> Ok, I commited more work for the TS-7200 to NetBSD-current today and was
> able to write drivers/test existing support for the following other
> peripherals:

His achivement is impressive.  It's worth announce in bsdnews.com with the
latest dmesg and a pointer to his note.  Several lines picked my eyebow;

"A single talented developer is typically more effectively utilitized in this (a single,
unified distribution) arrangement, and is part of the reason NetBSD can do with
a few hundred developers what Linux does with tens of thousands. "

"NetBSD recognizes that developer time and talent is finite and does not waste
energy with superfluous redesign."

"Linux gets its portability ... on top of patches to make hardware look like a i386
to the rest of the kernel. "  *see below for my note* 

"In Linux, the mantra is that developer time is 0-cost and available in infinite
quantities, ..."

** It's a technological agony.  Let me take a example.  Some may know that
HIGHMEM support by L* kernel.  It's a plain mistake, I believe, but the issue
cause is very deep in L* foundation and not fixable in near future.  The true
technology issue resides in kernel memory allocator, the way how kernel
manages memory.  The design flaw is serious and has been badly affecting
other type of processors indeed.  No i386 pretending, pls...

Toru Nishimura/ALKYL Technology, a NetBSD company