[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-smp
Subject:    Re: Solid freezes with 2.2.1 - 2.2.3-ac1 SMP
From:       Dave Cinege <dcinege () psychosis ! com>
Date:       1999-03-23 17:22:35
[Download RAW message or body]

"Robert G. Brown" wrote:
> 
> > Don't think so. I'm not seeing any such freezes with 2.0.36.
> > It also reports 450.56 bogomips for each CPU. 2.2.3 reports 448.92/450.56.
> 
> Have you tried putting the CPU clocks back to the rated values for your
> chips?  It may well be that 2.2.x is tighter in its timing tolerances
> than 2.0.x as it handles stacked interrupts.  I'm assuming that you're
> still overclocking everything as a matter of religion...

I moved the speed down to 450 and ran it for a week before even posting. They
normally run at 463, and have been tested much higher. With Celeron A's it's
easy for the novice to misidentify their true top end speed, because they will
remain idle without problems, and only show them during heavy load. That's not
the case here at all.Yesterday morning when I first turned the machine on, it
locked a few minutes after boot, not doing, and having not done much of
anything. (The room was also quite cold yesterday morning....) With 2.0.36
these chips have run fine sitting idle @503MHz for LONG periods of time. The
dice don't role at even that speed until system load gets high. 

This is not an OC problem.

I'd still like an explaination why the bogomips are off with 2.2. Why is there
a variation in the timing loop??

-- 
http://www.linkscape.net/       Linkscape Internet Services   732-541-4214
http://www.linuxrouter.org/     Linux Router Project

"If you're going to be stupid, don't be half-ass stupid, be all the way
stupid."
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to majordomo@vger.rutgers.edu

[prev in list] [next in list] [prev in thread] [next in thread]