[GRLUG] PCI VGA Help

[Michael Mol]

On Wed, Apr 20, 2011 at 10:19 AM, Adam Tauno Williams
<awilliam at whitemice.org> wrote:
> On Wed, 2011-04-20 at 10:06 -0400, Michael Mol wrote:
>> On Wed, Apr 20, 2011 at 9:14 AM, Adam Tauno Williams
>> <awilliam at whitemice.org> wrote:
>> > On Wed, 2011-04-20 at 07:54 -0500, L. V. Lammert wrote:
>> >> On Wed, 20 Apr 2011, Adam Tauno Williams wrote:
>> >> > On Tue, 2011-04-19 at 11:44 -0500, L. V. Lammert wrote:
>> >> > > Need someone with knowledge of hw - we have a passive backplane system
>> >> > > with ISA & PCI where we need a 2nd monitor (Ubuntu 10.04)
>> >> > Can you be more precise about "have a passive backplane system"?
>> >> In what way? 10 ISA + 6 PCI? Two slots are ISA/PCI for the CPU and a
>> >> spare.
>> > Is this some type of vertical appliance or a backplane extension to a
>> > "PC".  How does the backplane attach to the mainboard [simple PCI ribbon
>> > cable to the planar?  Or does the host motherboard have provisions for
>> > the attachment of the backplane?].
>> If it's the kind of system I had as a desktop box for several years,
>> here's how it would be laid out:
>> * You have a CPU card which has northbridge, southbridge, CPU, I/O,
>> etc on it. Will have headers for serial, USB, etc. May have onboard
>> network and video, but all of this on the CPU card.
>
> Yep, I've seen those.  I've also seen tethered backplanes.
>
>> * You have a "motherboard" in a 4U case which has three PCI-PCI
>> bridges on it and possibly some routing logic for a keyboard
>> connector.
>
> Right, which makes the "passive backplane" a bit of a lie; that is just
> an inherited/colloquial term.  There is nothing "passive" about a PCI
> bus - PCI does device capability and resource negotiation and
> 'understands' bus-to-bus data transfer.  The PCI bus controllers are a
> significant factor in getting something like multiple video cards spread
> over multiple busses to actually work.  If you can't access them, or at
> least interogate them to some degree, you might very well be 'screwed'.

Here's the output of lspci on my old system (from ages ago, found via
a bug report email)

shortcircuit at caterpillar.wonderland~
12:28:17 $ /sbin/lspci
00:00.0 Host bridge: Intel Corporation 82845G/GL[Brookdale-G]/GE/PE DRAM
Controller/Host-Hub Interface (rev 01)
00:02.0 VGA compatible controller: Intel Corporation 82845G/GL[Brookdale-G]/GE
Chipset Integrated Graphics Device (rev 01)
00:1d.0 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M)
USB UHCI Controller #1 (rev 01)
00:1d.1 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M)
USB UHCI Controller #2 (rev 01)
00:1d.2 USB Controller: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M)
USB UHCI Controller #3 (rev 01)
00:1d.7 USB Controller: Intel Corporation 82801DB/DBM (ICH4/ICH4-M) USB2 EHCI
Controller (rev 01)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev 81)
00:1f.0 ISA bridge: Intel Corporation 82801DB/DBL (ICH4/ICH4-L) LPC Interface
Bridge (rev 01)
00:1f.1 IDE interface: Intel Corporation 82801DB (ICH4) IDE Controller (rev 01)
00:1f.3 SMBus: Intel Corporation 82801DB/DBL/DBM (ICH4/ICH4-L/ICH4-M) SMBus
Controller (rev 01)
01:08.0 Ethernet controller: Intel Corporation 82801DB PRO/100 VE (CNR) Ethernet
Controller (rev 81)
01:0c.0 PCI bridge: Pericom Semiconductor PCI to PCI Bridge (rev 01)
01:0d.0 Ethernet controller: Accton Technology Corporation SMC2-1211TX (rev 10)
02:05.0 Ethernet controller: Atheros Communications, Inc. AR5005G 802.11abg NIC
(rev 01)
02:06.0 FireWire (IEEE 1394): VIA Technologies, Inc. IEEE 1394 Host Controller
(rev 46)
02:07.0 USB Controller: NEC Corporation USB (rev 41)
02:07.1 USB Controller: NEC Corporation USB (rev 41)
02:07.2 USB Controller: NEC Corporation USB 2.0 (rev 02)
02:08.0 Multimedia controller: Philips Semiconductors SAA7133/SAA7135 Video
Broadcast Decoder (rev d1)
02:0a.0 Multimedia audio controller: Ensoniq ES1371 [AudioPCI-97] (rev 06)
shortcircuit at caterpillar.wonderland~
12:28:22 $

Everything in the 00: range was on the CPU board. Everything in the
01: range was adjacent to the CPU board on the backplane. 02: was
after the first on-backplane PCI bridge. (I don't know where the
second bridge ran off to. I may be misremembering its layout)

So, your thought then is that if his video cards aren't immediately
adjacent to his CPU board (prior to any of the on-backplane PCI-PCI
bridges), they risk additional issues because of complications from
having a bridge. A recommended solution would then be to place the
cards as close to the CPU board as possible.

-- 
:wq

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.