Messages in this thread | | | Date | Wed, 2 Sep 1998 23:59:34 -0500 (CDT) | From | "Andre M. Hedrick" <> | Subject | Re: hdd: misaligned DMA buffer |
| |
On Wed, 2 Sep 1998, Sang Kang wrote:
> Hello, > > Under 2.1.119, I got the message "hdd: misaligned DMA buffer" while > I'm just cp'ing files from my hard drive to a md0 drive. since the > error is originated from hdd, which is my boot partition, I assume > it's got to do with my harddrive (6.4G, DMA/33 running on embedded > EIDE controller, MAXTOR). > > Would anyone kindly explain what this mean? >
Your hardware may have a non-standard entry point for the DMA request mapping. If you exceed the 64kB boundaries, you may/will rupture the table. This is unlikely.
My past experience with a 6.4G Maxtor (Diamond Series UDMA/33) was a so bad that I sent it back. I replaced it with a WD 6.4 UMDA drive. This was a mixed HD pair of a WD DMA mode 2 with a Maxtor UDMA mode 2. WD primary and Maxtor secondary on a PIIX3 ide0.
Try this silly test of hunting down timing errors. Recompile the kernel as a 386 and see if it goes away. If it does, repeat as a 486 and continue the process until you see the failures begin again. If they get progressively worse, you have a major timing problem and incompatable hardware combination. Solution, juggle the device combinations, preferred is to switch to Quantum UDMA drives.
First, would you supply a description of your hardware, before we start with the tests above. Since these tests can be "DISTRUCTIVE" to the filesystem.
Regards, Andre
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
| |