Advantages of pass-by-value and std::move over pass-by-reference
Clash Royale CLAN TAG#URR8PPP
up vote
45
down vote
favorite
I'm learning C++ at the moment and try avoid picking up bad habits.
From what I understand, clang-tidy contains many "best practices" and I try to stick to them as best as possible (even though I don't necessarily understand why they are considered good yet), but I'm not sure if I understand what's recommended here.
I used this class from the tutorial:
class Creature
private:
std::string m_name;
public:
Creature(const std::string &name)
: m_namename
;
This leads to a suggestion from clang-tidy that I should pass by value instead of reference and use std::move
.
If I do, I get the suggestion to make name
a reference (to ensure it does not get copied every time) and the warning that std::move
won't have any effect because name
is a const
so I should remove it.
The only way I don't get a warning is by removing const
altogether:
Creature(std::string name)
: m_namestd::move(name)
Which seems logical, as the only benefit of const
was to prevent messing with the original string (which doesn't happen because I passed by value).
But I read on CPlusPlus.com:
Although note that -in the standard library- moving implies that the moved-from object is left in a valid but unspecified state. Which means that, after such an operation, the value of the moved-from object should only be destroyed or assigned a new value; accessing it otherwise yields an unspecified value.
Now imagine this code:
std::string nameString("Alex");
Creature c(nameString);
Because nameString
gets passed by value, std::move
will only invalidate name
inside the constructor and not touch the original string. But what are the advantages of this? It seems like the content gets copied only once anyhow - if I pass by reference when I call m_namename
, if I pass by value when I pass it (and then it gets moved). I understand that this is better than passing by value and not using std::move
(because it gets copied twice).
So two questions:
- Did I understand correctly what is happening here?
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
c++
add a comment |Â
up vote
45
down vote
favorite
I'm learning C++ at the moment and try avoid picking up bad habits.
From what I understand, clang-tidy contains many "best practices" and I try to stick to them as best as possible (even though I don't necessarily understand why they are considered good yet), but I'm not sure if I understand what's recommended here.
I used this class from the tutorial:
class Creature
private:
std::string m_name;
public:
Creature(const std::string &name)
: m_namename
;
This leads to a suggestion from clang-tidy that I should pass by value instead of reference and use std::move
.
If I do, I get the suggestion to make name
a reference (to ensure it does not get copied every time) and the warning that std::move
won't have any effect because name
is a const
so I should remove it.
The only way I don't get a warning is by removing const
altogether:
Creature(std::string name)
: m_namestd::move(name)
Which seems logical, as the only benefit of const
was to prevent messing with the original string (which doesn't happen because I passed by value).
But I read on CPlusPlus.com:
Although note that -in the standard library- moving implies that the moved-from object is left in a valid but unspecified state. Which means that, after such an operation, the value of the moved-from object should only be destroyed or assigned a new value; accessing it otherwise yields an unspecified value.
Now imagine this code:
std::string nameString("Alex");
Creature c(nameString);
Because nameString
gets passed by value, std::move
will only invalidate name
inside the constructor and not touch the original string. But what are the advantages of this? It seems like the content gets copied only once anyhow - if I pass by reference when I call m_namename
, if I pass by value when I pass it (and then it gets moved). I understand that this is better than passing by value and not using std::move
(because it gets copied twice).
So two questions:
- Did I understand correctly what is happening here?
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
c++
With pass by reference,Creature c("John");
makes an extra copy
â immibis
Aug 7 at 5:46
1
This link might be a valuable read, it covers passingstd::string_view
and SSO, too.
â lubgr
Aug 7 at 7:34
add a comment |Â
up vote
45
down vote
favorite
up vote
45
down vote
favorite
I'm learning C++ at the moment and try avoid picking up bad habits.
From what I understand, clang-tidy contains many "best practices" and I try to stick to them as best as possible (even though I don't necessarily understand why they are considered good yet), but I'm not sure if I understand what's recommended here.
I used this class from the tutorial:
class Creature
private:
std::string m_name;
public:
Creature(const std::string &name)
: m_namename
;
This leads to a suggestion from clang-tidy that I should pass by value instead of reference and use std::move
.
If I do, I get the suggestion to make name
a reference (to ensure it does not get copied every time) and the warning that std::move
won't have any effect because name
is a const
so I should remove it.
The only way I don't get a warning is by removing const
altogether:
Creature(std::string name)
: m_namestd::move(name)
Which seems logical, as the only benefit of const
was to prevent messing with the original string (which doesn't happen because I passed by value).
But I read on CPlusPlus.com:
Although note that -in the standard library- moving implies that the moved-from object is left in a valid but unspecified state. Which means that, after such an operation, the value of the moved-from object should only be destroyed or assigned a new value; accessing it otherwise yields an unspecified value.
Now imagine this code:
std::string nameString("Alex");
Creature c(nameString);
Because nameString
gets passed by value, std::move
will only invalidate name
inside the constructor and not touch the original string. But what are the advantages of this? It seems like the content gets copied only once anyhow - if I pass by reference when I call m_namename
, if I pass by value when I pass it (and then it gets moved). I understand that this is better than passing by value and not using std::move
(because it gets copied twice).
So two questions:
- Did I understand correctly what is happening here?
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
c++
I'm learning C++ at the moment and try avoid picking up bad habits.
From what I understand, clang-tidy contains many "best practices" and I try to stick to them as best as possible (even though I don't necessarily understand why they are considered good yet), but I'm not sure if I understand what's recommended here.
I used this class from the tutorial:
class Creature
private:
std::string m_name;
public:
Creature(const std::string &name)
: m_namename
;
This leads to a suggestion from clang-tidy that I should pass by value instead of reference and use std::move
.
If I do, I get the suggestion to make name
a reference (to ensure it does not get copied every time) and the warning that std::move
won't have any effect because name
is a const
so I should remove it.
The only way I don't get a warning is by removing const
altogether:
Creature(std::string name)
: m_namestd::move(name)
Which seems logical, as the only benefit of const
was to prevent messing with the original string (which doesn't happen because I passed by value).
But I read on CPlusPlus.com:
Although note that -in the standard library- moving implies that the moved-from object is left in a valid but unspecified state. Which means that, after such an operation, the value of the moved-from object should only be destroyed or assigned a new value; accessing it otherwise yields an unspecified value.
Now imagine this code:
std::string nameString("Alex");
Creature c(nameString);
Because nameString
gets passed by value, std::move
will only invalidate name
inside the constructor and not touch the original string. But what are the advantages of this? It seems like the content gets copied only once anyhow - if I pass by reference when I call m_namename
, if I pass by value when I pass it (and then it gets moved). I understand that this is better than passing by value and not using std::move
(because it gets copied twice).
So two questions:
- Did I understand correctly what is happening here?
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
c++
asked Aug 6 at 10:53
Blackbot
22726
22726
With pass by reference,Creature c("John");
makes an extra copy
â immibis
Aug 7 at 5:46
1
This link might be a valuable read, it covers passingstd::string_view
and SSO, too.
â lubgr
Aug 7 at 7:34
add a comment |Â
With pass by reference,Creature c("John");
makes an extra copy
â immibis
Aug 7 at 5:46
1
This link might be a valuable read, it covers passingstd::string_view
and SSO, too.
â lubgr
Aug 7 at 7:34
With pass by reference,
Creature c("John");
makes an extra copyâ immibis
Aug 7 at 5:46
With pass by reference,
Creature c("John");
makes an extra copyâ immibis
Aug 7 at 5:46
1
1
This link might be a valuable read, it covers passing
std::string_view
and SSO, too.â lubgr
Aug 7 at 7:34
This link might be a valuable read, it covers passing
std::string_view
and SSO, too.â lubgr
Aug 7 at 7:34
add a comment |Â
4 Answers
4
active
oldest
votes
up vote
12
down vote
accepted
- Did I understand correctly what is happening here?
Yes.
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
An easy to grasp function signature without any additional overloads. The signature immediately reveals that the argument will be copied - this saves callers from wondering whether a const std::string&
reference might be stored as a data member, possibly becoming a dangling reference later on. And there is no need to overload on std::string&& name
and const std::string&
arguments to avoid unnecessary copies when rvalues are passed to the function. Passing an lvalue
std::string nameString("Alex");
Creature c(nameString);
to the function that takes its argument by value causes one copy and one move construction. Passing an rvalue to the same function
std::string nameString("Alex");
Creature c(std::move(nameString));
causes two move constructions. In contrast, when the function parameter is const std::string&
, there will always be a copy, even when passing an rvalue argument. This is clearly an advantage as long as the argument type is cheap to move-construct (this is the case for std::string
).
But there is a downside to consider: the reasoning doesn't work for functions that assign the function argument to another variable (instead of initializing it):
void setName(std::string name)
m_name = std::move(name);
will cause a deallocation of the resource that m_name
refers to before it's reassigned. I recommend reading Item 41 in Effective Modern C++ and also this question.
That makes sense, especially that this makes the declaration more intuitive to read. I'm not sure I fully grasp the deallocation part of your answer (and understand the linked thread), so just to check If I usemove
, the space gets deallocated. If I don't usemove
, it only gets deallocated if the allocated space is too small to hold the new string, leading to improved performance. Is that correct?
â Blackbot
Aug 6 at 12:36
1
Yes, that's exactly it. When assigning tom_name
from aconst std::string&
parameter, the internal memory is re-used as long asm_name
fits in. When move-assigningm_name
, the memory must be deallocated beforehand. Otherwise, it was impossible to "steal" the resources from the right hand side of the assignment.
â lubgr
Aug 6 at 12:38
add a comment |Â
up vote
39
down vote
/* (0) */
Creature(const std::string &name) : m_namename
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
name
, then is copied intom_name
.
/* (1) */
Creature(std::string name) : m_namestd::move(name)
A passed lvalue is copied into
name
, then is moved intom_name
.A passed rvalue is moved into
name
, then is moved intom_name
.
/* (2) */
Creature(const std::string &name) : m_namename
Creature(std::string &&rname) : m_namestd::move(rname)
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
rname
, then is moved intom_name
.
As move operations are usually faster than copies, (1) is better than (0) if you pass a lot of temporaries. (2) is optimal in terms of copies/moves, but requires code repetition.
The code repetition can be avoided with perfect forwarding:
/* (3) */
template <typename T,
std::enable_if_t<
std::is_convertible_v<std::remove_cvref_t<T>, std::string>,
int> = 0
>
Creature(T&& name) : m_namestd::forward<T>(name)
You might optionally want to constrain T
in order to restrict the domain of types that this constructor can be instantiated with (as shown above). C++20 aims to simplify this with Concepts.
In C++17, prvalues are affected by guaranteed copy elision, which - when applicable - will reduce the number of copies/moves when passing arguments to functions.
For (1) the pr-value and xvalue case are not identical since c++17 no?
â Oliv
Aug 6 at 11:55
1
Note that you don't need the SFINAE to perfect forward in this case. It's only needed to disambiguate. It's plausibly helpful for the potential error messages when passing bad arguments
â Caleth
Aug 6 at 15:57
@Oliv Yes. xvalues need to be moved, while prvalues can be ellided away :)
â Rakete1111
Aug 7 at 4:59
in (1), why do you need the std::move(rname)? std::move is a cast to a rvalue, rname is a xvalue; shouldn't the xvalue be moved anyway even without the explicit cast?
â Ant
Aug 7 at 9:19
@Ant: the rvalue reference itself is accessed through a name, which means it's an lvalue. You need to cast it to an rvalue to propagate its temporariness
â Vittorio Romeo
Aug 7 at 14:54
 |Â
show 2 more comments
up vote
1
down vote
How you pass is not the only variable here, what you pass makes the big difference between the two.
In C++, we have all kinds of value categories and this "idiom" exists for cases where you pass in an rvalue (such as "Alex-string-literal-that-constructs-temporary-std::string"
or std::move(nameString)
), which results in 0 copies of std::string
being made (the type does not even have to be copy-constructible for rvalue arguments), and only uses std::string
's move constructor.
Somewhat related Q&A.
add a comment |Â
up vote
1
down vote
There are several disadvantages of pass-by-value-and-move approach over pass-by-(rv)reference:
- it causes 3 objects to be spawned instead of 2;
- passing an object by value may lead to extra stack overhead, because even regular string class is typically at least 3 or 4 times larger than a pointer;
- argument objects construction is going to be done on the caller side, causing code bloat;
Could you clarify why it would cause 3 objects to spawn? From what I understand I can just pass "Peter" as a string. This would get spawned, copied and then moved, wouldn't it? And wouldn't the stack be used at some point regardless? Not at the point of the constructor call, but in them_namename
part where it gets copied?
â Blackbot
Aug 6 at 12:43
@Blackbot I was referring to your examplestd::string nameString("Alex"); Creature c(nameString);
one object isnameString
, another is function argument, and third one is a class field.
â VTT
Aug 6 at 17:57
add a comment |Â
4 Answers
4
active
oldest
votes
4 Answers
4
active
oldest
votes
active
oldest
votes
active
oldest
votes
up vote
12
down vote
accepted
- Did I understand correctly what is happening here?
Yes.
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
An easy to grasp function signature without any additional overloads. The signature immediately reveals that the argument will be copied - this saves callers from wondering whether a const std::string&
reference might be stored as a data member, possibly becoming a dangling reference later on. And there is no need to overload on std::string&& name
and const std::string&
arguments to avoid unnecessary copies when rvalues are passed to the function. Passing an lvalue
std::string nameString("Alex");
Creature c(nameString);
to the function that takes its argument by value causes one copy and one move construction. Passing an rvalue to the same function
std::string nameString("Alex");
Creature c(std::move(nameString));
causes two move constructions. In contrast, when the function parameter is const std::string&
, there will always be a copy, even when passing an rvalue argument. This is clearly an advantage as long as the argument type is cheap to move-construct (this is the case for std::string
).
But there is a downside to consider: the reasoning doesn't work for functions that assign the function argument to another variable (instead of initializing it):
void setName(std::string name)
m_name = std::move(name);
will cause a deallocation of the resource that m_name
refers to before it's reassigned. I recommend reading Item 41 in Effective Modern C++ and also this question.
That makes sense, especially that this makes the declaration more intuitive to read. I'm not sure I fully grasp the deallocation part of your answer (and understand the linked thread), so just to check If I usemove
, the space gets deallocated. If I don't usemove
, it only gets deallocated if the allocated space is too small to hold the new string, leading to improved performance. Is that correct?
â Blackbot
Aug 6 at 12:36
1
Yes, that's exactly it. When assigning tom_name
from aconst std::string&
parameter, the internal memory is re-used as long asm_name
fits in. When move-assigningm_name
, the memory must be deallocated beforehand. Otherwise, it was impossible to "steal" the resources from the right hand side of the assignment.
â lubgr
Aug 6 at 12:38
add a comment |Â
up vote
12
down vote
accepted
- Did I understand correctly what is happening here?
Yes.
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
An easy to grasp function signature without any additional overloads. The signature immediately reveals that the argument will be copied - this saves callers from wondering whether a const std::string&
reference might be stored as a data member, possibly becoming a dangling reference later on. And there is no need to overload on std::string&& name
and const std::string&
arguments to avoid unnecessary copies when rvalues are passed to the function. Passing an lvalue
std::string nameString("Alex");
Creature c(nameString);
to the function that takes its argument by value causes one copy and one move construction. Passing an rvalue to the same function
std::string nameString("Alex");
Creature c(std::move(nameString));
causes two move constructions. In contrast, when the function parameter is const std::string&
, there will always be a copy, even when passing an rvalue argument. This is clearly an advantage as long as the argument type is cheap to move-construct (this is the case for std::string
).
But there is a downside to consider: the reasoning doesn't work for functions that assign the function argument to another variable (instead of initializing it):
void setName(std::string name)
m_name = std::move(name);
will cause a deallocation of the resource that m_name
refers to before it's reassigned. I recommend reading Item 41 in Effective Modern C++ and also this question.
That makes sense, especially that this makes the declaration more intuitive to read. I'm not sure I fully grasp the deallocation part of your answer (and understand the linked thread), so just to check If I usemove
, the space gets deallocated. If I don't usemove
, it only gets deallocated if the allocated space is too small to hold the new string, leading to improved performance. Is that correct?
â Blackbot
Aug 6 at 12:36
1
Yes, that's exactly it. When assigning tom_name
from aconst std::string&
parameter, the internal memory is re-used as long asm_name
fits in. When move-assigningm_name
, the memory must be deallocated beforehand. Otherwise, it was impossible to "steal" the resources from the right hand side of the assignment.
â lubgr
Aug 6 at 12:38
add a comment |Â
up vote
12
down vote
accepted
up vote
12
down vote
accepted
- Did I understand correctly what is happening here?
Yes.
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
An easy to grasp function signature without any additional overloads. The signature immediately reveals that the argument will be copied - this saves callers from wondering whether a const std::string&
reference might be stored as a data member, possibly becoming a dangling reference later on. And there is no need to overload on std::string&& name
and const std::string&
arguments to avoid unnecessary copies when rvalues are passed to the function. Passing an lvalue
std::string nameString("Alex");
Creature c(nameString);
to the function that takes its argument by value causes one copy and one move construction. Passing an rvalue to the same function
std::string nameString("Alex");
Creature c(std::move(nameString));
causes two move constructions. In contrast, when the function parameter is const std::string&
, there will always be a copy, even when passing an rvalue argument. This is clearly an advantage as long as the argument type is cheap to move-construct (this is the case for std::string
).
But there is a downside to consider: the reasoning doesn't work for functions that assign the function argument to another variable (instead of initializing it):
void setName(std::string name)
m_name = std::move(name);
will cause a deallocation of the resource that m_name
refers to before it's reassigned. I recommend reading Item 41 in Effective Modern C++ and also this question.
- Did I understand correctly what is happening here?
Yes.
- Is there any upside of using
std::move
over passing by reference and just callingm_namename
?
An easy to grasp function signature without any additional overloads. The signature immediately reveals that the argument will be copied - this saves callers from wondering whether a const std::string&
reference might be stored as a data member, possibly becoming a dangling reference later on. And there is no need to overload on std::string&& name
and const std::string&
arguments to avoid unnecessary copies when rvalues are passed to the function. Passing an lvalue
std::string nameString("Alex");
Creature c(nameString);
to the function that takes its argument by value causes one copy and one move construction. Passing an rvalue to the same function
std::string nameString("Alex");
Creature c(std::move(nameString));
causes two move constructions. In contrast, when the function parameter is const std::string&
, there will always be a copy, even when passing an rvalue argument. This is clearly an advantage as long as the argument type is cheap to move-construct (this is the case for std::string
).
But there is a downside to consider: the reasoning doesn't work for functions that assign the function argument to another variable (instead of initializing it):
void setName(std::string name)
m_name = std::move(name);
will cause a deallocation of the resource that m_name
refers to before it's reassigned. I recommend reading Item 41 in Effective Modern C++ and also this question.
edited Aug 6 at 11:17
answered Aug 6 at 11:12
lubgr
4,606732
4,606732
That makes sense, especially that this makes the declaration more intuitive to read. I'm not sure I fully grasp the deallocation part of your answer (and understand the linked thread), so just to check If I usemove
, the space gets deallocated. If I don't usemove
, it only gets deallocated if the allocated space is too small to hold the new string, leading to improved performance. Is that correct?
â Blackbot
Aug 6 at 12:36
1
Yes, that's exactly it. When assigning tom_name
from aconst std::string&
parameter, the internal memory is re-used as long asm_name
fits in. When move-assigningm_name
, the memory must be deallocated beforehand. Otherwise, it was impossible to "steal" the resources from the right hand side of the assignment.
â lubgr
Aug 6 at 12:38
add a comment |Â
That makes sense, especially that this makes the declaration more intuitive to read. I'm not sure I fully grasp the deallocation part of your answer (and understand the linked thread), so just to check If I usemove
, the space gets deallocated. If I don't usemove
, it only gets deallocated if the allocated space is too small to hold the new string, leading to improved performance. Is that correct?
â Blackbot
Aug 6 at 12:36
1
Yes, that's exactly it. When assigning tom_name
from aconst std::string&
parameter, the internal memory is re-used as long asm_name
fits in. When move-assigningm_name
, the memory must be deallocated beforehand. Otherwise, it was impossible to "steal" the resources from the right hand side of the assignment.
â lubgr
Aug 6 at 12:38
That makes sense, especially that this makes the declaration more intuitive to read. I'm not sure I fully grasp the deallocation part of your answer (and understand the linked thread), so just to check If I use
move
, the space gets deallocated. If I don't use move
, it only gets deallocated if the allocated space is too small to hold the new string, leading to improved performance. Is that correct?â Blackbot
Aug 6 at 12:36
That makes sense, especially that this makes the declaration more intuitive to read. I'm not sure I fully grasp the deallocation part of your answer (and understand the linked thread), so just to check If I use
move
, the space gets deallocated. If I don't use move
, it only gets deallocated if the allocated space is too small to hold the new string, leading to improved performance. Is that correct?â Blackbot
Aug 6 at 12:36
1
1
Yes, that's exactly it. When assigning to
m_name
from a const std::string&
parameter, the internal memory is re-used as long as m_name
fits in. When move-assigning m_name
, the memory must be deallocated beforehand. Otherwise, it was impossible to "steal" the resources from the right hand side of the assignment.â lubgr
Aug 6 at 12:38
Yes, that's exactly it. When assigning to
m_name
from a const std::string&
parameter, the internal memory is re-used as long as m_name
fits in. When move-assigning m_name
, the memory must be deallocated beforehand. Otherwise, it was impossible to "steal" the resources from the right hand side of the assignment.â lubgr
Aug 6 at 12:38
add a comment |Â
up vote
39
down vote
/* (0) */
Creature(const std::string &name) : m_namename
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
name
, then is copied intom_name
.
/* (1) */
Creature(std::string name) : m_namestd::move(name)
A passed lvalue is copied into
name
, then is moved intom_name
.A passed rvalue is moved into
name
, then is moved intom_name
.
/* (2) */
Creature(const std::string &name) : m_namename
Creature(std::string &&rname) : m_namestd::move(rname)
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
rname
, then is moved intom_name
.
As move operations are usually faster than copies, (1) is better than (0) if you pass a lot of temporaries. (2) is optimal in terms of copies/moves, but requires code repetition.
The code repetition can be avoided with perfect forwarding:
/* (3) */
template <typename T,
std::enable_if_t<
std::is_convertible_v<std::remove_cvref_t<T>, std::string>,
int> = 0
>
Creature(T&& name) : m_namestd::forward<T>(name)
You might optionally want to constrain T
in order to restrict the domain of types that this constructor can be instantiated with (as shown above). C++20 aims to simplify this with Concepts.
In C++17, prvalues are affected by guaranteed copy elision, which - when applicable - will reduce the number of copies/moves when passing arguments to functions.
For (1) the pr-value and xvalue case are not identical since c++17 no?
â Oliv
Aug 6 at 11:55
1
Note that you don't need the SFINAE to perfect forward in this case. It's only needed to disambiguate. It's plausibly helpful for the potential error messages when passing bad arguments
â Caleth
Aug 6 at 15:57
@Oliv Yes. xvalues need to be moved, while prvalues can be ellided away :)
â Rakete1111
Aug 7 at 4:59
in (1), why do you need the std::move(rname)? std::move is a cast to a rvalue, rname is a xvalue; shouldn't the xvalue be moved anyway even without the explicit cast?
â Ant
Aug 7 at 9:19
@Ant: the rvalue reference itself is accessed through a name, which means it's an lvalue. You need to cast it to an rvalue to propagate its temporariness
â Vittorio Romeo
Aug 7 at 14:54
 |Â
show 2 more comments
up vote
39
down vote
/* (0) */
Creature(const std::string &name) : m_namename
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
name
, then is copied intom_name
.
/* (1) */
Creature(std::string name) : m_namestd::move(name)
A passed lvalue is copied into
name
, then is moved intom_name
.A passed rvalue is moved into
name
, then is moved intom_name
.
/* (2) */
Creature(const std::string &name) : m_namename
Creature(std::string &&rname) : m_namestd::move(rname)
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
rname
, then is moved intom_name
.
As move operations are usually faster than copies, (1) is better than (0) if you pass a lot of temporaries. (2) is optimal in terms of copies/moves, but requires code repetition.
The code repetition can be avoided with perfect forwarding:
/* (3) */
template <typename T,
std::enable_if_t<
std::is_convertible_v<std::remove_cvref_t<T>, std::string>,
int> = 0
>
Creature(T&& name) : m_namestd::forward<T>(name)
You might optionally want to constrain T
in order to restrict the domain of types that this constructor can be instantiated with (as shown above). C++20 aims to simplify this with Concepts.
In C++17, prvalues are affected by guaranteed copy elision, which - when applicable - will reduce the number of copies/moves when passing arguments to functions.
For (1) the pr-value and xvalue case are not identical since c++17 no?
â Oliv
Aug 6 at 11:55
1
Note that you don't need the SFINAE to perfect forward in this case. It's only needed to disambiguate. It's plausibly helpful for the potential error messages when passing bad arguments
â Caleth
Aug 6 at 15:57
@Oliv Yes. xvalues need to be moved, while prvalues can be ellided away :)
â Rakete1111
Aug 7 at 4:59
in (1), why do you need the std::move(rname)? std::move is a cast to a rvalue, rname is a xvalue; shouldn't the xvalue be moved anyway even without the explicit cast?
â Ant
Aug 7 at 9:19
@Ant: the rvalue reference itself is accessed through a name, which means it's an lvalue. You need to cast it to an rvalue to propagate its temporariness
â Vittorio Romeo
Aug 7 at 14:54
 |Â
show 2 more comments
up vote
39
down vote
up vote
39
down vote
/* (0) */
Creature(const std::string &name) : m_namename
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
name
, then is copied intom_name
.
/* (1) */
Creature(std::string name) : m_namestd::move(name)
A passed lvalue is copied into
name
, then is moved intom_name
.A passed rvalue is moved into
name
, then is moved intom_name
.
/* (2) */
Creature(const std::string &name) : m_namename
Creature(std::string &&rname) : m_namestd::move(rname)
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
rname
, then is moved intom_name
.
As move operations are usually faster than copies, (1) is better than (0) if you pass a lot of temporaries. (2) is optimal in terms of copies/moves, but requires code repetition.
The code repetition can be avoided with perfect forwarding:
/* (3) */
template <typename T,
std::enable_if_t<
std::is_convertible_v<std::remove_cvref_t<T>, std::string>,
int> = 0
>
Creature(T&& name) : m_namestd::forward<T>(name)
You might optionally want to constrain T
in order to restrict the domain of types that this constructor can be instantiated with (as shown above). C++20 aims to simplify this with Concepts.
In C++17, prvalues are affected by guaranteed copy elision, which - when applicable - will reduce the number of copies/moves when passing arguments to functions.
/* (0) */
Creature(const std::string &name) : m_namename
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
name
, then is copied intom_name
.
/* (1) */
Creature(std::string name) : m_namestd::move(name)
A passed lvalue is copied into
name
, then is moved intom_name
.A passed rvalue is moved into
name
, then is moved intom_name
.
/* (2) */
Creature(const std::string &name) : m_namename
Creature(std::string &&rname) : m_namestd::move(rname)
A passed lvalue binds to
name
, then is copied intom_name
.A passed rvalue binds to
rname
, then is moved intom_name
.
As move operations are usually faster than copies, (1) is better than (0) if you pass a lot of temporaries. (2) is optimal in terms of copies/moves, but requires code repetition.
The code repetition can be avoided with perfect forwarding:
/* (3) */
template <typename T,
std::enable_if_t<
std::is_convertible_v<std::remove_cvref_t<T>, std::string>,
int> = 0
>
Creature(T&& name) : m_namestd::forward<T>(name)
You might optionally want to constrain T
in order to restrict the domain of types that this constructor can be instantiated with (as shown above). C++20 aims to simplify this with Concepts.
In C++17, prvalues are affected by guaranteed copy elision, which - when applicable - will reduce the number of copies/moves when passing arguments to functions.
edited Aug 6 at 19:07
answered Aug 6 at 11:25
Vittorio Romeo
49.9k14134272
49.9k14134272
For (1) the pr-value and xvalue case are not identical since c++17 no?
â Oliv
Aug 6 at 11:55
1
Note that you don't need the SFINAE to perfect forward in this case. It's only needed to disambiguate. It's plausibly helpful for the potential error messages when passing bad arguments
â Caleth
Aug 6 at 15:57
@Oliv Yes. xvalues need to be moved, while prvalues can be ellided away :)
â Rakete1111
Aug 7 at 4:59
in (1), why do you need the std::move(rname)? std::move is a cast to a rvalue, rname is a xvalue; shouldn't the xvalue be moved anyway even without the explicit cast?
â Ant
Aug 7 at 9:19
@Ant: the rvalue reference itself is accessed through a name, which means it's an lvalue. You need to cast it to an rvalue to propagate its temporariness
â Vittorio Romeo
Aug 7 at 14:54
 |Â
show 2 more comments
For (1) the pr-value and xvalue case are not identical since c++17 no?
â Oliv
Aug 6 at 11:55
1
Note that you don't need the SFINAE to perfect forward in this case. It's only needed to disambiguate. It's plausibly helpful for the potential error messages when passing bad arguments
â Caleth
Aug 6 at 15:57
@Oliv Yes. xvalues need to be moved, while prvalues can be ellided away :)
â Rakete1111
Aug 7 at 4:59
in (1), why do you need the std::move(rname)? std::move is a cast to a rvalue, rname is a xvalue; shouldn't the xvalue be moved anyway even without the explicit cast?
â Ant
Aug 7 at 9:19
@Ant: the rvalue reference itself is accessed through a name, which means it's an lvalue. You need to cast it to an rvalue to propagate its temporariness
â Vittorio Romeo
Aug 7 at 14:54
For (1) the pr-value and xvalue case are not identical since c++17 no?
â Oliv
Aug 6 at 11:55
For (1) the pr-value and xvalue case are not identical since c++17 no?
â Oliv
Aug 6 at 11:55
1
1
Note that you don't need the SFINAE to perfect forward in this case. It's only needed to disambiguate. It's plausibly helpful for the potential error messages when passing bad arguments
â Caleth
Aug 6 at 15:57
Note that you don't need the SFINAE to perfect forward in this case. It's only needed to disambiguate. It's plausibly helpful for the potential error messages when passing bad arguments
â Caleth
Aug 6 at 15:57
@Oliv Yes. xvalues need to be moved, while prvalues can be ellided away :)
â Rakete1111
Aug 7 at 4:59
@Oliv Yes. xvalues need to be moved, while prvalues can be ellided away :)
â Rakete1111
Aug 7 at 4:59
in (1), why do you need the std::move(rname)? std::move is a cast to a rvalue, rname is a xvalue; shouldn't the xvalue be moved anyway even without the explicit cast?
â Ant
Aug 7 at 9:19
in (1), why do you need the std::move(rname)? std::move is a cast to a rvalue, rname is a xvalue; shouldn't the xvalue be moved anyway even without the explicit cast?
â Ant
Aug 7 at 9:19
@Ant: the rvalue reference itself is accessed through a name, which means it's an lvalue. You need to cast it to an rvalue to propagate its temporariness
â Vittorio Romeo
Aug 7 at 14:54
@Ant: the rvalue reference itself is accessed through a name, which means it's an lvalue. You need to cast it to an rvalue to propagate its temporariness
â Vittorio Romeo
Aug 7 at 14:54
 |Â
show 2 more comments
up vote
1
down vote
How you pass is not the only variable here, what you pass makes the big difference between the two.
In C++, we have all kinds of value categories and this "idiom" exists for cases where you pass in an rvalue (such as "Alex-string-literal-that-constructs-temporary-std::string"
or std::move(nameString)
), which results in 0 copies of std::string
being made (the type does not even have to be copy-constructible for rvalue arguments), and only uses std::string
's move constructor.
Somewhat related Q&A.
add a comment |Â
up vote
1
down vote
How you pass is not the only variable here, what you pass makes the big difference between the two.
In C++, we have all kinds of value categories and this "idiom" exists for cases where you pass in an rvalue (such as "Alex-string-literal-that-constructs-temporary-std::string"
or std::move(nameString)
), which results in 0 copies of std::string
being made (the type does not even have to be copy-constructible for rvalue arguments), and only uses std::string
's move constructor.
Somewhat related Q&A.
add a comment |Â
up vote
1
down vote
up vote
1
down vote
How you pass is not the only variable here, what you pass makes the big difference between the two.
In C++, we have all kinds of value categories and this "idiom" exists for cases where you pass in an rvalue (such as "Alex-string-literal-that-constructs-temporary-std::string"
or std::move(nameString)
), which results in 0 copies of std::string
being made (the type does not even have to be copy-constructible for rvalue arguments), and only uses std::string
's move constructor.
Somewhat related Q&A.
How you pass is not the only variable here, what you pass makes the big difference between the two.
In C++, we have all kinds of value categories and this "idiom" exists for cases where you pass in an rvalue (such as "Alex-string-literal-that-constructs-temporary-std::string"
or std::move(nameString)
), which results in 0 copies of std::string
being made (the type does not even have to be copy-constructible for rvalue arguments), and only uses std::string
's move constructor.
Somewhat related Q&A.
edited Aug 6 at 11:22
answered Aug 6 at 11:12
LogicStuff
15.4k63656
15.4k63656
add a comment |Â
add a comment |Â
up vote
1
down vote
There are several disadvantages of pass-by-value-and-move approach over pass-by-(rv)reference:
- it causes 3 objects to be spawned instead of 2;
- passing an object by value may lead to extra stack overhead, because even regular string class is typically at least 3 or 4 times larger than a pointer;
- argument objects construction is going to be done on the caller side, causing code bloat;
Could you clarify why it would cause 3 objects to spawn? From what I understand I can just pass "Peter" as a string. This would get spawned, copied and then moved, wouldn't it? And wouldn't the stack be used at some point regardless? Not at the point of the constructor call, but in them_namename
part where it gets copied?
â Blackbot
Aug 6 at 12:43
@Blackbot I was referring to your examplestd::string nameString("Alex"); Creature c(nameString);
one object isnameString
, another is function argument, and third one is a class field.
â VTT
Aug 6 at 17:57
add a comment |Â
up vote
1
down vote
There are several disadvantages of pass-by-value-and-move approach over pass-by-(rv)reference:
- it causes 3 objects to be spawned instead of 2;
- passing an object by value may lead to extra stack overhead, because even regular string class is typically at least 3 or 4 times larger than a pointer;
- argument objects construction is going to be done on the caller side, causing code bloat;
Could you clarify why it would cause 3 objects to spawn? From what I understand I can just pass "Peter" as a string. This would get spawned, copied and then moved, wouldn't it? And wouldn't the stack be used at some point regardless? Not at the point of the constructor call, but in them_namename
part where it gets copied?
â Blackbot
Aug 6 at 12:43
@Blackbot I was referring to your examplestd::string nameString("Alex"); Creature c(nameString);
one object isnameString
, another is function argument, and third one is a class field.
â VTT
Aug 6 at 17:57
add a comment |Â
up vote
1
down vote
up vote
1
down vote
There are several disadvantages of pass-by-value-and-move approach over pass-by-(rv)reference:
- it causes 3 objects to be spawned instead of 2;
- passing an object by value may lead to extra stack overhead, because even regular string class is typically at least 3 or 4 times larger than a pointer;
- argument objects construction is going to be done on the caller side, causing code bloat;
There are several disadvantages of pass-by-value-and-move approach over pass-by-(rv)reference:
- it causes 3 objects to be spawned instead of 2;
- passing an object by value may lead to extra stack overhead, because even regular string class is typically at least 3 or 4 times larger than a pointer;
- argument objects construction is going to be done on the caller side, causing code bloat;
answered Aug 6 at 11:36
VTT
20.8k32143
20.8k32143
Could you clarify why it would cause 3 objects to spawn? From what I understand I can just pass "Peter" as a string. This would get spawned, copied and then moved, wouldn't it? And wouldn't the stack be used at some point regardless? Not at the point of the constructor call, but in them_namename
part where it gets copied?
â Blackbot
Aug 6 at 12:43
@Blackbot I was referring to your examplestd::string nameString("Alex"); Creature c(nameString);
one object isnameString
, another is function argument, and third one is a class field.
â VTT
Aug 6 at 17:57
add a comment |Â
Could you clarify why it would cause 3 objects to spawn? From what I understand I can just pass "Peter" as a string. This would get spawned, copied and then moved, wouldn't it? And wouldn't the stack be used at some point regardless? Not at the point of the constructor call, but in them_namename
part where it gets copied?
â Blackbot
Aug 6 at 12:43
@Blackbot I was referring to your examplestd::string nameString("Alex"); Creature c(nameString);
one object isnameString
, another is function argument, and third one is a class field.
â VTT
Aug 6 at 17:57
Could you clarify why it would cause 3 objects to spawn? From what I understand I can just pass "Peter" as a string. This would get spawned, copied and then moved, wouldn't it? And wouldn't the stack be used at some point regardless? Not at the point of the constructor call, but in the
m_namename
part where it gets copied?â Blackbot
Aug 6 at 12:43
Could you clarify why it would cause 3 objects to spawn? From what I understand I can just pass "Peter" as a string. This would get spawned, copied and then moved, wouldn't it? And wouldn't the stack be used at some point regardless? Not at the point of the constructor call, but in the
m_namename
part where it gets copied?â Blackbot
Aug 6 at 12:43
@Blackbot I was referring to your example
std::string nameString("Alex"); Creature c(nameString);
one object is nameString
, another is function argument, and third one is a class field.â VTT
Aug 6 at 17:57
@Blackbot I was referring to your example
std::string nameString("Alex"); Creature c(nameString);
one object is nameString
, another is function argument, and third one is a class field.â VTT
Aug 6 at 17:57
add a comment |Â
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f51705967%2fadvantages-of-pass-by-value-and-stdmove-over-pass-by-reference%23new-answer', 'question_page');
);
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Sign up or log in
StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
With pass by reference,
Creature c("John");
makes an extra copyâ immibis
Aug 7 at 5:46
1
This link might be a valuable read, it covers passing
std::string_view
and SSO, too.â lubgr
Aug 7 at 7:34