[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: PCI slot lost on power up?
- To: pci-sig@znyx.com
- Subject: RE: PCI slot lost on power up?
- From: Ivor Bowden <ivor@peritek.com>
- Date: Thu, 07 Jun 2001 12:02:48 -0700
- Old-X-Envelope-From: ivor@peritek.com
- Old-X-Envelope-To: pci-sig@znyx.com
- Resent-Date: Thu, 7 Jun 2001 12:15:37 -0700
- Resent-From: pci-sig@znyx.com
- Resent-Message-ID: <cE_nFB.A.PgD.SC9H7@electra>
- Resent-Sender: pci-sig-request@znyx.com
I would like to thank everyone who replied! I got quite a few ideas.
It appears that the BIOS checks for a card, if there is no response
it disables the clock to that slot. Interesting that once the slot is
disabled it stays disabled until the power is cycled, even if you reboot
it will not turn any clocks back on.
Many people suggested a BIOS setting to disable this feature. I can not
find anything that affects it. The BIOS is
Award 4.60PGA - version 1.15JE35 - v1.0A PNP
or so it says on the boot up splash screen.
For PCI clock I only have the option for "spread spectrum" or not
and I choose not. All "power saving" features are turned off.
Many folks have suggested doing something with the clock generator
chip. Apparently it is common to use an I2C bus (2 wire serial bus)
(also called SM Bus?) to control the clock chip. It may be that
simply lifting the I2C control pins to the clock generator chip
will prevent it from being turned off. Or, if I can talk to it (I
do have a collection of I2C routines I have written, but I need to
know which bits to bang) I could turn it back on after boot up.
The motherboard is a
FIC VA-503+ using the VIA APOLLP MVP3 chipset
(VT82C598MVP northbridge and the VT82C586B south bridge).
It looks like the I2C bus is controlled by the south bridge chip.
The clock generator chip appears to be a "W127H". I can not identify
the mfg. Does anyone know where I can find a data sheet for this chip?
I would still like to figure out how to make the clock always run.
We want this test station to be able to boot up and run without
any cards, then they are plugged in and out (with the help of
the Catalyst PCIBX64 "hot swap" extender board. The OS is DOS v6.21
with test programs written in the djgpp DOS port of gnu C
( http://www.delorie.com/djgpp/ ).
If anyone knows where I can get a data sheet for the W127H clock chip,
or has any more ideas how to turn off this "feature" please let me know.
Once again, I would like to thank everyone for your thoughtful
responses.
Ivor
>>> -----Original Message-----
>>> From: Ivor Bowden [mailto:ivor@peritek.com]
>>> Sent: Wednesday, 06 June, 2001 10:05
>>> To: pci-sig@znyx.com
>>> Subject: PCI slot lost on power up?
>>>
>>>
>>> Hi PCI experts,
>>>
>>> I wonder if any one else has seen this problem and knows of a
>>> workaound:
>>>
>>> We are making a test station for PCI boards.
>>> We are using a Catalyst PCIBX64 "intelligent" extender board.
>>> It is a passive board with bus and power switching, that can
>>> be accessed via a parallel port.
>>>
>>> The problem is, if there is no board in the slot, or if the extender
>>> is turned off, AT POWER UP, then the slot is essentially dead.
>>>
>>> It is not a BIOS problem, I load the BARS after boot up. It does not
>>> make any difference which board is in the slot, as long as it is a
>>> valid PCI board.
<clip incorrect info)
>>>
>>> Any ideas?
>>>
>>> Thanks!
>>> -Ivor
>>>
>>> Ivor Bowden
>>> Peritek Corp
>>> 5550 Redwood Road
>>> Oakland, CA 94619 USA
>>> phone: (510) 531-6500
>>> FAX: (510) 530-8563
>>> url: www.peritek.com
>>> email: ivor@peritek.com