On Tue, Nov 13, 2001 at 09:51:21AM +0000, Alan Cox wrote:
> > I ask because I'm still having major problems with the OSB4 chipset on
> > 2.4.14. I've had to disable DMA completely on my boxes to avoid IDE errors
> > and fs corruption. Does anyone know of a patch that addresses the issues
> > with this chipset?
>
> No. I spent some time digging into this problem with both Serverworks and
> Red Hat customers. With certain disks, certain OSB4 revisions and UDMA
> the controller occasionally gets "stuck", the next DMA it issues starts
> by reissuing the last 4 bytes of the previous request and the entire thing
> goes totally to crap.
>
> The -ac tree will detect this case and panic and hang the machine solid to
> avoid actual disk corruption. The real fix appears to be "dont do UDMA on
> the OSB4". The CSB5 seems fine.
I am having problems with both UDMA and Multiword DMA. The problem doesn't
go away unless I disable CONFIG_IDEDMA_PCI_AUTO.
I don't know if there is actual FS corruption with MWord DMA, but there is
definitely a "hang" for a few seconds accompanied by a DMA error.
--Adam
-- Adam McKenna <adam@flounder.net> | GPG: 17A4 11F7 5E7E C2E7 08AA http://flounder.net/publickey.html | 38B0 05D0 8BF7 2C6D 110A - 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/
This archive was generated by hypermail 2b29 : Thu Nov 15 2001 - 21:00:34 EST