>   >>[*] I had planned to wait for dynamic-sysctl before adding a sysctl
>   >>tree for crypto; if dynamic sysctl is deferred I may revisit that.
>   >
>   >it's not very deferred...i'm just polishing it, doing testing,
>   >building kernels, etc.
>   >
>   >can you wait a few more days?
>   
>   dynamic sysctl was committed about ten days ago, and i think most of
>   the problems with it have been shaken out at this point, so besides
>   the fact that i haven't finished all the documentation yet...
>
>note that this really matters, but i noticed that new userland &
>old kernel doesn't like sysctl very much :-)

that might be true, but only for the sysctl(8) binary itself, since it
now relies on auto-discovery in order to function.  there isn't
anything else that's broken like that...is there?

besides...we can hardly provide forwards compatibility in the kernel
for newer userlands that don't exist yet...can we?  :)

-- 
|-----< "CODE WARRIOR" >-----|
codewarrior@daemon.org             * "ah!  i see you have the internet
twofsonet@graffiti.com (Andrew Brown)                that goes *ping*!"
werdna@squooshy.com       * "information is power -- share the wealth."