The base vfio has not much functionality but to provide a custom
container by opening this path.
See
https://www.kernel.org/doc/Documentation/vfio.txt for more.
Systems with static hostdevs will get /dev/vfio/vfio by virt-aa-
hotplug
right from the beginning. But if the guest initially had no hostdev
at
all it will run into the following deny before the security module
labelling callbacks will make the actual vfio device (like
/dev/vfio/93)
known.
Access by qemu is "wr" even thou in theory it could maybe be "r":
[ 2652.756712] audit: type=1400 audit(1491303691.719:25):
apparmor="DENIED" operation="open"
profile="libvirt-17a61b87-5132-497c-b928-421ac2ee0c8a"
name="/dev/vfio/vfio" pid=8486 comm="qemu-system-x86"
requested_mask="wr" denied_mask="wr" fsuid=64055 ouid=0
Bug-Ubuntu:
https://bugs.launchpad.net/bugs/1678322
Bug-Ubuntu:
https://bugs.launchpad.net/bugs/1775777
Signed-off-by: Christian Ehrhardt <christian.ehrhardt(a)canonical.com>
Signed-off-by: Stefan Bader <stefan.bader(a)canonical.com>
---
examples/apparmor/libvirt-qemu | 3 +++
1 file changed, 3 insertions(+)
diff --git a/examples/apparmor/libvirt-qemu
b/examples/apparmor/libvirt-qemu
index 2c47652250..874aca2092 100644
--- a/examples/apparmor/libvirt-qemu
+++ b/examples/apparmor/libvirt-qemu
@@ -193,6 +193,9 @@
deny /dev/shm/lttng-ust-wait-* r,
deny /run/shm/lttng-ust-wait-* r,
+ # for vfio hotplug on systems without static vfio (LP: #1775777)
+ /dev/vfio/vfio rw,
+
Makes sense. If the guest doesn't start with vfio then libvirtd is
going to have to give it to the guest and so libvirtd would need this
access. +1 to apply.
--
Jamie Strandboge |