Unneeded copy of string

[LemonParty]

Hello.

I have this simple code:

Code: Pascal  [Select][+][-]

1. {$mode objfpc}{$H+}
2.  
3. procedure P1(constref Ch: AnsiChar);
4. begin
5.   Writeln(Ch);
6. end;
7.  
8. var
9.   S: AnsiString = '12345';
10.  
11. begin
12.   P1(S[1]);
13. end.
14.  

When I disassebled this code I found this:

Code: ASM  [Select][+][-]

1. # [32] P1(S[1]);
2.   leaq  TC_$P$PROGRAM_$$_S(%rip),%rcx
3.   call  fpc_ansistr_unique
4.   movq  %rax,%rcx
5.   call  P$PROGRAM_$$_P1$CHAR
6.  

The question is why compiler make copy of a string?

[PascalDragon]

The question is why compiler make copy of a string?

constref enforces the passing of Ch as a reference aka a Pointer. To ensure that another potential reference to the string doesn't interfere with that passed in reference the compiler ensures a unique copy of the string. For a primitive argument type const is better. In general you should prefer to use const because the compiler will pick the most efficient way to pass the parameter - unless you really need a reference to the passed in value.

[LemonParty]

Code: Pascal  [Select][+][-]

1. {$mode objfpc}{$H+}
2.  
3. procedure P1(constref Ch: AnsiChar);
4. begin
5.   Writeln('P1 Char');
6. end;
7.  
8. procedure P1(constref Ch: WideChar);
9. begin
10.   Writeln('P1 Wide');
11. end;
12.  
13. procedure P2(P: PAnsiChar);
14. begin
15.   Writeln('P2 Char');
16. end;
17.  
18. procedure P2(P: PWideChar);
19. begin
20.   Writeln('P2 Wide');
21. end;
22.  
23. var
24.   S: AnsiString = '12345';
25.   W: WideString = '12345';
26.  
27. begin
28.   P1(S[1]);
29.   P1(W[1]);
30.   P2(@S[1]);
31.   P2(@W[1]);
32.   Writeln('1');
33. end.
34.  

This code demonstrate a problem when use something else than constref. It would be nice if compiler have an option to turn off such hidden copying.

[Thaddy]

well, then make it const as PascalDragon suggested.

[LemonParty]

I need a reference.

[Thaddy]

PascalDragon explained that in that case you can't avoid a copy.
That is related to the type being a managed type.

[Khrys]

Code: Pascal  [Select][+][-]

1. {$mode objfpc}{$H+}
2.  
3. procedure P1(constref Ch: AnsiChar);
4. begin
5.   Writeln('P1 Char');
6. end;
7.  
8. procedure P1(constref Ch: WideChar);
9. begin
10.   Writeln('P1 Wide');
11. end;
12.  
13. procedure P2(P: PAnsiChar);
14. begin
15.   Writeln('P2 Char');
16. end;
17.  
18. procedure P2(P: PWideChar);
19. begin
20.   Writeln('P2 Wide');
21. end;
22.  
23. var
24.   S: AnsiString = '12345';
25.   W: WideString = '12345';
26.  
27. begin
28.   P1(S[1]);
29.   P1(W[1]);
30.   P2(@S[1]);
31.   P2(@W[1]);
32.   Writeln('1');
33. end.
34.  

This code demonstrate a problem when use something else than constref. It would be nice if compiler have an option to turn off such hidden copying.

This can be solved by using  {$TYPEDADDRESS ON}  aka  {$T+}

[Thaddy]

Did you check the assembler output?

[Khrys]

Did you check the assembler output?

I assume the problem @LemonParty was referring to is the compiler selecting the "wrong" overload in line 31 (PWideChar  is supplied, but under  {$T-}  the compiler chooses the  PAnsiChar  version). It's not about string copying.

The output should be  P1 Char, P1 Wide, P2 Char, P2 Wide  but without  {$T+}  it's actually  P1 Char, P1 Wide, P2 Char, P2 Char.
With constref the @ operator could be avoided, preventing loss of type information in that case and allowing the compiler to select the correct overload.

@LemonParty is this the actual problem you were trying to solve? I've answered a similiar question about overloads before (that you posted).

[Thaddy]

In this case we need the assembler output.
Will check when I am home.

[LemonParty]

This can be solved by using  {$TYPEDADDRESS ON}  aka  {$T+}

Is this going to work when I place procedures in separate unit or whole project should be compiled with this switch?

[beria2]

I need a reference.

it is the link that is needed - use only the Pointer type. I always do that. And you always know exactly what you have.

[ASerge]

This can be solved by using  {$TYPEDADDRESS ON}  aka  {$T+}

Is this going to work when I place procedures in separate unit or whole project should be compiled with this switch?

A separate unit is enough.
Interestingly, the discussion turned to another issue, not the one stated in the topic.

[Khrys]

This can be solved by using  {$TYPEDADDRESS ON}  aka  {$T+}

Is this going to work when I place procedures in separate unit or whole project should be compiled with this switch?

It's a local directive, meaning that it can be toggled at the token level.
You could even enable it just for a single part of an expression if you really wanted to:

Code: Pascal  [Select][+][-]

1. P2({$push}{$T+} @W[1] {$pop});

Interestingly, the discussion turned to another issue, not the one stated in the topic.

Turns out it was an XY problem after all!   

[LemonParty]

I have built a test program
u1.pas:

Code: Pascal  [Select][+][-]

1. {$mode objfpc}{$H+}
2.  
3. {$T+}
4.  
5. unit u1;
6.  
7. interface
8.  
9. procedure P2(P: PAnsiChar);
10. procedure P2(P: PWideChar);
11.  
12. implementation
13.  
14. procedure P2(P: PAnsiChar);
15. begin
16.   Writeln('P2 Char');
17. end;
18.  
19. procedure P2(P: PWideChar);
20. begin
21.   Writeln('P2 Wide');
22. end;
23.  
24. end.

prog.pas:

Code: Pascal  [Select][+][-]

1. {$mode objfpc}{$H+}
2.  
3. uses u1;
4.  
5. var
6.   S: AnsiString = '12345';
7.   W: WideString = '12345';
8.  
9. begin
10.   {$Push}{$T+}
11.   P2(@S[1]);
12.   P2(@W[1]);
13.   {$Pop}
14.   {$Push}{$T-}
15.   P2(@S[1]);
16.   P2(@W[1]);
17.   {$Pop}
18. end.

Output is:

P2 Char
P2 Wide
P2 Char
P2 Char

This mean all project should be built with {T+} option. This also mean we can't use overloaded functions in units that rely on strings.
I think behavior of compiler around constref should be changed. As I suppouse idea of constref is that we reference something, but not change it. So compiler should not do copy of a string on constref.