Re: raw socket, IP_HDRINCL byteorder

Andrew Brown <[email protected]> wrote:
>>> I'd like to see a well-defined, portable raw socket interface (don't forget
>>> raw input, where the IP header is always included so you can't tell the
>>> desired behavior from IP_OHDRINCL / IP_HDRINCL).  I think that the right
>>> thing to do is introduce a new socket option that sets raw sockets to behave
>>> in the new way, and not try to overload IP_HDRINCL, whose behavior is
>>> too OS-specific.  IP_RAW_PKTS_HAVE_NETWORK_BYTE_ORDER, except that's too
>>> long =)
>>
>>Okay, if we have to pick a name, how about IP_HDR_RAW?
>
>that's just as undescriptive.  how about something that indicates that
>the headers in question are in network byte order?  isn't that the
>core issue here?
>
>IP_HDR_NETWORK?  IP_HDR_NBO?  gah.  i'm not so good at names...

IP_HDR_NETBYTEORDER?  or just IP_NETBYTEORDER?  How long is too long?

jdarrow

-- 
John Darrow - Senior Technical Specialist               Office: 630/752-5201
Computing Services, Wheaton College, Wheaton, IL 60187  Fax:    630/752-5968
Pager via email: [email protected]      Pager:  630/316-0707
Email: [email protected] (plain text please, no HTML or proprietary)

---------------------------------------------------------------------
The BSD APIs Discussion Mailing List
To unsubscribe:
send "unsubscribe bsd-api-discuss" to majordomo-DZEk9q7Sfju/3pe1ocb+swC/[email protected]