1: ; Copyright (c) 1981 Harvard-Radcliffe Student Timesharing System
   2: ; Science Center, Harvard University
   3: 
   4: ;this file has the routines to help out lisps that run on
   5: ;systems without floating hardware
   6: ;
   7: ; forrest howard, aug 75
   8: ;
   9:   .if ne,fpsim          ;bad name choice...
  10:         .globl  imul,idiv,sign,fpstuf
  11: 
  12: imul:
  13:         ;multiply the number in a+b * j1+j2
  14:         ;leave result in a+b
  15:         ;if v bit is set on return, then strict overflow
  16: 
  17: 
  18:         save1
  19:         call    fixsign         ;which leaves both args pos
  20:         mov     a,j3
  21:         mov     b,fpstuf
  22:         bit     #1,j2
  23:         bne     1$
  24:         clr     a
  25:         clr     b
  26: 1$:     bic     #1,j2
  27: 
  28: 
  29: domul:                  ;now we do stuff-- shift only when necessary
  30:         tst     j1
  31:         bne     1$
  32:         tst     j2
  33:         beq     donmul
  34: 1$:     asl     fpstuf
  35:         rol     j3
  36:         bvs     ovr
  37:         ashc    #-1,j1
  38:         bit     #1,j2
  39:         beq     2$
  40:         dec     j2
  41:         add     fpstuf,b
  42:         adc     a
  43:         bvs     ovr
  44:         add     j3,a
  45:         bvs     ovr
  46: 2$:     br      domul
  47: 
  48: ovr:    mov     2(sp),j3
  49:         cmp     (sp)+,(sp)+
  50:         sev
  51:         ret
  52: donmul:
  53: dfinup: mov     2(sp),j3
  54:         cmp     (sp)+,(sp)+
  55:         tst     sign
  56:         beq     1$
  57:         call    negs
  58:         call    negr
  59: 1$:     tst     a
  60:         bne     3$
  61:         tst     b
  62:         cln             ;clear the n-bit if set (want pos result)
  63: 3$:
  64:         rts     pc
  65: 
  66: fixsign:
  67:         mov     #sign,j3
  68:         clrb    (j3)
  69:         tst     j1
  70:         bge     1$
  71:         comb    (j3)
  72:         call    negs
  73: 1$:
  74:         tst     a
  75:         bge     2$
  76:         comb    (j3)
  77: negr= .
  78:         com     a
  79:         com     b
  80:         add     #1,b
  81:         adc     a
  82: 2$:     ret
  83: 
  84: negs:
  85:         com     j1
  86:         com     j2
  87:         add     #1,j2
  88:         adc     j1
  89:         rts     pc
  90: 
  91: 
  92: 
  93: 
  94: idiv:           ;idiv divides (a+b)/(j1+j2)
  95:                 ;quoitient in a+b
  96:                 ;rem in j1+j2
  97:                 ;rem is same sign as quo
  98:                 ;v bit is set if overflow occurs
  99:                 ;or on divide check
 100:                 ;z bit is set according to the ans
 101: 
 102: 
 103:         save1
 104:         call    fixsign
 105:                         ;ok; now we have all registers, and
 106:                         ;can muck around
 107:                         ;first check for zero's
 108:         mov     j1,j3
 109:         bne     10$
 110:         tst     j2
 111:         beq     ovr
 112: 10$:    mov     j2,fpstuf
 113:         mov     a,j1
 114:         bne     11$
 115:         tst     b
 116:         beq     retz
 117: 11$:    mov     b,j2
 118:         clr     fpstuf+2
 119:                         ;left justify the two ints, keeping count
 120: 1$:     ashc    #1,j1
 121:         bvs     2$
 122:         dec     fpstuf+2
 123:         br      1$
 124: 2$:     ror     j1
 125:         ror     j2                      ;recover stuff...
 126:         mov     fpstuf+2,fpstuf+4       ;save to form remainder
 127: 3$:     asl     fpstuf
 128:         rol     j3
 129:         bvs     4$
 130:         inc     fpstuf+4
 131:         br      3$
 132: 4$:     ror     j3
 133:         ror     fpstuf
 134:         clr     a
 135:         clr     b                       ;clear for answer
 136: 
 137:         ;fpstuf+4 now has count of number of things.....
 138:         ;if it is neg, we can go home
 139: 
 140:         tst     fpstuf+4
 141:         blt     divret
 142: actdiv:
 143:         sub     fpstuf,j2
 144:         sbc     j1
 145:         sub     j3,j1
 146:         blt     1$
 147:         inc     b
 148:         br      2$
 149: 1$:     add     fpstuf,j2
 150:         adc     j1
 151:         add     j3,j1
 152: 2$:     asr     j3
 153:         ror     fpstuf          ;shift right for next
 154:         dec     fpstuf+4
 155:         blt     divret
 156:         ashc    #1,a            ;for next
 157:         br      actdiv
 158: divret:
 159:         ashc    fpstuf+2,j1
 160:         br      dfinup
 161: 
 162: 
 163: retz:   clr     a
 164:         clr     b
 165:         clr     j1
 166:         mov     2(sp),j3
 167:         cmp     (sp)+,(sp)+
 168:         clr     j2
 169:         ret
 170: 
 171: .endc
Last modified: 1981-05-10
Generated: 2016-12-26
Generated by src2html V0.67
page hit count: 361
Valid CSS Valid XHTML 1.0 Strict