>> I have hearsay evidence that MIPS does not have the hardware to do
>> this.  If that is true, it does not mean that the pmax will never
>> run PostgreSQL, just that it may never run it well, or quickly, if
>> someone goes to the trouble of fixing it.
>
> I'm tempted to do something like the atomic_op syscall (called through
> the sysarch mechanism), at least for MIPS (I don't know what other CPUs
> are missing TAS primitives).  The reason I'm leaning this way is that we
> can use this to also implement the atomic_op Ultrix-compat syscall.

I would also lean toward having an additional syscall.  I point out
here, however, it might be complicated somehow to implement the ULTRIX
compatible system call which can work just like as it.  According to
research, ULTRIX/RISC kernel looks to have a set of code to emulate
certain VAX insns inside.  It's no surprise.  The atomic_op is a
backdoor gate plumbing user hose with their own MP locking primitive. 

Now what necessary here is the fine initiative endorsed by good will
and sophisticated technical feasibility combined with the-state-of-art
engineering sense of balance to make realized, as Simon said, NetBSD
has a common, clean, useful and light-weight for any implemention of
processors, architectures and possible user programs.

Tohru Nishimura