Rui Paulo rpaulo at
Tue Sep 4 11:22:44 PDT 2007

Andre Oppermann wrote:
> Rui Paulo wrote:
>> Hi,
>> I'm working on TCP ECN support and I would like to kill these defines 
>> from netinet/ip.h
>> #if 1
>> /* ECN RFC3168 obsoletes RFC2481, and these will be deprecated soon. */
>> #define IPTOS_CE                0x01
>> #define IPTOS_ECT               0x02
>> #endif
>> The are outdated and should no longer be used because that RFC has 
>> been obsoleted. RFC 3168 uses a bit different bits:
>> #define IPTOS_ECN_NOTECT        0x00    /* not-ECT */
>> #define IPTOS_ECN_ECT1          0x01    /* ECN-capable transport (1) */
>> #define IPTOS_ECN_ECT0          0x02    /* ECN-capable transport (0) */
>> #define IPTOS_ECN_CE            0x03    /* congestion experienced */
>> #define IPTOS_ECN_MASK          0x03    /* ECN field mask */
>> The only consumer of the RFC 2481 defines is sbin/ipfw/ipfw2.c and 
>> luigi@ accepted the attached patch.
>> Does anyone have any objection against the removal ?
> No.  Please produce a complete unified diff patch including both
> the changes to ip.h and ipfw2.c, a sufficiently verbose commit
> message, the result a complete make universe build run with the
> patch applied to otherwise clean sources and send it to re@ and
> me in CC.  If approved I commit it for you.

Well, I was asking for comments regarding on the usage of these flags. I 
was hoping to commit ip.h along with TCP ECN.

This doesn't really need to be before the branch, I think.
Rui Paulo

More information about the freebsd-net mailing list