x64 SwapEndian(<16bit>) via ROL is faster?

[AlexTP]

This page suggests to do SwapEndian via ROL
https://stackoverflow.com/questions/3065335/how-to-convert-big-endian-numbers-to-native-numbers-delphi

Code: Pascal  [Select][+][-]

1. function SwapEndian16(Value: smallint): smallint; register;
2. asm
3.   rol   ax, 8
4. end;
5.  

and we now have this in  rtl/x86_64/x86_64.inc

Code: Pascal  [Select][+][-]

1. { SwapEndian(<16 Bit>) being inlined is faster than using assembler }
2. function SwapEndian(const AValue: SmallInt): SmallInt;{$ifdef SYSTEMINLINE}inline;{$endif}
3.   begin
4.     { the extra Word type cast is necessary because the "AValue shr 8" }
5.     { is turned into "longint(AValue) shr 8", so if AValue < 0 then    }
6.     { the sign bits from the upper 16 bits are shifted in rather than  }
7.     { zeroes.                                                          }
8.     Result := SmallInt(((Word(AValue) shr 8) or (Word(AValue) shl 8)) and $ffff);
9.   end;
10.  
11.  
12. function SwapEndian(const AValue: Word): Word;{$ifdef SYSTEMINLINE}inline;{$endif}
13.   begin
14.     Result := ((AValue shr 8) or (AValue shl 8)) and $ffff;
15.   end;
16.  

Which is faster?

[birin]

If SwapEndian16 are not inlined, then she needs CALL and RET.
Instruction flow interruption is time consuming.

I think the second version of SwapEndian doesn't need the and $ffff because all values are Word, and Word are unsigned.