yet more info re:PCI-DIO-96

All -

Sorry about the many messages, it's just that now that I noticed the
"lack of valid address space" problem, I've been running around trying
to rapidly understand the PCI bus. Plus this is really bizarre...

OK: I've tried yanking the PCI-6025E, and seeing if the PCI-DIO-96
starts working. No such luck. It DOES, however, start requesting memory,
but only if "Plug-and-Play OS" is set to "YES" in the BIOS. The reason
behind that will quickly become clear (I think...)

Here's the output of lspci -vv -s 0:a (pci device 0:a is the PCI-DIO-96)

00:0a.0 Class ff00: National Instruments PCI-DIO-96
        Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
        Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
        Interrupt: pin A routed to IRQ 7
        Region 0: [virtual] Memory at d9000000 (32-bit,
non-prefetchable) [disabled]
        Region 1: [virtual] Memory at d9001000 (32-bit,
non-prefetchable) [disabled]

It is now listing memory in two locations, but listing them as disabled.
It's also listing itself as non-I/O and non-Memory capable (I/O-, Mem-).
Here the MITE routine does map those locations to other locations, but
it still works the same as before. Might it have something to do with
the fact that it's claiming they're 'disabled'?

With both cards plugged in, the PCI-DIO-96 no longer requests any space
whatsoever: lspci -vv -s 0:a reports

00:0a.0 Class ff00: National Instruments PCI-DIO-96
        Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
        Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
        Interrupt: pin A routed to IRQ 0

Hrm. Very confusing. In this case the MITE driver does what it was doing
before, and maps non-existent memory space to somewhere else.

comedi1: nidio:PCI: Enabling bus mastering for device 00:50
PCI: Increasing latency timer of device 00:50 to 64
MITE:0x00000000 mapped to d00e3000 DAQ:0x00000000 mapped to d00e5000
 pci-dio-96

This doesn't look good. :) I have this nagging feeling that there was a
bug in this board and Nat. Inst. worked around it for the Windows
drivers. Has anyone really gotten this to work with Comedi? I saw on
Nat. Inst.'s web site that other people have had the same problem:

http://exchange.ni.com/servlet/ProcessRequest?RHIVEID=101&RNAME=ViewQuestion&HOID=506500000008000000630D0000&ECategory=Measurement+Hardware.Multifunction+I%2FO

Does anyone have any idea what's going on here? Or does anyone have a
working PCI-DIO-96 and can send out their lspci output?

Patrick

Received on 2002-06-04Z20:30:17