Commit 7c8e606b64 attempted to fix
the specification of the ramfb property for vfio-pci devices, but it
failed when ramfb is explicitly set to 'off'. This is because only the
'vfio-pci-nohotplug' device supports the 'ramfb' property. Since we use
the base 'vfio-pci' device unless ramfb is enabled, attempting to set
the 'ramfb' parameter to 'off' this will result in an error like the
following:
error: internal error: QEMU unexpectedly closed the monitor
(vm='rhel'): 2024-06-06T04:43:22.896795Z qemu-kvm: -device
{"driver":"vfio-pci","host":"0000:b1:00.4","id":"hostdev0","display":"on
","ramfb":false,"bus":"pci.7","addr":"0x0"}: Property 'vfio-pci.ramfb'
not found.
This also more closely matches what is done for mdev devices.
Resolves: https://issues.redhat.com/browse/RHEL-28808
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
There was no test for this and we mistakenly used 'B' rather than 'T'
when constructing the json value for this parameter. Thus, a value of
'off' was VIR_TRISTATE_SWITCH_OFF=2, which was translated to a boolean
value of 'true'.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
We already allow the user to specify display="on" and ramfb="on" for
mdev host devices. But newer GPU models will no longer use the mdev
framework, so we should enable this same functionality for other
non-mdev passthrough PCI devices.
Resolves: https://issues.redhat.com/browse/RHEL-28808
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Previously we were only starting or stopping nbdkit when the guest was
started or stopped or when hotplugging/unplugging a disk. But when doing
block operations, the disk backing store sources can also be be added or
removed independently of the disk device. When this happens the nbdkit
backend was not being handled properly. For example, when doing a
blockcopy from a nbdkit-backed disk to a new disk and pivoting to that
new location, the nbdkit process did not get cleaned up properly. Add
some functionality to qemuDomainStorageSourceAccessModify() to handle
this scenario.
Since we're now starting nbdkit from the ChainAccessAllow/Revoke()
functions, we no longer need to explicitly start nbdkit in hotplug code
paths because the hotplug functions already call these allow/revoke
functions and will start/stop nbdkit if necessary.
Add a check to qemuNbdkitProcessStart() to report an error if we
are trying to start nbdkit for a disk source that already has a running
nbdkit process. This shouldn't happen, and if it does it indicates an
error in another part of our code.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
When starting nbdkit processes for the backing store of a disk, we were
returning an error if any backing store failed, but we were not cleaning
up processes that succeeded higher in the chain. Make sure that if we
return a failure status from qemuNbdkitStartStorageSource() that we roll
back any processes that had been started.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
This will allow us to start or stop nbdkit for just a single disk source
or for every source in the backing chain. This will be used in following
patches.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Previously this test only tested the generated nbdkit command for the
top level disk source. Update it to test the generated commmands for all
sources in the chain.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
These values are currently unsupported for ssh disks, and in fact aren't
even parsed for ssh disks. So while this didn't result in any test
errors, we can remove them from the test input files.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
When trying to start nbdkit-backed disks in backing chains, we were
accidentally always checking the private data of the top of the chain
instead of using the loop variable.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Currently when we build with nbdkit support, libvirt will always try to
use nbdkit to access remote disk sources when it is available. But
without an up-to-date selinux policy allowing this, it will fail.
because the required selinux policies are not yet widely available, we
have disabled nbdkit support on rpm builds for all distributions before
Fedora 40.
Unfortunately, this makes it more difficult to test nbdkit support.
After someone updates to the necessary selinux policies, they would also
need to rebuild libvirt to enable nbdkit support. By introducing a
configure option (nbdkit_config_default), we can build packages with
nbdkit support but have it disabled by default.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Suggested-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
DO_TEST_CPUID(arch, host, json) is a multipart test. It consists of the
following tests:
- cpuTestHostCPUID()
- cpuTestGuestCPUID(with JSON_* flag)
- cpuTestCPUIDSignature()
- DO_TEST_JSON():
- if json==JSON_MODELS:
- cpuTestGuestCPUID(without JSON_* flag)
- cpuTestJSONCPUID()
- cputestJSONSignature()
Notice that for tests with json==JSON_MODELS, cpuTestGuestCPUID() is
actually called twice but with different arguments. The first one passes
JSON_MODELS to the test function, while the second one passes 0.
The main difference in behavior when calling cpuTestGuestCPUID() with or
without the flag is that in the first case, it parses the captured qemu
output from $ARCH-cpuid-$CPU.json. It extracts the cpu model list from
that JSON, and uses that to filter out possible cpu models to match.
In other words, it tries to match the cpu to a model that was supported
by the qemu version that was used to generate this JSON file. When it
finds a match, it generates a cpu definition and compares the xml form
of that definition with the file $ARCH-cpuid-$CPU-guest.xml.
When called without the JSON_MODELS flag, it simply attempts to match it
against the full libvirt cpu map and doesn't attempt to filter out any
matches based on the JSON qemu cpu model list. After it finds a match,
it generates an xml definition for the cpu and compares it to the same
file listed above. So if these two invocations disagree on the cpu match
(e.g. because libvirt has added a cpu model to its cpu map that matches
better than one that was supported by the version of qemu that generated
the JSON file) the test will fail.
This duplicate call to cpuTestGuestCPUID() was originally added in
commit 49c945a6f5. The original
justification for that commit was to fix test failures when the Qemu
driver was disabled. But since DO_TEST_JSON() is #defined empty when
qemu is disabled, this particular invocation would not even be executed
in this scenario, so it doesn't seem relevant.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
virProcessKillPainfullyDelay() currently almost always returns 1 or -1,
even though the documentation indicates that it should return 0 if the
process was terminated gracefully. But the computation of the return
code is faulty and the only case where it currently returns 0 is when it
is called with the pid of a process that does not exist.
Since no callers ever even distinguish between the 0 and 1 response
codes, simply get rid of the distinction and return 0 for both cases.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Commit 93af79fb removed a cleanup label in favor of returning error
values directly in certain cases. But the final return value was changed
from -1 to 0. If we get to the end of the function, that means that
we've waited for the process to exit but it still exists. So we should
return -1. The error message was still being set correctly, but we were
returning a success status (0).
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Require libnbd-devel when building the qemu driver, recommend nbdkit
packages.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
It's not possible to use password-protected ssh keys directly with
libvirt because libvirt doesn't have any way to prompt a user for the
password. To accomodate password-protected key files, an administrator
can add these keys to an ssh agent and then configure the domain with
the path to the ssh-agent socket.
Note that this requires an administrator or management app to
configure the ssh-agent with an appropriate socket path and add the
necessary keys to it. In addition, it does not currently work with
selinux enabled. The ssh-agent socket would need a label that libvirt
would be allowed to access rather than unconfined_t.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Add the ability to specify a path to a ssh-agent socket in order to use
the ssh-agent to authenticate to remote ssh disks. Example
configuration:
<disk type='network'>
</source protocol='ssh' ...>
<identity username='myusername' agentsock='/path/to/socket'/>
...
</source>
...
</disk>
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
For ssh disks that are served by nbdkit, we can support logging in with
an ssh key file. Pass the path to the configured key file and the
username to the nbdkit process.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Authenticating via key file to an ssh server is often preferable to
logging in via password. In order to support this functionality add a
new <identity> xml element for ssh disks that allows the user to specify
a keyfile and username. Example configuration:
<disk type='network'>
<source protocol='ssh' ...>
<identity keyfile='/path/to/id_rsa' username='myusername'/>
...
</source>
...
</disk>
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
For ssh disks that are served by nbdkit, use the configured value for
knownHosts and pass it to the nbdkit process.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
In order to make ssh disks usable, we need to be able to validate a
remote host. To do this, add a <knownHosts> xml element for ssh disks to
allow the user to specify a location for a file that contains known host
keys. Implementation to follow.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
For ssh disks that are served by nbdkit, lookup the password from the
configured secret and securely pass it to the nbdkit process using fd
passing.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Right now, ssh network disks are not usable. There is some basic support
in libvirt that is meant to support disk chains that have backing disks
located at ssh urls, but there is no real way for a user to configure a
ssh-based disk. This commit allows users to configure an ssh disk with
password authentication. Implementation will follow.
<disk type='network'>
<source protocol='ssh' ...>
<auth username='myusername'>
<secret type='iscsi' usage='secretname'/>
</auth>
</disk>
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
When using nbdkit to serve a network disk source, the nbdkit process
will start and wait for an nbd connection before actually attempting to
connect to the (remote) disk location. Because of this, nbdkit will not
report an error until after qemu is launched and tries to read from the
disk. This results in a fairly user-unfriendly error saying that qemu
was unable to start because "Requested export not available".
Ideally we'd like to be able to tell the user *why* the export is not
available, but this sort of information is only available to nbdkit, not
qemu. It could be because the url was incorrect, or because of an
authentication failure, or one of many other possibilities.
To make this friendlier for users and easier to detect
misconfigurations, try to connect to nbdkit immediately after starting
nbdkit and before we try to start qemu. This requires adding a
dependency on libnbd. If an error occurs when connecting to nbdkit, read
back from the nbdkit error log and provide that information in the error
report from qemuNbdkitProcessStart().
User-visible change demonstrated below:
Previous error:
$ virsh start nbdkit-test
2023-01-18 19:47:45.778+0000: 30895: error : virNetClientProgramDispatchError:172 : internal
error: process exited while connecting to monitor: 2023-01-18T19:47:45.704658Z
qemu-system-x86_64: -blockdev {"driver":"nbd","server":{"type":"unix",
"path":"/var/lib/libvirt/qemu/domain-1-nbdkit-test/nbdkit-libvirt-1-storage.socket"},
"node-name":"libvirt-1-storage","auto-read-only":true,"discard":"unmap"}: Requested export not
available
error: Failed to start domain 'nbdkit-test'
error: internal error: process exited while connecting to monitor: 2023-01-18T19:47:45.704658Z
qemu-system-x86_64: -blockdev {"driver":"nbd","server":{"type":"unix",
"path":"/var/lib/libvirt/qemu/domain-1-nbdkit-test/nbdkit-libvirt-1-storage.socket"},
"node-name":"libvirt-1-storage","auto-read-only":true,"discard":"unmap"}: Requested export not
available
After this change:
$ virsh start nbdkit-test
2023-01-18 19:44:36.242+0000: 30895: error : virNetClientProgramDispatchError:172 : internal
error: Failed to connect to nbdkit for 'http://localhost:8888/nonexistent.iso': nbdkit: curl[1]:
error: problem doing HEAD request to fetch size of URL [http://localhost:8888/nonexistent.iso]:
HTTP response code said error: The requested URL returned error: 404
error: Failed to start domain 'nbdkit-test'
error: internal error: Failed to connect to nbdkit for 'http://localhost:8888/nonexistent.iso]:
error: problem doing HEAD request to fetch size of URL [http://localhost:8888/nonexistent.iso]:
HTTP response code said error: The requested URL returned error: 404
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Adds the ability to monitor the nbdkit process so that we can take
action in case the child exits unexpectedly.
When the nbdkit process exits, we pause the vm, restart nbdkit, and then
resume the vm. This allows the vm to continue working in the event of a
nbdkit failure.
Eventually we may want to generalize this functionality since we may
need something similar for e.g. qemu-storage-daemon, etc.
The process is monitored with the pidfd_open() syscall if it exists
(since linux 5.3). Otherwise it resorts to checking whether the process
is alive once a second. The one-second time period was chosen somewhat
arbitrarily.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Since the restart handler will trigger at an arbitrary time (when the
nbdkit process crashes, for instance), it's difficult to provide
feedback to the user if the restart is unsuccessful. Rather than just
relying on a warning in the log, taint the domain so that there will be
a slightly more user-visible notification.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
We were testing the arguments that were being passed to qemu when a disk
was being served by nbdkit, but the arguments used to start nbdkit
itself were not testable. This adds a test to ensure that we're invoking
nbdkit correctly for various disk source definitions.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Add a private function to peek at the list of send buffers in virCommand
so that it is testable
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
For virStorageSource objects that contain an nbdkitProcess, start that
nbdkit process to serve that network drive and then pass the nbdkit
socket to qemu rather than sending the network url to qemu directly.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Rather than passing passwords and cookies (which could contain
passwords) to nbdkit via commandline arguments, use the alternate format
that nbdkit supports where we can specify a file descriptor which nbdkit
will read to get the password or cookies.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
All users of virCommandSetSendBuffer() are using it to send sensitive
data to a child process. So, since these buffers contain sensitive
information, clear it with virSecureErase().
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Add xml to the private data for a disk source to represent the nbdkit
process so that the state can be re-created if the libvirt daemon is
restarted. Format:
<nbdkit>
<pidfile>/path/to/nbdkit.pid</pidfile>
<socketfile>/path/to/nbdkit.socket</socketfile>
</nbdkit>
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
This prepares encryption secrets and authentication secrets. When we add
nbdkit-backed network storage sources, we will not need to send
authentication secrets to qemu, since they will be sent to nbdkit
instead. So split this into two different functions.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Add new DO_TEST_CAPS_LATEST_NBDKIT macro to test xml2argv for various
nbdkit capability scenarios.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
log stderr and stdout from nbdkit into its own log so that
nbdkit-related issues can be debugged more easily.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
This code can be used by the nbdkit implementation for reading back
filtered log data for error reporting. Move it to qemuLogContext so that
it can be shared. Renamed to qemuLogContextReadFiltered().
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
This will allow us to use it for nbdkit logging in upcoming commits.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Allow to specify a basename for the log file so that
qemuDomainLogContextNew() can be used to create log contexts for
secondary loggers.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Add some helper functions to build a virCommand object and run the
nbdkit process for a given virStorageSource.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Rather than hard-coding the nbdkit module directory, query the nbdkit
binary for the location to these directories. nbdkit provides a
--dump-config optiont that outputs this information and can be easily
parsed. We can also get the version from this output rather than
executing `nbdkit --version` separately.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
An object for storing information about a nbdkit process that is serving
a specific virStorageSource. At the moment, this information is just
stored in the private data of virStorageSource and not used at all.
Future commits will use this data to actually start a nbdkit process.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Add the virFileCache implementation for nbdkit capabilities to the qemu
driver. This allows us to determine whether nbdkit is installed and
which plugins are supported. it also has persistent caching and the
capabilities are re-queried whenever something changes.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Implement the loadFile and saveFile virFileCacheHandlers callbacks so
that nbdkit capabilities are cached perstistently across daemon
restarts. The format and implementation is modeled on the qemu
capabilities, but simplified slightly.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Preparatory step for caching nbdkit capabilities. This patch implements
the newData and isValid virFileCacheHandlers callback functions.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Since the libvirt documentation suggests to prefer GObject over
virObject, and since virObject is a GObject, change virFileCache to
allow GObjects as data.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
In order to add caching of the nbdkit capabilities, we will need to
compare against file modification times, etc. So look up this
information when creating the nbdkit caps.
Add a nbdkit_moddir build option to allow the builder to specify the
location to look for nbdkit plugins and filters.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
In future commits, we will optionally use nbdkit to serve some remote
disk sources. This patch queries to see whether nbdkit is installed on
the host and queries it for capabilities. The data will be used in later
commits.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
There was support in the code for parsing protocol='ssh' on network disk
sources, but it was not present in the xml schema. Add this to the
schema.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
vDPA block devices will also need the same consideration for memlock
limits as other vdpa devices, so consider these devices when calculating
memlock limits.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
qemuInterfaceVDPAConnect() was a helper function for connecting to the
vdpa device file. But in order to support other vdpa devices besides
network interfaces (e.g. vdpa block devices) make this function a bit
more generic.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Check whether the qemu binary supports the vdpa block driver. We can't
rely simply on the existence of the virtio-blk-vhost-vdpa block driver
since the first releases of qemu didn't support fd-passing for this
driver. So we have to check for the 'fdset' feature on the driver
object. This feature will be present in the qemu 8.1.0 release and was
merged to qemu in commit 98b126f5.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
When running libvirt from the build directory with the 'run' script, it
will run as unconfined_t. This can result in unexpected behavior when
selinux is enforcing due to the fact that the selinux policies are
written assuming that libvirt is running with the
system_u:system_r:virtd_t context. This patch adds a new --selinux
option to the run script. When this option is specified, it will launch
the specified binary using the 'runcon' utility to set its selinux
context to the one mentioned above. Since this may require root
privileges, setting the selinux context is not the default behavior and
must be enabled with the command line switch.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
Since commit 44a0f2f0, we now query mdevctl for transient (active) mdevs
in order to gather attributes for the mdev. Unfortunately, this commit
introduced a regression because nodeDeviceUpdateMediatedDevice() assumed
that all mdevs returned from mdevctl were actually persistent mdevs but
we were using it to update transient mdevs. Refactor the function so
that we can use it to update both persistent and transient mdevs.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
The virtio-gpu 'blob' support was insufficiently validated. Qemu
requires a memfd memory backing in order to use udmabuf and enable blob
support. Example error:
$ virsh start rhel9
error: Failed to start domain 'rhel9'
error: internal error: qemu unexpectedly closed the monitor: 2023-07-18T02:33:57.083178Z qemu-kvm: -device {"driver":"virtio-vga","id":"video0","max_outputs":1,"blob":true,"bus":"pcie.0","addr":"0x1"}: cannot enable blob resources without udmabuf
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Add some additional information about running clangd for LSP when clang
is not your normal compiler.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-By: Tim Small <tim@seoss.co.uk>
Rather than directly executing mdevctl from the udev event thread when
we determine that we need to re-query, schedule the mdevctl thread to
run. This also helps to coalesce multiple back-to-back updates into a
single one when there are multiple updates in a row or at startup when a
host has a very large number of mdevs.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
Factor out a new scheduleMdevctlUpdate() function so that we can re-use
it from other places. Now that other events can make it necessary to
re-query mdevctl for mdev updates, this function will be useful for
coalescing multiple updates in quick succession into a single mdevctl
query.
Also rename a couple functions. The names weren't very descriptive of
their behavior. For example, the old scheduleMdevctlHandler() function
didn't actually schedule anything, it just started a thread. So rename
it to free up the 'schedule' name for the above refactored function.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Boris Fiuczynski <fiuczy@linux.ibm.com>
This can improve performance for some guests since it reduces copying of
display data between host and guest. Requires udmabuf on the host.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Capability to determine whether this qemu supports the 'blob' option for
virtio-gpu.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Rather than storing the video type as an integer, use the proper enum
type within the struct.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
In virDomainVideoModelDefParseXML(), use the virXMLProp* functions
rather than reimplementing them with virXPath* functions.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
This function allows you to specify a default value to return if the
property is not found rather than always setting *result to 0.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Factor out a separate function to parse out the <model> element for
video devices.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Ensure that new virDomainVideoDef objects have their 'type' set to
VIR_DOMAIN_VIDEO_TYPE_DEFAULT and remove places that this value is set
after construction. Since virDomainVideoDefNew() uses g_new0() allocate
the instance, all fields are initialized to 0.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
When there is no vIOMMU, vfio devices don't need to lock the entire guest
memory per-device, but they still need to lock the entire guest memory to
share between all vfio devices. This memory accounting is not shared
with vDPA devices, so it should be added to the memlock limit separately.
Commit 8d5704e2 added support for multiple vfio/vdpa devices but
calculated the limits incorrectly when there were both vdpa and vfio
devices and no vIOMMU. In this case, the memory lock limit was not
increased separately for the vfio devices.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2143838
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
Openstack developers reported that newly-created mdevs were not
recognized by libvirt until after a libvirt daemon restart. The source
of the problem appears to be that when libvirt gets the udev 'add'
event, the sysfs tree for that device might not be ready and so libvirt
waits 100ms for it to appear (max 100 waits of 1ms each). But in the
OpenStack environment, the sysfs tree for new mediated devices was
taking closer to 250ms to appear and therefore libvirt gave up waiting
and didn't add these new devices to its list of nodedevs.
By changing the wait time to 1 second (max 100 waits of 10ms each), this
should provide enough time to enable these deployments to recognize
newly-created mediated devices, but it shouldn't increase the delay for
more traditional deployments too much.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2109450
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
When multiple VFIO or VDPA devices are assigned to a guest, the guest
can fail to start because the guest fails to map enough memory. For
example, the case mentioned in
https://bugzilla.redhat.com/show_bug.cgi?id=2111317 results in this
failure:
2021-08-05T09:51:47.692578Z qemu-kvm: failed to write, fd=31, errno=14 (Bad address)
2021-08-05T09:51:47.692590Z qemu-kvm: vhost vdpa map fail!
2021-08-05T09:51:47.692594Z qemu-kvm: vhost-vdpa: DMA mapping failed, unable to continue
The current memlock limit calculation does not work for scenarios where
there are multiple such devices assigned to a guest. The root causes are
a little bit different between VFIO and VDPA devices.
For VFIO devices, the issue only occurs when a vIOMMU is present. In
this scenario, each vfio device is assigned a separate AddressSpace
fully mapping guest RAM. When there is no vIOMMU, the devices are all
within the same AddressSpace so no additional memory limit is needed.
For VDPA devices, each device requires the full memory to be mapped
regardless of whether there is a vIOMMU or not.
In order to enable these scenarios, we need to multiply memlock limit
by the number of VDPA devices plus the number of VFIO devices for guests
with a vIOMMU. This has the potential for pushing the memlock limit
above the host physical memory and negating any protection that these
locked memory limits are providing, but there is no other short-term
solution.
In the future, there should be have a revised userspace iommu interface
(iommufd) that the VFIO and VDPA backends can make use of. This will be
able to share locked memory limits between both vfio and vdpa use cases
and address spaces and then we can disable these short term hacks. But
this is still in development upstream.
Resolves: https://bugzilla.redhat.com/2111317
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
It would be nice to be able to test the mediated device capabilities
without having physical hardware which supports it. The 'mtty' kernel
module presents a virtual parent device which is capable of creating
'fake' mediated devices, and as such it would be useful for testing.
However, the 'mtty' device is not part of an existing device subsystem
(e.g. PCI, etc), so libvirt ignores it and it does not get added to the
node device list. And because it does not get added to the node device
list, it cannot be used to create child mdevs using `virsh
nodedev-create`.
There is already a node device type capability
VIR_NODE_DEV_CAP_MDEV_TYPES that indicates whether a device supports
creating child mediated devices, but libvirt assumes that this is a
nested capability (in other words, it assumes that the primary
capability of a device is something like PCI). If we allow this
MDEV_TYPES capability to be a primary device capability, then we can
support virtual devices like 'mtty' as a parent for mediated devices.
See https://bugzilla.redhat.com/show_bug.cgi?id=2107031
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
since qemu 6.0, if migration is blocked for some reason, 'query-migrate'
will return an array of error strings describing the migration blockers.
This can be used to check whether there are any devices blocking
migration, etc.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
Reviewed-by: Laine Stump <laine@redhat.com>
This was not causing any problems because all cases below were empty,
but in order to avoid future misbehavior, add a break to this case.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
These wrapper functions were used to adapt the virObjectUnref() function
signature for different callbacks. But in commit 0d184072, the
virObjectUnref() function was changed to return a void instead of a
bool, so these adapters are no longer necessary.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
Add <interleave> to allow the subproperties to be specified in any
order.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Document the format of the 'readahead' and 'timeout' XML elements more
accurately.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
The last use of this function was removed in commit 6d161bcc, so the
function is no longer used except as an internal implementation for
virCommandPassFD().
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Add the ability to configure a qemu-vdagent in guest domains. This
device is similar to the spice vdagent channel except that qemu handles
the spice-vdagent protocol messages itself rather than routing them over
a spice protocol channel.
The qemu-vdagent device has two notable configuration options which
determine whether qemu will handle particular vdagent features:
'clipboard' and 'mouse'.
The 'clipboard' option allows qemu to synchronize its internal clipboard
manager with the guest clipboard, which enables client<->guest clipboard
synchronization for non-spice guests such as vnc.
The 'mouse' option allows absolute mouse positioning to be sent over the
vdagent channel rather than using a usb or virtio tablet device.
Sample configuration:
<channel type='qemu-vdagent'>
<target type='virtio' name='com.redhat.spice.0'/>
<source>
<clipboard copypaste='yes'/>
<mouse mode='client'/>
</source>
</channel>
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
This enumeration will be useful for vnc with the upcoming qemu-vdagent
device so make the name more generic.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Detect whether qemu supports the qemu-vdagent character device. This
enables support for copy/paste with VNC graphics.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Validate the domain configuration to ensure that if there are more than
one vgpu assigned to a domain, only one of them has 'ramfb' enabled.
This was never a supported configuration. QEMU failed confusingly when
attempting to start a domain with this configuration. This change
attempts to provide better information about the error.
https://bugzilla.redhat.com/show_bug.cgi?id=2079760
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
While commit a5e659f0 removed the restriction against multiple queues
for the vdpa net device, there were some missing pieces. Configuring a
device statically and then starting the domain worked as expected, but
hotplugging a device didn't have the expected multiqueue support
enabled. Add the missing bits.
Consider the following device xml:
<interface type="vdpa">
<mac address="00:11:22:33:44:03" />
<source dev="/dev/vhost-vdpa-0" />
<model type="virtio" />
<driver queues='2' />
</interface>
Without this patch, hotplugging the above XML description resulted in
the following:
{"execute":"netdev_add","arguments":{"type":"vhost-vdpa","vhostdev":"/dev/fdset/0","id":"hostnet1"},"id":"libvirt-392"}
{"execute":"device_add","arguments":{"driver":"virtio-net-pci","netdev":"hostnet1","id":"net1","mac":"00:11:22:33:44:03","bus":"pci.5","addr":"0x0"},"id":"libvirt-393"}
With the patch, hotplugging results in the following:
{"execute":"netdev_add","arguments":{"type":"vhost-vdpa","vhostdev":"/dev/fdset/0","queues":2,"id":"hostnet1"},"id":"libvirt-392"}
{"execute":"device_add","arguments":{"driver":"virtio-net-pci","mq":true,"vectors":6,"netdev":"hostnet1","id":"net1","mac":"00:11:22:33:44:03","bus":"pci.5","addr":"0x0"},"id":"libvirt-393"}
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2024406
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Presumably the result of a copy/paste mistake, the the argument for the
`nodedev-start` command was described as a 'network' rather than a
'device'.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Add up-to-date information about creating and defining mediated devices
in libvirt.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Bring the documentation for nodedev-list up to date with the latest
code, especially documenting the --active and -all options.
Also add documentation for the nodedev-define, nodedev-undefine, and
nodedev-start commands.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
This is currently the only way to view the 'autostart' property for a
node device in virsh.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Implement these new API functions in the nodedev driver.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
These two public APIs are implemented for almost all other objects that
have a concept of persistent definition and activatability. Now that we
have node devices (mdevs) that can be defined and inactive, it will be
useful to query the persistent/active state of node devices as well.
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Jonathon Jongsma