[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v2] intel_iommu: add "eim" property
From: |
Peter Xu |
Subject: |
Re: [Qemu-devel] [PATCH v2] intel_iommu: add "eim" property |
Date: |
Thu, 1 Sep 2016 09:31:05 +0800 |
User-agent: |
Mutt/1.5.24 (2015-08-30) |
On Fri, Aug 12, 2016 at 05:41:27PM +0800, Peter Xu wrote:
> Adding one extra property for intel-iommu device to decide whether we
> should support EIM bit for IR.
>
> Now we are throwing high 24 bits of dest_id away directly. This will
> cause interrupt issues with guests that:
>
> - enabled x2apic with cluster mode
> - have more than 8 vcpus (so dest_id[31:8] might be nonzero)
>
> Let's make xapic the default one, and for the brave people who would
> like to try EIM and know the side effects, we can do it by explicitly
> enabling EIM using:
>
> -device intel-iommu,intremap=on,eim=on
>
> Even after we have x2apic support, it'll still be good if we can provide
> a way to switch xapic/x2apic from QEMU side for e.g. debugging purpose,
> which is an alternative for tuning guest kernel boot parameters.
>
> We can switch the default to "on" after x2apic fully supported.
>
> Signed-off-by: Peter Xu <address@hidden>
Ping.
I'd really appreciate if someone can help have a look on this patch,
and to merge it if possible. Since current x2apic is broken. We should
not allow broken system boot, at least not by default.
Thanks,
-- peterx