There's a time in a man's life when, confonted with several piles of
evidence, that he has to back down :-)

Seriously, I know a lot of people using RiscPCs without too much trouble
(running gimp, and other stuff). There _were_ bad hardware bugs in those
machines. Eg the NetBSD idle loop, when spinning, could cause
interference which cause the CPU bus to stall. The only solution was to
swap SIMMs around.

I can see no reason however why someone hasn't just applied the PRs if
they are legitimate fixes to problems. The only machine I have here are
EBSA-385 derived systems which, after the linker breakages were fixed,
have been stable to a tee (outlasting NetBSD/i386 in most cases). While
the ld.so was still screwed up, however, processes did bomb randomly,
especially under load. I don't know about Sharks and RiscPCs however.

The installations are planned to be fixed for NetBSD 1.4, hpoefully
using sysinst. Before then, it wasn't worth contributing to the other
1.3 releases, and there wasn't the time to spend to fix sysinst itself.

	Neil

-- 
Neil A. Carson