Related
In what circumstances should I prefer pass-by-reference? Pass-by-value?
There are four main cases where you should use pass-by-reference over pass-by-value:
If you are calling a function that needs to modify its arguments, use pass-by-reference or pass-by-pointer. Otherwise, you’ll get a copy of the argument.
If you're calling a function that needs to take a large object as a parameter, pass it by const reference to avoid making an unnecessary copy of that object and taking a large efficiency hit.
If you're writing a copy or move constructor which by definition must take a reference, use pass by reference.
If you're writing a function that wants to operate on a polymorphic class, use pass by reference or pass by pointer to avoid slicing.
There are several considerations, including:
Performance
Passing by value copies the data, so passing large data structures by value can inhibit performance. Passing by reference passes only a reference (basically the address) to the data. For large data structures, this can greatly improve performance. For smaller data structures (like an int), passing by reference can inhibit performance.
Modifications
Passing by value copies the data so if the target code modifies that copy, it will not affect the original. Passing by reference passes only the address of the data, so modifications made against that reference will be "visible" to the calling code.
Yes.
Pass by value for things like native types that are small enough that passing them directly is efficient. Otherwise use pass by (const) reference.
The hard part is writing a template that could apply to either (in which case, you usually want to use pass by reference -- the potential penalty for passing a large object by value is much worse than the potential penalty for passing by reference when passing by value would have been preferred).
Edit: this, of course, is assuming a situation where the required semantics would allow either one -- obviously if you're working with something like polymorphic objects, there's no real "preference" involved, because you must use a pointer or reference to get correct behavior.
As others already have replied to your question sufficiently well, I would like to add an important point:
If the class does not have public copy-constructor, then you don't have choice to pass by value; you have to pass by reference (or you can pass pointer).
The following program would not compile:
class A
{
public:
A(){}
private:
A(const A&) {}
};
//source of error : pass by value
void f(A ) {}
int main() {
A a;
f(a);
return 0;
}
Error:
prog.cpp: In function ‘int main()’:
prog.cpp:10: error: ‘A::A(const A&)’ is private
prog.cpp:18: error: within this context
prog.cpp:18: error: initializing argument 1 of ‘void f(A)’
See yourself at ideone : http://www.ideone.com/b2WLi
But once you make function f pass by reference, then it compiles fine : http://www.ideone.com/i6XXB
here's the simple rule:
pass by reference when the value is large.
the other answers are amazing. Just trying to make this simplest.
You have tagged your question with both C and C++.
Therefore, I suggest that you consider using pass by reference in C++ which supports this feature and that you do not consider using it in C which does not support this feature.
pass by reference can be called only in below conditions:
Pass-by-references is more efficient than pass-by-value, because it does not copy the arguments. The formal parameter is an alias for the argument. When the called function read or write the formal parameter, it is actually read or write the argument itself.
The difference between pass-by-reference and pass-by-value is that modifications made to arguments passed in by reference in the called function have effect in the calling function, whereas modifications made to arguments passed in by value in the called function can not affect the calling function.
Use pass-by-reference if you want to modify the argument value in the calling function. Otherwise, use pass-by-value to pass arguments.
The difference between pass-by-reference and pass-by-pointer is
that pointers can be NULL or reassigned whereas references cannot.
Use pass-by-pointer if NULL is a valid parameter value or if you want to reassign the pointer.
Otherwise, use constant or non-constant references to pass arguments.
While pointers are references, "reference" in c++ usually refers to the practice of tagging a parameter of SomeType&.
Which you should never do. The only place it is appropriate is as a magic syntax required to implement the various pre-defined operators. Otherwise:
You should never pass out parameters by reference - pass by pointer, otherwise you make code reviews all but impossible. Pass by reference makes it impossible to tell by examining a call which parameters can be expected to be changed.
You should never pass in parameter by reference either. Again, this means you are performing a meta optimization. You should always just pass-by-value, otherwise you are guilty of peeking inside an object, examining its implementation and deciding that pass-by-reference is preferred for some reason.
Any c++ class should implement all the copy and assignment constructors and overloads necessary to be passed around by value. Otherwise it has not done its job, of abstracting the programmer from the implementation details of the class.
If a have a function that takes a Eigen matrix as an argument, what would be the difference between:
void foo(Eigen::MatrixXd& container){
for(i=0;i<container.rows();i++){
for(j=0;j<container.cols();j++){
container(i,j)=47;
}
}
}
and
void foo(Eigen::MatrixXd* container){
for(i=0;i<container->rows();i++){
for(j=0;j<container->cols();j++){
container->coeffRef(i,j)=47;
}
}
}
In Eigen documentation, they only present the first method - does that mean that there are any advantages to that approach? And what are the drawbacks of not using const when passing the Matrix reference in the first case?
References are nice because there is no such thing as a null reference, so using a reference parameter reduces the risk of someone calling your function with an invalid value.
On the other hand some coding standards recommend making parameters you intend to modify pointers instead of non-const references. This forces the caller to explicitly take the address of any value they pass in making it more obvious the value will be modified. The choice of pointer vs. non-const reference is up to you.
However, if you do not intend to modify the parameter then making it a const reference is definitely the way to go. It avoids the problem of passing invalid pointers, allows you to pass in temporaries, and the caller doesn't care if the parameter is taken by reference since it isn't going to be modified.
With C++ code, there is the expectation that if a parameter is passed as pointer rather than reference, then the null pointer is a valid argument.
That is, by default you should use reference parameters. Only use pointers if the parameter is in some way "optional" and you want the caller to be able to pass the null pointer to signify "no value".
see the line:
container(i,j)=47.
That's not a constant operation, so you're not going to be able to set it to const.
One way a reference is different than a pointer is that your container reference can't be null. Pass by reference is a good way to avoid some errors while getting the benefits of not copying.
With C++ how do i decide if i should pass an argument by value or by reference/pointer? (tell me the answer for both 32 and 64bits) Lets take A. Is 2 32bit values more less or equal work as a pointer to a 32bit value?
B to me seems like i always should pass by value. C i think i should pass by value but someone told me (however i haven't seen proof) that processors don't handle values not their bitsize and so it is more work. So if i were passing them around would it be more work to pass by value thus byref is faster? Finally i threw in an enum. I think enums should always be by value
Note: When i say by ref i mean a const reference or pointer (can't forget the const...)
struct A { int a, b; }
struct B { int a; }
struct C { char a, b; }
enum D { a,b,c }
void fn(T a);
Now tell me the answer if i were pushing the parameters many times and the code doesn't use a tail call? (lets say the values isnt used until 4 or so calls deep)
Forget the stack size. You should pass by reference if you want to change it, otherwise you should pass by value.
Preventing the sort of bugs introduced by allowing functions to change your data unexpectedly is far more important than a few bytes of wasted stack space.
If stack space becomes a problem, stop using so many levels (such as replacing a recursive solution with an iterative one) or expand your stack. Four levels of recursion isn't usually that onerous, unless your structures are massive or you're operating in the embedded world.
If performance becomes a problem, find a faster algorithm :-) If that's not possible, then you can look at passing by reference, but you need to understand that it's breaking the contract between caller and callee. If you can live with that, that's okay. I generally can't :-)
The intent of the value/reference dichotomy is to control what happens to the thing you pass as a parameter at the language level, not to fiddle with the way an implementation of the language works.
I pass all parameters by reference for consistency, including builtins (of course, const is used where possible).
I did test this in performance critical domains -- worst case loss compared to builtins was marginal. Reference can be quite a bit faster, for non-builtins, and when the calls are deep (as a generalization). This was important for me as I was doing quite a bit of deep TMP, where function bodies were tiny.
You might consider breaking that convention if you're counting instructions, the hardware is register-starved (e.g. embedded), or if the function is not a good candidate for inlining.
Unfortunately, the question you ask is more complex than it appears -- the answer may vary greatly by your platform, ABI, calling conventions, register counts, etc.
A lot depends on your requirement but best practice is to pass by reference as it reduces the memory foot print.
If you pass large objects by value, a copy of it is made in memory andthe copy constructor is called for making a copy of this.
So it will take more machine cycles and also, if you pass by value, changes are not reflected in the original object.
So try passing them by reference.
Hope this has been helpful to you.
Regards, Ken
First, reference and pointers aren't the same.
Pass by pointer
Pass parameters by pointers if any/some of these apply:
The passed element could be null.
The resource is allocated inside the called function and the caller is responsible should be responsible for freeing such a resource. Remember in this case to provide a free() function for that resource.
The value is of a variable type, like for example void*. When it's type is determined at runtime or depending on the usage pattern (or hiding implementation - i.e Win32 HANDLE), such as a thread procedure argument. (Here favor c++ templates and std::function, and use pointers for this purpose only if your environment does not permit otherwise.
Pass by reference
Pass parameters by reference if any/some of these apply:
Most of the time. (prefer passing by const reference)
If you want the modifications to the passed arguments to be visible to the caller. (unless const reference is used).
If the passed argument is never null.
If you know what is the passed argument type and you have control over function's signature.
Pass by copy
Pass a copy if any/some of these apply:
Generally try to avoid this.
If you want to operate on a copy of the passed argument. i.e you know that the called function would create a copy anyway.
With primitive types smaller than the system's pointer size - as it makes no performance/memory difference compared to a const ref.
This is tricky - when you know that the type implements a move constructor (such as std::string in C++11). It then looks as if you're passing by copy.
Any of these three lists can go more longer, but these are - I would say - the basic rules of thumb.
Your complete question is a bit unclear to me, but I can answer when you would use passing by value or by reference.
When passing by value, you have a complete copy of the parameter into the call stack. It's like you're making a local variable in the function call initialized with whatever you passed into it.
When passing by reference, you... well, pass by reference. The main difference is that you can modify the external object.
There is the benefit of reducing memory load for large objects passing by reference. For basic data types (32-bit or 64-bit integers, for example), the performance is negligible.
Generally, if you're going to work in C/C++ you should learn to use pointers. Passing objects as parameters will almost always be passed via a pointer (vs reference). The few instances you absolutely must use references is in the copy constructor. You'll want to use it in the operators as well, but it's not required.
Copying objects by value is usually a bad idea - more CPU to do the constructor function; more memory for the actual object. Use const to prevent the function modifying the object. The function signature should tell the caller what might happen to the referenced object.
Things like int, char, pointers are usually passed by value.
As to the structures you outlined, passing by value will not really matter. You need to do profiling to find out, but on the grand scheme of a program you be better off looking elsewhere for increasing performance in terms of CPU and/or memory.
I would consider whether you want value or reference semantics before you go worrying about optimizations. Generally you would pass by reference if you want the method you are calling to be able to modify the parameter. You can pass a pointer in this case, like you would in C, but idiomatic C++ tends to use references.
There is no rule that says that small types or enums should always be passed by value. There is plenty of code that passes int& parameters, because they rely on the semantics of passing by reference. Also, you should keep in mind that for any relatively small data type, you won't notice a difference in speed between passing by reference and by value.
That said, if you have a very large structure, you probably don't want to make lots of copies of it. This is where const references are handy. Do keep in mind though that const in C++ is not strictly enforced (even if it's considered bad practice, you can always const_cast it away). There is no reason to pass a const int& over an int, although there is a reason to pass a const ClassWithManyMembers& over a ClassWithManyMembers.
All of the structs that you listed I would say are fine to pass by value if you are intending them to be treated as values. Consider that if you call a function that takes one parameter of type struct Rectangle{int x, y, w, h}, this is the same as passing those 4 parameters independently, which is really not a big deal. Generally you should be more worried about the work that the copy constructor has to do - for example, passing a vector by value is probably not such a good idea, because it will have to dynamically allocate memory and iterate through a list whose size you don't know, and invoke many more copy constructors.
While you should keep all this in mind, a good general rule is: if you want refence semantics, pass by refence. Otherwise, pass intrinsics by value, and other things by const reference.
Also, C++11 introduced r-value references which complicate things even further. But that's a different topic.
These are the rules that I use:
for native types:
by value when they are input arguments
by non-const reference when they are mandatory output arguments
for structs or classes:
by const reference when they are input arguments
by non-const reference when they are output arguments
for arrays:
by const pointer when they are input arguments (const applies to the data, not the pointer here, i.e. const TYPE *)
by pointer when they are output arguments (const applies to the data, not the pointer)
I've found that there are very few times that require making an exception to the above rules. The one exception that comes to mind is for a struct or class argument that is optional, in which case a reference would not work. In that case I use a const pointer (input) or a non-const pointer (output), so that you can also pass 0.
If you want a copy, then pass by value. If you want to change it and you want those changes to be seen outside the function, then pass by reference. If you want speed and don't want to change it, pass by const reference.
In what circumstances should I prefer pass-by-reference? Pass-by-value?
There are four main cases where you should use pass-by-reference over pass-by-value:
If you are calling a function that needs to modify its arguments, use pass-by-reference or pass-by-pointer. Otherwise, you’ll get a copy of the argument.
If you're calling a function that needs to take a large object as a parameter, pass it by const reference to avoid making an unnecessary copy of that object and taking a large efficiency hit.
If you're writing a copy or move constructor which by definition must take a reference, use pass by reference.
If you're writing a function that wants to operate on a polymorphic class, use pass by reference or pass by pointer to avoid slicing.
There are several considerations, including:
Performance
Passing by value copies the data, so passing large data structures by value can inhibit performance. Passing by reference passes only a reference (basically the address) to the data. For large data structures, this can greatly improve performance. For smaller data structures (like an int), passing by reference can inhibit performance.
Modifications
Passing by value copies the data so if the target code modifies that copy, it will not affect the original. Passing by reference passes only the address of the data, so modifications made against that reference will be "visible" to the calling code.
Yes.
Pass by value for things like native types that are small enough that passing them directly is efficient. Otherwise use pass by (const) reference.
The hard part is writing a template that could apply to either (in which case, you usually want to use pass by reference -- the potential penalty for passing a large object by value is much worse than the potential penalty for passing by reference when passing by value would have been preferred).
Edit: this, of course, is assuming a situation where the required semantics would allow either one -- obviously if you're working with something like polymorphic objects, there's no real "preference" involved, because you must use a pointer or reference to get correct behavior.
As others already have replied to your question sufficiently well, I would like to add an important point:
If the class does not have public copy-constructor, then you don't have choice to pass by value; you have to pass by reference (or you can pass pointer).
The following program would not compile:
class A
{
public:
A(){}
private:
A(const A&) {}
};
//source of error : pass by value
void f(A ) {}
int main() {
A a;
f(a);
return 0;
}
Error:
prog.cpp: In function ‘int main()’:
prog.cpp:10: error: ‘A::A(const A&)’ is private
prog.cpp:18: error: within this context
prog.cpp:18: error: initializing argument 1 of ‘void f(A)’
See yourself at ideone : http://www.ideone.com/b2WLi
But once you make function f pass by reference, then it compiles fine : http://www.ideone.com/i6XXB
here's the simple rule:
pass by reference when the value is large.
the other answers are amazing. Just trying to make this simplest.
You have tagged your question with both C and C++.
Therefore, I suggest that you consider using pass by reference in C++ which supports this feature and that you do not consider using it in C which does not support this feature.
pass by reference can be called only in below conditions:
Pass-by-references is more efficient than pass-by-value, because it does not copy the arguments. The formal parameter is an alias for the argument. When the called function read or write the formal parameter, it is actually read or write the argument itself.
The difference between pass-by-reference and pass-by-value is that modifications made to arguments passed in by reference in the called function have effect in the calling function, whereas modifications made to arguments passed in by value in the called function can not affect the calling function.
Use pass-by-reference if you want to modify the argument value in the calling function. Otherwise, use pass-by-value to pass arguments.
The difference between pass-by-reference and pass-by-pointer is
that pointers can be NULL or reassigned whereas references cannot.
Use pass-by-pointer if NULL is a valid parameter value or if you want to reassign the pointer.
Otherwise, use constant or non-constant references to pass arguments.
While pointers are references, "reference" in c++ usually refers to the practice of tagging a parameter of SomeType&.
Which you should never do. The only place it is appropriate is as a magic syntax required to implement the various pre-defined operators. Otherwise:
You should never pass out parameters by reference - pass by pointer, otherwise you make code reviews all but impossible. Pass by reference makes it impossible to tell by examining a call which parameters can be expected to be changed.
You should never pass in parameter by reference either. Again, this means you are performing a meta optimization. You should always just pass-by-value, otherwise you are guilty of peeking inside an object, examining its implementation and deciding that pass-by-reference is preferred for some reason.
Any c++ class should implement all the copy and assignment constructors and overloads necessary to be passed around by value. Otherwise it has not done its job, of abstracting the programmer from the implementation details of the class.
Provided, I want to pass a modifiable parameter to a function, what should I choose: to pass it by pointer or to pass it by reference?
bool GetFoo ( Foo& whereToPlaceResult );
bool GetFoo ( Foo* whereToPlaceResult );
I am asking this because I always considered it the best practice to pass parameter by reference (1), but after examining some local code database, I came to a conclusion, that the most common way is (2). Moreover, the man himself (Bjarne Stroustrup) recommends using (2). What are the [dis]advantages of (1) and (2), or is it just a matter of personal taste?
I prefer a reference instead of a pointer when:
It can't be null
It can't be changed (to point to something else)
It mustn't be deleted (by whoever receives the pointer)
Some people say though that the difference between a reference and a const reference is too subtle for many people, and is invisible in the code which calls the method (i.e., if you read the calling code which passes a parameter by reference, you can't see whether it's a const or a non-const reference), and that therefore you should make it a pointer (to make it explicit in the calling code that you're giving away the address of your variable, and that therefore the value of your variable may be altered by the callee).
I personally prefer a reference, for the following reason:
I think that a routine should know what subroutine it's calling
A subroutine shouldn't assume anything about what routine it's being called from.
[1.] implies that making the mutability visible to the caller doesn't matter much, because the caller should already (by other means) understand what the subroutine does (including the fact that it will modify the parameter).
[2.] implies that if it's a pointer then the subroutine should handle the possibility of the parameter's being a null pointer, which may be extra and IMO useless code.
Furthermore, whenever I see a pointer I think, "who's going to delete this, and when?", so whenever/wherever ownership/lifetime/deletion isn't an issue I prefer to use a reference.
For what it's worth I'm in the habit of writing const-correct code: so if I declare that a method has a non-const reference parameter, the fact that it's non-const is significant. If people weren't writing const-correct code then maybe it would be harder to tell whether a parameter will be modified in a subroutine, and the argument for another mechanism (e.g. a pointer instead of a reference) would be a bit stronger.
Advantages to passing by reference:
Forces user to supply a value.
Less error-prone: Handles pointer dereferencing itself. Don't have to check for null inside.
Makes the calling code look much cleaner.
Advantages to passing pointer by value:
Allows null to be passed for "optional" parameters. Kinda an ugly hack, but sometimes useful.
Forces caller to know what is being done w/ the parameter.
Gives the reader half a clue of what might be being done w/ the parameter without having to read the API.
Since reference passing is in the language, any non-pointer parameters might be getting modified too, and you don't know that pointer values are being changed. I've seen APIs where they are treated as constants. So pointer passing doesn't really give readers any info that they can count on. For some people that might be good enough, but for me it isn't.
Really, pointer passing is just an error-prone messy hack leftover from C which had no other way to pass values by reference. C++ has a way, so the hack is no longer needed.
One advantage to passing by reference is that they cannot be null (unlike pointers), obviating the need to null-check every out parameter.
I'd recommend that you consider (may not be best for every situation) returning Foo from the function rather than modifying a parameter. Your function prototype would look like this:
Foo GetFoo() // const (if a member function)
As you appear to be returning a success/failure flag, using an exception might be a better strategy.
Advantages:
You avoid all of the pointer/reference issues
Simplifies life for the caller. Can pass the return value to other functions without using a local variable, for example.
Caller cannot ignore error status if you throw an exception.
Return value optimization means that it may be as efficient as modifying a parameter.
I choose #2 because it obvious at the point of call that the parameter will be changed.
GetFoo(&var) rather than GetFoo(var)
I prefer pass by reference for just const references, where I am trying to avoid a copy constructor call.
Pass by reference, and avoid the whole NULL pointer problem.
I seem to recall that in c++ references where not null and pointers could be. Now I've not done c++ for a long time so my memory could be rusty.
The difference here is relatively minor.
A reference cannot be NULL.
A nullpointer may be passed.
Thus you can check if that happens and react accordingly.
I personally can't think of a real advantage of one of the two possibilities.
I find this a matter of personal taste. I actually prefer to pass by reference because pointers give more freedom but they also tend to cause a lot of problems.
The benefit to a pointer is that you can pass nothing, ie. use it as if the parameter was completely optional and not have a variable the caller passes in.
References otherwise are safer, if you have one its guaranteed to exist and be writeable (unless const of course)
I think its a matter of preference otherwise, but I don't like mixing the two as I think it makes maintainace and readability of your code harder to do (especially as your 2 functions look the same to the caller)
These days I use const references for input parameters and pointers for out parameters. FWIIW, Google C++ Style Guide recommends the same approach (not that I always agree with their style guide - for instance they don't use exceptions, which usually does not make much sense)
My preference is a reference. First, because it rhymes. :) Also because of the issues pointed out by other answers: no need to dereference, and no possibility of a reference being NULL. Another reason, which I have not seen mentioned, is that when you see a pointer you cannot be sure whether or not it points to dynamically allocated memory, and you may be tempted to call delete on it. A reference, on the other hand, dispenses with any ambiguity regarding memory management.
Having said that, there are of course many cases when passing a pointer is preferable, or even necessary. If you know in advance that the parameter is optional, then allowing it to be NULL is very useful. Similarly, you may know in advance that the parameter is always dynamically allocated and have the memory management all worked out.