On Aug 21, 2007, at 11:43 PM, Tejun Heo wrote:
Rusty Conover wrote:
Putting just one hard disk into the PMP slots, works great all by
itself. I created an ext3 fs on it, used dd to dump lots of data
to it,
no problems in all of my testing.
Hmmmm..
One hard disk in a PMP slot and another hard disk in a native slot
on a
different SATA port work great together.
Okay.
Two hard disks in the same PMP slot range (slots 1-2), timeout
right away.
Could it be a problem with timing the PMP requests?
Very unlikely.
I've tested this on a second separate machine and I'm seeing the same
timeout errors.
Can you try different combinations (e.g. use different slots of the
same
PMP, use the slots from the other PMP)? I still think it's some sort
of hardware issue.
Hi Tejun,
I've tested this on both PMPs, when there is more then 1 disk on the
PMP ports the timeouts occur.
For instance I tested:
Bays all on PMP1
1,2
1,3
2,4
3,4
Bays on PMP2
6,7
7,8
6,9
6,10
All of those combinations timeout very quickly, just when creating
the raid device. On the other attempts I was able to go past
creating the raid device and attempt to create a ext3 fs. These just
timeout doing the RAID-10 initial sync after about 5 seconds.
When I test just two disks in different PMP ports on two different
SATA ports everything works.
When I test three disks, 2 PMP ports of one SATA port and 1 in
another PMP port of a different SATA port I get the same timeout
error on the PMP with more then one disk.
It seems to be any arrangement with more then one disk in a set of
PMP ports leads to timeouts.
If it would help to have access to the machine to possibly debug
what's going on with the errors, I'd be more then happy to arrange that.
Thanks for your help,
Rusty
-
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
