Re: nmap on sparc/linux

[Tom Duffy <tduffy () sun com>]

On Mon, 2002-12-16 at 17:09, Cliff Woolley wrote:

> I assume it's a valid assumption among this group that everybody knows
> that SIGBUS on a Sparc usually means an unaligned access was attempted,
> right?

As Cliff pointed out, it is probably an unaligned access error.  Here is
the culprit line after the preprocessor has had its way:

A = (ntohl(*(bpf_u_int32 *)&p[k]));

(gdb) disassemble ntohl
Dump of assembler code for function ntohl:
0x702ce540 <ntohl>:     nop 
0x702ce544 <ntohl+4>:   retl 
0x702ce548 <ntohl+8>:   nop 

ntohl is a noop because we are on a big endian architecture.

(gdb) print k
$15 = 30
(gdb) print &p[k]
$16 = (u_char *) 0x7267be "\n\aÆ\034\b"

Can we just make a char* an int* and dereference it?  Only if it is 4
byte aligned.

-tduffy

-- 
YOO-ESS-AYE! YOO-ESS-AYE!


---------------------------------------------------------------------
For help using this (nmap-dev) mailing list, send a blank email to 
nmap-dev-help () insecure org . List run by ezmlm-idx (www.ezmlm.org).