Ext3 users - Apr 2007 - Compact Flash, EXT3, and IDE - a bad idea?

Hi All,

 

  I'm currently running an embedded linux system (arm) with a CF card
mounted as EXT3. I'll copy some system details below... Anyway, I'm
getting some drive errors and I'm wondering if they're due to bad/old
drivers, something with EXT3 and CF latency, or something else entirely.

 

These are the errors:

 

hda: write_intr error1: nr_sectors=1, stat=0x51

hda: write_intr: status=0x51 { DriveReady SeekComplete Error }

hda: write_intr: error=0x04 { DriveStatusError }

end_request: I/O error, dev 03:01 (hda), sector 528431

 

Let me know if you have any idea what I could do to fix these issues. Ideally it
wouldn't involve a recompile of the kernel!

 

 

Thanks in advance... 

 

 

 

Some system/configuration info:

 

kernel 2.4.9-ac10

PCMCIA CS 3.1.22

hda:SA1110 SPI v1.3 

EXT3 FS 2.4-0.9.6, 11 Aug 2001

 

CONFIG_IDEDISK_MULTI_MODE is not set

 

Proc/interrupts:

  2:      12103   xr788

  3:      34879   xr788

 11:     197994   GPIO 11-27

 15:      19318   serial_sa1100

 17:      21510   serial_sa1100

 26:   13907065   timer

 27:          0   rtc timer

 30:          1   rtc 1Hz

 31:          0   rtc Alrm

 38:          0   wdog snarl

 40:          0   CF_BVD2

 41:          0   CF_BVD1

 43:     897044   ide0

 44:          0   CF_CD

 47:      76675   eth0

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://listman.redhat.com/archives/ext3-users/attachments/20070423/319de149/attachment.htm>

Just off the cuff, I can't see it being likely an Ext3 problem -- as
long as the driver is passing along the proper device size ext3
shouldn't have any problems -- However, if you have any questions
about that, you should explicitly pass mkfs (or mkfs.ext[23] an
explicit size parameter).

( in that respect, I'd be more interested in seeing /proc/partitions
than /proc/interrupts).

check to see if you can do a raw write of the entire device (dd
if=/dev/zero of=/dev/hda bs=10k). It's about as simple as  you can
get, and it chould tell you more about the style of te problem.

On 4/23/07, Kieft, Brian <bkieft at mbari.org> wrote:
>   I'm currently running an embedded linux system (arm) with a CF card
> mounted as EXT3. I'll copy some system details below? Anyway, I'm
getting
> some drive errors and I'm wondering if they're due to bad/old
drivers,
> something with EXT3 and CF latency, or something else entirely.
>
> hda: write_intr error1: nr_sectors=1, stat=0x51
>
> hda: write_intr: status=0x51 { DriveReady SeekComplete Error }
>
> hda: write_intr: error=0x04 { DriveStatusError }
>
> end_request: I/O error, dev 03:01 (hda), sector 528431
-- 
Stephen Samuel http://www.bcgreen.com
778-861-7641