Re: IO-APIC on nforce2 [PATCH] + [PATCH] for nmi_debug=1 + [PATCH]for idle=C1halt, 2.6.5

[Prakash K. Cheemplavam]

Craig Bradney wrote:
> On Mon, 2004-05-17 at 17:26, Prakash K. Cheemplavam wrote:
>
> > Hi all,
> >
> > I just made an interesting finding and would like to have comments from 
> > NVidia:
> >
> > Chip   Current Value   New Value
> > C17       1F0FFF01     1F01FF01
> > C18D      9F0FFF01     9F01FF01
> >
> > In fact I have the newer chip revision (lspci says c1), but due to a 
> > post at Abit Forums I tried to use the value for the older revision on 
> > my board, and guess what: I never had such low idle temps! I am 
> > currently even using nvidia binary graphics driver and usually I would 
> > be having around 49-51°C idle temp, but now it is around 45°C, and it 
> > was not the first boot (then the mobo usually shows 5°C less). Instead 
> > the temp steadily fell from >50°C to 45°C.
> >
> > (esp @nvidia:) Is there anything evil using the old chip's value for the 
> > new chip? So far I haven't noticed any bad thing about it. Perhaps some 
> > daring nforce2 user with the new revision should try as well.
>
>
> Isnt it the case that that change is the one that brings about
> stability? Was indicated before to be the main causing c1halt crashes.

Nope, I am changing the 9F to 1F. The "stability byte" was changing the 
0F to 01. I am no using 1F01FF01 instead of 9F01FF01. I guess I wasn't 
clear enough.

Prakash
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/