Albert, Thanks for the driver. I am not sure whether you are correct about the PLL being initialized (I don't know enough about the internals to say for certain), but I have attached the dmesg output from 2.6.10 compiled with your patch and one other change to libata.h--inserted at line 342:
int (*check_atapi_dma) (struct ata_queued_cmd *qc);
This was done entirely so it would compile cleanly. I
have yet to test ATAPI devices on it.
However, the good news is that all eight drives have been
detected and configured at UDMA 5. I will do the stress
tests on the drive overnight and let you know how things
work out.
Thanks again,
Drew
dmesg
Description: dmesg
