virtio-iommu: Remove probe_done

Now we have switched to PCIIOMMUOps to convey host IOMMU information,
the host reserved regions are transmitted when the PCIe topology is
built. This happens way befo

virtio-iommu: Remove probe_done

Now we have switched to PCIIOMMUOps to convey host IOMMU information,
the host reserved regions are transmitted when the PCIe topology is
built. This happens way before the virtio-iommu driver calls the probe
request. So let's remove the probe_done flag that allowed to check
the probe was not done before the IOMMU MR got enabled. Besides this
probe_done flag had a flaw wrt migration since it was not saved/restored.

The only case at risk is if 2 devices were plugged to a
PCIe to PCI bridge and thus aliased. First of all we
discovered in the past this case was not properly supported for
neither SMMU nor virtio-iommu on guest kernel side: see

[RFC] virtio-iommu: Take into account possible aliasing in virtio_iommu_mr()
https://lore.kernel.org/all/20230116124709.793084-1-eric.auger@redhat.com/

If this were supported by the guest kernel, it is unclear what the call
sequence would be from a virtio-iommu driver point of view.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Message-Id: <20240716094619.1713905-3-eric.auger@redhat.com>
Tested-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Implement set|unset]_iommu_device() callbacks

Implement PCIIOMMUOPs [set|unset]_iommu_device() callbacks.
In set(), the HostIOMMUDevice handle is stored in a hash
table indexed by PCI

virtio-iommu: Implement set|unset]_iommu_device() callbacks

Implement PCIIOMMUOPs [set|unset]_iommu_device() callbacks.
In set(), the HostIOMMUDevice handle is stored in a hash
table indexed by PCI BDF. The object will allow to retrieve
information related to the physical IOMMU.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add an option to define the input range width

aw-bits is a new option that allows to set the bit width of
the input address range. This value will be used as a default for
the device c

virtio-iommu: Add an option to define the input range width

aw-bits is a new option that allows to set the bit width of
the input address range. This value will be used as a default for
the device config input_range.end. By default it is set to 64 bits
which is the current value.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Message-Id: <20240307134445.92296-7-eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add a granule property

This allows to choose which granule will be used by
default by the virtio-iommu. Current page size mask
default is qemu_target_page_mask so this translates
into

virtio-iommu: Add a granule property

This allows to choose which granule will be used by
default by the virtio-iommu. Current page size mask
default is qemu_target_page_mask so this translates
into a 4k granule on ARM and x86_64 where virtio-iommu
is supported.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>

Message-Id: <20240307134445.92296-3-eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add an option to define the input range width

aw-bits is a new option that allows to set the bit width of
the input address range. This value will be used as a default for
the device c

virtio-iommu: Add an option to define the input range width

aw-bits is a new option that allows to set the bit width of
the input address range. This value will be used as a default for
the device config input_range.end. By default it is set to 64 bits
which is the current value.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Message-Id: <20240307134445.92296-7-eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add a granule property

This allows to choose which granule will be used by
default by the virtio-iommu. Current page size mask
default is qemu_target_page_mask so this translates
into

virtio-iommu: Add a granule property

This allows to choose which granule will be used by
default by the virtio-iommu. Current page size mask
default is qemu_target_page_mask so this translates
into a 4k granule on ARM and x86_64 where virtio-iommu
is supported.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>

Message-Id: <20240307134445.92296-3-eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add an option to define the input range width

aw-bits is a new option that allows to set the bit width of
the input address range. This value will be used as a default for
the device c

virtio-iommu: Add an option to define the input range width

aw-bits is a new option that allows to set the bit width of
the input address range. This value will be used as a default for
the device config input_range.end. By default it is set to 64 bits
which is the current value.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Message-Id: <20240307134445.92296-7-eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add a granule property

This allows to choose which granule will be used by
default by the virtio-iommu. Current page size mask
default is qemu_target_page_mask so this translates
into

virtio-iommu: Add a granule property

This allows to choose which granule will be used by
default by the virtio-iommu. Current page size mask
default is qemu_target_page_mask so this translates
into a 4k granule on ARM and x86_64 where virtio-iommu
is supported.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>

Message-Id: <20240307134445.92296-3-eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Implement set_iova_ranges() callback

The implementation populates the array of per IOMMUDevice
host reserved ranges.

It is forbidden to have conflicting sets of host IOVA ranges
to be

virtio-iommu: Implement set_iova_ranges() callback

The implementation populates the array of per IOMMUDevice
host reserved ranges.

It is forbidden to have conflicting sets of host IOVA ranges
to be applied onto the same IOMMU MR (implied by different
host devices).

In case the callback is called after the probe request has
been issues by the driver, a warning is issued.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: "Michael S. Tsirkin" <mst@redhat.com>
Tested-by: Yanghang Liu <yanghliu@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>

show more ...

virtio-iommu: Record whether a probe request has been issued

Add an IOMMUDevice 'probe_done' flag to record that the driver
already issued a probe request on that device.

This will be useful to dou

virtio-iommu: Record whether a probe request has been issued

Add an IOMMUDevice 'probe_done' flag to record that the driver
already issued a probe request on that device.

This will be useful to double check host reserved regions aren't
notified after the probe and hence are not taken into account
by the driver.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Suggested-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: "Michael S. Tsirkin" <mst@redhat.com>
Tested-by: Yanghang Liu <yanghliu@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>

show more ...

virtio-iommu: Introduce per IOMMUDevice reserved regions

For the time being the per device reserved regions are
just a duplicate of IOMMU wide reserved regions. Subsequent
patches will combine those

virtio-iommu: Introduce per IOMMUDevice reserved regions

For the time being the per device reserved regions are
just a duplicate of IOMMU wide reserved regions. Subsequent
patches will combine those with host reserved regions, if any.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Yanghang Liu <yanghliu@redhat.com>
Reviewed-by: "Michael S. Tsirkin" <mst@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>

show more ...

virtio-iommu: Rename reserved_regions into prop_resv_regions

Rename VirtIOIOMMU (nb_)reserved_regions fields with the "prop_" prefix
to highlight those fields are set through a property, at machine

virtio-iommu: Rename reserved_regions into prop_resv_regions

Rename VirtIOIOMMU (nb_)reserved_regions fields with the "prop_" prefix
to highlight those fields are set through a property, at machine level.
They are IOMMU wide.

A subsequent patch will introduce per IOMMUDevice reserved regions
that will include both those IOMMU wide property reserved
regions plus, sometimes, host reserved regions, if the device is
backed by a host device protected by a physical IOMMU. Also change
nb_ prefix by nr_.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: "Michael S. Tsirkin" <mst@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>

show more ...

virtio-iommu: Implement set_iova_ranges() callback

The implementation populates the array of per IOMMUDevice
host reserved ranges.

It is forbidden to have conflicting sets of host IOVA ranges
to be

virtio-iommu: Implement set_iova_ranges() callback

The implementation populates the array of per IOMMUDevice
host reserved ranges.

It is forbidden to have conflicting sets of host IOVA ranges
to be applied onto the same IOMMU MR (implied by different
host devices).

In case the callback is called after the probe request has
been issues by the driver, a warning is issued.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: "Michael S. Tsirkin" <mst@redhat.com>
Tested-by: Yanghang Liu <yanghliu@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>

show more ...

virtio-iommu: Record whether a probe request has been issued

Add an IOMMUDevice 'probe_done' flag to record that the driver
already issued a probe request on that device.

This will be useful to dou

virtio-iommu: Record whether a probe request has been issued

Add an IOMMUDevice 'probe_done' flag to record that the driver
already issued a probe request on that device.

This will be useful to double check host reserved regions aren't
notified after the probe and hence are not taken into account
by the driver.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Suggested-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: "Michael S. Tsirkin" <mst@redhat.com>
Tested-by: Yanghang Liu <yanghliu@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>

show more ...

virtio-iommu: Introduce per IOMMUDevice reserved regions

For the time being the per device reserved regions are
just a duplicate of IOMMU wide reserved regions. Subsequent
patches will combine those

virtio-iommu: Introduce per IOMMUDevice reserved regions

For the time being the per device reserved regions are
just a duplicate of IOMMU wide reserved regions. Subsequent
patches will combine those with host reserved regions, if any.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Yanghang Liu <yanghliu@redhat.com>
Reviewed-by: "Michael S. Tsirkin" <mst@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>

show more ...

virtio-iommu: Rename reserved_regions into prop_resv_regions

Rename VirtIOIOMMU (nb_)reserved_regions fields with the "prop_" prefix
to highlight those fields are set through a property, at machine

virtio-iommu: Rename reserved_regions into prop_resv_regions

Rename VirtIOIOMMU (nb_)reserved_regions fields with the "prop_" prefix
to highlight those fields are set through a property, at machine level.
They are IOMMU wide.

A subsequent patch will introduce per IOMMUDevice reserved regions
that will include both those IOMMU wide property reserved
regions plus, sometimes, host reserved regions, if the device is
backed by a host device protected by a physical IOMMU. Also change
nb_ prefix by nr_.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: "Michael S. Tsirkin" <mst@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>

show more ...

virtio-iommu: Fix 64kB host page size VFIO device assignment

When running on a 64kB page size host and protecting a VFIO device
with the virtio-iommu, qemu crashes with this kind of message:

qemu-k

virtio-iommu: Fix 64kB host page size VFIO device assignment

When running on a 64kB page size host and protecting a VFIO device
with the virtio-iommu, qemu crashes with this kind of message:

qemu-kvm: virtio-iommu page mask 0xfffffffffffff000 is incompatible
with mask 0x20010000
qemu: hardware error: vfio: DMA mapping failed, unable to continue

This is due to the fact the IOMMU MR corresponding to the VFIO device
is enabled very late on domain attach, after the machine init.
The device reports a minimal 64kB page size but it is too late to be
applied. virtio_iommu_set_page_size_mask() fails and this causes
vfio_listener_region_add() to end up with hw_error();

To work around this issue, we transiently enable the IOMMU MR on
machine init to collect the page size requirements and then restore
the bypass state.

Fixes: 90519b9053 ("virtio-iommu: Add bypass mode support to assigned device")
Signed-off-by: Eric Auger <eric.auger@redhat.com>

Message-Id: <20230705165118.28194-2-eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Tested-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: Zhenzhong Duan <zhenzhong.duan@intel.com>

show more ...

virtio-iommu: Use recursive lock to avoid deadlock

When switching address space with mutex lock hold, mapping will be
replayed for assigned device. This will trigger relock deadlock.

Also release t

virtio-iommu: Use recursive lock to avoid deadlock

When switching address space with mutex lock hold, mapping will be
replayed for assigned device. This will trigger relock deadlock.

Also release the mutex resource in unrealize routine.

Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Message-Id: <20220613061010.2674054-3-zhenzhong.duan@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add bypass mode support to assigned device

Currently assigned devices can not work in virtio-iommu bypass mode.
Guest driver fails to probe the device due to DMA failure. And the
reaso

virtio-iommu: Add bypass mode support to assigned device

Currently assigned devices can not work in virtio-iommu bypass mode.
Guest driver fails to probe the device due to DMA failure. And the
reason is because of lacking GPA -> HPA mappings when VM is created.

Add a root container memory region to hold both bypass memory region
and iommu memory region, so the switch between them is supported
just like the implementation in virtual VT-d.

Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Message-Id: <20220613061010.2674054-2-zhenzhong.duan@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Use recursive lock to avoid deadlock

When switching address space with mutex lock hold, mapping will be
replayed for assigned device. This will trigger relock deadlock.

Also release t

virtio-iommu: Use recursive lock to avoid deadlock

When switching address space with mutex lock hold, mapping will be
replayed for assigned device. This will trigger relock deadlock.

Also release the mutex resource in unrealize routine.

Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Message-Id: <20220613061010.2674054-3-zhenzhong.duan@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add bypass mode support to assigned device

Currently assigned devices can not work in virtio-iommu bypass mode.
Guest driver fails to probe the device due to DMA failure. And the
reaso

virtio-iommu: Add bypass mode support to assigned device

Currently assigned devices can not work in virtio-iommu bypass mode.
Guest driver fails to probe the device due to DMA failure. And the
reason is because of lacking GPA -> HPA mappings when VM is created.

Add a root container memory region to hold both bypass memory region
and iommu memory region, so the switch between them is supported
just like the implementation in virtual VT-d.

Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Message-Id: <20220613061010.2674054-2-zhenzhong.duan@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Use recursive lock to avoid deadlock

When switching address space with mutex lock hold, mapping will be
replayed for assigned device. This will trigger relock deadlock.

Also release t

virtio-iommu: Use recursive lock to avoid deadlock

When switching address space with mutex lock hold, mapping will be
replayed for assigned device. This will trigger relock deadlock.

Also release the mutex resource in unrealize routine.

Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Message-Id: <20220613061010.2674054-3-zhenzhong.duan@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Add bypass mode support to assigned device

Currently assigned devices can not work in virtio-iommu bypass mode.
Guest driver fails to probe the device due to DMA failure. And the
reaso

virtio-iommu: Add bypass mode support to assigned device

Currently assigned devices can not work in virtio-iommu bypass mode.
Guest driver fails to probe the device due to DMA failure. And the
reason is because of lacking GPA -> HPA mappings when VM is created.

Add a root container memory region to hold both bypass memory region
and iommu memory region, so the switch between them is supported
just like the implementation in virtual VT-d.

Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Message-Id: <20220613061010.2674054-2-zhenzhong.duan@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Default to bypass during boot

Currently the virtio-iommu device must be programmed before it allows
DMA from any PCI device. This can make the VM entirely unusable when a
virtio-iommu

virtio-iommu: Default to bypass during boot

Currently the virtio-iommu device must be programmed before it allows
DMA from any PCI device. This can make the VM entirely unusable when a
virtio-iommu driver isn't present, for example in a bootloader that
loads the OS from storage.

Similarly to the other vIOMMU implementations, default to DMA bypassing
the IOMMU during boot. Add a "boot-bypass" property, defaulting to true,
that lets users change this behavior.

Replace the VIRTIO_IOMMU_F_BYPASS feature, which didn't support bypass
before feature negotiation, with VIRTIO_IOMMU_F_BYPASS_CONFIG.

We add the bypass field to the migration stream without introducing
subsections, based on the assumption that this virtio-iommu device isn't
being used in production enough to require cross-version migration at
the moment (all previous version required workarounds since they didn't
support ACPI and boot-bypass).

Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Message-Id: <20220214124356.872985-3-jean-philippe@linaro.org>
Acked-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...

virtio-iommu: Drop base_name and change generic_name

Drop base_name and turn generic_name into
"virtio-iommu-pci". This is more in line with
other modern-only devices.

Signed-off-by: Eric Auger <er

virtio-iommu: Drop base_name and change generic_name

Drop base_name and turn generic_name into
"virtio-iommu-pci". This is more in line with
other modern-only devices.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Suggested-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20211013191755.767468-3-eric.auger@redhat.com>
Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

show more ...