lkml.org 
[lkml]   [1999]   [Dec]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: memory detection problems: 2.0.x vs. 2.2.x
In <Pine.LNX.4.10.9912040001020.16504-100000@squish.home.org> Paul (set@pobox.com) wrote:
> Dear All;

> I have an odd 386 machine running a 2.0.33 kernel. This kernel boots
> fine, and detects the correct amount of ram, as did many 2.0.x kernels
> before it.
> Recently, I needed the old beast for some tests, and it required a
> 2.2.x kernel. However, the 2.2.[12, 13] kernels that I tried detect
> a strange large amount of ram and quickly have a general protection
> fault. Using mem=xxxx solves the problem, but I am curious
> why there should be a problem.
> The only (relevant?) lines that differ on the good vs. bad boot are:

> ---2.0.33---
> Memory: 5920k/7424k available (688k kernel code, 354k reserved, 432k data)

> ---2.2.13--
> Memory: 37852k/39876k available (932k kernel code, 408k reserved, 652k data,
> 32k init)
> Checking if the processor honors the WP bit even in supervisor mode... No

> I can provide the panic dump, but dont see how that would help.
> An explaination of why the 2.2.x kernels tested fail to detect the memory
> correctly would be nice, and I am willing to do experiments to find out,
> if anyone wants to suggest them.

Linux 2.2 uses new int 15 subfunction to detect more then 64MiB of RAM (and
fallback to old method if that subfunction is not there) and you have broken
BIOS where that same call is used for completely other purpose. That's all.





-
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/

\
 
 \ /
  Last update: 2005-03-22 13:55    [W:0.045 / U:0.296 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site