Hi Eric!
On Fri, 25 Jul 2003, Mudama, Eric wrote:
> 0x64 has the seek complete, write fault, and corrected data bits set...
>
> you get the short stalls, does the drive then keep functioning, or are the
> stalls fatal/permanent? The error recovery paths in our disk are generally
> configured to be "heroic", as in "do everything possible to try to get that
> data safely on the media"... this can take several seconds at least,
> depending on the type of error.
At least no reported loss of data. Filesystem is reiserfs.
> Do these same errors happen with other kernels?
It's ok with 2.4.21-pre6-ac1 and 2.6.0-test1-bk. So I go for the kernel
problem. But my knowledge of IDE and harddisks is about null.
> I'd suggest that your drive appears to be having trouble writing to the
> media... I'd back up your data if you can and do a full-pack zeroing/write
> of the drive. If that completes with no issues, then it's probably fine, if
> that has problems the drive ought to be RMA'd for a replacement... it could
> be one of those few out of a million that dies in the field within a year.
>
> --eric
>
>
> -----Original Message-----
> From: Axel Siebenwirth [mailto:axel@pearbough.net]
> Sent: Friday, July 25, 2003 1:08 PM
> To: linux-kernel@vger.kernel.org
> Subject: [2.4.22-pre8-bk] dma_timer_expiry: dma status == 0x64
>
>
> hi,
>
> this is a bug report?!
>
> with the linux kernel 2.4.22-pre8 from bk as of this day I get short stalls
> about 5 to 8 seconds resulting in nothing happening with my machine at all.
> (right now it happenend, but I can still type this)
>
> the kernel reports the following two messages:
>
> hda: dma_timer_expiry: dma status == 0x64
> hda: DMA interrupt recovery
> hda: lost interrupt
> ...
> hda: dma_timer_expiry: dma status == 0x64
> hda: DMA interrupt recovery
> hda: lost interrupt
> hda: dma_timer_expiry: dma status == 0x64
> hda: DMA interrupt recovery
> hda: lost interrupt
> hda: dma_timer_expiry: dma status == 0x64
> hda: DMA interrupt recovery
> hda: lost interrupt
>
> and so on.
>
> so maybe this helps you find an error in the kernel.
>
> best wishes,
> axel siebenwirth
>
> P.S. I attached gzipped dmesg and .config
-
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 Jul 31 2003 - 22:00:26 EST