mirror of
https://gitlab.com/freepascal.org/fpc/source.git
synced 2025-04-12 14:29:34 +02:00
56 lines
1.5 KiB
ObjectPascal
56 lines
1.5 KiB
ObjectPascal
{ %version=1.1 }
|
|
{ Source provided for Free Pascal Bug Report 2129 }
|
|
{ Submitted by "Bill Rayer" on 2002-09-18 }
|
|
{ e-mail: lingolanguage@hotmail.com }
|
|
|
|
{$mode delphi}
|
|
|
|
(*
|
|
Comp() cast has different effect in FPC.
|
|
Compiles using Delphi4:
|
|
dcc32 -CC fpc19
|
|
Compiles in FPC 1.0.6:
|
|
ppc386 -WC fpc19
|
|
|
|
When run, the Delphi version shows -6.5E+18, but the FPC version
|
|
shows zero. In Delphi, the comp() cast actually moves 8 bytes from
|
|
the double into the comp without converting the data, but FPC uses
|
|
floating point instructions to convert the data and therefore prints
|
|
zero.
|
|
|
|
In Delphi, if you want to convert a double to a float, you just use
|
|
the assignment "comp1 := dbl1" which corresponds to the FLD/FIST
|
|
opcodes. FPC should not use the comp() cast for doing this, since
|
|
it introduces a subtle incompatibility with Delphi.
|
|
*)
|
|
|
|
program fpc19;
|
|
uses
|
|
sysutils;
|
|
|
|
var
|
|
comp1 : comp;
|
|
dbl1 : double;
|
|
s : string;
|
|
begin
|
|
dbl1 := -1e-128;
|
|
comp1 := comp(dbl1);
|
|
{$ifdef FPC_DOUBLE_HILO_SWAPPED}
|
|
comp1:=swap(comp1);
|
|
{$endif FPC_DOUBLE_HILO_SWAPPED}
|
|
str(comp1:23,s);
|
|
{$ifdef FPC_HAS_TYPE_EXTENDED}
|
|
if s<>'-6.53142228756617E+0018' then
|
|
{$else FPC_HAS_TYPE_EXTENDED}
|
|
{ this constant has been verified and is correct (FK) }
|
|
{ doubles have slightly different precision on processors <> x86, because }
|
|
{ intermediate calculations are not performed in 80 bit there (JM) }
|
|
if trim(s)<>'-6531422287566170211' then
|
|
{$endif FPC_HAS_TYPE_EXTENDED}
|
|
begin
|
|
writeln(s);
|
|
writeln('error');
|
|
halt(1);
|
|
end;
|
|
end.
|