Re: Problem loading SCSI Host (kernel 2.6.9)

Patrick Percot wrote:
>>Maybe there should be an option for the ieee1394 driver that disables IRM
>>functionality.
>
> Something like this patch ? I have added a parameter named
> inhibit_bus_reset for the module ieee1394.

Yes, like that. But I'd call it differently because it does not inhibit
bus resets, it only disables certain resets. If only that else{} in the
IRM duties function is disabled, I'd suggest something like this:
     ensure_cmc_root, "As IRM, ensure that root is cycle master capable
     (default = 1, disable = 0)"
and the 'else' becomes 'else if (ensure_cmc_root)'. Of course nobody
expects to fix sbp2 problems with such an option, but that can be
documented elsewhere.

However when going to add another module parameter, it should perhaps
have bigger impact and disable all of the IRM functionality. That would
of course require changes at several places in the driver and I don't
know how difficult that is. The IRM implementation is not completely
IEEE 1394 compliant yet, therefore a thorough revision of it would be
beneficial anyway. (Not in your case, since SBP-2 does not use the IRM.)

> + * Some firmware don't behave correctly when the bus is reset. If you
> + * find messages like the following ones in kern.log : try setting
> + * this parameter to 1 when loading ieee1394 module.
> + *
> + * ieee1394: The root node is not cycle master capable; selecting a
> new root node and resetting...
> + * ieee1394: Node changed: 0-01:1023 -> 0-00:1023

This comment does not describe the problem very well. The log messages
and the situation in which they appear are for example perfectly normal.
Suggestion:
"Some (buggy?) SBP-2 devices cannot be scanned after the bus was reset
with the log message 'The root node is not cycle master capable;
selecting a new root'. Switch this parameter to disable such resets."

BTW, the problem and the patch apply to kernel 2.6 as well as to kernel
2.4. A good archive for patches (even if only useful to a few people) is
the tracker at the sourceforge homepage of the linux1394 project.
Patches of broader relevance should of course be discussed at
linux1394-devel.

Again, thumbs up for the good diagnostics.

-- 
Stefan Richter
-=====-=-=-- ==-- ---=-
http://arcgraph.de/sr/
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now. 
http://productguide.itmanagersjournal.com/
_______________________________________________
mailing list Linux1394-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux1394-user