[SunRescue] any 3B2 users out there?

[rescue at sunhelp.org]

woods at weird.com writes:

>Not to worry -- the WE32000 processors are basically a slightly
>modernised 32-bit VAX architecture on a chip.  There's more MMU support
>than you really need.  It was, after all, a chip designed specifically
>for running Unix and only Unix; and they run SysVr4 just fine too (which
>has an early version of the newer SunOS-5 VM).

>(I do have the processor manual :-)

>The only hard part would be interfacing to the "firmware" loaded into
>the various device adapter cards.  Even source licensees did not receive
>source for the firmware or for its development system (most "modern" 3b2
>cards have on-board 80186 CPUs), and mostly not even device drivers
>either.  That would mean keeping the same firmware and trying to figure
>out what the API for the driver was.

Were there any 3B2 IO boards that *didn't* have an onboard 186? I've
never seen one. (I've only dealt with the 300 and 400.)

One would think that given DMA & an 186 on every IO board, performance 
would be killer, but I have heard that the 186s spent most of their time
working around hardware bugs. Do you know if there's any truth to that?

How do the free unixes deal with downloaded "firmware" to boards on
microchannel machines? Do they just use the original files or have
free versions of any of that code actually been written?

-- david fischer -- dave at cca.org -- www.cca.org -- Cthulhu told me to. --