aboutsummaryrefslogtreecommitdiffstats
path: root/scripts/kallsyms.c
blob: fe11df83d1fc91b3e6de8a7bdc4a06e25f3d0e2f (plain) (blame)
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
/* Generate assembler source containing symbol information
 *
 * Copyright 2002       by Kai Germaschewski
 *
 * This software may be used and distributed according to the terms
 * of the GNU General Public License, incorporated herein by reference.
 *
 * Usage: nm -n vmlinux | scripts/kallsyms [--all-symbols] > symbols.S
 *
 * ChangeLog:
 *
 * (25/Aug/2004) Paulo Marques <pmarques@grupopie.com>
 *      Changed the compression method from stem compression to "table lookup"
 *      compression
 *
 *      Table compression uses all the unused char codes on the symbols and
 *  maps these to the most used substrings (tokens). For instance, it might
 *  map char code 0xF7 to represent "write_" and then in every symbol where
 *  "write_" appears it can be replaced by 0xF7, saving 5 bytes.
 *      The used codes themselves are also placed in the table so that the
 *  decompresion can work without "special cases".
 *      Applied to kernel symbols, this usually produces a compression ratio
 *  of about 50%.
 *
 */

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>

/* maximum token length used. It doesn't pay to increase it a lot, because
 * very long substrings probably don't repeat themselves too often. */
#define MAX_TOK_SIZE		11
#define KSYM_NAME_LEN		127

/* we use only a subset of the complete symbol table to gather the token count,
 * to speed up compression, at the expense of a little compression ratio */
#define WORKING_SET		1024

/* first find the best token only on the list of tokens that would profit more
 * than GOOD_BAD_THRESHOLD. Only if this list is empty go to the "bad" list.
 * Increasing this value will put less tokens on the "good" list, so the search
 * is faster. However, if the good list runs out of tokens, we must painfully
 * search the bad list. */
#define GOOD_BAD_THRESHOLD	10

/* token hash parameters */
#define HASH_BITS		18
#define HASH_TABLE_SIZE		(1 << HASH_BITS)
#define HASH_MASK		(HASH_TABLE_SIZE - 1)
#define HASH_BASE_OFFSET	2166136261U
#define HASH_FOLD(a)		((a)&(HASH_MASK))

/* flags to mark symbols */
#define SYM_FLAG_VALID		1
#define SYM_FLAG_SAMPLED	2

struct sym_entry {
	unsigned long long addr;
	char type;
	unsigned char flags;
	unsigned char len;
	unsigned char *sym;
};


static struct sym_entry *table;
static int size, cnt;
static unsigned long long _stext, _etext, _sinittext, _einittext;
static int all_symbols = 0;
static char symbol_prefix_char = '\0';

struct token {
	unsigned char data[MAX_TOK_SIZE];
	unsigned char len;
	/* profit: the number of bytes that could be saved by inserting this
	 * token into the table */
	int profit;
	struct token *next;	/* next token on the hash list */
	struct token *right;	/* next token on the good/bad list */
	struct token *left;    /* previous token on the good/bad list */
	struct token *smaller; /* token that is less one letter than this one */
	};

struct token bad_head, good_head;
struct token *hash_table[HASH_TABLE_SIZE];

/* the table that holds the result of the compression */
unsigned char best_table[256][MAX_TOK_SIZE+1];
unsigned char best_table_len[256];


static void
usage(void)
{
	fprintf(stderr, "Usage: kallsyms [--all-symbols] [--symbol-prefix=<prefix char>] < in.map > out.S\n");
	exit(1);
}

/*
 * This ignores the intensely annoying "mapping symbols" found
 * in ARM ELF files: $a, $t and $d.
 */
static inline int
is_arm_mapping_symbol(const char *str)
{
	return str[0] == '$' && strchr("atd", str[1])
	       && (str[2] == '\0' || str[2] == '.');
}

static int
read_symbol(FILE *in, struct sym_entry *s)
{
	char str[500];
	char *sym;
	int rc;

	rc = fscanf(in, "%llx %c %499s\n", &s->addr, &s->type, str);
	if (rc != 3) {
		if (rc != EOF) {
			/* skip line */
			fgets(str, 500, in);
		}
		return -1;
	}

	sym = str;
	/* skip prefix char */
	if (symbol_prefix_char && str[0] == symbol_prefix_char)
		sym++;

	/* Ignore most absolute/undefined (?) symbols. */
	if (strcmp(sym, "_stext") == 0)
		_stext = s->addr;
	else if (strcmp(sym, "_etext") == 0)
		_etext = s->addr;
	else if (strcmp(sym, "_sinittext") == 0)
		_sinittext = s->addr;
	else if (strcmp(sym, "_einittext") == 0)
		_einittext = s->addr;
	else if (toupper(s->type) == 'A')
	{
		/* Keep these useful absolute symbols */
		if (strcmp(sym, "__kernel_syscall_via_break") &&
		    strcmp(sym, "__kernel_syscall_via_epc") &&
		    strcmp(sym, "__kernel_sigtramp") &&
		    strcmp(sym, "__gp"))
			return -1;

	}
	else if (toupper(s->type) == 'U' ||
		 is_arm_mapping_symbol(sym))
		return -1;

	/* include the type field in the symbol name, so that it gets
	 * compressed together */
	s->len = strlen(str) + 1;
	s->sym = (char *) malloc(s->len + 1);
	strcpy(s->sym + 1, str);
	s->sym[0] = s->type;

	return 0;
}

static int
symbol_valid(struct sym_entry *s)
{
	/* Symbols which vary between passes.  Passes 1 and 2 must have
	 * identical symbol lists.  The kallsyms_* symbols below are only added
	 * after pass 1, they would be included in pass 2 when --all-symbols is
	 * specified so exclude them to get a stable symbol list.
	 */
	static char *special_symbols[] = {
		"kallsyms_addresses",
		"kallsyms_num_syms",
		"kallsyms_names",
		"kallsyms_markers",
		"kallsyms_token_table",
		"kallsyms_token_index",

	/* Exclude linker generated symbols which vary between passes */
		"_SDA_BASE_",		/* ppc */
		"_SDA2_BASE_",		/* ppc */
		NULL };
	int i;
	int offset = 1;

	/* skip prefix char */
	if (symbol_prefix_char && *(s->sym + 1) == symbol_prefix_char)
		offset++;

	/* if --all-symbols is not specified, then symbols outside the text
	 * and inittext sections are discarded */
	if (!all_symbols) {
		if ((s->addr < _stext || s->addr > _etext)
		    && (s->addr < _sinittext || s->addr > _einittext))
			return 0;
		/* Corner case.  Discard any symbols with the same value as
		 * _etext or _einittext, they can move between pass 1 and 2
		 * when the kallsyms data is added.  If these symbols move then
		 * they may get dropped in pass 2, which breaks the kallsyms
		 * rules.
		 */
		if ((s->addr == _etext && strcmp(s->sym + offset, "_etext")) ||
		    (s->addr == _einittext && strcmp(s->sym + offset, "_einittext")))
			return 0;
	}

	/* Exclude symbols which vary between passes. */
	if (strstr(s->sym + offset, "_compiled."))
		return 0;

	for (i = 0; special_symbols[i]; i++)
		if( strcmp(s->sym + offset, special_symbols[i]) == 0 )
			return 0;

	return 1;
}

static void
read_map(FILE *in)
{
	while (!feof(in)) {
		if (cnt >= size) {
			size += 10000;
			table = realloc(table, sizeof(*table) * size);
			if (!table) {
				fprintf(stderr, "out of memory\n");
				exit (1);
			}
		}
		if (read_symbol(in, &table[cnt]) == 0)
			cnt++;
	}
}

static void output_label(char *label)
{
	if (symbol_prefix_char)
		printf(".globl %c%s\n", symbol_prefix_char, label);
	else
		printf(".globl %s\n", label);
	printf("\tALGN\n");
	if (symbol_prefix_char)
		printf("%c%s:\n", symbol_prefix_char, label);
	else
		printf("%s:\n", label);
}

/* uncompress a compressed symbol. When this function is called, the best table
 * might still be compressed itself, so the function needs to be recursive */
static int expand_symbol(unsigned char *data, int len, char *result)
{
	int c, rlen, total=0;

	while (len) {
		c = *data;
		/* if the table holds a single char that is the same as the one
		 * we are looking for, then end the search */
		if (best_table[c][0]==c && best_table_len[c]==1) {
			*result++ = c;
			total++;
		} else {
			/* if not, recurse and expand */
			rlen = expand_symbol(best_table[c], best_table_len[c], result);
			total += rlen;
			result += rlen;
		}
		data++;
		len--;
	}
	*result=0;

	return total;
}

static void
write_src(void)
{
	int i, k, off, valid;
	unsigned int best_idx[256];
	unsigned int *markers;
	char buf[KSYM_NAME_LEN+1];

	printf("#include <asm/types.h>\n");
	printf("#if BITS_PER_LONG == 64\n");
	printf("#define PTR .quad\n");
	printf("#define ALGN .align 8\n");
	printf("#else\n");
	printf("#define PTR .long\n");
	printf("#define ALGN .align 4\n");
	printf("#endif\n");

	printf(".data\n");

	output_label("kallsyms_addresses");
	valid = 0;
	for (i = 0; i < cnt; i++) {
		if (table[i].flags & SYM_FLAG_VALID) {
			printf("\tPTR\t%#llx\n", table[i].addr);
			valid++;
		}
	}
	printf("\n");

	output_label("kallsyms_num_syms");
	printf("\tPTR\t%d\n", valid);
	printf("\n");

	/* table of offset markers, that give the offset in the compressed stream
	 * every 256 symbols */
	markers = (unsigned int *) malloc(sizeof(unsigned int)*((valid + 255) / 256));

	output_label("kallsyms_names");
	valid = 0;
	off = 0;
	for (i = 0; i < cnt; i++) {

		if (!table[i].flags & SYM_FLAG_VALID)
			continue;

		if ((valid & 0xFF) == 0)
			markers[valid >> 8] = off;

		printf("\t.byte 0x%02x", table[i].len);
		for (k = 0; k < table[i].len; k++)
			printf(", 0x%02x", table[i].sym[k]);
		printf("\n");

		off += table[i].len + 1;
		valid++;
	}
	printf("\n");

	output_label("kallsyms_markers");
	for (i = 0; i < ((valid + 255) >> 8); i++)
		printf("\tPTR\t%d\n", markers[i]);
	printf("\n");

	free(markers);

	output_label("kallsyms_token_table");
	off = 0;
	for (i = 0; i < 256; i++) {
		best_idx[i] = off;
		expand_symbol(best_table[i],best_table_len[i],buf);
		printf("\t.asciz\t\"%s\"\n", buf);
		off += strlen(buf) + 1;
	}
	printf("\n");

	output_label("kallsyms_token_index");
	for (i = 0; i < 256; i++)
		printf("\t.short\t%d\n", best_idx[i]);
	printf("\n");
}


/* table lookup compression functions */

static inline unsigned int rehash_token(unsigned int hash, unsigned char data)
{
	return ((hash * 16777619) ^ data);
}

static unsigned int hash_token(unsigned char *data, int len)
{
	unsigned int hash=HASH_BASE_OFFSET;
	int i;

	for (i = 0; i < len; i++)
		hash = rehash_token(hash, data[i]);

	return HASH_FOLD(hash);
}

/* find a token given its data and hash value */
static struct token *find_token_hash(unsigned char *data, int len, unsigned int hash)
{
	struct token *ptr;

	ptr = hash_table[hash];

	while (ptr) {
		if ((ptr->len == len) && (memcmp(ptr->data, data, len) == 0))
			return ptr;
		ptr=ptr->next;
	}

	return NULL;
}

static inline void insert_token_in_group(struct token *head, struct token *ptr)
{
	ptr->right = head->right;
	ptr->right->left = ptr;
	head->right = ptr;
	ptr->left = head;
}

static inline void remove_token_from_group(struct token *ptr)
{
	ptr->left->right = ptr->right;
	ptr->right->left = ptr->left;
}


/* build the counts for all the tokens that start with "data", and have lenghts
 * from 2 to "len" */
static void learn_token(unsigned char *data, int len)
{
	struct token *ptr,*last_ptr;
	int i, newprofit;
	unsigned int hash = HASH_BASE_OFFSET;
	unsigned int hashes[MAX_TOK_SIZE + 1];

	if (len > MAX_TOK_SIZE)
		len = MAX_TOK_SIZE;

	/* calculate and store the hash values for all the sub-tokens */
	hash = rehash_token(hash, data[0]);
	for (i = 2; i <= len; i++) {
		hash = rehash_token(hash, data[i-1]);
		hashes[i] = HASH_FOLD(hash);
	}

	last_ptr = NULL;
	ptr = NULL;

	for (i = len; i >= 2; i--) {
		hash = hashes[i];

		if (!ptr) ptr = find_token_hash(data, i, hash);

		if (!ptr) {
			/* create a new token entry */
			ptr = (struct token *) malloc(sizeof(*ptr));

			memcpy(ptr->data, data, i);
			ptr->len = i;

			/* when we create an entry, it's profit is 0 because
			 * we also take into account the size of the token on
			 * the compressed table. We then subtract GOOD_BAD_THRESHOLD
			 * so that the test to see if this token belongs to
			 * the good or bad list, is a comparison to zero */
			ptr->profit = -GOOD_BAD_THRESHOLD;

			ptr->next = hash_table[hash];
			hash_table[hash] = ptr;

			insert_token_in_group(&bad_head, ptr);

			ptr->smaller = NULL;
		} else {
			newprofit = ptr->profit + (ptr->len - 1);
			/* check to see if this token needs to be moved to a
			 * different list */
			if((ptr->profit < 0) && (newprofit >= 0)) {
				remove_token_from_group(ptr);
				insert_token_in_group(&good_head,ptr);
			}
			ptr->profit = newprofit;
		}

		if (last_ptr) last_ptr->smaller = ptr;
		last_ptr = ptr;

		ptr = ptr->smaller;
	}
}

/* decrease the counts for all the tokens that start with "data", and have lenghts
 * from 2 to "len". This function is much simpler than learn_token because we have
 * more guarantees (tho tokens exist, the ->smaller pointer is set, etc.)
 * The two separate functions exist only because of compression performance */
static void forget_token(unsigned char *data, int len)
{
	struct token *ptr;
	int i, newprofit;
	unsigned int hash=0;

	if (len > MAX_TOK_SIZE) len = MAX_TOK_SIZE;

	hash = hash_token(data, len);
	ptr = find_token_hash(data, len, hash);

	for (i = len; i >= 2; i--) {

		newprofit = ptr->profit - (ptr->len - 1);
		if ((ptr->profit >= 0) && (newprofit < 0)) {
			remove_token_from_group(ptr);
			insert_token_in_group(&bad_head, ptr);
		}
		ptr->profit=newprofit;

		ptr=ptr->smaller;
	}
}

/* count all the possible tokens in a symbol */
static void learn_symbol(unsigned char *symbol, int len)
{
	int i;

	for (i = 0; i < len - 1; i++)
		learn_token(symbol + i, len - i);
}

/* decrease the count for all the possible tokens in a symbol */
static void forget_symbol(unsigned char *symbol, int len)
{
	int i;

	for (i = 0; i < len - 1; i++)
		forget_token(symbol + i, len - i);
}

/* set all the symbol flags and do the initial token count */
static void build_initial_tok_table(void)
{
	int i, use_it, valid;

	valid = 0;
	for (i = 0; i < cnt; i++) {
		table[i].flags = 0;
		if ( symbol_valid(&table[i]) ) {
			table[i].flags |= SYM_FLAG_VALID;
			valid++;
		}
	}

	use_it = 0;
	for (i = 0; i < cnt; i++) {

		/* subsample the available symbols. This method is almost like
		 * a Bresenham's algorithm to get uniformly distributed samples
		 * across the symbol table */
		if (table[i].flags & SYM_FLAG_VALID) {

			use_it += WORKING_SET;

			if (use_it >= valid) {
				table[i].flags |= SYM_FLAG_SAMPLED;
				use_it -= valid;
			}
		}
		if (table[i].flags & SYM_FLAG_SAMPLED)
			learn_symbol(table[i].sym, table[i].len);
	}
}

/* replace a given token in all the valid symbols. Use the sampled symbols
 * to update the counts */
static void compress_symbols(unsigned char *str, int tlen, int idx)
{
	int i, len, learn, size;
	unsigned char *p;

	for (i = 0; i < cnt; i++) {

		if (!(table[i].flags & SYM_FLAG_VALID)) continue;

		len = table[i].len;
		learn = 0;
		p = table[i].sym;

		do {
			/* find the token on the symbol */
			p = (unsigned char *) strstr((char *) p, (char *) str);
			if (!p) break;

			if (!learn) {
				/* if this symbol was used to count, decrease it */
				if (table[i].flags & SYM_FLAG_SAMPLED)
					forget_symbol(table[i].sym, len);
				learn = 1;
			}

			*p = idx;
			size = (len - (p - table[i].sym)) - tlen + 1;
			memmove(p + 1, p + tlen, size);
			p++;
			len -= tlen - 1;

		} while (size >= tlen);

		if(learn) {
			table[i].len = len;
			/* if this symbol was used to count, learn it again */
			if(table[i].flags & SYM_FLAG_SAMPLED)
				learn_symbol(table[i].sym, len);
		}
	}
}

/* search the token with the maximum profit */
static struct token *find_best_token(void)
{
	struct token *ptr,*best,*head;
	int bestprofit;

	bestprofit=-10000;

	/* failsafe: if the "good" list is empty search from the "bad" list */
	if(good_head.right == &good_head) head = &bad_head;
	else head = &good_head;

	ptr = head->right;
	best = NULL;
	while (ptr != head) {
		if (ptr->profit > bestprofit) {
			bestprofit = ptr->profit;
			best = ptr;
		}
		ptr = ptr->right;
	}

	return best;
}

/* this is the core of the algorithm: calculate the "best" table */
static void optimize_result(void)
{
	struct token *best;
	int i;

	/* using the '\0' symbol last allows compress_symbols to use standard
	 * fast string functions */
	for (i = 255; i >= 0; i--) {

		/* if this table slot is empty (it is not used by an actual
		 * original char code */
		if (!best_table_len[i]) {

			/* find the token with the breates profit value */
			best = find_best_token();

			/* place it in the "best" table */
			best_table_len[i] = best->len;
			memcpy(best_table[i], best->data, best_table_len[i]);
			/* zero terminate the token so that we can use strstr
			   in compress_symbols */
			best_table[i][best_table_len[i]]='\0';

			/* replace this token in all the valid symbols */
			compress_symbols(best_table[i], best_table_len[i], i);
		}
	}
}

/* start by placing the symbols that are actually used on the table */
static void insert_real_symbols_in_table(void)
{
	int i, j, c;

	memset(best_table, 0, sizeof(best_table));
	memset(best_table_len, 0, sizeof(best_table_len));

	for (i = 0; i < cnt; i++) {
		if (table[i].flags & SYM_FLAG_VALID) {
			for (j = 0; j < table[i].len; j++) {
				c = table[i].sym[j];
				best_table[c][0]=c;
				best_table_len[c]=1;
			}
		}
	}
}

static void optimize_token_table(void)
{
	memset(hash_table, 0, sizeof(hash_table));

	good_head.left = &good_head;
	good_head.right = &good_head;

	bad_head.left = &bad_head;
	bad_head.right = &bad_head;

	build_initial_tok_table();

	insert_real_symbols_in_table();

	/* When valid symbol is not registered, exit to error */
	if (good_head.left == good_head.right &&
	    bad_head.left == bad_head.right) {
		fprintf(stderr, "No valid symbol.\n");
		exit(1);
	}

	optimize_result();
}


int
main(int argc, char **argv)
{
	if (argc >= 2) {
		int i;
		for (i = 1; i < argc; i++) {
			if(strcmp(argv[i], "--all-symbols") == 0)
				all_symbols = 1;
			else if (strncmp(argv[i], "--symbol-prefix=", 16) == 0) {
				char *p = &argv[i][16];
				/* skip quote */
				if ((*p == '"' && *(p+2) == '"') || (*p == '\'' && *(p+2) == '\''))
					p++;
				symbol_prefix_char = *p;
			} else
				usage();
		}
	} else if (argc != 1)
		usage();

	read_map(stdin);
	optimize_token_table();
	write_src();

	return 0;
}