?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
[PATCH] shpchp - cleanup init_slots()
[linux-2.6/linux-loongson.git] / arch / arm / vfp / vfpdouble.c
blob9b367a65cb4d65caec8afedf39b53c2cb99b0159
1 /*
2 * linux/arch/arm/vfp/vfpdouble.c
3 *
4 * This code is derived in part from John R. Housers softfloat library, which
5 * carries the following notice:
6 *
7 * ===========================================================================
8 * This C source file is part of the SoftFloat IEC/IEEE Floating-point
9 * Arithmetic Package, Release 2.
10 *
11 * Written by John R. Hauser. This work was made possible in part by the
12 * International Computer Science Institute, located at Suite 600, 1947 Center
13 * Street, Berkeley, California 94704. Funding was partially provided by the
14 * National Science Foundation under grant MIP-9311980. The original version
15 * of this code was written as part of a project to build a fixed-point vector
16 * processor in collaboration with the University of California at Berkeley,
17 * overseen by Profs. Nelson Morgan and John Wawrzynek. More information
18 * is available through the web page `http://HTTP.CS.Berkeley.EDU/~jhauser/
19 * arithmetic/softfloat.html'.
20 *
21 * THIS SOFTWARE IS DISTRIBUTED AS IS, FOR FREE. Although reasonable effort
22 * has been made to avoid it, THIS SOFTWARE MAY CONTAIN FAULTS THAT WILL AT
23 * TIMES RESULT IN INCORRECT BEHAVIOR. USE OF THIS SOFTWARE IS RESTRICTED TO
24 * PERSONS AND ORGANIZATIONS WHO CAN AND WILL TAKE FULL RESPONSIBILITY FOR ANY
25 * AND ALL LOSSES, COSTS, OR OTHER PROBLEMS ARISING FROM ITS USE.
26 *
27 * Derivative works are acceptable, even for commercial purposes, so long as
28 * (1) they include prominent notice that the work is derivative, and (2) they
29 * include prominent notice akin to these three paragraphs for those parts of
30 * this code that are retained.
31 * ===========================================================================
32 */
33 #include <linux/kernel.h>
34 #include <linux/bitops.h>
35
36 #include <asm/div64.h>
37 #include <asm/ptrace.h>
38 #include <asm/vfp.h>
39
40 #include "vfpinstr.h"
41 #include "vfp.h"
42
43 static struct vfp_double vfp_double_default_qnan = {
44 .exponent = 2047,
45 .sign = 0,
46 .significand = VFP_DOUBLE_SIGNIFICAND_QNAN,
47 };
48
49 static void vfp_double_dump(const char *str, struct vfp_double *d)
50 {
51 pr_debug("VFP: %s: sign=%d exponent=%d significand=%016llx\n",
52 str, d->sign != 0, d->exponent, d->significand);
53 }
54
55 static void vfp_double_normalise_denormal(struct vfp_double *vd)
56 {
57 int bits = 31 - fls(vd->significand >> 32);
58 if (bits == 31)
59 bits = 62 - fls(vd->significand);
60
61 vfp_double_dump("normalise_denormal: in", vd);
62
63 if (bits) {
64 vd->exponent -= bits - 1;
65 vd->significand <<= bits;
66 }
67
68 vfp_double_dump("normalise_denormal: out", vd);
69 }
70
71 u32 vfp_double_normaliseround(int dd, struct vfp_double *vd, u32 fpscr, u32 exceptions, const char *func)
72 {
73 u64 significand, incr;
74 int exponent, shift, underflow;
75 u32 rmode;
76
77 vfp_double_dump("pack: in", vd);
78
79 /*
80 * Infinities and NaNs are a special case.
81 */
82 if (vd->exponent == 2047 && (vd->significand == 0 || exceptions))
83 goto pack;
84
85 /*
86 * Special-case zero.
87 */
88 if (vd->significand == 0) {
89 vd->exponent = 0;
90 goto pack;
91 }
92
93 exponent = vd->exponent;
94 significand = vd->significand;
95
96 shift = 32 - fls(significand >> 32);
97 if (shift == 32)
98 shift = 64 - fls(significand);
99 if (shift) {
100 exponent -= shift;
101 significand <<= shift;
102 }
103
104 #ifdef DEBUG
105 vd->exponent = exponent;
106 vd->significand = significand;
107 vfp_double_dump("pack: normalised", vd);
108 #endif
109
110 /*
111 * Tiny number?
112 */
113 underflow = exponent < 0;
114 if (underflow) {
115 significand = vfp_shiftright64jamming(significand, -exponent);
116 exponent = 0;
117 #ifdef DEBUG
118 vd->exponent = exponent;
119 vd->significand = significand;
120 vfp_double_dump("pack: tiny number", vd);
121 #endif
122 if (!(significand & ((1ULL << (VFP_DOUBLE_LOW_BITS + 1)) - 1)))
123 underflow = 0;
124 }
125
126 /*
127 * Select rounding increment.
128 */
129 incr = 0;
130 rmode = fpscr & FPSCR_RMODE_MASK;
131
132 if (rmode == FPSCR_ROUND_NEAREST) {
133 incr = 1ULL << VFP_DOUBLE_LOW_BITS;
134 if ((significand & (1ULL << (VFP_DOUBLE_LOW_BITS + 1))) == 0)
135 incr -= 1;
136 } else if (rmode == FPSCR_ROUND_TOZERO) {
137 incr = 0;
138 } else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vd->sign != 0))
139 incr = (1ULL << (VFP_DOUBLE_LOW_BITS + 1)) - 1;
140
141 pr_debug("VFP: rounding increment = 0x%08llx\n", incr);
142
143 /*
144 * Is our rounding going to overflow?
145 */
146 if ((significand + incr) < significand) {
147 exponent += 1;
148 significand = (significand >> 1) | (significand & 1);
149 incr >>= 1;
150 #ifdef DEBUG
151 vd->exponent = exponent;
152 vd->significand = significand;
153 vfp_double_dump("pack: overflow", vd);
154 #endif
155 }
156
157 /*
158 * If any of the low bits (which will be shifted out of the
159 * number) are non-zero, the result is inexact.
160 */
161 if (significand & ((1 << (VFP_DOUBLE_LOW_BITS + 1)) - 1))
162 exceptions |= FPSCR_IXC;
163
164 /*
165 * Do our rounding.
166 */
167 significand += incr;
168
169 /*
170 * Infinity?
171 */
172 if (exponent >= 2046) {
173 exceptions |= FPSCR_OFC | FPSCR_IXC;
174 if (incr == 0) {
175 vd->exponent = 2045;
176 vd->significand = 0x7fffffffffffffffULL;
177 } else {
178 vd->exponent = 2047; /* infinity */
179 vd->significand = 0;
180 }
181 } else {
182 if (significand >> (VFP_DOUBLE_LOW_BITS + 1) == 0)
183 exponent = 0;
184 if (exponent || significand > 0x8000000000000000ULL)
185 underflow = 0;
186 if (underflow)
187 exceptions |= FPSCR_UFC;
188 vd->exponent = exponent;
189 vd->significand = significand >> 1;
190 }
191
192 pack:
193 vfp_double_dump("pack: final", vd);
194 {
195 s64 d = vfp_double_pack(vd);
196 pr_debug("VFP: %s: d(d%d)=%016llx exceptions=%08x\n", func,
197 dd, d, exceptions);
198 vfp_put_double(dd, d);
199 }
200 return exceptions & ~VFP_NAN_FLAG;
201 }
202
203 /*
204 * Propagate the NaN, setting exceptions if it is signalling.
205 * 'n' is always a NaN. 'm' may be a number, NaN or infinity.
206 */
207 static u32
208 vfp_propagate_nan(struct vfp_double *vdd, struct vfp_double *vdn,
209 struct vfp_double *vdm, u32 fpscr)
210 {
211 struct vfp_double *nan;
212 int tn, tm = 0;
213
214 tn = vfp_double_type(vdn);
215
216 if (vdm)
217 tm = vfp_double_type(vdm);
218
219 if (fpscr & FPSCR_DEFAULT_NAN)
220 /*
221 * Default NaN mode - always returns a quiet NaN
222 */
223 nan = &vfp_double_default_qnan;
224 else {
225 /*
226 * Contemporary mode - select the first signalling
227 * NAN, or if neither are signalling, the first
228 * quiet NAN.
229 */
230 if (tn == VFP_SNAN || (tm != VFP_SNAN && tn == VFP_QNAN))
231 nan = vdn;
232 else
233 nan = vdm;
234 /*
235 * Make the NaN quiet.
236 */
237 nan->significand |= VFP_DOUBLE_SIGNIFICAND_QNAN;
238 }
239
240 *vdd = *nan;
241
242 /*
243 * If one was a signalling NAN, raise invalid operation.
244 */
245 return tn == VFP_SNAN || tm == VFP_SNAN ? FPSCR_IOC : VFP_NAN_FLAG;
246 }
247
248 /*
249 * Extended operations
250 */
251 static u32 vfp_double_fabs(int dd, int unused, int dm, u32 fpscr)
252 {
253 vfp_put_double(dd, vfp_double_packed_abs(vfp_get_double(dm)));
254 return 0;
255 }
256
257 static u32 vfp_double_fcpy(int dd, int unused, int dm, u32 fpscr)
258 {
259 vfp_put_double(dd, vfp_get_double(dm));
260 return 0;
261 }
262
263 static u32 vfp_double_fneg(int dd, int unused, int dm, u32 fpscr)
264 {
265 vfp_put_double(dd, vfp_double_packed_negate(vfp_get_double(dm)));
266 return 0;
267 }
268
269 static u32 vfp_double_fsqrt(int dd, int unused, int dm, u32 fpscr)
270 {
271 struct vfp_double vdm, vdd;
272 int ret, tm;
273
274 vfp_double_unpack(&vdm, vfp_get_double(dm));
275 tm = vfp_double_type(&vdm);
276 if (tm & (VFP_NAN|VFP_INFINITY)) {
277 struct vfp_double *vdp = &vdd;
278
279 if (tm & VFP_NAN)
280 ret = vfp_propagate_nan(vdp, &vdm, NULL, fpscr);
281 else if (vdm.sign == 0) {
282 sqrt_copy:
283 vdp = &vdm;
284 ret = 0;
285 } else {
286 sqrt_invalid:
287 vdp = &vfp_double_default_qnan;
288 ret = FPSCR_IOC;
289 }
290 vfp_put_double(dd, vfp_double_pack(vdp));
291 return ret;
292 }
293
294 /*
295 * sqrt(+/- 0) == +/- 0
296 */
297 if (tm & VFP_ZERO)
298 goto sqrt_copy;
299
300 /*
301 * Normalise a denormalised number
302 */
303 if (tm & VFP_DENORMAL)
304 vfp_double_normalise_denormal(&vdm);
305
306 /*
307 * sqrt(<0) = invalid
308 */
309 if (vdm.sign)
310 goto sqrt_invalid;
311
312 vfp_double_dump("sqrt", &vdm);
313
314 /*
315 * Estimate the square root.
316 */
317 vdd.sign = 0;
318 vdd.exponent = ((vdm.exponent - 1023) >> 1) + 1023;
319 vdd.significand = (u64)vfp_estimate_sqrt_significand(vdm.exponent, vdm.significand >> 32) << 31;
320
321 vfp_double_dump("sqrt estimate1", &vdd);
322
323 vdm.significand >>= 1 + (vdm.exponent & 1);
324 vdd.significand += 2 + vfp_estimate_div128to64(vdm.significand, 0, vdd.significand);
325
326 vfp_double_dump("sqrt estimate2", &vdd);
327
328 /*
329 * And now adjust.
330 */
331 if ((vdd.significand & VFP_DOUBLE_LOW_BITS_MASK) <= 5) {
332 if (vdd.significand < 2) {
333 vdd.significand = ~0ULL;
334 } else {
335 u64 termh, terml, remh, reml;
336 vdm.significand <<= 2;
337 mul64to128(&termh, &terml, vdd.significand, vdd.significand);
338 sub128(&remh, &reml, vdm.significand, 0, termh, terml);
339 while ((s64)remh < 0) {
340 vdd.significand -= 1;
341 shift64left(&termh, &terml, vdd.significand);
342 terml |= 1;
343 add128(&remh, &reml, remh, reml, termh, terml);
344 }
345 vdd.significand |= (remh | reml) != 0;
346 }
347 }
348 vdd.significand = vfp_shiftright64jamming(vdd.significand, 1);
349
350 return vfp_double_normaliseround(dd, &vdd, fpscr, 0, "fsqrt");
351 }
352
353 /*
354 * Equal := ZC
355 * Less than := N
356 * Greater than := C
357 * Unordered := CV
358 */
359 static u32 vfp_compare(int dd, int signal_on_qnan, int dm, u32 fpscr)
360 {
361 s64 d, m;
362 u32 ret = 0;
363
364 m = vfp_get_double(dm);
365 if (vfp_double_packed_exponent(m) == 2047 && vfp_double_packed_mantissa(m)) {
366 ret |= FPSCR_C | FPSCR_V;
367 if (signal_on_qnan || !(vfp_double_packed_mantissa(m) & (1ULL << (VFP_DOUBLE_MANTISSA_BITS - 1))))
368 /*
369 * Signalling NaN, or signalling on quiet NaN
370 */
371 ret |= FPSCR_IOC;
372 }
373
374 d = vfp_get_double(dd);
375 if (vfp_double_packed_exponent(d) == 2047 && vfp_double_packed_mantissa(d)) {
376 ret |= FPSCR_C | FPSCR_V;
377 if (signal_on_qnan || !(vfp_double_packed_mantissa(d) & (1ULL << (VFP_DOUBLE_MANTISSA_BITS - 1))))
378 /*
379 * Signalling NaN, or signalling on quiet NaN
380 */
381 ret |= FPSCR_IOC;
382 }
383
384 if (ret == 0) {
385 if (d == m || vfp_double_packed_abs(d | m) == 0) {
386 /*
387 * equal
388 */
389 ret |= FPSCR_Z | FPSCR_C;
390 } else if (vfp_double_packed_sign(d ^ m)) {
391 /*
392 * different signs
393 */
394 if (vfp_double_packed_sign(d))
395 /*
396 * d is negative, so d < m
397 */
398 ret |= FPSCR_N;
399 else
400 /*
401 * d is positive, so d > m
402 */
403 ret |= FPSCR_C;
404 } else if ((vfp_double_packed_sign(d) != 0) ^ (d < m)) {
405 /*
406 * d < m
407 */
408 ret |= FPSCR_N;
409 } else if ((vfp_double_packed_sign(d) != 0) ^ (d > m)) {
410 /*
411 * d > m
412 */
413 ret |= FPSCR_C;
414 }
415 }
416
417 return ret;
418 }
419
420 static u32 vfp_double_fcmp(int dd, int unused, int dm, u32 fpscr)
421 {
422 return vfp_compare(dd, 0, dm, fpscr);
423 }
424
425 static u32 vfp_double_fcmpe(int dd, int unused, int dm, u32 fpscr)
426 {
427 return vfp_compare(dd, 1, dm, fpscr);
428 }
429
430 static u32 vfp_double_fcmpz(int dd, int unused, int dm, u32 fpscr)
431 {
432 return vfp_compare(dd, 0, VFP_REG_ZERO, fpscr);
433 }
434
435 static u32 vfp_double_fcmpez(int dd, int unused, int dm, u32 fpscr)
436 {
437 return vfp_compare(dd, 1, VFP_REG_ZERO, fpscr);
438 }
439
440 static u32 vfp_double_fcvts(int sd, int unused, int dm, u32 fpscr)
441 {
442 struct vfp_double vdm;
443 struct vfp_single vsd;
444 int tm;
445 u32 exceptions = 0;
446
447 vfp_double_unpack(&vdm, vfp_get_double(dm));
448
449 tm = vfp_double_type(&vdm);
450
451 /*
452 * If we have a signalling NaN, signal invalid operation.
453 */
454 if (tm == VFP_SNAN)
455 exceptions = FPSCR_IOC;
456
457 if (tm & VFP_DENORMAL)
458 vfp_double_normalise_denormal(&vdm);
459
460 vsd.sign = vdm.sign;
461 vsd.significand = vfp_hi64to32jamming(vdm.significand);
462
463 /*
464 * If we have an infinity or a NaN, the exponent must be 255
465 */
466 if (tm & (VFP_INFINITY|VFP_NAN)) {
467 vsd.exponent = 255;
468 if (tm & VFP_NAN)
469 vsd.significand |= VFP_SINGLE_SIGNIFICAND_QNAN;
470 goto pack_nan;
471 } else if (tm & VFP_ZERO)
472 vsd.exponent = 0;
473 else
474 vsd.exponent = vdm.exponent - (1023 - 127);
475
476 return vfp_single_normaliseround(sd, &vsd, fpscr, exceptions, "fcvts");
477
478 pack_nan:
479 vfp_put_float(sd, vfp_single_pack(&vsd));
480 return exceptions;
481 }
482
483 static u32 vfp_double_fuito(int dd, int unused, int dm, u32 fpscr)
484 {
485 struct vfp_double vdm;
486 u32 m = vfp_get_float(dm);
487
488 vdm.sign = 0;
489 vdm.exponent = 1023 + 63 - 1;
490 vdm.significand = (u64)m;
491
492 return vfp_double_normaliseround(dd, &vdm, fpscr, 0, "fuito");
493 }
494
495 static u32 vfp_double_fsito(int dd, int unused, int dm, u32 fpscr)
496 {
497 struct vfp_double vdm;
498 u32 m = vfp_get_float(dm);
499
500 vdm.sign = (m & 0x80000000) >> 16;
501 vdm.exponent = 1023 + 63 - 1;
502 vdm.significand = vdm.sign ? -m : m;
503
504 return vfp_double_normaliseround(dd, &vdm, fpscr, 0, "fsito");
505 }
506
507 static u32 vfp_double_ftoui(int sd, int unused, int dm, u32 fpscr)
508 {
509 struct vfp_double vdm;
510 u32 d, exceptions = 0;
511 int rmode = fpscr & FPSCR_RMODE_MASK;
512 int tm;
513
514 vfp_double_unpack(&vdm, vfp_get_double(dm));
515
516 /*
517 * Do we have a denormalised number?
518 */
519 tm = vfp_double_type(&vdm);
520 if (tm & VFP_DENORMAL)
521 exceptions |= FPSCR_IDC;
522
523 if (tm & VFP_NAN)
524 vdm.sign = 0;
525
526 if (vdm.exponent >= 1023 + 32) {
527 d = vdm.sign ? 0 : 0xffffffff;
528 exceptions = FPSCR_IOC;
529 } else if (vdm.exponent >= 1023 - 1) {
530 int shift = 1023 + 63 - vdm.exponent;
531 u64 rem, incr = 0;
532
533 /*
534 * 2^0 <= m < 2^32-2^8
535 */
536 d = (vdm.significand << 1) >> shift;
537 rem = vdm.significand << (65 - shift);
538
539 if (rmode == FPSCR_ROUND_NEAREST) {
540 incr = 0x8000000000000000ULL;
541 if ((d & 1) == 0)
542 incr -= 1;
543 } else if (rmode == FPSCR_ROUND_TOZERO) {
544 incr = 0;
545 } else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vdm.sign != 0)) {
546 incr = ~0ULL;
547 }
548
549 if ((rem + incr) < rem) {
550 if (d < 0xffffffff)
551 d += 1;
552 else
553 exceptions |= FPSCR_IOC;
554 }
555
556 if (d && vdm.sign) {
557 d = 0;
558 exceptions |= FPSCR_IOC;
559 } else if (rem)
560 exceptions |= FPSCR_IXC;
561 } else {
562 d = 0;
563 if (vdm.exponent | vdm.significand) {
564 exceptions |= FPSCR_IXC;
565 if (rmode == FPSCR_ROUND_PLUSINF && vdm.sign == 0)
566 d = 1;
567 else if (rmode == FPSCR_ROUND_MINUSINF && vdm.sign) {
568 d = 0;
569 exceptions |= FPSCR_IOC;
570 }
571 }
572 }
573
574 pr_debug("VFP: ftoui: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions);
575
576 vfp_put_float(sd, d);
577
578 return exceptions;
579 }
580
581 static u32 vfp_double_ftouiz(int sd, int unused, int dm, u32 fpscr)
582 {
583 return vfp_double_ftoui(sd, unused, dm, FPSCR_ROUND_TOZERO);
584 }
585
586 static u32 vfp_double_ftosi(int sd, int unused, int dm, u32 fpscr)
587 {
588 struct vfp_double vdm;
589 u32 d, exceptions = 0;
590 int rmode = fpscr & FPSCR_RMODE_MASK;
591
592 vfp_double_unpack(&vdm, vfp_get_double(dm));
593 vfp_double_dump("VDM", &vdm);
594
595 /*
596 * Do we have denormalised number?
597 */
598 if (vfp_double_type(&vdm) & VFP_DENORMAL)
599 exceptions |= FPSCR_IDC;
600
601 if (vdm.exponent >= 1023 + 32) {
602 d = 0x7fffffff;
603 if (vdm.sign)
604 d = ~d;
605 exceptions |= FPSCR_IOC;
606 } else if (vdm.exponent >= 1023 - 1) {
607 int shift = 1023 + 63 - vdm.exponent; /* 58 */
608 u64 rem, incr = 0;
609
610 d = (vdm.significand << 1) >> shift;
611 rem = vdm.significand << (65 - shift);
612
613 if (rmode == FPSCR_ROUND_NEAREST) {
614 incr = 0x8000000000000000ULL;
615 if ((d & 1) == 0)
616 incr -= 1;
617 } else if (rmode == FPSCR_ROUND_TOZERO) {
618 incr = 0;
619 } else if ((rmode == FPSCR_ROUND_PLUSINF) ^ (vdm.sign != 0)) {
620 incr = ~0ULL;
621 }
622
623 if ((rem + incr) < rem && d < 0xffffffff)
624 d += 1;
625 if (d > 0x7fffffff + (vdm.sign != 0)) {
626 d = 0x7fffffff + (vdm.sign != 0);
627 exceptions |= FPSCR_IOC;
628 } else if (rem)
629 exceptions |= FPSCR_IXC;
630
631 if (vdm.sign)
632 d = -d;
633 } else {
634 d = 0;
635 if (vdm.exponent | vdm.significand) {
636 exceptions |= FPSCR_IXC;
637 if (rmode == FPSCR_ROUND_PLUSINF && vdm.sign == 0)
638 d = 1;
639 else if (rmode == FPSCR_ROUND_MINUSINF && vdm.sign)
640 d = -1;
641 }
642 }
643
644 pr_debug("VFP: ftosi: d(s%d)=%08x exceptions=%08x\n", sd, d, exceptions);
645
646 vfp_put_float(sd, (s32)d);
647
648 return exceptions;
649 }
650
651 static u32 vfp_double_ftosiz(int dd, int unused, int dm, u32 fpscr)
652 {
653 return vfp_double_ftosi(dd, unused, dm, FPSCR_ROUND_TOZERO);
654 }
655
656
657 static u32 (* const fop_extfns[32])(int dd, int unused, int dm, u32 fpscr) = {
658 [FEXT_TO_IDX(FEXT_FCPY)] = vfp_double_fcpy,
659 [FEXT_TO_IDX(FEXT_FABS)] = vfp_double_fabs,
660 [FEXT_TO_IDX(FEXT_FNEG)] = vfp_double_fneg,
661 [FEXT_TO_IDX(FEXT_FSQRT)] = vfp_double_fsqrt,
662 [FEXT_TO_IDX(FEXT_FCMP)] = vfp_double_fcmp,
663 [FEXT_TO_IDX(FEXT_FCMPE)] = vfp_double_fcmpe,
664 [FEXT_TO_IDX(FEXT_FCMPZ)] = vfp_double_fcmpz,
665 [FEXT_TO_IDX(FEXT_FCMPEZ)] = vfp_double_fcmpez,
666 [FEXT_TO_IDX(FEXT_FCVT)] = vfp_double_fcvts,
667 [FEXT_TO_IDX(FEXT_FUITO)] = vfp_double_fuito,
668 [FEXT_TO_IDX(FEXT_FSITO)] = vfp_double_fsito,
669 [FEXT_TO_IDX(FEXT_FTOUI)] = vfp_double_ftoui,
670 [FEXT_TO_IDX(FEXT_FTOUIZ)] = vfp_double_ftouiz,
671 [FEXT_TO_IDX(FEXT_FTOSI)] = vfp_double_ftosi,
672 [FEXT_TO_IDX(FEXT_FTOSIZ)] = vfp_double_ftosiz,
673 };
674
675
676
677
678 static u32
679 vfp_double_fadd_nonnumber(struct vfp_double *vdd, struct vfp_double *vdn,
680 struct vfp_double *vdm, u32 fpscr)
681 {
682 struct vfp_double *vdp;
683 u32 exceptions = 0;
684 int tn, tm;
685
686 tn = vfp_double_type(vdn);
687 tm = vfp_double_type(vdm);
688
689 if (tn & tm & VFP_INFINITY) {
690 /*
691 * Two infinities. Are they different signs?
692 */
693 if (vdn->sign ^ vdm->sign) {
694 /*
695 * different signs -> invalid
696 */
697 exceptions = FPSCR_IOC;
698 vdp = &vfp_double_default_qnan;
699 } else {
700 /*
701 * same signs -> valid
702 */
703 vdp = vdn;
704 }
705 } else if (tn & VFP_INFINITY && tm & VFP_NUMBER) {
706 /*
707 * One infinity and one number -> infinity
708 */
709 vdp = vdn;
710 } else {
711 /*
712 * 'n' is a NaN of some type
713 */
714 return vfp_propagate_nan(vdd, vdn, vdm, fpscr);
715 }
716 *vdd = *vdp;
717 return exceptions;
718 }
719
720 static u32
721 vfp_double_add(struct vfp_double *vdd, struct vfp_double *vdn,
722 struct vfp_double *vdm, u32 fpscr)
723 {
724 u32 exp_diff;
725 u64 m_sig;
726
727 if (vdn->significand & (1ULL << 63) ||
728 vdm->significand & (1ULL << 63)) {
729 pr_info("VFP: bad FP values in %s\n", __func__);
730 vfp_double_dump("VDN", vdn);
731 vfp_double_dump("VDM", vdm);
732 }
733
734 /*
735 * Ensure that 'n' is the largest magnitude number. Note that
736 * if 'n' and 'm' have equal exponents, we do not swap them.
737 * This ensures that NaN propagation works correctly.
738 */
739 if (vdn->exponent < vdm->exponent) {
740 struct vfp_double *t = vdn;
741 vdn = vdm;
742 vdm = t;
743 }
744
745 /*
746 * Is 'n' an infinity or a NaN? Note that 'm' may be a number,
747 * infinity or a NaN here.
748 */
749 if (vdn->exponent == 2047)
750 return vfp_double_fadd_nonnumber(vdd, vdn, vdm, fpscr);
751
752 /*
753 * We have two proper numbers, where 'vdn' is the larger magnitude.
754 *
755 * Copy 'n' to 'd' before doing the arithmetic.
756 */
757 *vdd = *vdn;
758
759 /*
760 * Align 'm' with the result.
761 */
762 exp_diff = vdn->exponent - vdm->exponent;
763 m_sig = vfp_shiftright64jamming(vdm->significand, exp_diff);
764
765 /*
766 * If the signs are different, we are really subtracting.
767 */
768 if (vdn->sign ^ vdm->sign) {
769 m_sig = vdn->significand - m_sig;
770 if ((s64)m_sig < 0) {
771 vdd->sign = vfp_sign_negate(vdd->sign);
772 m_sig = -m_sig;
773 } else if (m_sig == 0) {
774 vdd->sign = (fpscr & FPSCR_RMODE_MASK) ==
775 FPSCR_ROUND_MINUSINF ? 0x8000 : 0;
776 }
777 } else {
778 m_sig += vdn->significand;
779 }
780 vdd->significand = m_sig;
781
782 return 0;
783 }
784
785 static u32
786 vfp_double_multiply(struct vfp_double *vdd, struct vfp_double *vdn,
787 struct vfp_double *vdm, u32 fpscr)
788 {
789 vfp_double_dump("VDN", vdn);
790 vfp_double_dump("VDM", vdm);
791
792 /*
793 * Ensure that 'n' is the largest magnitude number. Note that
794 * if 'n' and 'm' have equal exponents, we do not swap them.
795 * This ensures that NaN propagation works correctly.
796 */
797 if (vdn->exponent < vdm->exponent) {
798 struct vfp_double *t = vdn;
799 vdn = vdm;
800 vdm = t;
801 pr_debug("VFP: swapping M <-> N\n");
802 }
803
804 vdd->sign = vdn->sign ^ vdm->sign;
805
806 /*
807 * If 'n' is an infinity or NaN, handle it. 'm' may be anything.
808 */
809 if (vdn->exponent == 2047) {
810 if (vdn->significand || (vdm->exponent == 2047 && vdm->significand))
811 return vfp_propagate_nan(vdd, vdn, vdm, fpscr);
812 if ((vdm->exponent | vdm->significand) == 0) {
813 *vdd = vfp_double_default_qnan;
814 return FPSCR_IOC;
815 }
816 vdd->exponent = vdn->exponent;
817 vdd->significand = 0;
818 return 0;
819 }
820
821 /*
822 * If 'm' is zero, the result is always zero. In this case,
823 * 'n' may be zero or a number, but it doesn't matter which.
824 */
825 if ((vdm->exponent | vdm->significand) == 0) {
826 vdd->exponent = 0;
827 vdd->significand = 0;
828 return 0;
829 }
830
831 /*
832 * We add 2 to the destination exponent for the same reason
833 * as the addition case - though this time we have +1 from
834 * each input operand.
835 */
836 vdd->exponent = vdn->exponent + vdm->exponent - 1023 + 2;
837 vdd->significand = vfp_hi64multiply64(vdn->significand, vdm->significand);
838
839 vfp_double_dump("VDD", vdd);
840 return 0;
841 }
842
843 #define NEG_MULTIPLY (1 << 0)
844 #define NEG_SUBTRACT (1 << 1)
845
846 static u32
847 vfp_double_multiply_accumulate(int dd, int dn, int dm, u32 fpscr, u32 negate, char *func)
848 {
849 struct vfp_double vdd, vdp, vdn, vdm;
850 u32 exceptions;
851
852 vfp_double_unpack(&vdn, vfp_get_double(dn));
853 if (vdn.exponent == 0 && vdn.significand)
854 vfp_double_normalise_denormal(&vdn);
855
856 vfp_double_unpack(&vdm, vfp_get_double(dm));
857 if (vdm.exponent == 0 && vdm.significand)
858 vfp_double_normalise_denormal(&vdm);
859
860 exceptions = vfp_double_multiply(&vdp, &vdn, &vdm, fpscr);
861 if (negate & NEG_MULTIPLY)
862 vdp.sign = vfp_sign_negate(vdp.sign);
863
864 vfp_double_unpack(&vdn, vfp_get_double(dd));
865 if (negate & NEG_SUBTRACT)
866 vdn.sign = vfp_sign_negate(vdn.sign);
867
868 exceptions |= vfp_double_add(&vdd, &vdn, &vdp, fpscr);
869
870 return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, func);
871 }
872
873 /*
874 * Standard operations
875 */
876
877 /*
878 * sd = sd + (sn * sm)
879 */
880 static u32 vfp_double_fmac(int dd, int dn, int dm, u32 fpscr)
881 {
882 return vfp_double_multiply_accumulate(dd, dn, dm, fpscr, 0, "fmac");
883 }
884
885 /*
886 * sd = sd - (sn * sm)
887 */
888 static u32 vfp_double_fnmac(int dd, int dn, int dm, u32 fpscr)
889 {
890 return vfp_double_multiply_accumulate(dd, dn, dm, fpscr, NEG_MULTIPLY, "fnmac");
891 }
892
893 /*
894 * sd = -sd + (sn * sm)
895 */
896 static u32 vfp_double_fmsc(int dd, int dn, int dm, u32 fpscr)
897 {
898 return vfp_double_multiply_accumulate(dd, dn, dm, fpscr, NEG_SUBTRACT, "fmsc");
899 }
900
901 /*
902 * sd = -sd - (sn * sm)
903 */
904 static u32 vfp_double_fnmsc(int dd, int dn, int dm, u32 fpscr)
905 {
906 return vfp_double_multiply_accumulate(dd, dn, dm, fpscr, NEG_SUBTRACT | NEG_MULTIPLY, "fnmsc");
907 }
908
909 /*
910 * sd = sn * sm
911 */
912 static u32 vfp_double_fmul(int dd, int dn, int dm, u32 fpscr)
913 {
914 struct vfp_double vdd, vdn, vdm;
915 u32 exceptions;
916
917 vfp_double_unpack(&vdn, vfp_get_double(dn));
918 if (vdn.exponent == 0 && vdn.significand)
919 vfp_double_normalise_denormal(&vdn);
920
921 vfp_double_unpack(&vdm, vfp_get_double(dm));
922 if (vdm.exponent == 0 && vdm.significand)
923 vfp_double_normalise_denormal(&vdm);
924
925 exceptions = vfp_double_multiply(&vdd, &vdn, &vdm, fpscr);
926 return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fmul");
927 }
928
929 /*
930 * sd = -(sn * sm)
931 */
932 static u32 vfp_double_fnmul(int dd, int dn, int dm, u32 fpscr)
933 {
934 struct vfp_double vdd, vdn, vdm;
935 u32 exceptions;
936
937 vfp_double_unpack(&vdn, vfp_get_double(dn));
938 if (vdn.exponent == 0 && vdn.significand)
939 vfp_double_normalise_denormal(&vdn);
940
941 vfp_double_unpack(&vdm, vfp_get_double(dm));
942 if (vdm.exponent == 0 && vdm.significand)
943 vfp_double_normalise_denormal(&vdm);
944
945 exceptions = vfp_double_multiply(&vdd, &vdn, &vdm, fpscr);
946 vdd.sign = vfp_sign_negate(vdd.sign);
947
948 return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fnmul");
949 }
950
951 /*
952 * sd = sn + sm
953 */
954 static u32 vfp_double_fadd(int dd, int dn, int dm, u32 fpscr)
955 {
956 struct vfp_double vdd, vdn, vdm;
957 u32 exceptions;
958
959 vfp_double_unpack(&vdn, vfp_get_double(dn));
960 if (vdn.exponent == 0 && vdn.significand)
961 vfp_double_normalise_denormal(&vdn);
962
963 vfp_double_unpack(&vdm, vfp_get_double(dm));
964 if (vdm.exponent == 0 && vdm.significand)
965 vfp_double_normalise_denormal(&vdm);
966
967 exceptions = vfp_double_add(&vdd, &vdn, &vdm, fpscr);
968
969 return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fadd");
970 }
971
972 /*
973 * sd = sn - sm
974 */
975 static u32 vfp_double_fsub(int dd, int dn, int dm, u32 fpscr)
976 {
977 struct vfp_double vdd, vdn, vdm;
978 u32 exceptions;
979
980 vfp_double_unpack(&vdn, vfp_get_double(dn));
981 if (vdn.exponent == 0 && vdn.significand)
982 vfp_double_normalise_denormal(&vdn);
983
984 vfp_double_unpack(&vdm, vfp_get_double(dm));
985 if (vdm.exponent == 0 && vdm.significand)
986 vfp_double_normalise_denormal(&vdm);
987
988 /*
989 * Subtraction is like addition, but with a negated operand.
990 */
991 vdm.sign = vfp_sign_negate(vdm.sign);
992
993 exceptions = vfp_double_add(&vdd, &vdn, &vdm, fpscr);
994
995 return vfp_double_normaliseround(dd, &vdd, fpscr, exceptions, "fsub");
996 }
997
998 /*
999 * sd = sn / sm
1000 */
1001 static u32 vfp_double_fdiv(int dd, int dn, int dm, u32 fpscr)
1002 {
1003 struct vfp_double vdd, vdn, vdm;
1004 u32 exceptions = 0;
1005 int tm, tn;
1006
1007 vfp_double_unpack(&vdn, vfp_get_double(dn));
1008 vfp_double_unpack(&vdm, vfp_get_double(dm));
1009
1010 vdd.sign = vdn.sign ^ vdm.sign;
1011
1012 tn = vfp_double_type(&vdn);
1013 tm = vfp_double_type(&vdm);
1014
1015 /*
1016 * Is n a NAN?
1017 */
1018 if (tn & VFP_NAN)
1019 goto vdn_nan;
1020
1021 /*
1022 * Is m a NAN?
1023 */
1024 if (tm & VFP_NAN)
1025 goto vdm_nan;
1026
1027 /*
1028 * If n and m are infinity, the result is invalid
1029 * If n and m are zero, the result is invalid
1030 */
1031 if (tm & tn & (VFP_INFINITY|VFP_ZERO))
1032 goto invalid;
1033
1034 /*
1035 * If n is infinity, the result is infinity
1036 */
1037 if (tn & VFP_INFINITY)
1038 goto infinity;
1039
1040 /*
1041 * If m is zero, raise div0 exceptions
1042 */
1043 if (tm & VFP_ZERO)
1044 goto divzero;
1045
1046 /*
1047 * If m is infinity, or n is zero, the result is zero
1048 */
1049 if (tm & VFP_INFINITY || tn & VFP_ZERO)
1050 goto zero;
1051
1052 if (tn & VFP_DENORMAL)
1053 vfp_double_normalise_denormal(&vdn);
1054 if (tm & VFP_DENORMAL)
1055 vfp_double_normalise_denormal(&vdm);
1056
1057 /*
1058 * Ok, we have two numbers, we can perform division.
1059 */
1060 vdd.exponent = vdn.exponent - vdm.exponent + 1023 - 1;
1061 vdm.significand <<= 1;
1062 if (vdm.significand <= (2 * vdn.significand)) {
1063 vdn.significand >>= 1;
1064 vdd.exponent++;
1065 }
1066 vdd.significand = vfp_estimate_div128to64(vdn.significand, 0, vdm.significand);
1067 if ((vdd.significand & 0x1ff) <= 2) {
1068 u64 termh, terml, remh, reml;
1069 mul64to128(&termh, &terml, vdm.significand, vdd.significand);
1070 sub128(&remh, &reml, vdn.significand, 0, termh, terml);
1071 while ((s64)remh < 0) {
1072 vdd.significand -= 1;
1073 add128(&remh, &reml, remh, reml, 0, vdm.significand);
1074 }
1075 vdd.significand |= (reml != 0);
1076 }
1077 return vfp_double_normaliseround(dd, &vdd, fpscr, 0, "fdiv");
1078
1079 vdn_nan:
1080 exceptions = vfp_propagate_nan(&vdd, &vdn, &vdm, fpscr);
1081 pack:
1082 vfp_put_double(dd, vfp_double_pack(&vdd));
1083 return exceptions;
1084
1085 vdm_nan:
1086 exceptions = vfp_propagate_nan(&vdd, &vdm, &vdn, fpscr);
1087 goto pack;
1088
1089 zero:
1090 vdd.exponent = 0;
1091 vdd.significand = 0;
1092 goto pack;
1093
1094 divzero:
1095 exceptions = FPSCR_DZC;
1096 infinity:
1097 vdd.exponent = 2047;
1098 vdd.significand = 0;
1099 goto pack;
1100
1101 invalid:
1102 vfp_put_double(dd, vfp_double_pack(&vfp_double_default_qnan));
1103 return FPSCR_IOC;
1104 }
1105
1106 static u32 (* const fop_fns[16])(int dd, int dn, int dm, u32 fpscr) = {
1107 [FOP_TO_IDX(FOP_FMAC)] = vfp_double_fmac,
1108 [FOP_TO_IDX(FOP_FNMAC)] = vfp_double_fnmac,
1109 [FOP_TO_IDX(FOP_FMSC)] = vfp_double_fmsc,
1110 [FOP_TO_IDX(FOP_FNMSC)] = vfp_double_fnmsc,
1111 [FOP_TO_IDX(FOP_FMUL)] = vfp_double_fmul,
1112 [FOP_TO_IDX(FOP_FNMUL)] = vfp_double_fnmul,
1113 [FOP_TO_IDX(FOP_FADD)] = vfp_double_fadd,
1114 [FOP_TO_IDX(FOP_FSUB)] = vfp_double_fsub,
1115 [FOP_TO_IDX(FOP_FDIV)] = vfp_double_fdiv,
1116 };
1117
1118 #define FREG_BANK(x) ((x) & 0x0c)
1119 #define FREG_IDX(x) ((x) & 3)
1120
1121 u32 vfp_double_cpdo(u32 inst, u32 fpscr)
1122 {
1123 u32 op = inst & FOP_MASK;
1124 u32 exceptions = 0;
1125 unsigned int dd = vfp_get_sd(inst);
1126 unsigned int dn = vfp_get_sn(inst);
1127 unsigned int dm = vfp_get_sm(inst);
1128 unsigned int vecitr, veclen, vecstride;
1129 u32 (*fop)(int, int, s32, u32);
1130
1131 veclen = fpscr & FPSCR_LENGTH_MASK;
1132 vecstride = (1 + ((fpscr & FPSCR_STRIDE_MASK) == FPSCR_STRIDE_MASK)) * 2;
1133
1134 /*
1135 * If destination bank is zero, vector length is always '1'.
1136 * ARM DDI0100F C5.1.3, C5.3.2.
1137 */
1138 if (FREG_BANK(dd) == 0)
1139 veclen = 0;
1140
1141 pr_debug("VFP: vecstride=%u veclen=%u\n", vecstride,
1142 (veclen >> FPSCR_LENGTH_BIT) + 1);
1143
1144 fop = (op == FOP_EXT) ? fop_extfns[dn] : fop_fns[FOP_TO_IDX(op)];
1145 if (!fop)
1146 goto invalid;
1147
1148 for (vecitr = 0; vecitr <= veclen; vecitr += 1 << FPSCR_LENGTH_BIT) {
1149 u32 except;
1150
1151 if (op == FOP_EXT)
1152 pr_debug("VFP: itr%d (d%u.%u) = op[%u] (d%u.%u)\n",
1153 vecitr >> FPSCR_LENGTH_BIT,
1154 dd >> 1, dd & 1, dn,
1155 dm >> 1, dm & 1);
1156 else
1157 pr_debug("VFP: itr%d (d%u.%u) = (d%u.%u) op[%u] (d%u.%u)\n",
1158 vecitr >> FPSCR_LENGTH_BIT,
1159 dd >> 1, dd & 1,
1160 dn >> 1, dn & 1,
1161 FOP_TO_IDX(op),
1162 dm >> 1, dm & 1);
1163
1164 except = fop(dd, dn, dm, fpscr);
1165 pr_debug("VFP: itr%d: exceptions=%08x\n",
1166 vecitr >> FPSCR_LENGTH_BIT, except);
1167
1168 exceptions |= except;
1169
1170 /*
1171 * This ensures that comparisons only operate on scalars;
1172 * comparisons always return with one FPSCR status bit set.
1173 */
1174 if (except & (FPSCR_N|FPSCR_Z|FPSCR_C|FPSCR_V))
1175 break;
1176
1177 /*
1178 * CHECK: It appears to be undefined whether we stop when
1179 * we encounter an exception. We continue.
1180 */
1181
1182 dd = FREG_BANK(dd) + ((FREG_IDX(dd) + vecstride) & 6);
1183 dn = FREG_BANK(dn) + ((FREG_IDX(dn) + vecstride) & 6);
1184 if (FREG_BANK(dm) != 0)
1185 dm = FREG_BANK(dm) + ((FREG_IDX(dm) + vecstride) & 6);
1186 }
1187 return exceptions;
1188
1189 invalid:
1190 return ~0;
1191 }
linux kernel for loongson