Re: beta1

[Guy Harris <guy () alum mit edu>]

On Nov 16, 2003, at 4:59 PM, Neil Spring wrote:

> I've attached a patch against current cvs that adds _U_ to unused 
> function parameters.  The only warnings that remain on my system are 
> of signed / unsigned comparisons and no-op statements on a big-endian 
> machine.  I don't know if it's better to apply this patch or to remove 
> the unused parameters of functions that are not called indirectly via 
> function pointers.

In the case of "dhcp6_print()", the right thing to do, as far as I 
know, is to get rid of the port number arguments - if it matters 
whether it's being called on the server or client port, that should be 
passed as a Boolean flag.  I've checked in that change; in a quick look 
at RFC 3315, I didn't see any reason why we needed to know what port 
the packet was going to or from.

In the case of "udp6_cksum()" and "tcp6_cksum()", the value passed to 
it as "len" should be the length of the UDP or TCP packet (I've cleaned 
up the code in "print-ip6.c" to make sure it is), so it doesn't need to 
recompute that length itself.  I.e., that argument is now used in the 
code I checked in.

I also changed the IPv4 and IPv6 printers to set "snaplen" to the 
minimum of its current value and a pointer to the end of the IPv4/IPv6 
packet's payload, so that printers called by them don't run into any 
trailer or FCS data.

In the case of "icmp6_rrenum_print()", the "icmp6len" argument isn't 
needed because of the previous change, so I got rid of it.

In the case of "vjc_print()", I'm not sure what it's supposed to do 
with the "proto" argument, so I left it there, without a _U_, so that 
we'll get warnings reminding us that we should do something about it; I 
also put in a comment explaining why I'm not sure what it's supposed to 
do.

In the case of "Parse_fh()", it arguably *should* use "len", rather 
than blithely fetching data from the file handle - if the length is too 
short for a given type of file handle, or perhaps even if it's too 
*long* for that type of file handle, it shouldn't check for that type 
of file handle.  Therefore, I've again left the argument there, without 
a _U_, as a reminder that we should clean that up at some point.

These changes are in the main and 3.8 branches.

-
This is the TCPDUMP workers list. It is archived at
http://www.tcpdump.org/lists/workers/index.html
To unsubscribe use mailto:tcpdump-workers-request () tcpdump org?body=unsubscribe