Re: 2.1.110 and newer, SB AWE64

=?iso-8859-1?Q?Mikko_Hyv=E4rinen?= (mikko.hyvarinen@zoo-gate.fi)
Sun, 6 Sep 1998 18:08:46 +0300

>At 09:00 PM 9/2/98 +0300, you wrote:
> >Description of the hang:
>.
>.
>.
> >Description of hardware:
> >- Intel Pentium 200MMX running at 3 x 75MHz (225Mhz)
>
>What happens if you run it at 200Mhz??
>
> >- the processor has run at 225MHz for 10 months without any
> >problems on 2 different motherboards (just to shut up the
> >"don't overclock" people)
>
>That may be true but over clocking takes the chip closer to the edge of its
>performance envelope. Small changes like temperature, voltage, what
>instructions are executed or even what sequence the instructions are
>executed in (ie. cached vs. external) can cause transient timing glitches.
>The systems you describe are typical of such a glitch.
>
>I would at least try it at the correct clock speed to see if that fixes
>your problem.
>
>Dave

Please...

What has changed in the sound code since 2.0.xx to make it extremely hypersensitive to timing???

Linux 2.0.35, WinNT, Win95 and DOS 6.22 games have NO problems handling the card at all. To me this is a clear indication of the brokenness of 2.1.xx sound code.

I'm not ruling out the possibility that overclocking has a detrimental effect on the machine (I know it has) but I'm convinced that this particular bug is not directly related to overclocking.

But to prove my point, I did change my system to run at 3x66MHz (200MHz), let it run overnight cracking RC5 and then tried all my operating systems in the morning. No change in behaviour, 2.1.119 hangs the machine, 2.0.35, WinNT and Win95 all work like a dream.

I have seen discussion about a sound code memory leak still present in 2.1.120pre3 on this mailing list. Could it be the bug hanging my machine?

-Mikko Hyvärinen

-
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/faq.html