Which is why Im collecting statistics. And right now they say that you
are most likely to have disk corruption if you run SMP and you are
as likely to have disk corruption if you run SCSI or IDE.
The reports collated so far suggest there is no IDE DMA problem, there
is possibly a generic SMP problem.
> The argument is proven false by the fact that people who saw corruption
> with DMA on didn't see it with DMA off. In short, my counter-argument is
> that we _know_ that DMA is unsafe. Rather than wild and idle speculation,
> we have FACTS. And when it comes to fact vs wild speculation, I'll go with
> the facts any day.
You haven't even attempted to measure the "does turning DMA off cause
problems" case. Currently Im trying to measure them all. The more messages
I get from people detailing corruption cases the better.
> And yes, Alan, it's trivial to get 10 times throughput improvement by
> being buggy.
Or just run FreeBSD, which doesn't seem to be having any such problems.
The only known 'buggy' case we have is some WD drives, which are noted
so on their web site according to one poster. But then our 2.0 IDE code
destroyed older WD 1Gig drives due to a firmware bug in the drives anyway.
But given the numbers collated so far from my stats gathering exercise
you would be better turning off SCSI and SMP than IDE. And thats assuming
an even split between SCSI and IDE users - whereas it must be > 10:1 IDE
over SCSI.
Alan
-
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