<s>
In	O
C	B-Language
and	O
related	O
programming	O
languages	O
,	O
long	B-Language
double	I-Language
refers	O
to	O
a	O
floating-point	B-Algorithm
data	O
type	O
that	O
is	O
often	O
more	O
precise	O
than	O
double	O
precision	O
though	O
the	O
language	O
standard	O
only	O
requires	O
it	O
to	O
be	O
at	O
least	O
as	O
precise	O
as	O
double	O
.	O
</s>
<s>
As	O
with	O
C	B-Language
's	O
other	O
floating-point	B-Algorithm
types	O
,	O
it	O
may	O
not	O
necessarily	O
map	O
to	O
an	O
IEEE	O
format	O
.	O
</s>
<s>
The	O
long	B-Language
double	I-Language
type	O
was	O
present	O
in	O
the	O
original	O
1989	O
C	B-Language
standard	O
,	O
but	O
support	O
was	O
improved	O
by	O
the	O
1999	O
revision	O
of	O
the	O
C	B-Language
standard	O
,	O
or	O
C99	B-Language
,	O
which	O
extended	O
the	O
standard	O
library	B-Library
to	O
include	O
functions	O
operating	O
on	O
long	B-Language
double	I-Language
such	O
as	O
sinl( )	O
and	O
strtold( )	O
.	O
</s>
<s>
Long	B-Language
double	I-Language
constants	O
are	O
floating-point	B-Algorithm
constants	O
suffixed	O
with	O
"	O
L	O
"	O
or	O
"	O
l	O
"	O
(	O
lower-case	O
L	O
)	O
,	O
e.g.	O
,	O
0.3333333333333333333333333333333333L	O
or	O
3.1415926535897932384626433832795029L	O
for	O
quadruple	O
precision	O
.	O
</s>
<s>
On	O
the	O
x86	B-Operating_System
architecture	I-Operating_System
,	O
most	O
C	B-Language
compilers	O
implement	O
long	B-Language
double	I-Language
as	O
the	O
80-bit	O
extended	B-Algorithm
precision	I-Algorithm
type	O
supported	O
by	O
x86	B-Operating_System
hardware	O
(	O
generally	O
stored	O
as	O
12	O
or	O
16	O
bytes	O
to	O
maintain	O
data	B-Application
structure	I-Application
alignment	I-Application
)	O
,	O
as	O
specified	O
in	O
the	O
C99	B-Language
/	O
C11	O
standards	O
(	O
IEC	O
60559	O
floating-point	B-Algorithm
arithmetic	I-Algorithm
(	O
Annex	O
F	O
)	O
)	O
.	O
</s>
<s>
An	O
exception	O
is	O
Microsoft	B-Application
Visual	I-Application
C++	I-Application
for	O
x86	B-Operating_System
,	O
which	O
makes	O
long	B-Language
double	I-Language
a	O
synonym	O
for	O
double	O
.	O
</s>
<s>
The	O
Intel	B-Language
C++	I-Language
compiler	I-Language
on	O
Microsoft	O
Windows	O
supports	O
extended	B-Algorithm
precision	I-Algorithm
,	O
but	O
requires	O
the	O
/Qlong	O
‑double	O
switch	O
for	O
long	B-Language
double	I-Language
to	O
correspond	O
to	O
the	O
hardware	O
's	O
extended	B-Algorithm
precision	I-Algorithm
format	I-Algorithm
.	O
</s>
<s>
Compilers	O
may	O
also	O
use	O
long	B-Language
double	I-Language
for	O
the	O
IEEE	O
754	O
quadruple-precision	O
binary	B-Algorithm
floating-point	I-Algorithm
format	O
(	O
binary128	O
)	O
.	O
</s>
<s>
This	O
is	O
the	O
case	O
on	O
HP-UX	B-Application
,	O
Solaris/SPARC	O
,	O
MIPS	B-Device
with	O
the	O
64-bit	O
or	O
n32	O
ABI	B-Operating_System
,	O
64-bit	O
ARM	B-Architecture
(	O
AArch64	O
)	O
(	O
on	O
operating	O
systems	O
using	O
the	O
standard	O
AAPCS	O
calling	O
conventions	O
,	O
such	O
as	O
Linux	B-Application
)	O
,	O
and	O
z/OS	B-Application
with	O
FLOAT(IEEE )	O
.	O
</s>
<s>
On	O
some	O
PowerPC	B-Architecture
systems	O
,	O
long	B-Language
double	I-Language
is	O
implemented	O
as	O
a	O
double-double	O
arithmetic	O
,	O
where	O
a	O
long	B-Language
double	I-Language
value	O
is	O
regarded	O
as	O
the	O
exact	O
sum	O
of	O
two	O
double-precision	O
values	O
,	O
giving	O
at	O
least	O
a	O
106-bit	O
precision	O
;	O
with	O
such	O
a	O
format	O
,	O
the	O
long	B-Language
double	I-Language
type	O
does	O
not	O
conform	O
to	O
the	O
IEEE	O
floating-point	B-Algorithm
standard	O
.	O
</s>
<s>
Otherwise	O
,	O
long	B-Language
double	I-Language
is	O
simply	O
a	O
synonym	O
for	O
double	O
(	O
double	O
precision	O
)	O
,	O
e.g.	O
</s>
<s>
on	O
32-bit	O
ARM	B-Architecture
,	O
64-bit	O
ARM	B-Architecture
(	O
AArch64	O
)	O
(	O
on	O
Windows	O
and	O
macOS	O
)	O
and	O
on	O
32-bit	O
MIPS	B-Device
(	O
old	O
ABI	B-Operating_System
,	O
a.k.a.	O
</s>
<s>
With	O
the	O
GNU	B-Application
C	I-Application
Compiler	I-Application
,	O
long	B-Language
double	I-Language
is	O
80-bit	O
extended	B-Algorithm
precision	I-Algorithm
on	O
x86	B-Operating_System
processors	O
regardless	O
of	O
the	O
physical	O
storage	O
used	O
for	O
the	O
type	O
(	O
which	O
can	O
be	O
either	O
96	O
or	O
128	O
bits	O
)	O
,	O
On	O
some	O
other	O
architectures	O
,	O
long	B-Language
double	I-Language
can	O
be	O
double-double	O
(	O
e.g.	O
</s>
<s>
on	O
PowerPC	B-Architecture
)	O
or	O
128-bit	O
quadruple	O
precision	O
(	O
e.g.	O
</s>
<s>
on	O
SPARC	B-Architecture
)	O
.	O
</s>
<s>
As	O
of	O
gcc	B-Application
4.3	O
,	O
a	O
quadruple	O
precision	O
is	O
also	O
supported	O
on	O
x86	B-Operating_System
,	O
but	O
as	O
the	O
nonstandard	O
type	O
__float128	O
rather	O
than	O
long	B-Language
double	I-Language
.	O
</s>
<s>
Although	O
the	O
x86	B-Operating_System
architecture	I-Operating_System
,	O
and	O
specifically	O
the	O
x87	B-Application
floating-point	B-Algorithm
instructions	O
on	O
x86	B-Operating_System
,	O
supports	O
80-bit	O
extended-precision	O
operations	O
,	O
it	O
is	O
possible	O
to	O
configure	O
the	O
processor	O
to	O
automatically	O
round	O
operations	O
to	O
double	O
(	O
or	O
even	O
single	O
)	O
precision	O
.	O
</s>
<s>
Conversely	O
,	O
in	O
extended-precision	O
mode	O
,	O
extended	B-Algorithm
precision	I-Algorithm
may	O
be	O
used	O
for	O
intermediate	O
compiler-generated	O
calculations	O
even	O
when	O
the	O
final	O
results	O
are	O
stored	O
at	O
a	O
lower	O
precision	O
(	O
i.e.	O
</s>
<s>
With	O
gcc	B-Application
on	O
Linux	B-Application
,	O
80-bit	O
extended	B-Algorithm
precision	I-Algorithm
is	O
the	O
default	O
;	O
on	O
several	O
BSD	B-Operating_System
operating	I-Operating_System
systems	I-Operating_System
(	O
FreeBSD	B-Operating_System
and	O
OpenBSD	B-Operating_System
)	O
,	O
double-precision	O
mode	O
is	O
the	O
default	O
,	O
and	O
long	B-Language
double	I-Language
operations	O
are	O
effectively	O
reduced	O
to	O
double	O
precision	O
.	O
</s>
<s>
(	O
NetBSD	B-Device
7.0	O
and	O
later	O
,	O
however	O
,	O
defaults	O
to	O
80-bit	O
extended	B-Algorithm
precision	I-Algorithm
)	O
.	O
</s>
<s>
However	O
,	O
it	O
is	O
possible	O
to	O
override	O
this	O
within	O
an	O
individual	O
program	O
via	O
the	O
FLDCW	O
"	O
floating-point	B-Algorithm
load	O
control-word	O
"	O
instruction	O
.	O
</s>
<s>
On	O
x86_64	O
,	O
the	O
BSDs	B-Operating_System
default	O
to	O
80-bit	O
extended	B-Algorithm
precision	I-Algorithm
.	O
</s>
<s>
Microsoft	O
Windows	O
with	O
Visual	B-Application
C++	I-Application
also	O
sets	O
the	O
processor	O
in	O
double-precision	O
mode	O
by	O
default	O
,	O
but	O
this	O
can	O
again	O
be	O
overridden	O
within	O
an	O
individual	O
program	O
(	O
e.g.	O
</s>
<s>
by	O
the	O
_controlfp_s	O
function	O
in	O
Visual	B-Application
C++	I-Application
)	O
.	O
</s>
<s>
The	O
Intel	B-Language
C++	I-Language
Compiler	I-Language
for	O
x86	B-Operating_System
,	O
on	O
the	O
other	O
hand	O
,	O
enables	O
extended-precision	O
mode	O
by	O
default	O
.	O
</s>
<s>
On	O
IA-32	O
OS	O
X	O
,	O
long	B-Language
double	I-Language
is	O
80-bit	O
extended	B-Algorithm
precision	I-Algorithm
.	O
</s>
<s>
In	O
CORBA	B-Architecture
(	O
from	O
specification	O
of	O
3.0	O
,	O
which	O
uses	O
"	O
ANSI/IEEE	O
Standard	O
754-1985	O
"	O
as	O
its	O
reference	O
)	O
,	O
"	O
the	O
long	B-Language
double	I-Language
data	O
type	O
represents	O
an	O
IEEE	O
double-extended	O
floating-point	B-Algorithm
number	I-Algorithm
,	O
which	O
has	O
an	O
exponent	O
of	O
at	O
least	O
15	O
bits	O
in	O
length	O
and	O
a	O
signed	O
fraction	O
of	O
at	O
least	O
64	O
bits	O
"	O
,	O
with	O
GIOP/IIOP	O
CDR	O
,	O
whose	O
floating-point	B-Algorithm
types	O
"	O
exactly	O
follow	O
the	O
IEEE	O
standard	O
formats	O
for	O
floating	B-Algorithm
point	I-Algorithm
numbers	I-Algorithm
"	O
,	O
marshalling	O
this	O
as	O
what	O
seems	O
to	O
be	O
IEEE	O
754-2008	O
binary128	O
a.k.a.	O
</s>
