mirror of
https://github.com/radareorg/radare2.git
synced 2024-12-03 10:51:01 +00:00
323 lines
7.7 KiB
C
323 lines
7.7 KiB
C
/* radare - LGPL - Copyright 2012-2014 - pancake */
|
|
|
|
#include <r_anal.h>
|
|
|
|
#define MINLEN 1
|
|
static int is_string (const ut8 *buf, int size, int *len) {
|
|
int i;
|
|
if (size<1)
|
|
return 0;
|
|
if (size>3 && buf[0] &&!buf[1]&&buf[2]&&!buf[3]) {
|
|
*len = 1; // XXX: TODO: Measure wide string length
|
|
return 2; // is wide
|
|
}
|
|
for (i=0; i<size; i++) {
|
|
if (!buf[i] && i>MINLEN) {
|
|
*len = i;
|
|
return 1;
|
|
}
|
|
if (buf[i]<32 || buf[i]>127) {
|
|
// not ascii text
|
|
return 0;
|
|
}
|
|
if (buf[i]==10||buf[i]==13||buf[i]==9) {
|
|
continue;
|
|
}
|
|
if (!IS_PRINTABLE (buf[i])) {
|
|
*len = i;
|
|
return 0;
|
|
}
|
|
}
|
|
*len = i;
|
|
return 1;
|
|
}
|
|
|
|
static int is_number (const ut8 *buf, int endian, int size) {
|
|
ut64 n = r_mem_get_num (buf, size, endian);
|
|
return (n<UT32_MAX)? (int)n: 0;
|
|
}
|
|
|
|
static int is_null (const ut8 *buf, int size) {
|
|
const char zero[8] = {0,0,0,0,0,0,0,0};
|
|
return (!memcmp (buf, &zero, size))? 1: 0;
|
|
}
|
|
|
|
static int is_invalid (const ut8 *buf, int size) {
|
|
if (size<1) return 1;
|
|
if (size>8) size = 8;
|
|
return (!memcmp (buf, "\xff\xff\xff\xff\xff\xff\xff\xff", size))? 1: 0;
|
|
}
|
|
|
|
#define USE_IS_VALID_OFFSET 1
|
|
static ut64 is_pointer(RIOBind *iob, const ut8 *buf, int endian, int size) {
|
|
ut64 n;
|
|
ut8 buf2[32];
|
|
if (size > sizeof (buf2))
|
|
size = sizeof (buf2);
|
|
n = r_mem_get_num (buf, size, endian);
|
|
if (!n) return 1; // null pointer
|
|
#if USE_IS_VALID_OFFSET
|
|
int r = iob->is_valid_offset (iob->io, n);
|
|
return r? n: 0LL;
|
|
#else
|
|
// optimization to ignore very low and very high pointers
|
|
// this makes disasm 5x faster, but can result in some false positives
|
|
// we should compare with current offset, to avoid
|
|
// short/long references. and discard invalid ones
|
|
if (n<0x1000) return 0; // probably wrong
|
|
if (n>0xffffffffffffLL) return 0; // probably wrong
|
|
|
|
if (iob->read_at (iob->io, n, buf2, size) != size) return 0;
|
|
return is_invalid (buf2, size)? 0: n;
|
|
#endif
|
|
}
|
|
|
|
static int is_bin(const ut8 *buf, int size) {
|
|
// TODO: add more
|
|
if((size >= 4 && !memcmp (buf, "\xcf\xfa\xed\xfe", 4))
|
|
|| (size >= 4 && !memcmp (buf, "\x7e""ELF", 4))
|
|
|| (size >= 2 && !memcmp (buf, "MZ", 2)))
|
|
return 1;
|
|
return 0;
|
|
}
|
|
|
|
// TODO : add is_flag, is comment?
|
|
|
|
// XXX: optimize by removing all strlens here
|
|
R_API char *r_anal_data_to_string (RAnalData *d) {
|
|
int i, len, idx, mallocsz = 1024;
|
|
ut32 n32;
|
|
char *line;
|
|
|
|
if (!d) return NULL;
|
|
|
|
line = malloc (mallocsz);
|
|
snprintf (line, mallocsz, "0x%08"PFMT64x" ", d->addr);
|
|
n32 = (ut32)d->ptr;
|
|
len = R_MIN (d->len, 8);
|
|
for (i=0, idx = strlen (line); i<len; i++) {
|
|
int msz = mallocsz-idx;
|
|
if (msz>1) {
|
|
snprintf (line+idx, msz, "%02x", d->buf[i]);
|
|
idx += 2;
|
|
}
|
|
}
|
|
if (i>0 && d->len> len) {
|
|
int msz = mallocsz-idx;
|
|
snprintf (line+idx, msz, "..");
|
|
idx += 2;
|
|
msz -= 2;
|
|
}
|
|
strcat (line, " ");
|
|
idx += 2;
|
|
if ((mallocsz-idx)>12)
|
|
switch (d->type) {
|
|
case R_ANAL_DATA_TYPE_STRING:
|
|
snprintf (line+idx, mallocsz-idx, "string \"%s\"", d->str);
|
|
idx = strlen (line);
|
|
break;
|
|
case R_ANAL_DATA_TYPE_WIDE_STRING:
|
|
strcat (line, "wide string");
|
|
break;
|
|
case R_ANAL_DATA_TYPE_NUMBER:
|
|
if (n32 == d->ptr)
|
|
snprintf (line+idx, mallocsz-idx, "number %d 0x%x", n32, n32);
|
|
else snprintf (line+idx, mallocsz-idx, "number %"PFMT64d" 0x%"PFMT64x,
|
|
d->ptr, d->ptr);
|
|
break;
|
|
case R_ANAL_DATA_TYPE_POINTER:
|
|
strcat (line, "pointer ");
|
|
sprintf (line+strlen (line), " 0x%08"PFMT64x, d->ptr);
|
|
break;
|
|
case R_ANAL_DATA_TYPE_INVALID:
|
|
strcat (line, "invalid");
|
|
break;
|
|
case R_ANAL_DATA_TYPE_HEADER:
|
|
strcat (line, "header");
|
|
break;
|
|
case R_ANAL_DATA_TYPE_SEQUENCE:
|
|
strcat (line, "sequence");
|
|
break;
|
|
case R_ANAL_DATA_TYPE_PATTERN:
|
|
strcat (line, "pattern");
|
|
break;
|
|
case R_ANAL_DATA_TYPE_UNKNOWN:
|
|
strcat (line, "unknown");
|
|
break;
|
|
default:
|
|
strcat (line, "(null)");
|
|
break;
|
|
}
|
|
return line;
|
|
}
|
|
|
|
R_API RAnalData *r_anal_data_new_string (ut64 addr, const char *p, int len, int type) {
|
|
RAnalData *ad = R_NEW0 (RAnalData);
|
|
ad->str = NULL;
|
|
ad->addr = addr;
|
|
ad->type = type;
|
|
if (len == 0)
|
|
len = strlen (p);
|
|
if (type == R_ANAL_DATA_TYPE_WIDE_STRING) {
|
|
/* TODO: add support for wide strings */
|
|
//eprintf ("r_anal_data_new_string: wide string not supported yet\n");
|
|
} else {
|
|
ad->str = malloc (len+1);
|
|
memcpy (ad->str, p, len);
|
|
ad->str[len] = 0;
|
|
ad->buf = malloc (len+1);
|
|
memcpy (ad->buf, ad->str, len+1);
|
|
ad->len = len+1; // string length + \x00
|
|
}
|
|
ad->ptr = 0L;
|
|
return ad;
|
|
}
|
|
|
|
R_API RAnalData *r_anal_data_new (ut64 addr, int type, ut64 n, const ut8 *buf, int len) {
|
|
RAnalData *ad = R_NEW0 (RAnalData);
|
|
int l = R_MIN (len, 8);
|
|
ad->buf = (ut8*) &(ad->sbuf);
|
|
memset (ad->buf, 0, 8);
|
|
if (l<1) {
|
|
r_anal_data_free (ad);
|
|
return NULL;
|
|
}
|
|
if (buf) {
|
|
memcpy (ad->buf, buf, l);
|
|
}
|
|
ad->addr = addr;
|
|
ad->type = type;
|
|
ad->str = NULL;
|
|
switch (type) {
|
|
case R_ANAL_DATA_TYPE_PATTERN:
|
|
case R_ANAL_DATA_TYPE_SEQUENCE:
|
|
ad->len = len;
|
|
break;
|
|
default:
|
|
ad->len = l;
|
|
}
|
|
ad->ptr = n;
|
|
return ad;
|
|
}
|
|
|
|
R_API void r_anal_data_free (RAnalData *d) {
|
|
if (d) {
|
|
if (d->buf != (ut8*)&(d->sbuf)) free (d->buf);
|
|
if (d->str != NULL) free (d->str);
|
|
free (d);
|
|
}
|
|
}
|
|
|
|
R_API RAnalData *r_anal_data (RAnal *anal, ut64 addr, const ut8 *buf, int size) {
|
|
ut64 dst = 0;
|
|
int n, nsize = 0;
|
|
int bits = anal->bits;
|
|
int endi = !anal->big_endian;
|
|
int word = R_MIN (8, bits/8);
|
|
|
|
if (size<4)
|
|
return NULL;
|
|
if (size >= word && is_invalid (buf, word))
|
|
return r_anal_data_new (addr, R_ANAL_DATA_TYPE_INVALID,
|
|
-1, buf, word);
|
|
{
|
|
int i, len = R_MIN (size, 64);
|
|
int is_pattern = 0;
|
|
int is_sequence = 0;
|
|
char ch = buf[0];
|
|
char ch2 = ch+1;
|
|
for (i=1; i<len; i++) {
|
|
if (ch2 == buf[i]) {
|
|
ch2++;
|
|
is_sequence++;
|
|
} else is_sequence = 0;
|
|
if (ch==buf[i]) {
|
|
is_pattern++;
|
|
}
|
|
}
|
|
//eprintf ("%d %d %d %d\n", is_sequence, is_pattern , len, size);
|
|
if (is_sequence>len-2) {
|
|
return r_anal_data_new (addr, R_ANAL_DATA_TYPE_SEQUENCE, -1,
|
|
buf, is_sequence);
|
|
}
|
|
if (is_pattern>len-2) {
|
|
return r_anal_data_new (addr, R_ANAL_DATA_TYPE_PATTERN, -1,
|
|
buf, is_pattern);
|
|
}
|
|
}
|
|
if (size >= word && is_null (buf, word))
|
|
return r_anal_data_new (addr, R_ANAL_DATA_TYPE_NULL,
|
|
-1, buf, word);
|
|
if (is_bin (buf, size))
|
|
return r_anal_data_new (addr, R_ANAL_DATA_TYPE_HEADER, -1,
|
|
buf, word);
|
|
if (size>=word) {
|
|
dst = is_pointer (&anal->iob, buf, endi, word);
|
|
if (dst) return r_anal_data_new (addr,
|
|
R_ANAL_DATA_TYPE_POINTER, dst, buf, word);
|
|
}
|
|
switch (is_string (buf, size, &nsize)) {
|
|
case 1: return r_anal_data_new_string (addr, (const char *)buf,
|
|
nsize, R_ANAL_DATA_TYPE_STRING);
|
|
case 2: return r_anal_data_new_string (addr, (const char *)buf,
|
|
nsize, R_ANAL_DATA_TYPE_WIDE_STRING);
|
|
}
|
|
if (size >= word) {
|
|
n = is_number (buf, endi, word);
|
|
if (n) return r_anal_data_new (addr, R_ANAL_DATA_TYPE_NUMBER,
|
|
n, buf, word);
|
|
}
|
|
return r_anal_data_new (addr, R_ANAL_DATA_TYPE_UNKNOWN, dst,
|
|
buf, R_MIN(word, size));
|
|
}
|
|
|
|
R_API const char *r_anal_data_kind (RAnal *a, ut64 addr, const ut8 *buf, int len) {
|
|
int inv = 0;
|
|
int unk = 0;
|
|
int str = 0;
|
|
int num = 0;
|
|
int i, j;
|
|
RAnalData *data;
|
|
int word = a->bits /8;
|
|
for (i = j = 0; i<len; j++) {
|
|
if (str && !buf[i])
|
|
str ++;
|
|
data = r_anal_data (a, addr+i, buf+i, len-i);
|
|
if (data == NULL) {
|
|
i+= word;
|
|
continue;
|
|
}
|
|
switch (data->type) {
|
|
case R_ANAL_DATA_TYPE_INVALID:
|
|
inv++;
|
|
i += word;
|
|
break;
|
|
case R_ANAL_DATA_TYPE_NUMBER:
|
|
if (data->ptr> 1000) num++;
|
|
i += word;
|
|
break;
|
|
case R_ANAL_DATA_TYPE_UNKNOWN:
|
|
unk++;
|
|
i += word;
|
|
break;
|
|
case R_ANAL_DATA_TYPE_STRING:
|
|
if (data->len>0) {
|
|
i += data->len; //strlen ((const char*)buf+i)+1;
|
|
} else i+=word;
|
|
str++;
|
|
break;
|
|
default:
|
|
i += word;
|
|
}
|
|
r_anal_data_free (data);
|
|
}
|
|
//eprintf ("%d %d %d %d\n", inv, unk, num, str);
|
|
if (j<1) return "unknown";
|
|
if ((inv*100/j)>60) return "invalid";
|
|
if ((unk*100/j)>60) return "code";
|
|
if ((num*100/j)>60) return "code";
|
|
//return "text";
|
|
if ((str*100/j)>40) return "text";
|
|
return "data";
|
|
}
|