2001-02-14 21:12:37

by David D.W. Downey

[permalink] [raw]
Subject: MP-Table mappings


In my dmesg I'm getting duplicate table reservations.

found SMP MP-table at 000f5770
hm, page 000f5000 reserved twice.
hm, page 000f6000 reserved twice.
hm, page 000f1000 reserved twice.
hm, page 000f2000 reserved twice.
On node 0 totalpages: 262128
zone(0): 4096 pages.
zone(1): 225280 pages.
zone(2): 32752 pages.
Intel MultiProcessor Specification v1.4
Virtual Wire compatibility mode.
OEM ID: OEM00000 Product ID: PROD00000000 APIC at: 0xFEE00000



Is this an issue?


David



2001-02-14 21:27:40

by Alan

[permalink] [raw]
Subject: Re: MP-Table mappings

> In my dmesg I'm getting duplicate table reservations.

Just a crap bios

> OEM ID: OEM00000 Product ID: PROD00000000 APIC at: 0xFEE00000

I think the required OEM ID and product id speak volumes for the rest
of the quality issues

> Is this an issue?

Once is correct, twice is fine, zero times would be bad.

Its ok

2001-02-15 14:13:17

by Maciej W. Rozycki

[permalink] [raw]
Subject: Re: MP-Table mappings

On Wed, 14 Feb 2001, Alan Cox wrote:

> > In my dmesg I'm getting duplicate table reservations.
>
> Just a crap bios

That's unrelated -- duplicate reservations are due to the MP table being
located in memory areas marked as "reserved" (ROM, ususally) in the map.
Thus the area is never freed in the first place and when smp_scan_config()
calls reserve_bootmem() for the pages a warning is issued. Harmless,
indeed.

--
+ Maciej W. Rozycki, Technical University of Gdansk, Poland +
+--------------------------------------------------------------+
+ e-mail: [email protected], PGP key available +

2001-02-15 14:15:57

by Alan

[permalink] [raw]
Subject: Re: MP-Table mappings

> > Just a crap bios
>
> That's unrelated -- duplicate reservations are due to the MP table being
> located in memory areas marked as "reserved" (ROM, ususally) in the map.

Ah. Ok I'd not seen that specific case

> Thus the area is never freed in the first place and when smp_scan_config()
> calls reserve_bootmem() for the pages a warning is issued. Harmless,
> indeed.

Umm probably worth cleaning up.