1 //===-- APFloat.cpp - Implement APFloat class -----------------------------===// 2 // 3 // The LLVM Compiler Infrastructure 4 // 5 // This file is distributed under the University of Illinois Open Source 6 // License. See LICENSE.TXT for details. 7 // 8 //===----------------------------------------------------------------------===// 9 // 10 // This file implements a class to represent arbitrary precision floating 11 // point values and provide a variety of arithmetic operations on them. 12 // 13 //===----------------------------------------------------------------------===// 14 15 #include "llvm/ADT/APFloat.h" 16 #include "llvm/ADT/APSInt.h" 17 #include "llvm/ADT/FoldingSet.h" 18 #include "llvm/ADT/Hashing.h" 19 #include "llvm/ADT/StringExtras.h" 20 #include "llvm/ADT/StringRef.h" 21 #include "llvm/Support/ErrorHandling.h" 22 #include "llvm/Support/MathExtras.h" 23 #include <cstring> 24 #include <limits.h> 25 26 using namespace llvm; 27 28 /// A macro used to combine two fcCategory enums into one key which can be used 29 /// in a switch statement to classify how the interaction of two APFloat's 30 /// categories affects an operation. 31 /// 32 /// TODO: If clang source code is ever allowed to use constexpr in its own 33 /// codebase, change this into a static inline function. 34 #define PackCategoriesIntoKey(_lhs, _rhs) ((_lhs) * 4 + (_rhs)) 35 36 /* Assumed in hexadecimal significand parsing, and conversion to 37 hexadecimal strings. */ 38 #define COMPILE_TIME_ASSERT(cond) extern int CTAssert[(cond) ? 1 : -1] 39 COMPILE_TIME_ASSERT(integerPartWidth % 4 == 0); 40 41 namespace llvm { 42 43 /* Represents floating point arithmetic semantics. */ 44 struct fltSemantics { 45 /* The largest E such that 2^E is representable; this matches the 46 definition of IEEE 754. */ 47 APFloat::ExponentType maxExponent; 48 49 /* The smallest E such that 2^E is a normalized number; this 50 matches the definition of IEEE 754. */ 51 APFloat::ExponentType minExponent; 52 53 /* Number of bits in the significand. This includes the integer 54 bit. */ 55 unsigned int precision; 56 }; 57 58 const fltSemantics APFloat::IEEEhalf = { 15, -14, 11 }; 59 const fltSemantics APFloat::IEEEsingle = { 127, -126, 24 }; 60 const fltSemantics APFloat::IEEEdouble = { 1023, -1022, 53 }; 61 const fltSemantics APFloat::IEEEquad = { 16383, -16382, 113 }; 62 const fltSemantics APFloat::x87DoubleExtended = { 16383, -16382, 64 }; 63 const fltSemantics APFloat::Bogus = { 0, 0, 0 }; 64 65 /* The PowerPC format consists of two doubles. It does not map cleanly 66 onto the usual format above. It is approximated using twice the 67 mantissa bits. Note that for exponents near the double minimum, 68 we no longer can represent the full 106 mantissa bits, so those 69 will be treated as denormal numbers. 70 71 FIXME: While this approximation is equivalent to what GCC uses for 72 compile-time arithmetic on PPC double-double numbers, it is not able 73 to represent all possible values held by a PPC double-double number, 74 for example: (long double) 1.0 + (long double) 0x1p-106 75 Should this be replaced by a full emulation of PPC double-double? */ 76 const fltSemantics APFloat::PPCDoubleDouble = { 1023, -1022 + 53, 53 + 53 }; 77 78 /* A tight upper bound on number of parts required to hold the value 79 pow(5, power) is 80 81 power * 815 / (351 * integerPartWidth) + 1 82 83 However, whilst the result may require only this many parts, 84 because we are multiplying two values to get it, the 85 multiplication may require an extra part with the excess part 86 being zero (consider the trivial case of 1 * 1, tcFullMultiply 87 requires two parts to hold the single-part result). So we add an 88 extra one to guarantee enough space whilst multiplying. */ 89 const unsigned int maxExponent = 16383; 90 const unsigned int maxPrecision = 113; 91 const unsigned int maxPowerOfFiveExponent = maxExponent + maxPrecision - 1; 92 const unsigned int maxPowerOfFiveParts = 2 + ((maxPowerOfFiveExponent * 815) 93 / (351 * integerPartWidth)); 94 } 95 96 /* A bunch of private, handy routines. */ 97 98 static inline unsigned int 99 partCountForBits(unsigned int bits) 100 { 101 return ((bits) + integerPartWidth - 1) / integerPartWidth; 102 } 103 104 /* Returns 0U-9U. Return values >= 10U are not digits. */ 105 static inline unsigned int 106 decDigitValue(unsigned int c) 107 { 108 return c - '0'; 109 } 110 111 /* Return the value of a decimal exponent of the form 112 [+-]ddddddd. 113 114 If the exponent overflows, returns a large exponent with the 115 appropriate sign. */ 116 static int 117 readExponent(StringRef::iterator begin, StringRef::iterator end) 118 { 119 bool isNegative; 120 unsigned int absExponent; 121 const unsigned int overlargeExponent = 24000; /* FIXME. */ 122 StringRef::iterator p = begin; 123 124 assert(p != end && "Exponent has no digits"); 125 126 isNegative = (*p == '-'); 127 if (*p == '-' || *p == '+') { 128 p++; 129 assert(p != end && "Exponent has no digits"); 130 } 131 132 absExponent = decDigitValue(*p++); 133 assert(absExponent < 10U && "Invalid character in exponent"); 134 135 for (; p != end; ++p) { 136 unsigned int value; 137 138 value = decDigitValue(*p); 139 assert(value < 10U && "Invalid character in exponent"); 140 141 value += absExponent * 10; 142 if (absExponent >= overlargeExponent) { 143 absExponent = overlargeExponent; 144 p = end; /* outwit assert below */ 145 break; 146 } 147 absExponent = value; 148 } 149 150 assert(p == end && "Invalid exponent in exponent"); 151 152 if (isNegative) 153 return -(int) absExponent; 154 else 155 return (int) absExponent; 156 } 157 158 /* This is ugly and needs cleaning up, but I don't immediately see 159 how whilst remaining safe. */ 160 static int 161 totalExponent(StringRef::iterator p, StringRef::iterator end, 162 int exponentAdjustment) 163 { 164 int unsignedExponent; 165 bool negative, overflow; 166 int exponent = 0; 167 168 assert(p != end && "Exponent has no digits"); 169 170 negative = *p == '-'; 171 if (*p == '-' || *p == '+') { 172 p++; 173 assert(p != end && "Exponent has no digits"); 174 } 175 176 unsignedExponent = 0; 177 overflow = false; 178 for (; p != end; ++p) { 179 unsigned int value; 180 181 value = decDigitValue(*p); 182 assert(value < 10U && "Invalid character in exponent"); 183 184 unsignedExponent = unsignedExponent * 10 + value; 185 if (unsignedExponent > 32767) { 186 overflow = true; 187 break; 188 } 189 } 190 191 if (exponentAdjustment > 32767 || exponentAdjustment < -32768) 192 overflow = true; 193 194 if (!overflow) { 195 exponent = unsignedExponent; 196 if (negative) 197 exponent = -exponent; 198 exponent += exponentAdjustment; 199 if (exponent > 32767 || exponent < -32768) 200 overflow = true; 201 } 202 203 if (overflow) 204 exponent = negative ? -32768: 32767; 205 206 return exponent; 207 } 208 209 static StringRef::iterator 210 skipLeadingZeroesAndAnyDot(StringRef::iterator begin, StringRef::iterator end, 211 StringRef::iterator *dot) 212 { 213 StringRef::iterator p = begin; 214 *dot = end; 215 while (*p == '0' && p != end) 216 p++; 217 218 if (*p == '.') { 219 *dot = p++; 220 221 assert(end - begin != 1 && "Significand has no digits"); 222 223 while (*p == '0' && p != end) 224 p++; 225 } 226 227 return p; 228 } 229 230 /* Given a normal decimal floating point number of the form 231 232 dddd.dddd[eE][+-]ddd 233 234 where the decimal point and exponent are optional, fill out the 235 structure D. Exponent is appropriate if the significand is 236 treated as an integer, and normalizedExponent if the significand 237 is taken to have the decimal point after a single leading 238 non-zero digit. 239 240 If the value is zero, V->firstSigDigit points to a non-digit, and 241 the return exponent is zero. 242 */ 243 struct decimalInfo { 244 const char *firstSigDigit; 245 const char *lastSigDigit; 246 int exponent; 247 int normalizedExponent; 248 }; 249 250 static void 251 interpretDecimal(StringRef::iterator begin, StringRef::iterator end, 252 decimalInfo *D) 253 { 254 StringRef::iterator dot = end; 255 StringRef::iterator p = skipLeadingZeroesAndAnyDot (begin, end, &dot); 256 257 D->firstSigDigit = p; 258 D->exponent = 0; 259 D->normalizedExponent = 0; 260 261 for (; p != end; ++p) { 262 if (*p == '.') { 263 assert(dot == end && "String contains multiple dots"); 264 dot = p++; 265 if (p == end) 266 break; 267 } 268 if (decDigitValue(*p) >= 10U) 269 break; 270 } 271 272 if (p != end) { 273 assert((*p == 'e' || *p == 'E') && "Invalid character in significand"); 274 assert(p != begin && "Significand has no digits"); 275 assert((dot == end || p - begin != 1) && "Significand has no digits"); 276 277 /* p points to the first non-digit in the string */ 278 D->exponent = readExponent(p + 1, end); 279 280 /* Implied decimal point? */ 281 if (dot == end) 282 dot = p; 283 } 284 285 /* If number is all zeroes accept any exponent. */ 286 if (p != D->firstSigDigit) { 287 /* Drop insignificant trailing zeroes. */ 288 if (p != begin) { 289 do 290 do 291 p--; 292 while (p != begin && *p == '0'); 293 while (p != begin && *p == '.'); 294 } 295 296 /* Adjust the exponents for any decimal point. */ 297 D->exponent += static_cast<APFloat::ExponentType>((dot - p) - (dot > p)); 298 D->normalizedExponent = (D->exponent + 299 static_cast<APFloat::ExponentType>((p - D->firstSigDigit) 300 - (dot > D->firstSigDigit && dot < p))); 301 } 302 303 D->lastSigDigit = p; 304 } 305 306 /* Return the trailing fraction of a hexadecimal number. 307 DIGITVALUE is the first hex digit of the fraction, P points to 308 the next digit. */ 309 static lostFraction 310 trailingHexadecimalFraction(StringRef::iterator p, StringRef::iterator end, 311 unsigned int digitValue) 312 { 313 unsigned int hexDigit; 314 315 /* If the first trailing digit isn't 0 or 8 we can work out the 316 fraction immediately. */ 317 if (digitValue > 8) 318 return lfMoreThanHalf; 319 else if (digitValue < 8 && digitValue > 0) 320 return lfLessThanHalf; 321 322 // Otherwise we need to find the first non-zero digit. 323 while (p != end && (*p == '0' || *p == '.')) 324 p++; 325 326 assert(p != end && "Invalid trailing hexadecimal fraction!"); 327 328 hexDigit = hexDigitValue(*p); 329 330 /* If we ran off the end it is exactly zero or one-half, otherwise 331 a little more. */ 332 if (hexDigit == -1U) 333 return digitValue == 0 ? lfExactlyZero: lfExactlyHalf; 334 else 335 return digitValue == 0 ? lfLessThanHalf: lfMoreThanHalf; 336 } 337 338 /* Return the fraction lost were a bignum truncated losing the least 339 significant BITS bits. */ 340 static lostFraction 341 lostFractionThroughTruncation(const integerPart *parts, 342 unsigned int partCount, 343 unsigned int bits) 344 { 345 unsigned int lsb; 346 347 lsb = APInt::tcLSB(parts, partCount); 348 349 /* Note this is guaranteed true if bits == 0, or LSB == -1U. */ 350 if (bits <= lsb) 351 return lfExactlyZero; 352 if (bits == lsb + 1) 353 return lfExactlyHalf; 354 if (bits <= partCount * integerPartWidth && 355 APInt::tcExtractBit(parts, bits - 1)) 356 return lfMoreThanHalf; 357 358 return lfLessThanHalf; 359 } 360 361 /* Shift DST right BITS bits noting lost fraction. */ 362 static lostFraction 363 shiftRight(integerPart *dst, unsigned int parts, unsigned int bits) 364 { 365 lostFraction lost_fraction; 366 367 lost_fraction = lostFractionThroughTruncation(dst, parts, bits); 368 369 APInt::tcShiftRight(dst, parts, bits); 370 371 return lost_fraction; 372 } 373 374 /* Combine the effect of two lost fractions. */ 375 static lostFraction 376 combineLostFractions(lostFraction moreSignificant, 377 lostFraction lessSignificant) 378 { 379 if (lessSignificant != lfExactlyZero) { 380 if (moreSignificant == lfExactlyZero) 381 moreSignificant = lfLessThanHalf; 382 else if (moreSignificant == lfExactlyHalf) 383 moreSignificant = lfMoreThanHalf; 384 } 385 386 return moreSignificant; 387 } 388 389 /* The error from the true value, in half-ulps, on multiplying two 390 floating point numbers, which differ from the value they 391 approximate by at most HUE1 and HUE2 half-ulps, is strictly less 392 than the returned value. 393 394 See "How to Read Floating Point Numbers Accurately" by William D 395 Clinger. */ 396 static unsigned int 397 HUerrBound(bool inexactMultiply, unsigned int HUerr1, unsigned int HUerr2) 398 { 399 assert(HUerr1 < 2 || HUerr2 < 2 || (HUerr1 + HUerr2 < 8)); 400 401 if (HUerr1 + HUerr2 == 0) 402 return inexactMultiply * 2; /* <= inexactMultiply half-ulps. */ 403 else 404 return inexactMultiply + 2 * (HUerr1 + HUerr2); 405 } 406 407 /* The number of ulps from the boundary (zero, or half if ISNEAREST) 408 when the least significant BITS are truncated. BITS cannot be 409 zero. */ 410 static integerPart 411 ulpsFromBoundary(const integerPart *parts, unsigned int bits, bool isNearest) 412 { 413 unsigned int count, partBits; 414 integerPart part, boundary; 415 416 assert(bits != 0); 417 418 bits--; 419 count = bits / integerPartWidth; 420 partBits = bits % integerPartWidth + 1; 421 422 part = parts[count] & (~(integerPart) 0 >> (integerPartWidth - partBits)); 423 424 if (isNearest) 425 boundary = (integerPart) 1 << (partBits - 1); 426 else 427 boundary = 0; 428 429 if (count == 0) { 430 if (part - boundary <= boundary - part) 431 return part - boundary; 432 else 433 return boundary - part; 434 } 435 436 if (part == boundary) { 437 while (--count) 438 if (parts[count]) 439 return ~(integerPart) 0; /* A lot. */ 440 441 return parts[0]; 442 } else if (part == boundary - 1) { 443 while (--count) 444 if (~parts[count]) 445 return ~(integerPart) 0; /* A lot. */ 446 447 return -parts[0]; 448 } 449 450 return ~(integerPart) 0; /* A lot. */ 451 } 452 453 /* Place pow(5, power) in DST, and return the number of parts used. 454 DST must be at least one part larger than size of the answer. */ 455 static unsigned int 456 powerOf5(integerPart *dst, unsigned int power) 457 { 458 static const integerPart firstEightPowers[] = { 1, 5, 25, 125, 625, 3125, 459 15625, 78125 }; 460 integerPart pow5s[maxPowerOfFiveParts * 2 + 5]; 461 pow5s[0] = 78125 * 5; 462 463 unsigned int partsCount[16] = { 1 }; 464 integerPart scratch[maxPowerOfFiveParts], *p1, *p2, *pow5; 465 unsigned int result; 466 assert(power <= maxExponent); 467 468 p1 = dst; 469 p2 = scratch; 470 471 *p1 = firstEightPowers[power & 7]; 472 power >>= 3; 473 474 result = 1; 475 pow5 = pow5s; 476 477 for (unsigned int n = 0; power; power >>= 1, n++) { 478 unsigned int pc; 479 480 pc = partsCount[n]; 481 482 /* Calculate pow(5,pow(2,n+3)) if we haven't yet. */ 483 if (pc == 0) { 484 pc = partsCount[n - 1]; 485 APInt::tcFullMultiply(pow5, pow5 - pc, pow5 - pc, pc, pc); 486 pc *= 2; 487 if (pow5[pc - 1] == 0) 488 pc--; 489 partsCount[n] = pc; 490 } 491 492 if (power & 1) { 493 integerPart *tmp; 494 495 APInt::tcFullMultiply(p2, p1, pow5, result, pc); 496 result += pc; 497 if (p2[result - 1] == 0) 498 result--; 499 500 /* Now result is in p1 with partsCount parts and p2 is scratch 501 space. */ 502 tmp = p1, p1 = p2, p2 = tmp; 503 } 504 505 pow5 += pc; 506 } 507 508 if (p1 != dst) 509 APInt::tcAssign(dst, p1, result); 510 511 return result; 512 } 513 514 /* Zero at the end to avoid modular arithmetic when adding one; used 515 when rounding up during hexadecimal output. */ 516 static const char hexDigitsLower[] = "0123456789abcdef0"; 517 static const char hexDigitsUpper[] = "0123456789ABCDEF0"; 518 static const char infinityL[] = "infinity"; 519 static const char infinityU[] = "INFINITY"; 520 static const char NaNL[] = "nan"; 521 static const char NaNU[] = "NAN"; 522 523 /* Write out an integerPart in hexadecimal, starting with the most 524 significant nibble. Write out exactly COUNT hexdigits, return 525 COUNT. */ 526 static unsigned int 527 partAsHex (char *dst, integerPart part, unsigned int count, 528 const char *hexDigitChars) 529 { 530 unsigned int result = count; 531 532 assert(count != 0 && count <= integerPartWidth / 4); 533 534 part >>= (integerPartWidth - 4 * count); 535 while (count--) { 536 dst[count] = hexDigitChars[part & 0xf]; 537 part >>= 4; 538 } 539 540 return result; 541 } 542 543 /* Write out an unsigned decimal integer. */ 544 static char * 545 writeUnsignedDecimal (char *dst, unsigned int n) 546 { 547 char buff[40], *p; 548 549 p = buff; 550 do 551 *p++ = '0' + n % 10; 552 while (n /= 10); 553 554 do 555 *dst++ = *--p; 556 while (p != buff); 557 558 return dst; 559 } 560 561 /* Write out a signed decimal integer. */ 562 static char * 563 writeSignedDecimal (char *dst, int value) 564 { 565 if (value < 0) { 566 *dst++ = '-'; 567 dst = writeUnsignedDecimal(dst, -(unsigned) value); 568 } else 569 dst = writeUnsignedDecimal(dst, value); 570 571 return dst; 572 } 573 574 /* Constructors. */ 575 void 576 APFloat::initialize(const fltSemantics *ourSemantics) 577 { 578 unsigned int count; 579 580 semantics = ourSemantics; 581 count = partCount(); 582 if (count > 1) 583 significand.parts = new integerPart[count]; 584 } 585 586 void 587 APFloat::freeSignificand() 588 { 589 if (needsCleanup()) 590 delete [] significand.parts; 591 } 592 593 void 594 APFloat::assign(const APFloat &rhs) 595 { 596 assert(semantics == rhs.semantics); 597 598 sign = rhs.sign; 599 category = rhs.category; 600 exponent = rhs.exponent; 601 if (isFiniteNonZero() || category == fcNaN) 602 copySignificand(rhs); 603 } 604 605 void 606 APFloat::copySignificand(const APFloat &rhs) 607 { 608 assert(isFiniteNonZero() || category == fcNaN); 609 assert(rhs.partCount() >= partCount()); 610 611 APInt::tcAssign(significandParts(), rhs.significandParts(), 612 partCount()); 613 } 614 615 /* Make this number a NaN, with an arbitrary but deterministic value 616 for the significand. If double or longer, this is a signalling NaN, 617 which may not be ideal. If float, this is QNaN(0). */ 618 void APFloat::makeNaN(bool SNaN, bool Negative, const APInt *fill) 619 { 620 category = fcNaN; 621 sign = Negative; 622 623 integerPart *significand = significandParts(); 624 unsigned numParts = partCount(); 625 626 // Set the significand bits to the fill. 627 if (!fill || fill->getNumWords() < numParts) 628 APInt::tcSet(significand, 0, numParts); 629 if (fill) { 630 APInt::tcAssign(significand, fill->getRawData(), 631 std::min(fill->getNumWords(), numParts)); 632 633 // Zero out the excess bits of the significand. 634 unsigned bitsToPreserve = semantics->precision - 1; 635 unsigned part = bitsToPreserve / 64; 636 bitsToPreserve %= 64; 637 significand[part] &= ((1ULL << bitsToPreserve) - 1); 638 for (part++; part != numParts; ++part) 639 significand[part] = 0; 640 } 641 642 unsigned QNaNBit = semantics->precision - 2; 643 644 if (SNaN) { 645 // We always have to clear the QNaN bit to make it an SNaN. 646 APInt::tcClearBit(significand, QNaNBit); 647 648 // If there are no bits set in the payload, we have to set 649 // *something* to make it a NaN instead of an infinity; 650 // conventionally, this is the next bit down from the QNaN bit. 651 if (APInt::tcIsZero(significand, numParts)) 652 APInt::tcSetBit(significand, QNaNBit - 1); 653 } else { 654 // We always have to set the QNaN bit to make it a QNaN. 655 APInt::tcSetBit(significand, QNaNBit); 656 } 657 658 // For x87 extended precision, we want to make a NaN, not a 659 // pseudo-NaN. Maybe we should expose the ability to make 660 // pseudo-NaNs? 661 if (semantics == &APFloat::x87DoubleExtended) 662 APInt::tcSetBit(significand, QNaNBit + 1); 663 } 664 665 APFloat APFloat::makeNaN(const fltSemantics &Sem, bool SNaN, bool Negative, 666 const APInt *fill) { 667 APFloat value(Sem, uninitialized); 668 value.makeNaN(SNaN, Negative, fill); 669 return value; 670 } 671 672 APFloat & 673 APFloat::operator=(const APFloat &rhs) 674 { 675 if (this != &rhs) { 676 if (semantics != rhs.semantics) { 677 freeSignificand(); 678 initialize(rhs.semantics); 679 } 680 assign(rhs); 681 } 682 683 return *this; 684 } 685 686 bool 687 APFloat::isDenormal() const { 688 return isFiniteNonZero() && (exponent == semantics->minExponent) && 689 (APInt::tcExtractBit(significandParts(), 690 semantics->precision - 1) == 0); 691 } 692 693 bool 694 APFloat::isSmallest() const { 695 // The smallest number by magnitude in our format will be the smallest 696 // denormal, i.e. the floating point number with exponent being minimum 697 // exponent and significand bitwise equal to 1 (i.e. with MSB equal to 0). 698 return isFiniteNonZero() && exponent == semantics->minExponent && 699 significandMSB() == 0; 700 } 701 702 bool APFloat::isSignificandAllOnes() const { 703 // Test if the significand excluding the integral bit is all ones. This allows 704 // us to test for binade boundaries. 705 const integerPart *Parts = significandParts(); 706 const unsigned PartCount = partCount(); 707 for (unsigned i = 0; i < PartCount - 1; i++) 708 if (~Parts[i]) 709 return false; 710 711 // Set the unused high bits to all ones when we compare. 712 const unsigned NumHighBits = 713 PartCount*integerPartWidth - semantics->precision + 1; 714 assert(NumHighBits <= integerPartWidth && "Can not have more high bits to " 715 "fill than integerPartWidth"); 716 const integerPart HighBitFill = 717 ~integerPart(0) << (integerPartWidth - NumHighBits); 718 if (~(Parts[PartCount - 1] | HighBitFill)) 719 return false; 720 721 return true; 722 } 723 724 bool APFloat::isSignificandAllZeros() const { 725 // Test if the significand excluding the integral bit is all zeros. This 726 // allows us to test for binade boundaries. 727 const integerPart *Parts = significandParts(); 728 const unsigned PartCount = partCount(); 729 730 for (unsigned i = 0; i < PartCount - 1; i++) 731 if (Parts[i]) 732 return false; 733 734 const unsigned NumHighBits = 735 PartCount*integerPartWidth - semantics->precision + 1; 736 assert(NumHighBits <= integerPartWidth && "Can not have more high bits to " 737 "clear than integerPartWidth"); 738 const integerPart HighBitMask = ~integerPart(0) >> NumHighBits; 739 740 if (Parts[PartCount - 1] & HighBitMask) 741 return false; 742 743 return true; 744 } 745 746 bool 747 APFloat::isLargest() const { 748 // The largest number by magnitude in our format will be the floating point 749 // number with maximum exponent and with significand that is all ones. 750 return isFiniteNonZero() && exponent == semantics->maxExponent 751 && isSignificandAllOnes(); 752 } 753 754 bool 755 APFloat::bitwiseIsEqual(const APFloat &rhs) const { 756 if (this == &rhs) 757 return true; 758 if (semantics != rhs.semantics || 759 category != rhs.category || 760 sign != rhs.sign) 761 return false; 762 if (category==fcZero || category==fcInfinity) 763 return true; 764 else if (isFiniteNonZero() && exponent!=rhs.exponent) 765 return false; 766 else { 767 int i= partCount(); 768 const integerPart* p=significandParts(); 769 const integerPart* q=rhs.significandParts(); 770 for (; i>0; i--, p++, q++) { 771 if (*p != *q) 772 return false; 773 } 774 return true; 775 } 776 } 777 778 APFloat::APFloat(const fltSemantics &ourSemantics, integerPart value) { 779 initialize(&ourSemantics); 780 sign = 0; 781 category = fcNormal; 782 zeroSignificand(); 783 exponent = ourSemantics.precision - 1; 784 significandParts()[0] = value; 785 normalize(rmNearestTiesToEven, lfExactlyZero); 786 } 787 788 APFloat::APFloat(const fltSemantics &ourSemantics) { 789 initialize(&ourSemantics); 790 category = fcZero; 791 sign = false; 792 } 793 794 APFloat::APFloat(const fltSemantics &ourSemantics, uninitializedTag tag) { 795 // Allocates storage if necessary but does not initialize it. 796 initialize(&ourSemantics); 797 } 798 799 APFloat::APFloat(const fltSemantics &ourSemantics, StringRef text) { 800 initialize(&ourSemantics); 801 convertFromString(text, rmNearestTiesToEven); 802 } 803 804 APFloat::APFloat(const APFloat &rhs) { 805 initialize(rhs.semantics); 806 assign(rhs); 807 } 808 809 APFloat::~APFloat() 810 { 811 freeSignificand(); 812 } 813 814 // Profile - This method 'profiles' an APFloat for use with FoldingSet. 815 void APFloat::Profile(FoldingSetNodeID& ID) const { 816 ID.Add(bitcastToAPInt()); 817 } 818 819 unsigned int 820 APFloat::partCount() const 821 { 822 return partCountForBits(semantics->precision + 1); 823 } 824 825 unsigned int 826 APFloat::semanticsPrecision(const fltSemantics &semantics) 827 { 828 return semantics.precision; 829 } 830 831 const integerPart * 832 APFloat::significandParts() const 833 { 834 return const_cast<APFloat *>(this)->significandParts(); 835 } 836 837 integerPart * 838 APFloat::significandParts() 839 { 840 if (partCount() > 1) 841 return significand.parts; 842 else 843 return &significand.part; 844 } 845 846 void 847 APFloat::zeroSignificand() 848 { 849 APInt::tcSet(significandParts(), 0, partCount()); 850 } 851 852 /* Increment an fcNormal floating point number's significand. */ 853 void 854 APFloat::incrementSignificand() 855 { 856 integerPart carry; 857 858 carry = APInt::tcIncrement(significandParts(), partCount()); 859 860 /* Our callers should never cause us to overflow. */ 861 assert(carry == 0); 862 (void)carry; 863 } 864 865 /* Add the significand of the RHS. Returns the carry flag. */ 866 integerPart 867 APFloat::addSignificand(const APFloat &rhs) 868 { 869 integerPart *parts; 870 871 parts = significandParts(); 872 873 assert(semantics == rhs.semantics); 874 assert(exponent == rhs.exponent); 875 876 return APInt::tcAdd(parts, rhs.significandParts(), 0, partCount()); 877 } 878 879 /* Subtract the significand of the RHS with a borrow flag. Returns 880 the borrow flag. */ 881 integerPart 882 APFloat::subtractSignificand(const APFloat &rhs, integerPart borrow) 883 { 884 integerPart *parts; 885 886 parts = significandParts(); 887 888 assert(semantics == rhs.semantics); 889 assert(exponent == rhs.exponent); 890 891 return APInt::tcSubtract(parts, rhs.significandParts(), borrow, 892 partCount()); 893 } 894 895 /* Multiply the significand of the RHS. If ADDEND is non-NULL, add it 896 on to the full-precision result of the multiplication. Returns the 897 lost fraction. */ 898 lostFraction 899 APFloat::multiplySignificand(const APFloat &rhs, const APFloat *addend) 900 { 901 unsigned int omsb; // One, not zero, based MSB. 902 unsigned int partsCount, newPartsCount, precision; 903 integerPart *lhsSignificand; 904 integerPart scratch[4]; 905 integerPart *fullSignificand; 906 lostFraction lost_fraction; 907 bool ignored; 908 909 assert(semantics == rhs.semantics); 910 911 precision = semantics->precision; 912 newPartsCount = partCountForBits(precision * 2); 913 914 if (newPartsCount > 4) 915 fullSignificand = new integerPart[newPartsCount]; 916 else 917 fullSignificand = scratch; 918 919 lhsSignificand = significandParts(); 920 partsCount = partCount(); 921 922 APInt::tcFullMultiply(fullSignificand, lhsSignificand, 923 rhs.significandParts(), partsCount, partsCount); 924 925 lost_fraction = lfExactlyZero; 926 omsb = APInt::tcMSB(fullSignificand, newPartsCount) + 1; 927 exponent += rhs.exponent; 928 929 // Assume the operands involved in the multiplication are single-precision 930 // FP, and the two multiplicants are: 931 // *this = a23 . a22 ... a0 * 2^e1 932 // rhs = b23 . b22 ... b0 * 2^e2 933 // the result of multiplication is: 934 // *this = c47 c46 . c45 ... c0 * 2^(e1+e2) 935 // Note that there are two significant bits at the left-hand side of the 936 // radix point. Move the radix point toward left by one bit, and adjust 937 // exponent accordingly. 938 exponent += 1; 939 940 if (addend) { 941 // The intermediate result of the multiplication has "2 * precision" 942 // signicant bit; adjust the addend to be consistent with mul result. 943 // 944 Significand savedSignificand = significand; 945 const fltSemantics *savedSemantics = semantics; 946 fltSemantics extendedSemantics; 947 opStatus status; 948 unsigned int extendedPrecision; 949 950 /* Normalize our MSB. */ 951 extendedPrecision = 2 * precision; 952 if (omsb != extendedPrecision) { 953 assert(extendedPrecision > omsb); 954 APInt::tcShiftLeft(fullSignificand, newPartsCount, 955 extendedPrecision - omsb); 956 exponent -= extendedPrecision - omsb; 957 } 958 959 /* Create new semantics. */ 960 extendedSemantics = *semantics; 961 extendedSemantics.precision = extendedPrecision; 962 963 if (newPartsCount == 1) 964 significand.part = fullSignificand[0]; 965 else 966 significand.parts = fullSignificand; 967 semantics = &extendedSemantics; 968 969 APFloat extendedAddend(*addend); 970 status = extendedAddend.convert(extendedSemantics, rmTowardZero, &ignored); 971 assert(status == opOK); 972 (void)status; 973 lost_fraction = addOrSubtractSignificand(extendedAddend, false); 974 975 /* Restore our state. */ 976 if (newPartsCount == 1) 977 fullSignificand[0] = significand.part; 978 significand = savedSignificand; 979 semantics = savedSemantics; 980 981 omsb = APInt::tcMSB(fullSignificand, newPartsCount) + 1; 982 } 983 984 // Convert the result having "2 * precision" significant-bits back to the one 985 // having "precision" significant-bits. First, move the radix point from 986 // poision "2*precision - 1" to "precision - 1". The exponent need to be 987 // adjusted by "2*precision - 1" - "precision - 1" = "precision". 988 exponent -= precision; 989 990 // In case MSB resides at the left-hand side of radix point, shift the 991 // mantissa right by some amount to make sure the MSB reside right before 992 // the radix point (i.e. "MSB . rest-significant-bits"). 993 // 994 // Note that the result is not normalized when "omsb < precision". So, the 995 // caller needs to call APFloat::normalize() if normalized value is expected. 996 if (omsb > precision) { 997 unsigned int bits, significantParts; 998 lostFraction lf; 999 1000 bits = omsb - precision; 1001 significantParts = partCountForBits(omsb); 1002 lf = shiftRight(fullSignificand, significantParts, bits); 1003 lost_fraction = combineLostFractions(lf, lost_fraction); 1004 exponent += bits; 1005 } 1006 1007 APInt::tcAssign(lhsSignificand, fullSignificand, partsCount); 1008 1009 if (newPartsCount > 4) 1010 delete [] fullSignificand; 1011 1012 return lost_fraction; 1013 } 1014 1015 /* Multiply the significands of LHS and RHS to DST. */ 1016 lostFraction 1017 APFloat::divideSignificand(const APFloat &rhs) 1018 { 1019 unsigned int bit, i, partsCount; 1020 const integerPart *rhsSignificand; 1021 integerPart *lhsSignificand, *dividend, *divisor; 1022 integerPart scratch[4]; 1023 lostFraction lost_fraction; 1024 1025 assert(semantics == rhs.semantics); 1026 1027 lhsSignificand = significandParts(); 1028 rhsSignificand = rhs.significandParts(); 1029 partsCount = partCount(); 1030 1031 if (partsCount > 2) 1032 dividend = new integerPart[partsCount * 2]; 1033 else 1034 dividend = scratch; 1035 1036 divisor = dividend + partsCount; 1037 1038 /* Copy the dividend and divisor as they will be modified in-place. */ 1039 for (i = 0; i < partsCount; i++) { 1040 dividend[i] = lhsSignificand[i]; 1041 divisor[i] = rhsSignificand[i]; 1042 lhsSignificand[i] = 0; 1043 } 1044 1045 exponent -= rhs.exponent; 1046 1047 unsigned int precision = semantics->precision; 1048 1049 /* Normalize the divisor. */ 1050 bit = precision - APInt::tcMSB(divisor, partsCount) - 1; 1051 if (bit) { 1052 exponent += bit; 1053 APInt::tcShiftLeft(divisor, partsCount, bit); 1054 } 1055 1056 /* Normalize the dividend. */ 1057 bit = precision - APInt::tcMSB(dividend, partsCount) - 1; 1058 if (bit) { 1059 exponent -= bit; 1060 APInt::tcShiftLeft(dividend, partsCount, bit); 1061 } 1062 1063 /* Ensure the dividend >= divisor initially for the loop below. 1064 Incidentally, this means that the division loop below is 1065 guaranteed to set the integer bit to one. */ 1066 if (APInt::tcCompare(dividend, divisor, partsCount) < 0) { 1067 exponent--; 1068 APInt::tcShiftLeft(dividend, partsCount, 1); 1069 assert(APInt::tcCompare(dividend, divisor, partsCount) >= 0); 1070 } 1071 1072 /* Long division. */ 1073 for (bit = precision; bit; bit -= 1) { 1074 if (APInt::tcCompare(dividend, divisor, partsCount) >= 0) { 1075 APInt::tcSubtract(dividend, divisor, 0, partsCount); 1076 APInt::tcSetBit(lhsSignificand, bit - 1); 1077 } 1078 1079 APInt::tcShiftLeft(dividend, partsCount, 1); 1080 } 1081 1082 /* Figure out the lost fraction. */ 1083 int cmp = APInt::tcCompare(dividend, divisor, partsCount); 1084 1085 if (cmp > 0) 1086 lost_fraction = lfMoreThanHalf; 1087 else if (cmp == 0) 1088 lost_fraction = lfExactlyHalf; 1089 else if (APInt::tcIsZero(dividend, partsCount)) 1090 lost_fraction = lfExactlyZero; 1091 else 1092 lost_fraction = lfLessThanHalf; 1093 1094 if (partsCount > 2) 1095 delete [] dividend; 1096 1097 return lost_fraction; 1098 } 1099 1100 unsigned int 1101 APFloat::significandMSB() const 1102 { 1103 return APInt::tcMSB(significandParts(), partCount()); 1104 } 1105 1106 unsigned int 1107 APFloat::significandLSB() const 1108 { 1109 return APInt::tcLSB(significandParts(), partCount()); 1110 } 1111 1112 /* Note that a zero result is NOT normalized to fcZero. */ 1113 lostFraction 1114 APFloat::shiftSignificandRight(unsigned int bits) 1115 { 1116 /* Our exponent should not overflow. */ 1117 assert((ExponentType) (exponent + bits) >= exponent); 1118 1119 exponent += bits; 1120 1121 return shiftRight(significandParts(), partCount(), bits); 1122 } 1123 1124 /* Shift the significand left BITS bits, subtract BITS from its exponent. */ 1125 void 1126 APFloat::shiftSignificandLeft(unsigned int bits) 1127 { 1128 assert(bits < semantics->precision); 1129 1130 if (bits) { 1131 unsigned int partsCount = partCount(); 1132 1133 APInt::tcShiftLeft(significandParts(), partsCount, bits); 1134 exponent -= bits; 1135 1136 assert(!APInt::tcIsZero(significandParts(), partsCount)); 1137 } 1138 } 1139 1140 APFloat::cmpResult 1141 APFloat::compareAbsoluteValue(const APFloat &rhs) const 1142 { 1143 int compare; 1144 1145 assert(semantics == rhs.semantics); 1146 assert(isFiniteNonZero()); 1147 assert(rhs.isFiniteNonZero()); 1148 1149 compare = exponent - rhs.exponent; 1150 1151 /* If exponents are equal, do an unsigned bignum comparison of the 1152 significands. */ 1153 if (compare == 0) 1154 compare = APInt::tcCompare(significandParts(), rhs.significandParts(), 1155 partCount()); 1156 1157 if (compare > 0) 1158 return cmpGreaterThan; 1159 else if (compare < 0) 1160 return cmpLessThan; 1161 else 1162 return cmpEqual; 1163 } 1164 1165 /* Handle overflow. Sign is preserved. We either become infinity or 1166 the largest finite number. */ 1167 APFloat::opStatus 1168 APFloat::handleOverflow(roundingMode rounding_mode) 1169 { 1170 /* Infinity? */ 1171 if (rounding_mode == rmNearestTiesToEven || 1172 rounding_mode == rmNearestTiesToAway || 1173 (rounding_mode == rmTowardPositive && !sign) || 1174 (rounding_mode == rmTowardNegative && sign)) { 1175 category = fcInfinity; 1176 return (opStatus) (opOverflow | opInexact); 1177 } 1178 1179 /* Otherwise we become the largest finite number. */ 1180 category = fcNormal; 1181 exponent = semantics->maxExponent; 1182 APInt::tcSetLeastSignificantBits(significandParts(), partCount(), 1183 semantics->precision); 1184 1185 return opInexact; 1186 } 1187 1188 /* Returns TRUE if, when truncating the current number, with BIT the 1189 new LSB, with the given lost fraction and rounding mode, the result 1190 would need to be rounded away from zero (i.e., by increasing the 1191 signficand). This routine must work for fcZero of both signs, and 1192 fcNormal numbers. */ 1193 bool 1194 APFloat::roundAwayFromZero(roundingMode rounding_mode, 1195 lostFraction lost_fraction, 1196 unsigned int bit) const 1197 { 1198 /* NaNs and infinities should not have lost fractions. */ 1199 assert(isFiniteNonZero() || category == fcZero); 1200 1201 /* Current callers never pass this so we don't handle it. */ 1202 assert(lost_fraction != lfExactlyZero); 1203 1204 switch (rounding_mode) { 1205 case rmNearestTiesToAway: 1206 return lost_fraction == lfExactlyHalf || lost_fraction == lfMoreThanHalf; 1207 1208 case rmNearestTiesToEven: 1209 if (lost_fraction == lfMoreThanHalf) 1210 return true; 1211 1212 /* Our zeroes don't have a significand to test. */ 1213 if (lost_fraction == lfExactlyHalf && category != fcZero) 1214 return APInt::tcExtractBit(significandParts(), bit); 1215 1216 return false; 1217 1218 case rmTowardZero: 1219 return false; 1220 1221 case rmTowardPositive: 1222 return sign == false; 1223 1224 case rmTowardNegative: 1225 return sign == true; 1226 } 1227 llvm_unreachable("Invalid rounding mode found"); 1228 } 1229 1230 APFloat::opStatus 1231 APFloat::normalize(roundingMode rounding_mode, 1232 lostFraction lost_fraction) 1233 { 1234 unsigned int omsb; /* One, not zero, based MSB. */ 1235 int exponentChange; 1236 1237 if (!isFiniteNonZero()) 1238 return opOK; 1239 1240 /* Before rounding normalize the exponent of fcNormal numbers. */ 1241 omsb = significandMSB() + 1; 1242 1243 if (omsb) { 1244 /* OMSB is numbered from 1. We want to place it in the integer 1245 bit numbered PRECISION if possible, with a compensating change in 1246 the exponent. */ 1247 exponentChange = omsb - semantics->precision; 1248 1249 /* If the resulting exponent is too high, overflow according to 1250 the rounding mode. */ 1251 if (exponent + exponentChange > semantics->maxExponent) 1252 return handleOverflow(rounding_mode); 1253 1254 /* Subnormal numbers have exponent minExponent, and their MSB 1255 is forced based on that. */ 1256 if (exponent + exponentChange < semantics->minExponent) 1257 exponentChange = semantics->minExponent - exponent; 1258 1259 /* Shifting left is easy as we don't lose precision. */ 1260 if (exponentChange < 0) { 1261 assert(lost_fraction == lfExactlyZero); 1262 1263 shiftSignificandLeft(-exponentChange); 1264 1265 return opOK; 1266 } 1267 1268 if (exponentChange > 0) { 1269 lostFraction lf; 1270 1271 /* Shift right and capture any new lost fraction. */ 1272 lf = shiftSignificandRight(exponentChange); 1273 1274 lost_fraction = combineLostFractions(lf, lost_fraction); 1275 1276 /* Keep OMSB up-to-date. */ 1277 if (omsb > (unsigned) exponentChange) 1278 omsb -= exponentChange; 1279 else 1280 omsb = 0; 1281 } 1282 } 1283 1284 /* Now round the number according to rounding_mode given the lost 1285 fraction. */ 1286 1287 /* As specified in IEEE 754, since we do not trap we do not report 1288 underflow for exact results. */ 1289 if (lost_fraction == lfExactlyZero) { 1290 /* Canonicalize zeroes. */ 1291 if (omsb == 0) 1292 category = fcZero; 1293 1294 return opOK; 1295 } 1296 1297 /* Increment the significand if we're rounding away from zero. */ 1298 if (roundAwayFromZero(rounding_mode, lost_fraction, 0)) { 1299 if (omsb == 0) 1300 exponent = semantics->minExponent; 1301 1302 incrementSignificand(); 1303 omsb = significandMSB() + 1; 1304 1305 /* Did the significand increment overflow? */ 1306 if (omsb == (unsigned) semantics->precision + 1) { 1307 /* Renormalize by incrementing the exponent and shifting our 1308 significand right one. However if we already have the 1309 maximum exponent we overflow to infinity. */ 1310 if (exponent == semantics->maxExponent) { 1311 category = fcInfinity; 1312 1313 return (opStatus) (opOverflow | opInexact); 1314 } 1315 1316 shiftSignificandRight(1); 1317 1318 return opInexact; 1319 } 1320 } 1321 1322 /* The normal case - we were and are not denormal, and any 1323 significand increment above didn't overflow. */ 1324 if (omsb == semantics->precision) 1325 return opInexact; 1326 1327 /* We have a non-zero denormal. */ 1328 assert(omsb < semantics->precision); 1329 1330 /* Canonicalize zeroes. */ 1331 if (omsb == 0) 1332 category = fcZero; 1333 1334 /* The fcZero case is a denormal that underflowed to zero. */ 1335 return (opStatus) (opUnderflow | opInexact); 1336 } 1337 1338 APFloat::opStatus 1339 APFloat::addOrSubtractSpecials(const APFloat &rhs, bool subtract) 1340 { 1341 switch (PackCategoriesIntoKey(category, rhs.category)) { 1342 default: 1343 llvm_unreachable(0); 1344 1345 case PackCategoriesIntoKey(fcNaN, fcZero): 1346 case PackCategoriesIntoKey(fcNaN, fcNormal): 1347 case PackCategoriesIntoKey(fcNaN, fcInfinity): 1348 case PackCategoriesIntoKey(fcNaN, fcNaN): 1349 case PackCategoriesIntoKey(fcNormal, fcZero): 1350 case PackCategoriesIntoKey(fcInfinity, fcNormal): 1351 case PackCategoriesIntoKey(fcInfinity, fcZero): 1352 return opOK; 1353 1354 case PackCategoriesIntoKey(fcZero, fcNaN): 1355 case PackCategoriesIntoKey(fcNormal, fcNaN): 1356 case PackCategoriesIntoKey(fcInfinity, fcNaN): 1357 sign = false; 1358 category = fcNaN; 1359 copySignificand(rhs); 1360 return opOK; 1361 1362 case PackCategoriesIntoKey(fcNormal, fcInfinity): 1363 case PackCategoriesIntoKey(fcZero, fcInfinity): 1364 category = fcInfinity; 1365 sign = rhs.sign ^ subtract; 1366 return opOK; 1367 1368 case PackCategoriesIntoKey(fcZero, fcNormal): 1369 assign(rhs); 1370 sign = rhs.sign ^ subtract; 1371 return opOK; 1372 1373 case PackCategoriesIntoKey(fcZero, fcZero): 1374 /* Sign depends on rounding mode; handled by caller. */ 1375 return opOK; 1376 1377 case PackCategoriesIntoKey(fcInfinity, fcInfinity): 1378 /* Differently signed infinities can only be validly 1379 subtracted. */ 1380 if (((sign ^ rhs.sign)!=0) != subtract) { 1381 makeNaN(); 1382 return opInvalidOp; 1383 } 1384 1385 return opOK; 1386 1387 case PackCategoriesIntoKey(fcNormal, fcNormal): 1388 return opDivByZero; 1389 } 1390 } 1391 1392 /* Add or subtract two normal numbers. */ 1393 lostFraction 1394 APFloat::addOrSubtractSignificand(const APFloat &rhs, bool subtract) 1395 { 1396 integerPart carry; 1397 lostFraction lost_fraction; 1398 int bits; 1399 1400 /* Determine if the operation on the absolute values is effectively 1401 an addition or subtraction. */ 1402 subtract ^= (sign ^ rhs.sign) ? true : false; 1403 1404 /* Are we bigger exponent-wise than the RHS? */ 1405 bits = exponent - rhs.exponent; 1406 1407 /* Subtraction is more subtle than one might naively expect. */ 1408 if (subtract) { 1409 APFloat temp_rhs(rhs); 1410 bool reverse; 1411 1412 if (bits == 0) { 1413 reverse = compareAbsoluteValue(temp_rhs) == cmpLessThan; 1414 lost_fraction = lfExactlyZero; 1415 } else if (bits > 0) { 1416 lost_fraction = temp_rhs.shiftSignificandRight(bits - 1); 1417 shiftSignificandLeft(1); 1418 reverse = false; 1419 } else { 1420 lost_fraction = shiftSignificandRight(-bits - 1); 1421 temp_rhs.shiftSignificandLeft(1); 1422 reverse = true; 1423 } 1424 1425 if (reverse) { 1426 carry = temp_rhs.subtractSignificand 1427 (*this, lost_fraction != lfExactlyZero); 1428 copySignificand(temp_rhs); 1429 sign = !sign; 1430 } else { 1431 carry = subtractSignificand 1432 (temp_rhs, lost_fraction != lfExactlyZero); 1433 } 1434 1435 /* Invert the lost fraction - it was on the RHS and 1436 subtracted. */ 1437 if (lost_fraction == lfLessThanHalf) 1438 lost_fraction = lfMoreThanHalf; 1439 else if (lost_fraction == lfMoreThanHalf) 1440 lost_fraction = lfLessThanHalf; 1441 1442 /* The code above is intended to ensure that no borrow is 1443 necessary. */ 1444 assert(!carry); 1445 (void)carry; 1446 } else { 1447 if (bits > 0) { 1448 APFloat temp_rhs(rhs); 1449 1450 lost_fraction = temp_rhs.shiftSignificandRight(bits); 1451 carry = addSignificand(temp_rhs); 1452 } else { 1453 lost_fraction = shiftSignificandRight(-bits); 1454 carry = addSignificand(rhs); 1455 } 1456 1457 /* We have a guard bit; generating a carry cannot happen. */ 1458 assert(!carry); 1459 (void)carry; 1460 } 1461 1462 return lost_fraction; 1463 } 1464 1465 APFloat::opStatus 1466 APFloat::multiplySpecials(const APFloat &rhs) 1467 { 1468 switch (PackCategoriesIntoKey(category, rhs.category)) { 1469 default: 1470 llvm_unreachable(0); 1471 1472 case PackCategoriesIntoKey(fcNaN, fcZero): 1473 case PackCategoriesIntoKey(fcNaN, fcNormal): 1474 case PackCategoriesIntoKey(fcNaN, fcInfinity): 1475 case PackCategoriesIntoKey(fcNaN, fcNaN): 1476 sign = false; 1477 return opOK; 1478 1479 case PackCategoriesIntoKey(fcZero, fcNaN): 1480 case PackCategoriesIntoKey(fcNormal, fcNaN): 1481 case PackCategoriesIntoKey(fcInfinity, fcNaN): 1482 sign = false; 1483 category = fcNaN; 1484 copySignificand(rhs); 1485 return opOK; 1486 1487 case PackCategoriesIntoKey(fcNormal, fcInfinity): 1488 case PackCategoriesIntoKey(fcInfinity, fcNormal): 1489 case PackCategoriesIntoKey(fcInfinity, fcInfinity): 1490 category = fcInfinity; 1491 return opOK; 1492 1493 case PackCategoriesIntoKey(fcZero, fcNormal): 1494 case PackCategoriesIntoKey(fcNormal, fcZero): 1495 case PackCategoriesIntoKey(fcZero, fcZero): 1496 category = fcZero; 1497 return opOK; 1498 1499 case PackCategoriesIntoKey(fcZero, fcInfinity): 1500 case PackCategoriesIntoKey(fcInfinity, fcZero): 1501 makeNaN(); 1502 return opInvalidOp; 1503 1504 case PackCategoriesIntoKey(fcNormal, fcNormal): 1505 return opOK; 1506 } 1507 } 1508 1509 APFloat::opStatus 1510 APFloat::divideSpecials(const APFloat &rhs) 1511 { 1512 switch (PackCategoriesIntoKey(category, rhs.category)) { 1513 default: 1514 llvm_unreachable(0); 1515 1516 case PackCategoriesIntoKey(fcZero, fcNaN): 1517 case PackCategoriesIntoKey(fcNormal, fcNaN): 1518 case PackCategoriesIntoKey(fcInfinity, fcNaN): 1519 category = fcNaN; 1520 copySignificand(rhs); 1521 case PackCategoriesIntoKey(fcNaN, fcZero): 1522 case PackCategoriesIntoKey(fcNaN, fcNormal): 1523 case PackCategoriesIntoKey(fcNaN, fcInfinity): 1524 case PackCategoriesIntoKey(fcNaN, fcNaN): 1525 sign = false; 1526 case PackCategoriesIntoKey(fcInfinity, fcZero): 1527 case PackCategoriesIntoKey(fcInfinity, fcNormal): 1528 case PackCategoriesIntoKey(fcZero, fcInfinity): 1529 case PackCategoriesIntoKey(fcZero, fcNormal): 1530 return opOK; 1531 1532 case PackCategoriesIntoKey(fcNormal, fcInfinity): 1533 category = fcZero; 1534 return opOK; 1535 1536 case PackCategoriesIntoKey(fcNormal, fcZero): 1537 category = fcInfinity; 1538 return opDivByZero; 1539 1540 case PackCategoriesIntoKey(fcInfinity, fcInfinity): 1541 case PackCategoriesIntoKey(fcZero, fcZero): 1542 makeNaN(); 1543 return opInvalidOp; 1544 1545 case PackCategoriesIntoKey(fcNormal, fcNormal): 1546 return opOK; 1547 } 1548 } 1549 1550 APFloat::opStatus 1551 APFloat::modSpecials(const APFloat &rhs) 1552 { 1553 switch (PackCategoriesIntoKey(category, rhs.category)) { 1554 default: 1555 llvm_unreachable(0); 1556 1557 case PackCategoriesIntoKey(fcNaN, fcZero): 1558 case PackCategoriesIntoKey(fcNaN, fcNormal): 1559 case PackCategoriesIntoKey(fcNaN, fcInfinity): 1560 case PackCategoriesIntoKey(fcNaN, fcNaN): 1561 case PackCategoriesIntoKey(fcZero, fcInfinity): 1562 case PackCategoriesIntoKey(fcZero, fcNormal): 1563 case PackCategoriesIntoKey(fcNormal, fcInfinity): 1564 return opOK; 1565 1566 case PackCategoriesIntoKey(fcZero, fcNaN): 1567 case PackCategoriesIntoKey(fcNormal, fcNaN): 1568 case PackCategoriesIntoKey(fcInfinity, fcNaN): 1569 sign = false; 1570 category = fcNaN; 1571 copySignificand(rhs); 1572 return opOK; 1573 1574 case PackCategoriesIntoKey(fcNormal, fcZero): 1575 case PackCategoriesIntoKey(fcInfinity, fcZero): 1576 case PackCategoriesIntoKey(fcInfinity, fcNormal): 1577 case PackCategoriesIntoKey(fcInfinity, fcInfinity): 1578 case PackCategoriesIntoKey(fcZero, fcZero): 1579 makeNaN(); 1580 return opInvalidOp; 1581 1582 case PackCategoriesIntoKey(fcNormal, fcNormal): 1583 return opOK; 1584 } 1585 } 1586 1587 /* Change sign. */ 1588 void 1589 APFloat::changeSign() 1590 { 1591 /* Look mummy, this one's easy. */ 1592 sign = !sign; 1593 } 1594 1595 void 1596 APFloat::clearSign() 1597 { 1598 /* So is this one. */ 1599 sign = 0; 1600 } 1601 1602 void 1603 APFloat::copySign(const APFloat &rhs) 1604 { 1605 /* And this one. */ 1606 sign = rhs.sign; 1607 } 1608 1609 /* Normalized addition or subtraction. */ 1610 APFloat::opStatus 1611 APFloat::addOrSubtract(const APFloat &rhs, roundingMode rounding_mode, 1612 bool subtract) 1613 { 1614 opStatus fs; 1615 1616 fs = addOrSubtractSpecials(rhs, subtract); 1617 1618 /* This return code means it was not a simple case. */ 1619 if (fs == opDivByZero) { 1620 lostFraction lost_fraction; 1621 1622 lost_fraction = addOrSubtractSignificand(rhs, subtract); 1623 fs = normalize(rounding_mode, lost_fraction); 1624 1625 /* Can only be zero if we lost no fraction. */ 1626 assert(category != fcZero || lost_fraction == lfExactlyZero); 1627 } 1628 1629 /* If two numbers add (exactly) to zero, IEEE 754 decrees it is a 1630 positive zero unless rounding to minus infinity, except that 1631 adding two like-signed zeroes gives that zero. */ 1632 if (category == fcZero) { 1633 if (rhs.category != fcZero || (sign == rhs.sign) == subtract) 1634 sign = (rounding_mode == rmTowardNegative); 1635 } 1636 1637 return fs; 1638 } 1639 1640 /* Normalized addition. */ 1641 APFloat::opStatus 1642 APFloat::add(const APFloat &rhs, roundingMode rounding_mode) 1643 { 1644 return addOrSubtract(rhs, rounding_mode, false); 1645 } 1646 1647 /* Normalized subtraction. */ 1648 APFloat::opStatus 1649 APFloat::subtract(const APFloat &rhs, roundingMode rounding_mode) 1650 { 1651 return addOrSubtract(rhs, rounding_mode, true); 1652 } 1653 1654 /* Normalized multiply. */ 1655 APFloat::opStatus 1656 APFloat::multiply(const APFloat &rhs, roundingMode rounding_mode) 1657 { 1658 opStatus fs; 1659 1660 sign ^= rhs.sign; 1661 fs = multiplySpecials(rhs); 1662 1663 if (isFiniteNonZero()) { 1664 lostFraction lost_fraction = multiplySignificand(rhs, 0); 1665 fs = normalize(rounding_mode, lost_fraction); 1666 if (lost_fraction != lfExactlyZero) 1667 fs = (opStatus) (fs | opInexact); 1668 } 1669 1670 return fs; 1671 } 1672 1673 /* Normalized divide. */ 1674 APFloat::opStatus 1675 APFloat::divide(const APFloat &rhs, roundingMode rounding_mode) 1676 { 1677 opStatus fs; 1678 1679 sign ^= rhs.sign; 1680 fs = divideSpecials(rhs); 1681 1682 if (isFiniteNonZero()) { 1683 lostFraction lost_fraction = divideSignificand(rhs); 1684 fs = normalize(rounding_mode, lost_fraction); 1685 if (lost_fraction != lfExactlyZero) 1686 fs = (opStatus) (fs | opInexact); 1687 } 1688 1689 return fs; 1690 } 1691 1692 /* Normalized remainder. This is not currently correct in all cases. */ 1693 APFloat::opStatus 1694 APFloat::remainder(const APFloat &rhs) 1695 { 1696 opStatus fs; 1697 APFloat V = *this; 1698 unsigned int origSign = sign; 1699 1700 fs = V.divide(rhs, rmNearestTiesToEven); 1701 if (fs == opDivByZero) 1702 return fs; 1703 1704 int parts = partCount(); 1705 integerPart *x = new integerPart[parts]; 1706 bool ignored; 1707 fs = V.convertToInteger(x, parts * integerPartWidth, true, 1708 rmNearestTiesToEven, &ignored); 1709 if (fs==opInvalidOp) 1710 return fs; 1711 1712 fs = V.convertFromZeroExtendedInteger(x, parts * integerPartWidth, true, 1713 rmNearestTiesToEven); 1714 assert(fs==opOK); // should always work 1715 1716 fs = V.multiply(rhs, rmNearestTiesToEven); 1717 assert(fs==opOK || fs==opInexact); // should not overflow or underflow 1718 1719 fs = subtract(V, rmNearestTiesToEven); 1720 assert(fs==opOK || fs==opInexact); // likewise 1721 1722 if (isZero()) 1723 sign = origSign; // IEEE754 requires this 1724 delete[] x; 1725 return fs; 1726 } 1727 1728 /* Normalized llvm frem (C fmod). 1729 This is not currently correct in all cases. */ 1730 APFloat::opStatus 1731 APFloat::mod(const APFloat &rhs, roundingMode rounding_mode) 1732 { 1733 opStatus fs; 1734 fs = modSpecials(rhs); 1735 1736 if (isFiniteNonZero() && rhs.isFiniteNonZero()) { 1737 APFloat V = *this; 1738 unsigned int origSign = sign; 1739 1740 fs = V.divide(rhs, rmNearestTiesToEven); 1741 if (fs == opDivByZero) 1742 return fs; 1743 1744 int parts = partCount(); 1745 integerPart *x = new integerPart[parts]; 1746 bool ignored; 1747 fs = V.convertToInteger(x, parts * integerPartWidth, true, 1748 rmTowardZero, &ignored); 1749 if (fs==opInvalidOp) 1750 return fs; 1751 1752 fs = V.convertFromZeroExtendedInteger(x, parts * integerPartWidth, true, 1753 rmNearestTiesToEven); 1754 assert(fs==opOK); // should always work 1755 1756 fs = V.multiply(rhs, rounding_mode); 1757 assert(fs==opOK || fs==opInexact); // should not overflow or underflow 1758 1759 fs = subtract(V, rounding_mode); 1760 assert(fs==opOK || fs==opInexact); // likewise 1761 1762 if (isZero()) 1763 sign = origSign; // IEEE754 requires this 1764 delete[] x; 1765 } 1766 return fs; 1767 } 1768 1769 /* Normalized fused-multiply-add. */ 1770 APFloat::opStatus 1771 APFloat::fusedMultiplyAdd(const APFloat &multiplicand, 1772 const APFloat &addend, 1773 roundingMode rounding_mode) 1774 { 1775 opStatus fs; 1776 1777 /* Post-multiplication sign, before addition. */ 1778 sign ^= multiplicand.sign; 1779 1780 /* If and only if all arguments are normal do we need to do an 1781 extended-precision calculation. */ 1782 if (isFiniteNonZero() && 1783 multiplicand.isFiniteNonZero() && 1784 addend.isFiniteNonZero()) { 1785 lostFraction lost_fraction; 1786 1787 lost_fraction = multiplySignificand(multiplicand, &addend); 1788 fs = normalize(rounding_mode, lost_fraction); 1789 if (lost_fraction != lfExactlyZero) 1790 fs = (opStatus) (fs | opInexact); 1791 1792 /* If two numbers add (exactly) to zero, IEEE 754 decrees it is a 1793 positive zero unless rounding to minus infinity, except that 1794 adding two like-signed zeroes gives that zero. */ 1795 if (category == fcZero && sign != addend.sign) 1796 sign = (rounding_mode == rmTowardNegative); 1797 } else { 1798 fs = multiplySpecials(multiplicand); 1799 1800 /* FS can only be opOK or opInvalidOp. There is no more work 1801 to do in the latter case. The IEEE-754R standard says it is 1802 implementation-defined in this case whether, if ADDEND is a 1803 quiet NaN, we raise invalid op; this implementation does so. 1804 1805 If we need to do the addition we can do so with normal 1806 precision. */ 1807 if (fs == opOK) 1808 fs = addOrSubtract(addend, rounding_mode, false); 1809 } 1810 1811 return fs; 1812 } 1813 1814 /* Rounding-mode corrrect round to integral value. */ 1815 APFloat::opStatus APFloat::roundToIntegral(roundingMode rounding_mode) { 1816 opStatus fs; 1817 1818 // If the exponent is large enough, we know that this value is already 1819 // integral, and the arithmetic below would potentially cause it to saturate 1820 // to +/-Inf. Bail out early instead. 1821 if (isFiniteNonZero() && exponent+1 >= (int)semanticsPrecision(*semantics)) 1822 return opOK; 1823 1824 // The algorithm here is quite simple: we add 2^(p-1), where p is the 1825 // precision of our format, and then subtract it back off again. The choice 1826 // of rounding modes for the addition/subtraction determines the rounding mode 1827 // for our integral rounding as well. 1828 // NOTE: When the input value is negative, we do subtraction followed by 1829 // addition instead. 1830 APInt IntegerConstant(NextPowerOf2(semanticsPrecision(*semantics)), 1); 1831 IntegerConstant <<= semanticsPrecision(*semantics)-1; 1832 APFloat MagicConstant(*semantics); 1833 fs = MagicConstant.convertFromAPInt(IntegerConstant, false, 1834 rmNearestTiesToEven); 1835 MagicConstant.copySign(*this); 1836 1837 if (fs != opOK) 1838 return fs; 1839 1840 // Preserve the input sign so that we can handle 0.0/-0.0 cases correctly. 1841 bool inputSign = isNegative(); 1842 1843 fs = add(MagicConstant, rounding_mode); 1844 if (fs != opOK && fs != opInexact) 1845 return fs; 1846 1847 fs = subtract(MagicConstant, rounding_mode); 1848 1849 // Restore the input sign. 1850 if (inputSign != isNegative()) 1851 changeSign(); 1852 1853 return fs; 1854 } 1855 1856 1857 /* Comparison requires normalized numbers. */ 1858 APFloat::cmpResult 1859 APFloat::compare(const APFloat &rhs) const 1860 { 1861 cmpResult result; 1862 1863 assert(semantics == rhs.semantics); 1864 1865 switch (PackCategoriesIntoKey(category, rhs.category)) { 1866 default: 1867 llvm_unreachable(0); 1868 1869 case PackCategoriesIntoKey(fcNaN, fcZero): 1870 case PackCategoriesIntoKey(fcNaN, fcNormal): 1871 case PackCategoriesIntoKey(fcNaN, fcInfinity): 1872 case PackCategoriesIntoKey(fcNaN, fcNaN): 1873 case PackCategoriesIntoKey(fcZero, fcNaN): 1874 case PackCategoriesIntoKey(fcNormal, fcNaN): 1875 case PackCategoriesIntoKey(fcInfinity, fcNaN): 1876 return cmpUnordered; 1877 1878 case PackCategoriesIntoKey(fcInfinity, fcNormal): 1879 case PackCategoriesIntoKey(fcInfinity, fcZero): 1880 case PackCategoriesIntoKey(fcNormal, fcZero): 1881 if (sign) 1882 return cmpLessThan; 1883 else 1884 return cmpGreaterThan; 1885 1886 case PackCategoriesIntoKey(fcNormal, fcInfinity): 1887 case PackCategoriesIntoKey(fcZero, fcInfinity): 1888 case PackCategoriesIntoKey(fcZero, fcNormal): 1889 if (rhs.sign) 1890 return cmpGreaterThan; 1891 else 1892 return cmpLessThan; 1893 1894 case PackCategoriesIntoKey(fcInfinity, fcInfinity): 1895 if (sign == rhs.sign) 1896 return cmpEqual; 1897 else if (sign) 1898 return cmpLessThan; 1899 else 1900 return cmpGreaterThan; 1901 1902 case PackCategoriesIntoKey(fcZero, fcZero): 1903 return cmpEqual; 1904 1905 case PackCategoriesIntoKey(fcNormal, fcNormal): 1906 break; 1907 } 1908 1909 /* Two normal numbers. Do they have the same sign? */ 1910 if (sign != rhs.sign) { 1911 if (sign) 1912 result = cmpLessThan; 1913 else 1914 result = cmpGreaterThan; 1915 } else { 1916 /* Compare absolute values; invert result if negative. */ 1917 result = compareAbsoluteValue(rhs); 1918 1919 if (sign) { 1920 if (result == cmpLessThan) 1921 result = cmpGreaterThan; 1922 else if (result == cmpGreaterThan) 1923 result = cmpLessThan; 1924 } 1925 } 1926 1927 return result; 1928 } 1929 1930 /// APFloat::convert - convert a value of one floating point type to another. 1931 /// The return value corresponds to the IEEE754 exceptions. *losesInfo 1932 /// records whether the transformation lost information, i.e. whether 1933 /// converting the result back to the original type will produce the 1934 /// original value (this is almost the same as return value==fsOK, but there 1935 /// are edge cases where this is not so). 1936 1937 APFloat::opStatus 1938 APFloat::convert(const fltSemantics &toSemantics, 1939 roundingMode rounding_mode, bool *losesInfo) 1940 { 1941 lostFraction lostFraction; 1942 unsigned int newPartCount, oldPartCount; 1943 opStatus fs; 1944 int shift; 1945 const fltSemantics &fromSemantics = *semantics; 1946 1947 lostFraction = lfExactlyZero; 1948 newPartCount = partCountForBits(toSemantics.precision + 1); 1949 oldPartCount = partCount(); 1950 shift = toSemantics.precision - fromSemantics.precision; 1951 1952 bool X86SpecialNan = false; 1953 if (&fromSemantics == &APFloat::x87DoubleExtended && 1954 &toSemantics != &APFloat::x87DoubleExtended && category == fcNaN && 1955 (!(*significandParts() & 0x8000000000000000ULL) || 1956 !(*significandParts() & 0x4000000000000000ULL))) { 1957 // x86 has some unusual NaNs which cannot be represented in any other 1958 // format; note them here. 1959 X86SpecialNan = true; 1960 } 1961 1962 // If this is a truncation of a denormal number, and the target semantics 1963 // has larger exponent range than the source semantics (this can happen 1964 // when truncating from PowerPC double-double to double format), the 1965 // right shift could lose result mantissa bits. Adjust exponent instead 1966 // of performing excessive shift. 1967 if (shift < 0 && isFiniteNonZero()) { 1968 int exponentChange = significandMSB() + 1 - fromSemantics.precision; 1969 if (exponent + exponentChange < toSemantics.minExponent) 1970 exponentChange = toSemantics.minExponent - exponent; 1971 if (exponentChange < shift) 1972 exponentChange = shift; 1973 if (exponentChange < 0) { 1974 shift -= exponentChange; 1975 exponent += exponentChange; 1976 } 1977 } 1978 1979 // If this is a truncation, perform the shift before we narrow the storage. 1980 if (shift < 0 && (isFiniteNonZero() || category==fcNaN)) 1981 lostFraction = shiftRight(significandParts(), oldPartCount, -shift); 1982 1983 // Fix the storage so it can hold to new value. 1984 if (newPartCount > oldPartCount) { 1985 // The new type requires more storage; make it available. 1986 integerPart *newParts; 1987 newParts = new integerPart[newPartCount]; 1988 APInt::tcSet(newParts, 0, newPartCount); 1989 if (isFiniteNonZero() || category==fcNaN) 1990 APInt::tcAssign(newParts, significandParts(), oldPartCount); 1991 freeSignificand(); 1992 significand.parts = newParts; 1993 } else if (newPartCount == 1 && oldPartCount != 1) { 1994 // Switch to built-in storage for a single part. 1995 integerPart newPart = 0; 1996 if (isFiniteNonZero() || category==fcNaN) 1997 newPart = significandParts()[0]; 1998 freeSignificand(); 1999 significand.part = newPart; 2000 } 2001 2002 // Now that we have the right storage, switch the semantics. 2003 semantics = &toSemantics; 2004 2005 // If this is an extension, perform the shift now that the storage is 2006 // available. 2007 if (shift > 0 && (isFiniteNonZero() || category==fcNaN)) 2008 APInt::tcShiftLeft(significandParts(), newPartCount, shift); 2009 2010 if (isFiniteNonZero()) { 2011 fs = normalize(rounding_mode, lostFraction); 2012 *losesInfo = (fs != opOK); 2013 } else if (category == fcNaN) { 2014 *losesInfo = lostFraction != lfExactlyZero || X86SpecialNan; 2015 2016 // For x87 extended precision, we want to make a NaN, not a special NaN if 2017 // the input wasn't special either. 2018 if (!X86SpecialNan && semantics == &APFloat::x87DoubleExtended) 2019 APInt::tcSetBit(significandParts(), semantics->precision - 1); 2020 2021 // gcc forces the Quiet bit on, which means (float)(double)(float_sNan) 2022 // does not give you back the same bits. This is dubious, and we 2023 // don't currently do it. You're really supposed to get 2024 // an invalid operation signal at runtime, but nobody does that. 2025 fs = opOK; 2026 } else { 2027 *losesInfo = false; 2028 fs = opOK; 2029 } 2030 2031 return fs; 2032 } 2033 2034 /* Convert a floating point number to an integer according to the 2035 rounding mode. If the rounded integer value is out of range this 2036 returns an invalid operation exception and the contents of the 2037 destination parts are unspecified. If the rounded value is in 2038 range but the floating point number is not the exact integer, the C 2039 standard doesn't require an inexact exception to be raised. IEEE 2040 854 does require it so we do that. 2041 2042 Note that for conversions to integer type the C standard requires 2043 round-to-zero to always be used. */ 2044 APFloat::opStatus 2045 APFloat::convertToSignExtendedInteger(integerPart *parts, unsigned int width, 2046 bool isSigned, 2047 roundingMode rounding_mode, 2048 bool *isExact) const 2049 { 2050 lostFraction lost_fraction; 2051 const integerPart *src; 2052 unsigned int dstPartsCount, truncatedBits; 2053 2054 *isExact = false; 2055 2056 /* Handle the three special cases first. */ 2057 if (category == fcInfinity || category == fcNaN) 2058 return opInvalidOp; 2059 2060 dstPartsCount = partCountForBits(width); 2061 2062 if (category == fcZero) { 2063 APInt::tcSet(parts, 0, dstPartsCount); 2064 // Negative zero can't be represented as an int. 2065 *isExact = !sign; 2066 return opOK; 2067 } 2068 2069 src = significandParts(); 2070 2071 /* Step 1: place our absolute value, with any fraction truncated, in 2072 the destination. */ 2073 if (exponent < 0) { 2074 /* Our absolute value is less than one; truncate everything. */ 2075 APInt::tcSet(parts, 0, dstPartsCount); 2076 /* For exponent -1 the integer bit represents .5, look at that. 2077 For smaller exponents leftmost truncated bit is 0. */ 2078 truncatedBits = semantics->precision -1U - exponent; 2079 } else { 2080 /* We want the most significant (exponent + 1) bits; the rest are 2081 truncated. */ 2082 unsigned int bits = exponent + 1U; 2083 2084 /* Hopelessly large in magnitude? */ 2085 if (bits > width) 2086 return opInvalidOp; 2087 2088 if (bits < semantics->precision) { 2089 /* We truncate (semantics->precision - bits) bits. */ 2090 truncatedBits = semantics->precision - bits; 2091 APInt::tcExtract(parts, dstPartsCount, src, bits, truncatedBits); 2092 } else { 2093 /* We want at least as many bits as are available. */ 2094 APInt::tcExtract(parts, dstPartsCount, src, semantics->precision, 0); 2095 APInt::tcShiftLeft(parts, dstPartsCount, bits - semantics->precision); 2096 truncatedBits = 0; 2097 } 2098 } 2099 2100 /* Step 2: work out any lost fraction, and increment the absolute 2101 value if we would round away from zero. */ 2102 if (truncatedBits) { 2103 lost_fraction = lostFractionThroughTruncation(src, partCount(), 2104 truncatedBits); 2105 if (lost_fraction != lfExactlyZero && 2106 roundAwayFromZero(rounding_mode, lost_fraction, truncatedBits)) { 2107 if (APInt::tcIncrement(parts, dstPartsCount)) 2108 return opInvalidOp; /* Overflow. */ 2109 } 2110 } else { 2111 lost_fraction = lfExactlyZero; 2112 } 2113 2114 /* Step 3: check if we fit in the destination. */ 2115 unsigned int omsb = APInt::tcMSB(parts, dstPartsCount) + 1; 2116 2117 if (sign) { 2118 if (!isSigned) { 2119 /* Negative numbers cannot be represented as unsigned. */ 2120 if (omsb != 0) 2121 return opInvalidOp; 2122 } else { 2123 /* It takes omsb bits to represent the unsigned integer value. 2124 We lose a bit for the sign, but care is needed as the 2125 maximally negative integer is a special case. */ 2126 if (omsb == width && APInt::tcLSB(parts, dstPartsCount) + 1 != omsb) 2127 return opInvalidOp; 2128 2129 /* This case can happen because of rounding. */ 2130 if (omsb > width) 2131 return opInvalidOp; 2132 } 2133 2134 APInt::tcNegate (parts, dstPartsCount); 2135 } else { 2136 if (omsb >= width + !isSigned) 2137 return opInvalidOp; 2138 } 2139 2140 if (lost_fraction == lfExactlyZero) { 2141 *isExact = true; 2142 return opOK; 2143 } else 2144 return opInexact; 2145 } 2146 2147 /* Same as convertToSignExtendedInteger, except we provide 2148 deterministic values in case of an invalid operation exception, 2149 namely zero for NaNs and the minimal or maximal value respectively 2150 for underflow or overflow. 2151 The *isExact output tells whether the result is exact, in the sense 2152 that converting it back to the original floating point type produces 2153 the original value. This is almost equivalent to result==opOK, 2154 except for negative zeroes. 2155 */ 2156 APFloat::opStatus 2157 APFloat::convertToInteger(integerPart *parts, unsigned int width, 2158 bool isSigned, 2159 roundingMode rounding_mode, bool *isExact) const 2160 { 2161 opStatus fs; 2162 2163 fs = convertToSignExtendedInteger(parts, width, isSigned, rounding_mode, 2164 isExact); 2165 2166 if (fs == opInvalidOp) { 2167 unsigned int bits, dstPartsCount; 2168 2169 dstPartsCount = partCountForBits(width); 2170 2171 if (category == fcNaN) 2172 bits = 0; 2173 else if (sign) 2174 bits = isSigned; 2175 else 2176 bits = width - isSigned; 2177 2178 APInt::tcSetLeastSignificantBits(parts, dstPartsCount, bits); 2179 if (sign && isSigned) 2180 APInt::tcShiftLeft(parts, dstPartsCount, width - 1); 2181 } 2182 2183 return fs; 2184 } 2185 2186 /* Same as convertToInteger(integerPart*, ...), except the result is returned in 2187 an APSInt, whose initial bit-width and signed-ness are used to determine the 2188 precision of the conversion. 2189 */ 2190 APFloat::opStatus 2191 APFloat::convertToInteger(APSInt &result, 2192 roundingMode rounding_mode, bool *isExact) const 2193 { 2194 unsigned bitWidth = result.getBitWidth(); 2195 SmallVector<uint64_t, 4> parts(result.getNumWords()); 2196 opStatus status = convertToInteger( 2197 parts.data(), bitWidth, result.isSigned(), rounding_mode, isExact); 2198 // Keeps the original signed-ness. 2199 result = APInt(bitWidth, parts); 2200 return status; 2201 } 2202 2203 /* Convert an unsigned integer SRC to a floating point number, 2204 rounding according to ROUNDING_MODE. The sign of the floating 2205 point number is not modified. */ 2206 APFloat::opStatus 2207 APFloat::convertFromUnsignedParts(const integerPart *src, 2208 unsigned int srcCount, 2209 roundingMode rounding_mode) 2210 { 2211 unsigned int omsb, precision, dstCount; 2212 integerPart *dst; 2213 lostFraction lost_fraction; 2214 2215 category = fcNormal; 2216 omsb = APInt::tcMSB(src, srcCount) + 1; 2217 dst = significandParts(); 2218 dstCount = partCount(); 2219 precision = semantics->precision; 2220 2221 /* We want the most significant PRECISION bits of SRC. There may not 2222 be that many; extract what we can. */ 2223 if (precision <= omsb) { 2224 exponent = omsb - 1; 2225 lost_fraction = lostFractionThroughTruncation(src, srcCount, 2226 omsb - precision); 2227 APInt::tcExtract(dst, dstCount, src, precision, omsb - precision); 2228 } else { 2229 exponent = precision - 1; 2230 lost_fraction = lfExactlyZero; 2231 APInt::tcExtract(dst, dstCount, src, omsb, 0); 2232 } 2233 2234 return normalize(rounding_mode, lost_fraction); 2235 } 2236 2237 APFloat::opStatus 2238 APFloat::convertFromAPInt(const APInt &Val, 2239 bool isSigned, 2240 roundingMode rounding_mode) 2241 { 2242 unsigned int partCount = Val.getNumWords(); 2243 APInt api = Val; 2244 2245 sign = false; 2246 if (isSigned && api.isNegative()) { 2247 sign = true; 2248 api = -api; 2249 } 2250 2251 return convertFromUnsignedParts(api.getRawData(), partCount, rounding_mode); 2252 } 2253 2254 /* Convert a two's complement integer SRC to a floating point number, 2255 rounding according to ROUNDING_MODE. ISSIGNED is true if the 2256 integer is signed, in which case it must be sign-extended. */ 2257 APFloat::opStatus 2258 APFloat::convertFromSignExtendedInteger(const integerPart *src, 2259 unsigned int srcCount, 2260 bool isSigned, 2261 roundingMode rounding_mode) 2262 { 2263 opStatus status; 2264 2265 if (isSigned && 2266 APInt::tcExtractBit(src, srcCount * integerPartWidth - 1)) { 2267 integerPart *copy; 2268 2269 /* If we're signed and negative negate a copy. */ 2270 sign = true; 2271 copy = new integerPart[srcCount]; 2272 APInt::tcAssign(copy, src, srcCount); 2273 APInt::tcNegate(copy, srcCount); 2274 status = convertFromUnsignedParts(copy, srcCount, rounding_mode); 2275 delete [] copy; 2276 } else { 2277 sign = false; 2278 status = convertFromUnsignedParts(src, srcCount, rounding_mode); 2279 } 2280 2281 return status; 2282 } 2283 2284 /* FIXME: should this just take a const APInt reference? */ 2285 APFloat::opStatus 2286 APFloat::convertFromZeroExtendedInteger(const integerPart *parts, 2287 unsigned int width, bool isSigned, 2288 roundingMode rounding_mode) 2289 { 2290 unsigned int partCount = partCountForBits(width); 2291 APInt api = APInt(width, makeArrayRef(parts, partCount)); 2292 2293 sign = false; 2294 if (isSigned && APInt::tcExtractBit(parts, width - 1)) { 2295 sign = true; 2296 api = -api; 2297 } 2298 2299 return convertFromUnsignedParts(api.getRawData(), partCount, rounding_mode); 2300 } 2301 2302 APFloat::opStatus 2303 APFloat::convertFromHexadecimalString(StringRef s, roundingMode rounding_mode) 2304 { 2305 lostFraction lost_fraction = lfExactlyZero; 2306 2307 category = fcNormal; 2308 zeroSignificand(); 2309 exponent = 0; 2310 2311 integerPart *significand = significandParts(); 2312 unsigned partsCount = partCount(); 2313 unsigned bitPos = partsCount * integerPartWidth; 2314 bool computedTrailingFraction = false; 2315 2316 // Skip leading zeroes and any (hexa)decimal point. 2317 StringRef::iterator begin = s.begin(); 2318 StringRef::iterator end = s.end(); 2319 StringRef::iterator dot; 2320 StringRef::iterator p = skipLeadingZeroesAndAnyDot(begin, end, &dot); 2321 StringRef::iterator firstSignificantDigit = p; 2322 2323 while (p != end) { 2324 integerPart hex_value; 2325 2326 if (*p == '.') { 2327 assert(dot == end && "String contains multiple dots"); 2328 dot = p++; 2329 continue; 2330 } 2331 2332 hex_value = hexDigitValue(*p); 2333 if (hex_value == -1U) 2334 break; 2335 2336 p++; 2337 2338 // Store the number while we have space. 2339 if (bitPos) { 2340 bitPos -= 4; 2341 hex_value <<= bitPos % integerPartWidth; 2342 significand[bitPos / integerPartWidth] |= hex_value; 2343 } else if (!computedTrailingFraction) { 2344 lost_fraction = trailingHexadecimalFraction(p, end, hex_value); 2345 computedTrailingFraction = true; 2346 } 2347 } 2348 2349 /* Hex floats require an exponent but not a hexadecimal point. */ 2350 assert(p != end && "Hex strings require an exponent"); 2351 assert((*p == 'p' || *p == 'P') && "Invalid character in significand"); 2352 assert(p != begin && "Significand has no digits"); 2353 assert((dot == end || p - begin != 1) && "Significand has no digits"); 2354 2355 /* Ignore the exponent if we are zero. */ 2356 if (p != firstSignificantDigit) { 2357 int expAdjustment; 2358 2359 /* Implicit hexadecimal point? */ 2360 if (dot == end) 2361 dot = p; 2362 2363 /* Calculate the exponent adjustment implicit in the number of 2364 significant digits. */ 2365 expAdjustment = static_cast<int>(dot - firstSignificantDigit); 2366 if (expAdjustment < 0) 2367 expAdjustment++; 2368 expAdjustment = expAdjustment * 4 - 1; 2369 2370 /* Adjust for writing the significand starting at the most 2371 significant nibble. */ 2372 expAdjustment += semantics->precision; 2373 expAdjustment -= partsCount * integerPartWidth; 2374 2375 /* Adjust for the given exponent. */ 2376 exponent = totalExponent(p + 1, end, expAdjustment); 2377 } 2378 2379 return normalize(rounding_mode, lost_fraction); 2380 } 2381 2382 APFloat::opStatus 2383 APFloat::roundSignificandWithExponent(const integerPart *decSigParts, 2384 unsigned sigPartCount, int exp, 2385 roundingMode rounding_mode) 2386 { 2387 unsigned int parts, pow5PartCount; 2388 fltSemantics calcSemantics = { 32767, -32767, 0 }; 2389 integerPart pow5Parts[maxPowerOfFiveParts]; 2390 bool isNearest; 2391 2392 isNearest = (rounding_mode == rmNearestTiesToEven || 2393 rounding_mode == rmNearestTiesToAway); 2394 2395 parts = partCountForBits(semantics->precision + 11); 2396 2397 /* Calculate pow(5, abs(exp)). */ 2398 pow5PartCount = powerOf5(pow5Parts, exp >= 0 ? exp: -exp); 2399 2400 for (;; parts *= 2) { 2401 opStatus sigStatus, powStatus; 2402 unsigned int excessPrecision, truncatedBits; 2403 2404 calcSemantics.precision = parts * integerPartWidth - 1; 2405 excessPrecision = calcSemantics.precision - semantics->precision; 2406 truncatedBits = excessPrecision; 2407 2408 APFloat decSig = APFloat::getZero(calcSemantics, sign); 2409 APFloat pow5(calcSemantics); 2410 2411 sigStatus = decSig.convertFromUnsignedParts(decSigParts, sigPartCount, 2412 rmNearestTiesToEven); 2413 powStatus = pow5.convertFromUnsignedParts(pow5Parts, pow5PartCount, 2414 rmNearestTiesToEven); 2415 /* Add exp, as 10^n = 5^n * 2^n. */ 2416 decSig.exponent += exp; 2417 2418 lostFraction calcLostFraction; 2419 integerPart HUerr, HUdistance; 2420 unsigned int powHUerr; 2421 2422 if (exp >= 0) { 2423 /* multiplySignificand leaves the precision-th bit set to 1. */ 2424 calcLostFraction = decSig.multiplySignificand(pow5, NULL); 2425 powHUerr = powStatus != opOK; 2426 } else { 2427 calcLostFraction = decSig.divideSignificand(pow5); 2428 /* Denormal numbers have less precision. */ 2429 if (decSig.exponent < semantics->minExponent) { 2430 excessPrecision += (semantics->minExponent - decSig.exponent); 2431 truncatedBits = excessPrecision; 2432 if (excessPrecision > calcSemantics.precision) 2433 excessPrecision = calcSemantics.precision; 2434 } 2435 /* Extra half-ulp lost in reciprocal of exponent. */ 2436 powHUerr = (powStatus == opOK && calcLostFraction == lfExactlyZero) ? 0:2; 2437 } 2438 2439 /* Both multiplySignificand and divideSignificand return the 2440 result with the integer bit set. */ 2441 assert(APInt::tcExtractBit 2442 (decSig.significandParts(), calcSemantics.precision - 1) == 1); 2443 2444 HUerr = HUerrBound(calcLostFraction != lfExactlyZero, sigStatus != opOK, 2445 powHUerr); 2446 HUdistance = 2 * ulpsFromBoundary(decSig.significandParts(), 2447 excessPrecision, isNearest); 2448 2449 /* Are we guaranteed to round correctly if we truncate? */ 2450 if (HUdistance >= HUerr) { 2451 APInt::tcExtract(significandParts(), partCount(), decSig.significandParts(), 2452 calcSemantics.precision - excessPrecision, 2453 excessPrecision); 2454 /* Take the exponent of decSig. If we tcExtract-ed less bits 2455 above we must adjust our exponent to compensate for the 2456 implicit right shift. */ 2457 exponent = (decSig.exponent + semantics->precision 2458 - (calcSemantics.precision - excessPrecision)); 2459 calcLostFraction = lostFractionThroughTruncation(decSig.significandParts(), 2460 decSig.partCount(), 2461 truncatedBits); 2462 return normalize(rounding_mode, calcLostFraction); 2463 } 2464 } 2465 } 2466 2467 APFloat::opStatus 2468 APFloat::convertFromDecimalString(StringRef str, roundingMode rounding_mode) 2469 { 2470 decimalInfo D; 2471 opStatus fs; 2472 2473 /* Scan the text. */ 2474 StringRef::iterator p = str.begin(); 2475 interpretDecimal(p, str.end(), &D); 2476 2477 /* Handle the quick cases. First the case of no significant digits, 2478 i.e. zero, and then exponents that are obviously too large or too 2479 small. Writing L for log 10 / log 2, a number d.ddddd*10^exp 2480 definitely overflows if 2481 2482 (exp - 1) * L >= maxExponent 2483 2484 and definitely underflows to zero where 2485 2486 (exp + 1) * L <= minExponent - precision 2487 2488 With integer arithmetic the tightest bounds for L are 2489 2490 93/28 < L < 196/59 [ numerator <= 256 ] 2491 42039/12655 < L < 28738/8651 [ numerator <= 65536 ] 2492 */ 2493 2494 // Test if we have a zero number allowing for strings with no null terminators 2495 // and zero decimals with non-zero exponents. 2496 // 2497 // We computed firstSigDigit by ignoring all zeros and dots. Thus if 2498 // D->firstSigDigit equals str.end(), every digit must be a zero and there can 2499 // be at most one dot. On the other hand, if we have a zero with a non-zero 2500 // exponent, then we know that D.firstSigDigit will be non-numeric. 2501 if (D.firstSigDigit == str.end() || decDigitValue(*D.firstSigDigit) >= 10U) { 2502 category = fcZero; 2503 fs = opOK; 2504 2505 /* Check whether the normalized exponent is high enough to overflow 2506 max during the log-rebasing in the max-exponent check below. */ 2507 } else if (D.normalizedExponent - 1 > INT_MAX / 42039) { 2508 fs = handleOverflow(rounding_mode); 2509 2510 /* If it wasn't, then it also wasn't high enough to overflow max 2511 during the log-rebasing in the min-exponent check. Check that it 2512 won't overflow min in either check, then perform the min-exponent 2513 check. */ 2514 } else if (D.normalizedExponent - 1 < INT_MIN / 42039 || 2515 (D.normalizedExponent + 1) * 28738 <= 2516 8651 * (semantics->minExponent - (int) semantics->precision)) { 2517 /* Underflow to zero and round. */ 2518 category = fcNormal; 2519 zeroSignificand(); 2520 fs = normalize(rounding_mode, lfLessThanHalf); 2521 2522 /* We can finally safely perform the max-exponent check. */ 2523 } else if ((D.normalizedExponent - 1) * 42039 2524 >= 12655 * semantics->maxExponent) { 2525 /* Overflow and round. */ 2526 fs = handleOverflow(rounding_mode); 2527 } else { 2528 integerPart *decSignificand; 2529 unsigned int partCount; 2530 2531 /* A tight upper bound on number of bits required to hold an 2532 N-digit decimal integer is N * 196 / 59. Allocate enough space 2533 to hold the full significand, and an extra part required by 2534 tcMultiplyPart. */ 2535 partCount = static_cast<unsigned int>(D.lastSigDigit - D.firstSigDigit) + 1; 2536 partCount = partCountForBits(1 + 196 * partCount / 59); 2537 decSignificand = new integerPart[partCount + 1]; 2538 partCount = 0; 2539 2540 /* Convert to binary efficiently - we do almost all multiplication 2541 in an integerPart. When this would overflow do we do a single 2542 bignum multiplication, and then revert again to multiplication 2543 in an integerPart. */ 2544 do { 2545 integerPart decValue, val, multiplier; 2546 2547 val = 0; 2548 multiplier = 1; 2549 2550 do { 2551 if (*p == '.') { 2552 p++; 2553 if (p == str.end()) { 2554 break; 2555 } 2556 } 2557 decValue = decDigitValue(*p++); 2558 assert(decValue < 10U && "Invalid character in significand"); 2559 multiplier *= 10; 2560 val = val * 10 + decValue; 2561 /* The maximum number that can be multiplied by ten with any 2562 digit added without overflowing an integerPart. */ 2563 } while (p <= D.lastSigDigit && multiplier <= (~ (integerPart) 0 - 9) / 10); 2564 2565 /* Multiply out the current part. */ 2566 APInt::tcMultiplyPart(decSignificand, decSignificand, multiplier, val, 2567 partCount, partCount + 1, false); 2568 2569 /* If we used another part (likely but not guaranteed), increase 2570 the count. */ 2571 if (decSignificand[partCount]) 2572 partCount++; 2573 } while (p <= D.lastSigDigit); 2574 2575 category = fcNormal; 2576 fs = roundSignificandWithExponent(decSignificand, partCount, 2577 D.exponent, rounding_mode); 2578 2579 delete [] decSignificand; 2580 } 2581 2582 return fs; 2583 } 2584 2585 bool 2586 APFloat::convertFromStringSpecials(StringRef str) { 2587 if (str.equals("inf") || str.equals("INFINITY")) { 2588 makeInf(false); 2589 return true; 2590 } 2591 2592 if (str.equals("-inf") || str.equals("-INFINITY")) { 2593 makeInf(true); 2594 return true; 2595 } 2596 2597 if (str.equals("nan") || str.equals("NaN")) { 2598 makeNaN(false, false); 2599 return true; 2600 } 2601 2602 if (str.equals("-nan") || str.equals("-NaN")) { 2603 makeNaN(false, true); 2604 return true; 2605 } 2606 2607 return false; 2608 } 2609 2610 APFloat::opStatus 2611 APFloat::convertFromString(StringRef str, roundingMode rounding_mode) 2612 { 2613 assert(!str.empty() && "Invalid string length"); 2614 2615 // Handle special cases. 2616 if (convertFromStringSpecials(str)) 2617 return opOK; 2618 2619 /* Handle a leading minus sign. */ 2620 StringRef::iterator p = str.begin(); 2621 size_t slen = str.size(); 2622 sign = *p == '-' ? 1 : 0; 2623 if (*p == '-' || *p == '+') { 2624 p++; 2625 slen--; 2626 assert(slen && "String has no digits"); 2627 } 2628 2629 if (slen >= 2 && p[0] == '0' && (p[1] == 'x' || p[1] == 'X')) { 2630 assert(slen - 2 && "Invalid string"); 2631 return convertFromHexadecimalString(StringRef(p + 2, slen - 2), 2632 rounding_mode); 2633 } 2634 2635 return convertFromDecimalString(StringRef(p, slen), rounding_mode); 2636 } 2637 2638 /* Write out a hexadecimal representation of the floating point value 2639 to DST, which must be of sufficient size, in the C99 form 2640 [-]0xh.hhhhp[+-]d. Return the number of characters written, 2641 excluding the terminating NUL. 2642 2643 If UPPERCASE, the output is in upper case, otherwise in lower case. 2644 2645 HEXDIGITS digits appear altogether, rounding the value if 2646 necessary. If HEXDIGITS is 0, the minimal precision to display the 2647 number precisely is used instead. If nothing would appear after 2648 the decimal point it is suppressed. 2649 2650 The decimal exponent is always printed and has at least one digit. 2651 Zero values display an exponent of zero. Infinities and NaNs 2652 appear as "infinity" or "nan" respectively. 2653 2654 The above rules are as specified by C99. There is ambiguity about 2655 what the leading hexadecimal digit should be. This implementation 2656 uses whatever is necessary so that the exponent is displayed as 2657 stored. This implies the exponent will fall within the IEEE format 2658 range, and the leading hexadecimal digit will be 0 (for denormals), 2659 1 (normal numbers) or 2 (normal numbers rounded-away-from-zero with 2660 any other digits zero). 2661 */ 2662 unsigned int 2663 APFloat::convertToHexString(char *dst, unsigned int hexDigits, 2664 bool upperCase, roundingMode rounding_mode) const 2665 { 2666 char *p; 2667 2668 p = dst; 2669 if (sign) 2670 *dst++ = '-'; 2671 2672 switch (category) { 2673 case fcInfinity: 2674 memcpy (dst, upperCase ? infinityU: infinityL, sizeof infinityU - 1); 2675 dst += sizeof infinityL - 1; 2676 break; 2677 2678 case fcNaN: 2679 memcpy (dst, upperCase ? NaNU: NaNL, sizeof NaNU - 1); 2680 dst += sizeof NaNU - 1; 2681 break; 2682 2683 case fcZero: 2684 *dst++ = '0'; 2685 *dst++ = upperCase ? 'X': 'x'; 2686 *dst++ = '0'; 2687 if (hexDigits > 1) { 2688 *dst++ = '.'; 2689 memset (dst, '0', hexDigits - 1); 2690 dst += hexDigits - 1; 2691 } 2692 *dst++ = upperCase ? 'P': 'p'; 2693 *dst++ = '0'; 2694 break; 2695 2696 case fcNormal: 2697 dst = convertNormalToHexString (dst, hexDigits, upperCase, rounding_mode); 2698 break; 2699 } 2700 2701 *dst = 0; 2702 2703 return static_cast<unsigned int>(dst - p); 2704 } 2705 2706 /* Does the hard work of outputting the correctly rounded hexadecimal 2707 form of a normal floating point number with the specified number of 2708 hexadecimal digits. If HEXDIGITS is zero the minimum number of 2709 digits necessary to print the value precisely is output. */ 2710 char * 2711 APFloat::convertNormalToHexString(char *dst, unsigned int hexDigits, 2712 bool upperCase, 2713 roundingMode rounding_mode) const 2714 { 2715 unsigned int count, valueBits, shift, partsCount, outputDigits; 2716 const char *hexDigitChars; 2717 const integerPart *significand; 2718 char *p; 2719 bool roundUp; 2720 2721 *dst++ = '0'; 2722 *dst++ = upperCase ? 'X': 'x'; 2723 2724 roundUp = false; 2725 hexDigitChars = upperCase ? hexDigitsUpper: hexDigitsLower; 2726 2727 significand = significandParts(); 2728 partsCount = partCount(); 2729 2730 /* +3 because the first digit only uses the single integer bit, so 2731 we have 3 virtual zero most-significant-bits. */ 2732 valueBits = semantics->precision + 3; 2733 shift = integerPartWidth - valueBits % integerPartWidth; 2734 2735 /* The natural number of digits required ignoring trailing 2736 insignificant zeroes. */ 2737 outputDigits = (valueBits - significandLSB () + 3) / 4; 2738 2739 /* hexDigits of zero means use the required number for the 2740 precision. Otherwise, see if we are truncating. If we are, 2741 find out if we need to round away from zero. */ 2742 if (hexDigits) { 2743 if (hexDigits < outputDigits) { 2744 /* We are dropping non-zero bits, so need to check how to round. 2745 "bits" is the number of dropped bits. */ 2746 unsigned int bits; 2747 lostFraction fraction; 2748 2749 bits = valueBits - hexDigits * 4; 2750 fraction = lostFractionThroughTruncation (significand, partsCount, bits); 2751 roundUp = roundAwayFromZero(rounding_mode, fraction, bits); 2752 } 2753 outputDigits = hexDigits; 2754 } 2755 2756 /* Write the digits consecutively, and start writing in the location 2757 of the hexadecimal point. We move the most significant digit 2758 left and add the hexadecimal point later. */ 2759 p = ++dst; 2760 2761 count = (valueBits + integerPartWidth - 1) / integerPartWidth; 2762 2763 while (outputDigits && count) { 2764 integerPart part; 2765 2766 /* Put the most significant integerPartWidth bits in "part". */ 2767 if (--count == partsCount) 2768 part = 0; /* An imaginary higher zero part. */ 2769 else 2770 part = significand[count] << shift; 2771 2772 if (count && shift) 2773 part |= significand[count - 1] >> (integerPartWidth - shift); 2774 2775 /* Convert as much of "part" to hexdigits as we can. */ 2776 unsigned int curDigits = integerPartWidth / 4; 2777 2778 if (curDigits > outputDigits) 2779 curDigits = outputDigits; 2780 dst += partAsHex (dst, part, curDigits, hexDigitChars); 2781 outputDigits -= curDigits; 2782 } 2783 2784 if (roundUp) { 2785 char *q = dst; 2786 2787 /* Note that hexDigitChars has a trailing '0'. */ 2788 do { 2789 q--; 2790 *q = hexDigitChars[hexDigitValue (*q) + 1]; 2791 } while (*q == '0'); 2792 assert(q >= p); 2793 } else { 2794 /* Add trailing zeroes. */ 2795 memset (dst, '0', outputDigits); 2796 dst += outputDigits; 2797 } 2798 2799 /* Move the most significant digit to before the point, and if there 2800 is something after the decimal point add it. This must come 2801 after rounding above. */ 2802 p[-1] = p[0]; 2803 if (dst -1 == p) 2804 dst--; 2805 else 2806 p[0] = '.'; 2807 2808 /* Finally output the exponent. */ 2809 *dst++ = upperCase ? 'P': 'p'; 2810 2811 return writeSignedDecimal (dst, exponent); 2812 } 2813 2814 hash_code llvm::hash_value(const APFloat &Arg) { 2815 if (!Arg.isFiniteNonZero()) 2816 return hash_combine((uint8_t)Arg.category, 2817 // NaN has no sign, fix it at zero. 2818 Arg.isNaN() ? (uint8_t)0 : (uint8_t)Arg.sign, 2819 Arg.semantics->precision); 2820 2821 // Normal floats need their exponent and significand hashed. 2822 return hash_combine((uint8_t)Arg.category, (uint8_t)Arg.sign, 2823 Arg.semantics->precision, Arg.exponent, 2824 hash_combine_range( 2825 Arg.significandParts(), 2826 Arg.significandParts() + Arg.partCount())); 2827 } 2828 2829 // Conversion from APFloat to/from host float/double. It may eventually be 2830 // possible to eliminate these and have everybody deal with APFloats, but that 2831 // will take a while. This approach will not easily extend to long double. 2832 // Current implementation requires integerPartWidth==64, which is correct at 2833 // the moment but could be made more general. 2834 2835 // Denormals have exponent minExponent in APFloat, but minExponent-1 in 2836 // the actual IEEE respresentations. We compensate for that here. 2837 2838 APInt 2839 APFloat::convertF80LongDoubleAPFloatToAPInt() const 2840 { 2841 assert(semantics == (const llvm::fltSemantics*)&x87DoubleExtended); 2842 assert(partCount()==2); 2843 2844 uint64_t myexponent, mysignificand; 2845 2846 if (isFiniteNonZero()) { 2847 myexponent = exponent+16383; //bias 2848 mysignificand = significandParts()[0]; 2849 if (myexponent==1 && !(mysignificand & 0x8000000000000000ULL)) 2850 myexponent = 0; // denormal 2851 } else if (category==fcZero) { 2852 myexponent = 0; 2853 mysignificand = 0; 2854 } else if (category==fcInfinity) { 2855 myexponent = 0x7fff; 2856 mysignificand = 0x8000000000000000ULL; 2857 } else { 2858 assert(category == fcNaN && "Unknown category"); 2859 myexponent = 0x7fff; 2860 mysignificand = significandParts()[0]; 2861 } 2862 2863 uint64_t words[2]; 2864 words[0] = mysignificand; 2865 words[1] = ((uint64_t)(sign & 1) << 15) | 2866 (myexponent & 0x7fffLL); 2867 return APInt(80, words); 2868 } 2869 2870 APInt 2871 APFloat::convertPPCDoubleDoubleAPFloatToAPInt() const 2872 { 2873 assert(semantics == (const llvm::fltSemantics*)&PPCDoubleDouble); 2874 assert(partCount()==2); 2875 2876 uint64_t words[2]; 2877 opStatus fs; 2878 bool losesInfo; 2879 2880 // Convert number to double. To avoid spurious underflows, we re- 2881 // normalize against the "double" minExponent first, and only *then* 2882 // truncate the mantissa. The result of that second conversion 2883 // may be inexact, but should never underflow. 2884 // Declare fltSemantics before APFloat that uses it (and 2885 // saves pointer to it) to ensure correct destruction order. 2886 fltSemantics extendedSemantics = *semantics; 2887 extendedSemantics.minExponent = IEEEdouble.minExponent; 2888 APFloat extended(*this); 2889 fs = extended.convert(extendedSemantics, rmNearestTiesToEven, &losesInfo); 2890 assert(fs == opOK && !losesInfo); 2891 (void)fs; 2892 2893 APFloat u(extended); 2894 fs = u.convert(IEEEdouble, rmNearestTiesToEven, &losesInfo); 2895 assert(fs == opOK || fs == opInexact); 2896 (void)fs; 2897 words[0] = *u.convertDoubleAPFloatToAPInt().getRawData(); 2898 2899 // If conversion was exact or resulted in a special case, we're done; 2900 // just set the second double to zero. Otherwise, re-convert back to 2901 // the extended format and compute the difference. This now should 2902 // convert exactly to double. 2903 if (u.isFiniteNonZero() && losesInfo) { 2904 fs = u.convert(extendedSemantics, rmNearestTiesToEven, &losesInfo); 2905 assert(fs == opOK && !losesInfo); 2906 (void)fs; 2907 2908 APFloat v(extended); 2909 v.subtract(u, rmNearestTiesToEven); 2910 fs = v.convert(IEEEdouble, rmNearestTiesToEven, &losesInfo); 2911 assert(fs == opOK && !losesInfo); 2912 (void)fs; 2913 words[1] = *v.convertDoubleAPFloatToAPInt().getRawData(); 2914 } else { 2915 words[1] = 0; 2916 } 2917 2918 return APInt(128, words); 2919 } 2920 2921 APInt 2922 APFloat::convertQuadrupleAPFloatToAPInt() const 2923 { 2924 assert(semantics == (const llvm::fltSemantics*)&IEEEquad); 2925 assert(partCount()==2); 2926 2927 uint64_t myexponent, mysignificand, mysignificand2; 2928 2929 if (isFiniteNonZero()) { 2930 myexponent = exponent+16383; //bias 2931 mysignificand = significandParts()[0]; 2932 mysignificand2 = significandParts()[1]; 2933 if (myexponent==1 && !(mysignificand2 & 0x1000000000000LL)) 2934 myexponent = 0; // denormal 2935 } else if (category==fcZero) { 2936 myexponent = 0; 2937 mysignificand = mysignificand2 = 0; 2938 } else if (category==fcInfinity) { 2939 myexponent = 0x7fff; 2940 mysignificand = mysignificand2 = 0; 2941 } else { 2942 assert(category == fcNaN && "Unknown category!"); 2943 myexponent = 0x7fff; 2944 mysignificand = significandParts()[0]; 2945 mysignificand2 = significandParts()[1]; 2946 } 2947 2948 uint64_t words[2]; 2949 words[0] = mysignificand; 2950 words[1] = ((uint64_t)(sign & 1) << 63) | 2951 ((myexponent & 0x7fff) << 48) | 2952 (mysignificand2 & 0xffffffffffffLL); 2953 2954 return APInt(128, words); 2955 } 2956 2957 APInt 2958 APFloat::convertDoubleAPFloatToAPInt() const 2959 { 2960 assert(semantics == (const llvm::fltSemantics*)&IEEEdouble); 2961 assert(partCount()==1); 2962 2963 uint64_t myexponent, mysignificand; 2964 2965 if (isFiniteNonZero()) { 2966 myexponent = exponent+1023; //bias 2967 mysignificand = *significandParts(); 2968 if (myexponent==1 && !(mysignificand & 0x10000000000000LL)) 2969 myexponent = 0; // denormal 2970 } else if (category==fcZero) { 2971 myexponent = 0; 2972 mysignificand = 0; 2973 } else if (category==fcInfinity) { 2974 myexponent = 0x7ff; 2975 mysignificand = 0; 2976 } else { 2977 assert(category == fcNaN && "Unknown category!"); 2978 myexponent = 0x7ff; 2979 mysignificand = *significandParts(); 2980 } 2981 2982 return APInt(64, ((((uint64_t)(sign & 1) << 63) | 2983 ((myexponent & 0x7ff) << 52) | 2984 (mysignificand & 0xfffffffffffffLL)))); 2985 } 2986 2987 APInt 2988 APFloat::convertFloatAPFloatToAPInt() const 2989 { 2990 assert(semantics == (const llvm::fltSemantics*)&IEEEsingle); 2991 assert(partCount()==1); 2992 2993 uint32_t myexponent, mysignificand; 2994 2995 if (isFiniteNonZero()) { 2996 myexponent = exponent+127; //bias 2997 mysignificand = (uint32_t)*significandParts(); 2998 if (myexponent == 1 && !(mysignificand & 0x800000)) 2999 myexponent = 0; // denormal 3000 } else if (category==fcZero) { 3001 myexponent = 0; 3002 mysignificand = 0; 3003 } else if (category==fcInfinity) { 3004 myexponent = 0xff; 3005 mysignificand = 0; 3006 } else { 3007 assert(category == fcNaN && "Unknown category!"); 3008 myexponent = 0xff; 3009 mysignificand = (uint32_t)*significandParts(); 3010 } 3011 3012 return APInt(32, (((sign&1) << 31) | ((myexponent&0xff) << 23) | 3013 (mysignificand & 0x7fffff))); 3014 } 3015 3016 APInt 3017 APFloat::convertHalfAPFloatToAPInt() const 3018 { 3019 assert(semantics == (const llvm::fltSemantics*)&IEEEhalf); 3020 assert(partCount()==1); 3021 3022 uint32_t myexponent, mysignificand; 3023 3024 if (isFiniteNonZero()) { 3025 myexponent = exponent+15; //bias 3026 mysignificand = (uint32_t)*significandParts(); 3027 if (myexponent == 1 && !(mysignificand & 0x400)) 3028 myexponent = 0; // denormal 3029 } else if (category==fcZero) { 3030 myexponent = 0; 3031 mysignificand = 0; 3032 } else if (category==fcInfinity) { 3033 myexponent = 0x1f; 3034 mysignificand = 0; 3035 } else { 3036 assert(category == fcNaN && "Unknown category!"); 3037 myexponent = 0x1f; 3038 mysignificand = (uint32_t)*significandParts(); 3039 } 3040 3041 return APInt(16, (((sign&1) << 15) | ((myexponent&0x1f) << 10) | 3042 (mysignificand & 0x3ff))); 3043 } 3044 3045 // This function creates an APInt that is just a bit map of the floating 3046 // point constant as it would appear in memory. It is not a conversion, 3047 // and treating the result as a normal integer is unlikely to be useful. 3048 3049 APInt 3050 APFloat::bitcastToAPInt() const 3051 { 3052 if (semantics == (const llvm::fltSemantics*)&IEEEhalf) 3053 return convertHalfAPFloatToAPInt(); 3054 3055 if (semantics == (const llvm::fltSemantics*)&IEEEsingle) 3056 return convertFloatAPFloatToAPInt(); 3057 3058 if (semantics == (const llvm::fltSemantics*)&IEEEdouble) 3059 return convertDoubleAPFloatToAPInt(); 3060 3061 if (semantics == (const llvm::fltSemantics*)&IEEEquad) 3062 return convertQuadrupleAPFloatToAPInt(); 3063 3064 if (semantics == (const llvm::fltSemantics*)&PPCDoubleDouble) 3065 return convertPPCDoubleDoubleAPFloatToAPInt(); 3066 3067 assert(semantics == (const llvm::fltSemantics*)&x87DoubleExtended && 3068 "unknown format!"); 3069 return convertF80LongDoubleAPFloatToAPInt(); 3070 } 3071 3072 float 3073 APFloat::convertToFloat() const 3074 { 3075 assert(semantics == (const llvm::fltSemantics*)&IEEEsingle && 3076 "Float semantics are not IEEEsingle"); 3077 APInt api = bitcastToAPInt(); 3078 return api.bitsToFloat(); 3079 } 3080 3081 double 3082 APFloat::convertToDouble() const 3083 { 3084 assert(semantics == (const llvm::fltSemantics*)&IEEEdouble && 3085 "Float semantics are not IEEEdouble"); 3086 APInt api = bitcastToAPInt(); 3087 return api.bitsToDouble(); 3088 } 3089 3090 /// Integer bit is explicit in this format. Intel hardware (387 and later) 3091 /// does not support these bit patterns: 3092 /// exponent = all 1's, integer bit 0, significand 0 ("pseudoinfinity") 3093 /// exponent = all 1's, integer bit 0, significand nonzero ("pseudoNaN") 3094 /// exponent = 0, integer bit 1 ("pseudodenormal") 3095 /// exponent!=0 nor all 1's, integer bit 0 ("unnormal") 3096 /// At the moment, the first two are treated as NaNs, the second two as Normal. 3097 void 3098 APFloat::initFromF80LongDoubleAPInt(const APInt &api) 3099 { 3100 assert(api.getBitWidth()==80); 3101 uint64_t i1 = api.getRawData()[0]; 3102 uint64_t i2 = api.getRawData()[1]; 3103 uint64_t myexponent = (i2 & 0x7fff); 3104 uint64_t mysignificand = i1; 3105 3106 initialize(&APFloat::x87DoubleExtended); 3107 assert(partCount()==2); 3108 3109 sign = static_cast<unsigned int>(i2>>15); 3110 if (myexponent==0 && mysignificand==0) { 3111 // exponent, significand meaningless 3112 category = fcZero; 3113 } else if (myexponent==0x7fff && mysignificand==0x8000000000000000ULL) { 3114 // exponent, significand meaningless 3115 category = fcInfinity; 3116 } else if (myexponent==0x7fff && mysignificand!=0x8000000000000000ULL) { 3117 // exponent meaningless 3118 category = fcNaN; 3119 significandParts()[0] = mysignificand; 3120 significandParts()[1] = 0; 3121 } else { 3122 category = fcNormal; 3123 exponent = myexponent - 16383; 3124 significandParts()[0] = mysignificand; 3125 significandParts()[1] = 0; 3126 if (myexponent==0) // denormal 3127 exponent = -16382; 3128 } 3129 } 3130 3131 void 3132 APFloat::initFromPPCDoubleDoubleAPInt(const APInt &api) 3133 { 3134 assert(api.getBitWidth()==128); 3135 uint64_t i1 = api.getRawData()[0]; 3136 uint64_t i2 = api.getRawData()[1]; 3137 opStatus fs; 3138 bool losesInfo; 3139 3140 // Get the first double and convert to our format. 3141 initFromDoubleAPInt(APInt(64, i1)); 3142 fs = convert(PPCDoubleDouble, rmNearestTiesToEven, &losesInfo); 3143 assert(fs == opOK && !losesInfo); 3144 (void)fs; 3145 3146 // Unless we have a special case, add in second double. 3147 if (isFiniteNonZero()) { 3148 APFloat v(IEEEdouble, APInt(64, i2)); 3149 fs = v.convert(PPCDoubleDouble, rmNearestTiesToEven, &losesInfo); 3150 assert(fs == opOK && !losesInfo); 3151 (void)fs; 3152 3153 add(v, rmNearestTiesToEven); 3154 } 3155 } 3156 3157 void 3158 APFloat::initFromQuadrupleAPInt(const APInt &api) 3159 { 3160 assert(api.getBitWidth()==128); 3161 uint64_t i1 = api.getRawData()[0]; 3162 uint64_t i2 = api.getRawData()[1]; 3163 uint64_t myexponent = (i2 >> 48) & 0x7fff; 3164 uint64_t mysignificand = i1; 3165 uint64_t mysignificand2 = i2 & 0xffffffffffffLL; 3166 3167 initialize(&APFloat::IEEEquad); 3168 assert(partCount()==2); 3169 3170 sign = static_cast<unsigned int>(i2>>63); 3171 if (myexponent==0 && 3172 (mysignificand==0 && mysignificand2==0)) { 3173 // exponent, significand meaningless 3174 category = fcZero; 3175 } else if (myexponent==0x7fff && 3176 (mysignificand==0 && mysignificand2==0)) { 3177 // exponent, significand meaningless 3178 category = fcInfinity; 3179 } else if (myexponent==0x7fff && 3180 (mysignificand!=0 || mysignificand2 !=0)) { 3181 // exponent meaningless 3182 category = fcNaN; 3183 significandParts()[0] = mysignificand; 3184 significandParts()[1] = mysignificand2; 3185 } else { 3186 category = fcNormal; 3187 exponent = myexponent - 16383; 3188 significandParts()[0] = mysignificand; 3189 significandParts()[1] = mysignificand2; 3190 if (myexponent==0) // denormal 3191 exponent = -16382; 3192 else 3193 significandParts()[1] |= 0x1000000000000LL; // integer bit 3194 } 3195 } 3196 3197 void 3198 APFloat::initFromDoubleAPInt(const APInt &api) 3199 { 3200 assert(api.getBitWidth()==64); 3201 uint64_t i = *api.getRawData(); 3202 uint64_t myexponent = (i >> 52) & 0x7ff; 3203 uint64_t mysignificand = i & 0xfffffffffffffLL; 3204 3205 initialize(&APFloat::IEEEdouble); 3206 assert(partCount()==1); 3207 3208 sign = static_cast<unsigned int>(i>>63); 3209 if (myexponent==0 && mysignificand==0) { 3210 // exponent, significand meaningless 3211 category = fcZero; 3212 } else if (myexponent==0x7ff && mysignificand==0) { 3213 // exponent, significand meaningless 3214 category = fcInfinity; 3215 } else if (myexponent==0x7ff && mysignificand!=0) { 3216 // exponent meaningless 3217 category = fcNaN; 3218 *significandParts() = mysignificand; 3219 } else { 3220 category = fcNormal; 3221 exponent = myexponent - 1023; 3222 *significandParts() = mysignificand; 3223 if (myexponent==0) // denormal 3224 exponent = -1022; 3225 else 3226 *significandParts() |= 0x10000000000000LL; // integer bit 3227 } 3228 } 3229 3230 void 3231 APFloat::initFromFloatAPInt(const APInt & api) 3232 { 3233 assert(api.getBitWidth()==32); 3234 uint32_t i = (uint32_t)*api.getRawData(); 3235 uint32_t myexponent = (i >> 23) & 0xff; 3236 uint32_t mysignificand = i & 0x7fffff; 3237 3238 initialize(&APFloat::IEEEsingle); 3239 assert(partCount()==1); 3240 3241 sign = i >> 31; 3242 if (myexponent==0 && mysignificand==0) { 3243 // exponent, significand meaningless 3244 category = fcZero; 3245 } else if (myexponent==0xff && mysignificand==0) { 3246 // exponent, significand meaningless 3247 category = fcInfinity; 3248 } else if (myexponent==0xff && mysignificand!=0) { 3249 // sign, exponent, significand meaningless 3250 category = fcNaN; 3251 *significandParts() = mysignificand; 3252 } else { 3253 category = fcNormal; 3254 exponent = myexponent - 127; //bias 3255 *significandParts() = mysignificand; 3256 if (myexponent==0) // denormal 3257 exponent = -126; 3258 else 3259 *significandParts() |= 0x800000; // integer bit 3260 } 3261 } 3262 3263 void 3264 APFloat::initFromHalfAPInt(const APInt & api) 3265 { 3266 assert(api.getBitWidth()==16); 3267 uint32_t i = (uint32_t)*api.getRawData(); 3268 uint32_t myexponent = (i >> 10) & 0x1f; 3269 uint32_t mysignificand = i & 0x3ff; 3270 3271 initialize(&APFloat::IEEEhalf); 3272 assert(partCount()==1); 3273 3274 sign = i >> 15; 3275 if (myexponent==0 && mysignificand==0) { 3276 // exponent, significand meaningless 3277 category = fcZero; 3278 } else if (myexponent==0x1f && mysignificand==0) { 3279 // exponent, significand meaningless 3280 category = fcInfinity; 3281 } else if (myexponent==0x1f && mysignificand!=0) { 3282 // sign, exponent, significand meaningless 3283 category = fcNaN; 3284 *significandParts() = mysignificand; 3285 } else { 3286 category = fcNormal; 3287 exponent = myexponent - 15; //bias 3288 *significandParts() = mysignificand; 3289 if (myexponent==0) // denormal 3290 exponent = -14; 3291 else 3292 *significandParts() |= 0x400; // integer bit 3293 } 3294 } 3295 3296 /// Treat api as containing the bits of a floating point number. Currently 3297 /// we infer the floating point type from the size of the APInt. The 3298 /// isIEEE argument distinguishes between PPC128 and IEEE128 (not meaningful 3299 /// when the size is anything else). 3300 void 3301 APFloat::initFromAPInt(const fltSemantics* Sem, const APInt& api) 3302 { 3303 if (Sem == &IEEEhalf) 3304 return initFromHalfAPInt(api); 3305 if (Sem == &IEEEsingle) 3306 return initFromFloatAPInt(api); 3307 if (Sem == &IEEEdouble) 3308 return initFromDoubleAPInt(api); 3309 if (Sem == &x87DoubleExtended) 3310 return initFromF80LongDoubleAPInt(api); 3311 if (Sem == &IEEEquad) 3312 return initFromQuadrupleAPInt(api); 3313 if (Sem == &PPCDoubleDouble) 3314 return initFromPPCDoubleDoubleAPInt(api); 3315 3316 llvm_unreachable(0); 3317 } 3318 3319 APFloat 3320 APFloat::getAllOnesValue(unsigned BitWidth, bool isIEEE) 3321 { 3322 switch (BitWidth) { 3323 case 16: 3324 return APFloat(IEEEhalf, APInt::getAllOnesValue(BitWidth)); 3325 case 32: 3326 return APFloat(IEEEsingle, APInt::getAllOnesValue(BitWidth)); 3327 case 64: 3328 return APFloat(IEEEdouble, APInt::getAllOnesValue(BitWidth)); 3329 case 80: 3330 return APFloat(x87DoubleExtended, APInt::getAllOnesValue(BitWidth)); 3331 case 128: 3332 if (isIEEE) 3333 return APFloat(IEEEquad, APInt::getAllOnesValue(BitWidth)); 3334 return APFloat(PPCDoubleDouble, APInt::getAllOnesValue(BitWidth)); 3335 default: 3336 llvm_unreachable("Unknown floating bit width"); 3337 } 3338 } 3339 3340 /// Make this number the largest magnitude normal number in the given 3341 /// semantics. 3342 void APFloat::makeLargest(bool Negative) { 3343 // We want (in interchange format): 3344 // sign = {Negative} 3345 // exponent = 1..10 3346 // significand = 1..1 3347 category = fcNormal; 3348 sign = Negative; 3349 exponent = semantics->maxExponent; 3350 3351 // Use memset to set all but the highest integerPart to all ones. 3352 integerPart *significand = significandParts(); 3353 unsigned PartCount = partCount(); 3354 memset(significand, 0xFF, sizeof(integerPart)*(PartCount - 1)); 3355 3356 // Set the high integerPart especially setting all unused top bits for 3357 // internal consistency. 3358 const unsigned NumUnusedHighBits = 3359 PartCount*integerPartWidth - semantics->precision; 3360 significand[PartCount - 1] = ~integerPart(0) >> NumUnusedHighBits; 3361 } 3362 3363 /// Make this number the smallest magnitude denormal number in the given 3364 /// semantics. 3365 void APFloat::makeSmallest(bool Negative) { 3366 // We want (in interchange format): 3367 // sign = {Negative} 3368 // exponent = 0..0 3369 // significand = 0..01 3370 category = fcNormal; 3371 sign = Negative; 3372 exponent = semantics->minExponent; 3373 APInt::tcSet(significandParts(), 1, partCount()); 3374 } 3375 3376 3377 APFloat APFloat::getLargest(const fltSemantics &Sem, bool Negative) { 3378 // We want (in interchange format): 3379 // sign = {Negative} 3380 // exponent = 1..10 3381 // significand = 1..1 3382 APFloat Val(Sem, uninitialized); 3383 Val.makeLargest(Negative); 3384 return Val; 3385 } 3386 3387 APFloat APFloat::getSmallest(const fltSemantics &Sem, bool Negative) { 3388 // We want (in interchange format): 3389 // sign = {Negative} 3390 // exponent = 0..0 3391 // significand = 0..01 3392 APFloat Val(Sem, uninitialized); 3393 Val.makeSmallest(Negative); 3394 return Val; 3395 } 3396 3397 APFloat APFloat::getSmallestNormalized(const fltSemantics &Sem, bool Negative) { 3398 APFloat Val(Sem, uninitialized); 3399 3400 // We want (in interchange format): 3401 // sign = {Negative} 3402 // exponent = 0..0 3403 // significand = 10..0 3404 3405 Val.category = fcNormal; 3406 Val.zeroSignificand(); 3407 Val.sign = Negative; 3408 Val.exponent = Sem.minExponent; 3409 Val.significandParts()[partCountForBits(Sem.precision)-1] |= 3410 (((integerPart) 1) << ((Sem.precision - 1) % integerPartWidth)); 3411 3412 return Val; 3413 } 3414 3415 APFloat::APFloat(const fltSemantics &Sem, const APInt &API) { 3416 initFromAPInt(&Sem, API); 3417 } 3418 3419 APFloat::APFloat(float f) { 3420 initFromAPInt(&IEEEsingle, APInt::floatToBits(f)); 3421 } 3422 3423 APFloat::APFloat(double d) { 3424 initFromAPInt(&IEEEdouble, APInt::doubleToBits(d)); 3425 } 3426 3427 namespace { 3428 void append(SmallVectorImpl<char> &Buffer, StringRef Str) { 3429 Buffer.append(Str.begin(), Str.end()); 3430 } 3431 3432 /// Removes data from the given significand until it is no more 3433 /// precise than is required for the desired precision. 3434 void AdjustToPrecision(APInt &significand, 3435 int &exp, unsigned FormatPrecision) { 3436 unsigned bits = significand.getActiveBits(); 3437 3438 // 196/59 is a very slight overestimate of lg_2(10). 3439 unsigned bitsRequired = (FormatPrecision * 196 + 58) / 59; 3440 3441 if (bits <= bitsRequired) return; 3442 3443 unsigned tensRemovable = (bits - bitsRequired) * 59 / 196; 3444 if (!tensRemovable) return; 3445 3446 exp += tensRemovable; 3447 3448 APInt divisor(significand.getBitWidth(), 1); 3449 APInt powten(significand.getBitWidth(), 10); 3450 while (true) { 3451 if (tensRemovable & 1) 3452 divisor *= powten; 3453 tensRemovable >>= 1; 3454 if (!tensRemovable) break; 3455 powten *= powten; 3456 } 3457 3458 significand = significand.udiv(divisor); 3459 3460 // Truncate the significand down to its active bit count. 3461 significand = significand.trunc(significand.getActiveBits()); 3462 } 3463 3464 3465 void AdjustToPrecision(SmallVectorImpl<char> &buffer, 3466 int &exp, unsigned FormatPrecision) { 3467 unsigned N = buffer.size(); 3468 if (N <= FormatPrecision) return; 3469 3470 // The most significant figures are the last ones in the buffer. 3471 unsigned FirstSignificant = N - FormatPrecision; 3472 3473 // Round. 3474 // FIXME: this probably shouldn't use 'round half up'. 3475 3476 // Rounding down is just a truncation, except we also want to drop 3477 // trailing zeros from the new result. 3478 if (buffer[FirstSignificant - 1] < '5') { 3479 while (FirstSignificant < N && buffer[FirstSignificant] == '0') 3480 FirstSignificant++; 3481 3482 exp += FirstSignificant; 3483 buffer.erase(&buffer[0], &buffer[FirstSignificant]); 3484 return; 3485 } 3486 3487 // Rounding up requires a decimal add-with-carry. If we continue 3488 // the carry, the newly-introduced zeros will just be truncated. 3489 for (unsigned I = FirstSignificant; I != N; ++I) { 3490 if (buffer[I] == '9') { 3491 FirstSignificant++; 3492 } else { 3493 buffer[I]++; 3494 break; 3495 } 3496 } 3497 3498 // If we carried through, we have exactly one digit of precision. 3499 if (FirstSignificant == N) { 3500 exp += FirstSignificant; 3501 buffer.clear(); 3502 buffer.push_back('1'); 3503 return; 3504 } 3505 3506 exp += FirstSignificant; 3507 buffer.erase(&buffer[0], &buffer[FirstSignificant]); 3508 } 3509 } 3510 3511 void APFloat::toString(SmallVectorImpl<char> &Str, 3512 unsigned FormatPrecision, 3513 unsigned FormatMaxPadding) const { 3514 switch (category) { 3515 case fcInfinity: 3516 if (isNegative()) 3517 return append(Str, "-Inf"); 3518 else 3519 return append(Str, "+Inf"); 3520 3521 case fcNaN: return append(Str, "NaN"); 3522 3523 case fcZero: 3524 if (isNegative()) 3525 Str.push_back('-'); 3526 3527 if (!FormatMaxPadding) 3528 append(Str, "0.0E+0"); 3529 else 3530 Str.push_back('0'); 3531 return; 3532 3533 case fcNormal: 3534 break; 3535 } 3536 3537 if (isNegative()) 3538 Str.push_back('-'); 3539 3540 // Decompose the number into an APInt and an exponent. 3541 int exp = exponent - ((int) semantics->precision - 1); 3542 APInt significand(semantics->precision, 3543 makeArrayRef(significandParts(), 3544 partCountForBits(semantics->precision))); 3545 3546 // Set FormatPrecision if zero. We want to do this before we 3547 // truncate trailing zeros, as those are part of the precision. 3548 if (!FormatPrecision) { 3549 // It's an interesting question whether to use the nominal 3550 // precision or the active precision here for denormals. 3551 3552 // FormatPrecision = ceil(significandBits / lg_2(10)) 3553 FormatPrecision = (semantics->precision * 59 + 195) / 196; 3554 } 3555 3556 // Ignore trailing binary zeros. 3557 int trailingZeros = significand.countTrailingZeros(); 3558 exp += trailingZeros; 3559 significand = significand.lshr(trailingZeros); 3560 3561 // Change the exponent from 2^e to 10^e. 3562 if (exp == 0) { 3563 // Nothing to do. 3564 } else if (exp > 0) { 3565 // Just shift left. 3566 significand = significand.zext(semantics->precision + exp); 3567 significand <<= exp; 3568 exp = 0; 3569 } else { /* exp < 0 */ 3570 int texp = -exp; 3571 3572 // We transform this using the identity: 3573 // (N)(2^-e) == (N)(5^e)(10^-e) 3574 // This means we have to multiply N (the significand) by 5^e. 3575 // To avoid overflow, we have to operate on numbers large 3576 // enough to store N * 5^e: 3577 // log2(N * 5^e) == log2(N) + e * log2(5) 3578 // <= semantics->precision + e * 137 / 59 3579 // (log_2(5) ~ 2.321928 < 2.322034 ~ 137/59) 3580 3581 unsigned precision = semantics->precision + (137 * texp + 136) / 59; 3582 3583 // Multiply significand by 5^e. 3584 // N * 5^0101 == N * 5^(1*1) * 5^(0*2) * 5^(1*4) * 5^(0*8) 3585 significand = significand.zext(precision); 3586 APInt five_to_the_i(precision, 5); 3587 while (true) { 3588 if (texp & 1) significand *= five_to_the_i; 3589 3590 texp >>= 1; 3591 if (!texp) break; 3592 five_to_the_i *= five_to_the_i; 3593 } 3594 } 3595 3596 AdjustToPrecision(significand, exp, FormatPrecision); 3597 3598 SmallVector<char, 256> buffer; 3599 3600 // Fill the buffer. 3601 unsigned precision = significand.getBitWidth(); 3602 APInt ten(precision, 10); 3603 APInt digit(precision, 0); 3604 3605 bool inTrail = true; 3606 while (significand != 0) { 3607 // digit <- significand % 10 3608 // significand <- significand / 10 3609 APInt::udivrem(significand, ten, significand, digit); 3610 3611 unsigned d = digit.getZExtValue(); 3612 3613 // Drop trailing zeros. 3614 if (inTrail && !d) exp++; 3615 else { 3616 buffer.push_back((char) ('0' + d)); 3617 inTrail = false; 3618 } 3619 } 3620 3621 assert(!buffer.empty() && "no characters in buffer!"); 3622 3623 // Drop down to FormatPrecision. 3624 // TODO: don't do more precise calculations above than are required. 3625 AdjustToPrecision(buffer, exp, FormatPrecision); 3626 3627 unsigned NDigits = buffer.size(); 3628 3629 // Check whether we should use scientific notation. 3630 bool FormatScientific; 3631 if (!FormatMaxPadding) 3632 FormatScientific = true; 3633 else { 3634 if (exp >= 0) { 3635 // 765e3 --> 765000 3636 // ^^^ 3637 // But we shouldn't make the number look more precise than it is. 3638 FormatScientific = ((unsigned) exp > FormatMaxPadding || 3639 NDigits + (unsigned) exp > FormatPrecision); 3640 } else { 3641 // Power of the most significant digit. 3642 int MSD = exp + (int) (NDigits - 1); 3643 if (MSD >= 0) { 3644 // 765e-2 == 7.65 3645 FormatScientific = false; 3646 } else { 3647 // 765e-5 == 0.00765 3648 // ^ ^^ 3649 FormatScientific = ((unsigned) -MSD) > FormatMaxPadding; 3650 } 3651 } 3652 } 3653 3654 // Scientific formatting is pretty straightforward. 3655 if (FormatScientific) { 3656 exp += (NDigits - 1); 3657 3658 Str.push_back(buffer[NDigits-1]); 3659 Str.push_back('.'); 3660 if (NDigits == 1) 3661 Str.push_back('0'); 3662 else 3663 for (unsigned I = 1; I != NDigits; ++I) 3664 Str.push_back(buffer[NDigits-1-I]); 3665 Str.push_back('E'); 3666 3667 Str.push_back(exp >= 0 ? '+' : '-'); 3668 if (exp < 0) exp = -exp; 3669 SmallVector<char, 6> expbuf; 3670 do { 3671 expbuf.push_back((char) ('0' + (exp % 10))); 3672 exp /= 10; 3673 } while (exp); 3674 for (unsigned I = 0, E = expbuf.size(); I != E; ++I) 3675 Str.push_back(expbuf[E-1-I]); 3676 return; 3677 } 3678 3679 // Non-scientific, positive exponents. 3680 if (exp >= 0) { 3681 for (unsigned I = 0; I != NDigits; ++I) 3682 Str.push_back(buffer[NDigits-1-I]); 3683 for (unsigned I = 0; I != (unsigned) exp; ++I) 3684 Str.push_back('0'); 3685 return; 3686 } 3687 3688 // Non-scientific, negative exponents. 3689 3690 // The number of digits to the left of the decimal point. 3691 int NWholeDigits = exp + (int) NDigits; 3692 3693 unsigned I = 0; 3694 if (NWholeDigits > 0) { 3695 for (; I != (unsigned) NWholeDigits; ++I) 3696 Str.push_back(buffer[NDigits-I-1]); 3697 Str.push_back('.'); 3698 } else { 3699 unsigned NZeros = 1 + (unsigned) -NWholeDigits; 3700 3701 Str.push_back('0'); 3702 Str.push_back('.'); 3703 for (unsigned Z = 1; Z != NZeros; ++Z) 3704 Str.push_back('0'); 3705 } 3706 3707 for (; I != NDigits; ++I) 3708 Str.push_back(buffer[NDigits-I-1]); 3709 } 3710 3711 bool APFloat::getExactInverse(APFloat *inv) const { 3712 // Special floats and denormals have no exact inverse. 3713 if (!isFiniteNonZero()) 3714 return false; 3715 3716 // Check that the number is a power of two by making sure that only the 3717 // integer bit is set in the significand. 3718 if (significandLSB() != semantics->precision - 1) 3719 return false; 3720 3721 // Get the inverse. 3722 APFloat reciprocal(*semantics, 1ULL); 3723 if (reciprocal.divide(*this, rmNearestTiesToEven) != opOK) 3724 return false; 3725 3726 // Avoid multiplication with a denormal, it is not safe on all platforms and 3727 // may be slower than a normal division. 3728 if (reciprocal.isDenormal()) 3729 return false; 3730 3731 assert(reciprocal.isFiniteNonZero() && 3732 reciprocal.significandLSB() == reciprocal.semantics->precision - 1); 3733 3734 if (inv) 3735 *inv = reciprocal; 3736 3737 return true; 3738 } 3739 3740 bool APFloat::isSignaling() const { 3741 if (!isNaN()) 3742 return false; 3743 3744 // IEEE-754R 2008 6.2.1: A signaling NaN bit string should be encoded with the 3745 // first bit of the trailing significand being 0. 3746 return !APInt::tcExtractBit(significandParts(), semantics->precision - 2); 3747 } 3748 3749 /// IEEE-754R 2008 5.3.1: nextUp/nextDown. 3750 /// 3751 /// *NOTE* since nextDown(x) = -nextUp(-x), we only implement nextUp with 3752 /// appropriate sign switching before/after the computation. 3753 APFloat::opStatus APFloat::next(bool nextDown) { 3754 // If we are performing nextDown, swap sign so we have -x. 3755 if (nextDown) 3756 changeSign(); 3757 3758 // Compute nextUp(x) 3759 opStatus result = opOK; 3760 3761 // Handle each float category separately. 3762 switch (category) { 3763 case fcInfinity: 3764 // nextUp(+inf) = +inf 3765 if (!isNegative()) 3766 break; 3767 // nextUp(-inf) = -getLargest() 3768 makeLargest(true); 3769 break; 3770 case fcNaN: 3771 // IEEE-754R 2008 6.2 Par 2: nextUp(sNaN) = qNaN. Set Invalid flag. 3772 // IEEE-754R 2008 6.2: nextUp(qNaN) = qNaN. Must be identity so we do not 3773 // change the payload. 3774 if (isSignaling()) { 3775 result = opInvalidOp; 3776 // For consistency, propogate the sign of the sNaN to the qNaN. 3777 makeNaN(false, isNegative(), 0); 3778 } 3779 break; 3780 case fcZero: 3781 // nextUp(pm 0) = +getSmallest() 3782 makeSmallest(false); 3783 break; 3784 case fcNormal: 3785 // nextUp(-getSmallest()) = -0 3786 if (isSmallest() && isNegative()) { 3787 APInt::tcSet(significandParts(), 0, partCount()); 3788 category = fcZero; 3789 exponent = 0; 3790 break; 3791 } 3792 3793 // nextUp(getLargest()) == INFINITY 3794 if (isLargest() && !isNegative()) { 3795 APInt::tcSet(significandParts(), 0, partCount()); 3796 category = fcInfinity; 3797 exponent = semantics->maxExponent + 1; 3798 break; 3799 } 3800 3801 // nextUp(normal) == normal + inc. 3802 if (isNegative()) { 3803 // If we are negative, we need to decrement the significand. 3804 3805 // We only cross a binade boundary that requires adjusting the exponent 3806 // if: 3807 // 1. exponent != semantics->minExponent. This implies we are not in the 3808 // smallest binade or are dealing with denormals. 3809 // 2. Our significand excluding the integral bit is all zeros. 3810 bool WillCrossBinadeBoundary = 3811 exponent != semantics->minExponent && isSignificandAllZeros(); 3812 3813 // Decrement the significand. 3814 // 3815 // We always do this since: 3816 // 1. If we are dealing with a non binade decrement, by definition we 3817 // just decrement the significand. 3818 // 2. If we are dealing with a normal -> normal binade decrement, since 3819 // we have an explicit integral bit the fact that all bits but the 3820 // integral bit are zero implies that subtracting one will yield a 3821 // significand with 0 integral bit and 1 in all other spots. Thus we 3822 // must just adjust the exponent and set the integral bit to 1. 3823 // 3. If we are dealing with a normal -> denormal binade decrement, 3824 // since we set the integral bit to 0 when we represent denormals, we 3825 // just decrement the significand. 3826 integerPart *Parts = significandParts(); 3827 APInt::tcDecrement(Parts, partCount()); 3828 3829 if (WillCrossBinadeBoundary) { 3830 // Our result is a normal number. Do the following: 3831 // 1. Set the integral bit to 1. 3832 // 2. Decrement the exponent. 3833 APInt::tcSetBit(Parts, semantics->precision - 1); 3834 exponent--; 3835 } 3836 } else { 3837 // If we are positive, we need to increment the significand. 3838 3839 // We only cross a binade boundary that requires adjusting the exponent if 3840 // the input is not a denormal and all of said input's significand bits 3841 // are set. If all of said conditions are true: clear the significand, set 3842 // the integral bit to 1, and increment the exponent. If we have a 3843 // denormal always increment since moving denormals and the numbers in the 3844 // smallest normal binade have the same exponent in our representation. 3845 bool WillCrossBinadeBoundary = !isDenormal() && isSignificandAllOnes(); 3846 3847 if (WillCrossBinadeBoundary) { 3848 integerPart *Parts = significandParts(); 3849 APInt::tcSet(Parts, 0, partCount()); 3850 APInt::tcSetBit(Parts, semantics->precision - 1); 3851 assert(exponent != semantics->maxExponent && 3852 "We can not increment an exponent beyond the maxExponent allowed" 3853 " by the given floating point semantics."); 3854 exponent++; 3855 } else { 3856 incrementSignificand(); 3857 } 3858 } 3859 break; 3860 } 3861 3862 // If we are performing nextDown, swap sign so we have -nextUp(-x) 3863 if (nextDown) 3864 changeSign(); 3865 3866 return result; 3867 } 3868 3869 void 3870 APFloat::makeInf(bool Negative) { 3871 category = fcInfinity; 3872 sign = Negative; 3873 exponent = semantics->maxExponent + 1; 3874 APInt::tcSet(significandParts(), 0, partCount()); 3875 } 3876 3877 void 3878 APFloat::makeZero(bool Negative) { 3879 category = fcZero; 3880 sign = Negative; 3881 exponent = semantics->minExponent-1; 3882 APInt::tcSet(significandParts(), 0, partCount()); 3883 } 3884