Given the following code :
for (int i=0; i<n; i++)
{
counter += myArray[i];
}
And the Loop unrolling version :
for (int i=0; i<n; i+=4)
{
counter1 += myArray[i+0];
counter2 += myArray[i+1];
counter3 += myArray[i+2];
counter4 += myArray[i+3];
}
total = counter1+ counter2 + counter3+ counter4;
Why do we have a cache miss in the first version ?
Is the second version has indeed a better performance than the 1st ? why ?
Regards
Why do we have a cache miss in the first version ?
As Oli points out in the comments. This question is unfounded. If the data is already in the cache, then there will be no cache misses.
That aside, there is no difference in memory access between your two examples. So that will not likely be a factor in any performance difference between them.
Is the second version has indeed a better performance than the 1st ? why ?
Usually, the thing to do is to actually measure. But in this particular example, I'd say that it will likely be faster. Not because of better cache access, but because of the loop-unrolling.
The optimization that you are doing is called "Node-Splitting", where you separate the counter variable for the purpose of breaking the dependency chain.
However, in this case, you are doing a trivial reduction operation. Many modern compilers are able to recognize this pattern and do this node-splitting for you.
So is it faster? Most likely. But you should check to see if the compiler does it for you.
For the record: I just tested this on Visual Studio 2010.And I am quite surprised that it is not able to do this optimization.
; 129 :
; 130 : int counter = 0;
; 131 :
; 132 : for (int i=0; i<n; i++)
mov ecx, DWORD PTR n$[rsp]
xor edx, edx
test ecx, ecx
jle SHORT $LN1#main
$LL3#main:
; 133 : {
; 134 : counter += myArray[i];
add edx, DWORD PTR [rax]
add rax, 4
dec rcx
jne SHORT $LL3#main
$LN1#main:
; 135 : }
Visual Studio 2010 does not seem to be capable of performing "Node Splitting" for this (trivial) example...
In my lecture of Design and Analysis of
Algorithms the instructor said the for loop will take less time then while loop for the following sample algo.
1. for(int i=0;i<5;i++)
{
2. print(i);
}
1. int i=0;
2. while(i<5)
{
3. print(i);
4. i++;
}
He said that the compiler will read the 1. of for while 5 times line 2. 4 times thus total time 5+4=9
But in the case of while loop. The compiler will read the 1. for 1 time,2. for 5 time, 3 for 4time and 4. for 4 time. Thus total time 1+5+4+4 = 14time
Please tell me is this right. Is for loop is faster than while loop?
Thanks.
At least with MSVC 16 (VS 2010) the code is pretty much the same in both cases:
for
; Line 5
xor esi, esi
$LL3#main:
; Line 6
push esi
push OFFSET ??_C#_03PMGGPEJJ#?$CFd?6?$AA#
call _printf
inc esi
add esp, 8
cmp esi, 5
jl SHORT $LL3#main
while
; Line 4
xor esi, esi
$LL2#main:
; Line 6
push esi
push OFFSET ??_C#_03PMGGPEJJ#?$CFd?6?$AA#
call _printf
; Line 7
inc esi
add esp, 8
cmp esi, 5
jl SHORT $LL2#main
Code in my Subversion repository.
In all the modern compilers loop analysis is done on a lower level intermediate representation (i.e., when all the high level loop constructs are expanded into labels and jumps). For a compiler both loops are absolutely equivalent.
I'll pass on performance (hint: no difference, check the generated IR or assembly for proof) however there are two important differences in syntax and maintenance.
Syntax
The scope of the i variable is different. In the for case, the i is only accessible within the for header and body, while in the while case it is available after the loop. As a general rule, it's better to have tighter scopes, less variables in-flight mean less context to worry about when coding.
Maintenance
The for loop has the neat advantage of grouping all the iterations operations close together, so they can be inspected in one shot and so checked.
Also, there is one important difference when introducing continue statements:
for(int i = 0; i != 10; ++i) {
if (array[i] == nullptr) { continue; }
// act on it
}
int i = 0;
while (i != 10) {
if (array[i] == nullptr) { continue; }
// act on it
++i;
}
In the while case, the introduction of continue has created a bug: an infinite loop, as the counter is no longer implemented.
Impact
for loops are more readable and all-around better for regular iteration patterns. Even better, in C++11 the range-for statement:
for (Item const& item : collection) {
}
where iteration is entirely taken care of by the compiler, so you are sure not to mess up! (it makes the for_each algorithm somewhat moot... and we can wish the older for form starts retreating)
By corrolary: while loops should be reserved to irregular iteration patterns, this way they will attract special care during code review and from future maintainer by highlighting the irregularity of the case.
There are three loops in C: for, while, and do-while. What's the difference between them?
For example, it seems nearly all while statements can be replaced by for statements, right? Then, what's the advantage using while?
A while loop will always evaluate the condition first.
while (condition) {
//gets executed after condition is checked
}
A do/while loop will always execute
the code in the do{} block first
and then evaluate the condition.
do {
//gets executed at least once
} while (condition);
A for loop allows you to initiate a counter variable, a check condition, and a way to increment your counter all in one line.
for (int x = 0; x < 100; x++) {
//executed until x >= 100
}
At the end of the day, they are all still loops, but they offer some flexibility as to how they are executed.
Here is a great explanation of the reasoning behind the use of each different type of loop that may help clear things up. Thanks clyfe
The main difference between the for's
and the while's is a matter of
pragmatics: we usually use for when
there is a known number of iterations,
and use while constructs when the
number of iterations in not known in
advance. The while vs do ... while
issue is also of pragmatics, the
second executes the instructions once
at start, and afterwards it behaves
just like the simple while.
For loops are especially nice because they are concise. In order for this for loop:
for (int x = 0; x < 100; x++) {
//executed until x >= 100
}
to be written as a while loop, you'd have to do the following:
int count = 0;
while (count < 100) {
//do stuff
count++;
}
In this case, there's just more stuff to keep up with and the count++; could get lost in the logic. This could end up being troublesome depending on where count gets incremented, and whether or not it should get incremented before or after the loop's logic. With a for loop, your counter variable is always incremented before the next iteration of the loop, which adds some uniformity to your code.
For the sake of completeness, it's probably meaningful to talk about break and continue statements here which come in handy when doing loop processing.
break will instantly terminate the current loop and no more iterations will be executed.
//will only run "do stuff" twice
for (int x = 0; x < 100; x++) {
if (x == 2) {
break;
}
//do stuff
}
continue will terminate the current iteration and move on to the next one.
//will run "do stuff" until x >= 100 except for when x = 2
for (int x = 0; x < 100; x++) {
if (x == 2) {
continue;
}
//do stuff
}
Note that in a for loop, continue evaluates the part3 expression of for (part1; part2; part3); in contrast, in a while loop, it just jumps to re-evaluate the loop condition.
If there is a strong concern about speed and performance, the best approach is to verify the code produced by the compiler at the assembly level.
For instance, the following code shows that the "do-while" is a bit faster. This because the "jmp" instruction is not used by the "do-while" loop.
BTW, in this specific example, the worst case is given by the "for" loop. :))
int main(int argc, char* argv[])
{
int i;
char x[100];
// "FOR" LOOP:
for (i=0; i<100; i++ )
{
x[i] = 0;
}
// "WHILE" LOOP:
i = 0;
while (i<100 )
{
x[i++] = 0;
}
// "DO-WHILE" LOOP:
i = 0;
do
{
x[i++] = 0;
}
while (i<100);
return 0;
}
// "FOR" LOOP:
010013C8 mov dword ptr [ebp-0Ch],0
010013CF jmp wmain+3Ah (10013DAh)
for (i=0; i<100; i++ )
{
x[i] = 0;
010013D1 mov eax,dword ptr [ebp-0Ch] <<< UPDATE i
010013D4 add eax,1
010013D7 mov dword ptr [ebp-0Ch],eax
010013DA cmp dword ptr [ebp-0Ch],64h <<< TEST
010013DE jge wmain+4Ah (10013EAh) <<< COND JUMP
010013E0 mov eax,dword ptr [ebp-0Ch] <<< DO THE JOB..
010013E3 mov byte ptr [ebp+eax-78h],0
010013E8 jmp wmain+31h (10013D1h) <<< UNCOND JUMP
}
// "WHILE" LOOP:
i = 0;
010013EA mov dword ptr [ebp-0Ch],0
while (i<100 )
{
x[i++] = 0;
010013F1 cmp dword ptr [ebp-0Ch],64h <<< TEST
010013F5 jge wmain+6Ah (100140Ah) <<< COND JUMP
010013F7 mov eax,dword ptr [ebp-0Ch] <<< DO THE JOB..
010013FA mov byte ptr [ebp+eax-78h],0
010013FF mov ecx,dword ptr [ebp-0Ch] <<< UPDATE i
01001402 add ecx,1
01001405 mov dword ptr [ebp-0Ch],ecx
01001408 jmp wmain+51h (10013F1h) <<< UNCOND JUMP
}
// "DO-WHILE" LOOP:
i = 0;
. 0100140A mov dword ptr [ebp-0Ch],0
do
{
x[i++] = 0;
01001411 mov eax,dword ptr [ebp-0Ch] <<< DO THE JOB..
01001414 mov byte ptr [ebp+eax-78h],0
01001419 mov ecx,dword ptr [ebp-0Ch] <<< UPDATE i
0100141C add ecx,1
0100141F mov dword ptr [ebp-0Ch],ecx
01001422 cmp dword ptr [ebp-0Ch],64h <<< TEST
01001426 jl wmain+71h (1001411h) <<< COND JUMP
}
while (i<100);
For the sake of readability
They're all interchangeable; you could pick one type and use nothing but that forever, but usually one is more convenient for a given task. It's like saying "why have switch, you can just use a bunch of if statements" -- true, but if it's a common pattern to check a variable for a set of values, it's convenient and much easier to read if there's a language feature to do that
If you want a loop to execute while a condition is true, and not for a certain number of iterations, it is much easier for someone else to understand:
while (cond_true)
than something like this:
for (; cond_true ; )
Remember, a for loop is essentially a fancy while loop. They're the same thing.
while <some condition is true> {
// do some stuff
// possibly do something to change the condition
}
for ( some var, <some condition is true>; increment var ) {
}
The advantage of a for loop is that it's harder to accidentally do an infinite loop. Or rather, it's more obvious when you do one because you generally put the loop var in the initial statement.
A while loop is more clear when you're not doing a standard incrementing pattern. For example:
int x = 1;
while( x != 10 ) {
if ( some condition )
x = 10;
else
x += 5;
}
You should use such a loop, that most fully conforms to your needs.
For example:
for(int i = 0; i < 10; i++)
{
print(i);
}
//or
int i = 0;
while(i < 10)
{
print(i);
i++;
}
Obviously, in such situation, "for" looks better, than "while".
And "do while" shoud be used when some operations must be done already before the moment when condition of your loop will be checked.
Sorry for my bad english).
One common misunderstanding withwhile/for loops I've seen is that their efficiency differs. While loops and for loops are equally efficient. I remember my computer teacher from highschool told me that for loops are more efficient for iteration when you have to increment a number. That is not the case.
For loops are simply syntactically sugared while loops, and make iteration code faster to write.
When the compiler takes your code and compiles it, it is translating it into a form that is easier for the computer to understand and execute on a lower level (assembly). During this translation, the subtle differences between the while and for syntaxes are lost, and they become exactly the same.
A for suggest a fixed iteration using an index or variants on this scheme.
A while and do... while are constructions you use when there is a condition that must be checked each time (apart from some index-alike construction, see above). They differ in when the first execution of the condition check is performed.
You can use either construct, but they have their advantages and disadvantages depending on your use case.
I noticed some time ago that a For loop typically generates several more machine instructions than a while loop. However, if you look closely at the examples, which mirror my observations, the difference is two or three machine instructions, hardly worth much consideration.
Note, too, that the initializer for a WHILE loop can be eliminated by baking it into the code, e. g.:
static int intStartWith = 100;
The static modifier bakes the initial value into the code, saving (drum roll) one MOV instruction. Of greater significance, marking a variable as static moves it outside the stack frame. Variable alignment permitting, it may also produce slightly smaller code, too, since the MOV instruction and its operands take more room than, for example an integer, Boolean, or character value (either ANSI or Unicode).
However, if variables are aligned on 8 byte boundaries, a common default setting, an int, bool, or TCHAR baked into code costs the same number of bytes as a MOV instruction.
They are all the same in the work they do. You can do the same things using any of them. But for readability, usability, convenience etc., they differ.
A difference between while and do-while is that while checks the loop condition and if this is true, the body is executed and the condition checked again. The do-while checks the condition after execution of the body, so with do-while the body is executed at least one time.
Of course you can write a while loop as a do-while and vv, but this usually requires some code duplication.
One peculiarity of the do while is that you need a semi-colon after the while to complete. It is often used in macro definitions to execute several statements only once while constraining the impact of the macro. If macros where defined as blocks, some parsing errors may occur.
One explanation among others
For loops (at least considering C99) are superior to while loops because they limit the scope of the incremented variable(s).
Do while loops are useful when the condition is dependant on some inputs. They are the most seldom used of the three loop types.
Between for and while: while does not need initialization nor update statement, so it may look better, more elegant; for can have statements missing, one two or all, so it is the most flexible and obvious if you need initialization, looping condition and "update" before looping. If you need only loop condition (tested at the beginning of the loop) then while is more elegant.
Between for/while and do-while: in do-while the condition is evaluated at the end of the loop. More confortable if the loop must be executed at least once.
WHILE is more flexible. FOR is more concise in those instances in which it applies.
FOR is great for loops which have a counter of some kind, like
for (int n=0; n<max; ++n)
You can accomplish the same thing with a WHILE, of course, as others have pointed out, but now the initialization, test, and increment are broken across three lines. Possibly three widely-separated lines if the body of the loop is large. This makes it harder for the reader to see what you're doing. After all, while "++n" is a very common third piece of the FOR, it's certainly not the only possibility. I've written many loops where I write "n+=increment" or some more complex expression.
FOR can also work nicely with things other than a counter, of course. Like
for (int n=getFirstElementFromList(); listHasMoreElements(); n=getNextElementFromList())
Etc.
But FOR breaks down when the "next time through the loop" logic gets more complicated. Consider:
initializeList();
while (listHasMoreElements())
{
n=getCurrentElement();
int status=processElement(n);
if (status>0)
{
skipElements(status);
advanceElementPointer();
}
else
{
n=-status;
findElement(n);
}
}
That is, if the process of advancing may be different depending on conditions encountered while processing, a FOR statement is impractical. Yes, sometimes you could make it work with a complicated enough expressions, use of the ternary ?: operator, etc, but that usually makes the code less readable rather than more readable.
In practice, most of my loops are either stepping through an array or structure of some kind, in which case I use a FOR loop; or are reading a file or a result set from a database, in which case I use a WHILE loop ("while (!eof())" or something of that sort).
They are pretty much same except for do-while loop. The for loop is good when you have a counter kind of variable. It makes it obvious. while loop makes sense in cases where a flag is being checked as show below :
while (!done) {
if (some condtion)
done = true;
}
while and for statements can both be used for looping in programming. It will depend on the programmer as to whether the while loop or for loop is used. Some are comfortable using while loop and some are with for loop.
Use any loop you like. However, the do...while loop can be somewhat tricky in C programming.
/*
while loop
5 bucks
1 chocolate = 1 bucks
while my money is greater than 1 bucks
select chocolate
pay 1 bucks to the shopkeeper
money = money - 1
end
come to home and cant go to while shop because my money = 0 bucks
*/
#include<stdio.h>
int main(){
int money = 5;
while( money >= 1){
printf("inside the shopk and selecting chocolate\n");
printf("after selecting chocolate paying 1 bucks\n");
money = money - 1 ;
printf("my remaining moeny = %d\n", money);
printf("\n\n");
}
printf("dont have money cant go inside the shop, money = %d", money);
return 0;
}
infinite money
while( codition ){ // condition will always true ....infinite loop
statement(s)
}
please visit this video for better understanding
https://www.youtube.com/watch?v=eqDv2wxDMJ8&t=25s
/*
for loop
5 bucks
for my money is greater than equal to 1 bucks 0 money >= 1
select chocolate
pay 1 bucks to the shopkeeper
money = money - 1 1-1 => 0
end
*/
#include<stdio.h>
int main(){
int money = 5;
for( ; money >= 1; ){ 0>=1 false
printf("select chocolate \n");
printf("paying 1 bucks to the shopkeeper\n");
money = money - 1; 1-1 = 0
printf(" remaining money =%d\n", money);
printf("\n\n");
}
return 0;
}
For better understanding please visit https://www.youtube.com/watch?v=_vdvyzzp-R4&t=25s