Re: "aborting sbp2 command" expected in kernel.org FireWire maybe

Stefan R:

> (check the init message from sbp2),

Thanks again for here teaching me to observe and change that
(sbp2_)?serialize_io parameter.

> > whether this bug we have found is
> > in the device or in Linux or both?
>
> If the device works under kernel 2.4
> without serialization
> (check the init message from sbp2),
> then it must be a bug in kernel 2.6.

Ah. Then now instead I see, with that FireWire DVD/CD I bought retail
and the "sbp2: $Rev: 1074 $ Ben Collins" of the 2.4.27 of Knoppix 3.6,
that:

Copies fail after 2.4 modprobe sbp2 sbp2_serialize_io=0, same as after
2.6 modprobe serialize_io=0, for stimulus as small as a couple of 64
KiB dd reads, via such dmesg as:

ieee1394: sbp2: aborting sbp2 command
0x28 00 00 00 00 24 00 00 3f 00

Are we therefore concluding I should claim the device has a bug?

I should now go borrow a bus analyser and collect a 1394 bus trace,
because that (sbp2_)?serialize_io=0 trace will show a compliant host
provoking a device to fail to comply in some way that doesn't bother
Mac/ Windows?

Pat LaVarre

-------------------------------------------------------
This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
Project Admins to receive an Apple iPod Mini FREE for your judgement on
who ports your project to Linux PPC the best. Sponsored by IBM.
Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
_______________________________________________
mailing list Linux1394-user@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux1394-user
Received on Sat Sep 18 15:33:01 2004