> Second item: People reporting bugs with Intel i820 chipset machines where
> they get oopses or crashes please don't bother. See the intel board
> recall reports and bug your board vendor not Linux people. Until the recall
> issue is sorted and we know who or what identifies problem boards i820 bug
> reports are going in the bitbucket.
>
> [http://www.intel.com/support/mth/wtdcc820.htm]
All of the boards that are being recalled have an MTH chip, which allows
them to use SDRAM (as opposed to RDRAM). However, my understanding (which
could be incorrect) is not all boards with an MTH are defective. I don't
know if there's a way to differentiate the bad ones from the good ones.
(I think Intel has some kind of DOS utility on their web site which
simply checks if your board has an MTH, but my 'Net connection is flaky
at the moment so I can't verify this.)
So, what does this mean?
-Boards with SIMM slots and SDRAM might have the problem, but it's not
guaranteed.
-I would also think that adaptor cards which let you plug SIMMs into RIMM
slots would also be vulnerable because those cards have the MTH on them. I
don't know for sure, though.
-I don't know if i820 boards with both SIMM and RIMM slots are available or
even possible to make. If they are out there, I don't know if just using
RIMMs would be enough to keep the problem from happening.
-Finally, boards that have only RIMM slots, and which are using RDRAM, are
known to not be vulnerable. Taking this a step further, bug reports where
the board is known to have only RIMM slots and use only RDRAM probably
shouldn't be thrown away, because they are not at all affected by the
recall.
-Barry K. Nathan <barryn@pobox.com>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Tue May 23 2000 - 21:00:20 EST