Copyright 2007, 2024 Anthony Howe. All rights reserved.
Post4 is written in ISO C11 using only one extension, Labels As Values, which is supported by gcc and clang compilers. The bulk of the floating point support is provided by ISO C11 libm, which is enabled by default. It can be disabled.
$ ./configure --disable-math
( caddr u -- F:f bool )
Convert the numeric floating point string caddr u into an internal floating point value f on the float stack. Return true if caddr u is valid string, otherwise false.
( F:f faddr -- )
Store f at faddr.
(F: f1 f2 -- f3 )
Multiply the top two cells on the float stack.
(F: f1 f2 -- f3 )
Raise f1 to the power f2, giving the product f3.
(F: f1 f2 -- f3 )
Add the top two cells on the float stack.
(F: f1 f2 -- f3 )
Subtract f2 from f1 placing the result f3 on the float stack.
(F: f1 f2 -- f3 )
Divide the top two cells on the float stack.
(F: f -- )
Display, with a trailing space, the top number on the float stack using fixed-point notation:
[-] 〈digits〉.〈digits0〉
(F: f1 f2 -- f3 )
bool is true if and only if f is less than zero (0).
(F: f -- ) ( -- bool )
bool is true if and only if f equals zero (0).
(F: f1 f2 -- ) ( -- bool)
bool is true if and only if f1 is less than f2.
(F: f -- )( -- n)
Convert the float f to a single-cell signed integer n, the fractional portion of f is discarded.
( faddr -- ) (F: -- f )
Fetch from faddr the value f stored there.
(F: f1 -- f2 )
f2 is the absolute value of f1.
( -- )
If the data-space pointer is not float aligned, reserve enough space to align it.
( addr -- faddr )
faddr is the first float aligned address greater than or equal to addr.
(F: f -- ) ( <spaces>name -- )
Define the word name to represent the float constant value f. When name is executed, the value f is pushed on to the float stack.
(F: f1 -- f2 )
f2 is the cosine of the radian angle f1.
( -- u )
Number of cells on the float stack before f was placed there.
(F: f -- )
Remove the top of the float stack.
(F: f -- f f )
Duplicate f on the float stack.
(F: f1 -- f2 )
Raise e to the power f1, giving f2.
( n1 <spaces>name -- n2 )
Skip leading space delimiters. Parse name delimited by a space. Offset is the first float aligned value greater than or equal to n1. n2 = offset + 1 float.
(F: f -- ) immediate
Compile f into the definition so that it is later pushed onto the float stack during execution of the definition.
(F: f1 -- f2 )
f2 is the natural logarithm of f1.
( faddr1 -- faddr2 )
Add the size in address units of a float to faddr1 giving faddr2.
( n1 -- n2 )
n2 is the size in address units of n1 floating point numbers.
(F: f1 -- f2 )
f2 is the base-10 logarithm of f1.
(F: f1 -- f2 )
Round f1 to an integral value using the "round toward negative infinity" rule, giving f2.
(F: f1 f2 -- f3 )
Float f3 is the greater of f1 and f2.
(F: f1 f2 -- f3 )
Float f3 is the lesser of f1 and f2.
(F: f1 -- f2 )
Negate the float f1, giving its arithmetic inverse f2.
(F: f1 f2 -- f1 f2 f1 )
Copy the second value f1 below the float stack to the top.
(F: f1 f2 f3 -- f2 f3 f1 )
Rotate the top three float stack entries.
(F: f1 -- f2 )
Round f1 to an integral value using the "round to nearest" rule, giving f2.
(F: f -- )
Display, with a trailing space, the top number on the float stack using scientific notation:
[-] digits[.digits] E digits
(F: f1 -- f2 )
f2 is the sine of the radian angle f1.
(F: f1 f2 -- f2 f1 )
Exchange the top two float stack items.
(F: f1 -- f2 )
f2 is the square root f1.
(F: f1 -- f2 )
f2 is the tangent of the radian angle f1.
( f <spaces>name -- )
Create name and reserve data-space to store f from the float stack. When name is executed, the value f is pushed to the float stack. See TO.
ok 123.45 FVALUE fizz
ok fizz f.
123.450000 ok
ok 0.314159e1 TO fizz
ok fizz f.
3.141590 ok
( <spaces>name -- )
Create name with one float of data. When name is executed push the faddr of the float.
( -- u )
Return the number of significant digits currently used by F., FE., or FS. as u.
(F: f -- )( caddr u -- n isneg isok )
Taking the float f return the first u digits of the significand, where the is an implied decimal point to the left of the first digit, the base-10 exponent n, TRUE if the value is negative, and TRUE if the number is valid (not ± infinity nor NaN). If f does not represent a number, where isok is FALSE, then the values of n and isneg are undefined.
(F: -- f )( n -- )
Convert the single-cell signed integer to a float f.
( u -- )
Set the number of significant digits currently used by F., FE., or FS. to u.
(F: -- f ) (S: f -- )
Move top of the data stack to the float stack without format conversion. S>F will convert formats.
( -- aaddr n s )
Push the float stack base aaddr address, depth n, and size s (before executing _fs). This stack is a fixed size and grows upward. See .fs and _stack_dump.
(F: f -- )
Align and reserve data-space to store f there.
(F: f1 f2 -- bool )
Return bool as true if f1 and f2 are exactly equal (same internal representation).
( -- u ) constant
Size of the float stack. Zero (0) if the float stack is combined with the data stack. This is a deviation from ENVIRONMENT? queries.
(F: f -- ) (S: -- f )
Move top of the float stack to the data stack without format conversion. F>S will convert formats.
(F: f -- ) (R: -- f )
Move top of the float stack to the return stack.
(R: f -- ) (F: -- f )
Move top of the return stack to the float stack.
( aaddr -- )
Store aaddr into the float stack pointer.
( -- aaddr )
Fetch the float stack pointer.