Re: ieee1394: host adapter disappears on 1394 bus reset

[Stefan Richter <stefanr@s5r6.in-berlin.de>]

Robert Crocombe wrote:
...
> Nov 27 13:06:37 spanky kernel: ohci1394: fw-host1: IntEvent: 00020010

busReset + RQPkt (packet sent)

...
> Nov 27 13:06:37 spanky kernel: ohci1394: fw-host1: IntEvent: 00010000

selfIDcomplete

...
> Nov 27 13:06:40 spanky kernel: ohci1394: fw-host1: IntEventClear
> 00000000 IntEventSet   04508000 IntMaskSet    838301f3

IntEventSet = phyRegRcvd + cycleLost + cycleSynch + selfIDcomplete2

> Nov 27 13:06:40 spanky kernel: ohci1394: fw-host1: IntEvent: 00020010
...
> Nov 27 13:06:41 spanky kernel: ohci1394: fw-host1: IntEvent: 00010000
...
> Nov 27 13:06:44 spanky kernel: ohci1394: fw-host1: IntEventClear
> 00000000 IntEventSet   6ffdc33f IntMaskSet    00000000

IntEventSet = vendorSpecific + softInterrupt + ack_tardy + phyRegRcvd +
cycleTooLong + unrecoverableError + cycleInconsistent + cycleLost +
cycle64Seconds + cycleSynch + phy + regAccessFail + selfIDComplete +
selfIDComplete2 + [reserved...?] + lockRespErr + postedWriteErr + RSPkt
+ RQPkt + ARRS + ARRQ + respTxComplete + reqTxComplete

(The bit 00004000 is not standardized and shouldn't be there. Whatever.)

That's a lot of stuff that happened right before this last print macro.
Maybe ages were spent in hpsb_selfid_complete. I will try to work on
getting non-critical parts out of hpsb_selfid_complete issue during the
week, but I don't know how fast I can do this and if this will help in
the first place.

...
> I didn't mention that I have:
> 
> options ieee1394 disable_nodemgr=1
> options ohci1394 phys_dma=0
> 
> in my /etc/modprobe.conf.  The Linux adapters are functioning as
> simulated peripherals to a piece of control hardware that always has a
> dest address of 0x0000 0000 0000 on all packets so I needed to get rid
> of posted writes and any bickering over bus master.

Posted writes are still enabled. phys_dma=0 disables only the physical
response unit. You have to change the source if you want to disable
posted writes. See the top of ohci_initialize. Should this be a module
load parameter too?
-- 
Stefan Richter
-=====-=-==- =-== ==-==
http://arcgraph.de/sr/