I was told that a while loop was more efficient than a for loop. (c/c++)
This seemed reasonable but I wanted to find a way to prove or disprove it.
I have tried three tests using analogous snippets of code. Each containing Nothing but a for or while loop with the same output:
Compile time - roughly the same
Run time - Same
Compiled to intel assembly code and compared - Same number of lines and virtually the same code
Should I have tried anything else, or can anyone confirm one way or the other?
All loops follow the same template:
{
// Initialize
LOOP:
if(!(/* Condition */) ) {
goto END
}
// Loop body
// Loop increment/decrement
goto LOOP
}
END:
Therefor the two loops are the same:
// A
for(int i=0; i<10; i++) {
// Do stuff
}
// B
int i=0;
while(i < 10) {
// Do stuff
i++;
}
// Or even
int i=0;
while(true) {
if(!(i < 10) ) {
break;
}
// Do stuff
i++;
}
Both are converted to something similar to:
{
int i=0;
LOOP:
if(!(i < 10) ) {
goto END
}
// Do stuff
i++;
goto LOOP
}
END:
Unused/unreachable code will be removed from the final executable/library.
Do-while loops skip the first conditional check and are left as an exercise for the reader. :)
Certainly LLVM will convert ALL types of loops to a consistent form (to the extent possible, of course). So as long as you have the same functionality, it doesn't really matter if you use for, while, do-while or goto to form the loop, if it's got the same initialization, exit condition, and update statement and body, it will produce the exact same machine code.
This is not terribly hard to do in a compiler if it's done early enough during the optimisation (so the compiler still understands what is actually being written). The purpose of such "make all loops equal" is that you then only need one way to optimise loops, rather than having one for while-loops, one for for-loops, one for do-while loops and one for "any other loops".
It's not guaranteed for ALL compilers, but I know that gcc/g++ will also generate nearly identical code whatever loop construct you use, and from what I've seen Microsoft also does the same.
C and C++ compilers actually convert high level C or C++ codes to assembly codes and in assembly we don't have while or for loops. We can only check a condition and jump to another location.
So, performance of for or while loop heavily depends on how strong the compiler is to optimize the codes.
This is good paper on code optimizations:
http://www.linux-kongress.org/2009/slides/compiler_survey_felix_von_leitner.pdf.
Related
I'm trying to carry out some loop optimization as described here: Optimizing a Loop vs Code Duplication
I have the additional complication that some code inside the loop only needs to be executed depending on a combination of run-time-known variables external to the loop (which can be replaced with template parameters for optimization, as discussed in the link above) and a run-time-known variable that is only known inside the loop.
Here is the completely un-optimized version of the code:
for (int i = 0; i < 100000, i++){
if (external_condition_1 || (external_condition_2 && internal_condition[i])){
run_some_code;
}
else{
run_some_other_code;
}
run_lots_of_other_code;
}
This is my attempt at wrapping the loop in a templated function as suggested in the question linked above to optimize performance and avoid code duplication by writing multiple versions of the loop:
template<bool external_condition_1, external_condition_2>myloop(){
for (int i = 0; i < 100000, i++){
if (external_condition_1 || (external_condition_2 && internal_condition[i]){
run_some_code;
}
else{
run_some_other_code;
}
run_lots_of_other_code;
}
My question is: how can the code be written to avoid branching and code duplication?
Note that the code is sufficiently complex that the function probably can't be inlined, and compiler optimization also likely wouldn't sort this out in general.
My question is: how can the code be written to avoid branching and code duplication?
Well, you already wrote your template to avoid code duplication, right? So let's look at what branching is left. To do this, we should look at each function that is generated from your template (there are four of them). We should also apply the expected compiler optimizations based upon the template parameters.
First up, set condition 1 to true. This should produce two functions that are essentially (using a bit of pseudo-syntax) the following:
myloop<true, bool external_condition_2>() {
for (int i = 0; i < 100000, i++){
// if ( true || whatever ) <-- optimized out
run_some_code;
run_lots_of_other_code;
}
}
No branching there. Good. Moving on to the first condition being false and the second condition being true.
myloop<false, true>(){
for (int i = 0; i < 100000, i++){
if ( internal_condition[i] ){ // simplified from (false || (true && i_c[i]))
run_some_code;
}
else{
run_some_other_code;
}
run_lots_of_other_code;
}
}
OK, there is some branching going on here. However, each i needs to be analyzed to see which code should execute. I think there is nothing more that can be done here without more information about internal_condition. I'll give some thoughts on that later, but let's move on to the fourth function for now.
myloop<false, false>() {
for (int i = 0; i < 100000, i++){
// if ( false || (false && whatever) ) <-- optimized out
run_some_other_code;
run_lots_of_other_code;
}
}
No branching here. You already have done a good job avoiding branching and code duplication.
OK, let's go back to myloop<false,true>, where there is branching. The branching is largely unavoidable simply because of how your situation is set up. You are going to iterate many times. Some iterations you want to do one thing while other iterations should do another. To get around this, you would need to re-envision your setup so that you can do the same thing each iteration. (The optimization you are working from is based upon doing the same thing each iteration, even though it might be a different thing the next time the loop starts.)
The simplest, yet unlikely, scenario would be where internal_condition[i] is equivalent to something like i < 5000. It would also be convenient if you could do all of the "some code" before any of the "lots of other code". Then you could loop from 0 to 4999, running "some code" each iteration. Then loop from 5000 to 99999, running "other code". Then a third loop to run "lots of other code".
Any solution I can think of would involve adapting your situation to make it more like the unlikely simple scenario. Can you calculate how many times internal_condition[i] is true? Can you iterate that many times and map your (new) loop control variable to the appropriate value of i (the old loop control variable)? (Or maybe the exact value of i is not important?) Then do a second loop to cover the remaining cases? In some scenarios, this might be trivial. In others, far from it.
There might be other tricks that could be done, but they depend on more details about what you are doing, what you need to do, and what you think you need to do but don't really. (It's possible that the required level of detail would overwhelm StackOverflow.) Is the order important? Is the exact value of i important?
In the end, I would opt for profiling the code. Profile the code without code duplication but with branching. Profile the code with minimal branching but with code duplication. Is there a measurable change? If so, think about how you can re-arrange your internal condition so that i can cover large ranges without changing the value of the internal condition. Then divide your loop into smaller pieces.
In C++17, to guaranty no extra branches evaluation, you might do:
template <bool external_condition_1, bool external_condition_2>
void myloop()
{
for (int i = 0; i < 100000, i++){
if constexpr (external_condition_1) {
run_some_code;
} else if constexpr (external_condition_2){
if (internal_condition[i]) {
run_some_code;
} else {
run_some_other_code;
}
} else {
run_some_other_code;
}
run_lots_of_other_code;
}
}
I am writing a program in C++ and I am debating whether to put the "if"s inside the loop. I would imagine doing one check and then looping would be overall more efficient rather than the constant loop and check, but I am not quite sure. Or does any of this not matter because the compiler will optimize it anyway?
Is this more efficient?
for(int i = 0; i < SOME_BOUND; i++){
if(SOME_CONDITION){
//Some actions
}
else {
//Some actions
}
}
or is this more efficient?
if(SOME_CONDITION){
for(int i = 0; i < SOME_BOUND; i++){
//Some Actions
}
}
else {
for(int i = 0; i < SOME_BOUND; i++){
//Some Actions
}
}
One check is definitely better, however there are features both in hardware (branch prediction) and compiler (hoisting expressions and conditionals outside the loop) which make it hard to predict whether there will actually be a runtime difference between the two pieces of source code.
Generally you should focus on correctness and maintainability, and only go duplicating loops for performance reasons if profiling shows that the optimizer is missing out on performance.
The compiler can probably figure this out for you so that it executes it in the most efficient way. You should do what is more intuitive for your problem.
At a high level, doing the if only once is faster than doing it on every iteration of a loop, but again you should do what is more clear.
In a loop like so,
for (int i = 0; i < 5; i++)
{
int i_foo;
i_foo = foo();
if (i < 5)
return; //<-------- Right here
footwo();
}
How would I return that one particular turn of the loop?
I know that I could make footwo() execute under the condition that i >= 5, but I'm wondering if there is a way to make the loop exit (just the once).
For more explanation, I would like the for loop to start back at the beginning and add 1 to i, as if it had just finished that particular "loop" of the loop.
(I could not find an answer to this based on the strange wording, but if there is one just direct me and I will happily take this down.)
Use continue:
if (i < 5)
continue;
This jumps straight to the next iteration of the loop.
I'm not entirely sure what you mean but if you're checking the condition of if (i < 5) then just use the keyword continue. If the expression is true the loop will continue.
I'm not exactly what you're saying, but to clear up a bit of terminology, I think you mean to say "iteration" when you're saying a "turn of the loop" or a "loop of the loop." Common terms allow for better clarity.
As to your issue:
If you use the continue keyword, it allows you to skip to the next iteration. If you use the break keyword, it will skip past the entire iteration structure (out of the for loop entirely). This also works with while statements.
You can use continue to terminate the current iteration of a loop without terminating the loop itself. But depending on how your code is structured, an if statement might be cleaner.
Given your example, you might want:
for (int i = 0; i < 5; i++)
{
int i_foo;
i_foo = foo();
if (i_foo >= 5) {
footwo();
}
}
I'm assuming that you meant to assign the result of foo() to i_foo, not to i.
A continue can be simpler if you need to bail out from the middle of some nested structure, or if you need to bail out very early in the body of the loop and there's a lot of code that would be shoved into the if.
But in the case of nested control structures, you need to remember that continue applies only to the innermost enclosing loop; there's no construct (other than goto) for bailing out of multiple nested loops. And break applies to the innermost enclosing loop or switch statement.
what is the between in writing a loop by branching jump statement and a normal loop
i.e for example
main()
{
int i=0;
while(i<9)
{
//do something
i++;
}
and
main()
{
int i=0;
label:
//do something
i++;
if(i<9)
goto label;
}
is the performance of both of them are equal?
These two loops are not equivalent: the second one is similar to a do/while loop, not a while loop:
main() {
int i=0;
do {
//do something
i++;
} while(i<9);
}
The two should be the same as far as the performance goes; however, the readability of the solution with the goto suffers considerably, so I would refrain from using it in any of your projects.
The difference is that the first one is a lot easier to read, which is why goto is generally avoided. Also as pointed out in dasblinkenlight's answer the semantics are not the same.
The performance should be about the same, since the CPU has to do jumps in order to implement both. Of course, with a higher-level description of intent (as in the first case) the chance of the compiler being able to optimize something increases.
Perhaps not in this particular case, but in general a more high-level description of what you want to do is to be preferred, and then the looping constructs are better than goto.
In this case, compiler should produce an equivalent output. But I don't think these two examples are equivalent - a while first checks the condition and then decides wheter to execute the body. Your example with the goto first executes the body and then checks the conditions - it behaves more like a do while statement.
You shouldn't use goto in this case, as it reduces readability. gotos are valid in only a limited number of scenarios.
We can use for loop and while loop for same purpose.
in what means they effect our code if I use for instead of while? same question arises between if-else and switch-case? how to decide what to use?
for example which one you would prefer?
This code:
int main()
{
int n = 10;
for(int i=0;i<n;i++)
{
do_something();
}
return 0;
}
Or this code:
int main()
{
int n=10,i=0;
while(i<n)
{
do_something();
i++;
}
return 0;
}
if using for or while loop does not effect the code by any means then may I know What was the need to make 2 solution for same problem?
Use whichever one makes the intention of your code clearest.
If you know the number of iterations the loop should run beforehand, I would recommend the for construct. While loops are good for when the loop's terminating condition happens at some yet-to-be determined time.
I try to prefer the for loop. Why? Because when I see a for loop, I can expect all of the loop bookeeping is kept in a single statement. I can insert break or continue statements without worrying about breaking how the loop operates. And most importantly, the body of the loop focuses on what you actually want the loop to be doing, rather than maintaining the loop itself. If I see a while, then I have to look at and understand the entire loop body before I can understand what iteration pattern the loop uses.
The only place I end up using while is for those few cases where the control of the loop is provided by some outside routine (i.e. FindFirstFileW)
It's all a matter of personal opinion though. Lots of people don't like what I end up doing with for loops because the loop statement often ends up spanning multiple lines.
There are some very subtle differences..
scope of loop variable(s), for example, with the for loop i has local scope, with a while this has to be defined before (which means it is available after, of course you can do that with for as well..)
continue, with a for loop, variable will be increment/decremented, with a while, you'd have to insert the operation before continue
Frankly, if you need to increment/decrement, a for loop makes sense, if you don't know the bounds, and there is no real increment/decrement, a while loop makes more sense, e.g.
while(some_stream >> input)
{
// do stuff...
}
In general, a for loop might be preferable for simple loops, since the logic of the loop is contained in a single line:
for (int i = 0; i < 10; ++i) {...}
However, sometimes we need more complex logic or flow control. A while loop allows us to implement more complicated loops. For example, suppose we only want to increment the counter variable under certain conditions:
int i = 0;
while (i < 10)
{
if (some_condition) ++i;
else if (some_other_condition) { ... }
else break;
}
Just use the one that makes the code readable and logical.
In some cases the compiler (gcc at least) will be able to optimize a very slightly better than a for loop doing the same thing. If I remember correctly that optimization is only about few clock cycles so it probably never will have any noticeable affect on the performance.
You cannot write while(int i=0, i < n); that is, you've to define i before the while loop; means i exists inside as well as outside the loop.
However, in case of for loop, you can define i right in the for loop itself; and so i doesn't exist outside the loop. That is one difference. Just because of this difference, I like for more than while. And use while rarely, when for makes thing more cumbersome!
By no means they affect your program the way it works ! Its the matter of ease to understand better.
switch(i) // Once finding your case, you can easily know where the switch ends
// and thus the next statement of execution
{
case 1: break ;
case 2: break ;
// .....
case 10: break ;
default:break ;
}
if( i==1 ) // Here you have the pain of finding where the last else if ends !
{}
else if( i==2)
{}
// ...
else if( i==10)
{}
However, it is a matter of taste. I prefer switch.