This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


[Xen-devel] [PATCH] 0/3: MCA/MCE correctable error handling

To: xen-devel@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-devel] [PATCH] 0/3: MCA/MCE correctable error handling
From: "Christoph Egger" <Christoph.Egger@xxxxxxx>
Date: Tue, 21 Aug 2007 15:31:17 +0200
Cc: Gavin.Maltby@xxxxxxx, Keir Fraser <keir@xxxxxxxxxxxxx>
Delivery-date: Tue, 21 Aug 2007 06:32:53 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: KMail/1.9.6
Hi Keir,

This is a full patchset you requested last week :-)
This patchset is for handling of correctable errors only. The other part 
related to the uncorrectable errors will be another patchset.

All patches apply against changeset 15772 (staging tree) in
the given order.

Patch 1/3:  MSR_CAP never reports a negative count of available 
error-reporting banks. Thus make nr_mce_banks unsigned.
While there, do some other minor indentation cleanups.

Patch 2/3:  Extend arch_shared_info with machine check structure.
This will be used to transfer machine check information to the guest
via a shared page. The guest will be notified via an event on correctable
errors and via NMI on uncorrectable errors. Former is patch 3/3, latter one
will be another patchset.
The structure is designed to be as small and as highly flexible as possible
at the same time.

Patch 3/3:  This introduces an architecture dependent VIRQ_MCA
and extends the polling service routine to notify the Dom0
via an event. The polling frequency is initially every 15 seconds. The 
frequency adjusts automatically at runtime. No error means, lower frequency,
correctable error means, increase frequency. It also uses hw threshold 
register (MSR_K8_MC4_MISC) to detect correctable errors between two polls.
If the Dom0 ignores the notification (= Dom0 has no handler for VIRQ_MCA 
registered), then Xen dumps the machine check information in a format,
mcelog can parse.

Signed-off-by: Christoph Egger <Christoph.Egger@xxxxxxx>

AMD Saxony, Dresden, Germany
Operating System Research Center

Legal Information:
AMD Saxony Limited Liability Company & Co. KG
Sitz (Geschäftsanschrift):
   Wilschdorfer Landstr. 101, 01109 Dresden, Deutschland
Registergericht Dresden: HRA 4896
vertretungsberechtigter Komplementär:
   AMD Saxony LLC (Sitz Wilmington, Delaware, USA)
Geschäftsführer der AMD Saxony LLC:
   Dr. Hans-R. Deppe, Thomas McCoy

Xen-devel mailing list

<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-devel] [PATCH] 0/3: MCA/MCE correctable error handling, Christoph Egger <=