summaryrefslogtreecommitdiffstats
path: root/target-alpha/fpu_helper.c
blob: b091aa842a33cc409591836902294ba66b5eb806 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
/*
 *  Helpers for floating point instructions.
 *
 *  Copyright (c) 2007 Jocelyn Mayer
 *
 * This library is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2 of the License, or (at your option) any later version.
 *
 * This library is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with this library; if not, see <http://www.gnu.org/licenses/>.
 */

#include "cpu.h"
#include "exec/helper-proto.h"
#include "fpu/softfloat.h"

#define FP_STATUS (env->fp_status)


void helper_setroundmode(CPUAlphaState *env, uint32_t val)
{
    set_float_rounding_mode(val, &FP_STATUS);
}

void helper_setflushzero(CPUAlphaState *env, uint32_t val)
{
    set_flush_to_zero(val, &FP_STATUS);
}

#define CONVERT_BIT(X, SRC, DST) \
    (SRC > DST ? (X) / (SRC / DST) & (DST) : ((X) & SRC) * (DST / SRC))

static uint32_t soft_to_fpcr_exc(CPUAlphaState *env)
{
    uint8_t exc = get_float_exception_flags(&FP_STATUS);
    uint32_t ret = 0;

    if (unlikely(exc)) {
        set_float_exception_flags(0, &FP_STATUS);
        ret |= CONVERT_BIT(exc, float_flag_invalid, FPCR_INV);
        ret |= CONVERT_BIT(exc, float_flag_divbyzero, FPCR_DZE);
        ret |= CONVERT_BIT(exc, float_flag_overflow, FPCR_OVF);
        ret |= CONVERT_BIT(exc, float_flag_underflow, FPCR_UNF);
        ret |= CONVERT_BIT(exc, float_flag_inexact, FPCR_INE);
    }

    return ret;
}

static void fp_exc_raise1(CPUAlphaState *env, uintptr_t retaddr,
                          uint32_t exc, uint32_t regno, uint32_t hw_exc)
{
    hw_exc |= CONVERT_BIT(exc, FPCR_INV, EXC_M_INV);
    hw_exc |= CONVERT_BIT(exc, FPCR_DZE, EXC_M_DZE);
    hw_exc |= CONVERT_BIT(exc, FPCR_OVF, EXC_M_FOV);
    hw_exc |= CONVERT_BIT(exc, FPCR_UNF, EXC_M_UNF);
    hw_exc |= CONVERT_BIT(exc, FPCR_INE, EXC_M_INE);
    hw_exc |= CONVERT_BIT(exc, FPCR_IOV, EXC_M_IOV);

    arith_excp(env, retaddr, hw_exc, 1ull << regno);
}

/* Raise exceptions for ieee fp insns without software completion.
   In that case there are no exceptions that don't trap; the mask
   doesn't apply.  */
void helper_fp_exc_raise(CPUAlphaState *env, uint32_t ignore, uint32_t regno)
{
    uint32_t exc = env->error_code;
    if (exc) {
        env->fpcr |= exc;
        exc &= ~ignore;
        if (exc) {
            fp_exc_raise1(env, GETPC(), exc, regno, 0);
        }
    }
}

/* Raise exceptions for ieee fp insns with software completion.  */
void helper_fp_exc_raise_s(CPUAlphaState *env, uint32_t ignore, uint32_t regno)
{
    uint32_t exc = env->error_code & ~ignore;
    if (exc) {
        env->fpcr |= exc;
        exc &= ~ignore;
        if (exc) {
            exc &= env->fpcr_exc_enable;
            fp_exc_raise1(env, GETPC(), exc, regno, EXC_M_SWC);
        }
    }
}

/* Input handing without software completion.  Trap for all
   non-finite numbers.  */
void helper_ieee_input(CPUAlphaState *env, uint64_t val)
{
    uint32_t exp = (uint32_t)(val >> 52) & 0x7ff;
    uint64_t frac = val & 0xfffffffffffffull;

    if (exp == 0) {
        /* Denormals without /S raise an exception.  */
        if (frac != 0) {
            arith_excp(env, GETPC(), EXC_M_INV, 0);
        }
    } else if (exp == 0x7ff) {
        /* Infinity or NaN.  */
        env->fpcr |= FPCR_INV;
        arith_excp(env, GETPC(), EXC_M_INV, 0);
    }
}

/* Similar, but does not trap for infinities.  Used for comparisons.  */
void helper_ieee_input_cmp(CPUAlphaState *env, uint64_t val)
{
    uint32_t exp = (uint32_t)(val >> 52) & 0x7ff;
    uint64_t frac = val & 0xfffffffffffffull;

    if (exp == 0) {
        /* Denormals without /S raise an exception.  */
        if (frac != 0) {
            arith_excp(env, GETPC(), EXC_M_INV, 0);
        }
    } else if (exp == 0x7ff && frac) {
        /* NaN.  */
        env->fpcr |= FPCR_INV;
        arith_excp(env, GETPC(), EXC_M_INV, 0);
    }
}

/* Input handing with software completion.  Trap for denorms, unless DNZ
   is set.  If we try to support DNOD (which none of the produced hardware
   did, AFAICS), we'll need to suppress the trap when FPCR.DNOD is set;
   then the code downstream of that will need to cope with denorms sans
   flush_input_to_zero.  Most of it should work sanely, but there's
   nothing to compare with.  */
void helper_ieee_input_s(CPUAlphaState *env, uint64_t val)
{
    if (unlikely(2 * val - 1 < 0x1fffffffffffffull)
        && !env->fp_status.flush_inputs_to_zero) {
        arith_excp(env, GETPC(), EXC_M_INV | EXC_M_SWC, 0);
    }
}

/* S floating (single) */

/* Taken from linux/arch/alpha/kernel/traps.c, s_mem_to_reg.  */
static inline uint64_t float32_to_s_int(uint32_t fi)
{
    uint32_t frac = fi & 0x7fffff;
    uint32_t sign = fi >> 31;
    uint32_t exp_msb = (fi >> 30) & 1;
    uint32_t exp_low = (fi >> 23) & 0x7f;
    uint32_t exp;

    exp = (exp_msb << 10) | exp_low;
    if (exp_msb) {
        if (exp_low == 0x7f) {
            exp = 0x7ff;
        }
    } else {
        if (exp_low != 0x00) {
            exp |= 0x380;
        }
    }

    return (((uint64_t)sign << 63)
            | ((uint64_t)exp << 52)
            | ((uint64_t)frac << 29));
}

static inline uint64_t float32_to_s(float32 fa)
{
    CPU_FloatU a;
    a.f = fa;
    return float32_to_s_int(a.l);
}

static inline uint32_t s_to_float32_int(uint64_t a)
{
    return ((a >> 32) & 0xc0000000) | ((a >> 29) & 0x3fffffff);
}

static inline float32 s_to_float32(uint64_t a)
{
    CPU_FloatU r;
    r.l = s_to_float32_int(a);
    return r.f;
}

uint32_t helper_s_to_memory(uint64_t a)
{
    return s_to_float32_int(a);
}

uint64_t helper_memory_to_s(uint32_t a)
{
    return float32_to_s_int(a);
}

uint64_t helper_adds(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float32 fa, fb, fr;

    fa = s_to_float32(a);
    fb = s_to_float32(b);
    fr = float32_add(fa, fb, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float32_to_s(fr);
}

uint64_t helper_subs(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float32 fa, fb, fr;

    fa = s_to_float32(a);
    fb = s_to_float32(b);
    fr = float32_sub(fa, fb, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float32_to_s(fr);
}

uint64_t helper_muls(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float32 fa, fb, fr;

    fa = s_to_float32(a);
    fb = s_to_float32(b);
    fr = float32_mul(fa, fb, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float32_to_s(fr);
}

uint64_t helper_divs(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float32 fa, fb, fr;

    fa = s_to_float32(a);
    fb = s_to_float32(b);
    fr = float32_div(fa, fb, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float32_to_s(fr);
}

uint64_t helper_sqrts(CPUAlphaState *env, uint64_t a)
{
    float32 fa, fr;

    fa = s_to_float32(a);
    fr = float32_sqrt(fa, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float32_to_s(fr);
}


/* T floating (double) */
static inline float64 t_to_float64(uint64_t a)
{
    /* Memory format is the same as float64 */
    CPU_DoubleU r;
    r.ll = a;
    return r.d;
}

static inline uint64_t float64_to_t(float64 fa)
{
    /* Memory format is the same as float64 */
    CPU_DoubleU r;
    r.d = fa;
    return r.ll;
}

uint64_t helper_addt(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float64 fa, fb, fr;

    fa = t_to_float64(a);
    fb = t_to_float64(b);
    fr = float64_add(fa, fb, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float64_to_t(fr);
}

uint64_t helper_subt(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float64 fa, fb, fr;

    fa = t_to_float64(a);
    fb = t_to_float64(b);
    fr = float64_sub(fa, fb, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float64_to_t(fr);
}

uint64_t helper_mult(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float64 fa, fb, fr;

    fa = t_to_float64(a);
    fb = t_to_float64(b);
    fr = float64_mul(fa, fb, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float64_to_t(fr);
}

uint64_t helper_divt(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float64 fa, fb, fr;

    fa = t_to_float64(a);
    fb = t_to_float64(b);
    fr = float64_div(fa, fb, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float64_to_t(fr);
}

uint64_t helper_sqrtt(CPUAlphaState *env, uint64_t a)
{
    float64 fa, fr;

    fa = t_to_float64(a);
    fr = float64_sqrt(fa, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float64_to_t(fr);
}

/* Comparisons */
uint64_t helper_cmptun(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float64 fa, fb;
    uint64_t ret = 0;

    fa = t_to_float64(a);
    fb = t_to_float64(b);

    if (float64_unordered_quiet(fa, fb, &FP_STATUS)) {
        ret = 0x4000000000000000ULL;
    }
    env->error_code = soft_to_fpcr_exc(env);

    return ret;
}

uint64_t helper_cmpteq(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float64 fa, fb;
    uint64_t ret = 0;

    fa = t_to_float64(a);
    fb = t_to_float64(b);

    if (float64_eq_quiet(fa, fb, &FP_STATUS)) {
        ret = 0x4000000000000000ULL;
    }
    env->error_code = soft_to_fpcr_exc(env);

    return ret;
}

uint64_t helper_cmptle(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float64 fa, fb;
    uint64_t ret = 0;

    fa = t_to_float64(a);
    fb = t_to_float64(b);

    if (float64_le(fa, fb, &FP_STATUS)) {
        ret = 0x4000000000000000ULL;
    }
    env->error_code = soft_to_fpcr_exc(env);

    return ret;
}

uint64_t helper_cmptlt(CPUAlphaState *env, uint64_t a, uint64_t b)
{
    float64 fa, fb;
    uint64_t ret = 0;

    fa = t_to_float64(a);
    fb = t_to_float64(b);

    if (float64_lt(fa, fb, &FP_STATUS)) {
        ret = 0x4000000000000000ULL;
    }
    env->error_code = soft_to_fpcr_exc(env);

    return ret;
}

/* Floating point format conversion */
uint64_t helper_cvtts(CPUAlphaState *env, uint64_t a)
{
    float64 fa;
    float32 fr;

    fa = t_to_float64(a);
    fr = float64_to_float32(fa, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float32_to_s(fr);
}

uint64_t helper_cvtst(CPUAlphaState *env, uint64_t a)
{
    float32 fa;
    float64 fr;

    fa = s_to_float32(a);
    fr = float32_to_float64(fa, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float64_to_t(fr);
}

uint64_t helper_cvtqs(CPUAlphaState *env, uint64_t a)
{
    float32 fr = int64_to_float32(a, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);

    return float32_to_s(fr);
}

/* Implement float64 to uint64 conversion without saturation -- we must
   supply the truncated result.  This behaviour is used by the compiler
   to get unsigned conversion for free with the same instruction.  */

static uint64_t do_cvttq(CPUAlphaState *env, uint64_t a, int roundmode)
{
    uint64_t frac, ret = 0;
    uint32_t exp, sign, exc = 0;
    int shift;

    sign = (a >> 63);
    exp = (uint32_t)(a >> 52) & 0x7ff;
    frac = a & 0xfffffffffffffull;

    if (exp == 0) {
        if (unlikely(frac != 0) && !env->fp_status.flush_inputs_to_zero) {
            goto do_underflow;
        }
    } else if (exp == 0x7ff) {
        exc = FPCR_INV;
    } else {
        /* Restore implicit bit.  */
        frac |= 0x10000000000000ull;

        shift = exp - 1023 - 52;
        if (shift >= 0) {
            /* In this case the number is so large that we must shift
               the fraction left.  There is no rounding to do.  */
            if (shift < 64) {
                ret = frac << shift;
            }
            /* Check for overflow.  Note the special case of -0x1p63.  */
            if (shift >= 11 && a != 0xC3E0000000000000ull) {
                exc = FPCR_IOV | FPCR_INE;
            }
        } else {
            uint64_t round;

            /* In this case the number is smaller than the fraction as
               represented by the 52 bit number.  Here we must think
               about rounding the result.  Handle this by shifting the
               fractional part of the number into the high bits of ROUND.
               This will let us efficiently handle round-to-nearest.  */
            shift = -shift;
            if (shift < 63) {
                ret = frac >> shift;
                round = frac << (64 - shift);
            } else {
                /* The exponent is so small we shift out everything.
                   Leave a sticky bit for proper rounding below.  */
            do_underflow:
                round = 1;
            }

            if (round) {
                exc = FPCR_INE;
                switch (roundmode) {
                case float_round_nearest_even:
                    if (round == (1ull << 63)) {
                        /* Fraction is exactly 0.5; round to even.  */
                        ret += (ret & 1);
                    } else if (round > (1ull << 63)) {
                        ret += 1;
                    }
                    break;
                case float_round_to_zero:
                    break;
                case float_round_up:
                    ret += 1 - sign;
                    break;
                case float_round_down:
                    ret += sign;
                    break;
                }
            }
        }
        if (sign) {
            ret = -ret;
        }
    }
    env->error_code = exc;

    return ret;
}

uint64_t helper_cvttq(CPUAlphaState *env, uint64_t a)
{
    return do_cvttq(env, a, FP_STATUS.float_rounding_mode);
}

uint64_t helper_cvttq_c(CPUAlphaState *env, uint64_t a)
{
    return do_cvttq(env, a, float_round_to_zero);
}

uint64_t helper_cvtqt(CPUAlphaState *env, uint64_t a)
{
    float64 fr = int64_to_float64(a, &FP_STATUS);
    env->error_code = soft_to_fpcr_exc(env);
    return float64_to_t(fr);
}

uint64_t helper_cvtql(CPUAlphaState *env, uint64_t val)
{
    uint32_t exc = 0;
    if (val != (int32_t)val) {
        exc = FPCR_IOV | FPCR_INE;
    }
    env->error_code = exc;

    return ((val & 0xc0000000) << 32) | ((val & 0x3fffffff) << 29);
}
OpenPOWER on IntegriCloud