I had the same thing happen to a brand new Fujitsu 10GB. I was trying to use the
utility hdparm to tweak my drives for best DVD performance. I think it was 'hdparm
-X66 -m16c1d1 /dev/hda' for ide-drive and 'hdparm -X33 -c1d1 /dev/hdd' for my
dvd-cdrom etc. Anyways, all seemed fine until I used those commands after backing down
from an unstable Debian dist. to a stable Potato(2.2.19pre21) then running those
commands would produce that same error you have spit back at me on the console... I
myself hope that this is a glitch in Linux ide drivers and not a bad drive... after I
upgraded to a Woody class box 2.4 kernel (progeny) I had no more problems and those
hdparm commands didn't return the errors any longer... I hope my drive is proper. I
must confess that when I use lilo commands at boot like Linux idebus=66 ide0=ata66 I
do not see promising remarks in dmesg. Crap, I want a new drive! At least hdparm
tells me that I have all I want enabled on my disk etc. Peace of mind!
... the answer might lie in Kernel version and driver code! I am now running a 2.4.5
Debian Sid box and getting about 22 MB/sec buffered-disk read and 162 MB/sec
buffered-cache read according to hdparm... Good luck!
On Wed, Jun 20, 2001 at 12:09:46AM -0700, Neil Parker wrote:
> Calling all you hard drive gurus out there...
>
> The hard drive: Quantum Fireball EL10.2A, 10GB, only drive on primary cable
> IDE chipset: ALi M5229 IDE interface, ALi M1533 PCI to ISA Bridge (on
> motherboard)
> Kernel: 2.2.19, using the alim15x3 driver
>
> For about a year, I've been running this setup with UDMA (mode 2) enabled
> without problems. But yesterday I started getting this message on the
> console at boot time:
>
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> hda: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> hda: dma_intr: error=0x84 { DriveStatusError BadCRC }
> ide0: reset: success
>
> ...after which the drive reverts to non-DMA operation. Trying to turn
> DMA back on (hdparm -d1 /dev/hda) just causes the error to repeat. I've
> tried several different hdparm -X modes to no avail.
>
> The errors first appear when the system is going to multi-user mode, right
> after the root partition get remouted read-write. I'm not sure, but I
> think it's happening the first time something tries to write to the disk.
>
>
> So...is there anything I can do to get my DMA working again? Or is my hard
> drive starting to die? (It's less than three years old, and seems to be
> working fine in non-DMA mode.)
>
> - Neil Parker, [EMAIL PROTECTED]
>
--
- Debian GNU/Linux ->
http://www.debian.org
