Home | History | Annotate | Line # | Download | only in fpsp
bindec.sa revision 1.2
      1  1.1  mycroft *	MOTOROLA MICROPROCESSOR & MEMORY TECHNOLOGY GROUP
      2  1.1  mycroft *	M68000 Hi-Performance Microprocessor Division
      3  1.1  mycroft *	M68040 Software Package 
      4  1.1  mycroft *
      5  1.1  mycroft *	M68040 Software Package Copyright (c) 1993, 1994 Motorola Inc.
      6  1.1  mycroft *	All rights reserved.
      7  1.1  mycroft *
      8  1.1  mycroft *	THE SOFTWARE is provided on an "AS IS" basis and without warranty.
      9  1.1  mycroft *	To the maximum extent permitted by applicable law,
     10  1.1  mycroft *	MOTOROLA DISCLAIMS ALL WARRANTIES WHETHER EXPRESS OR IMPLIED,
     11  1.1  mycroft *	INCLUDING IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A
     12  1.1  mycroft *	PARTICULAR PURPOSE and any warranty against infringement with
     13  1.1  mycroft *	regard to the SOFTWARE (INCLUDING ANY MODIFIED VERSIONS THEREOF)
     14  1.1  mycroft *	and any accompanying written materials. 
     15  1.1  mycroft *
     16  1.1  mycroft *	To the maximum extent permitted by applicable law,
     17  1.1  mycroft *	IN NO EVENT SHALL MOTOROLA BE LIABLE FOR ANY DAMAGES WHATSOEVER
     18  1.1  mycroft *	(INCLUDING WITHOUT LIMITATION, DAMAGES FOR LOSS OF BUSINESS
     19  1.1  mycroft *	PROFITS, BUSINESS INTERRUPTION, LOSS OF BUSINESS INFORMATION, OR
     20  1.1  mycroft *	OTHER PECUNIARY LOSS) ARISING OF THE USE OR INABILITY TO USE THE
     21  1.1  mycroft *	SOFTWARE.  Motorola assumes no responsibility for the maintenance
     22  1.1  mycroft *	and support of the SOFTWARE.  
     23  1.1  mycroft *
     24  1.1  mycroft *	You are hereby granted a copyright license to use, modify, and
     25  1.1  mycroft *	distribute the SOFTWARE so long as this entire notice is retained
     26  1.1  mycroft *	without alteration in any modified and/or redistributed versions,
     27  1.1  mycroft *	and that such modified versions are clearly identified as such.
     28  1.1  mycroft *	No licenses are granted by implication, estoppel or otherwise
     29  1.1  mycroft *	under any patents or trademarks of Motorola, Inc.
     30  1.1  mycroft 
     31  1.1  mycroft *
     32  1.1  mycroft *	bindec.sa 3.4 1/3/91
     33  1.1  mycroft *
     34  1.1  mycroft *	bindec
     35  1.1  mycroft *
     36  1.1  mycroft *	Description:
     37  1.1  mycroft *		Converts an input in extended precision format
     38  1.1  mycroft *		to bcd format.
     39  1.1  mycroft *
     40  1.1  mycroft *	Input:
     41  1.1  mycroft *		a0 points to the input extended precision value
     42  1.1  mycroft *		value in memory; d0 contains the k-factor sign-extended
     43  1.1  mycroft *		to 32-bits.  The input may be either normalized,
     44  1.1  mycroft *		unnormalized, or denormalized.
     45  1.1  mycroft *
     46  1.1  mycroft *	Output:	result in the FP_SCR1 space on the stack.
     47  1.1  mycroft *
     48  1.1  mycroft *	Saves and Modifies: D2-D7,A2,FP2
     49  1.1  mycroft *
     50  1.1  mycroft *	Algorithm:
     51  1.1  mycroft *
     52  1.1  mycroft *	A1.	Set RM and size ext;  Set SIGMA = sign of input.  
     53  1.1  mycroft *		The k-factor is saved for use in d7. Clear the
     54  1.1  mycroft *		BINDEC_FLG for separating normalized/denormalized
     55  1.1  mycroft *		input.  If input is unnormalized or denormalized,
     56  1.1  mycroft *		normalize it.
     57  1.1  mycroft *
     58  1.1  mycroft *	A2.	Set X = abs(input).
     59  1.1  mycroft *
     60  1.1  mycroft *	A3.	Compute ILOG.
     61  1.1  mycroft *		ILOG is the log base 10 of the input value.  It is
     62  1.1  mycroft *		approximated by adding e + 0.f when the original 
     63  1.1  mycroft *		value is viewed as 2^^e * 1.f in extended precision.  
     64  1.1  mycroft *		This value is stored in d6.
     65  1.1  mycroft *
     66  1.1  mycroft *	A4.	Clr INEX bit.
     67  1.1  mycroft *		The operation in A3 above may have set INEX2.  
     68  1.1  mycroft *
     69  1.1  mycroft *	A5.	Set ICTR = 0;
     70  1.1  mycroft *		ICTR is a flag used in A13.  It must be set before the 
     71  1.1  mycroft *		loop entry A6.
     72  1.1  mycroft *
     73  1.1  mycroft *	A6.	Calculate LEN.
     74  1.1  mycroft *		LEN is the number of digits to be displayed.  The
     75  1.1  mycroft *		k-factor can dictate either the total number of digits,
     76  1.1  mycroft *		if it is a positive number, or the number of digits
     77  1.1  mycroft *		after the decimal point which are to be included as
     78  1.1  mycroft *		significant.  See the 68882 manual for examples.
     79  1.1  mycroft *		If LEN is computed to be greater than 17, set OPERR in
     80  1.1  mycroft *		USER_FPSR.  LEN is stored in d4.
     81  1.1  mycroft *
     82  1.1  mycroft *	A7.	Calculate SCALE.
     83  1.1  mycroft *		SCALE is equal to 10^ISCALE, where ISCALE is the number
     84  1.1  mycroft *		of decimal places needed to insure LEN integer digits
     85  1.1  mycroft *		in the output before conversion to bcd. LAMBDA is the
     86  1.1  mycroft *		sign of ISCALE, used in A9. Fp1 contains
     87  1.1  mycroft *		10^^(abs(ISCALE)) using a rounding mode which is a
     88  1.1  mycroft *		function of the original rounding mode and the signs
     89  1.1  mycroft *		of ISCALE and X.  A table is given in the code.
     90  1.1  mycroft *
     91  1.1  mycroft *	A8.	Clr INEX; Force RZ.
     92  1.1  mycroft *		The operation in A3 above may have set INEX2.  
     93  1.1  mycroft *		RZ mode is forced for the scaling operation to insure
     94  1.1  mycroft *		only one rounding error.  The grs bits are collected in 
     95  1.1  mycroft *		the INEX flag for use in A10.
     96  1.1  mycroft *
     97  1.1  mycroft *	A9.	Scale X -> Y.
     98  1.1  mycroft *		The mantissa is scaled to the desired number of
     99  1.1  mycroft *		significant digits.  The excess digits are collected
    100  1.1  mycroft *		in INEX2.
    101  1.1  mycroft *
    102  1.1  mycroft *	A10.	Or in INEX.
    103  1.1  mycroft *		If INEX is set, round error occured.  This is
    104  1.1  mycroft *		compensated for by 'or-ing' in the INEX2 flag to
    105  1.1  mycroft *		the lsb of Y.
    106  1.1  mycroft *
    107  1.1  mycroft *	A11.	Restore original FPCR; set size ext.
    108  1.1  mycroft *		Perform FINT operation in the user's rounding mode.
    109  1.1  mycroft *		Keep the size to extended.
    110  1.1  mycroft *
    111  1.1  mycroft *	A12.	Calculate YINT = FINT(Y) according to user's rounding
    112  1.1  mycroft *		mode.  The FPSP routine sintd0 is used.  The output
    113  1.1  mycroft *		is in fp0.
    114  1.1  mycroft *
    115  1.1  mycroft *	A13.	Check for LEN digits.
    116  1.1  mycroft *		If the int operation results in more than LEN digits,
    117  1.1  mycroft *		or less than LEN -1 digits, adjust ILOG and repeat from
    118  1.1  mycroft *		A6.  This test occurs only on the first pass.  If the
    119  1.1  mycroft *		result is exactly 10^LEN, decrement ILOG and divide
    120  1.1  mycroft *		the mantissa by 10.
    121  1.1  mycroft *
    122  1.1  mycroft *	A14.	Convert the mantissa to bcd.
    123  1.1  mycroft *		The binstr routine is used to convert the LEN digit 
    124  1.1  mycroft *		mantissa to bcd in memory.  The input to binstr is
    125  1.1  mycroft *		to be a fraction; i.e. (mantissa)/10^LEN and adjusted
    126  1.1  mycroft *		such that the decimal point is to the left of bit 63.
    127  1.1  mycroft *		The bcd digits are stored in the correct position in 
    128  1.1  mycroft *		the final string area in memory.
    129  1.1  mycroft *
    130  1.1  mycroft *	A15.	Convert the exponent to bcd.
    131  1.1  mycroft *		As in A14 above, the exp is converted to bcd and the
    132  1.1  mycroft *		digits are stored in the final string.
    133  1.1  mycroft *		Test the length of the final exponent string.  If the
    134  1.1  mycroft *		length is 4, set operr.
    135  1.1  mycroft *
    136  1.1  mycroft *	A16.	Write sign bits to final string.
    137  1.1  mycroft *
    138  1.1  mycroft *	Implementation Notes:
    139  1.1  mycroft *
    140  1.1  mycroft *	The registers are used as follows:
    141  1.1  mycroft *
    142  1.1  mycroft *		d0: scratch; LEN input to binstr
    143  1.1  mycroft *		d1: scratch
    144  1.1  mycroft *		d2: upper 32-bits of mantissa for binstr
    145  1.1  mycroft *		d3: scratch;lower 32-bits of mantissa for binstr
    146  1.1  mycroft *		d4: LEN
    147  1.1  mycroft *      		d5: LAMBDA/ICTR
    148  1.1  mycroft *		d6: ILOG
    149  1.1  mycroft *		d7: k-factor
    150  1.1  mycroft *		a0: ptr for original operand/final result
    151  1.1  mycroft *		a1: scratch pointer
    152  1.1  mycroft *		a2: pointer to FP_X; abs(original value) in ext
    153  1.1  mycroft *		fp0: scratch
    154  1.1  mycroft *		fp1: scratch
    155  1.1  mycroft *		fp2: scratch
    156  1.1  mycroft *		F_SCR1:
    157  1.1  mycroft *		F_SCR2:
    158  1.1  mycroft *		L_SCR1:
    159  1.1  mycroft *		L_SCR2:
    160  1.1  mycroft *
    161  1.1  mycroft 
    162  1.1  mycroft BINDEC    IDNT    2,1 Motorola 040 Floating Point Software Package
    163  1.1  mycroft 
    164  1.1  mycroft 	include	fpsp.h
    165  1.1  mycroft 
    166  1.1  mycroft 	section	8
    167  1.1  mycroft 
    168  1.1  mycroft * Constants in extended precision
    169  1.1  mycroft LOG2 	dc.l	$3FFD0000,$9A209A84,$FBCFF798,$00000000
    170  1.1  mycroft LOG2UP1	dc.l	$3FFD0000,$9A209A84,$FBCFF799,$00000000
    171  1.1  mycroft 
    172  1.1  mycroft * Constants in single precision
    173  1.1  mycroft FONE 	dc.l	$3F800000,$00000000,$00000000,$00000000
    174  1.1  mycroft FTWO	dc.l	$40000000,$00000000,$00000000,$00000000
    175  1.1  mycroft FTEN 	dc.l	$41200000,$00000000,$00000000,$00000000
    176  1.1  mycroft F4933	dc.l	$459A2800,$00000000,$00000000,$00000000
    177  1.1  mycroft 
    178  1.1  mycroft RBDTBL 	dc.b	0,0,0,0
    179  1.1  mycroft 	dc.b	3,3,2,2
    180  1.1  mycroft 	dc.b	3,2,2,3
    181  1.1  mycroft 	dc.b	2,3,3,2
    182  1.1  mycroft 
    183  1.1  mycroft 	xref	binstr
    184  1.1  mycroft 	xref	sintdo
    185  1.1  mycroft 	xref	ptenrn,ptenrm,ptenrp
    186  1.1  mycroft 
    187  1.1  mycroft 	xdef	bindec
    188  1.1  mycroft 	xdef	sc_mul
    189  1.1  mycroft bindec:
    190  1.1  mycroft 	movem.l	d2-d7/a2,-(a7)
    191  1.1  mycroft 	fmovem.x fp0-fp2,-(a7)
    192  1.1  mycroft 
    193  1.1  mycroft * A1. Set RM and size ext. Set SIGMA = sign input;
    194  1.1  mycroft *     The k-factor is saved for use in d7.  Clear BINDEC_FLG for
    195  1.1  mycroft *     separating  normalized/denormalized input.  If the input
    196  1.1  mycroft *     is a denormalized number, set the BINDEC_FLG memory word
    197  1.1  mycroft *     to signal denorm.  If the input is unnormalized, normalize
    198  1.1  mycroft *     the input and test for denormalized result.  
    199  1.1  mycroft *
    200  1.1  mycroft 	fmove.l	#rm_mode,FPCR	;set RM and ext
    201  1.1  mycroft 	move.l	(a0),L_SCR2(a6)	;save exponent for sign check
    202  1.1  mycroft 	move.l	d0,d7		;move k-factor to d7
    203  1.1  mycroft 	clr.b	BINDEC_FLG(a6)	;clr norm/denorm flag
    204  1.1  mycroft 	move.w	STAG(a6),d0	;get stag
    205  1.1  mycroft 	andi.w	#$e000,d0	;isolate stag bits
    206  1.1  mycroft 	beq	A2_str		;if zero, input is norm
    207  1.1  mycroft *
    208  1.1  mycroft * Normalize the denorm
    209  1.1  mycroft *
    210  1.1  mycroft un_de_norm:
    211  1.1  mycroft 	move.w	(a0),d0
    212  1.1  mycroft 	andi.w	#$7fff,d0	;strip sign of normalized exp
    213  1.1  mycroft 	move.l	4(a0),d1
    214  1.1  mycroft 	move.l	8(a0),d2
    215  1.1  mycroft norm_loop:
    216  1.1  mycroft 	sub.w	#1,d0
    217  1.2  mycroft 	add.l	d2,d2
    218  1.2  mycroft 	addx.l	d1,d1
    219  1.1  mycroft 	tst.l	d1
    220  1.1  mycroft 	bge.b	norm_loop
    221  1.1  mycroft *
    222  1.1  mycroft * Test if the normalized input is denormalized
    223  1.1  mycroft *
    224  1.1  mycroft 	tst.w	d0
    225  1.1  mycroft 	bgt.b	pos_exp		;if greater than zero, it is a norm
    226  1.1  mycroft 	st	BINDEC_FLG(a6)	;set flag for denorm
    227  1.1  mycroft pos_exp:
    228  1.1  mycroft 	andi.w	#$7fff,d0	;strip sign of normalized exp
    229  1.1  mycroft 	move.w	d0,(a0)
    230  1.1  mycroft 	move.l	d1,4(a0)
    231  1.1  mycroft 	move.l	d2,8(a0)
    232  1.1  mycroft 
    233  1.1  mycroft * A2. Set X = abs(input).
    234  1.1  mycroft *
    235  1.1  mycroft A2_str:
    236  1.1  mycroft 	move.l	(a0),FP_SCR2(a6) ; move input to work space
    237  1.1  mycroft 	move.l	4(a0),FP_SCR2+4(a6) ; move input to work space
    238  1.1  mycroft 	move.l	8(a0),FP_SCR2+8(a6) ; move input to work space
    239  1.1  mycroft 	andi.l	#$7fffffff,FP_SCR2(a6) ;create abs(X)
    240  1.1  mycroft 
    241  1.1  mycroft * A3. Compute ILOG.
    242  1.1  mycroft *     ILOG is the log base 10 of the input value.  It is approx-
    243  1.1  mycroft *     imated by adding e + 0.f when the original value is viewed
    244  1.1  mycroft *     as 2^^e * 1.f in extended precision.  This value is stored
    245  1.1  mycroft *     in d6.
    246  1.1  mycroft *
    247  1.1  mycroft * Register usage:
    248  1.1  mycroft *	Input/Output
    249  1.1  mycroft *	d0: k-factor/exponent
    250  1.1  mycroft *	d2: x/x
    251  1.1  mycroft *	d3: x/x
    252  1.1  mycroft *	d4: x/x
    253  1.1  mycroft *	d5: x/x
    254  1.1  mycroft *	d6: x/ILOG
    255  1.1  mycroft *	d7: k-factor/Unchanged
    256  1.1  mycroft *	a0: ptr for original operand/final result
    257  1.1  mycroft *	a1: x/x
    258  1.1  mycroft *	a2: x/x
    259  1.1  mycroft *	fp0: x/float(ILOG)
    260  1.1  mycroft *	fp1: x/x
    261  1.1  mycroft *	fp2: x/x
    262  1.1  mycroft *	F_SCR1:x/x
    263  1.1  mycroft *	F_SCR2:Abs(X)/Abs(X) with $3fff exponent
    264  1.1  mycroft *	L_SCR1:x/x
    265  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    266  1.1  mycroft 
    267  1.1  mycroft 	tst.b	BINDEC_FLG(a6)	;check for denorm
    268  1.1  mycroft 	beq.b	A3_cont		;if clr, continue with norm
    269  1.1  mycroft 	move.l	#-4933,d6	;force ILOG = -4933
    270  1.1  mycroft 	bra.b	A4_str
    271  1.1  mycroft A3_cont:
    272  1.1  mycroft 	move.w	FP_SCR2(a6),d0	;move exp to d0
    273  1.1  mycroft 	move.w	#$3fff,FP_SCR2(a6) ;replace exponent with 0x3fff
    274  1.1  mycroft 	fmove.x	FP_SCR2(a6),fp0	;now fp0 has 1.f
    275  1.1  mycroft 	sub.w	#$3fff,d0	;strip off bias
    276  1.1  mycroft 	fadd.w	d0,fp0		;add in exp
    277  1.1  mycroft 	fsub.s	FONE,fp0	;subtract off 1.0
    278  1.1  mycroft 	fbge.w	pos_res		;if pos, branch 
    279  1.1  mycroft 	fmul.x	LOG2UP1,fp0	;if neg, mul by LOG2UP1
    280  1.1  mycroft 	fmove.l	fp0,d6		;put ILOG in d6 as a lword
    281  1.1  mycroft 	bra.b	A4_str		;go move out ILOG
    282  1.1  mycroft pos_res:
    283  1.1  mycroft 	fmul.x	LOG2,fp0	;if pos, mul by LOG2
    284  1.1  mycroft 	fmove.l	fp0,d6		;put ILOG in d6 as a lword
    285  1.1  mycroft 
    286  1.1  mycroft 
    287  1.1  mycroft * A4. Clr INEX bit.
    288  1.1  mycroft *     The operation in A3 above may have set INEX2.  
    289  1.1  mycroft 
    290  1.1  mycroft A4_str:	
    291  1.1  mycroft 	fmove.l	#0,FPSR		;zero all of fpsr - nothing needed
    292  1.1  mycroft 
    293  1.1  mycroft 
    294  1.1  mycroft * A5. Set ICTR = 0;
    295  1.1  mycroft *     ICTR is a flag used in A13.  It must be set before the 
    296  1.1  mycroft *     loop entry A6. The lower word of d5 is used for ICTR.
    297  1.1  mycroft 
    298  1.1  mycroft 	clr.w	d5		;clear ICTR
    299  1.1  mycroft 
    300  1.1  mycroft 
    301  1.1  mycroft * A6. Calculate LEN.
    302  1.1  mycroft *     LEN is the number of digits to be displayed.  The k-factor
    303  1.1  mycroft *     can dictate either the total number of digits, if it is
    304  1.1  mycroft *     a positive number, or the number of digits after the
    305  1.1  mycroft *     original decimal point which are to be included as
    306  1.1  mycroft *     significant.  See the 68882 manual for examples.
    307  1.1  mycroft *     If LEN is computed to be greater than 17, set OPERR in
    308  1.1  mycroft *     USER_FPSR.  LEN is stored in d4.
    309  1.1  mycroft *
    310  1.1  mycroft * Register usage:
    311  1.1  mycroft *	Input/Output
    312  1.1  mycroft *	d0: exponent/Unchanged
    313  1.1  mycroft *	d2: x/x/scratch
    314  1.1  mycroft *	d3: x/x
    315  1.1  mycroft *	d4: exc picture/LEN
    316  1.1  mycroft *	d5: ICTR/Unchanged
    317  1.1  mycroft *	d6: ILOG/Unchanged
    318  1.1  mycroft *	d7: k-factor/Unchanged
    319  1.1  mycroft *	a0: ptr for original operand/final result
    320  1.1  mycroft *	a1: x/x
    321  1.1  mycroft *	a2: x/x
    322  1.1  mycroft *	fp0: float(ILOG)/Unchanged
    323  1.1  mycroft *	fp1: x/x
    324  1.1  mycroft *	fp2: x/x
    325  1.1  mycroft *	F_SCR1:x/x
    326  1.1  mycroft *	F_SCR2:Abs(X) with $3fff exponent/Unchanged
    327  1.1  mycroft *	L_SCR1:x/x
    328  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    329  1.1  mycroft 
    330  1.1  mycroft A6_str:	
    331  1.1  mycroft 	tst.l	d7		;branch on sign of k
    332  1.1  mycroft 	ble.b	k_neg		;if k <= 0, LEN = ILOG + 1 - k
    333  1.1  mycroft 	move.l	d7,d4		;if k > 0, LEN = k
    334  1.1  mycroft 	bra.b	len_ck		;skip to LEN check
    335  1.1  mycroft k_neg:
    336  1.1  mycroft 	move.l	d6,d4		;first load ILOG to d4
    337  1.1  mycroft 	sub.l	d7,d4		;subtract off k
    338  1.1  mycroft 	addq.l	#1,d4		;add in the 1
    339  1.1  mycroft len_ck:
    340  1.1  mycroft 	tst.l	d4		;LEN check: branch on sign of LEN
    341  1.1  mycroft 	ble.b	LEN_ng		;if neg, set LEN = 1
    342  1.1  mycroft 	cmp.l	#17,d4		;test if LEN > 17
    343  1.1  mycroft 	ble.b	A7_str		;if not, forget it
    344  1.1  mycroft 	move.l	#17,d4		;set max LEN = 17
    345  1.1  mycroft 	tst.l	d7		;if negative, never set OPERR
    346  1.1  mycroft 	ble.b	A7_str		;if positive, continue
    347  1.1  mycroft 	or.l	#opaop_mask,USER_FPSR(a6) ;set OPERR & AIOP in USER_FPSR
    348  1.1  mycroft 	bra.b	A7_str		;finished here
    349  1.1  mycroft LEN_ng:
    350  1.1  mycroft 	moveq.l	#1,d4		;min LEN is 1
    351  1.1  mycroft 
    352  1.1  mycroft 
    353  1.1  mycroft * A7. Calculate SCALE.
    354  1.1  mycroft *     SCALE is equal to 10^ISCALE, where ISCALE is the number
    355  1.1  mycroft *     of decimal places needed to insure LEN integer digits
    356  1.1  mycroft *     in the output before conversion to bcd. LAMBDA is the sign
    357  1.1  mycroft *     of ISCALE, used in A9.  Fp1 contains 10^^(abs(ISCALE)) using
    358  1.1  mycroft *     the rounding mode as given in the following table (see
    359  1.1  mycroft *     Coonen, p. 7.23 as ref.; however, the SCALE variable is
    360  1.1  mycroft *     of opposite sign in bindec.sa from Coonen).
    361  1.1  mycroft *
    362  1.1  mycroft *	Initial					USE
    363  1.1  mycroft *	FPCR[6:5]	LAMBDA	SIGN(X)		FPCR[6:5]
    364  1.1  mycroft *	----------------------------------------------
    365  1.1  mycroft *	 RN	00	   0	   0		00/0	RN
    366  1.1  mycroft *	 RN	00	   0	   1		00/0	RN
    367  1.1  mycroft *	 RN	00	   1	   0		00/0	RN
    368  1.1  mycroft *	 RN	00	   1	   1		00/0	RN
    369  1.1  mycroft *	 RZ	01	   0	   0		11/3	RP
    370  1.1  mycroft *	 RZ	01	   0	   1		11/3	RP
    371  1.1  mycroft *	 RZ	01	   1	   0		10/2	RM
    372  1.1  mycroft *	 RZ	01	   1	   1		10/2	RM
    373  1.1  mycroft *	 RM	10	   0	   0		11/3	RP
    374  1.1  mycroft *	 RM	10	   0	   1		10/2	RM
    375  1.1  mycroft *	 RM	10	   1	   0		10/2	RM
    376  1.1  mycroft *	 RM	10	   1	   1		11/3	RP
    377  1.1  mycroft *	 RP	11	   0	   0		10/2	RM
    378  1.1  mycroft *	 RP	11	   0	   1		11/3	RP
    379  1.1  mycroft *	 RP	11	   1	   0		11/3	RP
    380  1.1  mycroft *	 RP	11	   1	   1		10/2	RM
    381  1.1  mycroft *
    382  1.1  mycroft * Register usage:
    383  1.1  mycroft *	Input/Output
    384  1.1  mycroft *	d0: exponent/scratch - final is 0
    385  1.1  mycroft *	d2: x/0 or 24 for A9
    386  1.1  mycroft *	d3: x/scratch - offset ptr into PTENRM array
    387  1.1  mycroft *	d4: LEN/Unchanged
    388  1.1  mycroft *	d5: 0/ICTR:LAMBDA
    389  1.1  mycroft *	d6: ILOG/ILOG or k if ((k<=0)&(ILOG<k))
    390  1.1  mycroft *	d7: k-factor/Unchanged
    391  1.1  mycroft *	a0: ptr for original operand/final result
    392  1.1  mycroft *	a1: x/ptr to PTENRM array
    393  1.1  mycroft *	a2: x/x
    394  1.1  mycroft *	fp0: float(ILOG)/Unchanged
    395  1.1  mycroft *	fp1: x/10^ISCALE
    396  1.1  mycroft *	fp2: x/x
    397  1.1  mycroft *	F_SCR1:x/x
    398  1.1  mycroft *	F_SCR2:Abs(X) with $3fff exponent/Unchanged
    399  1.1  mycroft *	L_SCR1:x/x
    400  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    401  1.1  mycroft 
    402  1.1  mycroft A7_str:	
    403  1.1  mycroft 	tst.l	d7		;test sign of k
    404  1.1  mycroft 	bgt.b	k_pos		;if pos and > 0, skip this
    405  1.1  mycroft 	cmp.l	d6,d7		;test k - ILOG
    406  1.1  mycroft 	blt.b	k_pos		;if ILOG >= k, skip this
    407  1.1  mycroft 	move.l	d7,d6		;if ((k<0) & (ILOG < k)) ILOG = k
    408  1.1  mycroft k_pos:	
    409  1.1  mycroft 	move.l	d6,d0		;calc ILOG + 1 - LEN in d0
    410  1.1  mycroft 	addq.l	#1,d0		;add the 1
    411  1.1  mycroft 	sub.l	d4,d0		;sub off LEN
    412  1.1  mycroft 	swap	d5		;use upper word of d5 for LAMBDA
    413  1.1  mycroft 	clr.w	d5		;set it zero initially
    414  1.1  mycroft 	clr.w	d2		;set up d2 for very small case
    415  1.1  mycroft 	tst.l	d0		;test sign of ISCALE
    416  1.1  mycroft 	bge.b	iscale		;if pos, skip next inst
    417  1.1  mycroft 	addq.w	#1,d5		;if neg, set LAMBDA true
    418  1.1  mycroft 	cmp.l	#$ffffecd4,d0	;test iscale <= -4908
    419  1.1  mycroft 	bgt.b	no_inf		;if false, skip rest
    420  1.1  mycroft 	addi.l	#24,d0		;add in 24 to iscale
    421  1.1  mycroft 	move.l	#24,d2		;put 24 in d2 for A9
    422  1.1  mycroft no_inf:	
    423  1.1  mycroft 	neg.l	d0		;and take abs of ISCALE
    424  1.1  mycroft iscale:	
    425  1.1  mycroft 	fmove.s	FONE,fp1	;init fp1 to 1
    426  1.1  mycroft 	bfextu	USER_FPCR(a6){26:2},d1 ;get initial rmode bits
    427  1.2  mycroft 	add.w	d1,d1		;put them in bits 2:1
    428  1.1  mycroft 	add.w	d5,d1		;add in LAMBDA
    429  1.2  mycroft 	add.w	d1,d1		;put them in bits 3:1
    430  1.1  mycroft 	tst.l	L_SCR2(a6)	;test sign of original x
    431  1.1  mycroft 	bge.b	x_pos		;if pos, don't set bit 0
    432  1.1  mycroft 	addq.l	#1,d1		;if neg, set bit 0
    433  1.1  mycroft x_pos:
    434  1.1  mycroft 	lea.l	RBDTBL,a2	;load rbdtbl base
    435  1.1  mycroft 	move.b	(a2,d1),d3	;load d3 with new rmode
    436  1.1  mycroft 	lsl.l	#4,d3		;put bits in proper position
    437  1.1  mycroft 	fmove.l	d3,fpcr		;load bits into fpu
    438  1.1  mycroft 	lsr.l	#4,d3		;put bits in proper position
    439  1.1  mycroft 	tst.b	d3		;decode new rmode for pten table
    440  1.1  mycroft 	bne.b	not_rn		;if zero, it is RN
    441  1.1  mycroft 	lea.l	PTENRN,a1	;load a1 with RN table base
    442  1.1  mycroft 	bra.b	rmode		;exit decode
    443  1.1  mycroft not_rn:
    444  1.1  mycroft 	lsr.b	#1,d3		;get lsb in carry
    445  1.1  mycroft 	bcc.b	not_rp		;if carry clear, it is RM
    446  1.1  mycroft 	lea.l	PTENRP,a1	;load a1 with RP table base
    447  1.1  mycroft 	bra.b	rmode		;exit decode
    448  1.1  mycroft not_rp:
    449  1.1  mycroft 	lea.l	PTENRM,a1	;load a1 with RM table base
    450  1.1  mycroft rmode:
    451  1.1  mycroft 	clr.l	d3		;clr table index
    452  1.1  mycroft e_loop:	
    453  1.1  mycroft 	lsr.l	#1,d0		;shift next bit into carry
    454  1.1  mycroft 	bcc.b	e_next		;if zero, skip the mul
    455  1.1  mycroft 	fmul.x	(a1,d3),fp1	;mul by 10**(d3_bit_no)
    456  1.1  mycroft e_next:	
    457  1.1  mycroft 	add.l	#12,d3		;inc d3 to next pwrten table entry
    458  1.1  mycroft 	tst.l	d0		;test if ISCALE is zero
    459  1.1  mycroft 	bne.b	e_loop		;if not, loop
    460  1.1  mycroft 
    461  1.1  mycroft 
    462  1.1  mycroft * A8. Clr INEX; Force RZ.
    463  1.1  mycroft *     The operation in A3 above may have set INEX2.  
    464  1.1  mycroft *     RZ mode is forced for the scaling operation to insure
    465  1.1  mycroft *     only one rounding error.  The grs bits are collected in 
    466  1.1  mycroft *     the INEX flag for use in A10.
    467  1.1  mycroft *
    468  1.1  mycroft * Register usage:
    469  1.1  mycroft *	Input/Output
    470  1.1  mycroft 
    471  1.1  mycroft 	fmove.l	#0,FPSR		;clr INEX 
    472  1.1  mycroft 	fmove.l	#rz_mode,FPCR	;set RZ rounding mode
    473  1.1  mycroft 
    474  1.1  mycroft 
    475  1.1  mycroft * A9. Scale X -> Y.
    476  1.1  mycroft *     The mantissa is scaled to the desired number of significant
    477  1.1  mycroft *     digits.  The excess digits are collected in INEX2. If mul,
    478  1.1  mycroft *     Check d2 for excess 10 exponential value.  If not zero, 
    479  1.1  mycroft *     the iscale value would have caused the pwrten calculation
    480  1.1  mycroft *     to overflow.  Only a negative iscale can cause this, so
    481  1.1  mycroft *     multiply by 10^(d2), which is now only allowed to be 24,
    482  1.1  mycroft *     with a multiply by 10^8 and 10^16, which is exact since
    483  1.1  mycroft *     10^24 is exact.  If the input was denormalized, we must
    484  1.1  mycroft *     create a busy stack frame with the mul command and the
    485  1.1  mycroft *     two operands, and allow the fpu to complete the multiply.
    486  1.1  mycroft *
    487  1.1  mycroft * Register usage:
    488  1.1  mycroft *	Input/Output
    489  1.1  mycroft *	d0: FPCR with RZ mode/Unchanged
    490  1.1  mycroft *	d2: 0 or 24/unchanged
    491  1.1  mycroft *	d3: x/x
    492  1.1  mycroft *	d4: LEN/Unchanged
    493  1.1  mycroft *	d5: ICTR:LAMBDA
    494  1.1  mycroft *	d6: ILOG/Unchanged
    495  1.1  mycroft *	d7: k-factor/Unchanged
    496  1.1  mycroft *	a0: ptr for original operand/final result
    497  1.1  mycroft *	a1: ptr to PTENRM array/Unchanged
    498  1.1  mycroft *	a2: x/x
    499  1.1  mycroft *	fp0: float(ILOG)/X adjusted for SCALE (Y)
    500  1.1  mycroft *	fp1: 10^ISCALE/Unchanged
    501  1.1  mycroft *	fp2: x/x
    502  1.1  mycroft *	F_SCR1:x/x
    503  1.1  mycroft *	F_SCR2:Abs(X) with $3fff exponent/Unchanged
    504  1.1  mycroft *	L_SCR1:x/x
    505  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    506  1.1  mycroft 
    507  1.1  mycroft A9_str:	
    508  1.1  mycroft 	fmove.x	(a0),fp0	;load X from memory
    509  1.1  mycroft 	fabs.x	fp0		;use abs(X)
    510  1.1  mycroft 	tst.w	d5		;LAMBDA is in lower word of d5
    511  1.1  mycroft 	bne.b	sc_mul		;if neg (LAMBDA = 1), scale by mul
    512  1.1  mycroft 	fdiv.x	fp1,fp0		;calculate X / SCALE -> Y to fp0
    513  1.1  mycroft 	bra.b	A10_st		;branch to A10
    514  1.1  mycroft 
    515  1.1  mycroft sc_mul:
    516  1.1  mycroft 	tst.b	BINDEC_FLG(a6)	;check for denorm
    517  1.1  mycroft 	beq.b	A9_norm		;if norm, continue with mul
    518  1.1  mycroft 	fmovem.x fp1,-(a7)	;load ETEMP with 10^ISCALE
    519  1.1  mycroft 	move.l	8(a0),-(a7)	;load FPTEMP with input arg
    520  1.1  mycroft 	move.l	4(a0),-(a7)
    521  1.1  mycroft 	move.l	(a0),-(a7)
    522  1.1  mycroft 	move.l	#18,d3		;load count for busy stack
    523  1.1  mycroft A9_loop:
    524  1.1  mycroft 	clr.l	-(a7)		;clear lword on stack
    525  1.1  mycroft 	dbf.w	d3,A9_loop	
    526  1.1  mycroft 	move.b	VER_TMP(a6),(a7) ;write current version number
    527  1.1  mycroft 	move.b	#BUSY_SIZE-4,1(a7) ;write current busy size 
    528  1.1  mycroft 	move.b	#$10,$44(a7)	;set fcefpte[15] bit
    529  1.1  mycroft 	move.w	#$0023,$40(a7)	;load cmdreg1b with mul command
    530  1.1  mycroft 	move.b	#$fe,$8(a7)	;load all 1s to cu savepc
    531  1.1  mycroft 	frestore (a7)+		;restore frame to fpu for completion
    532  1.1  mycroft 	fmul.x	36(a1),fp0	;multiply fp0 by 10^8
    533  1.1  mycroft 	fmul.x	48(a1),fp0	;multiply fp0 by 10^16
    534  1.1  mycroft 	bra.b	A10_st
    535  1.1  mycroft A9_norm:
    536  1.1  mycroft 	tst.w	d2		;test for small exp case
    537  1.1  mycroft 	beq.b	A9_con		;if zero, continue as normal
    538  1.1  mycroft 	fmul.x	36(a1),fp0	;multiply fp0 by 10^8
    539  1.1  mycroft 	fmul.x	48(a1),fp0	;multiply fp0 by 10^16
    540  1.1  mycroft A9_con:
    541  1.1  mycroft 	fmul.x	fp1,fp0		;calculate X * SCALE -> Y to fp0
    542  1.1  mycroft 
    543  1.1  mycroft 
    544  1.1  mycroft * A10. Or in INEX.
    545  1.1  mycroft *      If INEX is set, round error occured.  This is compensated
    546  1.1  mycroft *      for by 'or-ing' in the INEX2 flag to the lsb of Y.
    547  1.1  mycroft *
    548  1.1  mycroft * Register usage:
    549  1.1  mycroft *	Input/Output
    550  1.1  mycroft *	d0: FPCR with RZ mode/FPSR with INEX2 isolated
    551  1.1  mycroft *	d2: x/x
    552  1.1  mycroft *	d3: x/x
    553  1.1  mycroft *	d4: LEN/Unchanged
    554  1.1  mycroft *	d5: ICTR:LAMBDA
    555  1.1  mycroft *	d6: ILOG/Unchanged
    556  1.1  mycroft *	d7: k-factor/Unchanged
    557  1.1  mycroft *	a0: ptr for original operand/final result
    558  1.1  mycroft *	a1: ptr to PTENxx array/Unchanged
    559  1.1  mycroft *	a2: x/ptr to FP_SCR2(a6)
    560  1.1  mycroft *	fp0: Y/Y with lsb adjusted
    561  1.1  mycroft *	fp1: 10^ISCALE/Unchanged
    562  1.1  mycroft *	fp2: x/x
    563  1.1  mycroft 
    564  1.1  mycroft A10_st:	
    565  1.1  mycroft 	fmove.l	FPSR,d0		;get FPSR
    566  1.1  mycroft 	fmove.x	fp0,FP_SCR2(a6)	;move Y to memory
    567  1.1  mycroft 	lea.l	FP_SCR2(a6),a2	;load a2 with ptr to FP_SCR2
    568  1.1  mycroft 	btst.l	#9,d0		;check if INEX2 set
    569  1.1  mycroft 	beq.b	A11_st		;if clear, skip rest
    570  1.1  mycroft 	ori.l	#1,8(a2)	;or in 1 to lsb of mantissa
    571  1.1  mycroft 	fmove.x	FP_SCR2(a6),fp0	;write adjusted Y back to fpu
    572  1.1  mycroft 
    573  1.1  mycroft 
    574  1.1  mycroft * A11. Restore original FPCR; set size ext.
    575  1.1  mycroft *      Perform FINT operation in the user's rounding mode.  Keep
    576  1.1  mycroft *      the size to extended.  The sintdo entry point in the sint
    577  1.1  mycroft *      routine expects the FPCR value to be in USER_FPCR for
    578  1.1  mycroft *      mode and precision.  The original FPCR is saved in L_SCR1.
    579  1.1  mycroft 
    580  1.1  mycroft A11_st:	
    581  1.1  mycroft 	move.l	USER_FPCR(a6),L_SCR1(a6) ;save it for later
    582  1.1  mycroft 	andi.l	#$00000030,USER_FPCR(a6) ;set size to ext, 
    583  1.1  mycroft *					;block exceptions
    584  1.1  mycroft 
    585  1.1  mycroft 
    586  1.1  mycroft * A12. Calculate YINT = FINT(Y) according to user's rounding mode.
    587  1.1  mycroft *      The FPSP routine sintd0 is used.  The output is in fp0.
    588  1.1  mycroft *
    589  1.1  mycroft * Register usage:
    590  1.1  mycroft *	Input/Output
    591  1.1  mycroft *	d0: FPSR with AINEX cleared/FPCR with size set to ext
    592  1.1  mycroft *	d2: x/x/scratch
    593  1.1  mycroft *	d3: x/x
    594  1.1  mycroft *	d4: LEN/Unchanged
    595  1.1  mycroft *	d5: ICTR:LAMBDA/Unchanged
    596  1.1  mycroft *	d6: ILOG/Unchanged
    597  1.1  mycroft *	d7: k-factor/Unchanged
    598  1.1  mycroft *	a0: ptr for original operand/src ptr for sintdo
    599  1.1  mycroft *	a1: ptr to PTENxx array/Unchanged
    600  1.1  mycroft *	a2: ptr to FP_SCR2(a6)/Unchanged
    601  1.1  mycroft *	a6: temp pointer to FP_SCR2(a6) - orig value saved and restored
    602  1.1  mycroft *	fp0: Y/YINT
    603  1.1  mycroft *	fp1: 10^ISCALE/Unchanged
    604  1.1  mycroft *	fp2: x/x
    605  1.1  mycroft *	F_SCR1:x/x
    606  1.1  mycroft *	F_SCR2:Y adjusted for inex/Y with original exponent
    607  1.1  mycroft *	L_SCR1:x/original USER_FPCR
    608  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    609  1.1  mycroft 
    610  1.1  mycroft A12_st:
    611  1.1  mycroft 	movem.l	d0-d1/a0-a1,-(a7)	;save regs used by sintd0	
    612  1.1  mycroft 	move.l	L_SCR1(a6),-(a7)
    613  1.1  mycroft 	move.l	L_SCR2(a6),-(a7)
    614  1.1  mycroft 	lea.l	FP_SCR2(a6),a0		;a0 is ptr to F_SCR2(a6)
    615  1.1  mycroft 	fmove.x	fp0,(a0)		;move Y to memory at FP_SCR2(a6)
    616  1.1  mycroft 	tst.l	L_SCR2(a6)		;test sign of original operand
    617  1.1  mycroft 	bge.b	do_fint			;if pos, use Y 
    618  1.1  mycroft 	or.l	#$80000000,(a0)		;if neg, use -Y
    619  1.1  mycroft do_fint:
    620  1.1  mycroft 	move.l	USER_FPSR(a6),-(a7)
    621  1.1  mycroft 	bsr	sintdo			;sint routine returns int in fp0
    622  1.1  mycroft 	move.b	(a7),USER_FPSR(a6)
    623  1.1  mycroft 	add.l	#4,a7
    624  1.1  mycroft 	move.l	(a7)+,L_SCR2(a6)
    625  1.1  mycroft 	move.l	(a7)+,L_SCR1(a6)
    626  1.1  mycroft 	movem.l	(a7)+,d0-d1/a0-a1	;restore regs used by sint	
    627  1.1  mycroft 	move.l	L_SCR2(a6),FP_SCR2(a6)	;restore original exponent
    628  1.1  mycroft 	move.l	L_SCR1(a6),USER_FPCR(a6) ;restore user's FPCR
    629  1.1  mycroft 
    630  1.1  mycroft 
    631  1.1  mycroft * A13. Check for LEN digits.
    632  1.1  mycroft *      If the int operation results in more than LEN digits,
    633  1.1  mycroft *      or less than LEN -1 digits, adjust ILOG and repeat from
    634  1.1  mycroft *      A6.  This test occurs only on the first pass.  If the
    635  1.1  mycroft *      result is exactly 10^LEN, decrement ILOG and divide
    636  1.1  mycroft *      the mantissa by 10.  The calculation of 10^LEN cannot
    637  1.1  mycroft *      be inexact, since all powers of ten upto 10^27 are exact
    638  1.1  mycroft *      in extended precision, so the use of a previous power-of-ten
    639  1.1  mycroft *      table will introduce no error.
    640  1.1  mycroft *
    641  1.1  mycroft *
    642  1.1  mycroft * Register usage:
    643  1.1  mycroft *	Input/Output
    644  1.1  mycroft *	d0: FPCR with size set to ext/scratch final = 0
    645  1.1  mycroft *	d2: x/x
    646  1.1  mycroft *	d3: x/scratch final = x
    647  1.1  mycroft *	d4: LEN/LEN adjusted
    648  1.1  mycroft *	d5: ICTR:LAMBDA/LAMBDA:ICTR
    649  1.1  mycroft *	d6: ILOG/ILOG adjusted
    650  1.1  mycroft *	d7: k-factor/Unchanged
    651  1.1  mycroft *	a0: pointer into memory for packed bcd string formation
    652  1.1  mycroft *	a1: ptr to PTENxx array/Unchanged
    653  1.1  mycroft *	a2: ptr to FP_SCR2(a6)/Unchanged
    654  1.1  mycroft *	fp0: int portion of Y/abs(YINT) adjusted
    655  1.1  mycroft *	fp1: 10^ISCALE/Unchanged
    656  1.1  mycroft *	fp2: x/10^LEN
    657  1.1  mycroft *	F_SCR1:x/x
    658  1.1  mycroft *	F_SCR2:Y with original exponent/Unchanged
    659  1.1  mycroft *	L_SCR1:original USER_FPCR/Unchanged
    660  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    661  1.1  mycroft 
    662  1.1  mycroft A13_st:	
    663  1.1  mycroft 	swap	d5		;put ICTR in lower word of d5
    664  1.1  mycroft 	tst.w	d5		;check if ICTR = 0
    665  1.1  mycroft 	bne	not_zr		;if non-zero, go to second test
    666  1.1  mycroft *
    667  1.1  mycroft * Compute 10^(LEN-1)
    668  1.1  mycroft *
    669  1.1  mycroft 	fmove.s	FONE,fp2	;init fp2 to 1.0
    670  1.1  mycroft 	move.l	d4,d0		;put LEN in d0
    671  1.1  mycroft 	subq.l	#1,d0		;d0 = LEN -1
    672  1.1  mycroft 	clr.l	d3		;clr table index
    673  1.1  mycroft l_loop:	
    674  1.1  mycroft 	lsr.l	#1,d0		;shift next bit into carry
    675  1.1  mycroft 	bcc.b	l_next		;if zero, skip the mul
    676  1.1  mycroft 	fmul.x	(a1,d3),fp2	;mul by 10**(d3_bit_no)
    677  1.1  mycroft l_next:
    678  1.1  mycroft 	add.l	#12,d3		;inc d3 to next pwrten table entry
    679  1.1  mycroft 	tst.l	d0		;test if LEN is zero
    680  1.1  mycroft 	bne.b	l_loop		;if not, loop
    681  1.1  mycroft *
    682  1.1  mycroft * 10^LEN-1 is computed for this test and A14.  If the input was
    683  1.1  mycroft * denormalized, check only the case in which YINT > 10^LEN.
    684  1.1  mycroft *
    685  1.1  mycroft 	tst.b	BINDEC_FLG(a6)	;check if input was norm
    686  1.1  mycroft 	beq.b	A13_con		;if norm, continue with checking
    687  1.1  mycroft 	fabs.x	fp0		;take abs of YINT
    688  1.1  mycroft 	bra	test_2
    689  1.1  mycroft *
    690  1.1  mycroft * Compare abs(YINT) to 10^(LEN-1) and 10^LEN
    691  1.1  mycroft *
    692  1.1  mycroft A13_con:
    693  1.1  mycroft 	fabs.x	fp0		;take abs of YINT
    694  1.1  mycroft 	fcmp.x	fp2,fp0		;compare abs(YINT) with 10^(LEN-1)
    695  1.1  mycroft 	fbge.w	test_2		;if greater, do next test
    696  1.1  mycroft 	subq.l	#1,d6		;subtract 1 from ILOG
    697  1.1  mycroft 	move.w	#1,d5		;set ICTR
    698  1.1  mycroft 	fmove.l	#rm_mode,FPCR	;set rmode to RM
    699  1.1  mycroft 	fmul.s	FTEN,fp2	;compute 10^LEN 
    700  1.1  mycroft 	bra.w	A6_str		;return to A6 and recompute YINT
    701  1.1  mycroft test_2:
    702  1.1  mycroft 	fmul.s	FTEN,fp2	;compute 10^LEN
    703  1.1  mycroft 	fcmp.x	fp2,fp0		;compare abs(YINT) with 10^LEN
    704  1.1  mycroft 	fblt.w	A14_st		;if less, all is ok, go to A14
    705  1.1  mycroft 	fbgt.w	fix_ex		;if greater, fix and redo
    706  1.1  mycroft 	fdiv.s	FTEN,fp0	;if equal, divide by 10
    707  1.1  mycroft 	addq.l	#1,d6		; and inc ILOG
    708  1.1  mycroft 	bra.b	A14_st		; and continue elsewhere
    709  1.1  mycroft fix_ex:
    710  1.1  mycroft 	addq.l	#1,d6		;increment ILOG by 1
    711  1.1  mycroft 	move.w	#1,d5		;set ICTR
    712  1.1  mycroft 	fmove.l	#rm_mode,FPCR	;set rmode to RM
    713  1.1  mycroft 	bra.w	A6_str		;return to A6 and recompute YINT
    714  1.1  mycroft *
    715  1.1  mycroft * Since ICTR <> 0, we have already been through one adjustment, 
    716  1.1  mycroft * and shouldn't have another; this is to check if abs(YINT) = 10^LEN
    717  1.1  mycroft * 10^LEN is again computed using whatever table is in a1 since the
    718  1.1  mycroft * value calculated cannot be inexact.
    719  1.1  mycroft *
    720  1.1  mycroft not_zr:
    721  1.1  mycroft 	fmove.s	FONE,fp2	;init fp2 to 1.0
    722  1.1  mycroft 	move.l	d4,d0		;put LEN in d0
    723  1.1  mycroft 	clr.l	d3		;clr table index
    724  1.1  mycroft z_loop:
    725  1.1  mycroft 	lsr.l	#1,d0		;shift next bit into carry
    726  1.1  mycroft 	bcc.b	z_next		;if zero, skip the mul
    727  1.1  mycroft 	fmul.x	(a1,d3),fp2	;mul by 10**(d3_bit_no)
    728  1.1  mycroft z_next:
    729  1.1  mycroft 	add.l	#12,d3		;inc d3 to next pwrten table entry
    730  1.1  mycroft 	tst.l	d0		;test if LEN is zero
    731  1.1  mycroft 	bne.b	z_loop		;if not, loop
    732  1.1  mycroft 	fabs.x	fp0		;get abs(YINT)
    733  1.1  mycroft 	fcmp.x	fp2,fp0		;check if abs(YINT) = 10^LEN
    734  1.1  mycroft 	fbne.w	A14_st		;if not, skip this
    735  1.1  mycroft 	fdiv.s	FTEN,fp0	;divide abs(YINT) by 10
    736  1.1  mycroft 	addq.l	#1,d6		;and inc ILOG by 1
    737  1.1  mycroft 	addq.l	#1,d4		; and inc LEN
    738  1.1  mycroft 	fmul.s	FTEN,fp2	; if LEN++, the get 10^^LEN
    739  1.1  mycroft 
    740  1.1  mycroft 
    741  1.1  mycroft * A14. Convert the mantissa to bcd.
    742  1.1  mycroft *      The binstr routine is used to convert the LEN digit 
    743  1.1  mycroft *      mantissa to bcd in memory.  The input to binstr is
    744  1.1  mycroft *      to be a fraction; i.e. (mantissa)/10^LEN and adjusted
    745  1.1  mycroft *      such that the decimal point is to the left of bit 63.
    746  1.1  mycroft *      The bcd digits are stored in the correct position in 
    747  1.1  mycroft *      the final string area in memory.
    748  1.1  mycroft *
    749  1.1  mycroft *
    750  1.1  mycroft * Register usage:
    751  1.1  mycroft *	Input/Output
    752  1.1  mycroft *	d0: x/LEN call to binstr - final is 0
    753  1.1  mycroft *	d1: x/0
    754  1.1  mycroft *	d2: x/ms 32-bits of mant of abs(YINT)
    755  1.1  mycroft *	d3: x/ls 32-bits of mant of abs(YINT)
    756  1.1  mycroft *	d4: LEN/Unchanged
    757  1.1  mycroft *	d5: ICTR:LAMBDA/LAMBDA:ICTR
    758  1.1  mycroft *	d6: ILOG
    759  1.1  mycroft *	d7: k-factor/Unchanged
    760  1.1  mycroft *	a0: pointer into memory for packed bcd string formation
    761  1.1  mycroft *	    /ptr to first mantissa byte in result string
    762  1.1  mycroft *	a1: ptr to PTENxx array/Unchanged
    763  1.1  mycroft *	a2: ptr to FP_SCR2(a6)/Unchanged
    764  1.1  mycroft *	fp0: int portion of Y/abs(YINT) adjusted
    765  1.1  mycroft *	fp1: 10^ISCALE/Unchanged
    766  1.1  mycroft *	fp2: 10^LEN/Unchanged
    767  1.1  mycroft *	F_SCR1:x/Work area for final result
    768  1.1  mycroft *	F_SCR2:Y with original exponent/Unchanged
    769  1.1  mycroft *	L_SCR1:original USER_FPCR/Unchanged
    770  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    771  1.1  mycroft 
    772  1.1  mycroft A14_st:	
    773  1.1  mycroft 	fmove.l	#rz_mode,FPCR	;force rz for conversion
    774  1.1  mycroft 	fdiv.x	fp2,fp0		;divide abs(YINT) by 10^LEN
    775  1.1  mycroft 	lea.l	FP_SCR1(a6),a0
    776  1.1  mycroft 	fmove.x	fp0,(a0)	;move abs(YINT)/10^LEN to memory
    777  1.1  mycroft 	move.l	4(a0),d2	;move 2nd word of FP_RES to d2
    778  1.1  mycroft 	move.l	8(a0),d3	;move 3rd word of FP_RES to d3
    779  1.1  mycroft 	clr.l	4(a0)		;zero word 2 of FP_RES
    780  1.1  mycroft 	clr.l	8(a0)		;zero word 3 of FP_RES
    781  1.1  mycroft 	move.l	(a0),d0		;move exponent to d0
    782  1.1  mycroft 	swap	d0		;put exponent in lower word
    783  1.1  mycroft 	beq.b	no_sft		;if zero, don't shift
    784  1.1  mycroft 	subi.l	#$3ffd,d0	;sub bias less 2 to make fract
    785  1.1  mycroft 	tst.l	d0		;check if > 1
    786  1.1  mycroft 	bgt.b	no_sft		;if so, don't shift
    787  1.1  mycroft 	neg.l	d0		;make exp positive
    788  1.1  mycroft m_loop:
    789  1.1  mycroft 	lsr.l	#1,d2		;shift d2:d3 right, add 0s 
    790  1.1  mycroft 	roxr.l	#1,d3		;the number of places
    791  1.1  mycroft 	dbf.w	d0,m_loop	;given in d0
    792  1.1  mycroft no_sft:
    793  1.1  mycroft 	tst.l	d2		;check for mantissa of zero
    794  1.1  mycroft 	bne.b	no_zr		;if not, go on
    795  1.1  mycroft 	tst.l	d3		;continue zero check
    796  1.1  mycroft 	beq.b	zer_m		;if zero, go directly to binstr
    797  1.1  mycroft no_zr:
    798  1.1  mycroft 	clr.l	d1		;put zero in d1 for addx
    799  1.1  mycroft 	addi.l	#$00000080,d3	;inc at bit 7
    800  1.1  mycroft 	addx.l	d1,d2		;continue inc
    801  1.1  mycroft 	andi.l	#$ffffff80,d3	;strip off lsb not used by 882
    802  1.1  mycroft zer_m:
    803  1.1  mycroft 	move.l	d4,d0		;put LEN in d0 for binstr call
    804  1.1  mycroft 	addq.l	#3,a0		;a0 points to M16 byte in result
    805  1.1  mycroft 	bsr	binstr		;call binstr to convert mant
    806  1.1  mycroft 
    807  1.1  mycroft 
    808  1.1  mycroft * A15. Convert the exponent to bcd.
    809  1.1  mycroft *      As in A14 above, the exp is converted to bcd and the
    810  1.1  mycroft *      digits are stored in the final string.
    811  1.1  mycroft *
    812  1.1  mycroft *      Digits are stored in L_SCR1(a6) on return from BINDEC as:
    813  1.1  mycroft *
    814  1.1  mycroft *  	 32               16 15                0
    815  1.1  mycroft *	-----------------------------------------
    816  1.1  mycroft *  	|  0 | e3 | e2 | e1 | e4 |  X |  X |  X |
    817  1.1  mycroft *	-----------------------------------------
    818  1.1  mycroft *
    819  1.1  mycroft * And are moved into their proper places in FP_SCR1.  If digit e4
    820  1.1  mycroft * is non-zero, OPERR is signaled.  In all cases, all 4 digits are
    821  1.1  mycroft * written as specified in the 881/882 manual for packed decimal.
    822  1.1  mycroft *
    823  1.1  mycroft * Register usage:
    824  1.1  mycroft *	Input/Output
    825  1.1  mycroft *	d0: x/LEN call to binstr - final is 0
    826  1.1  mycroft *	d1: x/scratch (0);shift count for final exponent packing
    827  1.1  mycroft *	d2: x/ms 32-bits of exp fraction/scratch
    828  1.1  mycroft *	d3: x/ls 32-bits of exp fraction
    829  1.1  mycroft *	d4: LEN/Unchanged
    830  1.1  mycroft *	d5: ICTR:LAMBDA/LAMBDA:ICTR
    831  1.1  mycroft *	d6: ILOG
    832  1.1  mycroft *	d7: k-factor/Unchanged
    833  1.1  mycroft *	a0: ptr to result string/ptr to L_SCR1(a6)
    834  1.1  mycroft *	a1: ptr to PTENxx array/Unchanged
    835  1.1  mycroft *	a2: ptr to FP_SCR2(a6)/Unchanged
    836  1.1  mycroft *	fp0: abs(YINT) adjusted/float(ILOG)
    837  1.1  mycroft *	fp1: 10^ISCALE/Unchanged
    838  1.1  mycroft *	fp2: 10^LEN/Unchanged
    839  1.1  mycroft *	F_SCR1:Work area for final result/BCD result
    840  1.1  mycroft *	F_SCR2:Y with original exponent/ILOG/10^4
    841  1.1  mycroft *	L_SCR1:original USER_FPCR/Exponent digits on return from binstr
    842  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    843  1.1  mycroft 
    844  1.1  mycroft A15_st:	
    845  1.1  mycroft 	tst.b	BINDEC_FLG(a6)	;check for denorm
    846  1.1  mycroft 	beq.b	not_denorm
    847  1.1  mycroft 	ftst.x	fp0		;test for zero
    848  1.1  mycroft 	fbeq.w	den_zero	;if zero, use k-factor or 4933
    849  1.1  mycroft 	fmove.l	d6,fp0		;float ILOG
    850  1.1  mycroft 	fabs.x	fp0		;get abs of ILOG
    851  1.1  mycroft 	bra.b	convrt
    852  1.1  mycroft den_zero:
    853  1.1  mycroft 	tst.l	d7		;check sign of the k-factor
    854  1.1  mycroft 	blt.b	use_ilog	;if negative, use ILOG
    855  1.1  mycroft 	fmove.s	F4933,fp0	;force exponent to 4933
    856  1.1  mycroft 	bra.b	convrt		;do it
    857  1.1  mycroft use_ilog:
    858  1.1  mycroft 	fmove.l	d6,fp0		;float ILOG
    859  1.1  mycroft 	fabs.x	fp0		;get abs of ILOG
    860  1.1  mycroft 	bra.b	convrt
    861  1.1  mycroft not_denorm:
    862  1.1  mycroft 	ftst.x	fp0		;test for zero
    863  1.1  mycroft 	fbne.w	not_zero	;if zero, force exponent
    864  1.1  mycroft 	fmove.s	FONE,fp0	;force exponent to 1
    865  1.1  mycroft 	bra.b	convrt		;do it
    866  1.1  mycroft not_zero:	
    867  1.1  mycroft 	fmove.l	d6,fp0		;float ILOG
    868  1.1  mycroft 	fabs.x	fp0		;get abs of ILOG
    869  1.1  mycroft convrt:
    870  1.1  mycroft 	fdiv.x	24(a1),fp0	;compute ILOG/10^4
    871  1.1  mycroft 	fmove.x	fp0,FP_SCR2(a6)	;store fp0 in memory
    872  1.1  mycroft 	move.l	4(a2),d2	;move word 2 to d2
    873  1.1  mycroft 	move.l	8(a2),d3	;move word 3 to d3
    874  1.1  mycroft 	move.w	(a2),d0		;move exp to d0
    875  1.1  mycroft 	beq.b	x_loop_fin	;if zero, skip the shift
    876  1.1  mycroft 	subi.w	#$3ffd,d0	;subtract off bias
    877  1.1  mycroft 	neg.w	d0		;make exp positive
    878  1.1  mycroft x_loop:
    879  1.1  mycroft 	lsr.l	#1,d2		;shift d2:d3 right 
    880  1.1  mycroft 	roxr.l	#1,d3		;the number of places
    881  1.1  mycroft 	dbf.w	d0,x_loop	;given in d0
    882  1.1  mycroft x_loop_fin:
    883  1.1  mycroft 	clr.l	d1		;put zero in d1 for addx
    884  1.1  mycroft 	addi.l	#$00000080,d3	;inc at bit 6
    885  1.1  mycroft 	addx.l	d1,d2		;continue inc
    886  1.1  mycroft 	andi.l	#$ffffff80,d3	;strip off lsb not used by 882
    887  1.1  mycroft 	move.l	#4,d0		;put 4 in d0 for binstr call
    888  1.1  mycroft 	lea.l	L_SCR1(a6),a0	;a0 is ptr to L_SCR1 for exp digits
    889  1.1  mycroft 	bsr	binstr		;call binstr to convert exp
    890  1.1  mycroft 	move.l	L_SCR1(a6),d0	;load L_SCR1 lword to d0 
    891  1.1  mycroft 	move.l	#12,d1		;use d1 for shift count
    892  1.1  mycroft 	lsr.l	d1,d0		;shift d0 right by 12
    893  1.1  mycroft 	bfins	d0,FP_SCR1(a6){4:12} ;put e3:e2:e1 in FP_SCR1
    894  1.1  mycroft 	lsr.l	d1,d0		;shift d0 right by 12
    895  1.1  mycroft 	bfins	d0,FP_SCR1(a6){16:4} ;put e4 in FP_SCR1 
    896  1.1  mycroft 	tst.b	d0		;check if e4 is zero
    897  1.1  mycroft 	beq.b	A16_st		;if zero, skip rest
    898  1.1  mycroft 	or.l	#opaop_mask,USER_FPSR(a6) ;set OPERR & AIOP in USER_FPSR
    899  1.1  mycroft 
    900  1.1  mycroft 
    901  1.1  mycroft * A16. Write sign bits to final string.
    902  1.1  mycroft *	   Sigma is bit 31 of initial value; RHO is bit 31 of d6 (ILOG).
    903  1.1  mycroft *
    904  1.1  mycroft * Register usage:
    905  1.1  mycroft *	Input/Output
    906  1.1  mycroft *	d0: x/scratch - final is x
    907  1.1  mycroft *	d2: x/x
    908  1.1  mycroft *	d3: x/x
    909  1.1  mycroft *	d4: LEN/Unchanged
    910  1.1  mycroft *	d5: ICTR:LAMBDA/LAMBDA:ICTR
    911  1.1  mycroft *	d6: ILOG/ILOG adjusted
    912  1.1  mycroft *	d7: k-factor/Unchanged
    913  1.1  mycroft *	a0: ptr to L_SCR1(a6)/Unchanged
    914  1.1  mycroft *	a1: ptr to PTENxx array/Unchanged
    915  1.1  mycroft *	a2: ptr to FP_SCR2(a6)/Unchanged
    916  1.1  mycroft *	fp0: float(ILOG)/Unchanged
    917  1.1  mycroft *	fp1: 10^ISCALE/Unchanged
    918  1.1  mycroft *	fp2: 10^LEN/Unchanged
    919  1.1  mycroft *	F_SCR1:BCD result with correct signs
    920  1.1  mycroft *	F_SCR2:ILOG/10^4
    921  1.1  mycroft *	L_SCR1:Exponent digits on return from binstr
    922  1.1  mycroft *	L_SCR2:first word of X packed/Unchanged
    923  1.1  mycroft 
    924  1.1  mycroft A16_st:
    925  1.1  mycroft 	clr.l	d0		;clr d0 for collection of signs
    926  1.1  mycroft 	andi.b	#$0f,FP_SCR1(a6) ;clear first nibble of FP_SCR1 
    927  1.1  mycroft 	tst.l	L_SCR2(a6)	;check sign of original mantissa
    928  1.1  mycroft 	bge.b	mant_p		;if pos, don't set SM
    929  1.1  mycroft 	moveq.l	#2,d0		;move 2 in to d0 for SM
    930  1.1  mycroft mant_p:
    931  1.1  mycroft 	tst.l	d6		;check sign of ILOG
    932  1.1  mycroft 	bge.b	wr_sgn		;if pos, don't set SE
    933  1.1  mycroft 	addq.l	#1,d0		;set bit 0 in d0 for SE 
    934  1.1  mycroft wr_sgn:
    935  1.1  mycroft 	bfins	d0,FP_SCR1(a6){0:2} ;insert SM and SE into FP_SCR1
    936  1.1  mycroft 
    937  1.1  mycroft * Clean up and restore all registers used.
    938  1.1  mycroft 
    939  1.1  mycroft 	fmove.l	#0,FPSR		;clear possible inex2/ainex bits
    940  1.1  mycroft 	fmovem.x (a7)+,fp0-fp2
    941  1.1  mycroft 	movem.l	(a7)+,d2-d7/a2
    942  1.1  mycroft 	rts
    943  1.1  mycroft 
    944  1.1  mycroft 	end
    945