> - There have been reports of corruption with certain SCSI drivers too.
> But two or more bugs don't "cancel" each other.
For historical reasons, drivers and devices can be optionnaly configured
for no parity checking although parity checking is not an option in
SCSI but is mandatory.
So, before castrating any SCSI piece of software because of some silent
corruption report, it will be relevant to recommend people to only use
drivers, controllers and devices configured for SCSI parity checking.
On the other hand, the PCI BUS also implement parity since a DMA that
fails can corrupt the memory in any place (What about a scatter entry
that is corrupted when the controller reads it from the memory).
So, we also should be able to check that controllers and bridges check
PCI parity and _actually_ report errors or halt the system when such an
error occurs.
Any silent data corruption reported without at least all the above points
having been checked should be discarded in my opinion.
Gerard.
-
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.altern.org/andrebalsa/doc/lkml-faq.html