lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Mon, 02 Aug 2010 13:53:30 -0700
From:	Suresh Siddha <suresh.b.siddha@...el.com>
To:	Yinghai Lu <yinghai@...nel.org>
Cc:	Len Brown <lenb@...nel.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	"H. Peter Anvin" <hpa@...or.com>,
	"linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Ingo Molnar <mingo@...e.hu>
Subject: Re: [PATCH] acpi: x2apic entry with uid < 255 could use processor
 statement

On Mon, 2010-08-02 at 13:18 -0700, Yinghai Lu wrote:
> On Mon, Aug 2, 2010 at 12:06 PM, Suresh Siddha
> <suresh.b.siddha@...el.com> wrote:
> > On Sat, 2010-07-31 at 07:51 +0100, Yinghai Lu wrote:
> >> According to Intel x2apic spec page 46
> >>
> >> " The hand-off to
> >> OSPM will have processor IDs in the range of 0 to 254 for xAPIC/x2APIC and 0 to 255
> >> for SAPIC declared as either Processor() or Device() objects, but not both. Processor
> >> IDs outside these ranges must be declared as Device() objects."
> >>
> >> So only check if Device is used when acpi_id >=255.
> >>
> >> that will help system with less 255 cpus, but some cpus apic id > 255,
> >> still can use Processor statement instead of Device() objects.
> >
> > But the entries with apic_id < 255 are supposed to use local APIC
> > structure and not local x2apic structure. So entries with apic id < 255
> > must be processed using map_lapic_id() which doesn't have any
> > device_declaration checks.
> >
> > Only for apic ids > 255, we use map_x2apic_id() which needs device
> > declaration. So this patch is not needed. or Am I missing something?
> 
> it is acpi_id aka  Processor id.
> 
> the system has less than 255 cpus, but some cpus apic_id > 255.
> BIOS have apic entries for apic_id < 255, and some x2apic entries for
> apic_id > 255.
> 
> but BIOS still use Processor statement for all cpus.

Ok. I think there might be some confusion or mis-interpretation of the
words here. You referred to x2apic spec page 46, perhaps this is an
older version. Newer x2apic version leaves all the ACPI definitions to
the ACPI 4.0 spec.

And here is what ACPI 4.0 spec says:

In Table5-33 for processor local x2apic structure:

ACPI Processor UID
4
12
OSPM associates the X2APIC Structure with a processor object declared in
the namespace using the Device statement, when the _UID child object of
the processor device evaluates to a numeric value, by matching the
numeric value with this field

And in page 312:

<snip>
The platform may declare processors with IDs in the range of 0-254 for
APIC/x2APIC implementations and 0-255 for SAPIC implementations using
either the ASL Processor statement or the ASL Device statement but not
both. Processors with IDs outside these ranges must be declared using
the ASL Device statement.
</snip>

And in the above paragraph "processors with IDs" are APIC id's and not
ACPI Id's.

So I think your bios need to implement ACPI device objects for the
x2apic entries.

thanks,
suresh

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

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ