oops hit the send button too quick.
It would be nice to have a "design set" to test out different configurations
with a unified set of "stress tests". e.g. everyone that can, running the
tests with and w/o apic, nosmp, diff bios.., etc, etc. And then also have a
formalized way of reporting back your test data, using xml or something ;)
What does everyone think?
-charlie.
> ----------
> From: Fulton CR (Charles)
> Sent: Friday, March 17, 2000 12:53 PM
> To: 'linux-abit'
> Subject: RE: [LINUX-ABIT] Best stress test strategies? (was Re:
> Details)
>
> I'm not sure if you have posted this yet or not.. but would you mind
> posting all of your "stress test" strategies? I would be happy to run
> through them this weekend and to test my board better.
>
> thanks,
>
> -Charlie.
>
> ----------
> From: Valentijn Sessink[SMTP:[EMAIL PROTECTED]]
> Reply To: Discussion List for Linux on Abit Motherboards
> Sent: Friday, March 17, 2000 10:12 AM
> To: Discussion List for Linux on Abit Motherboards
> Subject: Re: [LINUX-ABIT] Re: Details
>
> Chris Chiappa wrote:
>
>
> > We're starting to go through the same stuff we did months ago:
> "SCSI vs IDE"
> > "Powersupplies" etc. Please read the FAQ I've drawn up at
>
> I know. So I don't have extra cooling, I dont' have a 300W
> powersupply.
> However, there are a couple of instances where a large disk IO on
> one
> plus some activity on the other processor makes the thing crash. Be
> it
> SCSI or IDE. (Had a couple of reports on SCSI bus resetting). I
> *also*
> saw a kernel mailing list message talking about a Dell SMP box that
> crashed... so maybe it is not the BP6 after all. Remember that there
> are
> many more Linux hackers with a BP6 than there are with other
> multiprocessor boxes, so any SMP error will be very BP6 biased...
>
> The stress testing so far seems mostly kernel compilations or
> processor
> burn in tests; I had the box running kernel compile for 8 hours, and
> it
> kept running. I also had a load 80 and it kept running. However,
> large
> disk IO operations seem to make it crash.
>
> Also, I'd still like to know about the "nosmp" option - anyone have
> the
> same results, apart from the reports so far?
>
> V.
> --
> --
> =- To unsubscribe, email [EMAIL PROTECTED] with the
> -=
> =- body of "unsubscribe linux-abit".
> -=
>
>
--
=- To unsubscribe, email [EMAIL PROTECTED] with the -=
=- body of "unsubscribe linux-abit". -=
