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


Re: [Xen-devel] [PATCH 0 of 6] Fix kexec in Xen (take 3)

To: <xen-devel@xxxxxxxxxxxxxxxxxxx>
Subject: Re: [Xen-devel] [PATCH 0 of 6] Fix kexec in Xen (take 3)
From: Wei Wang2 <wei.wang2@xxxxxxx>
Date: Wed, 25 May 2011 17:01:40 +0200
Cc: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
Delivery-date: Wed, 25 May 2011 08:08:32 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <patchbomb.1306333922@andrewcoop>
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/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <patchbomb.1306333922@andrewcoop>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: KMail/1.9.6 (enterprise 20070904.708012)
On Wednesday 25 May 2011 16:32:02 Andrew Cooper wrote:
> This set of patches is designed to get the kexec path working again on Xen
> 4.x
> kdump kernels can't boot if x2apic mode is enabled and the ACPI tables dont
> state this fact.  They also cant boot at all with interrupt remapping
> enabled.
> These patches cause xen to track the BSP local APIC boot state and return
> to it before kexec'ing to a new kernel.  It also makes sure to disable IO
> virtualisation.
> One area which is problematic is disabling interrupt remapping. 
> lapic_suspend() calls iommu_disable_x2apic_IR() which in a previous thread
> was deemed to be Intel specific and only works by chance on AMD boxes by
> effectivly being a NOP.  As lapic_suspend() is generic code, does this mean
> that we can't/don't ever disable interrupt remapping on AMD boxes?

Yes, there is no explicit way to disable amd iommu like disable_intremap() for 
vtd. The reason is that interrupt remapping is enabled per device and there 
is no global flag to disable it. We need to visit every device entry and then 
disable per device, but it still sounds doable... Maybe we could make 
disable_intremap() generic for both Intel and AMD?

> Signed-off-by: Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@xxxxxxxxxxxxxxxxxxx
> http://lists.xensource.com/xen-devel

Xen-devel mailing list