Mike Christie wrote:
> On 08/04/2009 01:25 AM, Hannes Reinecke wrote:
>>> Also what is up with the simple patch? Did that work? I am not sure
>>> there is any point in doing the above patch, if we do not also fix the
>>> queueing. When the above patch hits EACCESS then because that task is at
>>> the head we will just retry the same one and no new IO to any other lun
>>> or task is sent. This is the same as the simple patch.
>> Actually I only tested with this patch, and it looked to me as if the
>> spurious PDUs I've seen are now gone.
> Did you have a patch for this that I can add for 2.6.32?
> I think I can fix the problem where tasks behind one that gets a EACCESS 
> get stuck. I think if we move the list deltion code around and add some 
> code to track the head, we can do this. The problem is that the ordering 
> of execution could get mangled, but that happens in the eh and other 
> paths right now already.
I don't think this matters as the command PDUs marked with -EACCESS will
be deleted later during fail_scsi_tasks, so we won't be sending PDUs
with cmdsn out-of-order. We might have holes in the CmdSn sequence,
though. Not sure if it matters; if we do have problems there we can
always send a non-immediate NOP instead to plug the CmdSn hole :-)
Or modify the SCSI command PDU to be a TUR; the trick with the
final bit doesn't really work here.

Anyway, I've finished off my patchset and will be sending it shortly.


Dr. Hannes Reinecke                   zSeries & Storage
h...@suse.de                          +49 911 74053 688
SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg
GF: Markus Rex, HRB 16746 (AG Nürnberg)

You received this message because you are subscribed to the Google Groups 
"open-iscsi" group.
To post to this group, send email to open-iscsi@googlegroups.com
To unsubscribe from this group, send email to 
For more options, visit this group at http://groups.google.com/group/open-iscsi

Reply via email to