Tobias Nygren wrote:
> I can reproduce this, with a LOCKDEBUG kernel. I don't have any swap
> enabled so instead of death by vm thrashing I get killed processes.
> I guess this could be a problem with the sparc64 atomic ops, since
> the uvmexp accounting doesn't add up.

That would make sense; if you look back at my top output in the original 
email (or the earlier followups), you'll notice that the memory usage 
doesn't add up -- there's almost nothing used for the file cache and yet 
no memory free.

--rafal