Re: [BUG?] Xeon with HyperThreading and linux-2.4.20-rc2

From: Steffen Persvold (sp@scali.com)
Date: Wed Nov 20 2002 - 08:27:20 EST


On Wed, 20 Nov 2002, Arjan van de Ven wrote:

> On Wed, Nov 20, 2002 at 12:53:04PM +0000, Hugh Dickins wrote:
> >
> > I know too little to comment definitively, but it's my understanding
> > that a dual HT machine should only show 2 processors in its MP table,
> > their siblings only appearing through analysis of the ACPI tables.
> >
> > Whether it's that your MP table has been wrongly set up, or that
> > you've really been given 4 processors when you only asked for 2
> > (sue your supplier!), I cannot say. I've copied Jun at Intel
> > and Arjan at RedHat, and hope they can shed more light on this.
>
> Linux has zero problem with a sane MP table that lists all
> CPU's. Intel normally seems to recommend against it (maybe N3.51 doesn't
> like it or so) but it's all fair as far as I'm concerned.
> The bios is supposed to offer you a choice
> to disable hyperthreading, use that ;)
>

Sure, the bios has this option (and it works). I just believed the 'noht'
option would disable it from a kernel perspective. I understand that if
the MP table lists 4 processors, the kernel must think it is 4 processors
and enable them. But what is the purpose of the 'noht' option ? If it is
to avoid scanning the ACPI table for CPUs, wouldn't it be less confusing
to call it something like 'acpismp=disable', since you apparently can't
disable the siblings anyway (when they are also listed in the MP table) ?

Regards,
 --
  Steffen Persvold | Scali AS
 mailto:sp@scali.com | http://www.scali.com
Tel: (+47) 2262 8950 | Olaf Helsets vei 6
Fax: (+47) 2262 8951 | N0621 Oslo, NORWAY

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



This archive was generated by hypermail 2b29 : Sat Nov 23 2002 - 22:00:31 EST