In article <20050320094253.GB56296@sigusr1.org>,
Kentaro A. Kurahone <kurahone@sigusr1.org> wrote:
>-=-=-=-=-=-
>
>On Thu, Mar 17, 2005 at 07:57:30AM +0000, Kentaro A. Kurahone wrote:
>> I went and ported over the changes that we were missing, I'd appreciate
>> some feedback/review before I go and check it in. (Test rig is at work so
>> it's kind of hard to test extensively right now.)
> 
>Ok, I just went and cleaned up the sysctl stuff in that patch, mainly
>by moving off the SACK related options into their own node.
>
>net.inet.tcp.sack becomes net.inet.tcp.sack.enable under this new scheme.
>Will anybody here have any objections to that?  (Since it's a relatively new
>feature, I don't think that many people are using the sysctl yet but, it never
>hurts to ask.)

Are there other nodes under net.inet.tcp.sack? If the enable node is the
only one, I'd say no, because there is no prior art in naming booleans
have an enable child. If there are other nodes, I'd say yes.

christos