In preparation for splitting up the CPU map data file, move it into a
dedicated directory of its own.
Reviewed-by: NJiri Denemark <jdenemar@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Get rid of the separate 'error:' label, so all code paths jump straight
to the 'cleanup:' label.
Reviewed-by: NJiri Denemark <jdenemar@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

The x86 and ppc impls both duplicate some logic when parsing CPU
features. Change the callback signature so that this duplication can be
pushed up a level to common code.
Reviewed-by: NJiri Denemark <jdenemar@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Two pieces of code accidentally jumped to the wrong label when they
failed causing incorrect cleanup, returning a partially initialized
CPU model struct.
Reviewed-by: NJiri Denemark <jdenemar@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Allow for syntax

    <include filename="subdir/fooo.xml"/>

to reference other files in the CPU database directory
Reviewed-by: NJiri Denemark <jdenemar@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

Commit deb057fd added a switch without a default case.
Add it and call virReportEnumRangeError for _LAST too.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

Commit 6700062f introduced a jump to error which skipped the
initialization of def:

qemu/qemu_parse_command.c:1870:9: error: variable 'def' is
used uninitialized whenever 'if' condition is true
      [-Werror,-Wsometimes-uninitialized]
    if (!(qemuCaps = virQEMUCapsCacheLookup(capsCache, progargv[0])))

Initialize def to fix this warning and qemuCaps, to prevent
a future error like this.
Signed-off-by: NJán Tomko <jtomko@redhat.com>

Historically the argv -> xml convertor wanted the same default machine
as we'd set when parsing xml. The latter has now changed, however, to
use a default defined by libvirt. The former needs fixing to again
honour the default QEMU machine.

This exposed a bug in handling for the aarch64 target, as QEMU does not
define any default machine. Thus we should not having been accepting
argv without a -machine provided.
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

The virQEMUCapsGetDefaultMachine() method doesn't get QEMU's default
machine any more, instead it gets the historical default that libvirt
prefers for each arch. Rename it, so that the old name can be used for
getting QEMU's default.
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

We don't honour the QEMU default machine type anymore, always using the
libvirt chosen default instead. The QEMU argv parser, however, will need
to know the exacty QEMU default, so we must record that info.
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

'metadata' and 'leases' are features internal to libvirt and thus don't
influence the generated QEMU command line. As they are not tested we
don't need the output files.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

Now we assume the flag always so there's no use for this test. Probably
a leftover from the cleanup of the capability.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

The test files were unused, but we don't have any other test for this
feature. Make use of the existing files by removing disks and using
DO_TEST_CAPS_LATEST to execute them. The legacy output files will be
dropped.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

Output file does not make sense for those.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

Turn

  virPCIDeviceAddressIsEmpty()
  virDeviceInfoPCIAddressIsWanted()
  virDeviceInfoPCIAddressIsPresent()

from inline functions to regular functions.
Signed-off-by: NAndrea Bolognani <abologna@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

The affected functions are

  virDeviceInfoPCIAddressWanted()
  virDeviceInfoPCIAddressPresent()

which get renamed to

  virDeviceInfoPCIAddressIsWanted()
  virDeviceInfoPCIAddressIsPresent()

to comply with the naming convention used for other
predicates.
Signed-off-by: NAndrea Bolognani <abologna@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

The function wcwidth() doesn't exist on mingw. However, it does exist in
gnulib, so we can rely on that.
Signed-off-by: NSimon Kobyda <skobyda@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

Reviewed-by: NJán Tomko <jtomko@redhat.com>
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

Reviewed-by: NJán Tomko <jtomko@redhat.com>
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

This fixes virCgroupEnableMissingControllers where virCgroupRemove
was not called in case virCgroupMakeGroup failed.
Reviewed-by: NJán Tomko <jtomko@redhat.com>
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

The case where we need path of any controller is only for internal use
so move it out to a different function.
Reviewed-by: NJán Tomko <jtomko@redhat.com>
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

Reviewed-by: NJán Tomko <jtomko@redhat.com>
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

Reviewed-by: NJán Tomko <jtomko@redhat.com>
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

The 'mntDir' is part of 'struct mntent' as a result of getmntent_r
therefore we should not mangle with it.
Reviewed-by: NJán Tomko <jtomko@redhat.com>
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

aarch64-acpi-nouefi and hostdev-scsi-boot are unused. Noticed when
checking whether '-nodefconfig' is still used by libvirt.

Unused since their introduction in commit deb38c45 and bab6ee6b
respectively.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

This patch ensures that changes in attributes of interfaces will emit
errors except if they are missing from the XML.
Previously we were falsely reporting successful updates, because some
changed attributes got overwritten before the validity checks.

https://bugzilla.redhat.com/show_bug.cgi?id=1599513Signed-off-by: NKaterina Koukiou <kkoukiou@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Due to kernel upstream change 338d0be4 ("apparmor: fix ptrace read check")
libvirt now hits apparmor denies like:
  apparmor="DENIED" operation="ptrace" profile="/usr/sbin/libvirtd"
  pid=4409 comm="libvirtd" requested_mask="read" denied_mask="read"
  peer="libvirt-14e92a75-7668-4b97-8f92-322fc1b9c78a"

Extend the ptrace rule to also allow 'ptrace (read)' for libvirtd to work
with these newer kernels.

Fixes: https://bugs.launchpad.net/bugs/1788603Reported-by: NThadeu Lima de Souza Cascardo <thadeu.cascardo@canonical.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>
Acked-by: NJamie Strandboge <jamie@canonical.com>
Signed-off-by: NChristian Ehrhardt <christian.ehrhardt@canonical.com>

qemuProcessInitCpuAffinity prevents a VM from getting started on a
platform that uses cpu affinity wrapper stubs e.g. macOS.

The patch adds qemuProcessInitCpuAffinity stub on all platforms without
HAVE_SCHED_GETAFFINITY or HAVE_BSD_CPU_AFFINITY.
Signed-off-by: NRoman Bolshakov <r.bolshakov@yadro.com>
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Fixes:
https://www.redhat.com/archives/libvir-list/2017-January/msg00978.html

QEMU is probed through monitor fd to check capabilities during libvirtd init.
The monitor fd is closed after probing by virQEMUCapsInitQMPCommandFree
that calls virQEMUCapsInitQMPCommandAbort that calls qemuMonitorClose,
the latter one notifies the event loop via an interrupt handle in
qemuMonitorUnregister and after then closes monitor fd.

There could be a case when interrupt is sent after eventLoop is unlocked
but before virEventPollRunOnce blocks in poll, shortly before file
descriptor is closed by qemuMonitorClose. Then poll receives closed monitor
fd in fdset and returns EBADF.

EBADF is not mentioned as a valid errno on macOS poll man-page but such
behaviour can appear release-to-release, according to cpython:
https://github.com/python/cpython/blob/master/Modules/selectmodule.c#L1161

The change also fixes the issue in qemucapabilitiestest. It returns
Bad file descriptor message 25 times without the fix.
Signed-off-by: NRoman Bolshakov <r.bolshakov@yadro.com>
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Save us hassle in the list if anybody would read this.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

Now that they are not linked any more remove them.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

The patches used as an example for the api_extension manual don't hold
up to the current standards any more. Carefully remove links and
mentions of the patches from the docs.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

The capability was usable since qemu 1.3 so we can remove all the
detection code.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

All supported qemus support FD passing so modify the tests to test the
proper code path.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

Allow mocking of the file descriptor numbers used for the TPM
passthrough mode by extracting the relevant code into an exported
function.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

Allow FDs which are marked as safe for FD passing.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

For versions where we can probe that the arguments are optional we can
perform the probing by a schema query rather than sending a separate
command to do so.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

Add a new modifier character '*' which will select given schema entry
only when it is optional (denoted by the presence of the 'default' key).
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

Split it into a function that returns the whole schema entry so that we
can do additional checks and a helper getting the type string from the
schema entry.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>

Introduce a few variables so that we can easily access the modifier
character and also don't have to do pointer arithmetic when selecting
the schema entries. This will simplify adding of new modifier
characters.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJohn Ferlan <jferlan@redhat.com>