Brad Campbell wrote:
> Jeff Garzik wrote:
> 
>> Interesting.
>>
>> Well since it's not global behavior, but isolated to one port or card, 
>> I still worry about non-libata things:
>> 1) is a SATA cable bad, or not plugged in well?  I'm finding that it's 
>> easier to screw up SATA cabling than PATA.  It's more-convenient 
>> design is also less rugged.
> 
> 
> Nup, all seated fine and working perfectly well (and I mean raid-5 
> across all 10 disks and spraying
> data at maximum PCI speed for 8 hours at a time well) with 2.6.5, and 
> they are Supermicro cables
> SATA cables with a pair of 5 bay Supermicro Hotswap bays so the quality 
> is pretty good.
> 
>> 2) is a PCI slot bad, or not busmastering like it should?  have you 
>> tried moving the card to another PCI slot?

Ok, I have now reproduced this on another system. (New SATA cable and Drive)
I took all three cards and plugged them into another system, as I don't have another 3 cards to spare!
It is 100% reproducible if any drive is plugged into the last card to get initialised. I am testing 
it now with three cards and 1 SATA hard disk.

If I plug the drive into any of the ports on the first 2 cards, it boots and all is happy. If a 
drive is plugged into the last card it locks on bootup as before.

As before, a Vanilla 2.6.5 works perfectly.

This is the same set of cards, however I have shuffled them around and into different slots to try 
and isolate the fault to a card. Unfortunately they all work perfectly under Vanilla 2.6.5 and it's 
always the last initialised card that causes the problem, no matter what slot/irq/physical card it is.

I'm happy to do some more testing as I now have both systems in bits.

Both machines are VIA KT600 chips, I have a Gigabyte GA-7VT600 and an ASUS A7V600. I have removed 
most of the other cards and have played with these Promise controllers in both of them. The problem 
follows the kernel. 2.6.5 works perfectly on both machines. 2.6.5 with the patch you sent me to 
update libata locks the same way on both machines.

I have tried with acpi=off, noacpi, apic=off, noapic and all combinations of. The problem still 
persists.

I have attached 3 serial console boot outputs, one where the drive was plugged into each card.

Anything else I can do?

Regards,
Brad