FPU instructions that verify accuracy

Question

Using the fldcw command, you can change the accuracy of an FPU to 24 or more bits. However, after performing some tests, I begin to think that very few x87 operations actually use this option.

I have not tested all the operations, but on this test machine it still looks like fdiv and fsqrt stop the calculation with the selected accuracy and that all the other operations (fadd fsub fmul ...) always calculate the full extended accuracy.

If that were the case, I would expect it because these 2 commands (fdiv and fsqrt) are much slower than most other x87 FPU commands, so when lower accuracy is sufficient, you can speed them up, but in fact, m is just interesting whether this was always the case, or if it was a fad of the very last processor used in my test machine.

edit: here is the delphi code to show it

program Project1;

uses
  windows,dialogs,sysutils;

{$R *.res}

const
 test_mul:single=1234567890.0987654321;

var
 i:longint;
 s:single absolute i;
 s1,s2,s3:single;

procedure test_24;
asm
 mov word([esp-2]),$103f  // 24bit precision, trunc
 fldcw word([esp-2])

 fld [s]
 fmul [test_mul]
 fstp [s1]
end;

procedure test_53;
asm
 mov word([esp-2]),$123f  // 53bit precision, trunc
 fldcw word([esp-2])

 fld [s]
 fmul [test_mul]
 fstp [s2]
end;

procedure test_64;
asm
 mov word([esp-2]),$133f  // 64bit precision, trunc
 fldcw word([esp-2])

 fld [s]
 fmul [test_mul]
 fstp [s3]
end;

begin
 i:=0;
 repeat
  test_24;
  test_53;
  test_64;

  if (s1<>s2) or (s2<>s3) then begin
   showmessage('Error at step:'+inttostr(i));
   break;
  end;

  inc(i);
 until i=0;
 showmessage('No difference found between precisions');
end.

edit2: false alarm, I made a mistake, I saved as a single instead of an extended one, so I could not understand the difference, here is a fixed test, thanks to hans passant in order to catch my error:

program Project1;

uses
  windows,dialogs,sysutils;

{$R *.res}

const
 test_mul:single=1234567890.0987654321;

var
 i:longint;
 errors:cardinal;
 s:single absolute i;
 s1,s2,s3:extended;

procedure test_24;
asm
 mov word([esp-2]),$103f  // 24bit precision, trunc
 fldcw word([esp-2])

 fld [s]
 fmul [test_mul]
 fstp [s1]
end;

procedure test_53;
asm
 mov word([esp-2]),$123f  // 53bit precision, trunc
 fldcw word([esp-2])

 fld [s]
 fmul [test_mul]
 fstp [s2]
end;

procedure test_64;
asm
 mov word([esp-2]),$133f  // 64bit precision, trunc
 fldcw word([esp-2])

 fld [s]
 fmul [test_mul]
 fstp [s3]
end;

begin
 errors:=0;
 i:=0;
 repeat
  test_24;
  test_53;
  test_64;

  if (s1<>s2) or (s2<>s3) then begin
   inc(errors);
  end;

  inc(i);
 until i=0;
 showmessage('Number of differences between precisions: '+inttostr(errors));
end.