summaryrefslogtreecommitdiffstats
path: root/arch/sparc/math-emu/math_32.c
blob: 5ce8f2f64604a0399e792cada3d9c3026f6a36c0 (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
/*
 * arch/sparc/math-emu/math.c
 *
 * Copyright (C) 1998 Peter Maydell (pmaydell@chiark.greenend.org.uk)
 * Copyright (C) 1997, 1999 Jakub Jelinek (jj@ultra.linux.cz)
 * Copyright (C) 1999 David S. Miller (davem@redhat.com)
 *
 * This is a good place to start if you're trying to understand the
 * emulation code, because it's pretty simple. What we do is
 * essentially analyse the instruction to work out what the operation
 * is and which registers are involved. We then execute the appropriate
 * FXXXX function. [The floating point queue introduces a minor wrinkle;
 * see below...]
 * The fxxxxx.c files each emulate a single insn. They look relatively
 * simple because the complexity is hidden away in an unholy tangle
 * of preprocessor macros.
 *
 * The first layer of macros is single.h, double.h, quad.h. Generally
 * these files define macros for working with floating point numbers
 * of the three IEEE formats. FP_ADD_D(R,A,B) is for adding doubles,
 * for instance. These macros are usually defined as calls to more
 * generic macros (in this case _FP_ADD(D,2,R,X,Y) where the number
 * of machine words required to store the given IEEE format is passed
 * as a parameter. [double.h and co check the number of bits in a word
 * and define FP_ADD_D & co appropriately].
 * The generic macros are defined in op-common.h. This is where all
 * the grotty stuff like handling NaNs is coded. To handle the possible
 * word sizes macros in op-common.h use macros like _FP_FRAC_SLL_##wc()
 * where wc is the 'number of machine words' parameter (here 2).
 * These are defined in the third layer of macros: op-1.h, op-2.h
 * and op-4.h. These handle operations on floating point numbers composed
 * of 1,2 and 4 machine words respectively. [For example, on sparc64
 * doubles are one machine word so macros in double.h eventually use
 * constructs in op-1.h, but on sparc32 they use op-2.h definitions.]
 * soft-fp.h is on the same level as op-common.h, and defines some
 * macros which are independent of both word size and FP format.
 * Finally, sfp-machine.h is the machine dependent part of the
 * code: it defines the word size and what type a word is. It also
 * defines how _FP_MUL_MEAT_t() maps to _FP_MUL_MEAT_n_* : op-n.h
 * provide several possible flavours of multiply algorithm, most
 * of which require that you supply some form of asm or C primitive to
 * do the actual multiply. (such asm primitives should be defined
 * in sfp-machine.h too). udivmodti4.c is the same sort of thing.
 *
 * There may be some errors here because I'm working from a
 * SPARC architecture manual V9, and what I really want is V8...
 * Also, the insns which can generate exceptions seem to be a
 * greater subset of the FPops than for V9 (for example, FCMPED
 * has to be emulated on V8). So I think I'm going to have
 * to emulate them all just to be on the safe side...
 *
 * Emulation routines originate from soft-fp package, which is
 * part of glibc and has appropriate copyrights in it (allegedly).
 *
 * NB: on sparc int == long == 4 bytes, long long == 8 bytes.
 * Most bits of the kernel seem to go for long rather than int,
 * so we follow that practice...
 */

/* TODO:
 * fpsave() saves the FP queue but fpload() doesn't reload it.
 * Therefore when we context switch or change FPU ownership
 * we have to check to see if the queue had anything in it and
 * emulate it if it did. This is going to be a pain.
 */

#include <linux/types.h>
#include <linux/sched.h>
#include <linux/mm.h>
#include <linux/perf_event.h>
#include <asm/uaccess.h>

#include "sfp-util_32.h"
#include <math-emu/soft-fp.h>
#include <math-emu/single.h>
#include <math-emu/double.h>
#include <math-emu/quad.h>

#define FLOATFUNC(x) extern int x(void *,void *,void *)

/* The Vn labels indicate what version of the SPARC architecture gas thinks
 * each insn is. This is from the binutils source :->
 */
/* quadword instructions */
#define FSQRTQ	0x02b		/* v8 */
#define FADDQ	0x043		/* v8 */
#define FSUBQ	0x047		/* v8 */
#define FMULQ	0x04b		/* v8 */
#define FDIVQ	0x04f		/* v8 */
#define FDMULQ	0x06e		/* v8 */
#define FQTOS	0x0c7		/* v8 */
#define FQTOD	0x0cb		/* v8 */
#define FITOQ	0x0cc		/* v8 */
#define FSTOQ	0x0cd		/* v8 */
#define FDTOQ	0x0ce		/* v8 */
#define FQTOI	0x0d3		/* v8 */
#define FCMPQ	0x053		/* v8 */
#define FCMPEQ	0x057		/* v8 */
/* single/double instructions (subnormal): should all work */
#define FSQRTS	0x029		/* v7 */
#define FSQRTD	0x02a		/* v7 */
#define FADDS	0x041		/* v6 */
#define FADDD	0x042		/* v6 */
#define FSUBS	0x045		/* v6 */
#define FSUBD	0x046		/* v6 */
#define FMULS	0x049		/* v6 */
#define FMULD	0x04a		/* v6 */
#define FDIVS	0x04d		/* v6 */
#define FDIVD	0x04e		/* v6 */
#define FSMULD	0x069		/* v6 */
#define FDTOS	0x0c6		/* v6 */
#define FSTOD	0x0c9		/* v6 */
#define FSTOI	0x0d1		/* v6 */
#define FDTOI	0x0d2		/* v6 */
#define FABSS	0x009		/* v6 */
#define FCMPS	0x051		/* v6 */
#define FCMPES	0x055		/* v6 */
#define FCMPD	0x052		/* v6 */
#define FCMPED	0x056		/* v6 */
#define FMOVS	0x001		/* v6 */
#define FNEGS	0x005		/* v6 */
#define FITOS	0x0c4		/* v6 */
#define FITOD	0x0c8		/* v6 */

#define FSR_TEM_SHIFT	23UL
#define FSR_TEM_MASK	(0x1fUL << FSR_TEM_SHIFT)
#define FSR_AEXC_SHIFT	5UL
#define FSR_AEXC_MASK	(0x1fUL << FSR_AEXC_SHIFT)
#define FSR_CEXC_SHIFT	0UL
#define FSR_CEXC_MASK	(0x1fUL << FSR_CEXC_SHIFT)

static int do_one_mathemu(u32 insn, unsigned long *fsr, unsigned long *fregs);

/* Unlike the Sparc64 version (which has a struct fpustate), we
 * pass the taskstruct corresponding to the task which currently owns the
 * FPU. This is partly because we don't have the fpustate struct and
 * partly because the task owning the FPU isn't always current (as is
 * the case for the Sparc64 port). This is probably SMP-related...
 * This function returns 1 if all queued insns were emulated successfully.
 * The test for unimplemented FPop in kernel mode has been moved into
 * kernel/traps.c for simplicity.
 */
int do_mathemu(struct pt_regs *regs, struct task_struct *fpt)
{
	/* regs->pc isn't necessarily the PC at which the offending insn is sitting.
	 * The FPU maintains a queue of FPops which cause traps.
	 * When it hits an instruction that requires that the trapped op succeeded
	 * (usually because it reads a reg. that the trapped op wrote) then it
	 * causes this exception. We need to emulate all the insns on the queue
	 * and then allow the op to proceed.
	 * This code should also handle the case where the trap was precise,
	 * in which case the queue length is zero and regs->pc points at the
	 * single FPop to be emulated. (this case is untested, though :->)
	 * You'll need this case if you want to be able to emulate all FPops
	 * because the FPU either doesn't exist or has been software-disabled.
	 * [The UltraSPARC makes FP a precise trap; this isn't as stupid as it
	 * might sound because the Ultra does funky things with a superscalar
	 * architecture.]
	 */

	/* You wouldn't believe how often I typed 'ftp' when I meant 'fpt' :-> */

	int i;
	int retcode = 0;                               /* assume all succeed */
	unsigned long insn;

	perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS, 1, regs, 0);

#ifdef DEBUG_MATHEMU
	printk("In do_mathemu()... pc is %08lx\n", regs->pc);
	printk("fpqdepth is %ld\n", fpt->thread.fpqdepth);
	for (i = 0; i < fpt->thread.fpqdepth; i++)
		printk("%d: %08lx at %08lx\n", i, fpt->thread.fpqueue[i].insn,
		       (unsigned long)fpt->thread.fpqueue[i].insn_addr);
#endif

	if (fpt->thread.fpqdepth == 0) {                   /* no queue, guilty insn is at regs->pc */
#ifdef DEBUG_MATHEMU
		printk("precise trap at %08lx\n", regs->pc);
#endif
		if (!get_user(insn, (u32 __user *) regs->pc)) {
			retcode = do_one_mathemu(insn, &fpt->thread.fsr, fpt->thread.float_regs);
			if (retcode) {
				/* in this case we need to fix up PC & nPC */
				regs->pc = regs->npc;
				regs->npc += 4;
			}
		}
		return retcode;
	}

	/* Normal case: need to empty the queue... */
	for (i = 0; i < fpt->thread.fpqdepth; i++) {
		retcode = do_one_mathemu(fpt->thread.fpqueue[i].insn, &(fpt->thread.fsr), fpt->thread.float_regs);
		if (!retcode)                               /* insn failed, no point doing any more */
			break;
	}
	/* Now empty the queue and clear the queue_not_empty flag */
	if (retcode)
		fpt->thread.fsr &= ~(0x3000 | FSR_CEXC_MASK);
	else
		fpt->thread.fsr &= ~0x3000;
	fpt->thread.fpqdepth = 0;

	return retcode;
}

/* All routines returning an exception to raise should detect
 * such exceptions _before_ rounding to be consistent with
 * the behavior of the hardware in the implemented cases
 * (and thus with the recommendations in the V9 architecture
 * manual).
 *
 * We return 0 if a SIGFPE should be sent, 1 otherwise.
 */
static inline int record_exception(unsigned long *pfsr, int eflag)
{
	unsigned long fsr = *pfsr;
	int would_trap;

	/* Determine if this exception would have generated a trap. */
	would_trap = (fsr & ((long)eflag << FSR_TEM_SHIFT)) != 0UL;

	/* If trapping, we only want to signal one bit. */
	if (would_trap != 0) {
		eflag &= ((fsr & FSR_TEM_MASK) >> FSR_TEM_SHIFT);
		if ((eflag & (eflag - 1)) != 0) {
			if (eflag & FP_EX_INVALID)
				eflag = FP_EX_INVALID;
			else if (eflag & FP_EX_OVERFLOW)
				eflag = FP_EX_OVERFLOW;
			else if (eflag & FP_EX_UNDERFLOW)
				eflag = FP_EX_UNDERFLOW;
			else if (eflag & FP_EX_DIVZERO)
				eflag = FP_EX_DIVZERO;
			else if (eflag & FP_EX_INEXACT)
				eflag = FP_EX_INEXACT;
		}
	}

	/* Set CEXC, here is the rule:
	 *
	 *    In general all FPU ops will set one and only one
	 *    bit in the CEXC field, this is always the case
	 *    when the IEEE exception trap is enabled in TEM.
	 */
	fsr &= ~(FSR_CEXC_MASK);
	fsr |= ((long)eflag << FSR_CEXC_SHIFT);

	/* Set the AEXC field, rule is:
	 *
	 *    If a trap would not be generated, the
	 *    CEXC just generated is OR'd into the
	 *    existing value of AEXC.
	 */
	if (would_trap == 0)
		fsr |= ((long)eflag << FSR_AEXC_SHIFT);

	/* If trapping, indicate fault trap type IEEE. */
	if (would_trap != 0)
		fsr |= (1UL << 14);

	*pfsr = fsr;

	return (would_trap ? 0 : 1);
}

typedef union {
	u32 s;
	u64 d;
	u64 q[2];
} *argp;

static int do_one_mathemu(u32 insn, unsigned long *pfsr, unsigned long *fregs)
{
	/* Emulate the given insn, updating fsr and fregs appropriately. */
	int type = 0;
	/* r is rd, b is rs2 and a is rs1. The *u arg tells
	   whether the argument should be packed/unpacked (0 - do not unpack/pack, 1 - unpack/pack)
	   non-u args tells the size of the argument (0 - no argument, 1 - single, 2 - double, 3 - quad */
#define TYPE(dummy, r, ru, b, bu, a, au) type = (au << 2) | (a << 0) | (bu << 5) | (b << 3) | (ru << 8) | (r << 6)
	int freg;
	argp rs1 = NULL, rs2 = NULL, rd = NULL;
	FP_DECL_EX;
	FP_DECL_S(SA); FP_DECL_S(SB); FP_DECL_S(SR);
	FP_DECL_D(DA); FP_DECL_D(DB); FP_DECL_D(DR);
	FP_DECL_Q(QA); FP_DECL_Q(QB); FP_DECL_Q(QR);
	int IR;
	long fsr;

#ifdef DEBUG_MATHEMU
	printk("In do_mathemu(), emulating %08lx\n", insn);
#endif

	if ((insn & 0xc1f80000) == 0x81a00000)	/* FPOP1 */ {
		switch ((insn >> 5) & 0x1ff) {
		case FSQRTQ: TYPE(3,3,1,3,1,0,0); break;
		case FADDQ:
		case FSUBQ:
		case FMULQ:
		case FDIVQ: TYPE(3,3,1,3,1,3,1); break;
		case FDMULQ: TYPE(3,3,1,2,1,2,1); break;
		case FQTOS: TYPE(3,1,1,3,1,0,0); break;
		case FQTOD: TYPE(3,2,1,3,1,0,0); break;
		case FITOQ: TYPE(3,3,1,1,0,0,0); break;
		case FSTOQ: TYPE(3,3,1,1,1,0,0); break;
		case FDTOQ: TYPE(3,3,1,2,1,0,0); break;
		case FQTOI: TYPE(3,1,0,3,1,0,0); break;
		case FSQRTS: TYPE(2,1,1,1,1,0,0); break;
		case FSQRTD: TYPE(2,2,1,2,1,0,0); break;
		case FADDD:
		case FSUBD:
		case FMULD:
		case FDIVD: TYPE(2,2,1,2,1,2,1); break;
		case FADDS:
		case FSUBS:
		case FMULS:
		case FDIVS: TYPE(2,1,1,1,1,1,1); break;
		case FSMULD: TYPE(2,2,1,1,1,1,1); break;
		case FDTOS: TYPE(2,1,1,2,1,0,0); break;
		case FSTOD: TYPE(2,2,1,1,1,0,0); break;
		case FSTOI: TYPE(2,1,0,1,1,0,0); break;
		case FDTOI: TYPE(2,1,0,2,1,0,0); break;
		case FITOS: TYPE(2,1,1,1,0,0,0); break;
		case FITOD: TYPE(2,2,1,1,0,0,0); break;
		case FMOVS:
		case FABSS:
		case FNEGS: TYPE(2,1,0,1,0,0,0); break;
		}
	} else if ((insn & 0xc1f80000) == 0x81a80000)	/* FPOP2 */ {
		switch ((insn >> 5) & 0x1ff) {
		case FCMPS: TYPE(3,0,0,1,1,1,1); break;
		case FCMPES: TYPE(3,0,0,1,1,1,1); break;
		case FCMPD: TYPE(3,0,0,2,1,2,1); break;
		case FCMPED: TYPE(3,0,0,2,1,2,1); break;
		case FCMPQ: TYPE(3,0,0,3,1,3,1); break;
		case FCMPEQ: TYPE(3,0,0,3,1,3,1); break;
		}
	}

	if (!type) {	/* oops, didn't recognise that FPop */
#ifdef DEBUG_MATHEMU
		printk("attempt to emulate unrecognised FPop!\n");
#endif
		return 0;
	}

	/* Decode the registers to be used */
	freg = (*pfsr >> 14) & 0xf;

	*pfsr &= ~0x1c000;				/* clear the traptype bits */
	
	freg = ((insn >> 14) & 0x1f);
	switch (type & 0x3) {				/* is rs1 single, double or quad? */
	case 3:
		if (freg & 3) {				/* quadwords must have bits 4&5 of the */
							/* encoded reg. number set to zero. */
			*pfsr |= (6 << 14);
			return 0;			/* simulate invalid_fp_register exception */
		}
	/* fall through */
	case 2:
		if (freg & 1) {				/* doublewords must have bit 5 zeroed */
			*pfsr |= (6 << 14);
			return 0;
		}
	}
	rs1 = (argp)&fregs[freg];
	switch (type & 0x7) {
	case 7: FP_UNPACK_QP (QA, rs1); break;
	case 6: FP_UNPACK_DP (DA, rs1); break;
	case 5: FP_UNPACK_SP (SA, rs1); break;
	}
	freg = (insn & 0x1f);
	switch ((type >> 3) & 0x3) {			/* same again for rs2 */
	case 3:
		if (freg & 3) {				/* quadwords must have bits 4&5 of the */
							/* encoded reg. number set to zero. */
			*pfsr |= (6 << 14);
			return 0;			/* simulate invalid_fp_register exception */
		}
	/* fall through */
	case 2:
		if (freg & 1) {				/* doublewords must have bit 5 zeroed */
			*pfsr |= (6 << 14);
			return 0;
		}
	}
	rs2 = (argp)&fregs[freg];
	switch ((type >> 3) & 0x7) {
	case 7: FP_UNPACK_QP (QB, rs2); break;
	case 6: FP_UNPACK_DP (DB, rs2); break;
	case 5: FP_UNPACK_SP (SB, rs2); break;
	}
	freg = ((insn >> 25) & 0x1f);
	switch ((type >> 6) & 0x3) {			/* and finally rd. This one's a bit different */
	case 0:						/* dest is fcc. (this must be FCMPQ or FCMPEQ) */
		if (freg) {				/* V8 has only one set of condition codes, so */
							/* anything but 0 in the rd field is an error */
			*pfsr |= (6 << 14);		/* (should probably flag as invalid opcode */
			return 0;			/* but SIGFPE will do :-> ) */
		}
		break;
	case 3:
		if (freg & 3) {				/* quadwords must have bits 4&5 of the */
							/* encoded reg. number set to zero. */
			*pfsr |= (6 << 14);
			return 0;			/* simulate invalid_fp_register exception */
		}
	/* fall through */
	case 2:
		if (freg & 1) {				/* doublewords must have bit 5 zeroed */
			*pfsr |= (6 << 14);
			return 0;
		}
	/* fall through */
	case 1:
		rd = (void *)&fregs[freg];
		break;
	}
#ifdef DEBUG_MATHEMU
	printk("executing insn...\n");
#endif
	/* do the Right Thing */
	switch ((insn >> 5) & 0x1ff) {
	/* + */
	case FADDS: FP_ADD_S (SR, SA, SB); break;
	case FADDD: FP_ADD_D (DR, DA, DB); break;
	case FADDQ: FP_ADD_Q (QR, QA, QB); break;
	/* - */
	case FSUBS: FP_SUB_S (SR, SA, SB); break;
	case FSUBD: FP_SUB_D (DR, DA, DB); break;
	case FSUBQ: FP_SUB_Q (QR, QA, QB); break;
	/* * */
	case FMULS: FP_MUL_S (SR, SA, SB); break;
	case FSMULD: FP_CONV (D, S, 2, 1, DA, SA);
		     FP_CONV (D, S, 2, 1, DB, SB);
	case FMULD: FP_MUL_D (DR, DA, DB); break;
	case FDMULQ: FP_CONV (Q, D, 4, 2, QA, DA);
		     FP_CONV (Q, D, 4, 2, QB, DB);
	case FMULQ: FP_MUL_Q (QR, QA, QB); break;
	/* / */
	case FDIVS: FP_DIV_S (SR, SA, SB); break;
	case FDIVD: FP_DIV_D (DR, DA, DB); break;
	case FDIVQ: FP_DIV_Q (QR, QA, QB); break;
	/* sqrt */
	case FSQRTS: FP_SQRT_S (SR, SB); break;
	case FSQRTD: FP_SQRT_D (DR, DB); break;
	case FSQRTQ: FP_SQRT_Q (QR, QB); break;
	/* mov */
	case FMOVS: rd->s = rs2->s; break;
	case FABSS: rd->s = rs2->s & 0x7fffffff; break;
	case FNEGS: rd->s = rs2->s ^ 0x80000000; break;
	/* float to int */
	case FSTOI: FP_TO_INT_S (IR, SB, 32, 1); break;
	case FDTOI: FP_TO_INT_D (IR, DB, 32, 1); break;
	case FQTOI: FP_TO_INT_Q (IR, QB, 32, 1); break;
	/* int to float */
	case FITOS: IR = rs2->s; FP_FROM_INT_S (SR, IR, 32, int); break;
	case FITOD: IR = rs2->s; FP_FROM_INT_D (DR, IR, 32, int); break;
	case FITOQ: IR = rs2->s; FP_FROM_INT_Q (QR, IR, 32, int); break;
	/* float to float */
	case FSTOD: FP_CONV (D, S, 2, 1, DR, SB); break;
	case FSTOQ: FP_CONV (Q, S, 4, 1, QR, SB); break;
	case FDTOQ: FP_CONV (Q, D, 4, 2, QR, DB); break;
	case FDTOS: FP_CONV (S, D, 1, 2, SR, DB); break;
	case FQTOS: FP_CONV (S, Q, 1, 4, SR, QB); break;
	case FQTOD: FP_CONV (D, Q, 2, 4, DR, QB); break;
	/* comparison */
	case FCMPS:
	case FCMPES:
		FP_CMP_S(IR, SB, SA, 3);
		if (IR == 3 &&
		    (((insn >> 5) & 0x1ff) == FCMPES ||
		     FP_ISSIGNAN_S(SA) ||
		     FP_ISSIGNAN_S(SB)))
			FP_SET_EXCEPTION (FP_EX_INVALID);
		break;
	case FCMPD:
	case FCMPED:
		FP_CMP_D(IR, DB, DA, 3);
		if (IR == 3 &&
		    (((insn >> 5) & 0x1ff) == FCMPED ||
		     FP_ISSIGNAN_D(DA) ||
		     FP_ISSIGNAN_D(DB)))
			FP_SET_EXCEPTION (FP_EX_INVALID);
		break;
	case FCMPQ:
	case FCMPEQ:
		FP_CMP_Q(IR, QB, QA, 3);
		if (IR == 3 &&
		    (((insn >> 5) & 0x1ff) == FCMPEQ ||
		     FP_ISSIGNAN_Q(QA) ||
		     FP_ISSIGNAN_Q(QB)))
			FP_SET_EXCEPTION (FP_EX_INVALID);
	}
	if (!FP_INHIBIT_RESULTS) {
		switch ((type >> 6) & 0x7) {
		case 0: fsr = *pfsr;
			if (IR == -1) IR = 2;
			/* fcc is always fcc0 */
			fsr &= ~0xc00; fsr |= (IR << 10);
			*pfsr = fsr;
			break;
		case 1: rd->s = IR; break;
		case 5: FP_PACK_SP (rd, SR); break;
		case 6: FP_PACK_DP (rd, DR); break;
		case 7: FP_PACK_QP (rd, QR); break;
		}
	}
	if (_fex == 0)
		return 1;				/* success! */
	return record_exception(pfsr, _fex);
}