Re: Issues with TI XIO2213 chipset?

[Stefan R. <st...@s5...>]

On Dec 29 Charlie McMackin wrote:
> > Does you PCIe card have an internal power connector?  If yes, make sure
> > that it is attached to the PSU.  If not, power the ADVC110 by a
> > separate PSU.
> >
> 
> My card does not have an internal power connector but it does appear
> to be delivering power to the ADVC110 since the lights are coming on
> fine... Is it still possible that the ADVC110 needs a power adapter?

If you operated this the same way on the Audigy card (which AFAIK does
not have a power connector either), then you were lucky.  The 12 V
power rail of PCI is rather weak and not adequate for the minimum
requirements on a spe compliant 1394 bus power provider.

Ditto PCI Express:  The +12V power rail of a PCIe x1 slot is merely
specified for 0.5 A which may not be enough to power both the card and
the ADVC110.  The +12V rail of x4/x8 slots is specified for 2.1 A and
the one of x16 slots for 4.4 A but I don't know if an x1 card put into
an x4 or x8 or x16 slot can actually tap into the better supply of
those slots, if you can spare one.

Anyway, you should rather power the ADVC110 directly.

[...]
> I get a constant stream of messages which all
> look the same save for the trailing number. Here are the last few
> entries plus the event where I unplug the cable from the box:
> 
> Dec 29 15:02:07 onyx kernel: [  317.646314] firewire_ohci: IRQ 00000010 AR_req
> Dec 29 15:02:07 onyx kernel: [  317.646318] firewire_ohci: AR evt_bus_reset, generation 135
> Dec 29 15:02:07 onyx kernel: [  317.646646] firewire_ohci: IRQ 00000010 AR_req
> Dec 29 15:02:07 onyx kernel: [  317.646651] firewire_ohci: AR evt_bus_reset, generation 136
> Dec 29 15:02:07 onyx kernel: [  317.646980] firewire_ohci: IRQ 00000010 AR_req
> Dec 29 15:02:07 onyx kernel: [  317.646985] firewire_ohci: AR evt_bus_reset, generation 137
> Dec 29 15:02:07 onyx kernel: [  317.647314] firewire_ohci: IRQ 00000010 AR_req
> Dec 29 15:02:07 onyx kernel: [  317.647319] firewire_ohci: AR evt_bus_reset, generation 138
> Dec 29 15:02:07 onyx kernel: [  317.647616] firewire_ohci: IRQ 00000010 AR_req
> Dec 29 15:02:07 onyx kernel: [  317.647620] firewire_ohci: AR evt_bus_reset, generation 139
> Dec 29 15:02:07 onyx kernel: [  317.647795] firewire_ohci: IRQ 00010000 selfID
> Dec 29 15:02:07 onyx kernel: [  317.647809] firewire_ohci: 1 selfIDs, generation 140, local node ID ffc0
> Dec 29 15:02:07 onyx kernel: [  317.647812] firewire_ohci: selfID 0: 807fcc56, phy 0 [---] beta gc=63 -3W Lci
> Dec 29 15:02:07 onyx kernel: [  317.647815] firewire_core: skipped bus generations, destroying all nodes
> Dec 29 15:02:08 onyx kernel: [  318.140068] firewire_core: rediscovered device fw0

Given that this endless series of bus reset IRQs happens with a known
good physical layer chip, it is a very strong indication of a hardware
fault.  E.g. said power supply problem.
-- 
Stefan Richter
-=====-==-=- ==-- ===-=
http://arcgraph.de/sr/