Intel® C++ Compiler 16.0 User and Reference Guide
Lets you override the default configuration of the long double data type.
Not available on Intel® 64 architecture targeting the Intel® Xeon Phi™ coprocessor x100 product family (formerly code name Knights Corner), on IA-32 architecture targeting Intel® Graphics Technology, or on Intel® 64 architecture targeting Intel® Graphics Technology
Linux: | -mlong-double-n |
OS X: | None |
Windows: | None |
n |
Specifies the size of the long double data type. Possible values are:
|
-mlong-double-80 |
Specifies that the size of the long double data type is 80 bits. |
This option lets you override the default configuration of the long double data type.
When you specify -mlong-double-64, the size of the long double data type is 8 bytes and the macro __LONG_DOUBLE_64__ is defined.
When you specify -mlong-double-80, the size of the long double data type is 12 bytes on IA-32 architecture and 16 bytes on Intel® 64 architecture.
This option has no effect on floating-point significand precision. That must be specified by using the -pc64 or -pc80 option.
Note that this option has no effect when you pass arguments. When you pass arguments, the 64-bit long double data type is treated as the double data type and it is always 64-bit.
Remember to include the math.h and complex.h header files when you use this option.
The following restrictions apply to this option:
__bultin_* functions using the long double type should not be used in the non-default mode with Intel compiler libraries.
long double functions from the 'std' namespace should not be called from C++ sources when the non-default mode is set.
Linux and OS X: None
Windows: /Qlong-double