This lets the app expose the virtual SCSI or IDE disks as solid state
devices by setting a rate of '1', or rotational media by setting a
rate between 1025 and 65534.
https://bugzilla.redhat.com/show_bug.cgi?id=1498955
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Previously, we accepted empty bridge name, because some old versions of
VMWare Workstation did not put it into the config. But this doesn't make
much sense - to have an interface type bridge with no name. We
circumvented this problem by generating an empty name but that is
equally wrong.
Therefore, fill in missing bridge names (according to the documentation
[1] the default bridge name is VMnet0) and error out if bridge name is
missing.
This partially reverts f246cdb5ac
1: https://docs.vmware.com/en/VMware-Workstation-Player-for-Linux/16.0/com.vmware.player.linux.using.doc/GUID-BAFA66C3-81F0-4FCA-84C4-D9F7D258A60A.html
Signed-off-by: Kristina Hanicova <khanicov@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Signed-off-by: Ján Tomko <jtomko@redhat.com>
The
<os firmware='efi'>
<firmware type='efi'>
<feature enabled='no' name='enrolled-keys'/>
</firmware>
</os>
repeats the firmware attribute twice. This has no functional benefit, as
evidenced by fact that we use a single struct field to store both
attributes, while needlessly introducing an error scenario. The XML can
just be simplified to:
<os firmware='efi'>
<firmware>
<feature enabled='no' name='enrolled-keys'/>
</firmware>
</os>
which also means that we don't need to emit the empty element
<firmware type='efi'/> for all existing configs too.
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
This API talks to QEMU and changes its internal state. Therefore,
it should acquire QEMU_JOB_MODIFY instead of QEMU_JOB_QUERY.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Jiri Denemark <jdenemar@redhat.com>
This API interacts with the hypervisor and makes changes to its
behaviour, so must be protected by the write permission.
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Since the functions provided by libpciaccess are not thread-safe,
when the udev-event and nodedev-init threads of libvirt call the
pci_get_strings function provided by libpaciaccess at the same
time the following can happen:
nodedev-init thread:
nodeStateInitializeEnumerate ->
udevEnumerateDevices->
udevProcessDeviceListEntry ->
udevAddOneDevice ->
udevGetDeviceDetails->
udevProcessPCI ->
udevTranslatePCIIds ->
pci_get_strings -> (libpciaccess)
find_device_name ->
populate_vendor ->
d = realloc( vend->devices, (vend->num_devices + 1), * sizeof( struct pci_device_leaf ) );
vend->num_devices++;
udev-event thread:
udevEventHandleThread ->
udevHandleOneDevice ->
udevAddOneDevice->
udevGetDeviceDetails->
udevProcessPCI ->
udevTranslatePCIIds ->
pci_get_strings -> (libpciaccess)
find_device_name ->
populate_vendor ->
d = realloc( vend->devices, (vend->num_devices + 1), * sizeof( struct pci_device_leaf ) );
vend->num_devices++;
Signed-off-by: WangJian <wangjian161@huawei.com>
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Recently, a few commits back I've switched bunch of code to
g_steal_pointer() using coccinelle. Problem was that the semantic
patch used was slightly off:
@@
expression a, b;
@@
+ b = g_steal_pointer(&a);
- b = a;
... when != a
- a = NULL;
Problem is that, "... when != a" is supposed to jump over those
lines, which don't contain expression a. My idea was to replace
the following pattern too:
ptrX = ptrY;
if (something(ptrZ) < 0) goto error;
ptrY = NULL;
But what I missed is that the following pattern is also matched
and replaced:
ptrX = ptrY;
if (something(ptrX) < 0) goto error;
ptrY = NULL;
This is not necessarily correct - as demonstrated by our hotplug
code. The real problem is ambiguous memory ownership transfer
(functions which add device to domain def take ownership only on
success), but to not tackle the real issue let's revert those
parts.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
Turns out, the way that glib implements g_steal_pointer() is not
compatible with function callbacks. And that's what my recent
patch did in virNetSocketEventFree(). Revert that part.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Peter Krempa <pkrempa@redhat.com>
In one of my recent patches I've introduced new connection
feature VIR_DRV_FEATURE_NETWORK_UPDATE_HAS_CORRECT_ORDER.
However, I forgot to add corresponding case into a switch in
vzConnectSupportsFeature().
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
The order in which virNetworkUpdate() accepts @section and
@command arguments is not the same as in which it passes them
onto networkUpdate() callback. Until recently, it did not really
matter, because calling the API on client side meant arguments
were encoded in reversed order (compared to the public API), but
then on the server it was fixed again - because the server
decoded RPC (still swapped), called public API (still swapped)
and in turn called the network driver callback (with reversing
the order - so magically fixing the order).
Long story short, if the public API is called even number of
times those swaps cancel each other out. The problem is when the
API is called an odd numbed of times - which happens with split
daemons and the right URI. There's one call in the client (e.g.
virsh net-update), the other in a hypervisor daemon (say
virtqemud) which ends up calling the API in the virnetworkd.
The fix is obvious - fix the order in which arguments are passed
to the callback.
But, to maintain compatibility with older, yet unfixed, daemons
new connection feature is introduced. The feature is detected
just before calling the callback and allows client to pass
arguments in correct order (talking to fixed daemon) or in
reversed order (talking to older daemon).
Unfortunately, older client talking to newer daemon can't be
fixed. Let's hope that it's less frequent scenario.
Fixes: 574b9bc66b
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1870552
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
So far, it was not needed, but shortly a client will want to know
whether virNetworkUpdate() API is fixed or not. See next commits
for more info.
Side note, this driver's implementation is called only when using
sub-driver's connection, i.e. "network:///system". For any other
URI the corresponding hypervisor's driver callback is called.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
Just like VFIO devices, vDPA devices may need to have all guest memory
pages locked/pinned in order to operate properly. In the case of VFIO
devices (including mdev and NVME, which also use VFIO) libvirt
automatically increases the locked memory limit when one of those
devices is present. This patch modifies that code to also increase the
limit if there are any vDPA devices present.
Resolves: https://bugzilla.redhat.com/1939776
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
This function is a specialized version of
qemuDomainGetMemLockLimitBytes() for PPC64. Simplifying it in the same
manner as the previous patch has the nice side effect of accounting
for the possibility of an mdev device
(I don't know if mdev devices are supported on PPC, but even if not
then a) the additional check for mdev devices gained by using
qemuDomainNeedsVFIO() in place of open coding will be an effective
NOP, and b) if mdev devices are supported on PPC64 in the future, this
function will be prepared for it).
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
This function goes through a loop checking if each hostdev is a VFIO
or mdev device, and then later it calls virDomainDefHasNVMEDisk(). The
function qemuDomainNeedsVFIO() does exactly the same thing, so let's
just call that instead.
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
This function returns true if the domain has any interfaces that are
type='vdpa'.
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Generated by the following spatch:
@@
expression a, b;
@@
+ b = g_steal_pointer(&a);
- b = a;
... when != a
- a = NULL;
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
The ESX implementation of virConnectListAllDomains() follows
pretty much implementations in other drivers: it has local array
of virDomainPtr-s which (if requested by caller) is filled by
actual domains or not (if the caller is interested only in the
count of domains).
Anyway, in case of the former, the passed @domains argument is
set to the local array, which is then set to NULL to prevent it
from freeing under cleanup label. Pretty standard pattern.
Except, the local array is set to NULL always. Even if the local
array is not stolen. Fortunately, this doesn't lead to a memory
leak, because if caller is not interested in the array, none is
allocated. But it doesn't set good example and also breaks my
spatch rules.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
Some SRIOV PFs don't have a netdev associated with them (the spec
apparently doesn't require it). In most cases when libvirt is dealing
with an SRIOV VF, that VF must have a PF, and the PF *must* have an
associated netdev (the only way to set the MAC address of a VF is by
sending a netlink message to the netdev of that VF's PF). But there
are times when we don't need for the PF to have a netdev; in
particular, when we're just getting the Switchdev Features for a VF,
we don't need the PF netdev - the netdev of the VF (apparently) works
just as well.
Commit 6452e2f5 (libvirt 5.1.0) *kind of* made libvirt work around PFs
with no netdevs in this case - if virNetDevGetPhysicalFunction
returned an error when setting up to retrieve Switchdev feature info,
it would ignore the error, and then check if the PF netdev name was
NULL and, if so it would reset the error object and continue on rather
than returning early with a failure. The problem is that by the time
this special handling occured, the error message about missing netdev
had already been logged, which was harmless to proper operation, but
confused the user.
Fortunately there are only 2 users of virNetDevGetPhysicalFunction, so
it is easy to redefine it's API to state that a missing netdev name is
*not* an error - in that case it will still return success, but the
caller must be prepared for the PF netdev name to be NULL. After
making this change, we can modify the two callers to behave properly
with the new semantics (for one of the callers it *is* still an error,
so the error message is moved there, but for the other it is okay to
continue), and our spurious error messages are a thing of the past.
Resolves: https://bugzilla.redhat.com/1924616
Fixes: 6452e2f5e1
Signed-off-by: Laine Stump <laine@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
With this, incomplete XML without <frames/> for <rx/> in coalesce
won't raise error as before. It will leave the coalesce parameter
empty, thanks to passing it as a parameter and return an integer
to indicate error state - previously it returned pointer (or NULL
for both error and incomplete XML).
I also added a test case to test this functionality in the
qemuxml2xmltest.
The code went through some refactoring:
* change of a condition
* addition of a parameter
* change of order, that allowed removal of VIR_FREE
* removal of redundant labels and variables
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1535930
Signed-off-by: Kristina Hanicova <khanicov@redhat.com>
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
Commit ac87d3520a consolidated common cgroup code between the QEMU and
lxc drivers in domain_cgroup.c. In this process, in
virDomainCgroupSetupDomainBlkioParameters(), a call to
virCgroupGetBlkioWeight() went missing.
The result is that 'virsh blkiotune' is setting the blkio.weight for the
guest in the host cgroup, but not on the domain XML, because
virCgroupGetBlkioWeight() is also used to write the blkio.weight value
in the domain object.
Fix it by adding the virCgroupGetBlkioWeight() call in the
virDomainCgroupSetupDomainBlkioParameters() helper.
Fixes: ac87d3520a
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1941407
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Signed-off-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Previously we'd assign the default checkpoint bitmap names in
virDomainCheckpointAlignDisks. In cases when the checkpoint is redefined
without a domain XML virDomainCheckpointAlignDisks is not called.
Add an explicit call to virDomainCheckpointDefAssignBitmapNames to
restore functionality.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
When a checkpoint is redefined without providing the domain XML, we
might end up with a definition where the per-disk bitmap name is not
set. Trying to delete such checkpoint would lead to a crash.
Refuse such deletion.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1941600
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Base the detection on the presence of the 'secret' qom-type entry, which
isn't conditionally compiled in qemu.
All caps-based test now switch to using JSON for -object.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Skip the lossy conversion to legacy commandline arguments by using the
JSON props directly when -object is QAPIfied. This avoids issues with
conversion of bitmaps and also allows validation of the generated JSON
against the QMP schema in the tests.
Since the new approach is triggered by a qemu capability the code
from 'virQEMUBuildObjectCommandlineFromJSON' in util/virqemu.c was moved
to 'qemuBuildObjectCommandlineFromJSON' in qemu/qemu_command.c which has
the virQEMUCaps type.
Some functions needed to be modified to propagate qemuCaps.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Set 'objectAddNoWrap' when the capability is present.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
There's just one caller left. Since qemuBuildMemoryBackendProps is too
complex to be modified for now, just move the adding of 'id' and 'qom'
type directly into the function.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Construct the JSON object which is used for object-add without the
'props' wrapper and add the wrapper only in the monitor code.
This simplifies the JSON->commandline generator in the first place and
also prepares for upcoming qemu where 'props' will be removed.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Starting from qemu-6.0 the parameters of -object/object-add are formally
described by the QAPI schema. Additionally this changes the nesting of
the properties as the 'props' nested object will be flattened to the
parent.
We'll need to detect whether qemu switched to this new approach to
generate the objects with proper nesting and also allow testing.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
The firmwareFeatures member of virDomainOSDef struct is allocated
in virDomainDefParseBootFirmwareOptions() but never freed.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
The virDomainDefFree() function frees individual members of
virDomainDef struct. The function is already long enough, move
code that handles def->os member into a separate function.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
Our reallocation APIs already abort on OOM and thus can only return 0.
There's no need to force callers to check the result.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
virQEMUCapsGet checks for qemuCaps itself, no need to do it explicitly.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
In case an async job spans multiple APIs (e.g., incoming migration) the
API that started the job is recorded as the asyncOwnerAPI even though it
is no longer running and the owner thread is updated properly to the one
currently handling the job. Let's also update asyncOwnerAPI to make it
more obvious which is the current (or the most recent) API involved in
the job.
Signed-off-by: Jiri Denemark <jdenemar@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
Attempting to set the memlock limit might fail if we're running
in a containerized environment where CAP_SYS_RESOURCE is not
available, and if the limit is already high enough there's no
point in trying to raise it anyway.
https://bugzilla.redhat.com/show_bug.cgi?id=1916346
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Store the current memory locking limit and the desired one
separately, which will help with later changes.
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Now that we've implemented a fallback for the function that
obtains the information from /proc, there is no reason we would
get a failure unless there's something seriously wrong with the
environment we're running in, in which case we're better off
reporting the issue to the user rather than pretending
everything is fine.
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Calling prlimit() requires elevated privileges, specifically
CAP_SYS_RESOURCE, and getrlimit() only works for the current
process which is too limiting for our needs; /proc/$pid/limits,
on the other hand, can be read by any process, so implement
parsing that file as a fallback for when prlimit() fails.
This is useful in containerized environments.
Signed-off-by: Andrea Bolognani <abologna@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Implement "<os firmware='efi'>" support for bhyve driver.
As there are not really lot of options, try to find
"BHYVE_UEFI.fd" firmware which is installed by the
sysutils/uefi-edk2-bhyve FreeBSD port.
If not found, just use the first found firmware
in the firmwares directory (which is configurable via
config file).
Signed-off-by: Roman Bogorodskiy <bogorodskiy@gmail.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
When the backup job is terminated normally the security label is
restored by the blockjob finishing handler.
If the VM dies or is destroyed that wouldn't happen as the blockjob
handler wouldn't be called.
Restore the security label on disk store where we remember that the job
was running at the point when 'qemuBackupJobTerminate' was called.
Not resetting the security label means that we also leak the xattr
attributes remembering the label which prevents any further use of the
file, which is a problem for block devices.
This also requires that the call to 'qemuBackupJobTerminate' from
'qemuProcessStop' happens only after 'vm->pid' was reset as otherwise
the security subdrivers attempt to enter the process namespace which
fails if the process isn't running any more.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1939082
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
qemuBackupBegin can take a full backup of the disks (excluding any
operations with bitmaps) without the need to wait for the
blockdev-reopen support in qemu.
Add a check that no checkpoint creation is required and the disk backup
mode isn't VIR_DOMAIN_BACKUP_DISK_BACKUP_MODE_INCREMENTAL.
Call to virDomainBackupAlignDisks is moved earlier as it initializes the
disk backup mode if not present in user config.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Upcoming commit will enable full backup support (incremental part
requires blockdev-reopen, which won't happen in qemu for at least
another release).
Add a capability that the 'blockdev-backup' job is supported by qemu
capped, but limited to when qemu supports QEMU_CAPS_BLOCKDEV.
We can also use it in the expression to enable
QEMU_CAPS_INCREMENTAL_BACKUP since it's a pre-requisite too.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
When the firmware auto-selection was introduced it always picked first
usable firmware based on the JSON descriptions on the host. It is
possible to add/remove/change the JSON files but it will always be for
the whole host.
This patch introduces support for configuring the auto-selection per VM
by adding users an option to limit what features they would like to have
available in the firmware.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
The original code used a lot of conditions and was not that obvious
when each XML bits are parsed.
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
With this, XML fails if non-virtio video devices have virtio
options. Previously it didn't raise error.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1922093
Signed-off-by: Kristina Hanicova <khanicov@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Move this function in order to use it in the next patch before
its previous declaration.
Signed-off-by: Kristina Hanicova <khanicov@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Currently, virDomainDeviceInfoParseXML() uses node->children
evaluation which is too verbose. Use XPath evaluation which is
nicer.
Signed-off-by: Kristina Hanicova <khanicov@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Function virDomainDeviceInfoParseXML() will need it soon, because it
will be doing XPath evaluation.
Signed-off-by: Kristina Hanicova <khanicov@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
If libvirtd is terminated before the node driver finishes
initialization, it can crash with a backtrace similar to the following:
Stack trace of thread 1922933:
#0 0x00007f8515178774 g_hash_table_find (libglib-2.0.so.0)
#1 0x00007f851593ea98 virHashSearch (libvirt.so.0)
#2 0x00007f8515a1dd83 virNodeDeviceObjListSearch (libvirt.so.0)
#3 0x00007f84cceb40a1 udevAddOneDevice (libvirt_driver_nodedev.so)
#4 0x00007f84cceb5fae nodeStateInitializeEnumerate (libvirt_driver_nodedev.so)
#5 0x00007f85159840cb virThreadHelper (libvirt.so.0)
#6 0x00007f8511c7d14a start_thread (libpthread.so.0)
#7 0x00007f851442bdb3 __clone (libc.so.6)
Stack trace of thread 1922863:
#0 0x00007f851442651d syscall (libc.so.6)
#1 0x00007f85159842d4 virThreadSelfID (libvirt.so.0)
#2 0x00007f851594e240 virLogFormatString (libvirt.so.0)
#3 0x00007f851596635d vir_object_finalize (libvirt.so.0)
#4 0x00007f8514efe8e9 g_object_unref (libgobject-2.0.so.0)
#5 0x00007f85159667f8 virObjectUnref (libvirt.so.0)
#6 0x00007f851517755f g_hash_table_remove_all_nodes.part.0 (libglib-2.0.so.0)
#7 0x00007f8515177e62 g_hash_table_unref (libglib-2.0.so.0)
#8 0x00007f851596637e vir_object_finalize (libvirt.so.0)
#9 0x00007f8514efe8e9 g_object_unref (libgobject-2.0.so.0)
#10 0x00007f85159667f8 virObjectUnref (libvirt.so.0)
#11 0x00007f84cceb2b42 nodeStateCleanup (libvirt_driver_nodedev.so)
#12 0x00007f8515b37950 virStateCleanup (libvirt.so.0)
#13 0x00005648085348e8 main (libvirtd)
#14 0x00007f8514352493 __libc_start_main (libc.so.6)
#15 0x00005648085350fe _start (libvirtd)
This is because the initial population of the device list is done in a
separate initialization thread. If we attempt to exit libvirtd before
this init thread has completed, we'll try to free the device list while
accessing it from the other thread. In order to guarantee that this
init thread is not accessing the device list when we're cleaning up the
nodedev driver, make it joinable and wait for it to finish before
proceding with the cleanup. This is similar to how we handle the udev
event handler thread.
The separate initialization thread was added in commit
9f0ae0b1.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1836865
Signed-off-by: Jonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Previously, if xml node passed to the virXMLNodeContentString()
was not of type XML_ELEMENT_NODE, @ret could have caused a memory
leak because xmlNodeGetContent() works for other types of nodes
as well.
Signed-off-by: Kristina Hanicova <khanicov@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Extend dirtyrate statistics for domGetStats to display the information
of a domain's memory dirty rate produced by domainStartDirtyRateCalc.
Signed-off-by: Hao Wang <wanghao232@huawei.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Introduce virDomainStartDirtyRateCalc API for start calculation of
a domain's memory dirty rate with a specified time.
Signed-off-by: Hao Wang <wanghao232@huawei.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
We don't need to go to the trouble of telling users about existance of
insecure SASL mechanisms only to then say that they shouldn't be used.
We should only tell people about the GSSAPI mechanism for TCP sockets.
For the SCRAM mechanism we should be telling people about the SHA256
variant only, and also warning that the password database stores the
passwords in clear text.
Reviewed-by: Erik Skultety <eskultet@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
If running libvirtd via systemd, it gets a 64 MB memlock limit, but if
running from the shell it will only get 64 KB on a Fedora 33 system.
The latter low limit causes any attempt to use BPF to fail and it is
not obvious why.
This improves the error message thus:
# virsh -c lxc:/// start sh
error: Failed to start domain 'sh'
error: internal error: guest failed to start: Failure in libvirt_lxc startup: failed to initialize device BPF map; locked memory limit for libvirtd probably needs to be raised: Operation not permitted
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The g_idle_add function adds a callback to the primary GMainContext.
To workaround the GSource unref bugs, we need to add our callbacks
to the GMainContext that is associated with the GSource being
unref'd. Thus code using the per-VM virEventThread must use its
private GMainContext.
Reviewed-by: Martin Kletzander <mkletzan@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Somehow, command argument was not printed into debug logs. It is
imperative that all arguments are logged.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
When connecting to the monitor, a timeout is calculated that is
bigger the more memory guest has (because QEMU has to allocate
and possibly zero out the memory and what not, empirically
deducted). However, when computing the timeout the @total_memory
mmember is accessed directly even though
virDomainDefGetMemoryTotal() should have been used.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
When switching to g_autoptr this was incorrectly changed from
'continue;' into 'return -1;' resulting into an error when user tries
to set vcpu_quota of running VM:
error: An error occurred, but the cause is unknown
Fixes: e4a8bbfaf2
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Reviewed-by: Erik Skultety <eskultet@redhat.com>
In short, virXXXPtr type is going away. With big bang. And to
help us rewrite the code with a sed script, it's better if each
variable is declared on its own line.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
The name is supposed to be virCapsGuestArchPtr not ..ptr.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
The use of virXXXPtr is going away soon, therefore use 'virXXX *'
instead.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
What we are using really is heap allocated structure rather than
stack allocated. And for that it's better to use g_autoptr() +
G_DEFINE_AUTOPTR_CLEANUP_FUNC() combo, as Glib documentation for
g_auto() reads:
This is meant to be used with stack-allocated structures and
non-pointer types. For the (more commonly used) pointer
version, see g_autoptr().
This will be even more visible, when virSysinfoDefPtr type is
gone. Stay tuned.
Fixes: cee3a900a0
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
The qemu shim spawns a separate thread in which the event loop is
ran. The virEventRunDefaultImpl() call is wrapped in a while()
loop, just like it should. There are few lines of code around
which try to ensure that domain is destroyed (when quitting) and
that the last round of event loop is ran after the
virDomainDestroy() call. Only after that the loop is quit from
and the thread quits.
However, if domain creation fails, there is no @dom to call
destroy over, the @quit flag is never set and while() never
exits. Set the flag regardless of @dom pointer.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1920337
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
The commandline generator for 'iothread' objects has a private
implementation of the properties. Convert it to JSON so that it can be
later validated.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
While the 'sev0' sev-guest object will never be hotplugged, but we want
to generate it through JSON so that we'll be able to validate all
parameters of '-object' against the QAPI schema once 'object-add' is
qapified in qemu.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
While the 'masterKey0' secret object will never be hotplugged we want to
generate it through JSON so that we'll be able to validate all
parameters of '-object' against the QAPI schema once 'object-add' is
qapified in qemu.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Commit 94e45d1042 broke exec-restart of virtlogd and virtlockd as the
code waiting for the daemon shutdown closed the daemons before
exec-restarting.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1912243
Fixes: 94e45d1042
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Recent changes which meant to fix daemon shutdown broke the exec-restart
capability of virtlogd and virtlockd, since the code actually closed all
the sockets and shut down all the internals.
Add virNetDaemonQuitExecRestart, which requests a shutdown of the
process, but keeps all the services open and registered since they are
preserved across the restart.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
This problem is reproducible only with secret driver. When
starting a domain via virt-qemu-run and both secret and
(nonexistent) root directory specified this is what happens:
1) virt-qemu-run opens "secret:///embed?root=$rootdir"
connection, which results in the secret driver initialization
(done in secretStateInitialize()). During this process, the
driver creates its own configDir (derived from $rootdir)
including those parents which don't exists yet. This is all
done with the mode S_IRWXU and thus results in the $rootdir
being created with very restrictive mode (specifically, +x is
missing for group and others).
2) now, virt-qemu-run opens "qemu:///embed?root=$rootdir" and
calls virDomainCreateXML(). This results in the master-key.aes
being written somewhere under the $rootdir and telling qemu
where to find it.
But because the secret driver created $rootdir with too
restrictive mode, qemu can't access the file (even though it
knows the full path) and fails to start.
It looks like the best solution is to pre-create the root
directory before opening any connection (letting any driver
initialize itself) and set its mode to something less
restrictive.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1859873
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
In theory, users might want to use a relative path as a root
directory for embed drivers. But in practice, nothing in driver
initialization (specifically QEMU driver since it's the only one
that supports embedding now), is prepared for that. Document and
enforce absolute paths.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1883725
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Andrea Bolognani <abologna@redhat.com>
'res->owners' is allocated to 'res->nOwners' elements, but unfortunately
'res->nOwners' doesn't contain the proper value until after the
allocation so 0 elements are allocated. The following loop which assumes
that the array has the right number of elements then accesses the
pointer out of bounds. The bug was also faithfully converted from
VIR_ALLOC_N to g_new0.
Fixes: 4a3d6ed5ee
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Recent refactor marked 'object' which is returned from the function as
autofree but forgot to use g_steal_pointer in the return statement to
prevent freeing it.
Fixes: 9a1651f64d
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Commit cb29e4e801 didn't take into account that the VM can be inactive
when it's destroyed. This means that the job would remain active also
when the VM became inactive.
To fix this properly:
1) Remove the bogus VM liveness check and early return
(reverts the aforementioned commit)
2) Conditionalize the stats assignment only when the stats object is
present
(properly fix the crash when VM dies when reconnecting)
3) end the asyncjob only when it was already set
(prevent corruption of priv->jobs_queued)
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1937598
Fixes: cb29e4e801
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
'qemuBackupJobTerminate' needs the API flags to see whether
VIR_DOMAIN_BACKUP_BEGIN_REUSE_EXTERNAL. Unfortunately when called via
qemuProcessReconnect()->qemuProcessStop() early (e.g. if the qemu
process died while we were reconnecting) the job is cleared temporarily
so that other APIs can be called. This would mean that we couldn't clean
up the files in some cases.
Save the 'apiFlags' inside the backup object and set it from the
'qemuDomainJobObj' 'apiFlags' member when reconnecting to a VM.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
g_variant_new_parsed uses '%t' for a uint64_t rather than printf-like
%llu. Additionally ensure that the passed value is a uint64_t since the
argument used is a 'unsigned int'.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1937287
Fixes: bf5f2ed09c
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Pavel Hrdina <phrdina@redhat.com>
The function is now unused and motivated users to write crazy parsers
which were hard to understand, had pointless error paths just to avoid
few memory allocations.
Remove the function as we're fine with g_strndup and virStrcpy.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Use g_strsplit to split the string and avoid use of stack'd strings.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
The problem is that g_get_host_name() caches the hostname in a
thread local variable. Therefore, it doesn't reflect any
subsequent hostname changes. While this might be acceptable for
logs where the hostname is printed exactly once when the libvirtd
starts up, it is not optimal for virGetHostnameImpl() which is
what our public virConnectGetHostname() API calls. If the
hostname at the moment of the first API invocation happens to
start with "localhost" or contains a dot, then no further
hostname changes will ever be reflected.
This reverts 26d9748ff1, partially.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
We already assume that 'retr_passphrase.result' is a string, thus we can
use virStrcpy instead.
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
