Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 5 Jun 2000 17:11:07 -0500
From:      "Jason" <jsmethers@pdq.net>
To:        "Fred Clift" <fred@veriohosting.com>
Cc:        <freebsd-hackers@FreeBSD.ORG>, <freebsd-stable@FreeBSD.ORG>
Subject:   Re: changed pci bus probe order from 3.2 to 4.0 -- ideas?
Message-ID:  <011001bfcf3a$f4b9bfd0$0dcb1f40@mom>
References:  <Pine.BSF.4.21.0006051515340.14309-100000@vespa.orem.iserver.com>

next in thread | previous in thread | raw e-mail | index | archive | help
From: "Fred Clift" <fred@veriohosting.com>
> Hm so does that mean that non-bridge devices on those busses should also
> be done depth first?  I understand wanting to find all the busses first,
> but devices too?  All the bioses I've seen seem to find all the busses,
> then assign resources to devices starting with those on bus 0 then bus 1
> etc.  This is also how FreeBSD <=3.4 work.  Am I just missing something?
>
> Fred

No, I am. For some reason I didn't make a distinction between regular
devices and bridge devices in my mind when I though about this. =p

What needs to be changed is the assignment of resources based on a
combination of chassis, slot numbers, and fuction number of a device in the
slot, not based on the bus number. This would have the desired effect.

There are four ways of obtaining information that may be used to do this.
Useing the PCI interrupt routing table in BIOS memory, querying the PCI BIOS
for the same table, use the PCI 2.2 chassis/slot numbering capabilities
register, and using the function numbers of the device. Everything could be
done while enumerating the buses. Doing this would also potentially allow
the wiring down of devices based on the chassis, slot, and fuction of the
slot.

This would allow logical ordering of devices. Now it just needs to be
implemented.

- Jason




To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?011001bfcf3a$f4b9bfd0$0dcb1f40>