Re: [patch] smp-2.3.48-B2, MP config table in high memory

From: Lyle Coder (x_coder@hotmail.com)
Date: Fri Mar 03 2000 - 13:08:10 EST


Hi,
I will check with the BIOS engineers to see what they have to say... I will
get back to you very quickly :). Basically I want to see if they are infact
putting the tables somewhere > 1GB on a machine where so much RAM is
available.

If my understanding is correct, their algorithm trie sto put it at the
highest available RAM. So say on a machine with 4GB RAM I think we should
expect to find the MP and ACPI tables close to 4GB.

In any event let me confirm this.

If this is so, I have a fix for this (at least I think so :)

Best Wishes,
Lyle

>From: Ingo Molnar <mingo@chiara.csoma.elte.hu>
>To: Lyle Coder <x_coder@hotmail.com>
>CC: linux-kernel@vger.rutgers.edu
>Subject: Re: [patch] smp-2.3.48-B2, MP config table in high memory
>Date: Fri, 3 Mar 2000 12:10:01 +0100 (CET)
>
>
>On Thu, 2 Mar 2000, Lyle Coder wrote:
>
> > So I saw the latest kernel moves paging_init before reading the MP
>tables.
> > My question is this... on our systems, the MP tables will be in the BIOS
> > protected area (ie the area that the e820 call will report as unusable).
> > Will the de-reference of such an address still work once paging_init has
> > been called? [...]
>
>yes. We map all RAM from phys address 0 to phys address max linearly on
>x86. It will not however map more than 1GB physical - do you expect any
>such case to happen?
>
> > [...] Also, is this intended to work because paging_init installs
> > translations for all available RAM?
>
>is there any reason it should not work? We map all physical memory
>(including the ROMs below 1MB or any other unusable area mentioned in the
>e820 table). More accurately, on x86 we are mapping:
>
> [0 ... min(max(e820.end), ~1GB)]
>
>-- mingo
>

______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com

-
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/



This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:15 EST