On Thu, Dec 02, 1999 at 02:18:10PM -0600, phil_t@xxxxxxxxxxxxx wrote:
>
>
> I put the SNMP build issue off to the side for now since I can get around
> it. Now I'm up and running (almost) with 0-7.9. This is the first time
> I've run against captures with NTP packets since the NTP support has been
> added.
>
> There are a couple of problems with the NTP decoding:
>
> 1) It gets a SIGBUS on my Solaris box as it tries to do an unaligned access
> referencing a long value in the packet instead of copying those bytes to an
> aligned location then referencing it. I haven't finished looking at it,
> but most of the payload portion of the packet should probably be copied out
> to avoid this type of problem.
Right. Just in case you (or anyone else reading on this list) haven't seen them,
packet.h has some macros to help doing that. They take a pointer, and construct
either 16-bit integer or a 32-bit integer from them the data pointed to:
#define pntohs(p) ((guint16) \
((guint16)*((guint8 *)p+0)<<8| \
(guint16)*((guint8 *)p+1)<<0))
#define pntohl(p) ((guint32)*((guint8 *)p+0)<<24| \
(guint32)*((guint8 *)p+1)<<16| \
(guint32)*((guint8 *)p+2)<<8| \
(guint32)*((guint8 *)p+3)<<0)
#define pletohs(p) ((guint16) \
((guint16)*((guint8 *)p+1)<<8| \
(guint16)*((guint8 *)p+0)<<0))
#define pletohl(p) ((guint32)*((guint8 *)p+3)<<24| \
(guint32)*((guint8 *)p+2)<<16| \
(guint32)*((guint8 *)p+1)<<8| \
(guint32)*((guint8 *)p+0)<<0)
--gilbert
