1 | // SPDX-License-Identifier: GPL-2.0 |
2 | /* |
3 | * Copyright (C) 2009 Matt Fleming <matt@console-pimps.org> |
4 | * |
5 | * This is an implementation of a DWARF unwinder. Its main purpose is |
6 | * for generating stacktrace information. Based on the DWARF 3 |
7 | * specification from http://www.dwarfstd.org. |
8 | * |
9 | * TODO: |
10 | * - DWARF64 doesn't work. |
11 | * - Registers with DWARF_VAL_OFFSET rules aren't handled properly. |
12 | */ |
13 | |
14 | /* #define DEBUG */ |
15 | #include <linux/kernel.h> |
16 | #include <linux/io.h> |
17 | #include <linux/list.h> |
18 | #include <linux/mempool.h> |
19 | #include <linux/mm.h> |
20 | #include <linux/elf.h> |
21 | #include <linux/ftrace.h> |
22 | #include <linux/module.h> |
23 | #include <linux/slab.h> |
24 | #include <asm/dwarf.h> |
25 | #include <asm/unwinder.h> |
26 | #include <asm/sections.h> |
27 | #include <asm/unaligned.h> |
28 | #include <asm/stacktrace.h> |
29 | |
30 | /* Reserve enough memory for two stack frames */ |
31 | #define DWARF_FRAME_MIN_REQ 2 |
32 | /* ... with 4 registers per frame. */ |
33 | #define DWARF_REG_MIN_REQ (DWARF_FRAME_MIN_REQ * 4) |
34 | |
35 | static struct kmem_cache *dwarf_frame_cachep; |
36 | static mempool_t *dwarf_frame_pool; |
37 | |
38 | static struct kmem_cache *dwarf_reg_cachep; |
39 | static mempool_t *dwarf_reg_pool; |
40 | |
41 | static struct rb_root cie_root; |
42 | static DEFINE_SPINLOCK(dwarf_cie_lock); |
43 | |
44 | static struct rb_root fde_root; |
45 | static DEFINE_SPINLOCK(dwarf_fde_lock); |
46 | |
47 | static struct dwarf_cie *cached_cie; |
48 | |
49 | static unsigned int dwarf_unwinder_ready; |
50 | |
51 | /** |
52 | * dwarf_frame_alloc_reg - allocate memory for a DWARF register |
53 | * @frame: the DWARF frame whose list of registers we insert on |
54 | * @reg_num: the register number |
55 | * |
56 | * Allocate space for, and initialise, a dwarf reg from |
57 | * dwarf_reg_pool and insert it onto the (unsorted) linked-list of |
58 | * dwarf registers for @frame. |
59 | * |
60 | * Return the initialised DWARF reg. |
61 | */ |
62 | static struct dwarf_reg *dwarf_frame_alloc_reg(struct dwarf_frame *frame, |
63 | unsigned int reg_num) |
64 | { |
65 | struct dwarf_reg *reg; |
66 | |
67 | reg = mempool_alloc(pool: dwarf_reg_pool, GFP_ATOMIC); |
68 | if (!reg) { |
69 | printk(KERN_WARNING "Unable to allocate a DWARF register\n" ); |
70 | /* |
71 | * Let's just bomb hard here, we have no way to |
72 | * gracefully recover. |
73 | */ |
74 | UNWINDER_BUG(); |
75 | } |
76 | |
77 | reg->number = reg_num; |
78 | reg->addr = 0; |
79 | reg->flags = 0; |
80 | |
81 | list_add(new: ®->link, head: &frame->reg_list); |
82 | |
83 | return reg; |
84 | } |
85 | |
86 | static void dwarf_frame_free_regs(struct dwarf_frame *frame) |
87 | { |
88 | struct dwarf_reg *reg, *n; |
89 | |
90 | list_for_each_entry_safe(reg, n, &frame->reg_list, link) { |
91 | list_del(entry: ®->link); |
92 | mempool_free(element: reg, pool: dwarf_reg_pool); |
93 | } |
94 | } |
95 | |
96 | /** |
97 | * dwarf_frame_reg - return a DWARF register |
98 | * @frame: the DWARF frame to search in for @reg_num |
99 | * @reg_num: the register number to search for |
100 | * |
101 | * Lookup and return the dwarf reg @reg_num for this frame. Return |
102 | * NULL if @reg_num is an register invalid number. |
103 | */ |
104 | static struct dwarf_reg *dwarf_frame_reg(struct dwarf_frame *frame, |
105 | unsigned int reg_num) |
106 | { |
107 | struct dwarf_reg *reg; |
108 | |
109 | list_for_each_entry(reg, &frame->reg_list, link) { |
110 | if (reg->number == reg_num) |
111 | return reg; |
112 | } |
113 | |
114 | return NULL; |
115 | } |
116 | |
117 | /** |
118 | * dwarf_read_addr - read dwarf data |
119 | * @src: source address of data |
120 | * @dst: destination address to store the data to |
121 | * |
122 | * Read 'n' bytes from @src, where 'n' is the size of an address on |
123 | * the native machine. We return the number of bytes read, which |
124 | * should always be 'n'. We also have to be careful when reading |
125 | * from @src and writing to @dst, because they can be arbitrarily |
126 | * aligned. Return 'n' - the number of bytes read. |
127 | */ |
128 | static inline int dwarf_read_addr(unsigned long *src, unsigned long *dst) |
129 | { |
130 | u32 val = get_unaligned(src); |
131 | put_unaligned(val, dst); |
132 | return sizeof(unsigned long *); |
133 | } |
134 | |
135 | /** |
136 | * dwarf_read_uleb128 - read unsigned LEB128 data |
137 | * @addr: the address where the ULEB128 data is stored |
138 | * @ret: address to store the result |
139 | * |
140 | * Decode an unsigned LEB128 encoded datum. The algorithm is taken |
141 | * from Appendix C of the DWARF 3 spec. For information on the |
142 | * encodings refer to section "7.6 - Variable Length Data". Return |
143 | * the number of bytes read. |
144 | */ |
145 | static inline unsigned long dwarf_read_uleb128(char *addr, unsigned int *ret) |
146 | { |
147 | unsigned int result; |
148 | unsigned char byte; |
149 | int shift, count; |
150 | |
151 | result = 0; |
152 | shift = 0; |
153 | count = 0; |
154 | |
155 | while (1) { |
156 | byte = __raw_readb(addr); |
157 | addr++; |
158 | count++; |
159 | |
160 | result |= (byte & 0x7f) << shift; |
161 | shift += 7; |
162 | |
163 | if (!(byte & 0x80)) |
164 | break; |
165 | } |
166 | |
167 | *ret = result; |
168 | |
169 | return count; |
170 | } |
171 | |
172 | /** |
173 | * dwarf_read_leb128 - read signed LEB128 data |
174 | * @addr: the address of the LEB128 encoded data |
175 | * @ret: address to store the result |
176 | * |
177 | * Decode signed LEB128 data. The algorithm is taken from Appendix |
178 | * C of the DWARF 3 spec. Return the number of bytes read. |
179 | */ |
180 | static inline unsigned long dwarf_read_leb128(char *addr, int *ret) |
181 | { |
182 | unsigned char byte; |
183 | int result, shift; |
184 | int num_bits; |
185 | int count; |
186 | |
187 | result = 0; |
188 | shift = 0; |
189 | count = 0; |
190 | |
191 | while (1) { |
192 | byte = __raw_readb(addr); |
193 | addr++; |
194 | result |= (byte & 0x7f) << shift; |
195 | shift += 7; |
196 | count++; |
197 | |
198 | if (!(byte & 0x80)) |
199 | break; |
200 | } |
201 | |
202 | /* The number of bits in a signed integer. */ |
203 | num_bits = 8 * sizeof(result); |
204 | |
205 | if ((shift < num_bits) && (byte & 0x40)) |
206 | result |= (-1 << shift); |
207 | |
208 | *ret = result; |
209 | |
210 | return count; |
211 | } |
212 | |
213 | /** |
214 | * dwarf_read_encoded_value - return the decoded value at @addr |
215 | * @addr: the address of the encoded value |
216 | * @val: where to write the decoded value |
217 | * @encoding: the encoding with which we can decode @addr |
218 | * |
219 | * GCC emits encoded address in the .eh_frame FDE entries. Decode |
220 | * the value at @addr using @encoding. The decoded value is written |
221 | * to @val and the number of bytes read is returned. |
222 | */ |
223 | static int dwarf_read_encoded_value(char *addr, unsigned long *val, |
224 | char encoding) |
225 | { |
226 | unsigned long decoded_addr = 0; |
227 | int count = 0; |
228 | |
229 | switch (encoding & 0x70) { |
230 | case DW_EH_PE_absptr: |
231 | break; |
232 | case DW_EH_PE_pcrel: |
233 | decoded_addr = (unsigned long)addr; |
234 | break; |
235 | default: |
236 | pr_debug("encoding=0x%x\n" , (encoding & 0x70)); |
237 | UNWINDER_BUG(); |
238 | } |
239 | |
240 | if ((encoding & 0x07) == 0x00) |
241 | encoding |= DW_EH_PE_udata4; |
242 | |
243 | switch (encoding & 0x0f) { |
244 | case DW_EH_PE_sdata4: |
245 | case DW_EH_PE_udata4: |
246 | count += 4; |
247 | decoded_addr += get_unaligned((u32 *)addr); |
248 | __raw_writel(val: decoded_addr, addr: val); |
249 | break; |
250 | default: |
251 | pr_debug("encoding=0x%x\n" , encoding); |
252 | UNWINDER_BUG(); |
253 | } |
254 | |
255 | return count; |
256 | } |
257 | |
258 | /** |
259 | * dwarf_entry_len - return the length of an FDE or CIE |
260 | * @addr: the address of the entry |
261 | * @len: the length of the entry |
262 | * |
263 | * Read the initial_length field of the entry and store the size of |
264 | * the entry in @len. We return the number of bytes read. Return a |
265 | * count of 0 on error. |
266 | */ |
267 | static inline int dwarf_entry_len(char *addr, unsigned long *len) |
268 | { |
269 | u32 initial_len; |
270 | int count; |
271 | |
272 | initial_len = get_unaligned((u32 *)addr); |
273 | count = 4; |
274 | |
275 | /* |
276 | * An initial length field value in the range DW_LEN_EXT_LO - |
277 | * DW_LEN_EXT_HI indicates an extension, and should not be |
278 | * interpreted as a length. The only extension that we currently |
279 | * understand is the use of DWARF64 addresses. |
280 | */ |
281 | if (initial_len >= DW_EXT_LO && initial_len <= DW_EXT_HI) { |
282 | /* |
283 | * The 64-bit length field immediately follows the |
284 | * compulsory 32-bit length field. |
285 | */ |
286 | if (initial_len == DW_EXT_DWARF64) { |
287 | *len = get_unaligned((u64 *)addr + 4); |
288 | count = 12; |
289 | } else { |
290 | printk(KERN_WARNING "Unknown DWARF extension\n" ); |
291 | count = 0; |
292 | } |
293 | } else |
294 | *len = initial_len; |
295 | |
296 | return count; |
297 | } |
298 | |
299 | /** |
300 | * dwarf_lookup_cie - locate the cie |
301 | * @cie_ptr: pointer to help with lookup |
302 | */ |
303 | static struct dwarf_cie *dwarf_lookup_cie(unsigned long cie_ptr) |
304 | { |
305 | struct rb_node **rb_node = &cie_root.rb_node; |
306 | struct dwarf_cie *cie = NULL; |
307 | unsigned long flags; |
308 | |
309 | spin_lock_irqsave(&dwarf_cie_lock, flags); |
310 | |
311 | /* |
312 | * We've cached the last CIE we looked up because chances are |
313 | * that the FDE wants this CIE. |
314 | */ |
315 | if (cached_cie && cached_cie->cie_pointer == cie_ptr) { |
316 | cie = cached_cie; |
317 | goto out; |
318 | } |
319 | |
320 | while (*rb_node) { |
321 | struct dwarf_cie *cie_tmp; |
322 | |
323 | cie_tmp = rb_entry(*rb_node, struct dwarf_cie, node); |
324 | BUG_ON(!cie_tmp); |
325 | |
326 | if (cie_ptr == cie_tmp->cie_pointer) { |
327 | cie = cie_tmp; |
328 | cached_cie = cie_tmp; |
329 | goto out; |
330 | } else { |
331 | if (cie_ptr < cie_tmp->cie_pointer) |
332 | rb_node = &(*rb_node)->rb_left; |
333 | else |
334 | rb_node = &(*rb_node)->rb_right; |
335 | } |
336 | } |
337 | |
338 | out: |
339 | spin_unlock_irqrestore(lock: &dwarf_cie_lock, flags); |
340 | return cie; |
341 | } |
342 | |
343 | /** |
344 | * dwarf_lookup_fde - locate the FDE that covers pc |
345 | * @pc: the program counter |
346 | */ |
347 | struct dwarf_fde *dwarf_lookup_fde(unsigned long pc) |
348 | { |
349 | struct rb_node **rb_node = &fde_root.rb_node; |
350 | struct dwarf_fde *fde = NULL; |
351 | unsigned long flags; |
352 | |
353 | spin_lock_irqsave(&dwarf_fde_lock, flags); |
354 | |
355 | while (*rb_node) { |
356 | struct dwarf_fde *fde_tmp; |
357 | unsigned long tmp_start, tmp_end; |
358 | |
359 | fde_tmp = rb_entry(*rb_node, struct dwarf_fde, node); |
360 | BUG_ON(!fde_tmp); |
361 | |
362 | tmp_start = fde_tmp->initial_location; |
363 | tmp_end = fde_tmp->initial_location + fde_tmp->address_range; |
364 | |
365 | if (pc < tmp_start) { |
366 | rb_node = &(*rb_node)->rb_left; |
367 | } else { |
368 | if (pc < tmp_end) { |
369 | fde = fde_tmp; |
370 | goto out; |
371 | } else |
372 | rb_node = &(*rb_node)->rb_right; |
373 | } |
374 | } |
375 | |
376 | out: |
377 | spin_unlock_irqrestore(lock: &dwarf_fde_lock, flags); |
378 | |
379 | return fde; |
380 | } |
381 | |
382 | /** |
383 | * dwarf_cfa_execute_insns - execute instructions to calculate a CFA |
384 | * @insn_start: address of the first instruction |
385 | * @insn_end: address of the last instruction |
386 | * @cie: the CIE for this function |
387 | * @fde: the FDE for this function |
388 | * @frame: the instructions calculate the CFA for this frame |
389 | * @pc: the program counter of the address we're interested in |
390 | * |
391 | * Execute the Call Frame instruction sequence starting at |
392 | * @insn_start and ending at @insn_end. The instructions describe |
393 | * how to calculate the Canonical Frame Address of a stackframe. |
394 | * Store the results in @frame. |
395 | */ |
396 | static int dwarf_cfa_execute_insns(unsigned char *insn_start, |
397 | unsigned char *insn_end, |
398 | struct dwarf_cie *cie, |
399 | struct dwarf_fde *fde, |
400 | struct dwarf_frame *frame, |
401 | unsigned long pc) |
402 | { |
403 | unsigned char insn; |
404 | unsigned char *current_insn; |
405 | unsigned int count, delta, reg, expr_len, offset; |
406 | struct dwarf_reg *regp; |
407 | |
408 | current_insn = insn_start; |
409 | |
410 | while (current_insn < insn_end && frame->pc <= pc) { |
411 | insn = __raw_readb(addr: current_insn++); |
412 | |
413 | /* |
414 | * Firstly, handle the opcodes that embed their operands |
415 | * in the instructions. |
416 | */ |
417 | switch (DW_CFA_opcode(insn)) { |
418 | case DW_CFA_advance_loc: |
419 | delta = DW_CFA_operand(insn); |
420 | delta *= cie->code_alignment_factor; |
421 | frame->pc += delta; |
422 | continue; |
423 | /* NOTREACHED */ |
424 | case DW_CFA_offset: |
425 | reg = DW_CFA_operand(insn); |
426 | count = dwarf_read_uleb128(addr: current_insn, ret: &offset); |
427 | current_insn += count; |
428 | offset *= cie->data_alignment_factor; |
429 | regp = dwarf_frame_alloc_reg(frame, reg_num: reg); |
430 | regp->addr = offset; |
431 | regp->flags |= DWARF_REG_OFFSET; |
432 | continue; |
433 | /* NOTREACHED */ |
434 | case DW_CFA_restore: |
435 | reg = DW_CFA_operand(insn); |
436 | continue; |
437 | /* NOTREACHED */ |
438 | } |
439 | |
440 | /* |
441 | * Secondly, handle the opcodes that don't embed their |
442 | * operands in the instruction. |
443 | */ |
444 | switch (insn) { |
445 | case DW_CFA_nop: |
446 | continue; |
447 | case DW_CFA_advance_loc1: |
448 | delta = *current_insn++; |
449 | frame->pc += delta * cie->code_alignment_factor; |
450 | break; |
451 | case DW_CFA_advance_loc2: |
452 | delta = get_unaligned((u16 *)current_insn); |
453 | current_insn += 2; |
454 | frame->pc += delta * cie->code_alignment_factor; |
455 | break; |
456 | case DW_CFA_advance_loc4: |
457 | delta = get_unaligned((u32 *)current_insn); |
458 | current_insn += 4; |
459 | frame->pc += delta * cie->code_alignment_factor; |
460 | break; |
461 | case DW_CFA_offset_extended: |
462 | count = dwarf_read_uleb128(addr: current_insn, ret: ®); |
463 | current_insn += count; |
464 | count = dwarf_read_uleb128(addr: current_insn, ret: &offset); |
465 | current_insn += count; |
466 | offset *= cie->data_alignment_factor; |
467 | break; |
468 | case DW_CFA_restore_extended: |
469 | count = dwarf_read_uleb128(addr: current_insn, ret: ®); |
470 | current_insn += count; |
471 | break; |
472 | case DW_CFA_undefined: |
473 | count = dwarf_read_uleb128(addr: current_insn, ret: ®); |
474 | current_insn += count; |
475 | regp = dwarf_frame_alloc_reg(frame, reg_num: reg); |
476 | regp->flags |= DWARF_UNDEFINED; |
477 | break; |
478 | case DW_CFA_def_cfa: |
479 | count = dwarf_read_uleb128(addr: current_insn, |
480 | ret: &frame->cfa_register); |
481 | current_insn += count; |
482 | count = dwarf_read_uleb128(addr: current_insn, |
483 | ret: &frame->cfa_offset); |
484 | current_insn += count; |
485 | |
486 | frame->flags |= DWARF_FRAME_CFA_REG_OFFSET; |
487 | break; |
488 | case DW_CFA_def_cfa_register: |
489 | count = dwarf_read_uleb128(addr: current_insn, |
490 | ret: &frame->cfa_register); |
491 | current_insn += count; |
492 | frame->flags |= DWARF_FRAME_CFA_REG_OFFSET; |
493 | break; |
494 | case DW_CFA_def_cfa_offset: |
495 | count = dwarf_read_uleb128(addr: current_insn, ret: &offset); |
496 | current_insn += count; |
497 | frame->cfa_offset = offset; |
498 | break; |
499 | case DW_CFA_def_cfa_expression: |
500 | count = dwarf_read_uleb128(addr: current_insn, ret: &expr_len); |
501 | current_insn += count; |
502 | |
503 | frame->cfa_expr = current_insn; |
504 | frame->cfa_expr_len = expr_len; |
505 | current_insn += expr_len; |
506 | |
507 | frame->flags |= DWARF_FRAME_CFA_REG_EXP; |
508 | break; |
509 | case DW_CFA_offset_extended_sf: |
510 | count = dwarf_read_uleb128(addr: current_insn, ret: ®); |
511 | current_insn += count; |
512 | count = dwarf_read_leb128(addr: current_insn, ret: &offset); |
513 | current_insn += count; |
514 | offset *= cie->data_alignment_factor; |
515 | regp = dwarf_frame_alloc_reg(frame, reg_num: reg); |
516 | regp->flags |= DWARF_REG_OFFSET; |
517 | regp->addr = offset; |
518 | break; |
519 | case DW_CFA_val_offset: |
520 | count = dwarf_read_uleb128(addr: current_insn, ret: ®); |
521 | current_insn += count; |
522 | count = dwarf_read_leb128(addr: current_insn, ret: &offset); |
523 | offset *= cie->data_alignment_factor; |
524 | regp = dwarf_frame_alloc_reg(frame, reg_num: reg); |
525 | regp->flags |= DWARF_VAL_OFFSET; |
526 | regp->addr = offset; |
527 | break; |
528 | case DW_CFA_GNU_args_size: |
529 | count = dwarf_read_uleb128(addr: current_insn, ret: &offset); |
530 | current_insn += count; |
531 | break; |
532 | case DW_CFA_GNU_negative_offset_extended: |
533 | count = dwarf_read_uleb128(addr: current_insn, ret: ®); |
534 | current_insn += count; |
535 | count = dwarf_read_uleb128(addr: current_insn, ret: &offset); |
536 | offset *= cie->data_alignment_factor; |
537 | |
538 | regp = dwarf_frame_alloc_reg(frame, reg_num: reg); |
539 | regp->flags |= DWARF_REG_OFFSET; |
540 | regp->addr = -offset; |
541 | break; |
542 | default: |
543 | pr_debug("unhandled DWARF instruction 0x%x\n" , insn); |
544 | UNWINDER_BUG(); |
545 | break; |
546 | } |
547 | } |
548 | |
549 | return 0; |
550 | } |
551 | |
552 | /** |
553 | * dwarf_free_frame - free the memory allocated for @frame |
554 | * @frame: the frame to free |
555 | */ |
556 | void dwarf_free_frame(struct dwarf_frame *frame) |
557 | { |
558 | dwarf_frame_free_regs(frame); |
559 | mempool_free(element: frame, pool: dwarf_frame_pool); |
560 | } |
561 | |
562 | extern void ret_from_irq(void); |
563 | |
564 | /** |
565 | * dwarf_unwind_stack - unwind the stack |
566 | * |
567 | * @pc: address of the function to unwind |
568 | * @prev: struct dwarf_frame of the previous stackframe on the callstack |
569 | * |
570 | * Return a struct dwarf_frame representing the most recent frame |
571 | * on the callstack. Each of the lower (older) stack frames are |
572 | * linked via the "prev" member. |
573 | */ |
574 | struct dwarf_frame *dwarf_unwind_stack(unsigned long pc, |
575 | struct dwarf_frame *prev) |
576 | { |
577 | struct dwarf_frame *frame; |
578 | struct dwarf_cie *cie; |
579 | struct dwarf_fde *fde; |
580 | struct dwarf_reg *reg; |
581 | unsigned long addr; |
582 | |
583 | /* |
584 | * If we've been called in to before initialization has |
585 | * completed, bail out immediately. |
586 | */ |
587 | if (!dwarf_unwinder_ready) |
588 | return NULL; |
589 | |
590 | /* |
591 | * If we're starting at the top of the stack we need get the |
592 | * contents of a physical register to get the CFA in order to |
593 | * begin the virtual unwinding of the stack. |
594 | * |
595 | * NOTE: the return address is guaranteed to be setup by the |
596 | * time this function makes its first function call. |
597 | */ |
598 | if (!pc || !prev) |
599 | pc = _THIS_IP_; |
600 | |
601 | #ifdef CONFIG_FUNCTION_GRAPH_TRACER |
602 | /* |
603 | * If our stack has been patched by the function graph tracer |
604 | * then we might see the address of return_to_handler() where we |
605 | * expected to find the real return address. |
606 | */ |
607 | if (pc == (unsigned long)&return_to_handler) { |
608 | struct ftrace_ret_stack *ret_stack; |
609 | |
610 | ret_stack = ftrace_graph_get_ret_stack(current, idx: 0); |
611 | if (ret_stack) |
612 | pc = ret_stack->ret; |
613 | /* |
614 | * We currently have no way of tracking how many |
615 | * return_to_handler()'s we've seen. If there is more |
616 | * than one patched return address on our stack, |
617 | * complain loudly. |
618 | */ |
619 | WARN_ON(ftrace_graph_get_ret_stack(current, 1)); |
620 | } |
621 | #endif |
622 | |
623 | frame = mempool_alloc(pool: dwarf_frame_pool, GFP_ATOMIC); |
624 | if (!frame) { |
625 | printk(KERN_ERR "Unable to allocate a dwarf frame\n" ); |
626 | UNWINDER_BUG(); |
627 | } |
628 | |
629 | INIT_LIST_HEAD(list: &frame->reg_list); |
630 | frame->flags = 0; |
631 | frame->prev = prev; |
632 | frame->return_addr = 0; |
633 | |
634 | fde = dwarf_lookup_fde(pc); |
635 | if (!fde) { |
636 | /* |
637 | * This is our normal exit path. There are two reasons |
638 | * why we might exit here, |
639 | * |
640 | * a) pc has no asscociated DWARF frame info and so |
641 | * we don't know how to unwind this frame. This is |
642 | * usually the case when we're trying to unwind a |
643 | * frame that was called from some assembly code |
644 | * that has no DWARF info, e.g. syscalls. |
645 | * |
646 | * b) the DEBUG info for pc is bogus. There's |
647 | * really no way to distinguish this case from the |
648 | * case above, which sucks because we could print a |
649 | * warning here. |
650 | */ |
651 | goto bail; |
652 | } |
653 | |
654 | cie = dwarf_lookup_cie(cie_ptr: fde->cie_pointer); |
655 | |
656 | frame->pc = fde->initial_location; |
657 | |
658 | /* CIE initial instructions */ |
659 | dwarf_cfa_execute_insns(insn_start: cie->initial_instructions, |
660 | insn_end: cie->instructions_end, cie, fde, |
661 | frame, pc); |
662 | |
663 | /* FDE instructions */ |
664 | dwarf_cfa_execute_insns(insn_start: fde->instructions, insn_end: fde->end, cie, |
665 | fde, frame, pc); |
666 | |
667 | /* Calculate the CFA */ |
668 | switch (frame->flags) { |
669 | case DWARF_FRAME_CFA_REG_OFFSET: |
670 | if (prev) { |
671 | reg = dwarf_frame_reg(frame: prev, reg_num: frame->cfa_register); |
672 | UNWINDER_BUG_ON(!reg); |
673 | UNWINDER_BUG_ON(reg->flags != DWARF_REG_OFFSET); |
674 | |
675 | addr = prev->cfa + reg->addr; |
676 | frame->cfa = __raw_readl(addr); |
677 | |
678 | } else { |
679 | /* |
680 | * Again, we're starting from the top of the |
681 | * stack. We need to physically read |
682 | * the contents of a register in order to get |
683 | * the Canonical Frame Address for this |
684 | * function. |
685 | */ |
686 | frame->cfa = dwarf_read_arch_reg(frame->cfa_register); |
687 | } |
688 | |
689 | frame->cfa += frame->cfa_offset; |
690 | break; |
691 | default: |
692 | UNWINDER_BUG(); |
693 | } |
694 | |
695 | reg = dwarf_frame_reg(frame, reg_num: DWARF_ARCH_RA_REG); |
696 | |
697 | /* |
698 | * If we haven't seen the return address register or the return |
699 | * address column is undefined then we must assume that this is |
700 | * the end of the callstack. |
701 | */ |
702 | if (!reg || reg->flags == DWARF_UNDEFINED) |
703 | goto bail; |
704 | |
705 | UNWINDER_BUG_ON(reg->flags != DWARF_REG_OFFSET); |
706 | |
707 | addr = frame->cfa + reg->addr; |
708 | frame->return_addr = __raw_readl(addr); |
709 | |
710 | /* |
711 | * Ah, the joys of unwinding through interrupts. |
712 | * |
713 | * Interrupts are tricky - the DWARF info needs to be _really_ |
714 | * accurate and unfortunately I'm seeing a lot of bogus DWARF |
715 | * info. For example, I've seen interrupts occur in epilogues |
716 | * just after the frame pointer (r14) had been restored. The |
717 | * problem was that the DWARF info claimed that the CFA could be |
718 | * reached by using the value of the frame pointer before it was |
719 | * restored. |
720 | * |
721 | * So until the compiler can be trusted to produce reliable |
722 | * DWARF info when it really matters, let's stop unwinding once |
723 | * we've calculated the function that was interrupted. |
724 | */ |
725 | if (prev && prev->pc == (unsigned long)ret_from_irq) |
726 | frame->return_addr = 0; |
727 | |
728 | return frame; |
729 | |
730 | bail: |
731 | dwarf_free_frame(frame); |
732 | return NULL; |
733 | } |
734 | |
735 | static int dwarf_parse_cie(void *entry, void *p, unsigned long len, |
736 | unsigned char *end, struct module *mod) |
737 | { |
738 | struct rb_node **rb_node = &cie_root.rb_node; |
739 | struct rb_node *parent = *rb_node; |
740 | struct dwarf_cie *cie; |
741 | unsigned long flags; |
742 | int count; |
743 | |
744 | cie = kzalloc(sizeof(*cie), GFP_KERNEL); |
745 | if (!cie) |
746 | return -ENOMEM; |
747 | |
748 | cie->length = len; |
749 | |
750 | /* |
751 | * Record the offset into the .eh_frame section |
752 | * for this CIE. It allows this CIE to be |
753 | * quickly and easily looked up from the |
754 | * corresponding FDE. |
755 | */ |
756 | cie->cie_pointer = (unsigned long)entry; |
757 | |
758 | cie->version = *(char *)p++; |
759 | UNWINDER_BUG_ON(cie->version != 1); |
760 | |
761 | cie->augmentation = p; |
762 | p += strlen(cie->augmentation) + 1; |
763 | |
764 | count = dwarf_read_uleb128(addr: p, ret: &cie->code_alignment_factor); |
765 | p += count; |
766 | |
767 | count = dwarf_read_leb128(addr: p, ret: &cie->data_alignment_factor); |
768 | p += count; |
769 | |
770 | /* |
771 | * Which column in the rule table contains the |
772 | * return address? |
773 | */ |
774 | if (cie->version == 1) { |
775 | cie->return_address_reg = __raw_readb(addr: p); |
776 | p++; |
777 | } else { |
778 | count = dwarf_read_uleb128(addr: p, ret: &cie->return_address_reg); |
779 | p += count; |
780 | } |
781 | |
782 | if (cie->augmentation[0] == 'z') { |
783 | unsigned int length, count; |
784 | cie->flags |= DWARF_CIE_Z_AUGMENTATION; |
785 | |
786 | count = dwarf_read_uleb128(addr: p, ret: &length); |
787 | p += count; |
788 | |
789 | UNWINDER_BUG_ON((unsigned char *)p > end); |
790 | |
791 | cie->initial_instructions = p + length; |
792 | cie->augmentation++; |
793 | } |
794 | |
795 | while (*cie->augmentation) { |
796 | /* |
797 | * "L" indicates a byte showing how the |
798 | * LSDA pointer is encoded. Skip it. |
799 | */ |
800 | if (*cie->augmentation == 'L') { |
801 | p++; |
802 | cie->augmentation++; |
803 | } else if (*cie->augmentation == 'R') { |
804 | /* |
805 | * "R" indicates a byte showing |
806 | * how FDE addresses are |
807 | * encoded. |
808 | */ |
809 | cie->encoding = *(char *)p++; |
810 | cie->augmentation++; |
811 | } else if (*cie->augmentation == 'P') { |
812 | /* |
813 | * "R" indicates a personality |
814 | * routine in the CIE |
815 | * augmentation. |
816 | */ |
817 | UNWINDER_BUG(); |
818 | } else if (*cie->augmentation == 'S') { |
819 | UNWINDER_BUG(); |
820 | } else { |
821 | /* |
822 | * Unknown augmentation. Assume |
823 | * 'z' augmentation. |
824 | */ |
825 | p = cie->initial_instructions; |
826 | UNWINDER_BUG_ON(!p); |
827 | break; |
828 | } |
829 | } |
830 | |
831 | cie->initial_instructions = p; |
832 | cie->instructions_end = end; |
833 | |
834 | /* Add to list */ |
835 | spin_lock_irqsave(&dwarf_cie_lock, flags); |
836 | |
837 | while (*rb_node) { |
838 | struct dwarf_cie *cie_tmp; |
839 | |
840 | cie_tmp = rb_entry(*rb_node, struct dwarf_cie, node); |
841 | |
842 | parent = *rb_node; |
843 | |
844 | if (cie->cie_pointer < cie_tmp->cie_pointer) |
845 | rb_node = &parent->rb_left; |
846 | else if (cie->cie_pointer >= cie_tmp->cie_pointer) |
847 | rb_node = &parent->rb_right; |
848 | else |
849 | WARN_ON(1); |
850 | } |
851 | |
852 | rb_link_node(node: &cie->node, parent, rb_link: rb_node); |
853 | rb_insert_color(&cie->node, &cie_root); |
854 | |
855 | #ifdef CONFIG_MODULES |
856 | if (mod != NULL) |
857 | list_add_tail(new: &cie->link, head: &mod->arch.cie_list); |
858 | #endif |
859 | |
860 | spin_unlock_irqrestore(lock: &dwarf_cie_lock, flags); |
861 | |
862 | return 0; |
863 | } |
864 | |
865 | static int dwarf_parse_fde(void *entry, u32 entry_type, |
866 | void *start, unsigned long len, |
867 | unsigned char *end, struct module *mod) |
868 | { |
869 | struct rb_node **rb_node = &fde_root.rb_node; |
870 | struct rb_node *parent = *rb_node; |
871 | struct dwarf_fde *fde; |
872 | struct dwarf_cie *cie; |
873 | unsigned long flags; |
874 | int count; |
875 | void *p = start; |
876 | |
877 | fde = kzalloc(sizeof(*fde), GFP_KERNEL); |
878 | if (!fde) |
879 | return -ENOMEM; |
880 | |
881 | fde->length = len; |
882 | |
883 | /* |
884 | * In a .eh_frame section the CIE pointer is the |
885 | * delta between the address within the FDE |
886 | */ |
887 | fde->cie_pointer = (unsigned long)(p - entry_type - 4); |
888 | |
889 | cie = dwarf_lookup_cie(cie_ptr: fde->cie_pointer); |
890 | fde->cie = cie; |
891 | |
892 | if (cie->encoding) |
893 | count = dwarf_read_encoded_value(addr: p, val: &fde->initial_location, |
894 | encoding: cie->encoding); |
895 | else |
896 | count = dwarf_read_addr(src: p, dst: &fde->initial_location); |
897 | |
898 | p += count; |
899 | |
900 | if (cie->encoding) |
901 | count = dwarf_read_encoded_value(addr: p, val: &fde->address_range, |
902 | encoding: cie->encoding & 0x0f); |
903 | else |
904 | count = dwarf_read_addr(src: p, dst: &fde->address_range); |
905 | |
906 | p += count; |
907 | |
908 | if (fde->cie->flags & DWARF_CIE_Z_AUGMENTATION) { |
909 | unsigned int length; |
910 | count = dwarf_read_uleb128(addr: p, ret: &length); |
911 | p += count + length; |
912 | } |
913 | |
914 | /* Call frame instructions. */ |
915 | fde->instructions = p; |
916 | fde->end = end; |
917 | |
918 | /* Add to list. */ |
919 | spin_lock_irqsave(&dwarf_fde_lock, flags); |
920 | |
921 | while (*rb_node) { |
922 | struct dwarf_fde *fde_tmp; |
923 | unsigned long tmp_start, tmp_end; |
924 | unsigned long start, end; |
925 | |
926 | fde_tmp = rb_entry(*rb_node, struct dwarf_fde, node); |
927 | |
928 | start = fde->initial_location; |
929 | end = fde->initial_location + fde->address_range; |
930 | |
931 | tmp_start = fde_tmp->initial_location; |
932 | tmp_end = fde_tmp->initial_location + fde_tmp->address_range; |
933 | |
934 | parent = *rb_node; |
935 | |
936 | if (start < tmp_start) |
937 | rb_node = &parent->rb_left; |
938 | else if (start >= tmp_end) |
939 | rb_node = &parent->rb_right; |
940 | else |
941 | WARN_ON(1); |
942 | } |
943 | |
944 | rb_link_node(node: &fde->node, parent, rb_link: rb_node); |
945 | rb_insert_color(&fde->node, &fde_root); |
946 | |
947 | #ifdef CONFIG_MODULES |
948 | if (mod != NULL) |
949 | list_add_tail(new: &fde->link, head: &mod->arch.fde_list); |
950 | #endif |
951 | |
952 | spin_unlock_irqrestore(lock: &dwarf_fde_lock, flags); |
953 | |
954 | return 0; |
955 | } |
956 | |
957 | static void dwarf_unwinder_dump(struct task_struct *task, |
958 | struct pt_regs *regs, |
959 | unsigned long *sp, |
960 | const struct stacktrace_ops *ops, |
961 | void *data) |
962 | { |
963 | struct dwarf_frame *frame, *_frame; |
964 | unsigned long return_addr; |
965 | |
966 | _frame = NULL; |
967 | return_addr = 0; |
968 | |
969 | while (1) { |
970 | frame = dwarf_unwind_stack(pc: return_addr, prev: _frame); |
971 | |
972 | if (_frame) |
973 | dwarf_free_frame(frame: _frame); |
974 | |
975 | _frame = frame; |
976 | |
977 | if (!frame || !frame->return_addr) |
978 | break; |
979 | |
980 | return_addr = frame->return_addr; |
981 | ops->address(data, return_addr, 1); |
982 | } |
983 | |
984 | if (frame) |
985 | dwarf_free_frame(frame); |
986 | } |
987 | |
988 | static struct unwinder dwarf_unwinder = { |
989 | .name = "dwarf-unwinder" , |
990 | .dump = dwarf_unwinder_dump, |
991 | .rating = 150, |
992 | }; |
993 | |
994 | static void __init dwarf_unwinder_cleanup(void) |
995 | { |
996 | struct dwarf_fde *fde, *next_fde; |
997 | struct dwarf_cie *cie, *next_cie; |
998 | |
999 | /* |
1000 | * Deallocate all the memory allocated for the DWARF unwinder. |
1001 | * Traverse all the FDE/CIE lists and remove and free all the |
1002 | * memory associated with those data structures. |
1003 | */ |
1004 | rbtree_postorder_for_each_entry_safe(fde, next_fde, &fde_root, node) |
1005 | kfree(objp: fde); |
1006 | |
1007 | rbtree_postorder_for_each_entry_safe(cie, next_cie, &cie_root, node) |
1008 | kfree(objp: cie); |
1009 | |
1010 | mempool_destroy(pool: dwarf_reg_pool); |
1011 | mempool_destroy(pool: dwarf_frame_pool); |
1012 | kmem_cache_destroy(s: dwarf_reg_cachep); |
1013 | kmem_cache_destroy(s: dwarf_frame_cachep); |
1014 | } |
1015 | |
1016 | /** |
1017 | * dwarf_parse_section - parse DWARF section |
1018 | * @eh_frame_start: start address of the .eh_frame section |
1019 | * @eh_frame_end: end address of the .eh_frame section |
1020 | * @mod: the kernel module containing the .eh_frame section |
1021 | * |
1022 | * Parse the information in a .eh_frame section. |
1023 | */ |
1024 | static int dwarf_parse_section(char *eh_frame_start, char *eh_frame_end, |
1025 | struct module *mod) |
1026 | { |
1027 | u32 entry_type; |
1028 | void *p, *entry; |
1029 | int count, err = 0; |
1030 | unsigned long len = 0; |
1031 | unsigned int c_entries, f_entries; |
1032 | unsigned char *end; |
1033 | |
1034 | c_entries = 0; |
1035 | f_entries = 0; |
1036 | entry = eh_frame_start; |
1037 | |
1038 | while ((char *)entry < eh_frame_end) { |
1039 | p = entry; |
1040 | |
1041 | count = dwarf_entry_len(addr: p, len: &len); |
1042 | if (count == 0) { |
1043 | /* |
1044 | * We read a bogus length field value. There is |
1045 | * nothing we can do here apart from disabling |
1046 | * the DWARF unwinder. We can't even skip this |
1047 | * entry and move to the next one because 'len' |
1048 | * tells us where our next entry is. |
1049 | */ |
1050 | err = -EINVAL; |
1051 | goto out; |
1052 | } else |
1053 | p += count; |
1054 | |
1055 | /* initial length does not include itself */ |
1056 | end = p + len; |
1057 | |
1058 | entry_type = get_unaligned((u32 *)p); |
1059 | p += 4; |
1060 | |
1061 | if (entry_type == DW_EH_FRAME_CIE) { |
1062 | err = dwarf_parse_cie(entry, p, len, end, mod); |
1063 | if (err < 0) |
1064 | goto out; |
1065 | else |
1066 | c_entries++; |
1067 | } else { |
1068 | err = dwarf_parse_fde(entry, entry_type, start: p, len, |
1069 | end, mod); |
1070 | if (err < 0) |
1071 | goto out; |
1072 | else |
1073 | f_entries++; |
1074 | } |
1075 | |
1076 | entry = (char *)entry + len + 4; |
1077 | } |
1078 | |
1079 | printk(KERN_INFO "DWARF unwinder initialised: read %u CIEs, %u FDEs\n" , |
1080 | c_entries, f_entries); |
1081 | |
1082 | return 0; |
1083 | |
1084 | out: |
1085 | return err; |
1086 | } |
1087 | |
1088 | #ifdef CONFIG_MODULES |
1089 | int module_dwarf_finalize(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, |
1090 | struct module *me) |
1091 | { |
1092 | unsigned int i, err; |
1093 | unsigned long start, end; |
1094 | char *secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset; |
1095 | |
1096 | start = end = 0; |
1097 | |
1098 | for (i = 1; i < hdr->e_shnum; i++) { |
1099 | /* Alloc bit cleared means "ignore it." */ |
1100 | if ((sechdrs[i].sh_flags & SHF_ALLOC) |
1101 | && !strcmp(secstrings+sechdrs[i].sh_name, ".eh_frame" )) { |
1102 | start = sechdrs[i].sh_addr; |
1103 | end = start + sechdrs[i].sh_size; |
1104 | break; |
1105 | } |
1106 | } |
1107 | |
1108 | /* Did we find the .eh_frame section? */ |
1109 | if (i != hdr->e_shnum) { |
1110 | INIT_LIST_HEAD(list: &me->arch.cie_list); |
1111 | INIT_LIST_HEAD(list: &me->arch.fde_list); |
1112 | err = dwarf_parse_section(eh_frame_start: (char *)start, eh_frame_end: (char *)end, mod: me); |
1113 | if (err) { |
1114 | printk(KERN_WARNING "%s: failed to parse DWARF info\n" , |
1115 | me->name); |
1116 | return err; |
1117 | } |
1118 | } |
1119 | |
1120 | return 0; |
1121 | } |
1122 | |
1123 | /** |
1124 | * module_dwarf_cleanup - remove FDE/CIEs associated with @mod |
1125 | * @mod: the module that is being unloaded |
1126 | * |
1127 | * Remove any FDEs and CIEs from the global lists that came from |
1128 | * @mod's .eh_frame section because @mod is being unloaded. |
1129 | */ |
1130 | void module_dwarf_cleanup(struct module *mod) |
1131 | { |
1132 | struct dwarf_fde *fde, *ftmp; |
1133 | struct dwarf_cie *cie, *ctmp; |
1134 | unsigned long flags; |
1135 | |
1136 | spin_lock_irqsave(&dwarf_cie_lock, flags); |
1137 | |
1138 | list_for_each_entry_safe(cie, ctmp, &mod->arch.cie_list, link) { |
1139 | list_del(entry: &cie->link); |
1140 | rb_erase(&cie->node, &cie_root); |
1141 | kfree(objp: cie); |
1142 | } |
1143 | |
1144 | spin_unlock_irqrestore(lock: &dwarf_cie_lock, flags); |
1145 | |
1146 | spin_lock_irqsave(&dwarf_fde_lock, flags); |
1147 | |
1148 | list_for_each_entry_safe(fde, ftmp, &mod->arch.fde_list, link) { |
1149 | list_del(entry: &fde->link); |
1150 | rb_erase(&fde->node, &fde_root); |
1151 | kfree(objp: fde); |
1152 | } |
1153 | |
1154 | spin_unlock_irqrestore(lock: &dwarf_fde_lock, flags); |
1155 | } |
1156 | #endif /* CONFIG_MODULES */ |
1157 | |
1158 | /** |
1159 | * dwarf_unwinder_init - initialise the dwarf unwinder |
1160 | * |
1161 | * Build the data structures describing the .dwarf_frame section to |
1162 | * make it easier to lookup CIE and FDE entries. Because the |
1163 | * .eh_frame section is packed as tightly as possible it is not |
1164 | * easy to lookup the FDE for a given PC, so we build a list of FDE |
1165 | * and CIE entries that make it easier. |
1166 | */ |
1167 | static int __init dwarf_unwinder_init(void) |
1168 | { |
1169 | int err = -ENOMEM; |
1170 | |
1171 | dwarf_frame_cachep = kmem_cache_create("dwarf_frames" , |
1172 | sizeof(struct dwarf_frame), 0, |
1173 | SLAB_PANIC | SLAB_HWCACHE_ALIGN, NULL); |
1174 | |
1175 | dwarf_reg_cachep = kmem_cache_create("dwarf_regs" , |
1176 | sizeof(struct dwarf_reg), 0, |
1177 | SLAB_PANIC | SLAB_HWCACHE_ALIGN, NULL); |
1178 | |
1179 | dwarf_frame_pool = mempool_create_slab_pool(DWARF_FRAME_MIN_REQ, |
1180 | kc: dwarf_frame_cachep); |
1181 | if (!dwarf_frame_pool) |
1182 | goto out; |
1183 | |
1184 | dwarf_reg_pool = mempool_create_slab_pool(DWARF_REG_MIN_REQ, |
1185 | kc: dwarf_reg_cachep); |
1186 | if (!dwarf_reg_pool) |
1187 | goto out; |
1188 | |
1189 | err = dwarf_parse_section(__start_eh_frame, __stop_eh_frame, NULL); |
1190 | if (err) |
1191 | goto out; |
1192 | |
1193 | err = unwinder_register(&dwarf_unwinder); |
1194 | if (err) |
1195 | goto out; |
1196 | |
1197 | dwarf_unwinder_ready = 1; |
1198 | |
1199 | return 0; |
1200 | |
1201 | out: |
1202 | printk(KERN_ERR "Failed to initialise DWARF unwinder: %d\n" , err); |
1203 | dwarf_unwinder_cleanup(); |
1204 | return err; |
1205 | } |
1206 | early_initcall(dwarf_unwinder_init); |
1207 | |