Compilers like GCC and Clang treat C's "long double" type by default as 80-bit wide and result in x87 generated code. This can be overridden to use either 64-bi

Compilers like GCC and Clang treat C's "long double" type by default as 80-bit wide and result in x87 generated code. This can be overridden to use either 64-bit or 128-bit floating point values.