http://bugs.wireshark.org/bugzilla/show_bug.cgi?id=1956
ellocodelassembler@xxxxxxxxx changed:
What |Removed |Added
----------------------------------------------------------------------------
OS/Version|Windows XP |All
------- Comment #3 from ellocodelassembler@xxxxxxxxx 2007-11-06 04:40 GMT -------
Now I've tried on a Debian Linux box.
Take a look at this when I use "ip.addr & 1.2.3.4":
(gdb) frame
#0 cmp_bytes_bitwise_and (fv_a=0x8a672d0, fv_b=0x8800e80) at ftype-ipv4.c:178
178 if (b->len != a->len) {
(gdb) info args
fv_a = (fvalue_t *) 0x8a672d0
fv_b = (fvalue_t *) 0x8800e80
(gdb) info locals
a = (GByteArray *) 0xc0a800fc
b = (GByteArray *) 0x1020304
i = <value optimized out>
p_a = <value optimized out>
p_b = <value optimized out>
Do you see somewhat strange pointers? ;)
c0.a8.00.fc = 192.168.0.252
01.02.03.04 = 1.2.3.4
I have replaced the function completely with the code below but since the
incoming data type is so much unexpected (and because this is my first time
digging in WireShark sources), I can't assure that this patch is valid and
although it worked well with the tested expression, I really don't know if the
same function can be reached from some code that actually sends the expected
type.
static gboolean cmp_bytes_bitwise_and(fvalue_t *fv_a, fvalue_t *fv_b)
{
guint i, and;
gboolean res;
and = (guint)fv_a->value.bytes & (guint)fv_b->value.bytes;
/* I can't remember the magic trick so I use a looped version of the
find-zero-byte-on-dword algo */
for (i = 4, res = TRUE; i; i--){
res &= (char)and != 0;
and >>= 8;
}
return res;
}
I also copy the original function for reference:
static gboolean cmp_bytes_bitwise_and(fvalue_t *fv_a, fvalue_t *fv_b)
{
GByteArray *a = fv_a->value.bytes;
GByteArray *b = fv_b->value.bytes;
guint i = 0;
unsigned char *p_a, *p_b;
if (b->len != a->len) {
return FALSE;
}
p_a = a->data;
p_b = b->data;
while (i < b->len) {
if (p_a[i] & p_b[i])
i++;
else
return FALSE;
}
return TRUE;
}
Cheers
--
Configure bugmail: http://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.
