Re: [Qemu-devel] [PATCH v2 00/13] IOMMU: Enable interrupt remapping for Intel IOMMU

[Peter Xu]

On Mon, Apr 11, 2016 at 04:19:01PM -0600, Alex Williamson wrote:
> On Mon, 11 Apr 2016 17:19:10 +0800
> Peter Xu <address@hidden> wrote:
> 
> > v2 changes:
> > - patch 1
> >   - rename "int_remap" to "intr" in several places [Marcel]
> >   - remove "Intel" specific words in desc or commit message, prepare
> >     itself with further AMD support [Marcel]
> >   - avoid using object_property_get_bool() [Marcel]
> > - patch 5
> >   - use PCI bus number 0xff rather than 0xf0 for the IOAPIC scope
> >     definition. (please let me know if anyone knows how I can avoid
> >     user using PCI bus number 0xff... TIA)
> > - patch 11
> >   - fix comments [Marcel]
> > - all
> >   - remove intr_supported variable [Marcel]
> > 
> > This patchset provide very basic functionalities for interrupt
> > remapping (IR) support of the emulated Intel IOMMU device.
> > 
> > By default, IR is disabled to be better compatible with current
> > QEMU. To enable IR, we can using the following command to boot a
> > IR-supported VM with basic network (still do not support kvm-ioapic,
> > so we need to specify kernel-irqchip=off here):
> > 
> > $ qemu-system-x86_64 -M q35,iommu=on,intr=on,kernel-irqchip=off \
> >      -enable-kvm -m 1024 -s \
> >      -monitor telnet::3333,server,nowait \
> >      -netdev user,id=user.0,hostfwd=tcp::5555-:22 \
> >      -device virtio-net-pci,netdev=user.0 \
> >      /var/lib/libvirt/images/vm1.qcow2
> > 
> > When guest boots, we can verify whether IR enabled by grepping the
> > dmesg like:
> > 
> > address@hidden ~]# journalctl -k | grep "DMAR-IR"
> > Feb 19 11:21:23 localhost.localdomain kernel: DMAR-IR: IOAPIC id 0 under 
> > DRHD base  0xfed90000 IOMMU 0
> > Feb 19 11:21:23 localhost.localdomain kernel: DMAR-IR: Enabled IRQ 
> > remapping in xapic mode
> > 
> > Currently only two devices are supported:
> > 
> > - Emulated IOAPIC device
> > - PCI Devices
> > 
> > TODO List:
> > 
> > - kvm-ioapic support
> > - vhost support
> > - pass through device support

Hi, Alex,

> 
> vhost and assigned devices should be identical, which boils down to
> irqfd support.  I imagine a vfio-pci device works with what you've got
> here since kvm irqfd support will be disabled without kernel irqchip
> (vfio doesn't depend on it).

Thanks for the confirmation on vfio-pci part. That would be great if
we can make it work without special care. :)

> 
> Also, patches 1, 2, and 4 seem mis-ordered, I would think we only want
> to advertise IR support both to the VM and to the user once it's
> complete, otherwise we have versions where the options are enabled yet
> don't work, even if only in git.  Otherwise very nice patch breakdown.

Do you mean I should put patch 1 to the end of the series? Since
even with patch 2 and 4, IR won't work as well, and guest will hang
at kernel boot.  I can move patch 1 to the end of series
though. That will make sure that when people see the parameter, they
can use it as well.

Thanks!

-- peterx