For instance, the 8086 had a REP MOVS instruction that copies a block of bytes, the TP standard library used that for the copy data function. 80186 added REP MOVSW which would move 2 bytes at a time and be about two times faster on my 80286 machine. I had several replacements for library subroutines which took advantage of 80286 instructions to get much better performance.
procedure FillWord(var Dest,Count,Data: word);
begin
inline(
$C4/$BE/Dest/ { LES DI,Dest[BP] }
$8B/$8E/Count/ { MOV CX,Count[BP] }
$8B/$86/Data/ { MOV AX,Data[BP] }
$FC/ { CLD }
$F3/$AB); { REP STOSW }
end;