Since v5.6.0-48-g270583ed we try to cache domain capabilities,
i.e. store filled virDomainCaps in a hash table in virQEMUCaps
for future use. However, there's a race condition in the way it's
implemented. We use virQEMUCapsGetDomainCapsCache() to obtain the
pointer to the hash table, then we search the hash table for
cached data and if none is found the domcaps is constructed and
put into the table. Problem is that this is all done without any
locking, so if there are two threads trying to do the same, one
will succeed and the other will fail inserting the data into the
table.

Also, the API looks a bit fishy - obtaining pointer to the hash
table is dangerous.

The solution is to use a mutex that guards the whole operation
with the hash table. Then, the API can be changes to return
virDomainCapsPtr directly.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1791790Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NPeter Krempa <pkrempa@redhat.com>

When fixing [1] I've ran attached reproducer and had it spawn
1024 threads and query capabilities XML in each one of them. This
lead libvirtd to hit the RLIMIT_NOFILE limit which was kind of
expected. What wasn't expected was a subsequent segfault. It
happened because virCPUProbeHost failed and returned NULL. We've
taken the NULL and passed it to virCapabilitiesHostNUMARef()
which dereferenced it. Code inspection showed the same flas in
virQEMUDriverGetHostNUMACaps(), so I'm fixing both places.

1: https://bugzilla.redhat.com/show_bug.cgi?id=1791790Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NPeter Krempa <pkrempa@redhat.com>

Remove the 'gluster' part and decouple the return from
the gluster_debug_level parsing to allow adding more options
to this section.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

virGetUserRuntimeDirectory() *never* *ever* returns NULL, making the
checks for it completely unnecessary.
Signed-off-by: NFabiano Fidêncio <fidencio@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

virGetUserConfigDirectory() *never* *ever* returns NULL, making the
checks for it completely unnecessary.
Signed-off-by: NFabiano Fidêncio <fidencio@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

virGetUserCacheDirectory() *never* *ever* returns NULL, making the
checks for it completely unnecessary.
Signed-off-by: NFabiano Fidêncio <fidencio@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

When QEMU uid/gid is set to non-root this is pointless as if we just
used a regular setuid/setgid call, the process will have all its
capabilities cleared anyway by the kernel.

When QEMU uid/gid is set to root, this is almost (always?) never
what people actually want. People make QEMU run as root in order
to access some privileged resource that libvirt doesn't support
yet and this often requires capabilities. As a result they have
to go find the qemu.conf param to turn this off. This is not
viable for libguestfs - they want to control everything via the
XML security label to request running as root regardless of the
qemu.conf settings for user/group.

Clearing capabilities was implemented originally because there
was a proposal in Fedora to change permissions such that root,
with no capabilities would not be able to compromise the system.
ie a locked down root account. This never went anywhere though,
and as a result clearing capabilities when running as root does
not really get us any security benefit AFAICT. The root user
can easily do something like create a cronjob, which will then
faithfully be run with full capabilities, trivially bypassing
the restriction we place.

IOW, our clearing of capabilities is both useless from a security
POV, and breaks valid use cases when people need to run as root.

This removes the clear_emulator_capabilities configuration
option from qemu.conf, and always runs QEMU with capabilities
when root.  The behaviour when non-root is unchanged.
Reviewed-by: NCole Robinson <crobinso@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Now that nearly all internal APIs use the QEMU capabilities or other
QEMU driver data directly, there's no compelling benefit to create
virCapsPtr at driver startup.

Skipping this means we don't probe capabilities for all 30 system
emulator targets at startup, only those emulators which are referenced
by an XML doc. This massively improves libvirtd startup time when the
capabilities cache is not populated. It even improves startup time
when the cache is up to date, as we don't bother to load files from
the cache until we need them.
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Avoid grabbing the whole virCapsPtr object when we only need the
host CPU information.
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Avoid grabbing the whole virCapsPtr object when we only need the
NUMA information.
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Instead of using the virCapsPtr to get the default security model,
pass this in via the parser config.
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

As part of a goal to eliminate the need to use virCapsPtr for anything
other than the virConnectGetCapabilies() API impl, cache the host arch
against the QEMU driver struct and use that field directly.

In the tests we move virArchFromHost() globally in testutils.c so that
every test runs with a fixed default architecture reported.
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

This previous commit introduced a simpler free callback for
hash data with only 1 arg, the value to free:

  commit 49288fac
  Author: Peter Krempa <pkrempa@redhat.com>
  Date:   Wed Oct 9 15:26:37 2019 +0200

    util: hash: Add possibility to use simpler data free function in virHash

It missed two functions in the hash table code which need
to call the alternate data free function, virHashRemoveEntry
and virHashRemoveSet.

After the previous patch though, there is no code that
makes functional use of the 2nd key arg in the data
free function. There is merely one log message that can
be dropped.

We can thus purge the current virHashDataFree callback
entirely, and rename virHashDataFreeSimple to replace
it.
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NDaniel Henrique Barboza <danielhb413@gmail.com>

There are two ways for specifying loader:nvram pairs:

  1) --with-loader-nvram configure option
  2) nvram variable in qemu.conf

Since we have FW descriptors, using this old style is
discouraged, but not as strong as one would expect. Produce more
warnings:

  1) produce a warning if somebody tries the configure option
  2) produce a warning if somebody sets nvram variable and at
     least on FW descriptor was found

The reason for producing warning in case 1) is that package
maintainers, who set the configure option in the first place
should start moving towards FW descriptors and abandon the
configure option. After all, the warning is printed into config
output only in this case.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1763477Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

Replace all occurrences of
  if (VIR_STRDUP(a, b) < 0)
     /* effectively dead code */
with:
  a = g_strdup(b);
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

When searching qemuCaps->domCapsCache for existing domCaps data,
we check for a matching pair of arch+virttype+machine+emulator. However
for the hash table key we only use the machine string. So if the
cache already contains:

  x86_64 + kvm + pc + /usr/bin/qemu-kvm

But a new VM is defined with

  x86_64 + qemu + pc + /usr/bin/qemu-kvm

We correctly fail to find matching cached domCaps, but then attempt
to use a colliding key with virHashAddEntry

Fix this by building a hash key from the 4 values, not just machine
Reviewed-by: NJonathon Jongsma <jjongsma@redhat.com>
Reviewed-by: NDaniel Henrique Barboza <danielhb413@gmail.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

Prefer:
    return g_steal_pointer(&ptr);
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

Now that all the types using VIR_AUTOUNREF have a cleanup func defined
to virObjectUnref, use g_autoptr instead of VIR_AUTOUNREF.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

Since commit 44e7f029
    util: rewrite auto cleanup macros to use glib's equivalent

VIR_AUTOPTR aliases to g_autoptr. Replace all of its use by the GLib
macro version.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

Since commit 44e7f029
    util: rewrite auto cleanup macros to use glib's equivalent

VIR_AUTOFREE is just an alias for g_autofree. Use the GLib macros
directly instead of our custom aliases.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

Prefer the GLib version of the macro.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

Use G_GNUC_UNUSED from GLib instead of ATTRIBUTE_UNUSED.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

The macro VIR_DELETE_ELEMENT assume that the items being deleted
have already been cleared, so we must explicitly free domain name
from the list of domains using the shared device to prevent a
memory leak.
Signed-off-by: NXu Yandong <xuyandong2@huawei.com>
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NDaniel Henrique Barboza <danielhb413@gmail.com>

Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NDaniel Henrique Barboza <danielhb413@gmail.com>

Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NDaniel Henrique Barboza <danielhb413@gmail.com>

There are two 'cleanup' labels - one in
virQEMUDriverConfigHugeTLBFSInit() and the other in
virQEMUDriverConfigSetDefaults() that do nothing more than
return and integer value. No memory freeing or anything important
is done there. Drop them in favour of returning immediately.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NDaniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: NPavel Hrdina <phrdina@redhat.com>

Our naming rules prefer qemuObjectOperation() scheme rather than
qemuOperationObject() for function names. These were not honoured
in recent commits to qemu_conf.c.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NDaniel Henrique Barboza <danielhb413@gmail.com>

After the previous commits, qemuAddSharedDevice() and
qemuRemoveSharedDevice() are now the same code with a different
flag to call the internal functions.

This patch aggregates the common code into a new function called
qemuAddRemoveSharedDeviceInternal() to further reduce
code repetition. Both qemuAddSharedDevice() and
qemuRemoveSharedDevice() are kept since they are public
functions used elsewhere.

No functional change was made.
Signed-off-by: NDaniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: NPavel Hrdina <phrdina@redhat.com>

Following the same idea of avoid code repetition from the
previous patch, this commit introduces a new function that
aggregates the functions of qemuAddSharedDisk() and
qemuRemoveSharedDisk() into a single place, using a flag to
switch between add/remove operations.

Both qemuAddSharedDisk() and qemuRemoveSharedDisk() are
public, so keep them around to avoid changing other files
due to an internal qemu_conf.c refactory.

No functional change was made.
Signed-off-by: NDaniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: NPavel Hrdina <phrdina@redhat.com>

qemuAddSharedHostdev() has a code similar to
qemuRemoveSharedHostdev(), with exception of one line that
defines the operation (add or remove).

This patch introduces a new function that aggregates the common
code, using a flag to switch between the operations, avoiding
code repetition.

No functional change was made.
Signed-off-by: NDaniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: NPavel Hrdina <phrdina@redhat.com>

Signed-off-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

A slirp helper is a process that provides user-mode networking through
a unix domain socket. It is expected to follow the following
specification:
https://gitlab.freedesktop.org/slirp/libslirp-rs/blob/master/src/bin/README.rstSigned-off-by: NMarc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

'virQEMUDriverConfigPtr cfg' is declared, initiated, but never
used in virQEMUDriverCreateCapabilities().
Signed-off-by: NDaniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>
Reviewed-by: NChristophe de Dinechin <dinechin@redhat.com>

All code using LOCALSTATEDIR "/run" is updated to use RUNSTATEDIR
instead. The exception is the remote driver client which still
uses LOCALSTATEDIR "/run". The client needs to connect to remote
machines which may not be using /run, so /var/run is more portable
due to the /var/run -> /run symlink.

Some duplicate paths in the apparmor code are also purged.

There's no functional change by default yet since both expressions
expand to the same value.
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

qemuCaps is tied to a binary on disk. domCaps is tied to a combo
of binary+machine+arch+virttype values. For the qemu driver this almost
entirely translates to a permutation of qemuCaps though

Upcoming patches want to use the domCaps data store at XML validate
time, but we need to cache the data so we aren't repeatedly
regenerating it.

Add a domCapsCache hash table to qemuCaps. This ensures that the domCaps
cache is blown away whenever qemuCaps needs to be regenerated. Similarly
when qemuCaps is invalidated, the next call to virQEMUCapsCacheLookup
will unref qemuCaps and free our cache as well.

Adjust virQEMUDriverGetDomainCapabilities to search the cache and add
to it if we don't find a hit.
Signed-off-by: NCole Robinson <crobinso@redhat.com>
Reviewed-by: NReviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>

For now it's just a helper for building a qemu virDomainCapsPtr,
used in qemuConnectGetDomainCapabilities
Reviewed-by: NReviewed-by: Daniel Henrique Barboza <danielhb413@gmail.com>
Reviewed-by: NMichal Privoznik <mprivozn@redhat.com>
Signed-off-by: NCole Robinson <crobinso@redhat.com>

This is similar to the existing directory for snapshots; the domain
will save one xml file per checkpoint, for reloading on the next
libvirtd restart.  Fortunately, since checkpoints mandate RNG
validation, we are assured that the checkpoint name will be usable as
a file name (no abuse of '../escape' as a checkpoint name, for
example).
Signed-off-by: NEric Blake <eblake@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

Some VM configurations may result in a large number of threads created by
the associated qemu process which can exceed the system default limit. The
maximum number of threads allowed per process is controlled by the pids
cgroup controller and is set to 16k when creating VMs with systemd's
machined service. The maximum number of threads per process is recorded
in the pids.max file under the machine's pids controller cgroup hierarchy,
e.g.

$cgrp-mnt/pids/machine.slice/machine-qemu\\x2d1\\x2dtest.scope/pids.max

Maximum threads per process is controlled with the TasksMax property of
the systemd scope for the machine. This patch adds an option to qemu.conf
which can be used to override the maximum number of threads allowed per
qemu process. If the value of option is greater than zero, it will be set
in the TasksMax property of the machine's scope after creating the machine.
Signed-off-by: NJim Fehlig <jfehlig@suse.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>