Frank van der Linden <frank@wins.uva.nl> writes:
> On Sun, Feb 13, 2000 at 12:28:39AM -0500, Laine Stump wrote:
> > NetBSD supports other UDMA/66 controllers, but not the one on the BP6.
> > Neither does NetBSD-current as of mid-December it didn't, and I haven't
> > noticed any change since. (Someone correct me if my info is out of date.)
> > The UDMA/33 controllers on the BP6 still work fine, though.
> 
> Just to avoid confusion that I have seen occur before: "not supporting"
> in this case means that they won't be driven in full UDMA/66 mode,
> but that they will work just fine, albeit using "lower" modes.

Okay, to avoid confusion, I'll be more specific in my statement: This
isn't your everyday IDE controller. "not supporting" in this case means
"disks connected to this controller will not be seen by NetBSD (because
the controller itself is not recognized) and will not be usable in *any*
mode". Period.

Unless something has changed in the IDE support since mid December, the
HPT controller on the Abit BP6 motherboard is not supported *at all* on
NetBSD, not in any 33 modes, including plain old PIO. (I just tried it
again to verify that I hadn't missed something earlier).

The BP6 has 2 dual channel IDE controllers, each channel supporting one
master and one slave drive (theoretically, you could connect 8 IDE
devices to this board using just the controllers on the
motherboard!). The first controller is the one that's part of the BX
chipset, and it is of course supported just fine in all modes it has (it
is 33-only). The second is from HighPoint Technologies, it is labelled
as "HPT366" in the BIOS screen, and probed during boot as:

Triones Technologies product 0x0004 (miscellaneous mass storage, revision 0x01) at pci0 dev 19 function 0 not configured
Triones Technologies product 0x0004 (miscellaneous mass storage, revision 0x01) at pci0 dev 19 function 1 not configured

NetBSD does not recognize this controller as anything at all, and disks
connected to it are not seen either. This isn't a huge problem as long
as you're willing to have only 33 support, and "only" 4 devices
connected to the machine instead of 8.

> > Also don't expect to get any use out of the 2nd CPU on your BP6. NetBSD
> > still doesn't support more than one CPU (and I haven't seen any messages on
> > tech-smp in quite awhile).
> 
> It's being actively worked on, the preliminary code may be available
> soonish, but this is up to the person who's writing it (i.e. no
> sense in asking yet..)

Not to be impolite, but that's the same thing everyone was told over 2
years ago. About once every six months, a short thread starts up on
tech-smp where some people say something like "Okay, it's time to
actually do something about this. What if we simplify our goals?" or
something like that. After that there's a flurry of messages from a few
people with ideas, and other people volunteering to help if they're
given direction from someone more in the know (I've been in the latter
category), followed by a deafening silence. Any project that takes this
long to complete could *surely* be helped by more people, yet we never
(at least I never) see any requests for assistance. I'd be willing to do
any sort of work on this, from writing documentation and testing stuff
all the way to writing bits of code or auditing existing code to check
for MP-safeness, but my inquiries have fallen on deaf ears (and another
6 months has passed). Have I been listening in the wrong place, or is
this development going on "in secret" for some reason? If the latter,
why is that, when there is a large community of people just dying to
help?