Take any two: motherboard performance, compatibility, value
Josip Loncaric
josip at icase.edu
Thu Jun 29 05:39:13 PDT 2000
Bob Drzyzgula wrote:
>
> On Wed, Jun 28, 2000 at 10:28:58PM +0200, Jakob Østergaard wrote:
> >
> > Are you absolutely certain that ECC RAM on PC hardware actually *corrects*
> > bit errors ?
[ Intel says yes ]
> I would expect, for example, that the chipset
> would raise some sort of alert if a single-bit ECC error
> was detected and corrected; certainly the OS would want
> to log such an event.
After BIOS activates ECC, the 440BX chipset logs all corrected single
bit errors (and all detected multiple bit errors); but Linux kernel does
*not* automatically monitor the relevant 440BX register. You need to
donload, compile and insert the 'ecc' module to get ECC logging. This
module (which currently works *only* on uniprocessor machines) is
available from
http://www.anime.net/~goemon/linux-ecc/
This module produces the following kind of output:
Jun 8 18:47:11 n009 kernel: ECC: monitor version 0.9 (Oct 15 1999)
Jun 27 19:04:30 n009 kernel: ECC: SBE detected in DRAM row 3
Jun 27 19:04:30 n009 kernel: ECC: SBE at memory address 8000
Sincerely,
Josip
--
Dr. Josip Loncaric, Senior Staff Scientist mailto:josip at icase.edu
ICASE, Mail Stop 132C PGP key at http://www.icase.edu./~josip/
NASA Langley Research Center mailto:j.loncaric at larc.nasa.gov
Hampton, VA 23681-2199, USA Tel. +1 757 864-2192 Fax +1 757 864-6134
More information about the Beowulf
mailing list