets to remove them in case
they serve no purpose. Eliminating these will definitely optimize the code.
renaming temporary variables is a straightforward but important process
 Renaming of temporary variables: in compiler design to ensure clarity, prevent naming conflicts
A statement t:=b+c where t is a temporary name can be changed to u:=b+c where u is
another temporary name, and change all uses of t to u.
In this we can transform a basic block to its equivalent block called normal-form block.
^Transforming a basic block into its normal-form
 Interchange of two independent adjacent statements: block means converting it into a standardized or
simplified form where temporary variables are
Two statements consistently named and used

t1:=b+c
This swapping doesn't change the semantics of the code but might improve efficiency or other
t2:=x+y performance metrics, especially in scenarios like pipelined processors or parallel execution
environments.
can be interchanged or reordered in its computation in the basic block when value of t 1
does not affect the value of t 2.

Algebraic Transformations:
Algebraic identities represent another important class of optimizations on basic blocks.
This includes simplifying expressions or replacing expensive operation by cheaper ones
i.e. reduction in strength.
Another class of related optimizations is constant folding. Here we evaluate constant
expressions at compile time and replace the constant expressions by their values. Thus
the expression 2*3.14 would be replaced by 6.28.
The relational operators <=, >=, <, >, + and = sometimes generate unexpected common
sub expressions.
Associative laws may also be applied to expose common sub expressions. For example, if
the source code has the assignments Some of the algebraic transformation on basic blocks includes:

Constant Folding
a :=b+c Copy Propagation
Strength Reduction
e :=c+d+b
1. Constant Folding:
Solve the constant terms which are continuous so that compiler does
the following intermediate code may be generated: not need to solve this expression.
x = 2 * 3 + y x = 6 + y (Optimized code)
a :=b+c 2. Copy Propagation:
t :=c+d It is of two types, Variable Propagation, and Constant Propagation.
Variable Propagation:
e :=t+b x=y z = y + 2 (Optimized code)
z=x+2

Example: Constant Propagation:

x=3 z = 3 + a (Optimized code)
z=x+a\
x:=x+0 can be removed
3. Strength Reduction:
Replace expensive statement/ instruction with cheaper ones.
x:=y**2 can be replaced by a cheaper statement x:=y*y
x = 2 * y (costly) x = y + y (cheaper)
x = 2 * y (costly) x = y << 1 (cheaper)
SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY COURSE MATERIAL

The compiler writer should examine the language carefully to determine what
rearrangements of computations are permitted, since computer arithmetic does not always
obey the algebraic identities of mathematics. Thus, a compiler may evaluate x*y-x*z as
x*(y-z) but it may not evaluate a+(b-c) as (a+b)-c.

LOOPS IN FLOW GRAPH

A graph representation of three-address statements, called a flow graph, is useful for

understanding code-generation algorithms, even if the graph is not explicitly constructed by a
code-generation algorithm. Nodes in the flow graph represent computations, and the edges
represent the flow of control.

Dominators:
In a flow graph, a node d dominates node n, if every path from initial node of the flow
graph to n goes through d. This will be denoted by d dom n. Every initial node dominates all the
remaining nodes in the flow graph and the entry of a loop dominates all nodes in the loop.
Similarly every node dominates itself.

Example:

*In the flow graph below,

*Initial node,node1 dominates every node.
*node 2 dominates itself
*node 3 dominates all but 1 and 2.
*node 4 dominates all but 1,2 and 3.
*node 5 and 6 dominates only themselves,since flow of control can skip around either by goin
through the other.
*node 7 dominates 7,8 ,9 and 10.
*node 8 dominates 8,9 and 10.
*node 9 and 10 dominates only themselves.
SRI VIDYA COLLEGE OF ENGINEERING AND TECHNOLOGY COURSE MATERIAL

The way of presenting dominator information is in a tree, called the dominator tree in
which the initial node is the root.
The parent of each other node is its immediate dominator.
Each node d dominates only its descendents in the tree.
The existence of dominator tree follows from a property of dominators; each node has a
unique immediate dominator in that is the last dominator of n on any path from the initial
node to n.
In terms of the dom relation, the immediate dominator m has the property is d=!n and d
dom n, then d dom m.

D(1)={1} D(2)={1,2}

D(3)={1,3}

D(4)={1,3,4}

D(5)={1,3,4,5}

D(6)={1,3,4,6}

D(7)={1,3,4,7}

D(8)={1,3,4,7,8}

D(9)={1,3,4,7,8,9}

D(10)={1,3,4,7,8,10}
Natural Loop:

One application of dominator information is in determining the loops of a flow graph suitable
for improvement.

The properties of loops are

 A loop must have a single entry point, called the header. This entry point-dominates all
nodes in the loop, or it would not be the sole entry to the loop.
 There must be at least one way to iterate the loop(i.e.)at least one path back to the header.

One way to find all the loops in a flow graph is to search for edges in the flow graph whose
heads dominate their tails. If a→b is an edge, b is the head and a is the tail. These types of
edges are called as back edges.

 Example:

In the above graph,

7→4 4 DOM 7
10 →7 7 DOM 10
4→3
8→3
9 →1

The above edges will form loop in flow graph.

Given a back edge n → d, we define the natural loop of the edge to be d plus the set of nodes
that can reach n without going through d. Node d is the header of the loop.

Algorithm: Constructing the natural loop of a back edge.

Input: A flow graph G and a back edge n→d.

Output: The set loop consisting of all nodes in the natural loop n→d.

Method: Beginning with node n, we consider each node m*d that we know is in loop, to make
sure that m‟s predecessors are also placed in loop. Each node in loop, except for d, is placed once
on stack, so its predecessors will be examined. Note that because d is put in the loop initially, we
never examine its predecessors, and thus find only those nodes that reach n without going
through d.

Procedure insert(m);
if m is not in loop then begin
loop := loop U {m};
push m onto stack
end;

stack : = empty;
loop : = {d};
insert(n);
while stack is not empty do begin
pop m, the first element of stack, off stack;
for each predecessor p of m do insert(p)
end

Inner loop:

If we use the natural loops as “the loops”, then we have the useful property that unless
two loops have the same header, they are either disjointed or one is entirely contained in
the other. Thus, neglecting loops with the same header for the moment, we have a natural
notion of inner loop: one that contains no other loop.
When two natural loops have the same header, but neither is nested within the other, they
are combined and treated as a single loop.

Pre-Headers:
Several transformations require us to move statements “before the header”. Therefore
begin treatment of a loop L by creating a new block, called the preheater.

The pre-header has only the header as successor, and all edges which formerly entered
the header of L from outside L instead enter the pre-header.

Edges from inside loop L to the header are not changed.

Initially the pre-header is empty, but transformations on L may place statements in it.

header pre-header

loop L

header
loop L

(a) Before (b) After

Reducible flow graphs:

Reducible flow graphs are special flow graphs, for which several code optimization
transformations are especially easy to perform, loops are unambiguously defined,
dominators can be easily calculated, data flow analysis problems can also be solved
efficiently.

Exclusive use of structured flow-of-control statements such as if-then-else, while-do,

continue, and break statements produces programs whose flow graphs are always
reducible.
 The most important properties of reducible flow graphs are that there are no jumps into
the middle of loops from outside; the only entry to a loop is through its header.

Definition:
A flow graph G is reducible if and only if we can partition the edges into two disjoint
groups, forward edges and back edges, with the following properties.

 The forward edges from an acyclic graph in which every node can be reached from initial
node of G.

 The back edges consist only of edges where heads dominate theirs tails.

 Example: The above flow graph is reducible.

If we know the relation DOM for a flow graph, we can find and remove all the back
edges.

The remaining edges are forward edges.

If the forward edges form an acyclic graph, then we can say the flow graph reducible.

In the above example remove the five back edges 4→3, 7→4, 8→3, 9→1 and 10→7
whose heads dominate their tails, the remaining graph is acyclic.

The key property of reducible flow graphs for loop analysis is that in such flow graphs
every set of nodes that we would informally regard as a loop must contain a back edge.

PEEPHOLE OPTIMIZATION

A statement-by-statement code-generations strategy often produce target code that

contains redundant instructions and suboptimal constructs .The quality of such target
code can be improved by applying “optimizing” transformations to the target program.
A simple but effective technique for improving the target code is peephole optimization,
a method for trying to improving the performance of the target program by examining a
short sequence of target instructions (called the peephole) and replacing these
instructions by a shorter or faster sequence, whenever possible.
The peephole is a small, moving window on the target program. The code in the peephole
need not contiguous, although some implementations do require this.it is characteristic of
peephole optimization that each improvement may spawn opportunities for additional
improvements.
We shall give the following examples of program transformations that are characteristic
of peephole optimizations:

 Redundant-instructions elimination
 Flow-of-control optimizations
 Algebraic simplifications
 Use of machine idioms
 Unreachable Code
Redundant Loads And Stores:

If we see the instructions sequence

(1) MOV R0,a

(2) MOV a,R0

we can delete instructions (2) because whenever (2) is executed. (1) will ensure that the value of
a is already in register R0.If (2) had a label we could not be sure that (1) was always executed
immediately before (2) and so we could not remove (2).

Unreachable Code:
Another opportunity for peephole optimizations is the removal of unreachable instructions.
An unlabeled instruction immediately following an unconditional jump may be removed.
This operation can be repeated to eliminate a sequence of instructions. For ex ample, for
debugging purposes, a large program may have within it certain segments that are executed
only if a variable debug is 1. In C, the source code might look like:

#define debug 0

….

If ( debug ) {

Print debugging information

In the intermediate representations the if-statement may be translated as:

If debug =1 goto L2

goto L2

L1: print debugging information

L2: …………………………(a)

One obvious peephole optimization is to eliminate jumps over jumps .Thus no matter what
the value of debug; (a) can be replaced by:

If debug ≠1 goto L2

Print debugging information

L2: ……………………………(b)

As the argument of the statement of (b) evaluates to a constant true it can be replaced
by
 If debug ≠0 goto L2

Print debugging information

L2: ……………………………(c)

As the argument of the first statement of (c) evaluates to a constant true, it can be replaced by
goto L2. Then all the statement that print debugging aids are manifestly unreachable and
can be eliminated one at a time.

Flows-Of-Control Optimizations:

The unnecessary jumps can be eliminated in either the intermediate code or th e target code
by the following types of peephole optimizations. We can replace the jump sequence

goto L1

….

L1: gotoL2

by the sequence

goto L2

….

L1: goto L2

If there are now no jumps to L1, then it may be possible to eliminate the statement L1:goto
L2 provided it is preceded by an unconditional jump .Similarly, the sequence

if a < b goto L1

….

L1: goto L2

can be replaced by

If a < b goto L2

….

L1: goto L2

Finally, suppose there is only one jump to L1 and L1 is preceded by an unconditional goto.
Then the sequence

goto L1

……..
 L1: if a < b goto L2

L3: …………………………………..(1)

May be replaced by

If a < b goto L2

goto L3

…….

L3: ………………………………….(2)

While the number of instructions in (1) and (2) is the same, we sometimes skip the
unconditional jump in (2), but never in (1).Thus (2) is superior to (1) in execution time

Algebraic Simplification:

There is no end to the amount of algebraic simplification that can be attempted through
peephole optimization. Only a few algebraic identities occur frequently enough that it is
worth considering implementing them .For example, statements such as

x := x+0

Or

x := x * 1

Are often produced by straightforward intermediate code-generation algorithms, and they can
be eliminated easily through peephole optimization.

Reduction in Strength:
Reduction in strength replaces expensive operations by equivalent cheaper ones on the target
machine. Certain machine instructions are considerably cheaper than others and can often be
used as special cases of more expensive operators.
For example, x² is invariably cheaper to implement as x*x than as a call to an exponentiation
routine. Fixed-point multiplication or division by a power of two is cheaper to implement as
a shift. Floating-point division by a constant can be implemented as multiplication by a
constant, which may be cheaper.

X2 → X*X

Use of Machine Idioms:

The target machine may have hardware instructions to implement certain specific operations
efficiently. For example, some machines have auto-increment and auto-decrement addressing
modes. These add or subtract one from an operand before or after using its value.
The use of these modes greatly improves the quality of code when pushing or popping a
stack, as in parameter passing. These modes can also be used in code for statements like i :
=i+1.
 i:=i+1 → i++
i:=i-1 → i- -

INTRODUCTION TO GLOBAL DATAFLOW ANALYSIS

In order to do code optimization and a good job of code generation , compiler needs to
collect information about the program as a whole and to distribute this information to
each block in the flow graph.

A compiler could take advantage of “reaching definitions” , such as knowing where a

variable like debug was last defined before reaching a given block, in order to perform
transformations are just a few examples of data-flow information that an optimizing
compiler collects by a process known as data-flow analysis.

Data-flow information can be collected by setting up and solving systems of equations of

the form :

out [S] = gen [S] U ( in [S] – kill [S] )

This equation can be read as “ the information at the end of a statement is either generated
within the statement , or enters at the beginning and is not killed as control flows through
the statement.”

The details of how data-flow equations are set and solved depend on three factors.

 The notions of generating and killing depend on the desired information, i.e., on the data
flow analysis problem to be solved. Moreover, for some problems, instead of proceeding
along with flow of control and defining out[s] in terms of in[s], we need to proceed
backwards and define in[s] in terms of out[s].

 Since data flows along control paths, data-flow analysis is affected by the constructs in a
program. In fact, when we write out[s] we implicitly assume that there is unique end
point where control leaves the statement; in general, equations are set up at the level of
basic blocks rather than statements, because blocks do have unique end points.

 There are subtleties that go along with such statements as procedure calls, assignments
through pointer variables, and even assignments to array variables.

Points and Paths:

Within a basic block, we talk of the point between two adjacent statements, as well as the
point before the first statement and after the last. Thus, block B1 has four points: one
before any of the assignments and one after each of the three assignments.
 B1

d1 : i :=m-1
d2: j :=n
d3: a := u1
B2
d4 : I := i+1
B3
d5: j := j-1

B4

B5 B6

d6 :a :=u2

Now let us take a global view and consider all the points in all the blocks. A path from p 1
to pn is a sequence of points p1, p2,….,pn such that for each i between 1 and n-1, either

 Pi is the point immediately preceding a statement and p i+1 is the point immediately
following that statement in the same block, or

 Pi is the end of some block and pi+1 is the beginning of a successor block.

Reaching definitions:
A definition of variable x is a statement that assigns, or may assign, a value to x. The
most common forms of definition are assignments to x and statements that read a value
from an i/o device and store it in x.

These statements certainly define a value for x, and they are referred to as unambiguous
definitions of x. There are certain kinds of statements that may define a value for x; they
are called ambiguous definitions. The most usual forms of ambiguous definitions of x
are:

 A call of a procedure with x as a parameter or a procedure that can access x because x is

in the scope of the procedure.

 An assignment through a pointer that could refer to x. For example, the assignment * q: = y
is a definition of x if it is possible that q points to x. we must assume that an assignment
through a pointer is a definition of every variable.

We say a definition d reaches a point p if there is a path from the point immediately
following d to p, such that d is not “killed” along that path. Thus a point can be reached
 by an unambiguous definition and an ambiguous definition of the same variable
appearing later along one path.

Data-flow analysis of structured programs:

Flow graphs for control flow constructs such as do-while statements have a useful
property: there is a single beginning point at which control enters and a single end point
that control leaves from when execution of the statement is over. We exploit this property
when we talk of the definitions reaching the beginning and the end of statements wit h the
following syntax.

S id: = E| S; S | if E then S else S | do S while E

E id + id| id

Expressions in this language are similar to those in the intermediate code, but the flow
graphs for statements have restricted forms.

S1
S1
If E goto s1

S2
S1 S2 If E goto s1

S1 ; S2

IF E then S1 else S2 do S1 while E

We define a portion of a flow graph called a region to be a set of nodes N that includes a
header, which dominates all other nodes in the region. All edges between nodes in N are
in the region, except for some that enter the header.
The portion of flow graph corresponding to a statement S is a region that obeys the
further restriction that control can flow to just one outside block when it leaves the
region.
 We say that the beginning points of the dummy blocks at the entry and exit of a
statement‟s region are the beginning and end points, respectively, of the statement. The
equations are inductive, or syntax-directed, definition of the sets in[S], out[S], gen[S],
and kill[S] for all statements S.
gen[S] is the set of definitions “generated” by S while kill[S] is the set of definitions
that never reach the end of S.
Consider the following data-flow equations for reaching definitions :

i)

S d: a:=b+c

gen [S] = { d }
kill [S] = Da – { d }
out [S] = gen [S] U ( in[S] – kill[S] )

Observe the rules for a single assignment of variable a. Surely that assignment is a
definition of a, say d. Thus
Gen[S]={d}
On the other hand, d “kills” all other definitions of a, so we write
Kill[S] = Da – {d}
Where, Da is the set of all definitions in the program for variable a. ii )

S S1

S2

gen[S]=gen[S2] U (gen[S1]-kill[S2])
Kill[S] = kill[S2] U (kill[S1] – gen[S2])

in [S1] = in [S]
in [S2] = out [S1]
out [S] = out [S2]
 Under what circumstances is definition d generated by S=S 1; S2? First of all, if it is
generated by S2, then it is surely generated by S. if d is generated by S1, it will reach the
end of S provided it is not killed by S 2. Thus, we write
gen[S]=gen[S2] U (gen[S1]-kill[S2]
Similar reasoning applies to the killing of a definition, so we have
Kill[S] = kill[S2] U (kill[S1] – gen[S2])

Conservative estimation of data-flow information:

There is a subtle miscalculation in the rules for gen and kill. We have made the
assumption that the conditional expression E in the if and do statements are
“uninterpreted”; that is, there exists inputs to the program that make their branches go
either way.

We assume that any graph-theoretic path in the flow graph is also an execution path, i.e.,
a path that is executed when the program is run with least one possible input.

When we compare the computed gen with the “true” gen we discover that the true gen is
always a subset of the computed gen. on the other hand, the true kill is always a superset
of the computed kill.

These containments hold even after we consider the other rules. It is natural to wonder
whether these differences between the true and computed gen and kill sets present a
serious obstacle to data-flow analysis. The answer lies in the use intended for these data.

Overestimating the set of definitions reaching a point does not seem serious; it merely
stops us from doing an optimization that we could legitimately do. On the other hand,
underestimating the set of definitions is a fatal error; it could lead us into making a
change in the program that changes what the program computes. For the case of reaching
definitions, then, we call a set of definitions safe or conservative if the estimate is a
superset of the true set of reaching definitions. We call the estimate unsafe, if it is not
necessarily a superset of the truth.

Returning now to the implications of safety on the estimation of gen and kill for reaching
definitions, note that our discrepancies, supersets for gen and subsets for kill are both in
the safe direction. Intuitively, increasing gen adds to the set of definitions that can reach a
point, and cannot prevent a definition from reaching a place that it truly reached.
Decreasing kill can only increase the set of definitions reaching any given point.

Computation of in and out:

Many data-flow problems can be solved by synthesized translations similar to those used
to compute gen and kill. It can be used, for example, to determine loop-invariant
computations.

However, there are other kinds of data-flow information, such as the reaching-definitions
problem. It turns out that in is an inherited attribute, and out is a synthesized attribute
depending on in. we intend that in[S] be the set of definitions reaching the beginning of
 S, taking into account the flow of control throughout the entire program, including
statements outside of S or within which S is nested.

The set out[S] is defined similarly for the end of s. it is important to note the distinction
between out[S] and gen[S]. The latter is the set of definitions that reach the end of S
without following paths outside S.

Assuming we know in[S] we compute out by equation, that is

Out[S] = gen[S] U (in[S] - kill[S])

Considering cascade of two statements S 1; S2, as in the second case. We start by

observing in[S1]=in[S]. Then, we recursively compute out[S 1], which gives us in[S2],
since a definition reaches the beginning of S 2 if and only if it reaches the end of S 1. Now
we can compute out[S2], and this set is equal to out[S].

Considering if-statement we have conservatively assumed that control can follow either
branch, a definition reaches the beginning of S1 or S2 exactly when it reaches the
beginning of S.

In[S1] = in[S2] = in[S]

If a definition reaches the end of S if and only if it reaches the end of one or both sub
statements; i.e,

Out[S]=out[S1] U out[S2]

Representation of sets:

Sets of definitions, such as gen[S] and kill[S], can be represented compactly using bit
vectors. We assign a number to each definition of interest in the flow graph. Then bit
vector representing a set of definitions will have 1 in position I if and only if the
definition numbered I is in the set.

The number of definition statement can be taken as the index of statement in an array
holding pointers to statements. However, not all definitions may be of interest during
global data-flow analysis. Therefore the number of definitions of interest will typically be
recorded in a separate table.

A bit vector representation for sets also allows set operations to be implemented
efficiently. The union and intersection of two sets can be implemented by logical or and
logical and, respectively, basic operations in most systems-oriented programming
languages. The difference A-B of sets A and B can be implemented by taking the
complement of B and then using logical and to compute A

.
Local reaching definitions:

Space for data-flow information can be traded for time, by saving information only at
certain points and, as needed, recomputing information at intervening points. Basic
blocks are usually treated as a unit during global flow analysis, with attention restricted to
only those points that are the beginnings of blocks.

Since there are usually many more points than blocks, restricting our effort to blocks is a
significant savings. When needed, the reaching definitions for all points in a block can be
calculated from the reaching definitions for the beginning of a block.

Use-definition chains:

It is often convenient to store the reaching definition information as” use-definition

chains” or “ud-chains”, which are lists, for each use of a variable, of all the definitions
that reaches that use. If a use of variable a in block B is preceded by no unambiguous
definition of a, then ud-chain for that use of a is the set of definitions in in[B] that are
definitions of a.in addition, if there are ambiguous definitions of a ,then all of these for
which no unambiguous definition of a lies between it and the use of a are on the ud -chain
for this use of a.

Evaluation order:

The techniques for conserving space during attribute evaluation, also apply to the
computation of data-flow information using specifications. Specifically, the only
constraint on the evaluation order for the gen, kill, in and out sets for statements is that
imposed by dependencies between these sets. Having chosen an evaluation order, we are
free to release the space for a set after all uses of it have occurred.

Earlier circular dependencies between attributes were not allowed, but we have seen that
data-flow equations may have circular dependencies.

General control flow:

Data-flow analysis must take all control paths into account. If the control paths are
evident from the syntax, then data-flow equations can be set up and solved in a syntax-
directed manner.

When programs can contain goto statements or even the more disciplined break and
continue statements, the approach we have taken must be modified to take the actual
control paths into account.

Several approaches may be taken. The iterative method works arbitrary flow graphs.
Since the flow graphs obtained in the presence of break and continue statements are
reducible, such constraints can be handled systematically using the interval -based
methods
 However, the syntax-directed approach need not be abandoned when break and continue
statements are allowed.
CODE GENERATION
The final phase in compiler model is the code generator. It takes as input an intermediate
representation of the source program and produces as output an equivalent target program. The
code generation techniques presented below can be used whether or not an optimizing phase
occurs before code generation.

Position of code generator

source front end intermediate code intermediate code target

program code optimizer code generator program

symbol
table

ISSUES IN THE DESIGN OF A CODE GENERATOR

The following issues arise during the code generation phase :

1. Input to code generator

2. Target program
3. Memory management
4. Instruction selection
5. Register allocation
6. Evaluation order

1. Input to code generator:

The input to the code generation consists of the intermediate representation of the source
program produced by front end , together with information in the symbol table to
determine run-time addresses of the data objects denoted by the names in the
intermediate representation.
Intermediate representation can be :
a. Linear representation such as postfix notation
b. Three address representation such as quadruples
c. Virtual machine representation such as stack machine code
d. Graphical representations such as syntax trees and dags.
Prior to code generation, the front end must be scanned, parsed and translated into
intermediate representation along with necessary type checking. Therefore, input to code
generation is assumed to be error-free.
2. Target program:
The output of the code generator is the target program. The output may be :
a. Absolute machine language
- It can be placed in a fixed memory location and can be executed immediately.

CS6660 COMPILER DESIGN UNIT V

 b. Relocatable machine language
- It allows subprograms to be compiled separately.

c. Assembly language
- Code generation is made easier.

3. Memory management:
Names in the source program are mapped to addresses of data objects in run-time
memory by the front end and code generator.
It makes use of symbol table, that is, a name in a three-address statement refers to a
symbol-table entry for the name.
Labels in three-address statements have to be converted to addresses of instructions.
For example,
j : goto i generates jump instruction as follows :
 if i < j, a backward jump instruction with target address equal to location of
code for quadruple i is generated.
 if i > j, the jump is forward. We must store on a list for quadruple i the
location of the first machine instruction generated for quadruple j. When i is
processed, the machine locations for all instructions that forward jumps to i
are filled.

4. Instruction selection:
The instructions of target machine should be complete and uniform.
Instruction speeds and machine idioms are important factors when efficiency of target
program is considered.
The quality of the generated code is determined by its speed and size.
The former statement can be translated into the latter statement as shown below:

5. Register allocation
Instructions involving register operands are shorter and faster than those involving
operands in memory.
The use of registers is subdivided into two subproblems :
 Register allocation – the set of variables that will reside in registers at a point in
the program is selected.

CS6660 COMPILER DESIGN UNIT V

  Register assignment – the specific register that a variable will reside in is
picked
Certain machine requires even-odd register pairs for some operands and results.
For example , consider the division instruction of the form :
D x, y

where, x – dividend even register in even/odd register pair

y – divisor
even register holds the remainder
odd register holds the quotient

6. Evaluation order
The order in which the computations are performed can affect the efficiency of the
target code. Some computation orders require fewer registers to hold intermediate
results than others.

TARGET MACHINE
Familiarity with the target machine and its instruction set is a prerequisite for designing a
good code generator.
The target computer is a byte-addressable machine with 4 bytes to a word.
It has n general-purpose registers, R0, R1, . . . , Rn-1.
It has two-address instructions of the form:
op source, destination
where, op is an op-code, and source and destination are data fields.

It has the following op-codes :

MOV (move source to destination)
ADD (add source to destination)
SUB (subtract source from destination)

The source and destination of an instruction are specified by combining registers and
memory locations with address modes.
Address modes with their assembly-language forms

MODE FORM ADDRESS ADDED COST

absolute M M 1

register R R 0

indexed c(R) c+contents(R) 1

indirect register *R contents (R) 0

indirect indexed *c(R) contents(c+ 1

contents(R))

literal #c c 1

CS6660 COMPILER DESIGN UNIT V

 For example : MOV R0, M stores contents of Register R0 into memory location M ;
MOV 4(R0), M stores the value contents(4+contents(R0)) into M.

Instruction costs :
Instruction cost = 1+cost for source and destination address modes. This cost corresponds
to the length of the instruction.
Address modes involving registers have cost zero.
Address modes involving memory location or literal have cost one.
Instruction length should be minimized if space is important. Doing so also minimizes the
time taken to fetch and perform the instruction.
For example : MOV R0, R1 copies the contents of register R0 into R1. It has cost one,
since it occupies only one word of memory.
The three-address statement a : = b + c can be implemented by many different instruction
sequences :

i) MOV b, R0
ADD c, R0 cost = 6
MOV R0, a

ii) MOV b, a
ADD c, a cost = 6

iii) Assuming R0, R1 and R2 contain the addresses of a, b, and c :

MOV *R1, *R0
ADD *R2, *R0 cost = 2

In order to generate good code for target machine, we must utilize its addressing
capabilities efficiently.

RUN-TIME STORAGE MANAGEMENT

Information needed during an execution of a procedure is kept in a block of storage
called an activation record, which includes storage for names local to the procedure.
The two standard storage allocation strategies are:
1. Static allocation
2. Stack allocation
In static allocation, the position of an activation record in memory is fixed at compile
time.
In stack allocation, a new activation record is pushed onto the stack for each execution of
a procedure. The record is popped when the activation ends.
The following three-address statements are associated with the run-time allocation and
deallocation of activation records:
1. Call,
2. Return,
3. Halt, and
4. Action, a placeholder for other statements.
We assume that the run-time memory is divided into areas for:
1. Code
2. Static data
3. Stack

CS6660 COMPILER DESIGN UNIT V

Static allocation

Implementation of call statement:

The codes needed to implement static allocation are as follows:

MOV #here + 20, callee.static_area /*It saves return address*/

GOTO callee.code_area /*It transfers control to the target code for the called procedure */

where,
callee.static_area – Address of the activation record
callee.code_area – Address of the first instruction for called procedure
#here + 20 – Literal return address which is the address of the instruction following GOTO.

Implementation of return statement:

A return from procedure callee is implemented by :
GOTO *callee.static_area

This transfers control to the address saved at the beginning of the activation record.

Implementation of action statement:

The instruction ACTION is used to implement action statement.

Implementation of halt statement:

The statement HALT is the final instruction that returns control to the operating system.

Stack allocation

Static allocation can become stack allocation by using relative addresses for storage in
activation records. In stack allocation, the position of activation record is stored in register so
words in activation records can be accessed as offsets from the value in this register.

The codes needed to implement stack allocation are as follows:

Initialization of stack:

MOV #stackstart , SP /* initializes stack */

Code for the first procedure

HALT /* terminate execution */

Implementation of Call statement:

ADD #caller.recordsize, SP /* increment stack pointer */

MOV #here + 16, *SP /*Save return address */

GOTO callee.code_area

CS6660 COMPILER DESIGN UNIT V

where,
caller.recordsize – size of the activation record
#here + 16 – address of the instruction following the GOTO

Implementation of Return statement:

GOTO *0 ( SP ) /*return to the caller */

SUB #caller.recordsize, SP /* decrement SP and restore to previous value */

BASIC BLOCKS AND FLOW GRAPHS

Basic Blocks

A basic block is a sequence of consecutive statements in which flow of control enters at

the beginning and leaves at the end without any halt or possibility of branching except at
the end.
The following sequence of three-address statements forms a basic block:
t1 : = a * a
t2 : = a * b
t3 : = 2 * t2
t4 : = t1 + t3
t5 : = b * b
t6 : = t4 + t5

Basic Block Construction:

Algorithm: Partition into basic blocks

Input: A sequence of three-address statements

Output: A list of basic blocks with each three-address statement in exactly one block

Method:

1. We first determine the set of leaders, the first statements of basic blocks. The rules
we use are of the following:
a. The first statement is a leader.
b. Any statement that is the target of a conditional or unconditional goto is a
leader.
c. Any statement that immediately follows a goto or conditional goto statement
is a leader.
2. For each leader, its basic block consists of the leader and all statements up to but not
including the next leader or the end of the program.

CS6660 COMPILER DESIGN UNIT V

 Consider the following source code for dot product of two vectors a and b of length 20

begin

prod :=0;

i:=1;

do begin

prod :=prod+ a[i] * b[i];

i :=i+1;

end

while i <= 20

end

The three-address code for the above source program is given as :

(1) prod := 0

(2) i := 1

(3) t1 := 4* i

(4) t2 := a[t1] /*compute a[i] */

(5) t3 := 4* i

(6) t4 := b[t3] /*compute b[i] */

(7) t5 := t2*t4

(8) t6 := prod+t5

(9) prod := t6

(10) t7 := i+1

(11) i := t7

(12) if i<=20 goto (3)

Basic block 1: Statement (1) to (2)

Basic block 2: Statement (3) to (12)

CS6660 COMPILER DESIGN UNIT V

Transformations on Basic Blocks:

A number of transformations can be applied to a basic block without changing the set of
expressions computed by the block. Two important classes of transformation are :

Structure-preserving transformations

Algebraic transformations

1. Structure preserving transformations:

a) Common subexpression elimination:

a:=b+c a:=b+c
b:=a–d b:=a-d
c:=b+c c:=b+c
d:=a–d d:=b

Since the second and fourth expressions compute the same expression, the basic block can be
transformed as above.

b) Dead-code elimination:

Suppose x is dead, that is, never subsequently used, at the point where the statement x : =
y + z appears in a basic block. Then this statement may be safely removed without changing
the value of the basic block.

c) Renaming temporary variables:

A statement t : = b + c ( t is a temporary ) can be changed to u : = b + c (u is a new

temporary) and all uses of this instance of t can be changed to u without changing the value of
the basic block.
Such a block is called a normal-form block.

d) Interchange of statements:

Suppose a block has the following two adjacent statements:

t1 : = b + c
t2 : = x + y

We can interchange the two statements without affecting the value of the block if and
only if neither x nor y is t1 and neither b nor c is t2.

2. Algebraic transformations:

Algebraic transformations can be used to change the set of expressions computed by a basic
block into an algebraically equivalent set.
Examples:
i) x : = x + 0 or x : = x * 1 can be eliminated from a basic block without changing the set of
expressions it computes.
ii) The exponential statement x : = y * * 2 can be replaced by x : = y * y.

CS6660 COMPILER DESIGN UNIT V

Flow Graphs

Flow graph is a directed graph containing the flow-of-control information for the set of
basic blocks making up a program.
The nodes of the flow graph are basic blocks. It has a distinguished initial node.
E.g.: Flow graph for the vector dot product is given as follows:

prod : = 0 B1
i:=1

t1 : = 4 * i
t2 : = a [ t1 ]
t3 : = 4 * i
B2
t4 : = b [ t3 ]
t5 : = t2 * t4
t6 : = prod + t5
prod : = t6
t7 : = i + 1
i : = t7
if i <= 20 goto B2

B1 is the initial node. B2 immediately follows B1, so there is an edge from B1 to B2. The
target of jump from last statement of B1 is the first statement B2, so there is an edge from
B1 (last statement) to B2 (first statement).
B1 is the predecessor of B2, and B2 is a successor of B1.

Loops

A loop is a collection of nodes in a flow graph such that

1. All nodes in the collection are strongly connected.
2. The collection of nodes has a unique entry.
A loop that contains no other loops is called an inner loop.

NEXT-USE INFORMATION

If the name in a register is no longer needed, then we remove the name from the register
and the register can be used to store some other names.

CS6660 COMPILER DESIGN UNIT V

 Input: Basic block B of three-address statements

Output: At each statement i: x= y op z, we attach to i the liveliness and next-uses of x,

y and z.

Method: We start at the last statement of B and scan backwards.

1. Attach to statement i the information currently found in the symbol table

regarding the next-use and liveliness of x, y and z.
2. In the symbol table, set x to “not live” and “no next use”.
3. In the symbol table, set y and z to “live”, and next-uses of y and z to i.

Symbol Table:

Names Liveliness Next-use

x not live no next-use

y Live i

z Live i

A SIMPLE CODE GENERATOR

A code generator generates target code for a sequence of three- address statements and
effectively uses registers to store operands of the statements.

For example: consider the three-address statement a := b+c

It can have the following sequence of codes:

ADD Rj, Ri Cost = 1 // if Ri contains b and Rj contains c

(or)

ADD c, Ri Cost = 2 // if c is in a memory location

(or)

MOV c, Rj Cost = 3 // move c from memory to Rj and add

ADD Rj, Ri

Register and Address Descriptors:

A register descriptor is used to keep track of what is currently in each registers. The
register descriptors show that initially all the registers are empty.
An address descriptor stores the location where the current value of the name can be
found at run time.

CS6660 COMPILER DESIGN UNIT V

 A code-generation algorithm:

The algorithm takes as input a sequence of three-address statements constituting a basic block.
For each three-address statement of the form x : = y op z, perform the following actions:

1. Invoke a function getreg to determine the location L where the result of the computation y op
z should be stored.

2. Consult the address descriptor for y to determine y‟, the current location of y. Prefer the
register for y‟ if the value of y is currently both in memory and a register. If the value of y is
not already in L, generate the instruction MOV y’ , L to place a copy of y in L.

3. Generate the instruction OP z’ , L where z‟ is a current location of z. Prefer a register to a

memory location if z is in both. Update the address descriptor of x to indicate that x is in
location L. If x is in L, update its descriptor and remove x from all other descriptors.

4. If the current values of y or z have no next uses, are not live on exit from the block, and are in
registers, alter the register descriptor to indicate that, after execution of x : = y op z , those
registers will no longer contain y or z.

Generating Code for Assignment Statements:

The assignment d : = (a-b) + (a-c) + (a-c) might be translated into the following three-
address code sequence:
t:=a–b
u:=a–c
v:=t+u
d:=v+u
with d live at the end.

Code sequence for the example is:

Statements Code Generated Register descriptor Address descriptor

Register empty

t:=a-b MOV a, R0 R0 contains t t in R0

SUB b, R0

u:=a-c MOV a , R1 R0 contains t t in R0

SUB c , R1 R1 contains u u in R1

v:=t+u ADD R1, R0 R0 contains v u in R1

R1 contains u v in R0

d:=v+u ADD R1, R0 R0 contains d d in R0

d in R0 and memory
MOV R0, d

CS6660 COMPILER DESIGN UNIT V

 Generating Code for Indexed Assignments

The table shows the code sequences generated for the indexed assignment statements
a : = b [ i ] and a [ i ] : = b

Statements Code Generated Cost

a : = b[i] MOV b(Ri), R 2

a[i] : = b MOV b, a(Ri) 3

Generating Code for Pointer Assignments

The table shows the code sequences generated for the pointer assignments
a : = *p and *p : = a

Statements Code Generated Cost

a : = *p MOV *Rp, a 2

*p : = a MOV a, *Rp 2

Generating Code for Conditional Statements

Statement Code

if x < y goto z CMP x, y

CJ< z /* jump to z if condition code
is negative */

x : = y +z MOV y, R0
if x < 0 goto z ADD z, R0
MOV R0,x
CJ< z

THE DAG REPRESENTATION FOR BASIC BLOCKS

A DAG for a basic block is a directed acyclic graph with the following labels on nodes:
1. Leaves are labeled by unique identifiers, either variable names or constants.
2. Interior nodes are labeled by an operator symbol.
3. Nodes are also optionally given a sequence of identifiers for labels to store the
computed values.
DAGs are useful data structures for implementing transformations on basic blocks.
It gives a picture of how the value computed by a statement is used in subsequent
statements.
It provides a good way of determining common sub - expressions.
CS6660 COMPILER DESIGN UNIT V
Algorithm for construction of DAG

Input: A basic block

Output: A DAG for the basic block containing the following information:

1. A label for each node. For leaves, the label is an identifier. For interior nodes, an
operator symbol.
2. For each node a list of attached identifiers to hold the computed values.
Case (i) x : = y OP z

Case (ii) x : = OP y

Case (iii) x : = y

Method:

Step 1: If y is undefined then create node(y).

If z is undefined, create node(z) for case(i).

Step 2: For the case(i), create a node(OP) whose left child is node(y) and right child is

node(z). ( Checking for common sub expression). Let n be this node.

For case(ii), determine whether there is node(OP) with one child node(y). If not create such
a node.

For case(iii), node n will be node(y).

Step 3: Delete x from the list of identifiers for node(x). Append x to the list of attached

identifiers for the node n found in step 2 and set node(x) to n.

Example: Consider the block of three- address statements:

1. t1 := 4* i
2. t2 := a[t1]
3. t3 := 4* i
4. t4 := b[t3]
5. t5 := t2*t4
6. t6 := prod+t5
7. prod := t6
8. t7 := i+1
9. i := t7
10. if i<=20 goto (1)

CS6660 COMPILER DESIGN UNIT V

Stages in DAG Construction
Application of DAGs:

1. We can automatically detect common sub expressions.

2. We can determine which identifiers have their values used in the block.
3. We can determine which statements compute values that could be used outside the block.
GENERATING CODE FROM DAGs

The advantage of generating code for a basic block from its dag representation is that,
from a dag we can easily see how to rearrange the order of the final computation sequence than
we can starting from a linear sequence of three-address statements or quadruples.

Rearranging the order

The order in which computations are done can affect the cost of resulting object code.

For example, consider the following basic block:

t1 : = a + b
t2 : = c + d
t3 : = e – t2
t4 : = t1 – t3

Generated code sequence for basic block:

MOV a , R0
ADD b , R0
MOV c , R1
ADD d , R1
MOV R0 , t1
MOV e , R0
SUB R1 , R0
MOV t1 , R1
SUB R0 , R1
MOV R1 , t4

Rearranged basic block:

Now t1 occurs immediately before t4.

t2 : = c + d
t3 : = e – t2
t1 : = a + b
t4 : = t1 – t3

Revised code sequence:

MOV c , R0
ADD d , R0
MOV a , R0
SUB R0 , R1
MOV a , R0
ADD b , R0
SUB R1 , R0
MOV R0 , t4

In this order, two instructions MOV R0 , t1 and MOV t1 , R1 have been saved.
A Heuristic ordering for Dags

The heuristic ordering algorithm attempts to make the evaluation of a node immediately follow
the evaluation of its leftmost argument.

The algorithm shown below produces the ordering in reverse.

Algorithm:

1) while unlisted interior nodes remain do begin

2) select an unlisted node n, all of whose parents have been listed;
3) list n;
4) while the leftmost child m of n has no unlisted parents and is not a leaf do
begin
5) list m;
6) n:=m
end
end

Example: Consider the DAG shown below:

1
*

2 + - 3

4
*

5 - + 8

6 + 7 c d 11 e 12

a b
9 10

Initially, the only node with no unlisted parents is 1 so set n=1 at line (2) and list 1 at line (3).

Now, the left argument of 1, which is 2, has its parents listed, so we list 2 and set n=2 at line (6).

Now, at line (4) we find the leftmost child of 2, which is 6, has an unlisted parent 5. Thus we
select a new n at line (2), and node 3 is the only candidate. We list 3 and proceed down its left
chain, listing 4, 5 and 6. This leaves only 8 among the interior nodes so we list that.

The resulting list is 1234568 and the order of evaluation is 8654321.

Code
sequence:

t8 : = d + e
t6 : = a + b
t5 : = t6 – c
t4 : = t5 * t8
t3 : = t4 – e
t2 : = t6 + t4
t1 : = t2 * t3

This will yield an optimal code for the DAG on machine whatever be the n umber of registers.