Michal Privoznik [Thu, 11 Feb 2016 10:14:11 +0000 (11:14 +0100)]
dbus: Don't unref NULL messages
Apparently we are not the only ones with dumb free functions
because dbus_message_unref() does not accept NULL either. But if
I were to vote, this one is even more evil. Instead of returning
an error just like we do it immediately dereference any pointer
passed and thus crash you app. Well done DBus!
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f878ebda700 (LWP 31264)]
0x00007f87be4016e5 in ?? () from /usr/lib64/libdbus-1.so.3
(gdb) bt
#0 0x00007f87be4016e5 in ?? () from /usr/lib64/libdbus-1.so.3
#1 0x00007f87be3f004e in dbus_message_unref () from /usr/lib64/libdbus-1.so.3
#2 0x00007f87bf6ecf95 in virSystemdGetMachineNameByPID (pid=9849) at util/virsystemd.c:228
#3 0x00007f879761bd4d in qemuConnectCgroup (driver=0x7f87600a32a0, vm=0x7f87600c7550) at qemu/qemu_cgroup.c:909
#4 0x00007f87976386b7 in qemuProcessReconnect (opaque=0x7f87600db840) at qemu/qemu_process.c:3386
#5 0x00007f87bf6edfff in virThreadHelper (data=0x7f87600d5580) at util/virthread.c:206
#6 0x00007f87bb602334 in start_thread (arg=0x7f878ebda700) at pthread_create.c:333
#7 0x00007f87bb3481bd in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:109
(gdb) frame 2
#2 0x00007f87bf6ecf95 in virSystemdGetMachineNameByPID (pid=9849) at util/virsystemd.c:228
228 dbus_message_unref(reply);
(gdb) p reply
$1 = (DBusMessage *) 0x0
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Ján Tomko [Thu, 28 Jan 2016 16:44:33 +0000 (17:44 +0100)]
Clean up usage of 'ret' variable
Do not store the return value of called functions in the same variable
as the (future) return value of the current function.
This makes tracking the origin of the value easier and reduces
the chance of introducing a new point of exit without resetting
the return value back to -1.
Since we already have virtio channel events, we know when guest
agent within guest has (dis-)connected. Instead of us blindly
connecting to a socket that no one is listening to, we can just
follow what qemu-ga does. This has a nice benefit that we don't
need to 'guest-ping' the agent just to timeout and find out
nobody is listening.
The way that this commit is implemented:
- don't connect in qemuProcessLaunch directly, defer that to event
callback (which already follows the agent) -
processSerialChangedEvent
- after migration is settled, before we resume vCPUs, ask qemu
whether somebody is listening on the socket and if so, connect
to it.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
John Ferlan [Tue, 9 Feb 2016 11:24:33 +0000 (06:24 -0500)]
qemu: Check return status for virDomainGraphicsListenSetAddress
Recent refactors in the vbox code to check the return status for the
function tipped Coverity's scales of justice for any functions that
do not check status - such as this one.
While I'm at it, since the call is essentially the same other than
whether starting from val or val+1 when val[0] = '[', just adjust
the val pointer by one and have one call instead of two.
Additionally, the call to virDomainGraphicsListenGetAddress is redundant
since it checking that the address field got filled. It's a leftover
from the strndup -> ListenSetAddress conversion (commit id 'ef79fb5b5')
John Ferlan [Tue, 9 Feb 2016 11:36:17 +0000 (06:36 -0500)]
qemu: Introduce qemuParseCommandLineVnc
Refactor qemuParseCommandLine to pull out the "-vnc" argument parsing
into its own helper function. Modify the code to use "cleanup" instead
of "error" and use the standard return processing to indicate success
or failure by using ret
Turns out that not linking against libvirt and gnulib is okay for
regular Linux (and FreeBSD) builds, but makes mingw very unhappy.
.../virnetserverclientmock_la-virnetserverclientmock.o:
In function `virNetSocketGetSELinuxContext':
.../virnetserverclientmock.c:61: undefined reference to `rpl_strdup'
.../libvirportallocatormock_la-virportallocatortest.o:
In function `init_syms':
.../virportallocatortest.c:61: undefined reference to `virFileClose'
Andrea Bolognani [Wed, 10 Feb 2016 15:51:51 +0000 (16:51 +0100)]
tests: Don't link mock libraries against libvirt and gnulib
Mock libraries are used with LD_PRELOAD from test binaries that
are already linked against those libraries, so they will be able
to resolve the symbols anyway.
Michal Privoznik [Wed, 10 Feb 2016 08:34:59 +0000 (09:34 +0100)]
xenconfig: Properly check retval of virDomainGraphicsListenSetAddress
The function, like others in our code, returns zero on success
and a negative value on error. However, there are two places in
xenconfig source code where we check for non-zero value. While
the function can't currently return a positive value, those
checks look okay, but does not really follow our style.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Joao Martins [Fri, 5 Feb 2016 20:45:03 +0000 (20:45 +0000)]
libxl: keepalive messages support
This patch introduces keep alive messages support for P2P migration
and it adds two new configuration entries namely 'keepalive_interval'
'keepalive_count' to control it. Behavior of these entries is the
same as qemu driver thus the description is copied from there
with just a few simplifications.
Joao Martins [Fri, 5 Feb 2016 20:45:02 +0000 (20:45 +0000)]
libxl: add p2p migration
Introduce support for VIR_MIGRATE_PEER2PEER in libvirt migration.
Most of the changes occur at the source and no modifications at
the receiver.
In P2P mode there is only the Perform phase so we must handle the
connection with the destination and actually perform the
migration. libxlDomainPerformP2P implements the connection to the
destination and libxlDoMigrateP2P implements the actual migration
logic with virConnectPtr. In this function we take care of doing
all phases of migration in the destination similar to
virDomainMigrateVersion3Full. We appropriately save the last
error reported in each of the phases to provide proper reporting.
We don't yet support VIR_MIGRATE_TUNNELED and we always use V3
with extensible params, thus it also makes the implementation
simpler.
It is worth noting that the receiver didn't have any changes, and
since it's still the v3 sequence thus it is possible to migrate
from a P2P to non-P2P host.
Cole Robinson [Wed, 27 Jan 2016 21:03:52 +0000 (16:03 -0500)]
tests: qemuxml2xml: assign device addresses
We use the PreFormat callback for this. Many test cases need to be extended
to pass in proper qemuCaps flags so AssignAddresses doesn't throw errors.
One test case (pcie-root-port-too-many) is dropped, since it was meant
only for checking an error condition in qemuxml2argv, and one we add in
AssignAddresses it errors here too.
Long term I think AssignAddresses should be handled in qemu's PostParse
callback, but that's not entirely straightforward. Handling it here
means we can get the test suite churn over with.
The virStringListLength function does not ever modify the passed
string list. It merely counts the items in it. Make sure that we
reflect this bit in the function header.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
(crobinso: fix up spacing and squash in sheepdog bit suggested
by Andrea)
Cole Robinson [Wed, 13 Jan 2016 01:41:49 +0000 (20:41 -0500)]
tests: qemuxml2argv: remove some QEMU_CAPS_DEVICE problem cases
When we unconditionally enable QEMU_CAPS_DEVICE, these tests need
some massaging, so do it ahead of time to not mix it in with the
big test refresh.
- minimal-s390 is not a real world working config, so drop it
- disk-usb was testing for an old code path that will be removed.
instead use it to test lack of USB disk support, and rename it
to disk-usb-nosupport. Switch xml2xml to use disk-usb-device for
input.
- cputune-numatune was needlessly using q35, switch it to an older
machine type
Older gcc fails to see that the variable is set iff @hasPriority
== true in which case the former is set a value. Initialize the
value while declaring it to make the compiler shut up.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Cole Robinson [Wed, 13 Jan 2016 15:43:27 +0000 (10:43 -0500)]
tests: qemuxml2xml: Always use different output file
Most qemuxml2xml tests expect that the input XML is unchanged after
parsing. This is unlike 99% of new qemu configs in the wild, which after
initial parsing end up with stable PCI device addresses. The xml2xml bit
doesn't currently hit that code path though, so most XML testing indeed
does not change.
Future patches will add that PCI address bits, which means most test cases
will have different output. So let's do away with the hardcoded same vs
different test split, and always track a separate output file. Tests can
still have same input and output, it just necessitates 2 separate XML files.
Commit 97e70a593589e8f0432a748b4b93c3cdc18b9e5a added the option -pie to
CFLAGS and LDFLAGS, however '-pie' is just a linker option. That
wouldn't be a problem. However, clang is checking for that and outputs
an error or unused argument:
error: argument unused during compilation: '-pie'
Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
Ján Tomko [Fri, 5 Feb 2016 14:44:19 +0000 (15:44 +0100)]
vboxDumpDisplay: add addDesktop bool
When FRONTEND/Type is not any of "sdl", "gui", "vrdp", we add a DESKTOP.
Use a bool to track this, instead of checking that both
totalPresent ("sdl" or "gui" present) and vrdpPresent are zero.
Ján Tomko [Thu, 14 Jan 2016 16:15:19 +0000 (17:15 +0100)]
leaseshelper: split out virLeaseNew
For the actions ADD and OLD, split out creating the new lease object,
as well as getting the environment variables that do not affect
the parsing of command line arguments.
Peter Krempa [Thu, 14 Jan 2016 09:38:02 +0000 (10:38 +0100)]
qemu: iothread: Aggregate code to set IOThread tuning
Rather than iterating 3 times for various settings this function
aggregates all the code into single place. One of the other advantages
is that it can then be reused for properly setting IOThread info on
hotplug.
Peter Krempa [Wed, 13 Jan 2016 16:00:54 +0000 (17:00 +0100)]
qemu: vcpu: Reuse qemuProcessSetupVcpu in vcpu hotplug
Since majority of the steps is shared, the function can be reused to
simplify code.
Additionally this resolves
https://bugzilla.redhat.com/show_bug.cgi?id=1244128 since the cpu
bandwidth limiting with cgroups would not be set on the hotplug path.
Additionally the code now sets the thread affinity and honors autoCpuset
as in the regular startup code path.
Peter Krempa [Wed, 13 Jan 2016 15:36:52 +0000 (16:36 +0100)]
qemu: vcpu: Aggregate code to set vCPU tuning
Rather than iterating 3 times for various settings this function
aggregates all the code into single place. One of the other advantages
is that it can then be reused for properly setting vCPU info on hotplug.
With this approach autoCpuset is also used when setting the process
affinity rather than just via cgroups.
Joao Martins [Fri, 5 Feb 2016 19:24:05 +0000 (19:24 +0000)]
remote: enforce VIR_TYPED_PARAM_STRING_OKAY flag on client side serialization
Commit 8cd1d54 consolidates both daemon and remote driver typed param
serialization functions. The consolidation now enforces client to use
VIR_TYPED_PARAM_STRING_OKAY flag to properly serialize string parameters, which
server has used for quite some time now. And this caused an issue, since the
commit had not adjusted client remote calls appropriately, thus causing a
failure in blkiotune, numatune and migration APIs (as per Xen CI tests). This
patch adjusts both remote_driver.c and gendispatch.pl to properly address this
issue.
In the commit 7938b533 we've changed the function signature,
however forgot to update stump that's used on systems without
CGroups causing a build failure.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Peter Krempa [Tue, 12 Jan 2016 12:12:05 +0000 (13:12 +0100)]
conf: Don't store vcpusched orthogonally to other vcpu info
Due to bad design the vcpu sched element is orthogonal to the way how
the data belongs to the corresponding objects. Now that vcpus are a
struct that allow to store other info too, let's convert the data to the
sane structure.
The helpers for the conversion are made universal so that they can be
reused for iothreads too.
This patch also resolves https://bugzilla.redhat.com/show_bug.cgi?id=1235180
since with the correct storage approach you can't have dangling data.
Peter Krempa [Tue, 15 Dec 2015 07:48:46 +0000 (08:48 +0100)]
qemu: Reuse qemuDomainDetectVcpuPids in cpu hot(un)plug
Now that qemuDomainDetectVcpuPids is able to refresh the vCPU pid
information it can be reused in the hotplug and hotunplug code paths
rather than open-coding a very similar algorithm.
A slight algorithm change is necessary for unplug since the vCPU needs
to be marked offline prior to calling the thread detector function and
eventually rolled back if something fails.
Peter Krempa [Thu, 4 Feb 2016 13:24:53 +0000 (14:24 +0100)]
conf: Move and optimize disk target duplicity checking
Move the logic from virDomainDiskDefDstDuplicates into
virDomainDiskDefCheckDuplicateInfo so that we don't have to loop
multiple times through the array of disks. Since the original function
was called in qemuBuildDriveDevStr, it was actually called for every
single disk which was quite wasteful.
Additionally the target uniqueness check needed to be duplicated in
the disk hotplug case, since the disk was inserted into the domain
definition after the device string was formatted and thus
virDomainDiskDefDstDuplicates didn't do anything in that case.
Peter Krempa [Thu, 4 Feb 2016 14:29:49 +0000 (15:29 +0100)]
tests: Integrate startup checks to qemuxml2argvtest
Some of the tests that are not a part of qemuBuildCommandLine were not
executed in the test suite. We can now reuse qemuProcessStartValidate to
integrate these tests.
Peter Krempa [Thu, 4 Feb 2016 14:25:29 +0000 (15:25 +0100)]
qemu: process: Extract pre-start checks into a function
When starting a qemu process there are certain checks done to ensure
that the configuration makes sense. Extract them into a separate
function so that they can be reused in the test code.
Peter Krempa [Fri, 5 Feb 2016 15:57:58 +0000 (16:57 +0100)]
qemu: snapshot: Avoid infinite loop if vCPUs can't be resumed
In b3d2a42e I've refactored the code and moved the 'cleanup' label.
Unfortunately the code that was originally in the 'endjob' label and
wanted to jump to cleanup is now in the cleanup label. Remove the jump
and let the function finish.
Peter Krempa [Fri, 5 Feb 2016 15:55:09 +0000 (16:55 +0100)]
qemu: snapshot: Don't overwrite existing errors when thawing filesystems
If we are attempting to thaw the filesystems on error, the code would
overwrite the error code that caused the snapshot to fail with the error
of thawing the filesystem. Since the thawing function allows control of
error reporting behavior we can use this feature.
Add a stub for nodeDeviceSysfsGetPCIRelatedDevCaps() for non-Linux
platforms. It allows nodedev driver to work on non-Linux platoforms
that, however, have HAL.
cppi: src/bhyve/bhyve_driver.h: line 26: not properly indented
cppi: src/bhyve/bhyve_driver.h: line 27: not properly indented
maint.mk: incorrect preprocessor indentation
After 1036ddadb276e we use bhyveDriverGetCapabilities from other
sources too, not only from bhyve_driver.c. However, the function
was static so not properly expose to other files. In order to
expose it, we need to move couple of #include-s too.
Then, there has been a copy paste error in
virBhyveProcessReconnect: s/privconn/data->driver/.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Cole Robinson [Wed, 13 Jan 2016 16:02:03 +0000 (11:02 -0500)]
tests: qemuargv2xml: separate from qemuxml2argv data
Most of the qemuargv2xml tests are parsing old style qemu command
lines (with -disk, -serial, etc), and it gets its input from
qemuxml2argv output.
But since we've raise the minimum supported qemu version to 0.12.0,
which supports -device, once that changes propagates through libvirt
the vast majority of qemuxml2argv output is _not_ going to be using
old style qemu options.
In preparation for this, switch qemuargv2xml to use its own copies
of input and output, so it's not tied to qemuxml2argv results.
This is just a straight copy of the current tests.
I've noticed that variable @reply is not initialized and if
something at the beginning of the function fails, e.g.
virDBusGetSystemBus(), the control jump straight to cleanup label
where dbus_message_unref() is then called over this uninitialized
variable.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
I've noticed couple of warning in dmesg while debugging
something:
[ 9683.973754] HTB: quantum of class 10001 is big. Consider r2q change.
[ 9683.976460] HTB: quantum of class 10002 is big. Consider r2q change.
I've read the HTB documentation and linux kernel code to find out
what's wrong. Basically we need to pass another argument
"quantum" to our tc cmd line because the default computed by HTB
does not always work in which case the warning message is printed
out.
Peter Krempa [Tue, 12 Jan 2016 12:11:56 +0000 (13:11 +0100)]
conf: Extract code for parsing thread resource scheduler info
As the scheduler info elements are represented orthogonally to how it
makes sense to actually store the information, the extracted code will
be later used when converting between XML and internal definitions.
So, systemd-machined has this philosophy that machine names are like
hostnames and hence should follow the same rules. But we always allowed
international characters in domain names. Thus we need to modify the
machine name we are passing to systemd.
In order to change some machine names that we will be passing to systemd,
we also need to call TerminateMachine at the end of a lifetime of a
domain. Even for domains that were started with older libvirt. That
can be achieved thanks to virSystemdGetMachineNameByPID(). And because
we can change machine names, we can get rid of the inconsistent and
pointless escaping of domain names when creating machine names.
So this patch modifies the naming in the following way. It creates the
name as <drivername>-<id>-<name> where invalid hostname characters are
stripped out of the name and if the resulting name is longer, it
truncates it to 64 characters. That way we can start domains we
couldn't start before. Well, at least on systemd.
To make it work all together, the machineName (which is needed only with
systemd) is saved in domain's private data. That way the generation is
moved to the driver and we don't need to pass various unnecessary
arguments to cgroup functions.
The only thing this complicates a bit is the scope generation when
validating a cgroup where we must check both old and new naming, so a
slight modification was needed there.
Joao Martins [Thu, 4 Feb 2016 22:55:05 +0000 (22:55 +0000)]
conf: add caps to virDomainSnapshotDefFormat
The virDomainSnapshotDefFormat calls into virDomainDefFormat,
so should be providing a non-NULL virCapsPtr instance. On the
qemu driver we change qemuDomainSnapshotWriteMetadata to also
include caps since it calls virDomainSnapshotDefFormat.
Joao Martins [Wed, 3 Feb 2016 21:40:37 +0000 (21:40 +0000)]
libxl: set net device prefix
Use the newly added virCapabilitiesSetNetPrefix to set
the network prefix for the driver. This in return will
be use by NetDefFormat() and NetDefParseXML() routines
to free any interface name that start with the registered
prefix.
Acked-by: Daniel P. Berrange <berrange@redhat.com> Signed-off-by: Joao Martins <joao.m.martins@oracle.com>
Joao Martins [Wed, 3 Feb 2016 21:40:36 +0000 (21:40 +0000)]
conf: add caps to virDomainSaveConfig
virDomainSaveConfig calls virDomainDefFormat which was setting the caps
to NULL, thus keeping the old behaviour (i.e. not looking at
netprefix). This patch adds the virCapsPtr to the function and allows
the configuration to be saved and skipping interface names that were
registered with virCapabilitiesSetNetPrefix().
Joao Martins [Wed, 3 Feb 2016 21:40:33 +0000 (21:40 +0000)]
conf: add net device prefix to capabilities
In the reverted commit d2e5538b1, the libxl driver was changed to copy
interface names autogenerated by libxl to the corresponding network def
in the domain's virDomainDef object. The copied name is freed when the
domain transitions to the shutoff state. But when migrating a domain,
the autogenerated name is included in the XML sent to the destination
host. It is possible an interface with the same name already exists on
the destination host, causing migration to fail.
This patch defines a new capability for setting the network device
prefix that will be used in the driver. Valid prefixes are
VIR_NET_GENERATED_PREFIX or the one announced by the driver.
ZFS-on-Linux implementation of ZFS starting with version 0.6.4
contains all the features we use. Additionally, as we support
'volmode' option handling that's not available on ZoL but is
available on FreeBSD, there is no need to block ZFS storage driver
on Linux anymore.