Messages in this thread |  | | Date | Wed, 27 Sep 2000 12:40:22 +0200 (MET DST) | From | "Maciej W. Rozycki" <> | Subject | Re: what causes Machine Check exception? |
| |
On Wed, 27 Sep 2000, Arjan van de Ven wrote:
> "This means that your CPU indicates that it is defective"
An MCE might also mean an external error. For example, the Mercury/Neptune chipsets used to report MB memory and PCI parity errors via the BUSCHK# CPU input, which in turn triggers an MCE in Pentia.
BTW, this is a better way of reporting such errors, than using NMI, in my opinion, especially as triggering an MCE latches the address and data buses of the affected cycle (chipset could possibly drive appropriate data if a parity error happens on a posted write, for example) so they can be examined directly later.
-- + Maciej W. Rozycki, Technical University of Gdansk, Poland + +--------------------------------------------------------------+ + e-mail: macro@ds2.pg.gda.pl, PGP key available +
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/
|  |