Loading...
  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
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
/*
 * Copyright (c) 2007 Apple Inc. All rights reserved.
 *
 * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
 * 
 * This file contains Original Code and/or Modifications of Original Code
 * as defined in and that are subject to the Apple Public Source License
 * Version 2.0 (the 'License'). You may not use this file except in
 * compliance with the License. The rights granted to you under the License
 * may not be used to create, or enable the creation or redistribution of,
 * unlawful or unlicensed copies of an Apple operating system, or to
 * circumvent, violate, or enable the circumvention or violation of, any
 * terms of an Apple operating system software license agreement.
 * 
 * Please obtain a copy of the License at
 * http://www.opensource.apple.com/apsl/ and read it before using this file.
 * 
 * The Original Code and all software distributed under the License are
 * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
 * Please see the License for the specific language governing rights and
 * limitations under the License.
 * 
 * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
 */

/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */

/*
 * Copyright 2008 Sun Microsystems, Inc.  All rights reserved.
 * Use is subject to license terms.
 */

/*
 * #pragma ident	"@(#)fasttrap_isa.c	1.27	08/04/09 SMI"
 */

#ifdef KERNEL
#ifndef _KERNEL
#define _KERNEL /* Solaris vs. Darwin */
#endif
#endif

#define MACH__POSIX_C_SOURCE_PRIVATE 1 /* pulls in suitable savearea from mach/ppc/thread_status.h */
#include <sys/fasttrap_isa.h>
#include <sys/fasttrap_impl.h>
#include <sys/dtrace.h>
#include <sys/dtrace_impl.h>
#include <sys/dtrace_ptss.h>
#include <kern/debug.h>
#include <ppc/decodePPC.h>
#include <kern/task.h>
#include <mach/vm_param.h>
#include <mach/mach_vm.h>
#include <mach/task.h>
#include <vm/pmap.h>
#include <vm/vm_map.h> /* All the bits we care about are guarded by MACH_KERNEL_PRIVATE :-( */
extern dtrace_id_t dtrace_probeid_error;

/* Solaris proc_t is the struct. Darwin's proc_t is a pointer to it. */
#define proc_t struct proc /* Steer clear of the Darwin typedef for proc_t */

static int32_t branchtaken(int32_t bo, int32_t bi, ppc_saved_state_t *sv);
static int32_t dtrace_decode_ppc(uint32_t inst);
int patchInst(task_t task, addr64_t vaddr, uint32_t inst);
kern_return_t dtrace_user_probe(ppc_saved_state_t *sv);

/*
 * Lossless User-Land Tracing on PPC
 * ---------------------------------
 *
 * PPC uses a different technique to emulate user-land instruction replaces by a probe
 * trap than x86.
 *
 * Like x86, it will emulate all forms of branch instructions.  We will not attempt
 * to emulate any instruction that we know will cause an interruption or exception
 * (system call, trap, privileged instruction, instruction that uses a privileged
 * register).
 *
 * NOTE: I am thinking that we should punish tight loopers, e.g., branch-to-dot.
 * Depending upon clock resolution and how fast we can process these guys, it is
 * possible that its quantum will never decrease.  Maybe we could just manually
 * end the guy's quantum and let the next guy go...
 *
 * When fasttrap_tracepoint_init is called, we fetch the instruction and decode it.
 * If we don't recognize it or find it is a "banned" instruction, we return -1,
 * telling our caller to forget it.  Otherwise we save the instruction image and
 * enough of the decode to quickly handle it at probe time.  We cram it into
 * the fasttrap_machtp_t structure.
 *
 * When the probe hits, we verify that the PC is still a probe point and if not,
 * we bail.  Otherwise we have a bit more to do.
 *
 * If DTFTP_ENTRY is set, we have an entry probe and need to call dtrace_probe.
 *
 * If DTFTP_IS_ENABLED is set, all we need to do is to return a 1.
 *
 * If ftp_argmap is NULL, we call dtrace_probe
 *
 * Otherwise, we figure out what the arguments are and pass them to dtrace_probe
 *
 * Next, we need to set up to emulate the probed instruction and here is where we are
 * the most different than the x86 code.
 *
 * Like x86, we first check to see if the instruction is any form of branch.  If so, 
 * we emulate it completely within the kernel and are done.
 *
 * If it is anything else, we build a code stream within the kernel to execute the
 * instruction.  Note that this is very different from x86 which build the code in
 * userland.
 *
 * The generated stream needs to be executed within the kernel's code space but with
 * the user address space and registers.  Because PPC allows different translation modes
 * for instruction fetch and data fetch, this is not too difficult.
 *
 * There are two kinds streams needed: execute and continue, and execute and return,
 * which are used for entry/offset and exit probes respectivily. 
 *
 * The probe code will copy the instruction image into the current user savearea (which
 * also contains the complete user state register context).  A flag that requests either
 * execute/continue or execute/return is also set in the savearea.
 *
 * We now exit the dtrace code and the marked context makes its way back to the point
 * where it will be dispatched on the processor.
 *
 * The exception return code will start to restore the user context, including registers
 * and address space.  However, before dispatching the user, it will notice that the
 * emulate flags are set.  At this point the code will build a code stream 
 * in an area in the per_proc that consists of
 * the original instruction followed by a trap instruction.  It will set the new MSR (in
 * SRR1) to have address translation enable for data, translation disabled for instruction
 * fetches, interruptions disabled, and supervisor state.
 *
 * The new PC and MSR are loaded via a RFID and the generated stream is executed. If a
 * synchronous fault occurs, it is either handled (PTE miss, FPU or vector unavailable),
 * emulated (alignment or denorm), or passed on to the user.
 *
 * Assuming the emulated instruction completes, the trap will execute.  When that happens, 
 * low-level trap handler will check its flags.  If the trap corresponds to an
 * execute/continue stream, the trap handler will adjust the PC and complete the
 * transition into user space. 
 *
 * If the trap corresponds to an execute/return stream, the handler will generate 
 * a T_DTRACE_RET exception and let the trap handler pass it along to dtrace_user_probe.
 *
 */


static uint64_t
fasttrap_anarg(ppc_saved_state_t *sv, int function_entry, int argno)
{
#pragma unused(function_entry)
	uint32_t farg;
 	uint64_t value;
 	
 	/* The first 8 arguments (argno 0-7) are in registers */
 	if (argno < 8) {
 		value = (&sv->save_r3)[argno];
 	} else {
 		if (sv->save_srr1 & 0x8000000000000000ULL) {
 			/* 64-bit */
 			/* Grab argument >= 8 from stack */
 			fasttrap_fuword64_noerr(sv->save_r1 + 48 + ((argno)* sizeof(uint64_t)), &value);
 		} else {
 			/* 32-bit */
			/* Grab argument >= 8 from stack */
 			fasttrap_fuword32_noerr(sv->save_r1 + 24 + ((argno) * sizeof(uint32_t)), &farg);
			value = (uint64_t)farg;
 		}
 	}
 	
 	return (value);
}

/*ARGSUSED*/
int
fasttrap_tracepoint_init(proc_t *p, fasttrap_tracepoint_t *tp, user_addr_t pc,
    fasttrap_probe_type_t type)
{
#pragma unused(type)

	uint32_t instr, testr1, testr2, testr3;
	user_addr_t targpc;
	int32_t target, optype;

	/*
	 * Read the instruction at the given address out of the process's
	 * address space. We don't have to worry about a debugger
	 * changing this instruction before we overwrite it with our trap
	 * instruction since P_PR_LOCK is set. Since instructions can span
	 * pages, we potentially read the instruction in two parts. If the
	 * second part fails, we just zero out that part of the instruction.
	 */
	/*
	 * APPLE NOTE: Of course, we do not have a P_PR_LOCK, so this is racey...
	 */

	if (uread(p, &instr, 4, pc) != 0) return (-1);	/* Grab instruction, return suddenly if read fails... */
		
	optype = dtrace_decode_ppc(instr);		/* See if we have an instruction we can probe */
	
	tp->ftt_instr = instr;					/* Save the instruction image */
	testr1 = tp->ftt_bo = (uint8_t)((instr >> (31 - 10)) & 0x1F);	/* Extract branch options */
	testr2 = tp->ftt_bi = (uint8_t)((instr >> (31 - 15)) & 0x1F);	/* Extract condition register bit */
	testr3 = (instr >> (31 - 20)) & 0x1F;	/* Get that last register */
	tp->ftt_flgs = (uint8_t)(instr & 3);	/* Set the absolute address and link flags */

	switch(optype) {						/* Do instruction specific decode */
		
		case diCMN:							/* Common instruction */
			tp->ftt_type = ftmtCommon;		/* Mark as common instruction */
			break;
			
		case diINV: 						/* Invalid */
		case diTRP:							/* Trap */
		case diSC:							/* System Call */
		case diRFI: 						/* Return from interrupt */
		case diPRV:							/* Priviliged instruction */
			return (-1);					/* We will not emulate these... */
			break;
		
		case diB:							/* Branch */
			tp->ftt_type = ftmtB;			/* Mark as branch instruction */
			target = instr & 0x03FFFFFC;	/* Extract address or offset */
			if(target & 0x02000000) target |= 0xFC000000;	/* Sign extend */
			tp->ftt_trgt = target;			/* Trim back down and save */
			
			targpc = (user_addr_t)((int64_t)target);	/* Generate a target address, hopefully we sign extend... */
			if(!(tp->ftt_flgs & ftmtAbs)) {	/* Are we dealing with an offset here? */
				targpc = targpc + pc;		/* Apply offset to get target address */
			}
			
			if(targpc == pc) return -1;		/* Branching to self is a sin and is forbidden... */
			break;
			
		case diBC:							/* Branch conditional */
			tp->ftt_type = ftmtBC;			/* Mark as branch conditional */
			target = instr & 0x0000FFFC;	/* Extract address or offset */
			if(target & 0x00008000) target |= 0xFFFF0000;	/* Sign extend */
			tp->ftt_trgt = target;			/* Trim back down and save */
			
			targpc = (user_addr_t)((int64_t)target);	/* Generate a target address, hopefully we sign extend... */
			if(!(tp->ftt_flgs & ftmtAbs)) {		/* Are we dealing with an offset here? */
				targpc = targpc + pc;		/* Apply offset to get target address */
			}
			
			if(targpc == pc) return -1;		/* Branching to self is a sin and is forbidden... */
			break;
			
		case diBLR:							/* Branch conditional to link register */
			tp->ftt_type = ftmtBLR;			/* Mark as branch conditional to link register */
			break;
			
		case diBCTR:						/* Branch conditional to count register */
			tp->ftt_type = ftmtBCTR;		/* Mark as branch conditional to count register */
			break;
			
		case diOR:							/* OR */
			if((instr >> 26) == 24) {		/* Is this the ORI nop? */
				if((testr1 == testr2) && ((instr & 0x0000FFFF) == 0)) tp->ftt_type = ftmtNOP;	/* Remember if this is a NOP instruction */
				else tp->ftt_type = ftmtCommon;	/* Otherwise it is a common ORI instruction */
			}
			else if((testr1 == testr2) && (testr1 == testr3)) tp->ftt_type = ftmtNOP;	/* If all three registers are the same, this is a NOP */
			else tp->ftt_type = ftmtCommon;	/* Otherwise it is a common OR instruction */

			break;
			
		default:
			panic("fasttrap_tracepoint_init: invalid branch decode, inst = %08X, optype = %d\n", instr, optype);
			break;
			
	}

	return (0);
}

int
fasttrap_tracepoint_install(proc_t *p, fasttrap_tracepoint_t *tp)
{
	return patchInst(p->task, tp->ftt_pc, FASTTRAP_INSTR);	/* Patch the instruction and flush it */
}

extern void dbgTrace(uint32_t, uint32_t, uint32_t, uint32_t, uint32_t);

int
fasttrap_tracepoint_remove(proc_t *p, fasttrap_tracepoint_t *tp)
{
	uint32_t instr;

	/*
	 * Distinguish between read or write failures and a changed
	 * instruction.
	 */
	if (uread(p, &instr, 4, tp->ftt_pc) != 0) return (0);	/* Get the instruction, but exit if not mapped */

//	dbgTrace(0x99999999, (uint32_t)tp->ftt_pc, tp->ftt_instr, instr, 0);	/* (TRACE/DEBUG) */

	if (instr != FASTTRAP_INSTR) return (0);	/* Did someone change it? If so, just leave */

	return patchInst(p->task, tp->ftt_pc, tp->ftt_instr);	/* Patch the old instruction back in and flush it */
}

static void
fasttrap_return_common(ppc_saved_state_t *sv, user_addr_t pc, pid_t pid, user_addr_t new_pc)
{

	fasttrap_tracepoint_t *tp;
	fasttrap_bucket_t *bucket;
	fasttrap_id_t *id;
	lck_mtx_t *pid_mtx;

	pid_mtx = &cpu_core[CPU->cpu_id].cpuc_pid_lock;
	lck_mtx_lock(pid_mtx);
	bucket = &fasttrap_tpoints.fth_table[FASTTRAP_TPOINTS_INDEX(pid, pc)];

	for (tp = bucket->ftb_data; tp != NULL; tp = tp->ftt_next) {
		if (pid == tp->ftt_pid && pc == tp->ftt_pc &&
		    tp->ftt_proc->ftpc_acount != 0)
			break;
	}

	/*
	 * Don't sweat it if we can't find the tracepoint again. Unlike
	 * when we're in fasttrap_pid_probe(), finding the tracepoint here
	 * is not essential to the correct execution of the process.
	 */
	if (tp == NULL) {
		lck_mtx_unlock(pid_mtx);
		return;
	}

	for (id = tp->ftt_retids; id != NULL; id = id->fti_next) {
		/*
		 * If there's a branch that could act as a return site, we
		 * need to trace it, and check here if the program counter is
		 * external to the function.
		 */
		if((new_pc - id->fti_probe->ftp_faddr) < id->fti_probe->ftp_fsize)	/* Is target within the function? */
			continue;							/* Yeah, skip this one... */

		DTRACE_CPUFLAG_SET(CPU_DTRACE_USTACK_FP);
		if (ISSET(current_proc()->p_lflag, P_LNOATTACH)) {
			dtrace_probe(dtrace_probeid_error, 0 /* state */, 
				     id->fti_probe->ftp_id, 1 /* ndx */, -1 /* offset */, 
				     DTRACEFLT_UPRIV);
		} else {
			dtrace_probe(id->fti_probe->ftp_id,
				pc - id->fti_probe->ftp_faddr,
				sv->save_r3, sv->save_r4, 0, 0);
		}
		DTRACE_CPUFLAG_CLEAR(CPU_DTRACE_USTACK_FP);
	}

	lck_mtx_unlock(pid_mtx);
}

static void
fasttrap_usdt_args(fasttrap_probe_t *probe, ppc_saved_state_t *sv, int argc,
    uint64_t *argv)
{
	int i, x, cap = MIN(argc, probe->ftp_nargs);
	uint32_t farg;

	for (i = 0; i < cap; i++) {
		x = probe->ftp_argmap[i];

		if (x <= 8) {							/* Is this argument in a register? */
			argv[i] = (&sv->save_r0)[x];
		} else {
			if(sv->save_srr1 & 0x8000000000000000ULL) {	/* Are we running in 64-bit? */
				fasttrap_fuword64_noerr(sv->save_r1 + 48 + (x * sizeof(uint64_t)), &argv[i]);	/* Grab argument > 8 from stack */
			}
			else {
				fasttrap_fuword32_noerr(sv->save_r1 + 24 + (x * sizeof(uint32_t)), &farg);	/* Grab argument > 8 from stack */
				argv[i] = (uint64_t)farg;		/* Convert to 64-bit */
			}
		}
	}

	for (; i < argc; i++) {
		argv[i] = 0;
	}
}

int
fasttrap_pid_probe(ppc_saved_state_t *sv)
{
	proc_t *p = current_proc();
	fasttrap_bucket_t *bucket;
	lck_mtx_t *pid_mtx;
	fasttrap_tracepoint_t *tp, tp_local;
	pid_t pid;
	dtrace_icookie_t cookie;
	uint_t is_enabled = 0;
	user_addr_t new_pc = 0;
	user_addr_t pc;
	user_addr_t addrmask;

	pc = sv->save_srr0;							/* Remember the PC for later */
	if(sv->save_srr1 & 0x8000000000000000ULL) addrmask = 0xFFFFFFFFFFFFFFFFULL;	/* Set 64-bit addressing if enabled */
	else addrmask = 0x00000000FFFFFFFFULL;		/* Otherwise set 32-bit */

	uthread_t uthread = (uthread_t)get_bsdthread_info(current_thread());

	/*
	 * Clear all user tracing flags.
	 */
	uthread->t_dtrace_ft = 0;

	/*
	 * Treat a child created by a call to vfork(2) as if it were its
	 * parent. We know that there's only one thread of control in such a
	 * process: this one.
	 */
	/*
	 * APPLE NOTE: Terry says: "You need to hold the process locks (currently: kernel funnel) for this traversal"
	 * FIXME: How do we assert this?
	 */
	while (p->p_lflag & P_LINVFORK) p = p->p_pptr;	/* Search the end */

	pid = p->p_pid;
	pid_mtx = &cpu_core[CPU->cpu_id].cpuc_pid_lock;
	lck_mtx_lock(pid_mtx);
	bucket = &fasttrap_tpoints.fth_table[FASTTRAP_TPOINTS_INDEX(pid, sv->save_srr0)];	/* Get the bucket that corresponds to out PC */

	/*
	 * Lookup the tracepoint that the process just hit.
	 */
	for (tp = bucket->ftb_data; tp != NULL; tp = tp->ftt_next) {
		if (pid == tp->ftt_pid && (sv->save_srr0 == tp->ftt_pc) &&
		    tp->ftt_proc->ftpc_acount != 0)
			break;
	}

	/*
	 * If we couldn't find a matching tracepoint, either a tracepoint has
	 * been inserted without using the pid<pid> ioctl interface (see
	 * fasttrap_ioctl), or somehow we have mislaid this tracepoint.
	 */
	if (tp == NULL) {
		lck_mtx_unlock(pid_mtx);
		return (-1);
	}

	if (tp->ftt_ids != NULL) {
		fasttrap_id_t *id;
		
		for (id = tp->ftt_ids; id != NULL; id = id->fti_next) {
			fasttrap_probe_t *probe = id->fti_probe;
			
			if (ISSET(current_proc()->p_lflag, P_LNOATTACH)) {
				dtrace_probe(dtrace_probeid_error, 0 /* state */, 
				     id->fti_probe->ftp_id, 1 /* ndx */, -1 /* offset */, 
				     DTRACEFLT_UPRIV);
			} else if (id->fti_ptype == DTFTP_ENTRY) {
				/*
				 * We note that this was an entry
				 * probe to help ustack() find the
				 * first caller.
				 */
				cookie = dtrace_interrupt_disable();
				DTRACE_CPUFLAG_SET(CPU_DTRACE_USTACK_FP | CPU_DTRACE_ENTRY);
				dtrace_probe(probe->ftp_id, sv->save_r3, sv->save_r4,	/* Call the main probe routine with the first 5 args */
					sv->save_r5, sv->save_r6, sv->save_r7);
				DTRACE_CPUFLAG_CLEAR(CPU_DTRACE_USTACK_FP | CPU_DTRACE_ENTRY);
				dtrace_interrupt_enable(cookie);
				
			} else if (id->fti_ptype == DTFTP_IS_ENABLED) {
				/*
				 * Note that in this case, we don't
				 * call dtrace_probe() since it's only
				 * an artificial probe meant to change
				 * the flow of control so that it
				 * encounters the true probe.
				 */
				is_enabled = 1;
				
			} else if (probe->ftp_argmap == NULL) {
				DTRACE_CPUFLAG_SET(CPU_DTRACE_USTACK_FP);
				dtrace_probe(probe->ftp_id, sv->save_r3, sv->save_r4,	/* Call the main probe routine with the first 5 args */
					     sv->save_r5, sv->save_r6, sv->save_r7);
				DTRACE_CPUFLAG_CLEAR(CPU_DTRACE_USTACK_FP);
					     
			} else {
				uint64_t t[5];
				
				fasttrap_usdt_args(probe, sv, 5, t);	/* Grab 5 arguments */
				
				DTRACE_CPUFLAG_SET(CPU_DTRACE_USTACK_FP);
				dtrace_probe(probe->ftp_id, t[0], t[1],
					     t[2], t[3], t[4]);
				DTRACE_CPUFLAG_CLEAR(CPU_DTRACE_USTACK_FP);
			}

			/* APPLE NOTE: Oneshot probes get one and only one chance... */
			if (probe->ftp_prov->ftp_provider_type == DTFTP_PROVIDER_ONESHOT) {
				fasttrap_tracepoint_remove(p, tp);
			}
		}
	}

	/*
	 * We're about to do a bunch of work so we cache a local copy of
	 * the tracepoint to emulate the instruction, and then find the
	 * tracepoint again later if we need to light up any return probes.
	 */
	tp_local = *tp;
	lck_mtx_unlock(pid_mtx);
	tp = &tp_local;

	/*
	 * If there's an is-enabled probe connected to this tracepoint it
	 * means that there was a 'xor r3,r3,r3'
	 * instruction that was placed there by DTrace when the binary was
	 * linked. As this probe is, in fact, enabled, we need to stuff 1
	 * into R3. Accordingly, we can bypass all the instruction
	 * emulation logic since we know the inevitable result. It's possible
	 * that a user could construct a scenario where the 'is-enabled'
	 * probe was on some other instruction, but that would be a rather
	 * exotic way to shoot oneself in the foot.
	 */
	if (is_enabled) {
		sv->save_r3 = 1;				/* Set condition to true */
		new_pc = (sv->save_srr0 + 4) & addrmask;		/* Just fall through to the next instruction */
		goto done;
	}

	/*
	 * We emulate certain types of instructions to ensure correctness
	 * (in the case of position dependent instructions) or optimize
	 * common cases. The rest we execute in the kernel, but with
	 * most of the user's context active.
	 */
	switch (tp->ftt_type) {
	
		case ftmtNOP:					/* NOP  */
			new_pc = (sv->save_srr0 + 4) & addrmask;	/* Just fall through to the next instruction */
			break;

		case ftmtB:						/* Plain unconditional branch */
			new_pc = (user_addr_t)((int64_t)tp->ftt_trgt);	/* Assume target is absolute address for the moment */
			if(!(tp->ftt_flgs & ftmtAbs)) new_pc = (new_pc + sv->save_srr0) & addrmask;	/* We don't have absolute address, use as offset from instruction address */

			if(tp->ftt_flgs & ftmtLink) sv->save_lr = (sv->save_srr0 + 4) & addrmask;	/* Set the LR to the next instruction if needed */
			break;
		
		case ftmtBC:					/* Conditional PC relative or absolute branch */
			new_pc = (user_addr_t)((int64_t)tp->ftt_trgt);	/* Assume target is absolute address for the moment */
			if(!(tp->ftt_flgs & ftmtAbs)) new_pc = new_pc + sv->save_srr0;	/* We don't have absolute address, use as offset from instruction address */

			if(tp->ftt_flgs & ftmtLink) sv->save_lr = (sv->save_srr0 + 4) & addrmask;	/* Set the LR to the next instruction if needed */
			if(!branchtaken(tp->ftt_bo, tp->ftt_bi, sv)) new_pc = (sv->save_srr0 + 4) & addrmask;	/* If branch was not taken, set PC to next address */
			break;
		
		case ftmtBLR:					/* Conditional branch to LR */
			new_pc = sv->save_lr;		/* Branch target comes from the LR */

			if(tp->ftt_flgs & ftmtLink) sv->save_lr = (sv->save_srr0 + 4) & addrmask;	/* Set the LR to the next instruction if needed */			
			if(!branchtaken(tp->ftt_bo, tp->ftt_bi, sv)) new_pc = (sv->save_srr0 + 4) & addrmask;	/* If branch was not taken, set PC to next address */
			break;
		
		case ftmtBCTR:					/* Conditional branch to CTR */
			new_pc = sv->save_ctr;		/* Branch target comes from the CTR */

			if(tp->ftt_flgs & ftmtLink) sv->save_lr = (sv->save_srr0 + 4) & addrmask;	/* Set the LR to the next instruction if needed */			
			if(!branchtaken(tp->ftt_bo, tp->ftt_bi, sv)) new_pc = (sv->save_srr0 + 4) & addrmask;	/* If branch was not taken, set PC to next address */
			break;
		
		case ftmtCommon:				/* Common, non-in-kernel emulated instruction */
			sv->save_instr[0] = 1;		/* We only have one instruction to inject */
			sv->save_instr[1] = tp->ftt_instr;	/* Set the instruction */
			sv->save_hdr.save_flags = sv->save_hdr.save_flags | SAVinject;	/* Tell low-level exception return to inject the instruction */
			uthread->t_dtrace_step = 1;	/* Let it be known that a trace return is imminent */
			return 0;					/* Go and don't dome back until you are done... */
			
		default:
			panic("fasttrap_pid_probe: invalid ftt_type = %08X\n", tp->ftt_type);	/* Huh, wha happened? */
			break;
	}
		

done:
	
	/*
	 * If there were no return probes when we first found the tracepoint,
	 * we should feel no obligation to honor any return probes that were
	 * subsequently enabled -- they'll just have to wait until the next
	 * time around.
	 */
	sv->save_srr0 = new_pc;				/* Set the new PC */
	if (tp->ftt_retids != NULL) fasttrap_return_common(sv, pc, pid, new_pc);

	return (0);
}


int
fasttrap_return_probe(ppc_saved_state_t *sv)
{

	user_addr_t pc, npc;
	
	proc_t *p = current_proc();


	/*
	 * Treat a child created by a call to vfork(2) as if it were its
	 * parent. We know that there's only one thread of control in such a
	 * process: this one.
	 */
	/*
	 * APPLE NOTE: Terry says: "You need to hold the process locks (currently: kernel funnel) for this traversal"
	 * How do we assert this?
	 */
	while (p->p_lflag & P_LINVFORK) {
		p = p->p_pptr;
	}

	pc = sv->save_srr0;		/* Get the PC of the probed instruction */
	npc = pc + 4;			/* Get next PC */	
	if(!(sv->save_srr1 & 0x8000000000000000ULL)) npc &= 0x00000000FFFFFFFF;	/* Wrap new PC if running 32-bit */
	fasttrap_return_common(sv, pc, p->p_pid, npc);

	return (0);
}

uint64_t
fasttrap_pid_getarg(void *arg, dtrace_id_t id, void *parg, int argno,
    int aframes)
{
#pragma unused(arg, id, parg, aframes)
	return (fasttrap_anarg((ppc_saved_state_t *)find_user_regs(current_thread()), 1, argno));
}

uint64_t
fasttrap_usdt_getarg(void *arg, dtrace_id_t id, void *parg, int argno,
    int aframes)
{
#pragma unused(arg, id, parg, aframes)
	return (fasttrap_anarg((ppc_saved_state_t *)find_user_regs(current_thread()), 0, argno));
}


static int32_t branchtaken(int32_t bo, int32_t bi, ppc_saved_state_t *sv) {
	int32_t bcond, czero, crmatch;
	uint64_t ctr;
	
	if((bo & 0x14) == 0x14) return 1;	/* If this is a branch always, exit with true... */
	
	czero = 0;							/* Assume that we have not just decremented the CTR to 0 */
	
	if(!(bo & 4)) {						/* Skip the next bit if we do NOT muck with the CTR */
		ctr = sv->save_ctr = sv->save_ctr - 1;	/* Decrement the CTR */
		if(!(sv->save_srr1 & 0x8000000000000000ULL)) ctr &= 0x00000000FFFFFFFF;	/* Only look at the bottom 32 bits if 32-bit mode */
		czero = (ctr == 0);				/* Remember if we just hit zero */
	}
	
	bcond = (bo >> 3);					/* If 1, branch if CR flag is 1.  If 0, branch if 0 */
	crmatch = bo >> 4;					/* If bo[0] is set, do not check CR flag */
	crmatch = crmatch | (((sv->save_cr >> (31 - bi)) ^ bcond) ^ 1);	/* Low bit is now set if CR flag matches or CR is not checked. Other bits are trash. */

//	dbgTrace(0x77777777, bo, bi, sv->save_cr, ((czero | crmatch) & 1));	/* (TRACE/DEBUG) */

	return ((czero | crmatch) & 1);		/* Return 1 if branch taken, 0 if not... */	
}

static int32_t dtrace_decode_ppc(uint32_t inst) {

	int32_t curdcd, lastmask, newmask, spr, bit, bito, word;
	uint16_t xop = 0;
	dcdtab *dcd;
	
	curdcd = inst >> 26;				/* Isolate major op code to start decode */
	lastmask = 99;						/* Always force a new xop at the start */
	
	while(1) {							/* Loop until we find instruction or fail */
		dcd = &insts[curdcd];			/* Point to the current decode table entry */
		if(dcd->dcdFlgs & dcdJump) {	/* Should we jump to a new spot in the decode table? */
			curdcd = dcd->dcdMatch;		/* Jump */
			continue;
		}
		
		newmask = dcd->dcdFlgs & dcdMask;	/* Isolate the mask index */
		if(lastmask != newmask) {		/* Are we changing masks? */
			if(!newmask) break;			/* If the mask is 0, we match everything and succeed... (note: lastmask can never be 0) */
			xop = inst & masktab[newmask];	/* Clear all extra bits to make match */
			lastmask = newmask;			/* Remember */
		}
		
		if(xop == dcd->dcdMatch) break;	/* We found our guy! */
		
		if(!(dcd->dcdFlgs & dcdStep)) {	/* No stepping, we failed */
			dcd = &dcdfail;				/* Point to a failure entry */
			break;						/* Leave... */
		}
		
		curdcd = curdcd + 1;			/* Step to the next decode entry */
	}

	if(dcd->dcdType != diSPR) return (int32_t)(dcd->dcdType);	/* Return what we found */
	
	spr = (inst >> (31 - 20)) & 0x3FF;	/* Get the source */
	spr = ((spr << 5) & 0x3E0) | ((spr >> 5) & 0x1F);	/* Flip to right order */
	
	word = spr >> 5;					/* Get word index into table */
	bito = spr & 0x1F;					/* Get bit offset into entry */
	bit = 0x80000000 >> bito;			/* Position bit for a test */
	
	if(!(sprtbl[word] & bit)) return (diINV);	/* Bogus SPR so whole instruction is invalid... */
	
	if(spr & 0x10) return (diPRV);		/* This is a priviliged SPR so instruction is priviliged... */
	return (diCMN);						/* Just a common SPR so instruction is the same... */
}