"Matthew Schnierle" <pyld@grove.iup.edu> writes:


>> There is a fix for this, it means redoing the spl machinery in the
>> kernel.  Jason Thorpe planned out most of the work and I'm still
>> waiting for him to finish it.  Meantime, if you have a 3MIN,
>> you *have* to pull unsupported video cards.
                     ^^^^^^^^^^^

>I'm curious: what about a supported card?  

Uh, that means it Just Works.  Both as console on framebuffer via the
rcons device, and with X11 using the X11R5 server (currently), or even
just ignored.  (My 5000/150 currently lacks a keyboard, and it boots
just fine with a serial console and a PMAG-B).

Come to think of it, I am interested in reports of multiple
framebuffers of the same type in a single machine.  AFAIK, nobody has
tried this since Ted Lemon was portmaster.


>I ask because I *thought* I booted one of our 3MINs without yanking the card
>(as mentioned below)...

I would be very skeptical.  On some cards you have to actively enable
vertical-blank interrupts (or at least, software can disable them),
and you'd already booted an OS which disabled the vertical-blank
interrupt.

That might work with an SFB. There's no way it can work with a mfb or
a CFB, and I don't recommend it even for the SFB: you will lose next
time the machine gets powercycled.