Re: Problems with DMA on IDE/ServerWorks/Seagate.
From: Bartlomiej Zolnierkiewicz (B.Zolnierkiewicz_at_elka.pw.edu.pl)
Date: 07/15/04
- Previous message: IWAMOTO Toshihiro: "rebinding buffer_head to new(remapped) page"
- In reply to: Dave Woods: "Problems with DMA on IDE/ServerWorks/Seagate."
- Next in thread: Jeff Garzik: "Re: Problems with DMA on IDE/ServerWorks/Seagate."
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
To: "Dave Woods" <dwoods@fastclick.com>, <linux-kernel@vger.kernel.org> Date: Thu, 15 Jul 2004 12:24:45 +0200
On Thursday 15 of July 2004 02:16, Dave Woods wrote:
> Please CC me on all replies, as I am not subscribed to the list.
>
> We have diagnosed a problem (file corruption) using Ultra DMA & IDE with
>
> ServerWorks OSB4 Chipset and Seagate drives under heavy disk I/O.
> We were able to find a thread that discusses this problem in detail,
> but dates back to 2001. We have nearly 100 boxes using this
> configuration
> with kernal 2.4.20, and we can reproduce the problem easily with a perl
> script using md5 checksums.
>
> The applicable thread is given below:
> http://www.uwsg.iu.edu/hypermail/linux/kernel/0109.3/1006.html
>
> We've disabled DMA on several machines, and that appears to fix
> the problem. However, despite seeing this workaround mentioned
> several times, we have not had any luck downshifting to mdma2.
> Every time I attempt it, all disk accesses freeze up, and the box
> has to be rebooted.
>
> Our question: short of disabling DMA on all of our machines, what
> other alternatives are there? I've played with using the hdparm -p
> parameter prior to issuing hdparm -X34 and hdparm -d1, but it
> doesn't like it. Is there a simple fix to this problem out there?
Upgrade to 2.4.26 and see if it fixes your problem.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
- Previous message: IWAMOTO Toshihiro: "rebinding buffer_head to new(remapped) page"
- In reply to: Dave Woods: "Problems with DMA on IDE/ServerWorks/Seagate."
- Next in thread: Jeff Garzik: "Re: Problems with DMA on IDE/ServerWorks/Seagate."
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|
