![[BACK]](/cvsweb/icons/back.gif){kind=icon}
Return to doprnt.c CVS log ![[TXT]](/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/cvsweb/icons/dir.gif){kind=icon}
Up to [CSRG BSD Unix] / 40BSD / cmd / pxp|
Return to doprnt.c CVS log | Up to [CSRG BSD Unix] / 40BSD / cmd / pxp |
1.1 ! root 1: # C library -- conversions ! 2: ! 3: .globl __doprnt ! 4: .globl __strout ! 5: ! 6: #define flags r10 ! 7: #define literb 0 ! 8: #define liter 1 ! 9: #define ndfndb 0 ! 10: #define ndfnd 1 ! 11: #define ljustb 1 ! 12: #define ljust 2 ! 13: #define zfillb 2 ! 14: #define zfill 4 ! 15: #define precb 3 ! 16: #define prec 8 ! 17: #define psignb 4 ! 18: #define psign 16 ! 19: #define gflagb 5 ! 20: #define gflag 32 ! 21: #define width r9 ! 22: #define ndigit r8 ! 23: #define fdesc -4(fp) ! 24: #define exp -8(fp) ! 25: #define sign -9(fp) ! 26: .set one,010 # 1.0 in floating immediate ! 27: .set ch.zer,'0 # cpp doesn't like single appostrophes ! 28: ! 29: .align 1 ! 30: __doprnt: ! 31: .word 0xfc0 # uses r11-r6 ! 32: subl2 $128,sp ! 33: movl 4(ap),r11 # addr of format string ! 34: movl 12(ap),fdesc # output FILE ptr ! 35: movl 8(ap),ap # addr of first arg ! 36: loop: ! 37: movl r11,r0 # current point in format ! 38: bicl2 $liter,flags # no literal characters yet ! 39: L1: movb (r11)+,width # next character of format ! 40: beql L2 # end of format string ! 41: cmpb width,$'% ! 42: beql L2 # warning character ! 43: bisl2 $liter,flags # literal character ! 44: jbr L1 ! 45: L2: blbc flags,L3 # bbc $literb,flags,L3 # no literals in format ! 46: pushl fdesc # file pointer ! 47: pushl $0 # no left/right adjust ! 48: pushl r0 # addr ! 49: subl3 r0,r11,r1 # length ! 50: subl3 $1,r1,-(sp) # % or null not part of literal ! 51: calls $4,__strout # dump the literal ! 52: L3: ! 53: blbs width,L4 # % is odd; end of format? ! 54: ret # yes ! 55: ! 56: # htab overlaps last 16 characters of ftab ! 57: ftab: .byte 0, 0, 0,'c,'d,'e,'f,'g, 0, 0, 0,'+,'l,'-,'.,'o ! 58: htab: .byte '0,'1,'2,'3,'4,'5,'6,'7,'8,'9,'a,'b,'c,'d,'e,'f ! 59: ! 60: L4: movl sp,r5 # reset output buffer pointer ! 61: clrq r9 # width; flags ljustb,ndfndb,zfillb ! 62: L4a: movzbl (r11)+,r0 # supposed format ! 63: extzv $0,$5,r0,r1 # bottom 5 bits ! 64: L4b: cmpb r0,ftab[r1] # good enough? ! 65: jneq L6 # no ! 66: L4c: casel r1,$3,$22 # yes ! 67: L5: .word charac-L5 # c ! 68: .word decimal-L5 # d ! 69: .word scien-L5 # e ! 70: .word float-L5 # f ! 71: .word general-L5 # g ! 72: .word L6-L5 # h ! 73: .word L6-L5 # i ! 74: .word L6-L5 # j ! 75: .word plus-L5 # + ! 76: .word longorunsg-L5 # l ! 77: .word minus-L5 # - ! 78: .word dot-L5 # . ! 79: .word octal-L5 # o ! 80: .word gnum0-L5 # 0 ! 81: .word gnum-L5 # 1 ! 82: .word gnum-L5 # 2 ! 83: .word gnum-L5 # 3 ! 84: .word gnum-L5 # 4 ! 85: .word gnum-L5 # 5 ! 86: .word gnum-L5 # 6 ! 87: .word gnum-L5 # 7 ! 88: .word gnum-L5 # 8 ! 89: .word gnum-L5 # 9 ! 90: ! 91: L6: jbcs $5,r0,L4b # capitals same as small ! 92: cmpb r0,$'s ! 93: jeql string ! 94: cmpb r0,$'x ! 95: jeql hex ! 96: cmpb r0,$'u ! 97: jeql unsigned ! 98: cmpb r0,$'r ! 99: jeql remote ! 100: movzbl -1(r11),r0 # orginal "format" character ! 101: cmpb r0,$'* ! 102: jeql indir ! 103: L9: movb r0,(r5)+ # print the unfound character ! 104: jbr prbuf ! 105: ! 106: nulstr: ! 107: .byte '(,'n,'u,'l,'l,'),0 ! 108: ! 109: string: ! 110: movl ndigit,r0 ! 111: jbs $precb,flags,L20 # max length was specified ! 112: mnegl $1,r0 # default max length ! 113: L20: movl (ap)+,r2 # addr first byte ! 114: bneq L21 ! 115: movab nulstr,r2 ! 116: L21: locc $0,r0,(r2) # find the zero at the end ! 117: movl r1,r5 # addr last byte +1 ! 118: movl r2,r1 # addr first byte ! 119: jbr prstr ! 120: ! 121: ! 122: longorunsg: ! 123: movb (r11)+,r0 ! 124: cmpb r0,$'o ! 125: jeql loct ! 126: cmpb r0,$'x ! 127: jeql lhex ! 128: cmpb r0,$'d ! 129: jeql long ! 130: cmpb r0,$'u ! 131: jeql lunsigned ! 132: decl r11 ! 133: jbr unsigned ! 134: ! 135: loct: ! 136: octal: ! 137: movl $30,r2 # init position ! 138: movl $3,r3 # field width ! 139: movl $10,r4 # result length -1 ! 140: jbr L10 ! 141: ! 142: lhex: ! 143: hex: ! 144: movl $28,r2 # init position ! 145: movl $4,r3 # field width ! 146: movl $7,r4 # result length -1 ! 147: L10: mnegl r3,r6 # increment ! 148: clrl r1 ! 149: movl (ap)+,r0 # fetch arg ! 150: L11: extzv r2,r3,r0,r1 # pull out a digit ! 151: movb htab[r1],(r5)+ # convert to character ! 152: L12: acbl $0,r6,r2,L11 # continue until done ! 153: clrb (r5) # flag end ! 154: skpc $'0,r4,(sp) # skip over leading zeroes ! 155: jbr prstr ! 156: ! 157: patdec: # editpc pattern for decimal printing ! 158: .byte 0xA9 # eo$float 9 ! 159: .byte 0x01 # eo$end_float ! 160: .byte 0x91 # eo$move 1 ! 161: .byte 0 # eo$end ! 162: ! 163: long: ! 164: decimal: ! 165: cvtlp (ap)+,$10,(sp) # 10 digits max ! 166: L14: editpc $10,(sp),patdec,8(sp) # ascii at 8(sp); r5=end+1 ! 167: skpc $' ,$10,8(sp) # skip leading blanks; r1=first ! 168: ! 169: prstr: # r1=addr first byte; r5=addr last byte +1 ! 170: cvtbl $' ,-(sp) # blank fill ! 171: jbc $zfillb,flags,L15 ! 172: cvtbl $'0,(sp) # zero fill ! 173: L15: pushl fdesc # FILE ! 174: subl2 r1,r5 # r5=actual length=end+1-first ! 175: subl3 r5,width,r0 # if >0, how much to fill ! 176: bgeq L24 ! 177: clrl r0 # no fill ! 178: L24: jbs $ljustb,flags,L25 ! 179: mnegl r0,r0 ! 180: L25: pushl r0 # fill count ! 181: pushl r1 # addr first byte ! 182: pushl r5 # length ! 183: calls $5,__strout ! 184: jbr loop ! 185: ! 186: pone: .byte 0x1C # packed 1 ! 187: ! 188: unsigned: ! 189: lunsigned: ! 190: extzv $1,$31,(ap),r0 # right shift logical 1 bit ! 191: cvtlp r0,$10,(sp) # convert [n/2] to packed ! 192: movp $10,(sp),8(sp) # copy packed ! 193: addp4 $10,8(sp),$10,(sp) # 2*[n/2] in packed, at (sp) ! 194: blbc (ap)+,L14 # n was even ! 195: addp4 $1,pone,$10,(sp) # n was odd ! 196: jbr L14 ! 197: ! 198: charac: ! 199: movl $4,r0 # chars per word ! 200: L18: movb (ap)+,(r5)+ # transfer char ! 201: bneq L19 ! 202: decl r5 # omit null characters ! 203: L19: sobgtr r0,L18 ! 204: ! 205: prbuf: ! 206: movl sp,r1 # addr first byte ! 207: jbr prstr ! 208: ! 209: plus: bisl2 $psign,flags # always print sign for floats ! 210: jbr L4a ! 211: minus: bisl2 $ljust,flags # left justification, please ! 212: jbr L4a ! 213: gnum0: jbs $ndfndb,flags,gnum ! 214: jbs $precb,flags,gnump # ignore when reading precision ! 215: bisl2 $zfill,flags # leading zero fill, please ! 216: gnum: jbs $precb,flags,gnump ! 217: moval (width)[width],width # width *= 5; ! 218: movaw -ch.zer(r0)[width],width # width = 2*witdh + r0 - '0'; ! 219: jbr gnumd ! 220: gnump: moval (ndigit)[ndigit],ndigit # ndigit *= 5; ! 221: movaw -ch.zer(r0)[ndigit],ndigit # ndigit = 2*ndigit + r0 - '0'; ! 222: gnumd: bisl2 $ndfnd,flags # digit seen ! 223: jbr L4a ! 224: dot: clrl ndigit # start on the precision ! 225: bisl2 $prec,flags ! 226: bicl2 $ndfnd,flags ! 227: jbr L4a ! 228: indir: movl (ap)+,ndigit # width specified by parameter ! 229: jbr gnumd ! 230: remote: movl (ap)+,ap ! 231: movl (ap)+,r11 ! 232: jbr loop ! 233: ! 234: float: ! 235: bsbw fltcvt ! 236: fltg: jbs $ndfndb,flags,float1 ! 237: movl $6,ndigit # default # digits to right of decpt. ! 238: float1: addl3 exp,ndigit,r7 ! 239: movl r7,r6 # for later "underflow" checking ! 240: bgeq fxplrd ! 241: clrl r7 # poor programmer planning ! 242: fxplrd: cmpl r7,$31 # expressible in packed decimal? ! 243: bleq fnarro # yes ! 244: movl $31,r7 ! 245: fnarro: subl3 $17,r7,r0 # where to round ! 246: ashp r0,$17,(sp),$5,r7,16(sp) # do it ! 247: bvc fnovfl ! 248: # band-aid for microcode error (spurious overflow) ! 249: clrl r0 # assume even length result ! 250: jlbc r7,fleven # right ! 251: movl $4,r0 # odd length result ! 252: fleven: cmpv r0,$4,16(sp),$0 # top digit zero iff true overflow ! 253: bneq fnovfl ! 254: # end band-aid ! 255: aobleq $0,r6,fnovfl # if "underflow" then jump ! 256: movl r7,r0 ! 257: incl exp ! 258: incl r7 ! 259: ashp r0,$1,pone,$0,r7,16(sp) ! 260: ashl $-1,r7,r0 # displ to last byte ! 261: bisb2 sign,16(sp)[r0] # insert sign ! 262: fnovfl: ! 263: movc3 $4,patsci,(sp) ! 264: clrl r6 # # digits moved so far ! 265: movl exp,r0 ! 266: bleq fexpng ! 267: bsbb patmov # digits to left of decpt. ! 268: fexpng: tstl ndigit ! 269: jeql fnodp ! 270: movc3 $2,fpatdp,(r3) ! 271: tstl exp ! 272: bgeq fxppos ! 273: addl3 exp,ndigit,r6 ! 274: bgeq flfakl ! 275: clrl r6 # it's all fill ! 276: flfakl: subl3 r6,$31,r6 # fake length for patmov ! 277: flfill: movc3 $2,fpatzf,(r3) # zero fill to right of dec.pt ! 278: fxppos: movl ndigit,r0 ! 279: bsbb patmov ! 280: fnodp: sobgeq r6,fledit # must move at least 1 digit ! 281: movl $31,r6 # none moved; fake it ! 282: aobleq $1,ndigit,flfill # with a one-character zero fill ! 283: fledit: editpc r7,16(sp),(sp),32(sp) ! 284: jbr prflt ! 285: ! 286: patexp: .byte 0x03 # eo$set_signif ! 287: .byte 0x44,'e # eo$insert 'e ! 288: .byte 0x42,'+ # eo$load_plus '+ ! 289: .byte 0x04 # eo$store_sign ! 290: .byte 0x92 # eo$move 2 ! 291: .byte 0 # eo$end ! 292: patsci: .byte 0x42,'+ # eo$load_plus '+ ! 293: .byte 0x03 # eo$set_signif ! 294: .byte 0x04 # eo$store_sign ! 295: .byte 0x91 # eo$move 1 ! 296: fpatdp: .byte 0x44,'. # eo$insert '. ! 297: fpatzf: .byte 0x40,'0 # eo$load_fill '0 ! 298: ! 299: # construct pattern at (r3) to move r0 digits in editpc; ! 300: # r6 digits already moved for this number ! 301: patmov: ! 302: movb $0x90,r2 # eo$move ! 303: subl3 r6,$31,r1 # # digits remaining in packed ! 304: addl2 r0,r6 ! 305: cmpl r0,r1 # enough digits remaining? ! 306: bleq patsml # yes ! 307: tstl exp # zero 'fill'; before or after rest? ! 308: bgeq pataft # after ! 309: pushl r1 # # digits remaining ! 310: movb $0x80,r2 # eo$fill ! 311: subl3 $31,r6,r0 # number of fill bytes ! 312: bsbb patsml # recursion! ! 313: movl (sp)+,r0 ! 314: movb $0x90,r2 # eo$move ! 315: jbr patsml ! 316: pataft: movl r1,r0 # last of the 31 ! 317: bsbb patsml # recursion! ! 318: subl3 $31,r6,r0 # number of fill bytes ! 319: movb $0x80,r2 # eo$fill ! 320: patsml: tstl r0 ! 321: bleq patzer # DEC doesn't like repetition counts of 0 ! 322: mnegl $15,r1 # 15 digits at a time ! 323: subl2 r1,r0 # counteract acbl ! 324: jbr pattst ! 325: patmlp: bisb3 r2,$15,(r3)+ # 15 ! 326: pattst: acbl $16,r1,r0,patmlp # until <= 15 left ! 327: bisb3 r2,r0,(r3)+ # rest ! 328: patzer: clrb (r3) # eo$end ! 329: rsb ! 330: ! 331: scien: ! 332: bsbw fltcvt # get packed digits ! 333: scig: incl ndigit ! 334: jbs $ndfndb,flags,L23 ! 335: movl $7,ndigit ! 336: L23: subl3 $17,ndigit,r0 # rounding position ! 337: ashp r0,$17,(sp),$5,ndigit,16(sp) # shift and round ! 338: bvc snovfl ! 339: # band-aid for microcode error (spurious overflow) ! 340: clrl r0 # assume even length result ! 341: jlbc ndigit,sceven # right ! 342: movl $4,r0 # odd length result ! 343: sceven: cmpv r0,$4,16(sp),$0 # top digit zero iff true overflow ! 344: bneq snovfl ! 345: # end band-aid ! 346: incl exp # rounding overflowed to 100... ! 347: subl3 $1,ndigit,r0 ! 348: ashp r0,$1,pone,$0,ndigit,16(sp) ! 349: ashl $-1,ndigit,r0 # displ to last byte ! 350: bisb2 sign,16(sp)[r0] # insert sign ! 351: snovfl: ! 352: jbc $gflagb,flags,enotg # not %g format ! 353: # find trailing zeroes in packed number ! 354: ashl $-1,ndigit,r0 ! 355: addl2 r3,r0 # addr of l.s.digit and sign ! 356: movl $4,r1 # bit position of digit ! 357: movl ndigit,r7 # current length of packed ! 358: jbr gtz ! 359: gtz1: xorl2 $4,r1 # position of next digit ! 360: bneq gtz # same byte ! 361: decl r0 # different byte ! 362: gtz: cmpv r1,$4,(r0),$0 # a trailing zero? ! 363: jneq gntz ! 364: sobgtr r7,gtz1 ! 365: incl r7 ! 366: gntz: # r7: minimum width of fraction ! 367: cmpl exp,$-4 ! 368: jleq eg # small exponents use %e ! 369: subl3 r7,exp,r0 ! 370: cmpl $5,r0 ! 371: jleq eg # so do (w+5) <= exp ! 372: tstl r0 # rest use %f ! 373: jleq fg # did we trim too many trailing zeroes? ! 374: movl exp,r7 # yes ! 375: fg: subl3 ndigit,r7,r0 ! 376: ashp r0,ndigit,16(sp),$0,r7,(sp) ! 377: movp r7,(sp),16(sp) ! 378: subl3 exp,r7,ndigit # correct ndigit for %f ! 379: jbr fnovfl ! 380: eg: subl3 ndigit,r7,r0 ! 381: ashp r0,ndigit,16(sp),$0,r7,(sp) ! 382: movp r7,(sp),16(sp) ! 383: movl r7,ndigit # packed number has been trimmed ! 384: enotg: ! 385: movc3 $7,patsci,(sp) ! 386: movl $1,r6 # 1P ! 387: subl3 $1,ndigit,r0 # digits after dec.pt ! 388: bsbw patmov ! 389: editpc ndigit,16(sp),(sp),32(sp) # 32(sp)->result, r5->(end+1) ! 390: decl exp # compensate: 1 digit left of dec.pt ! 391: cvtlp exp,$2,(sp) # exponent ! 392: editpc $2,(sp),patexp,(r5) ! 393: prflt: movab 32(sp),r1 ! 394: jbs $psignb,flags,prflt1 ! 395: cmpb (r1)+,$'+ ! 396: beql prflt1 ! 397: decl r1 ! 398: prflt1: skpc $' ,$63,(r1) ! 399: jbr prstr ! 400: ! 401: general: ! 402: jbcs $gflagb,flags,scien ! 403: jbr scien # safety net ! 404: ! 405: # convert double-floating at (ap) to 17-digit packed at (sp), ! 406: # set 'sign' and 'exp', advance ap. ! 407: fltcvt: ! 408: clrb sign ! 409: movd (ap)+,r5 ! 410: jeql fzero ! 411: bgtr fpos ! 412: mnegd r5,r5 ! 413: incb sign ! 414: fpos: ! 415: extzv $7,$8,r5,r2 # exponent of 2 ! 416: movaw -0600(r2)[r2],r2 # unbias and mult by 3 ! 417: bgeq epos ! 418: subl2 $9,r2 ! 419: epos: divl2 $10,r2 ! 420: bsbb expten ! 421: cmpd r0,r5 ! 422: bgtr ceil ! 423: incl r2 ! 424: ceil: movl r2,exp ! 425: mnegl r2,r2 ! 426: cmpl r2,$29 # 10^(29+9) is all we can handle ! 427: bleq getman ! 428: muld2 ten16,r5 ! 429: subl2 $16,r2 ! 430: getman: addl2 $9,r2 # -ceil(log10(x)) + 9 ! 431: bsbb expten ! 432: emodd r0,r4,r5,r0,r5 # (r0+r4)*r5; r0=int, r5=frac ! 433: fz1: cvtlp r0,$9,16(sp) # leading 9 digits ! 434: ashp $8,$9,16(sp),$0,$17,4(sp) # as top 9 of 17 ! 435: emodd ten8,$0,r5,r0,r5 ! 436: cvtlp r0,$8,16(sp) # trailing 8 digits ! 437: addp4 $8,16(sp),$17,4(sp) # combine leading and trailing ! 438: bisb2 sign,12(sp) # and insert sign ! 439: rsb ! 440: fzero: clrl r0 ! 441: movl $1,exp # 0.000e+00 and 0.000 rather than 0.000e-01 and .000 ! 442: jbr fz1 ! 443: ! 444: # return 10^r2 as a double float in r0||r1 and 8 extra bits of precision in r4 ! 445: # preserve r2, r5||r6 ! 446: expten: ! 447: movd $one,r0 # begin computing 10^exp10 ! 448: clrl r4 # bit counter ! 449: movad ten1,r3 # table address ! 450: tstl r2 ! 451: bgeq e10lp ! 452: mnegl r2,r2 # get absolute value ! 453: jbss $6,r2,e10lp # flag as negative ! 454: e10lp: jbc r4,r2,el1 # want this power? ! 455: muld2 (r3),r0 # yes ! 456: el1: addl2 $8,r3 # advance to next power ! 457: aobleq $5,r4,e10lp # through 10^32 ! 458: jbcc $6,r2,el2 # correct for negative exponent ! 459: divd3 r0,$one,r0 # by taking reciprocal ! 460: mnegl r2,r2 ! 461: el2: clrl r4 # 8 extra bits of precision ! 462: rsb ! 463: ! 464: # powers of ten ! 465: .align 3 ! 466: ten1: .word 0x4220,0,0,0 ! 467: ten2: .word 0x43c8,0,0,0 ! 468: ten4: .word 0x471c,0x4000,0,0 ! 469: ten8: .word 0x4dbe,0xbc20,0,0 ! 470: ten16: .word 0x5b0e,0x1bc9,0xbf04,0 ! 471: ten32: .word 0x759d,0xc5ad,0xa82b,0x70b6
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.