The first argument should be const char ** instead of
char **, because this is a search function and as such it
doesn't, and shouldn't, alter the haystack in any way.

This change means we no longer have to cast arrays of
immutable strings to arrays of mutable strings; we still
have to do the opposite, though, but that's reasonable.

When commit id '6dfb4507' refactored where the iothreadsched data was
stored, the error message for when the virDomainIOThreadIDFind failed
to find an iothreadid ("iothreadsched attribute 'iothreads' uses
undefined iothread ids") was lost. This led to the possibility that
someone would try to use it, but receive the generic message "An error
occurred, but the cause is unknown".

This patch adds the error message back so that someone will know that
they have an invalid configuration.
Signed-off-by: NJohn Ferlan <jferlan@redhat.com>

All other modes of qemuDomainSetVcpusFlags have helpers so finish the
work by splitting the regular code into a new function.

This patch also touches up the coding (spacing) style.

The live code does ugly things. Contain it in a separate function.

Setting of the maximum vcpu count is slightly semantically different
thus split it into a self-contained func.

qemu uses 'url' instead of 'uri'. They unfortunately look very similar.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1367260

If any of the devices referenced a USB hub that does not exist,
defining the domain would either fail with:
error: An error occurred, but the cause is unknown
(if only the last hub in the path is missing)
or crash.

Return a proper error instead of crashing.

https://bugzilla.redhat.com/show_bug.cgi?id=1367130

The array needs to be freed too, not just its members.

https://bugzilla.redhat.com/show_bug.cgi?id=1366097

Mention whether it was the live or persistent definition which caused an
error reported and explicitly error out in case when attempting to set
maximum vcpu count for a live domain.

  <filesystem type='ram' accessmode='passthrough'>
    <source usage='524288' units='KiB'/>
    <target dir='/dev/shm'/>
  </filesystem>

would lead to lxcContainerMountAllFS calling STRPREFIX
on a NLL pointer because it failed to check if fs->src->path
was non-NULL. This is a regression caused by

  commit da665fbd
  Author: Olga Krishtal <okrishtal@virtuozzo.com>
  Date:   Thu Jul 14 16:52:38 2016 +0300

    filesystem: adds possibility to use storage pool as fs source
Signed-off-by: NOlga Krishtal <okrishtal@virtuozzo.com>
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>

  <filesystem type='ram' accessmode='passthrough'>
    <source usage='524288' units='KiB'/>
    <target dir='/dev/shm'/>
  </filesystem>

would lead to lxcContainerResolveSymlinks calling
access(NULL) because it failed to check if fs->src->path
was non-NULL. This is a regression caused by

  commit da665fbd
  Author: Olga Krishtal <okrishtal@virtuozzo.com>
  Date:   Thu Jul 14 16:52:38 2016 +0300

    filesystem: adds possibility to use storage pool as fs source
Signed-off-by: NOlga Krishtal <okrishtal@virtuozzo.com>
Signed-off-by: NDaniel P. Berrange <berrange@redhat.com>

Because of change in caaa1bd3 this macro is no under
#ifdef block. That means it needs to be re-intended correctly.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Commit 11567cf6 added some libxl tests into domaincapstest and
added libvirt_driver_libxl_impl.la to domaincapstest_LDADD.

This causes link fail on systems without GNU regex implementation:

gmake[2]: Entering directory '/usr/home/novel/code/libvirt/tests'
  CCLD     domaincapstest
  ../src/.libs/libvirt_driver_libxl_impl.a(libvirt_driver_libxl_impl_la-libxl_capabilities.o):
  In function `libxlMakeCapabilities':
  libxl/libxl_capabilities.c:(.text+0x6b2): undefined reference to
  `rpl_regcomp'
  libxl/libxl_capabilities.c:(.text+0x6d0): undefined reference to
  `rpl_regerror'
  libxl/libxl_capabilities.c:(.text+0x803): undefined reference to
  `rpl_regexec'
  libxl/libxl_capabilities.c:(.text+0xa58): undefined reference to
  `rpl_regfree'
  clang-3.8: error: linker command failed with exit code 1 (use -v to
  see invocation)

This happens because on these system it tries to use gnulib's builtin
regex implementation, but doesn't link to gnulib.

Fix by adding $(GNULIB_LIBS) along with libvirt_driver_libxl_impl.la to
domaincapstest_LDADD.

Commit eee7bd4e introduced two functions: libxlDiskPathToID and
libxlDiskSectorSize.

However, as they're used only by code under #ifdef __linux__,
on non-Linux platforms it results in errors similar to this:

 CC       libxl/libvirt_driver_libxl_impl_la-libxl_driver.lo
libxl/libxl_driver.c:5263:1: error: unused function 'libxlDiskPathToID' [-Werror,-Wunused-function]
libxlDiskPathToID(const char *virtpath)
^
libxl/libxl_driver.c:5312:1: error: unused function 'libxlDiskSectorSize' [-Werror,-Wunused-function]
libxlDiskSectorSize(int domid, int devno)
^
2 errors generated.

Fix that by moving these functions under the #ifdef __linux__ block.

This event is emitted when a nodedev XML definition is updated,
like when cdrom media is changed in a cdrom block device.

Also includes node device update event implementation for udev
backend, virsh nodedev-event support, and event-test support

https://bugzilla.redhat.com/show_bug.cgi?id=1366611

When commit id 'cb2e3e50' reworked the cmdSecretGetValue call to use
VIR_DISPOSE_STRING for base64, it neglected to initialize the base64
value to NULL since the cleanup: label could be reached prior to the
base64 value being set or not.  This resulted in a core dump, adding
the initialization will avoid the issue.

Setting heads to 0 in case that *max_outputs* is not supported while building
command line doesn't have any real effect.  It only removes *heads* attribute
from live XML, but after restarting libvirt the default value is restored.
Signed-off-by: NPavel Hrdina <phrdina@redhat.com>

It may happen that a developer wants to run just a specific
subset of tests:

tests $ VIR_TEST_RANGE=22 ../run ./virschematest

This now fails miserably:

    ==6840== Invalid read of size 8
    ==6840==    at 0x4F397C0: virXMLValidatorValidate (virxml.c:1216)
    ==6840==    by 0x402B72: testSchemaFile (virschematest.c:53)
    ==6840==    by 0x403737: virTestRun (testutils.c:180)
    ==6840==    by 0x402CF5: testSchemaDir (virschematest.c:98)
    ==6840==    by 0x402EB1: testSchemaDirs (virschematest.c:131)
    ==6840==    by 0x40314D: mymain (virschematest.c:194)
    ==6840==    by 0x4051AF: virTestMain (testutils.c:982)
    ==6840==    by 0x4035A9: main (virschematest.c:217)
    ==6840==  Address 0x10 is not stack'd, malloc'd or (recently) free'd

Problem is, we are trying to do two types of tests here: validate
RNG schema itself, and validate XML files against RNG schemas.
And the latter tries to re-use a resource allocated in the
former. Therefore if the former is skipped (due to
VIR_TEST_RANGE) we have to allocate the resource manually.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

When starting a guest and copying host vendor cpuid to the guest
cpu, libvirtd would crash if the host cpu contained a NULL vendor
field. Avoid the crash by checking for a valid vendor in the host
cpu before copying the cpuid to the guest cpu.

For completeness, here is a backtrace from the crash

(gdb) bt
f0  0x00007ffff739bf33 in x86DataCpuid (cpuid=0x8, cpuid=0x8,
    data=data@entry=0x7fffb800ee78) at cpu/cpu_x86.c:287
f1  virCPUx86DataAddCPUID (data=data@entry=0x7fffb800ee78, cpuid=0x8)
    at cpu/cpu_x86.c:355
f2  0x00007ffff739ef47 in x86Compute (host=<optimized out>, cpu=0x7fffb8000cc0,
    guest=0x7fffecca7348, message=<optimized out>) at cpu/cpu_x86.c:1580
f3  0x00007fffd2b38e53 in qemuBuildCpuModelArgStr (migrating=false,
    hasHwVirt=<synthetic pointer>, qemuCaps=0x7fffb8001040, buf=0x7fffecca7360,
    def=0x7fffc400ce20, driver=0x1c) at qemu/qemu_command.c:6283
f4  qemuBuildCpuCommandLine (cmd=cmd@entry=0x7fffb8002f60,
    driver=driver@entry=0x7fffc80882c0, def=def@entry=0x7fffc400ce20,
    qemuCaps=qemuCaps@entry=0x7fffb8001040, migrating=<optimized out>)
    at qemu/qemu_command.c:6445
(gdb) f2
(gdb) p *host_model
$23 = {name = 0x7fffb800ec50 "qemu64", vendor = 0x0, signature = 0, data = {
    len = 2, data = 0x7fffb800e720}}

Since we now pick the default USB controller model when parsing
the guest XML, we can get rid of some duplicated code so that
the default model selection happens in one place only.

Add some comments as well.

Now that the default USB controller model is explicit rather
than implicit for i440fx machines, we have to tweak the
conditions for dropping it in order to keep migration towards
libvirt <= 0.9.4 working.

When the user doesn't specify any model for a USB controller,
we use an architecture-dependent default, but we don't reflect
it in the guest XML.

Pick the default USB controller model when parsing the guest
XML instead of when creating the QEMU command line, so that
our choice is saved back to disk.

==8630== Invalid read of size 8
==8630==    at 0x4EA4F0F: virFree (viralloc.c:582)
==8630==    by 0x4F398F0: virXMLValidatorFree (virxml.c:1257)
==8630==    by 0x40305C: mymain (virschematest.c:191)
==8630==    by 0x405159: virTestMain (testutils.c:982)
==8630==    by 0x403553: main (virschematest.c:215)
==8630==  Address 0xcd72243 is 131 bytes inside a block of size 177 free'd
==8630==    at 0x4C2B1F0: free (vg_replace_malloc.c:473)
==8630==    by 0x4EA4F19: virFree (viralloc.c:582)
==8630==    by 0x4ED0973: virFindFileInPath (virfile.c:1646)
==8630==    by 0x405149: virTestMain (testutils.c:980)
==8630==    by 0x403553: main (virschematest.c:215)
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

https://bugzilla.redhat.com/show_bug.cgi?id=1353296

On UNIX like systems there are no constraints on what characters
can be in file/dir names (except for NULL, obviously). Moreover,
some values that we think of as paths (e.g. disk source) are not
necessarily paths at all. For instance, some hypervisors take
that as an arbitrary identifier and corresponding file is then
looked up by hypervisor in its table. Instead of trying to fix
our regular expressions (and forgetting to include yet another
character there), lets drop the validation completely.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Usually, this variable is used to hold the return value for a
function of ours. Well, this is not the case. Its use does not
match our pattern and therefore it is very misleading. Drop it
and define an alternative @rc variable, but only in that single
block where it is needed.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

This variable is very misleading. We use VIR_FORCE_CLOSE to set
it to -1 and returning it even though it does not refer to a FD
at all. It merely holds 0 or -1. Drop it completely. Also, at the
same time some corner cases are fixed too.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

https://bugzilla.redhat.com/show_bug.cgi?id=1240439

In this function we create a macvtap device and open its tap
device. Possibly multiple times. Now the thing is, if opening the
tap device fails, that is virNetDevMacVLanTapOpen() returns a
negative value, we unroll all the changes BUT return 0 fooling
caller into thinking everything went okay.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Since a9331394 (first release v2.1.0), specifying a manual
security_driver setting in qemu.conf causes the daemon to fail to
start, erroring with 'Duplicate security driver X'.

The duplicate checking was incorrectly comparing every entry
against itself, guaranteeing a false positive.

https://bugzilla.redhat.com/show_bug.cgi?id=1365607

More misunderstanding/mistaken assumptions on my part - I had thought
that a pci-expander-bus could be plugged into any legacy PCI slot, and
that pcie-expander-bus could be plugged into any PCIe slot. This isn't
correct - they can both be plugged ontly into their respective root
buses. This patch adds that restriction.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1358712

libvirt had allowed a dmi-to-pci-bridge to be plugged in anywhere a
normal PCIe endpoint can be connected, but this is wrong - it will
only work if it's plugged into pcie-root (the PCIe root complex) or a
pcie-expander-bus (the qemu device pxb-pcie). This patch adjusts the
connection flags accordingly.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1363648

I apparently misunderstood Marcel's description of what could and
couldn't be plugged into qemu's pxb-pcie controller (known as
pcie-expander-bus in libvirt) - I specifically allowed directly
connecting a pcie-switch-upstream-port, and it turns out that causes
the guest kernel to crash.

This patch forbids such a connection, and updates the xml docs
appropriately.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1361172

The virDomainPCIAddressFlagsCompatible() error logs report that a
device required a controller that accepted standard PCI endpoint
devices, or PCI Express endpoint devices, and if hotplug was required
by the configuration but not provided by the selected controller. But
the wording of the error messages was apparently confusing (according
to the bugzilla report referenced below). On top of that, if the
device was something other than an endpoint device (e.g. a
pcie-switch-downstream-port) the error message was a complete punt -
it would just say that the flags were incorrect.

This patch makes the messages for PCI/PCIe endpoint and hotplug
requirements more clear, and also specifically indicates what was the
device type when it is other than an endpoint device.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1363627

After commit 9d479dd1 fiddled with the cmdConnect's output which used to be a
bit more verbose prior to the mentioned commit, the program flow would result
in a quite confusing error if an invalid URI has been provided:

    error: Failed to connect to the admin server
    Connected to the admin server
    error: <some error>

The problem is that the commit mentioned above relied on the fact that
connect routine always succeeds which is not true.
Signed-off-by: NErik Skultety <eskultet@redhat.com>

Since the introduction of CMT features (commit v1.3.5-461-gf294b83e)
starting a domain with host-model CPU on a host which supports CMT fails
because QEMU complains about unknown 'cmt' feature:

    qemu-system-x86_64: CPU feature cmt not found

https://bugzilla.redhat.com/show_bug.cgi?id=1355857Signed-off-by: NJiri Denemark <jdenemar@redhat.com>

The generated command line wouldn't work since QEMU doesn't know what
'cmt' is. The following patch will fix this issue.

https://bugzilla.redhat.com/show_bug.cgi?id=1355857Signed-off-by: NJiri Denemark <jdenemar@redhat.com>

By removing a non-migratable feature in a for loop we would fail to drop
every second non-migratable feature if the features array contained
several of them in a row.
Signed-off-by: NJiri Denemark <jdenemar@redhat.com>

Signed-off-by: NJiri Denemark <jdenemar@redhat.com>

In libvirt, snapshot means disk snapshot.
snapshot --live is more like VM checkpoint.
Make it clear in virsh.pod.
Signed-off-by: NChen Hanxiao <chenhanxiao@gmail.com>
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Add nodedev-event support for node device lifecycle events

Commit 30ce2f0e tried to fix the issue with an incorrect session URI to admin
server but it messed up the checks:

    if (geteuid == 0 && VIR_STRDUP(*uristr, "libvirtd:///system") < 0)
        return -1;
    else if (VIR_STRDUP(*uristr, "libvirtd:///session") < 0)
        return -1;

So if a client executed with root privileges tries to connect, its euid is
checked (true) and the correct URI is successfully copied to @uristr (false),
therefore the 'else' branch is taken and @uristr is replaced by the session URI
which for root results in:
    Failed to connect socket to '/root/.cache/libvirt/libvirt-admin-sock':
    No such file or directory
Signed-off-by: NErik Skultety <eskultet@redhat.com>