Floating Point
--------------

   All 80387 floating point types except packed BCD are supported.
(BCD support may be added without much difficulty).  These data types
are 16-, 32-, and 64- bit integers, and single (32-bit), double
(64-bit), and extended (80-bit) precision floating point.  Each
supported type has an opcode suffix and a constructor associated with
it.  Opcode suffixes specify operand's data types.  Constructors build
these data types into memory.

   * Floating point constructors are `.float' or `.single', `.double',
     and `.tfloat' for 32-, 64-, and 80-bit formats.  These correspond
     to opcode suffixes `s', `l', and `t'.  `t' stands for temporary
     real, and that the 80387 only supports this format via the `fldt'
     (load temporary real to stack top) and `fstpt' (store temporary
     real and pop stack) instructions.

   * Integer constructors are `.word', `.long' or `.int', and `.quad'
     for the 16-, 32-, and 64-bit integer formats.  The corresponding
     opcode suffixes are `s' (single), `l' (long), and `q' (quad).  As
     with the temporary real format the 64-bit `q' format is only
     present in the `fildq' (load quad integer to stack top) and
     `fistpq' (store quad integer and pop stack) instructions.

   Register to register operations do not require opcode suffixes, so
that `fst %st, %st(1)' is equivalent to `fstl %st, %st(1)'.

   Since the 80387 automatically synchronizes with the 80386 `fwait'
instructions are almost never needed (this is not the case for the
80286/80287 and 8086/8087 combinations).  Therefore, `as' suppresses
the `fwait' instruction whenever it is implicitly selected by one of
the `fn...' instructions.  For example, `fsave' and `fnsave' are
treated identically.  In general, all the `fn...' instructions are made
equivalent to `f...' instructions.  If `fwait' is desired it must be
explicitly coded.