Some of the callbacks in VFIOIOMMUOps pass VFIOContainerBase poiner,
those callbacks only need read access to the sub object of VFIOContainerBase.
So make VFIOContainerBase, VFIOContainer and VFIOIOMMUFDContainer as const
in these callbacks.
Local functions called by those callbacks also need same changes to avoid
build error.
Suggested-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
This gives management tools like libvirt a chance to open the vfio
cdev with privilege and pass FD to qemu. This way qemu never needs
to have privilege to open a VFIO or iommu cdev node.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Now we support two types of iommu backends, let's add the capability
to select one of them. This depends on whether an iommufd object has
been linked with the vfio-ccw device:
If the user wants to use the legacy backend, it shall not
link the vfio-ccw device with any iommufd object:
-device vfio-ccw,sysfsdev=/sys/bus/mdev/devices/XXX
This is called the legacy mode/backend.
If the user wants to use the iommufd backend (/dev/iommu) it
shall pass an iommufd object id in the vfio-ccw device options:
-object iommufd,id=iommufd0
-device vfio-ccw,sysfsdev=/sys/bus/mdev/devices/XXX,iommufd=iommufd0
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
This gives management tools like libvirt a chance to open the vfio
cdev with privilege and pass FD to qemu. This way qemu never needs
to have privilege to open a VFIO or iommu cdev node.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Now we support two types of iommu backends, let's add the capability
to select one of them. This depends on whether an iommufd object has
been linked with the vfio-ap device:
if the user wants to use the legacy backend, it shall not
link the vfio-ap device with any iommufd object:
-device vfio-ap,sysfsdev=/sys/bus/mdev/devices/XXX
This is called the legacy mode/backend.
If the user wants to use the iommufd backend (/dev/iommu) it
shall pass an iommufd object id in the vfio-ap device options:
-object iommufd,id=iommufd0
-device vfio-ap,sysfsdev=/sys/bus/mdev/devices/XXX,iommufd=iommufd0
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Matthew Rosato <mjrosato@linux.ibm.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
This gives management tools like libvirt a chance to open the vfio
cdev with privilege and pass FD to qemu. This way qemu never needs
to have privilege to open a VFIO or iommu cdev node.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Now we support two types of iommu backends, let's add the capability
to select one of them. This depends on whether an iommufd object has
been linked with the vfio-platform device:
If the user wants to use the legacy backend, it shall not
link the vfio-platform device with any iommufd object:
-device vfio-platform,host=XXX
This is called the legacy mode/backend.
If the user wants to use the iommufd backend (/dev/iommu) it
shall pass an iommufd object id in the vfio-platform device options:
-object iommufd,id=iommufd0
-device vfio-platform,host=XXX,iommufd=iommufd0
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
This gives management tools like libvirt a chance to open the vfio
cdev with privilege and pass FD to qemu. This way qemu never needs
to have privilege to open a VFIO or iommu cdev node.
Together with the earlier support of pre-opening /dev/iommu device,
now we have full support of passing a vfio device to unprivileged
qemu by management tool. This mode is no more considered for the
legacy backend. So let's remove the "TODO" comment.
Add helper functions vfio_device_set_fd() and vfio_device_get_name()
to set fd and get device name, they will also be used by other vfio
devices.
There is no easy way to check if a device is mdev with FD passing,
so fail the x-balloon-allowed check unconditionally in this case.
There is also no easy way to get BDF as name with FD passing, so
we fake a name by VFIO_FD[fd].
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Now we support two types of iommu backends, let's add the capability
to select one of them. This depends on whether an iommufd object has
been linked with the vfio-pci device:
If the user wants to use the legacy backend, it shall not
link the vfio-pci device with any iommufd object:
-device vfio-pci,host=0000:02:00.0
This is called the legacy mode/backend.
If the user wants to use the iommufd backend (/dev/iommu) it
shall pass an iommufd object id in the vfio-pci device options:
-object iommufd,id=iommufd0
-device vfio-pci,host=0000:02:00.0,iommufd=iommufd0
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Implement the newly introduced pci_hot_reset callback named
iommufd_cdev_pci_hot_reset to do iommufd specific check and
reset operation.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Legacy vfio pci and iommufd cdev have different process to hot reset
vfio device, expand current code to abstract out pci_hot_reset callback
for legacy vfio, this same interface will also be used by iommufd
cdev vfio device.
Rename vfio_pci_hot_reset to vfio_legacy_pci_hot_reset and move it
into container.c.
vfio_pci_[pre/post]_reset and vfio_pci_host_match are exported so
they could be called in legacy and iommufd pci_hot_reset callback.
Suggested-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
This helper will be used by both legacy and iommufd backends.
No functional changes intended.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Some vIOMMU such as virtio-iommu use IOVA ranges from host side to
setup reserved ranges for passthrough device, so that guest will not
use an IOVA range beyond host support.
Use an uAPI of IOMMUFD to get IOVA ranges of host side and pass to
vIOMMU just like the legacy backend, if this fails, fallback to
64bit IOVA range.
Also use out_iova_alignment returned from uAPI as pgsizes instead of
qemu_real_host_page_size() as a fallback.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Currently iommufd doesn't support dirty page sync yet,
but it will not block us doing live migration if VFIO
migration is force enabled.
So in this case we allow set_dirty_page_tracking to be NULL.
Note we don't need same change for query_dirty_bitmap because
when dirty page sync isn't supported, query_dirty_bitmap will
never be called.
Suggested-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
The iommufd backend is implemented based on the new /dev/iommu user API.
This backend obviously depends on CONFIG_IOMMUFD.
So far, the iommufd backend doesn't support dirty page sync yet.
Co-authored-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
This is a trivial optimization. If there is active container in space,
vfio_reset_handler will never be unregistered. So revert the check of
space->containers and return early.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
/dev/vfio/devices/vfioX may not exist. In that case it is still possible
to open /dev/char/$major:$minor instead. Add helper function to abstract
the cdev open.
Suggested-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Introduce an iommufd object which allows the interaction
with the host /dev/iommu device.
The /dev/iommu can have been already pre-opened outside of qemu,
in which case the fd can be passed directly along with the
iommufd object:
This allows the iommufd object to be shared accross several
subsystems (VFIO, VDPA, ...). For example, libvirt would open
the /dev/iommu once.
If no fd is passed along with the iommufd object, the /dev/iommu
is opened by the qemu code.
Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Tested-by: Eric Auger <eric.auger@redhat.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Introduce an empty spapr backend which will hold spapr specific
content, currently only prereg_listener and hostwin_list.
Also introduce two spapr specific callbacks add/del_window into
VFIOIOMMUOps. Instantiate a spapr ops with a helper setup_spapr_ops
and assign it to bcontainer->ops.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Meanwhile remove the helper function vfio_free_container as it
only calls g_free now.
No functional change intended.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Move listener to base container. Also error and initialized fields
are moved at the same time.
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
In the prospect to get rid of VFIOContainer refs
in common.c lets convert misc functions to use the base
container object instead:
vfio_devices_all_dirty_tracking
vfio_devices_all_device_dirty_tracking
vfio_devices_all_running_and_mig_active
vfio_devices_query_dirty_bitmap
vfio_get_dirty_bitmap
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
VFIO Device is also changed to point to base container instead of
legacy container.
No functional change intended.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
dirty_pages_supported field is also moved to the base container
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Move the space field to the base object. Also the VFIOAddressSpace
now contains a list of base containers.
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Move the giommu_list field in the base container and store
the base container in the VFIOGuestIOMMU.
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
This adds two helper functions vfio_container_init/destroy which will be
used by both legacy and iommufd containers to do base container specific
initialization and release.
No functional change intended.
Suggested-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
This empty VFIOIOMMUOps named vfio_legacy_ops will hold all general
IOMMU ops of legacy container.
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Introduce a dumb VFIOContainerBase object and its targeted interface.
This is willingly not a QOM object because we don't want it to be
visible from the user interface. The VFIOContainerBase will be
smoothly populated in subsequent patches as well as interfaces.
No functional change intended.
Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Yi Liu <yi.l.liu@intel.com>
Signed-off-by: Yi Sun <yi.y.sun@linux.intel.com>
Signed-off-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Signed-off-by: Cédric Le Goater <clg@redhat.com>
Reduce atomicity while emitting opcodes, instead of later
during code generation. This ensures that any helper called
also sees the reduced atomicity requirement.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2034
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20231212193542.149117-1-richard.henderson@linaro.org>
In 32-bit mode, pc = eip + cs_base is also 32-bit, and must wrap.
Failure to do so results in incorrect memory exceptions to the guest.
Before 732d548732, this was implicitly done via truncation to
target_ulong but only in qemu-system-i386, not qemu-system-x86_64.
To fix this, we must add conditional zero-extensions.
Since we have to test for 32 vs 64-bit anyway, note that cs_base
is always zero in 64-bit mode.
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2022
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20231212172510.103305-1-richard.henderson@linaro.org>
I noticed the code blocks where not rendering properly so thought I'd
better fix things up. So:
- Use better title for the machine type
- Explain why Xen is a little different
- Add a proper anchor to the tpm-device link
- add newline so code block properly renders
- add some indentation to make continuation clearer
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20231207130623.360473-1-alex.bennee@linaro.org>
The AioContext must be unlocked before calling blk_co_unref(), because
it takes the AioContext lock internally in blk_unref_bh(), which is
scheduled in the main thread. If we don't unlock, the AioContext is
locked twice and nested event loops such as in bdrv_graph_wrlock() will
deadlock.
Cc: <qemu-stable@nongnu.org>
Fixes: https://issues.redhat.com/browse/RHEL-15965
Fixes: 0c7d204f50
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20231208124352.30295-1-kwolf@redhat.com>
GUEST_VIRTIO_MMIO_* was added in Xen 4.17, so only define them
for CONFIG_XEN_CTRL_INTERFACE_VERSIONs up to 4.16.
Reported-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A misspelled condition in xen_native.h is hiding a bug in the enablement of
Xen for qemu-system-aarch64. The bug becomes apparent when building for
Xen 4.18.
While the i386 emulator provides the xenpv machine type for multiple architectures,
and therefore can be compiled with Xen enabled even when the host is Arm, the
opposite is not true: qemu-system-aarch64 can only be compiled with Xen support
enabled when the host is Arm.
Expand the computation of accelerator_targets['CONFIG_XEN'] similar to what is
already there for KVM.
Cc: Stefano Stabellini <stefano.stabellini@amd.com>
Cc: Richard W.M. Jones <rjones@redhat.com>
Cc: Daniel P. Berrangé <berrange@redhat.com>
Reported-by: Michael Young <m.a.young@durham.ac.uk>
Fixes: 0c8ab1cddd ("xen_arm: Create virtio-mmio devices during initialization", 2023-08-30)
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit 7191f24c7f ("accel/kvm/kvm-all: Handle register access errors")
added error checking for KVM_SET_SREGS/KVM_SET_SREGS2. In doing so, it
exposed a long-running bug in current KVM support for SEV-ES where the
kernel assumes that MSR_EFER_LMA will be set explicitly by the guest
kernel, in which case EFER write traps would result in KVM eventually
seeing MSR_EFER_LMA get set and recording it in such a way that it would
be subsequently visible when accessing it via KVM_GET_SREGS/etc.
However, guest kernels currently rely on MSR_EFER_LMA getting set
automatically when MSR_EFER_LME is set and paging is enabled via
CR0_PG_MASK. As a result, the EFER write traps don't actually expose the
MSR_EFER_LMA bit, even though it is set internally, and when QEMU
subsequently tries to pass this EFER value back to KVM via
KVM_SET_SREGS* it will fail various sanity checks and return -EINVAL,
which is now considered fatal due to the aforementioned QEMU commit.
This can be addressed by inferring the MSR_EFER_LMA bit being set when
paging is enabled and MSR_EFER_LME is set, and synthesizing it to ensure
the expected bits are all present in subsequent handling on the host
side.
Ultimately, this handling will be implemented in the host kernel, but to
avoid breaking QEMU's SEV-ES support when using older host kernels, the
same handling can be done in QEMU just after fetching the register
values via KVM_GET_SREGS*. Implement that here.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Tom Lendacky <thomas.lendacky@amd.com>
Cc: Akihiko Odaki <akihiko.odaki@daynix.com>
Cc: Philippe Mathieu-Daudé <philmd@linaro.org>
Cc: Lara Lazier <laramglazier@gmail.com>
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
Cc: Maxim Levitsky <mlevitsk@redhat.com>
Cc: <kvm@vger.kernel.org>
Fixes: 7191f24c7f ("accel/kvm/kvm-all: Handle register access errors")
Signed-off-by: Michael Roth <michael.roth@amd.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-ID: <20231206155821.1194551-1-michael.roth@amd.com>
Zhenzhong Duan