In message <[EMAIL PROTECTED]> Alex writes:
: Many thanks for your explanation. The question is why PNPBIOS tries to
: attach and assign resources to devices that have already been attached
: (things like AT DMA controller, keyboard controller, math coprocessor,
: printer port, etc.) Is it not smart enough to figure it out? And why
: do I need this option to make my non-PnP ISA sound card work? Am I
: right in assuming that some intrusive device probes (that break my sound
: card) are avoided when PNPBIOS is defined?
Because we don't have drivers for each fo these devices, in some
cases. The order in which these things probe is also incorrect. We
should create drivers for the few that don't exist and also teach the
current drivers about these pnp ids. I think that this part of the
PnP probe should happen BEFORE the rest of the isa device are probed.
They are fixed addresses on the mobo that can't be changed at all.
The pnp code likely needs to become smarter so that when presented
with a pnp device that is has the same resources as the hardwired
devices, the hardwire should "win" if you will (eg, fdc0 should attach
pnp or no and there shouldn't be a fdc1 that attaches to pnp and then
the nonpnp one fails to attach).
Don't know about the non-PnP isa question. Likely it is more pnp than
you think :-)
: P.S. Do you have the correct URL for PnP codes?
Yes:
http://members.hyperlink.net.au/~chart/download/pnpid.txt
Warner
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
