Coverage Report

Created: 2022-05-17 06:19

/Users/buildslave/jenkins/workspace/coverage/llvm-project/libcxx/src/ryu/d2s.cpp
Line
Count
Source (jump to first uncovered line)
1
//===----------------------------------------------------------------------===//
2
//
3
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
4
// See https://llvm.org/LICENSE.txt for license information.
5
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
6
//
7
//===----------------------------------------------------------------------===//
8
9
// Copyright (c) Microsoft Corporation.
10
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
11
12
// Copyright 2018 Ulf Adams
13
// Copyright (c) Microsoft Corporation. All rights reserved.
14
15
// Boost Software License - Version 1.0 - August 17th, 2003
16
17
// Permission is hereby granted, free of charge, to any person or organization
18
// obtaining a copy of the software and accompanying documentation covered by
19
// this license (the "Software") to use, reproduce, display, distribute,
20
// execute, and transmit the Software, and to prepare derivative works of the
21
// Software, and to permit third-parties to whom the Software is furnished to
22
// do so, all subject to the following:
23
24
// The copyright notices in the Software and this entire statement, including
25
// the above license grant, this restriction and the following disclaimer,
26
// must be included in all copies of the Software, in whole or in part, and
27
// all derivative works of the Software, unless such copies or derivative
28
// works are solely in the form of machine-executable object code generated by
29
// a source language processor.
30
31
// THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
32
// IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
33
// FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
34
// SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
35
// FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
36
// ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
37
// DEALINGS IN THE SOFTWARE.
38
39
// Avoid formatting to keep the changes with the original code minimal.
40
// clang-format off
41
42
#include <__assert>
43
#include <__config>
44
#include <charconv>
45
46
#include "include/ryu/common.h"
47
#include "include/ryu/d2fixed.h"
48
#include "include/ryu/d2s.h"
49
#include "include/ryu/d2s_full_table.h"
50
#include "include/ryu/d2s_intrinsics.h"
51
#include "include/ryu/digit_table.h"
52
#include "include/ryu/ryu.h"
53
54
_LIBCPP_BEGIN_NAMESPACE_STD
55
56
// We need a 64x128-bit multiplication and a subsequent 128-bit shift.
57
// Multiplication:
58
//   The 64-bit factor is variable and passed in, the 128-bit factor comes
59
//   from a lookup table. We know that the 64-bit factor only has 55
60
//   significant bits (i.e., the 9 topmost bits are zeros). The 128-bit
61
//   factor only has 124 significant bits (i.e., the 4 topmost bits are
62
//   zeros).
63
// Shift:
64
//   In principle, the multiplication result requires 55 + 124 = 179 bits to
65
//   represent. However, we then shift this value to the right by __j, which is
66
//   at least __j >= 115, so the result is guaranteed to fit into 179 - 115 = 64
67
//   bits. This means that we only need the topmost 64 significant bits of
68
//   the 64x128-bit multiplication.
69
//
70
// There are several ways to do this:
71
// 1. Best case: the compiler exposes a 128-bit type.
72
//    We perform two 64x64-bit multiplications, add the higher 64 bits of the
73
//    lower result to the higher result, and shift by __j - 64 bits.
74
//
75
//    We explicitly cast from 64-bit to 128-bit, so the compiler can tell
76
//    that these are only 64-bit inputs, and can map these to the best
77
//    possible sequence of assembly instructions.
78
//    x64 machines happen to have matching assembly instructions for
79
//    64x64-bit multiplications and 128-bit shifts.
80
//
81
// 2. Second best case: the compiler exposes intrinsics for the x64 assembly
82
//    instructions mentioned in 1.
83
//
84
// 3. We only have 64x64 bit instructions that return the lower 64 bits of
85
//    the result, i.e., we have to use plain C.
86
//    Our inputs are less than the full width, so we have three options:
87
//    a. Ignore this fact and just implement the intrinsics manually.
88
//    b. Split both into 31-bit pieces, which guarantees no internal overflow,
89
//       but requires extra work upfront (unless we change the lookup table).
90
//    c. Split only the first factor into 31-bit pieces, which also guarantees
91
//       no internal overflow, but requires extra work since the intermediate
92
//       results are not perfectly aligned.
93
#ifdef _LIBCPP_INTRINSIC128
94
95
0
[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline uint64_t __mulShift(const uint64_t __m, const uint64_t* const __mul, const int32_t __j) {
96
  // __m is maximum 55 bits
97
0
  uint64_t __high1;                                               // 128
98
0
  const uint64_t __low1 = __ryu_umul128(__m, __mul[1], &__high1); // 64
99
0
  uint64_t __high0;                                               // 64
100
0
  (void) __ryu_umul128(__m, __mul[0], &__high0);                  // 0
101
0
  const uint64_t __sum = __high0 + __low1;
102
0
  if (__sum < __high0) {
103
0
    ++__high1; // overflow into __high1
104
0
  }
105
0
  return __ryu_shiftright128(__sum, __high1, static_cast<uint32_t>(__j - 64));
106
0
}
107
108
[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline uint64_t __mulShiftAll(const uint64_t __m, const uint64_t* const __mul, const int32_t __j,
109
0
  uint64_t* const __vp, uint64_t* const __vm, const uint32_t __mmShift) {
110
0
  *__vp = __mulShift(4 * __m + 2, __mul, __j);
111
0
  *__vm = __mulShift(4 * __m - 1 - __mmShift, __mul, __j);
112
0
  return __mulShift(4 * __m, __mul, __j);
113
0
}
114
115
#else // ^^^ intrinsics available ^^^ / vvv intrinsics unavailable vvv
116
117
[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline _LIBCPP_ALWAYS_INLINE uint64_t __mulShiftAll(uint64_t __m, const uint64_t* const __mul, const int32_t __j,
118
  uint64_t* const __vp, uint64_t* const __vm, const uint32_t __mmShift) { // TRANSITION, VSO-634761
119
  __m <<= 1;
120
  // __m is maximum 55 bits
121
  uint64_t __tmp;
122
  const uint64_t __lo = __ryu_umul128(__m, __mul[0], &__tmp);
123
  uint64_t __hi;
124
  const uint64_t __mid = __tmp + __ryu_umul128(__m, __mul[1], &__hi);
125
  __hi += __mid < __tmp; // overflow into __hi
126
127
  const uint64_t __lo2 = __lo + __mul[0];
128
  const uint64_t __mid2 = __mid + __mul[1] + (__lo2 < __lo);
129
  const uint64_t __hi2 = __hi + (__mid2 < __mid);
130
  *__vp = __ryu_shiftright128(__mid2, __hi2, static_cast<uint32_t>(__j - 64 - 1));
131
132
  if (__mmShift == 1) {
133
    const uint64_t __lo3 = __lo - __mul[0];
134
    const uint64_t __mid3 = __mid - __mul[1] - (__lo3 > __lo);
135
    const uint64_t __hi3 = __hi - (__mid3 > __mid);
136
    *__vm = __ryu_shiftright128(__mid3, __hi3, static_cast<uint32_t>(__j - 64 - 1));
137
  } else {
138
    const uint64_t __lo3 = __lo + __lo;
139
    const uint64_t __mid3 = __mid + __mid + (__lo3 < __lo);
140
    const uint64_t __hi3 = __hi + __hi + (__mid3 < __mid);
141
    const uint64_t __lo4 = __lo3 - __mul[0];
142
    const uint64_t __mid4 = __mid3 - __mul[1] - (__lo4 > __lo3);
143
    const uint64_t __hi4 = __hi3 - (__mid4 > __mid3);
144
    *__vm = __ryu_shiftright128(__mid4, __hi4, static_cast<uint32_t>(__j - 64));
145
  }
146
147
  return __ryu_shiftright128(__mid, __hi, static_cast<uint32_t>(__j - 64 - 1));
148
}
149
150
#endif // ^^^ intrinsics unavailable ^^^
151
152
0
[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline uint32_t __decimalLength17(const uint64_t __v) {
153
  // This is slightly faster than a loop.
154
  // The average output length is 16.38 digits, so we check high-to-low.
155
  // Function precondition: __v is not an 18, 19, or 20-digit number.
156
  // (17 digits are sufficient for round-tripping.)
157
0
  _LIBCPP_ASSERT(__v < 100000000000000000u, "");
158
0
  if (__v >= 10000000000000000u) { return 17; }
159
0
  if (__v >= 1000000000000000u) { return 16; }
160
0
  if (__v >= 100000000000000u) { return 15; }
161
0
  if (__v >= 10000000000000u) { return 14; }
162
0
  if (__v >= 1000000000000u) { return 13; }
163
0
  if (__v >= 100000000000u) { return 12; }
164
0
  if (__v >= 10000000000u) { return 11; }
165
0
  if (__v >= 1000000000u) { return 10; }
166
0
  if (__v >= 100000000u) { return 9; }
167
0
  if (__v >= 10000000u) { return 8; }
168
0
  if (__v >= 1000000u) { return 7; }
169
0
  if (__v >= 100000u) { return 6; }
170
0
  if (__v >= 10000u) { return 5; }
171
0
  if (__v >= 1000u) { return 4; }
172
0
  if (__v >= 100u) { return 3; }
173
0
  if (__v >= 10u) { return 2; }
174
0
  return 1;
175
0
}
176
177
// A floating decimal representing m * 10^e.
178
struct __floating_decimal_64 {
179
  uint64_t __mantissa;
180
  int32_t __exponent;
181
};
182
183
0
[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline __floating_decimal_64 __d2d(const uint64_t __ieeeMantissa, const uint32_t __ieeeExponent) {
184
0
  int32_t __e2;
185
0
  uint64_t __m2;
186
0
  if (__ieeeExponent == 0) {
187
    // We subtract 2 so that the bounds computation has 2 additional bits.
188
0
    __e2 = 1 - __DOUBLE_BIAS - __DOUBLE_MANTISSA_BITS - 2;
189
0
    __m2 = __ieeeMantissa;
190
0
  } else {
191
0
    __e2 = static_cast<int32_t>(__ieeeExponent) - __DOUBLE_BIAS - __DOUBLE_MANTISSA_BITS - 2;
192
0
    __m2 = (1ull << __DOUBLE_MANTISSA_BITS) | __ieeeMantissa;
193
0
  }
194
0
  const bool __even = (__m2 & 1) == 0;
195
0
  const bool __acceptBounds = __even;
196
197
  // Step 2: Determine the interval of valid decimal representations.
198
0
  const uint64_t __mv = 4 * __m2;
199
  // Implicit bool -> int conversion. True is 1, false is 0.
200
0
  const uint32_t __mmShift = __ieeeMantissa != 0 || __ieeeExponent <= 1;
201
  // We would compute __mp and __mm like this:
202
  // uint64_t __mp = 4 * __m2 + 2;
203
  // uint64_t __mm = __mv - 1 - __mmShift;
204
205
  // Step 3: Convert to a decimal power base using 128-bit arithmetic.
206
0
  uint64_t __vr, __vp, __vm;
207
0
  int32_t __e10;
208
0
  bool __vmIsTrailingZeros = false;
209
0
  bool __vrIsTrailingZeros = false;
210
0
  if (__e2 >= 0) {
211
    // I tried special-casing __q == 0, but there was no effect on performance.
212
    // This expression is slightly faster than max(0, __log10Pow2(__e2) - 1).
213
0
    const uint32_t __q = __log10Pow2(__e2) - (__e2 > 3);
214
0
    __e10 = static_cast<int32_t>(__q);
215
0
    const int32_t __k = __DOUBLE_POW5_INV_BITCOUNT + __pow5bits(static_cast<int32_t>(__q)) - 1;
216
0
    const int32_t __i = -__e2 + static_cast<int32_t>(__q) + __k;
217
0
    __vr = __mulShiftAll(__m2, __DOUBLE_POW5_INV_SPLIT[__q], __i, &__vp, &__vm, __mmShift);
218
0
    if (__q <= 21) {
219
      // This should use __q <= 22, but I think 21 is also safe. Smaller values
220
      // may still be safe, but it's more difficult to reason about them.
221
      // Only one of __mp, __mv, and __mm can be a multiple of 5, if any.
222
0
      const uint32_t __mvMod5 = static_cast<uint32_t>(__mv) - 5 * static_cast<uint32_t>(__div5(__mv));
223
0
      if (__mvMod5 == 0) {
224
0
        __vrIsTrailingZeros = __multipleOfPowerOf5(__mv, __q);
225
0
      } else if (__acceptBounds) {
226
        // Same as min(__e2 + (~__mm & 1), __pow5Factor(__mm)) >= __q
227
        // <=> __e2 + (~__mm & 1) >= __q && __pow5Factor(__mm) >= __q
228
        // <=> true && __pow5Factor(__mm) >= __q, since __e2 >= __q.
229
0
        __vmIsTrailingZeros = __multipleOfPowerOf5(__mv - 1 - __mmShift, __q);
230
0
      } else {
231
        // Same as min(__e2 + 1, __pow5Factor(__mp)) >= __q.
232
0
        __vp -= __multipleOfPowerOf5(__mv + 2, __q);
233
0
      }
234
0
    }
235
0
  } else {
236
    // This expression is slightly faster than max(0, __log10Pow5(-__e2) - 1).
237
0
    const uint32_t __q = __log10Pow5(-__e2) - (-__e2 > 1);
238
0
    __e10 = static_cast<int32_t>(__q) + __e2;
239
0
    const int32_t __i = -__e2 - static_cast<int32_t>(__q);
240
0
    const int32_t __k = __pow5bits(__i) - __DOUBLE_POW5_BITCOUNT;
241
0
    const int32_t __j = static_cast<int32_t>(__q) - __k;
242
0
    __vr = __mulShiftAll(__m2, __DOUBLE_POW5_SPLIT[__i], __j, &__vp, &__vm, __mmShift);
243
0
    if (__q <= 1) {
244
      // {__vr,__vp,__vm} is trailing zeros if {__mv,__mp,__mm} has at least __q trailing 0 bits.
245
      // __mv = 4 * __m2, so it always has at least two trailing 0 bits.
246
0
      __vrIsTrailingZeros = true;
247
0
      if (__acceptBounds) {
248
        // __mm = __mv - 1 - __mmShift, so it has 1 trailing 0 bit iff __mmShift == 1.
249
0
        __vmIsTrailingZeros = __mmShift == 1;
250
0
      } else {
251
        // __mp = __mv + 2, so it always has at least one trailing 0 bit.
252
0
        --__vp;
253
0
      }
254
0
    } else if (__q < 63) { // TRANSITION(ulfjack): Use a tighter bound here.
255
      // We need to compute min(ntz(__mv), __pow5Factor(__mv) - __e2) >= __q - 1
256
      // <=> ntz(__mv) >= __q - 1 && __pow5Factor(__mv) - __e2 >= __q - 1
257
      // <=> ntz(__mv) >= __q - 1 (__e2 is negative and -__e2 >= __q)
258
      // <=> (__mv & ((1 << (__q - 1)) - 1)) == 0
259
      // We also need to make sure that the left shift does not overflow.
260
0
      __vrIsTrailingZeros = __multipleOfPowerOf2(__mv, __q - 1);
261
0
    }
262
0
  }
263
264
  // Step 4: Find the shortest decimal representation in the interval of valid representations.
265
0
  int32_t __removed = 0;
266
0
  uint8_t __lastRemovedDigit = 0;
267
0
  uint64_t _Output;
268
  // On average, we remove ~2 digits.
269
0
  if (__vmIsTrailingZeros || __vrIsTrailingZeros) {
270
    // General case, which happens rarely (~0.7%).
271
0
    for (;;) {
272
0
      const uint64_t __vpDiv10 = __div10(__vp);
273
0
      const uint64_t __vmDiv10 = __div10(__vm);
274
0
      if (__vpDiv10 <= __vmDiv10) {
275
0
        break;
276
0
      }
277
0
      const uint32_t __vmMod10 = static_cast<uint32_t>(__vm) - 10 * static_cast<uint32_t>(__vmDiv10);
278
0
      const uint64_t __vrDiv10 = __div10(__vr);
279
0
      const uint32_t __vrMod10 = static_cast<uint32_t>(__vr) - 10 * static_cast<uint32_t>(__vrDiv10);
280
0
      __vmIsTrailingZeros &= __vmMod10 == 0;
281
0
      __vrIsTrailingZeros &= __lastRemovedDigit == 0;
282
0
      __lastRemovedDigit = static_cast<uint8_t>(__vrMod10);
283
0
      __vr = __vrDiv10;
284
0
      __vp = __vpDiv10;
285
0
      __vm = __vmDiv10;
286
0
      ++__removed;
287
0
    }
288
0
    if (__vmIsTrailingZeros) {
289
0
      for (;;) {
290
0
        const uint64_t __vmDiv10 = __div10(__vm);
291
0
        const uint32_t __vmMod10 = static_cast<uint32_t>(__vm) - 10 * static_cast<uint32_t>(__vmDiv10);
292
0
        if (__vmMod10 != 0) {
293
0
          break;
294
0
        }
295
0
        const uint64_t __vpDiv10 = __div10(__vp);
296
0
        const uint64_t __vrDiv10 = __div10(__vr);
297
0
        const uint32_t __vrMod10 = static_cast<uint32_t>(__vr) - 10 * static_cast<uint32_t>(__vrDiv10);
298
0
        __vrIsTrailingZeros &= __lastRemovedDigit == 0;
299
0
        __lastRemovedDigit = static_cast<uint8_t>(__vrMod10);
300
0
        __vr = __vrDiv10;
301
0
        __vp = __vpDiv10;
302
0
        __vm = __vmDiv10;
303
0
        ++__removed;
304
0
      }
305
0
    }
306
0
    if (__vrIsTrailingZeros && __lastRemovedDigit == 5 && __vr % 2 == 0) {
307
      // Round even if the exact number is .....50..0.
308
0
      __lastRemovedDigit = 4;
309
0
    }
310
    // We need to take __vr + 1 if __vr is outside bounds or we need to round up.
311
0
    _Output = __vr + ((__vr == __vm && (!__acceptBounds || !__vmIsTrailingZeros)) || __lastRemovedDigit >= 5);
312
0
  } else {
313
    // Specialized for the common case (~99.3%). Percentages below are relative to this.
314
0
    bool __roundUp = false;
315
0
    const uint64_t __vpDiv100 = __div100(__vp);
316
0
    const uint64_t __vmDiv100 = __div100(__vm);
317
0
    if (__vpDiv100 > __vmDiv100) { // Optimization: remove two digits at a time (~86.2%).
318
0
      const uint64_t __vrDiv100 = __div100(__vr);
319
0
      const uint32_t __vrMod100 = static_cast<uint32_t>(__vr) - 100 * static_cast<uint32_t>(__vrDiv100);
320
0
      __roundUp = __vrMod100 >= 50;
321
0
      __vr = __vrDiv100;
322
0
      __vp = __vpDiv100;
323
0
      __vm = __vmDiv100;
324
0
      __removed += 2;
325
0
    }
326
    // Loop iterations below (approximately), without optimization above:
327
    // 0: 0.03%, 1: 13.8%, 2: 70.6%, 3: 14.0%, 4: 1.40%, 5: 0.14%, 6+: 0.02%
328
    // Loop iterations below (approximately), with optimization above:
329
    // 0: 70.6%, 1: 27.8%, 2: 1.40%, 3: 0.14%, 4+: 0.02%
330
0
    for (;;) {
331
0
      const uint64_t __vpDiv10 = __div10(__vp);
332
0
      const uint64_t __vmDiv10 = __div10(__vm);
333
0
      if (__vpDiv10 <= __vmDiv10) {
334
0
        break;
335
0
      }
336
0
      const uint64_t __vrDiv10 = __div10(__vr);
337
0
      const uint32_t __vrMod10 = static_cast<uint32_t>(__vr) - 10 * static_cast<uint32_t>(__vrDiv10);
338
0
      __roundUp = __vrMod10 >= 5;
339
0
      __vr = __vrDiv10;
340
0
      __vp = __vpDiv10;
341
0
      __vm = __vmDiv10;
342
0
      ++__removed;
343
0
    }
344
    // We need to take __vr + 1 if __vr is outside bounds or we need to round up.
345
0
    _Output = __vr + (__vr == __vm || __roundUp);
346
0
  }
347
0
  const int32_t __exp = __e10 + __removed;
348
349
0
  __floating_decimal_64 __fd;
350
0
  __fd.__exponent = __exp;
351
0
  __fd.__mantissa = _Output;
352
0
  return __fd;
353
0
}
354
355
[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline to_chars_result __to_chars(char* const _First, char* const _Last, const __floating_decimal_64 __v,
356
0
  chars_format _Fmt, const double __f) {
357
  // Step 5: Print the decimal representation.
358
0
  uint64_t _Output = __v.__mantissa;
359
0
  int32_t _Ryu_exponent = __v.__exponent;
360
0
  const uint32_t __olength = __decimalLength17(_Output);
361
0
  int32_t _Scientific_exponent = _Ryu_exponent + static_cast<int32_t>(__olength) - 1;
362
363
0
  if (_Fmt == chars_format{}) {
364
0
    int32_t _Lower;
365
0
    int32_t _Upper;
366
367
0
    if (__olength == 1) {
368
      // Value | Fixed   | Scientific
369
      // 1e-3  | "0.001" | "1e-03"
370
      // 1e4   | "10000" | "1e+04"
371
0
      _Lower = -3;
372
0
      _Upper = 4;
373
0
    } else {
374
      // Value   | Fixed       | Scientific
375
      // 1234e-7 | "0.0001234" | "1.234e-04"
376
      // 1234e5  | "123400000" | "1.234e+08"
377
0
      _Lower = -static_cast<int32_t>(__olength + 3);
378
0
      _Upper = 5;
379
0
    }
380
381
0
    if (_Lower <= _Ryu_exponent && _Ryu_exponent <= _Upper) {
382
0
      _Fmt = chars_format::fixed;
383
0
    } else {
384
0
      _Fmt = chars_format::scientific;
385
0
    }
386
0
  } else if (_Fmt == chars_format::general) {
387
    // C11 7.21.6.1 "The fprintf function"/8:
388
    // "Let P equal [...] 6 if the precision is omitted [...].
389
    // Then, if a conversion with style E would have an exponent of X:
390
    // - if P > X >= -4, the conversion is with style f [...].
391
    // - otherwise, the conversion is with style e [...]."
392
0
    if (-4 <= _Scientific_exponent && _Scientific_exponent < 6) {
393
0
      _Fmt = chars_format::fixed;
394
0
    } else {
395
0
      _Fmt = chars_format::scientific;
396
0
    }
397
0
  }
398
399
0
  if (_Fmt == chars_format::fixed) {
400
    // Example: _Output == 1729, __olength == 4
401
402
    // _Ryu_exponent | Printed  | _Whole_digits | _Total_fixed_length  | Notes
403
    // --------------|----------|---------------|----------------------|---------------------------------------
404
    //             2 | 172900   |  6            | _Whole_digits        | Ryu can't be used for printing
405
    //             1 | 17290    |  5            | (sometimes adjusted) | when the trimmed digits are nonzero.
406
    // --------------|----------|---------------|----------------------|---------------------------------------
407
    //             0 | 1729     |  4            | _Whole_digits        | Unified length cases.
408
    // --------------|----------|---------------|----------------------|---------------------------------------
409
    //            -1 | 172.9    |  3            | __olength + 1        | This case can't happen for
410
    //            -2 | 17.29    |  2            |                      | __olength == 1, but no additional
411
    //            -3 | 1.729    |  1            |                      | code is needed to avoid it.
412
    // --------------|----------|---------------|----------------------|---------------------------------------
413
    //            -4 | 0.1729   |  0            | 2 - _Ryu_exponent    | C11 7.21.6.1 "The fprintf function"/8:
414
    //            -5 | 0.01729  | -1            |                      | "If a decimal-point character appears,
415
    //            -6 | 0.001729 | -2            |                      | at least one digit appears before it."
416
417
0
    const int32_t _Whole_digits = static_cast<int32_t>(__olength) + _Ryu_exponent;
418
419
0
    uint32_t _Total_fixed_length;
420
0
    if (_Ryu_exponent >= 0) { // cases "172900" and "1729"
421
0
      _Total_fixed_length = static_cast<uint32_t>(_Whole_digits);
422
0
      if (_Output == 1) {
423
        // Rounding can affect the number of digits.
424
        // For example, 1e23 is exactly "99999999999999991611392" which is 23 digits instead of 24.
425
        // We can use a lookup table to detect this and adjust the total length.
426
0
        static constexpr uint8_t _Adjustment[309] = {
427
0
          0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,0,0,0,1,1,0,1,0,1,1,1,0,1,1,1,0,0,0,0,0,
428
0
          1,1,0,0,1,0,1,1,1,0,0,0,0,1,1,1,1,0,0,0,1,1,1,1,0,0,0,1,1,1,1,0,1,0,1,0,1,1,0,0,0,0,1,1,1,
429
0
          1,0,0,0,0,0,0,0,1,1,0,1,1,0,0,1,0,1,0,1,0,1,1,0,0,0,0,0,1,1,1,0,0,1,1,1,1,1,0,1,0,1,1,0,1,
430
0
          1,0,0,0,0,0,0,0,0,0,1,1,1,0,0,1,0,0,1,0,0,1,1,1,1,0,0,1,1,0,1,1,0,1,1,0,1,0,0,0,1,0,0,0,1,
431
0
          0,1,0,1,0,1,1,1,0,0,0,0,0,0,1,1,1,1,0,0,1,0,1,1,1,0,0,0,1,0,1,1,1,1,1,1,0,1,0,1,1,0,0,0,1,
432
0
          1,1,0,1,1,0,0,0,1,0,0,0,1,0,1,0,0,0,0,0,0,0,1,0,1,1,0,0,1,1,1,0,0,0,1,0,1,0,0,0,0,0,1,1,0,
433
0
          0,1,0,1,1,1,0,0,1,0,0,0,0,1,0,1,0,0,0,0,0,1,0,1,0,1,1,0,1,0,0,0,0,0,1,1,0,1,0 };
434
0
        _Total_fixed_length -= _Adjustment[_Ryu_exponent];
435
        // _Whole_digits doesn't need to be adjusted because these cases won't refer to it later.
436
0
      }
437
0
    } else if (_Whole_digits > 0) { // case "17.29"
438
0
      _Total_fixed_length = __olength + 1;
439
0
    } else { // case "0.001729"
440
0
      _Total_fixed_length = static_cast<uint32_t>(2 - _Ryu_exponent);
441
0
    }
442
443
0
    if (_Last - _First < static_cast<ptrdiff_t>(_Total_fixed_length)) {
444
0
      return { _Last, errc::value_too_large };
445
0
    }
446
447
0
    char* _Mid;
448
0
    if (_Ryu_exponent > 0) { // case "172900"
449
0
      bool _Can_use_ryu;
450
451
0
      if (_Ryu_exponent > 22) { // 10^22 is the largest power of 10 that's exactly representable as a double.
452
0
        _Can_use_ryu = false;
453
0
      } else {
454
        // Ryu generated X: __v.__mantissa * 10^_Ryu_exponent
455
        // __v.__mantissa == 2^_Trailing_zero_bits * (__v.__mantissa >> _Trailing_zero_bits)
456
        // 10^_Ryu_exponent == 2^_Ryu_exponent * 5^_Ryu_exponent
457
458
        // _Trailing_zero_bits is [0, 56] (aside: because 2^56 is the largest power of 2
459
        // with 17 decimal digits, which is double's round-trip limit.)
460
        // _Ryu_exponent is [1, 22].
461
        // Normalization adds [2, 52] (aside: at least 2 because the pre-normalized mantissa is at least 5).
462
        // This adds up to [3, 130], which is well below double's maximum binary exponent 1023.
463
464
        // Therefore, we just need to consider (__v.__mantissa >> _Trailing_zero_bits) * 5^_Ryu_exponent.
465
466
        // If that product would exceed 53 bits, then X can't be exactly represented as a double.
467
        // (That's not a problem for round-tripping, because X is close enough to the original double,
468
        // but X isn't mathematically equal to the original double.) This requires a high-precision fallback.
469
470
        // If the product is 53 bits or smaller, then X can be exactly represented as a double (and we don't
471
        // need to re-synthesize it; the original double must have been X, because Ryu wouldn't produce the
472
        // same output for two different doubles X and Y). This allows Ryu's output to be used (zero-filled).
473
474
        // (2^53 - 1) / 5^0 (for indexing), (2^53 - 1) / 5^1, ..., (2^53 - 1) / 5^22
475
0
        static constexpr uint64_t _Max_shifted_mantissa[23] = {
476
0
          9007199254740991u, 1801439850948198u, 360287970189639u, 72057594037927u, 14411518807585u,
477
0
          2882303761517u, 576460752303u, 115292150460u, 23058430092u, 4611686018u, 922337203u, 184467440u,
478
0
          36893488u, 7378697u, 1475739u, 295147u, 59029u, 11805u, 2361u, 472u, 94u, 18u, 3u };
479
480
0
        unsigned long _Trailing_zero_bits;
481
#ifdef _LIBCPP_HAS_BITSCAN64
482
        (void) _BitScanForward64(&_Trailing_zero_bits, __v.__mantissa); // __v.__mantissa is guaranteed nonzero
483
#else // ^^^ 64-bit ^^^ / vvv 32-bit vvv
484
0
        const uint32_t _Low_mantissa = static_cast<uint32_t>(__v.__mantissa);
485
0
        if (_Low_mantissa != 0) {
486
0
          (void) _BitScanForward(&_Trailing_zero_bits, _Low_mantissa);
487
0
        } else {
488
0
          const uint32_t _High_mantissa = static_cast<uint32_t>(__v.__mantissa >> 32); // nonzero here
489
0
          (void) _BitScanForward(&_Trailing_zero_bits, _High_mantissa);
490
0
          _Trailing_zero_bits += 32;
491
0
        }
492
0
#endif // ^^^ 32-bit ^^^
493
0
        const uint64_t _Shifted_mantissa = __v.__mantissa >> _Trailing_zero_bits;
494
0
        _Can_use_ryu = _Shifted_mantissa <= _Max_shifted_mantissa[_Ryu_exponent];
495
0
      }
496
497
0
      if (!_Can_use_ryu) {
498
        // Print the integer exactly.
499
        // Performance note: This will redundantly perform bounds checking.
500
        // Performance note: This will redundantly decompose the IEEE representation.
501
0
        return __d2fixed_buffered_n(_First, _Last, __f, 0);
502
0
      }
503
504
      // _Can_use_ryu
505
      // Print the decimal digits, left-aligned within [_First, _First + _Total_fixed_length).
506
0
      _Mid = _First + __olength;
507
0
    } else { // cases "1729", "17.29", and "0.001729"
508
      // Print the decimal digits, right-aligned within [_First, _First + _Total_fixed_length).
509
0
      _Mid = _First + _Total_fixed_length;
510
0
    }
511
512
    // We prefer 32-bit operations, even on 64-bit platforms.
513
    // We have at most 17 digits, and uint32_t can store 9 digits.
514
    // If _Output doesn't fit into uint32_t, we cut off 8 digits,
515
    // so the rest will fit into uint32_t.
516
0
    if ((_Output >> 32) != 0) {
517
      // Expensive 64-bit division.
518
0
      const uint64_t __q = __div1e8(_Output);
519
0
      uint32_t __output2 = static_cast<uint32_t>(_Output - 100000000 * __q);
520
0
      _Output = __q;
521
522
0
      const uint32_t __c = __output2 % 10000;
523
0
      __output2 /= 10000;
524
0
      const uint32_t __d = __output2 % 10000;
525
0
      const uint32_t __c0 = (__c % 100) << 1;
526
0
      const uint32_t __c1 = (__c / 100) << 1;
527
0
      const uint32_t __d0 = (__d % 100) << 1;
528
0
      const uint32_t __d1 = (__d / 100) << 1;
529
530
0
      _VSTD::memcpy(_Mid -= 2, __DIGIT_TABLE + __c0, 2);
531
0
      _VSTD::memcpy(_Mid -= 2, __DIGIT_TABLE + __c1, 2);
532
0
      _VSTD::memcpy(_Mid -= 2, __DIGIT_TABLE + __d0, 2);
533
0
      _VSTD::memcpy(_Mid -= 2, __DIGIT_TABLE + __d1, 2);
534
0
    }
535
0
    uint32_t __output2 = static_cast<uint32_t>(_Output);
536
0
    while (__output2 >= 10000) {
537
0
#ifdef __clang__ // TRANSITION, LLVM-38217
538
0
      const uint32_t __c = __output2 - 10000 * (__output2 / 10000);
539
#else
540
      const uint32_t __c = __output2 % 10000;
541
#endif
542
0
      __output2 /= 10000;
543
0
      const uint32_t __c0 = (__c % 100) << 1;
544
0
      const uint32_t __c1 = (__c / 100) << 1;
545
0
      _VSTD::memcpy(_Mid -= 2, __DIGIT_TABLE + __c0, 2);
546
0
      _VSTD::memcpy(_Mid -= 2, __DIGIT_TABLE + __c1, 2);
547
0
    }
548
0
    if (__output2 >= 100) {
549
0
      const uint32_t __c = (__output2 % 100) << 1;
550
0
      __output2 /= 100;
551
0
      _VSTD::memcpy(_Mid -= 2, __DIGIT_TABLE + __c, 2);
552
0
    }
553
0
    if (__output2 >= 10) {
554
0
      const uint32_t __c = __output2 << 1;
555
0
      _VSTD::memcpy(_Mid -= 2, __DIGIT_TABLE + __c, 2);
556
0
    } else {
557
0
      *--_Mid = static_cast<char>('0' + __output2);
558
0
    }
559
560
0
    if (_Ryu_exponent > 0) { // case "172900" with _Can_use_ryu
561
      // Performance note: it might be more efficient to do this immediately after setting _Mid.
562
0
      _VSTD::memset(_First + __olength, '0', static_cast<size_t>(_Ryu_exponent));
563
0
    } else if (_Ryu_exponent == 0) { // case "1729"
564
      // Done!
565
0
    } else if (_Whole_digits > 0) { // case "17.29"
566
      // Performance note: moving digits might not be optimal.
567
0
      _VSTD::memmove(_First, _First + 1, static_cast<size_t>(_Whole_digits));
568
0
      _First[_Whole_digits] = '.';
569
0
    } else { // case "0.001729"
570
      // Performance note: a larger memset() followed by overwriting '.' might be more efficient.
571
0
      _First[0] = '0';
572
0
      _First[1] = '.';
573
0
      _VSTD::memset(_First + 2, '0', static_cast<size_t>(-_Whole_digits));
574
0
    }
575
576
0
    return { _First + _Total_fixed_length, errc{} };
577
0
  }
578
579
0
  const uint32_t _Total_scientific_length = __olength + (__olength > 1) // digits + possible decimal point
580
0
    + (-100 < _Scientific_exponent && _Scientific_exponent < 100 ? 4 : 5); // + scientific exponent
581
0
  if (_Last - _First < static_cast<ptrdiff_t>(_Total_scientific_length)) {
582
0
    return { _Last, errc::value_too_large };
583
0
  }
584
0
  char* const __result = _First;
585
586
  // Print the decimal digits.
587
0
  uint32_t __i = 0;
588
  // We prefer 32-bit operations, even on 64-bit platforms.
589
  // We have at most 17 digits, and uint32_t can store 9 digits.
590
  // If _Output doesn't fit into uint32_t, we cut off 8 digits,
591
  // so the rest will fit into uint32_t.
592
0
  if ((_Output >> 32) != 0) {
593
    // Expensive 64-bit division.
594
0
    const uint64_t __q = __div1e8(_Output);
595
0
    uint32_t __output2 = static_cast<uint32_t>(_Output) - 100000000 * static_cast<uint32_t>(__q);
596
0
    _Output = __q;
597
598
0
    const uint32_t __c = __output2 % 10000;
599
0
    __output2 /= 10000;
600
0
    const uint32_t __d = __output2 % 10000;
601
0
    const uint32_t __c0 = (__c % 100) << 1;
602
0
    const uint32_t __c1 = (__c / 100) << 1;
603
0
    const uint32_t __d0 = (__d % 100) << 1;
604
0
    const uint32_t __d1 = (__d / 100) << 1;
605
0
    _VSTD::memcpy(__result + __olength - __i - 1, __DIGIT_TABLE + __c0, 2);
606
0
    _VSTD::memcpy(__result + __olength - __i - 3, __DIGIT_TABLE + __c1, 2);
607
0
    _VSTD::memcpy(__result + __olength - __i - 5, __DIGIT_TABLE + __d0, 2);
608
0
    _VSTD::memcpy(__result + __olength - __i - 7, __DIGIT_TABLE + __d1, 2);
609
0
    __i += 8;
610
0
  }
611
0
  uint32_t __output2 = static_cast<uint32_t>(_Output);
612
0
  while (__output2 >= 10000) {
613
0
#ifdef __clang__ // TRANSITION, LLVM-38217
614
0
    const uint32_t __c = __output2 - 10000 * (__output2 / 10000);
615
#else
616
    const uint32_t __c = __output2 % 10000;
617
#endif
618
0
    __output2 /= 10000;
619
0
    const uint32_t __c0 = (__c % 100) << 1;
620
0
    const uint32_t __c1 = (__c / 100) << 1;
621
0
    _VSTD::memcpy(__result + __olength - __i - 1, __DIGIT_TABLE + __c0, 2);
622
0
    _VSTD::memcpy(__result + __olength - __i - 3, __DIGIT_TABLE + __c1, 2);
623
0
    __i += 4;
624
0
  }
625
0
  if (__output2 >= 100) {
626
0
    const uint32_t __c = (__output2 % 100) << 1;
627
0
    __output2 /= 100;
628
0
    _VSTD::memcpy(__result + __olength - __i - 1, __DIGIT_TABLE + __c, 2);
629
0
    __i += 2;
630
0
  }
631
0
  if (__output2 >= 10) {
632
0
    const uint32_t __c = __output2 << 1;
633
    // We can't use memcpy here: the decimal dot goes between these two digits.
634
0
    __result[2] = __DIGIT_TABLE[__c + 1];
635
0
    __result[0] = __DIGIT_TABLE[__c];
636
0
  } else {
637
0
    __result[0] = static_cast<char>('0' + __output2);
638
0
  }
639
640
  // Print decimal point if needed.
641
0
  uint32_t __index;
642
0
  if (__olength > 1) {
643
0
    __result[1] = '.';
644
0
    __index = __olength + 1;
645
0
  } else {
646
0
    __index = 1;
647
0
  }
648
649
  // Print the exponent.
650
0
  __result[__index++] = 'e';
651
0
  if (_Scientific_exponent < 0) {
652
0
    __result[__index++] = '-';
653
0
    _Scientific_exponent = -_Scientific_exponent;
654
0
  } else {
655
0
    __result[__index++] = '+';
656
0
  }
657
658
0
  if (_Scientific_exponent >= 100) {
659
0
    const int32_t __c = _Scientific_exponent % 10;
660
0
    _VSTD::memcpy(__result + __index, __DIGIT_TABLE + 2 * (_Scientific_exponent / 10), 2);
661
0
    __result[__index + 2] = static_cast<char>('0' + __c);
662
0
    __index += 3;
663
0
  } else {
664
0
    _VSTD::memcpy(__result + __index, __DIGIT_TABLE + 2 * _Scientific_exponent, 2);
665
0
    __index += 2;
666
0
  }
667
668
0
  return { _First + _Total_scientific_length, errc{} };
669
0
}
670
671
[[nodiscard]] _LIBCPP_HIDE_FROM_ABI inline bool __d2d_small_int(const uint64_t __ieeeMantissa, const uint32_t __ieeeExponent,
672
0
  __floating_decimal_64* const __v) {
673
0
  const uint64_t __m2 = (1ull << __DOUBLE_MANTISSA_BITS) | __ieeeMantissa;
674
0
  const int32_t __e2 = static_cast<int32_t>(__ieeeExponent) - __DOUBLE_BIAS - __DOUBLE_MANTISSA_BITS;
675
676
0
  if (__e2 > 0) {
677
    // f = __m2 * 2^__e2 >= 2^53 is an integer.
678
    // Ignore this case for now.
679
0
    return false;
680
0
  }
681
682
0
  if (__e2 < -52) {
683
    // f < 1.
684
0
    return false;
685
0
  }
686
687
  // Since 2^52 <= __m2 < 2^53 and 0 <= -__e2 <= 52: 1 <= f = __m2 / 2^-__e2 < 2^53.
688
  // Test if the lower -__e2 bits of the significand are 0, i.e. whether the fraction is 0.
689
0
  const uint64_t __mask = (1ull << -__e2) - 1;
690
0
  const uint64_t __fraction = __m2 & __mask;
691
0
  if (__fraction != 0) {
692
0
    return false;
693
0
  }
694
695
  // f is an integer in the range [1, 2^53).
696
  // Note: __mantissa might contain trailing (decimal) 0's.
697
  // Note: since 2^53 < 10^16, there is no need to adjust __decimalLength17().
698
0
  __v->__mantissa = __m2 >> -__e2;
699
0
  __v->__exponent = 0;
700
0
  return true;
701
0
}
702
703
[[nodiscard]] to_chars_result __d2s_buffered_n(char* const _First, char* const _Last, const double __f,
704
0
  const chars_format _Fmt) {
705
706
  // Step 1: Decode the floating-point number, and unify normalized and subnormal cases.
707
0
  const uint64_t __bits = __double_to_bits(__f);
708
709
  // Case distinction; exit early for the easy cases.
710
0
  if (__bits == 0) {
711
0
    if (_Fmt == chars_format::scientific) {
712
0
      if (_Last - _First < 5) {
713
0
        return { _Last, errc::value_too_large };
714
0
      }
715
716
0
      _VSTD::memcpy(_First, "0e+00", 5);
717
718
0
      return { _First + 5, errc{} };
719
0
    }
720
721
    // Print "0" for chars_format::fixed, chars_format::general, and chars_format{}.
722
0
    if (_First == _Last) {
723
0
      return { _Last, errc::value_too_large };
724
0
    }
725
726
0
    *_First = '0';
727
728
0
    return { _First + 1, errc{} };
729
0
  }
730
731
  // Decode __bits into mantissa and exponent.
732
0
  const uint64_t __ieeeMantissa = __bits & ((1ull << __DOUBLE_MANTISSA_BITS) - 1);
733
0
  const uint32_t __ieeeExponent = static_cast<uint32_t>(__bits >> __DOUBLE_MANTISSA_BITS);
734
735
0
  if (_Fmt == chars_format::fixed) {
736
    // const uint64_t _Mantissa2 = __ieeeMantissa | (1ull << __DOUBLE_MANTISSA_BITS); // restore implicit bit
737
0
    const int32_t _Exponent2 = static_cast<int32_t>(__ieeeExponent)
738
0
      - __DOUBLE_BIAS - __DOUBLE_MANTISSA_BITS; // bias and normalization
739
740
    // Normal values are equal to _Mantissa2 * 2^_Exponent2.
741
    // (Subnormals are different, but they'll be rejected by the _Exponent2 test here, so they can be ignored.)
742
743
    // For nonzero integers, _Exponent2 >= -52. (The minimum value occurs when _Mantissa2 * 2^_Exponent2 is 1.
744
    // In that case, _Mantissa2 is the implicit 1 bit followed by 52 zeros, so _Exponent2 is -52 to shift away
745
    // the zeros.) The dense range of exactly representable integers has negative or zero exponents
746
    // (as positive exponents make the range non-dense). For that dense range, Ryu will always be used:
747
    // every digit is necessary to uniquely identify the value, so Ryu must print them all.
748
749
    // Positive exponents are the non-dense range of exactly representable integers. This contains all of the values
750
    // for which Ryu can't be used (and a few Ryu-friendly values). We can save time by detecting positive
751
    // exponents here and skipping Ryu. Calling __d2fixed_buffered_n() with precision 0 is valid for all integers
752
    // (so it's okay if we call it with a Ryu-friendly value).
753
0
    if (_Exponent2 > 0) {
754
0
      return __d2fixed_buffered_n(_First, _Last, __f, 0);
755
0
    }
756
0
  }
757
758
0
  __floating_decimal_64 __v;
759
0
  const bool __isSmallInt = __d2d_small_int(__ieeeMantissa, __ieeeExponent, &__v);
760
0
  if (__isSmallInt) {
761
    // For small integers in the range [1, 2^53), __v.__mantissa might contain trailing (decimal) zeros.
762
    // For scientific notation we need to move these zeros into the exponent.
763
    // (This is not needed for fixed-point notation, so it might be beneficial to trim
764
    // trailing zeros in __to_chars only if needed - once fixed-point notation output is implemented.)
765
0
    for (;;) {
766
0
      const uint64_t __q = __div10(__v.__mantissa);
767
0
      const uint32_t __r = static_cast<uint32_t>(__v.__mantissa) - 10 * static_cast<uint32_t>(__q);
768
0
      if (__r != 0) {
769
0
        break;
770
0
      }
771
0
      __v.__mantissa = __q;
772
0
      ++__v.__exponent;
773
0
    }
774
0
  } else {
775
0
    __v = __d2d(__ieeeMantissa, __ieeeExponent);
776
0
  }
777
778
0
  return __to_chars(_First, _Last, __v, _Fmt, __f);
779
0
}
780
781
_LIBCPP_END_NAMESPACE_STD
782
783
// clang-format on