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
|
// SPDX-License-Identifier: GPL-2.0
/*
* Kernel unwinding support
*
* (c) 2002-2004 Randolph Chung <tausq@debian.org>
*
* Derived partially from the IA64 implementation. The PA-RISC
* Runtime Architecture Document is also a useful reference to
* understand what is happening here
*/
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/sched.h>
#include <linux/slab.h>
#include <linux/sort.h>
#include <linux/sched/task_stack.h>
#include <linux/uaccess.h>
#include <asm/assembly.h>
#include <asm/asm-offsets.h>
#include <asm/ptrace.h>
#include <asm/unwind.h>
#include <asm/switch_to.h>
#include <asm/sections.h>
/* #define DEBUG 1 */
#ifdef DEBUG
#define dbg(x...) pr_debug(x)
#else
#define dbg(x...)
#endif
#define KERNEL_START (KERNEL_BINARY_TEXT_START)
extern struct unwind_table_entry __start___unwind[];
extern struct unwind_table_entry __stop___unwind[];
static DEFINE_SPINLOCK(unwind_lock);
/*
* the kernel unwind block is not dynamically allocated so that
* we can call unwind_init as early in the bootup process as
* possible (before the slab allocator is initialized)
*/
static struct unwind_table kernel_unwind_table __ro_after_init;
static LIST_HEAD(unwind_tables);
static inline const struct unwind_table_entry *
find_unwind_entry_in_table(const struct unwind_table *table, unsigned long addr)
{
const struct unwind_table_entry *e = NULL;
unsigned long lo, hi, mid;
lo = 0;
hi = table->length - 1;
while (lo <= hi) {
mid = (hi - lo) / 2 + lo;
e = &table->table[mid];
if (addr < e->region_start)
hi = mid - 1;
else if (addr > e->region_end)
lo = mid + 1;
else
return e;
}
return NULL;
}
static const struct unwind_table_entry *
find_unwind_entry(unsigned long addr)
{
struct unwind_table *table;
const struct unwind_table_entry *e = NULL;
if (addr >= kernel_unwind_table.start &&
addr <= kernel_unwind_table.end)
e = find_unwind_entry_in_table(&kernel_unwind_table, addr);
else {
unsigned long flags;
spin_lock_irqsave(&unwind_lock, flags);
list_for_each_entry(table, &unwind_tables, list) {
if (addr >= table->start &&
addr <= table->end)
e = find_unwind_entry_in_table(table, addr);
if (e) {
/* Move-to-front to exploit common traces */
list_move(&table->list, &unwind_tables);
break;
}
}
spin_unlock_irqrestore(&unwind_lock, flags);
}
return e;
}
static void
unwind_table_init(struct unwind_table *table, const char *name,
unsigned long base_addr, unsigned long gp,
void *table_start, void *table_end)
{
struct unwind_table_entry *start = table_start;
struct unwind_table_entry *end =
(struct unwind_table_entry *)table_end - 1;
table->name = name;
table->base_addr = base_addr;
table->gp = gp;
table->start = base_addr + start->region_start;
table->end = base_addr + end->region_end;
table->table = (struct unwind_table_entry *)table_start;
table->length = end - start + 1;
INIT_LIST_HEAD(&table->list);
for (; start <= end; start++) {
if (start < end &&
start->region_end > (start+1)->region_start) {
pr_warn("Out of order unwind entry! %px and %px\n",
start, start+1);
}
start->region_start += base_addr;
start->region_end += base_addr;
}
}
static int cmp_unwind_table_entry(const void *a, const void *b)
{
return ((const struct unwind_table_entry *)a)->region_start
- ((const struct unwind_table_entry *)b)->region_start;
}
static void
unwind_table_sort(struct unwind_table_entry *start,
struct unwind_table_entry *finish)
{
sort(start, finish - start, sizeof(struct unwind_table_entry),
cmp_unwind_table_entry, NULL);
}
struct unwind_table *
unwind_table_add(const char *name, unsigned long base_addr,
unsigned long gp,
void *start, void *end)
{
struct unwind_table *table;
unsigned long flags;
struct unwind_table_entry *s = (struct unwind_table_entry *)start;
struct unwind_table_entry *e = (struct unwind_table_entry *)end;
unwind_table_sort(s, e);
table = kmalloc(sizeof(struct unwind_table), GFP_USER);
if (table == NULL)
return NULL;
unwind_table_init(table, name, base_addr, gp, start, end);
spin_lock_irqsave(&unwind_lock, flags);
list_add_tail(&table->list, &unwind_tables);
spin_unlock_irqrestore(&unwind_lock, flags);
return table;
}
void unwind_table_remove(struct unwind_table *table)
{
unsigned long flags;
spin_lock_irqsave(&unwind_lock, flags);
list_del(&table->list);
spin_unlock_irqrestore(&unwind_lock, flags);
kfree(table);
}
/* Called from setup_arch to import the kernel unwind info */
int __init unwind_init(void)
{
long start, stop;
register unsigned long gp __asm__ ("r27");
start = (long)&__start___unwind[0];
stop = (long)&__stop___unwind[0];
dbg("unwind_init: start = 0x%lx, end = 0x%lx, entries = %lu\n",
start, stop,
(stop - start) / sizeof(struct unwind_table_entry));
unwind_table_init(&kernel_unwind_table, "kernel", KERNEL_START,
gp,
&__start___unwind[0], &__stop___unwind[0]);
#if 0
{
int i;
for (i = 0; i < 10; i++)
{
printk("region 0x%x-0x%x\n",
__start___unwind[i].region_start,
__start___unwind[i].region_end);
}
}
#endif
return 0;
}
static bool pc_is_kernel_fn(unsigned long pc, void *fn)
{
return (unsigned long)dereference_kernel_function_descriptor(fn) == pc;
}
static int unwind_special(struct unwind_frame_info *info, unsigned long pc, int frame_size)
{
/*
* We have to use void * instead of a function pointer, because
* function pointers aren't a pointer to the function on 64-bit.
* Make them const so the compiler knows they live in .text
* Note: We could use dereference_kernel_function_descriptor()
* instead but we want to keep it simple here.
*/
extern void * const handle_interruption;
extern void * const ret_from_kernel_thread;
extern void * const syscall_exit;
extern void * const intr_return;
extern void * const _switch_to_ret;
#ifdef CONFIG_IRQSTACKS
extern void * const _call_on_stack;
#endif /* CONFIG_IRQSTACKS */
if (pc_is_kernel_fn(pc, handle_interruption)) {
struct pt_regs *regs = (struct pt_regs *)(info->sp - frame_size - PT_SZ_ALGN);
dbg("Unwinding through handle_interruption()\n");
info->prev_sp = regs->gr[30];
info->prev_ip = regs->iaoq[0];
return 1;
}
if (pc_is_kernel_fn(pc, ret_from_kernel_thread) ||
pc_is_kernel_fn(pc, syscall_exit)) {
info->prev_sp = info->prev_ip = 0;
return 1;
}
if (pc_is_kernel_fn(pc, intr_return)) {
struct pt_regs *regs;
dbg("Found intr_return()\n");
regs = (struct pt_regs *)(info->sp - PT_SZ_ALGN);
info->prev_sp = regs->gr[30];
info->prev_ip = regs->iaoq[0];
info->rp = regs->gr[2];
return 1;
}
if (pc_is_kernel_fn(pc, _switch_to) ||
pc_is_kernel_fn(pc, _switch_to_ret)) {
info->prev_sp = info->sp - CALLEE_SAVE_FRAME_SIZE;
info->prev_ip = *(unsigned long *)(info->prev_sp - RP_OFFSET);
return 1;
}
#ifdef CONFIG_IRQSTACKS
if (pc_is_kernel_fn(pc, _call_on_stack)) {
info->prev_sp = *(unsigned long *)(info->sp - FRAME_SIZE - REG_SZ);
info->prev_ip = *(unsigned long *)(info->sp - FRAME_SIZE - RP_OFFSET);
return 1;
}
#endif
return 0;
}
static void unwind_frame_regs(struct unwind_frame_info *info)
{
const struct unwind_table_entry *e;
unsigned long npc;
unsigned int insn;
long frame_size = 0;
int looking_for_rp, rpoffset = 0;
e = find_unwind_entry(info->ip);
if (e == NULL) {
unsigned long sp;
dbg("Cannot find unwind entry for %pS; forced unwinding\n",
(void *) info->ip);
/* Since we are doing the unwinding blind, we don't know if
we are adjusting the stack correctly or extracting the rp
correctly. The rp is checked to see if it belongs to the
kernel text section, if not we assume we don't have a
correct stack frame and we continue to unwind the stack.
This is not quite correct, and will fail for loadable
modules. */
sp = info->sp & ~63;
do {
unsigned long tmp;
info->prev_sp = sp - 64;
info->prev_ip = 0;
/* Check if stack is inside kernel stack area */
if ((info->prev_sp - (unsigned long) task_stack_page(info->t))
>= THREAD_SIZE) {
info->prev_sp = 0;
break;
}
if (copy_from_kernel_nofault(&tmp,
(void *)info->prev_sp - RP_OFFSET, sizeof(tmp)))
break;
info->prev_ip = tmp;
sp = info->prev_sp;
} while (!kernel_text_address(info->prev_ip));
info->rp = 0;
dbg("analyzing func @ %lx with no unwind info, setting "
"prev_sp=%lx prev_ip=%lx\n", info->ip,
info->prev_sp, info->prev_ip);
} else {
dbg("e->start = 0x%x, e->end = 0x%x, Save_SP = %d, "
"Save_RP = %d, Millicode = %d size = %u\n",
e->region_start, e->region_end, e->Save_SP, e->Save_RP,
e->Millicode, e->Total_frame_size);
looking_for_rp = e->Save_RP;
for (npc = e->region_start;
(frame_size < (e->Total_frame_size << 3) ||
looking_for_rp) &&
npc < info->ip;
npc += 4) {
insn = *(unsigned int *)npc;
if ((insn & 0xffffc001) == 0x37de0000 ||
(insn & 0xffe00001) == 0x6fc00000) {
/* ldo X(sp), sp, or stwm X,D(sp) */
frame_size += (insn & 0x3fff) >> 1;
dbg("analyzing func @ %lx, insn=%08x @ "
"%lx, frame_size = %ld\n", info->ip,
insn, npc, frame_size);
} else if ((insn & 0xffe00009) == 0x73c00008) {
/* std,ma X,D(sp) */
frame_size += ((insn >> 4) & 0x3ff) << 3;
dbg("analyzing func @ %lx, insn=%08x @ "
"%lx, frame_size = %ld\n", info->ip,
insn, npc, frame_size);
} else if (insn == 0x6bc23fd9) {
/* stw rp,-20(sp) */
rpoffset = 20;
looking_for_rp = 0;
dbg("analyzing func @ %lx, insn=stw rp,"
"-20(sp) @ %lx\n", info->ip, npc);
} else if (insn == 0x0fc212c1) {
/* std rp,-16(sr0,sp) */
rpoffset = 16;
looking_for_rp = 0;
dbg("analyzing func @ %lx, insn=std rp,"
"-16(sp) @ %lx\n", info->ip, npc);
}
}
if (frame_size > e->Total_frame_size << 3)
frame_size = e->Total_frame_size << 3;
if (!unwind_special(info, e->region_start, frame_size)) {
info->prev_sp = info->sp - frame_size;
if (e->Millicode)
info->rp = info->r31;
else if (rpoffset)
info->rp = *(unsigned long *)(info->prev_sp - rpoffset);
info->prev_ip = info->rp;
info->rp = 0;
}
dbg("analyzing func @ %lx, setting prev_sp=%lx "
"prev_ip=%lx npc=%lx\n", info->ip, info->prev_sp,
info->prev_ip, npc);
}
}
void unwind_frame_init(struct unwind_frame_info *info, struct task_struct *t,
struct pt_regs *regs)
{
memset(info, 0, sizeof(struct unwind_frame_info));
info->t = t;
info->sp = regs->gr[30];
info->ip = regs->iaoq[0];
info->rp = regs->gr[2];
info->r31 = regs->gr[31];
dbg("(%d) Start unwind from sp=%08lx ip=%08lx\n",
t ? (int)t->pid : -1, info->sp, info->ip);
}
void unwind_frame_init_from_blocked_task(struct unwind_frame_info *info, struct task_struct *t)
{
struct pt_regs *r = &t->thread.regs;
struct pt_regs *r2;
r2 = kmalloc(sizeof(struct pt_regs), GFP_ATOMIC);
if (!r2)
return;
*r2 = *r;
r2->gr[30] = r->ksp;
r2->iaoq[0] = r->kpc;
unwind_frame_init(info, t, r2);
kfree(r2);
}
#define get_parisc_stackpointer() ({ \
unsigned long sp; \
__asm__("copy %%r30, %0" : "=r"(sp)); \
(sp); \
})
void unwind_frame_init_task(struct unwind_frame_info *info,
struct task_struct *task, struct pt_regs *regs)
{
task = task ? task : current;
if (task == current) {
struct pt_regs r;
if (!regs) {
memset(&r, 0, sizeof(r));
r.iaoq[0] = _THIS_IP_;
r.gr[2] = _RET_IP_;
r.gr[30] = get_parisc_stackpointer();
regs = &r;
}
unwind_frame_init(info, task, regs);
} else {
unwind_frame_init_from_blocked_task(info, task);
}
}
int unwind_once(struct unwind_frame_info *next_frame)
{
unwind_frame_regs(next_frame);
if (next_frame->prev_sp == 0 ||
next_frame->prev_ip == 0)
return -1;
next_frame->sp = next_frame->prev_sp;
next_frame->ip = next_frame->prev_ip;
next_frame->prev_sp = 0;
next_frame->prev_ip = 0;
dbg("(%d) Continue unwind to sp=%08lx ip=%08lx\n",
next_frame->t ? (int)next_frame->t->pid : -1,
next_frame->sp, next_frame->ip);
return 0;
}
int unwind_to_user(struct unwind_frame_info *info)
{
int ret;
do {
ret = unwind_once(info);
} while (!ret && !(info->ip & 3));
return ret;
}
unsigned long return_address(unsigned int level)
{
struct unwind_frame_info info;
/* initialize unwind info */
unwind_frame_init_task(&info, current, NULL);
/* unwind stack */
level += 2;
do {
if (unwind_once(&info) < 0 || info.ip == 0)
return 0;
if (!kernel_text_address(info.ip))
return 0;
} while (info.ip && level--);
return info.ip;
}
|