On Feb 22, 2005, at 10:01 PM, George Adkins wrote:
>> Hmm... SMP is more important then X IMOH.  I bought this machine 
>> specifically for
>> server usage, so it won't be running X anytime soon.  Anyways... 
>> whatever the case,
>> I'll use Solaris if Sparc64 SMP NetBSD remains unavailable.  I won't 
>> be getting
>> rid of the machine anytime soon, so even if I have to wait until 2007 
>> for NetBSD
>> 4 or whatever, then it doesn't really matter.
>
> I agree wholeheartedly here.  I own or am responsible for about 
> fifteen Ultrasparc Sun machines running NetBSD, most of them are 
> actually active machines on the 'net.  I'm running 1.6.x or 2.0RC1 on 
> the majority of them, and the U2s are (disappointingly) single-proc 
> for no reason other than the lack of SMP support in Sparc64.  if this 
> support were there, there are a couple U1/170's which I would replace 
> IMEEEDIATELY with U2-2/300s.
>
> I don't run X on a single one of these machines.  in fact, not a 
> single one of them even has a framebuffer installed.

   Well...I'm sure it must be irritating for the main kernel developers 
to hear whining about "when will this be done" and "when will that be 
fixed", but there's often a fine line between pestering and 
enthusiastic desire.

   Earlier someone stated that the main kernel people have all the SMP 
hardware they need to continue development...So I have to ask, what can 
the user community do to move SMP development forward?  There is 
clearly a great deal of desire, and the hardware is there.  The biggest 
problem from what I've seen over the years is lack of time.  So what 
would motivate the Powers That Be to take some time away from other 
projects, sleep, etc., to hack on this?

           -Dave

--
Dave McGuire              "I've watched Harley people throw up
Cape Coral, FL                      on the ceiling."    -Krissi