Signed-off-by: NJiri Denemark <jdenemar@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

This function gets snapshot XML (provided by used) as an
argument. It parses it into a local variable @def and then sets
some more members (e.g. it creates a copy of live domain XML).
Then it proceeds to checking if snapshot XML is valid (e.g. it
contains as many disks as currently in the domain). If this fails
then the control jumps to endjob label and subsequently return
from the function. This is where AUTOFREE function for @def is
ran. Well, because the code says to run plain VIR_FREE() we leak
some memory because @def is actually an object and therefore
it should have been declared as AUTOUNREF.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

A double free may occur in testCompareXMLToArgvFiles() when @def
is freed right after virStoragePoolObjNew() failed and the second
time at cleanup label.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

The idea of virCommand* APIs is that a possible error that
occurred while constructing cmd line is kept in virCommand
struct. If that's the case all subsequent calls to virCommand*()
are NO-OPs or they return an error. Well,
virCommandPassFDGetFDIndex() is not honoring that.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

If virStoragePoolObjNew() fails to create new volume object list
then virObjectUnref() is called and since refcounter is 1 then
virStoragePoolObjDispose() is called which in turn calls
virStoragePoolObjClearVols() which in turn dereferences
obj->volumes.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>
Reviewed-by: NErik Skultety <eskultet@redhat.com>

The qsort element is a pointer of virResctrlMonitorStats, and
the comparing function's arguments have a type of pointer of
virResctrlMonitorStatsPtr.
Signed-off-by: NHuaqiang <huaqiang.wang@intel.com>
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

The virDomainGetDiskErrors() API copies disk targets into @disks
array that we allocate. But we forgot to free it:

==140828== 16 bytes in 4 blocks are definitely lost in loss record 41 of 242
==140828==    at 0x4C2F08F: malloc (vg_replace_malloc.c:299)
==140828==    by 0x8C406D9: strdup (in /lib64/libc-2.28.so)
==140828==    by 0x5377DD3: virStrdup (virstring.c:966)
==140828==    by 0x54C112F: testDomainGetDiskErrors (test_driver.c:3068)
==140828==    by 0x55C863D: virDomainGetDiskErrors (libvirt-domain.c:10988)
==140828==    by 0x15D1FA: cmdDomBlkError (virsh-domain-monitor.c:1215)
==140828==    by 0x17F1A8: vshCommandRun (vsh.c:1335)
==140828==    by 0x13489E: main (virsh.c:920)
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Return the number of disks present in the configuration of the test
domain when called with @errors as NULL and @maxerrors as 0.

Otherwise report an error for every second disk, assigning available
error codes in a cyclic order.
Signed-off-by: NIlias Stamatis <stamatis.iliass@gmail.com>
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Our build system doesn't currently install the various
example programs provided along libvirt; however, both the
upstream .spec file and the Debian packaging go out of
their way to make sure these useful demos are included in
the respective documentation packages.

Moreover, doing so without help from the upstream build
system is easy to get wrong: the libvirt-docs RPM package,
for example, ends up missing one of the examples and
including a bunch of empty .deps/ directories.

Install the examples in $(docdir) as part of our regular
procedure, so that users and downstreams don't have to do
anything special about them.
Signed-off-by: NAndrea Bolognani <abologna@redhat.com>
Acked-by: NMichal Privoznik <mprivozn@redhat.com>

At the moment we allow the user to specify exactly where
they want the HTML documentation to be installed with an
extreme level of precision through the --with-html-dir and
--with-html-subdir configure options.

Most of the time, of course, the user will stick with the
default, that is $(datadir)/doc/$(PACKAGE)-$(VERSION)/html.

So close to $(docdir)! Including the version number in
the path, specifically, seems entirely unnecessary since
different releases of libvirt are not going to be able to
coexist on the same system anyway.

Drop all these custom flexibilty for flexibilty's sake
shenaningans in favor of the standard, well understood
$(docdir).
Signed-off-by: NAndrea Bolognani <abologna@redhat.com>
Acked-by: NMichal Privoznik <mprivozn@redhat.com>

Our XSLT magic generates one Devhelp-compatible HTML file
per documentation module, but so far we have only shipped
and installed documentation for virterror.

Now that we have $(modules), however, we can generate the
list of files the same way we do for regular documentation
and make sure we always ship and install everything.
Signed-off-by: NAndrea Bolognani <abologna@redhat.com>
Acked-by: NMichal Privoznik <mprivozn@redhat.com>

This variable contains a lists of documentation modules,
in a neutral format.

Right now is only used to define $(apihtml_generated), but
later on we're gonna reuse it.
Signed-off-by: NAndrea Bolognani <abologna@redhat.com>
Acked-by: NMichal Privoznik <mprivozn@redhat.com>

In refactoring the snapshot code to prepare for checkpoints, I changed
qemuDomainMomentDiscardAll to take a callback that would handle the
cleanup of either a snapshot or a checkpoint, but failed to set the
callback on one of the two snapshot callers.  As a result, 'virsh
undefine $dom --snapshots-metadata' crashed on a NULL function
dereference.

Fixes: a487890d
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1707708Signed-off-by: NEric Blake <eblake@redhat.com>
Acked-by: NPeter Krempa <pkrempa@redhat.com>

This way more users can run our CI builds locally.
Signed-off-by: NMartin Kletzander <mkletzan@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

If no board was detected then VIR_REALLOC_N() done at the end of
the function will actually free the memory (because nborads ==
0), but @boards will be set to a non-NULL pointer. This makes it
unnecessary harder for a caller to see if any board was detected.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Commit 10575666 was too eager and did
not consider SLES 12 which still has 2.0.1 that does not ship
a pkg-config file.

Similar to how we check for readline, prefer pkg-config if available
and fall back to the old detection code if not found.

NB: this is not a clean revert because we're not reintroducing support
for YAJL 1.
Signed-off-by: NJán Tomko <jtomko@redhat.com>
Reported-by: NOlaf Hering <olaf@aepfle.de>
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>

The way it works now the Makefile needs to be both make valid and automake
valid.  That is fine for now, but if we want to use anything more advanced, like
conditionals, we cannot have it like that any more.

So instead forward all ci-* rules to that file.
Signed-off-by: NMartin Kletzander <mkletzan@redhat.com>

This brings about a couple of benefits:
- use of VIR_AUTOUNREF() simplifies several callers
- Fixes a todo about virDomainMomentObjList not being polymorphic enough
Signed-off-by: NEric Blake <eblake@redhat.com>
Acked-by: NPeter Krempa <pkrempa@redhat.com>

In preparation for making virDomainSnapshotDef a descendant of
virObject, it is time to fix all callers that allocate an object to
use virDomainSnapshotDefNew() instead of VIR_ALLOC().  Fortunately,
there aren't very many :)
Signed-off-by: NEric Blake <eblake@redhat.com>
Acked-by: NPeter Krempa <pkrempa@redhat.com>

VIR_CLASS_NEW insists that descendents of virObject have 'parent' as
the name of their inherited base class member at offset 0. While it
would be possible to write a new class-creation macro that takes the
actual field name 'current', and rewrite VIR_CLASS_NEW to call the new
macro with the hard-coded name 'parent', it seems less confusing if
all object code uses similar naming. Thus, this is a mechanical rename
in preparation of making virDomainSnapshotDef a descendent of
virObject.
Signed-off-by: NEric Blake <eblake@redhat.com>
Acked-by: NPeter Krempa <pkrempa@redhat.com>

VIR_CLASS_NEW insists that descendents of virObject have 'parent' as
the name of their inherited base class member at offset 0. While it
would be possible to write a new class-creation macro that takes the
actual field name, and rewrite VIR_CLASS_NEW to call the new macro
with the hard-coded name 'parent', so that we could make
virDomainMomentDef use a custom name for its base class, it seems less
confusing if all object code uses similar naming. Thus, this is a
mechanical rename in preparation of making virDomainSnapshotDef a
descendent of virObject, when we can no longer use 'parent' for a
different purpose than the base class.
Signed-off-by: NEric Blake <eblake@redhat.com>
Acked-by: NPeter Krempa <pkrempa@redhat.com>

Use qemuDomainStorageSourceAccessModify with correct flags to do the
job.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Some operations e.g. namespace setup are not necessary when modifying
access to a file which the VM can already access. Add a flag which
allows to skip them.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

In some cases when we need to modify access permissions for a storage
source which is already used by the VM we should not revoke all
permissions on a failure. Allow this in qemuDomainStorageSourceAccessModify
by adding a new flag.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Rather than jumping to the correct label use a set of booleans to
determine which operation needs to be rolled back. This will allow more
flexibility when e.g. rollback after a failed operation will not be
necessary/desired.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Add a new flag which will set the image as read-only even if the image
data allows writing.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Use qemuDomainStorageSourceAccessModify instead of the individual calls.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Add a new flag QEMU_DOMAIN_STORAGE_SOURCE_ACCESS_CHAIN to select whether
to work on single image or full chain.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Upcoming patches will add a few more flags. Add an enum to collect them
so that we don't end up with multiple bools.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

The function will be able to deal with non-chains too so drop 'Chain'
and also change the suffix to 'Modify' as it's used both for setup and
teardown.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Introduce qemuDomainStorageSourceChainAccess(Allow|Revoke) as entry
points to qemuDomainStorageSourceChainAccessPrepare for symmetry with
the functions for single backing chain elements.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Move it to qemu_domain.c and call it
qemuDomainStorageSourceChainAccessPrepare.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Use qemuDomainStorageSourceAccess(Allow|Revoke) instead.
Signed-off-by: NPeter Krempa <pkrempa@redhat.com>
Reviewed-by: NJán Tomko <jtomko@redhat.com>

Fedora 30 is out, which means that Fedora 28 is going to be
EOL very soon. Let's get ahead of the game and drop support
for it right now.
Signed-off-by: NAndrea Bolognani <abologna@redhat.com>
Reviewed-by: NDaniel P. Berrangé <berrange@redhat.com>

Most releases don't need a "Removed features" section so don't include
it in the template by default, but leave a reminder in case it is
relevant.
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

GitLab CI provides some shared build runners that use Docker containers.
This resource can usefully run cross-compiled builds since all other CI
build testing is currently x86 only, and Travis CI is already very busy
testing native builds.
Reviewed-by: NAndrea Bolognani <abologna@redhat.com>
Signed-off-by: NDaniel P. Berrangé <berrange@redhat.com>

If the source tree was freshly configured and no objects are
built yet then 'make check-access' has no test to run. Build
the sources beforehand.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>

Introduced by ff376c62.

Previously, init_syms() was called from stat() mock and its
friends. This is crucial because checkPath() might call
printFile() which in turn calls real_fopen(). But if stat() or
one of its friends is the first function called then because of
lacking init_syms() call no real_* is initialized.

The other thing is that we really want the recorded action to be
"stat" instead of __FUNCTION__ because there's no good in
recording that it was __xstat64 who touched some file.
Signed-off-by: NMichal Privoznik <mprivozn@redhat.com>