Arne Redlich:
> I believe the cause of the problem lies within open-iscsi:
i think so too.
but i could not find the root cause in open-iscsi, and still searching.
> I've tried Junjiro's script with qla4xxx, unh-iscsi, linux-iscsi and
> open-iscsi. The data corruption only occurs with the latter in file f19.
yes it is. my environment too.
and the DataSegmentLength value in pdu is correct.
> open-iscsi code, I believe the problem is open-iscsi not respecting the
> BufferOffset of the DataIn-PDUs. Unless I'm completely mistaken, the
> offset is stored in ctask->data_offset when reading the DataIn-header,
> but is not taken into account when copying the data from the skb to the
> buffer.
so you are pointing out skb_copy_bits() call in iscsi_ctask_copy(),
right?
Junjiro Okajima
-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
|
