gforth/float.fs - diff

Return to float.fs CVS log

Up to [gforth] / gforth

Diff for /gforth/float.fs between versions 1.30 and 1.31

version 1.30, 2000/01/17 00:04:29	version 1.31, 2000/08/09 20:04:05
Line 52	Line 52
here dup dfaligned swap ?DO bl c, LOOP ;	here dup dfaligned swap ?DO bl c, LOOP ;

1 sfloats (Field) sfloat+ , ( sf-addr1 -- sf-addr2 ) \ float-ext s-float-plus	1 sfloats (Field) sfloat+ , ( sf-addr1 -- sf-addr2 ) \ float-ext s-float-plus
\G Increment @i{sf-addr1} by the number of address units corresponding to the size of	\G @code{1 sfloats +}.
\G a single-precision IEEE floating-point number, to give @i{sf-addr2}.""

1 dfloats (Field) dfloat+ , ( df-addr1 -- df-addr2 ) \ float-ext d-float-plus	1 dfloats (Field) dfloat+ , ( df-addr1 -- df-addr2 ) \ float-ext d-float-plus
\G Increment @i{df-addr1} by the number of address units corresponding to the size of	\G @code{1 dfloats +}.
\G a double-precision IEEE floating-point number, to give @i{df-addr2}.""

: f, ( f -- ) \ gforth	: f, ( f -- ) \ gforth
\G Reserve data space for one floating-point number and store	\G Reserve data space for one floating-point number and store
Line 201 IS interpreter-notfound	Line 199 IS interpreter-notfound
f- fabs frot frot f* f< ;	f- fabs frot frot f* f< ;

: f~ ( r1 r2 r3 -- flag ) \ float-ext f-proximate	: f~ ( r1 r2 r3 -- flag ) \ float-ext f-proximate
\G ANS Forth medley: r3>0: @code{f~abs}; r3=0: r1=r2; r3<0: @code{fnegate f~rel}.	\G ANS Forth medley for comparing r1 and r2 for equality: r3>0:
	\G @code{f~abs}; r3=0: bitwise comparison; r3<0: @code{fnegate f~rel}.
fdup f0=	fdup f0=
IF	IF \ bitwise comparison
fdrop f= \ !! this does not work, because 0=-0 with f= on Linux-Intel	fp@ float+ 1 floats over float+ -text 0=
\ the standard says they should compare unequal	fdrop fdrop fdrop
\ the comparison should be done with COMPARE
EXIT	EXIT
THEN	THEN
fdup f0>	fdup f0>

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>

Removed from v.1.30
changed lines
	Added in v.1.31