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

List:       linux-smp
Subject:    RE: unexpected IO APIC messages
From:       "Dunlap, Randy" <randy.dunlap () intel ! com>
Date:       2001-02-14 15:21:43
[Download RAW message or body]


> From: Maciej W. Rozycki [mailto:macro@ds2.pg.gda.pl]
> 
> On Tue, 13 Feb 2001, Dunlap, Randy wrote:
> 
> > b.  Modify the mptable program to do more than it already
> > does, i.e., to look at reserved fields and diagnose them.
> > (updated for Linux at ftp://virtualwire.org/mptable.c; originally
> > from http://people.freebsd.org/~fsmp/SMP/mptable/mptable.c).
> > And since the BIOS builds the MP table dynamically at boot
> > time and there may be multiple MP table signatures in
> > memory, also modify the mptable
> > to search all memory areas that an MP table could
> > reside so that it finds & prints multiple of them if found.
> > The kernel already prints the table address that it is using,
> > so that could be matched with the one(s) found by mptable.
> 
>  This isn't going to help you.  I/O APIC data (beside the ID, which is
> sometimes bogus anyway) is not kept in the MP table.  Trying 
> to get the MP
> table from Linux is unreliable anyway -- the proper copy 
> might have been
> present in EBDA and EBDA gets usually destroyed by Linux.  
> That's one of the reasons the initial dump is implemented. 

OK, thanks for the info.

~Randy

-
To unsubscribe from this list: send the line "unsubscribe linux-smp" in
the body of a message to majordomo@vger.kernel.org

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