Commit Graph

22910 Commits

Author SHA1 Message Date
Peter Krempa
d071d292ca qemu: process: Call the domain config validator when starting a new VM
To avoid duplicating all the checks when starting a fresh VM from a
possibly unchecked config, call the domain def validator.
2016-06-07 13:02:20 +02:00
Peter Krempa
0d1c17aa68 qemu: process: Convert multiple boolean args to a single flag
Validation of qemu process startup requires to know whether the process
is used for a fresh VM or whether it's reloaded from a
snapshot/migration. Pass this information in via a flag rather than
calculating it from a bunch of bools.
2016-06-07 13:02:20 +02:00
Peter Krempa
43e21b1f10 qemu: process: Unexport qemuProcessStartValidate 2016-06-07 13:02:20 +02:00
Peter Krempa
9890a7a183 conf: Add device def validation callback
Similarly to the domain definition validator add a device validator. The
change to the prototype of the domain validator is necessary as
virDomainDeviceInfoIterateInternal requires a non-const pointer.
2016-06-07 13:02:20 +02:00
Peter Krempa
0fc1fde20b conf: drop 'def' from struct virDomainDefPostParseDeviceIteratorData
It's passed to all places along with the structure.
2016-06-07 13:02:20 +02:00
Peter Krempa
b394af162a conf: Add infrastructure for adding configuration validation
Until now we weren't able to add checks that would reject configuration
once accepted by the parser. This patch adds a new callback and
infrastructure to add such checks. In this patch all the places where
rejecting a now-invalid configuration wouldn't be a good idea are marked
with a new parser flag.
2016-06-07 13:02:20 +02:00
Peter Krempa
998c9e34e6 conf: Rename VIR_DOMAIN_DEF_PARSE_VALIDATE to VIR_DOMAIN_DEF_PARSE_VALIDATE_SCHEMA
Make it obvious that the flag is controlling RNG schema validation.
2016-06-07 13:02:20 +02:00
Ján Tomko
0c8fe3d457 node_device_udev: remove yoda condition 2016-06-07 12:53:37 +02:00
Ján Tomko
29c2a9cc21 udevSetupSystemDev: return if allocation fails
There is no cleanup to be done.
2016-06-07 12:53:37 +02:00
Ján Tomko
d1d4719814 Reformat udevProcessRemoveableMedia
Remove unnecessary ret variable and return early if we have no media
to save on indentation.
2016-06-07 12:53:37 +02:00
Ján Tomko
e2859b9dcd udevProcessStorage: trim all whitespace from model and vendor
Use virTrimSpaces instead of a custom implementation.
2016-06-07 12:53:37 +02:00
Ján Tomko
cc1d0e2a0e node_device_udev: switch to using virReportError
Also use the more common "Unable to initialize mutex" string
and virReportSystemError instead of virStrerror.
2016-06-07 12:53:37 +02:00
Ján Tomko
c40ed871cc Remove PROPERTY_* constants
They are no longer used.
2016-06-07 12:53:37 +02:00
Ján Tomko
4ab526ef71 Only return two values in udevGetUintSysfsAttr
Open code the call to udev_device_get_sysattr_value
in the one place where it's needed.
2016-06-07 12:53:37 +02:00
Ján Tomko
3554492a08 Only return two values in udevGetIntSysfsAttr
Callers only check for an error or a specific integer value.
2016-06-07 12:53:37 +02:00
Ján Tomko
e545806db9 Only return two values in udevGetStringSysfsAttr
The callers only care for an error, and a missing attribute
is simply NULL.
2016-06-07 12:53:37 +02:00
Ján Tomko
470498de18 Remove extra allocation in udevGetDeviceSysfsAttr
Most of the code paths free it right after converting it to
an integer.
2016-06-07 12:53:37 +02:00
Ján Tomko
10427db779 Only return two values in udevGetUintProperty
We only care about the failure, not a missing property.
2016-06-07 12:53:37 +02:00
Ján Tomko
71402ef6f5 Only return two values in udevGetStringProperty
There is no need to differentiate between PROPERTY_FOUND
and PROPERTY_MISSING - we can just look if the string is non-NULL.
2016-06-07 12:53:37 +02:00
Martin Kletzander
3484c05ce0 qemu: Add support to QXL's max_outputs parameter
Historically, we added heads=1 to videos, but for example for qxl, we
did not reflect that on the command line.

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1283207

Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
2016-06-07 12:42:54 +02:00
Martin Kletzander
6e81a1f9ef qemu: Check for qxl's max_outputs parameter
Add capabilities for both qxl and qxl-vga devices.

Signed-off-by: Martin Kletzander <mkletzan@redhat.com>
2016-06-07 12:40:01 +02:00
Ján Tomko
71cddab042 Rewrite disk type checking in udevProcessStorage
Error out on parsing errors and use a local const char pointer
instead of chained ifs to check whether we found a match.
2016-06-07 12:14:24 +02:00
Ján Tomko
4ccf6886c8 Fix the return value in udevKludgeStorageType
Since the switch to VIR_STRDUP this function returns 1 on success,
but the caller treats any non-zero value as failure.
2016-06-07 12:13:59 +02:00
Ján Tomko
df7291c31c udevProcessFloppy; remove unnecessary allocation
Use udevHasDeviceProperty instead of udevGetStringProperty.
We do not need to copy the string since we do not need it.

Also add braces around the if body, since the change made
syntax check complain.
2016-06-07 12:13:38 +02:00
Ján Tomko
61cafffb2f Move udevHasDeviceProperty earlier 2016-06-07 12:13:33 +02:00
Ján Tomko
0d372687eb Do not VIR_STRDUP the string in udevGetDeviceProperty
Two out of three callers free it right after converting it to a number.

Also change the comment at the beginning of the function, because
the comment inside the function told me to.
2016-06-07 12:00:36 +02:00
Ján Tomko
fde3a38e03 Remove udevStrToLong_i
Open code the error message.
2016-06-07 12:00:36 +02:00
Ján Tomko
07202bf4cf Remove udevStrToLong_ui
Remove the debug message, open code the error in the two udevGetUint
callers and use a more specific error in SCSI and PCI processing.
2016-06-07 12:00:36 +02:00
Ján Tomko
1a97fb16ae Remove udevStrToLong_ull
The wrapper adds an error message or a debug log.

Since we already log the properties we get from udev as strings,
there is no much use for the debug logs.

Open code the error message and delete the function.
2016-06-07 12:00:36 +02:00
Ján Tomko
876a5da03e Rewrite usage of StrToLong_ui in udevProcess{PCI,SCSI}
Use virStrToLong_ui instead of udevStrToLong_ui, reformat the code
and report a more specific error message.
2016-06-07 12:00:36 +02:00
Ján Tomko
3775a2e174 udevProcessSCSIHost: use STRSKIP
Instead of separating it into STRPEFIX and str + strlen.
2016-06-07 10:51:36 +02:00
Ján Tomko
170c68c4a8 udevGetDMIData: remove unused variable
A variable without use is pointless.

Remove it, since we have no use for it.
2016-06-07 10:51:36 +02:00
Ján Tomko
fa89f6c4a3 Assign node device driver private data earlier 2016-06-07 10:51:36 +02:00
Ján Tomko
c8afb4e1ec Do not call nodeStateCleanup on early initialization error
If we have not allocated driver yet, there is nothing to cleanup.
2016-06-07 10:51:36 +02:00
Ján Tomko
d85e29dd82 Reformat nodeStateCleanup
Remove the ret variable and return early if there is no driver.
2016-06-07 10:51:36 +02:00
Ján Tomko
5c88b34afa node_device_udev: initialize libpciaccess after the driver lock
This will simplify cleanup.
2016-06-07 10:51:36 +02:00
Ján Tomko
feb876660e Split out pciaccess (de)initialization
Move pci_system_init and pci_system_cleanup into separate functions,
to make the conditional compilation easier to read.
2016-06-07 10:51:36 +02:00
Ján Tomko
b2a55dfd1f Initialize ret to -1 in nodeStateInitialize
Most of the code paths had to reset it to -1 and returning 0 was
only possible if we made it to the end of the function.

Initialize it to -1 and only set it to 0 if we reach the end, as we do
in most of libvirt code.
2016-06-07 10:51:36 +02:00
Daniel P. Berrange
c0bc172383 systemd: directly notify systemd instead of using sd_notify
The sd_notify method is used to tell systemd when libvirtd
has finished starting up. All it does is send a datagram
containing the string parameter to systemd on a UNIX socket
named in the NOTIFY_SOCKET environment variable. Rather than
pulling in the systemd libraries for this, just code the
notification directly in libvirt as this is a stable ABI
from systemd's POV which explicitly allows independant
implementations:

See "Reimplementable Independently" column in the
"$NOTIFY_SOCKET Daemon Notifications" row:

https://www.freedesktop.org/wiki/Software/systemd/InterfacePortabilityAndStabilityChart/

Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1314881

Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
2016-06-07 09:37:57 +01:00
John Ferlan
1b5f1884a2 qemu: Move and rename qemuBuildObjectCommandlineFromJSON
Move the module from qemu_command.c to a new module virqemu.c and
rename the API to virQEMUBuildObjectCommandline.

This API will then be shareable with qemu-img and the need to build
a security object for luks support.

Signed-off-by: John Ferlan <jferlan@redhat.com>
2016-06-06 13:44:51 -04:00
John Ferlan
77ad76b615 storage: Create helper to set input for CreateQemuImg code
Create helper virStorageBackendCreateQemuImgSetInput to set the input

Signed-off-by: John Ferlan <jferlan@redhat.com>
2016-06-06 12:15:43 -04:00
John Ferlan
4c6038a35e storage: Split out a helper for encryption checks
Split out a helper from virStorageBackendCreateQemuImgCmdFromVol
to check the encryption - soon a new encryption sheriff will be
patroling and that'll mean all sorts of new checks.

Signed-off-by: John Ferlan <jferlan@redhat.com>
2016-06-06 11:41:45 -04:00
John Ferlan
a2a7f7ede8 storage: Split out setting default secret for encryption
Split the qcow setting of encryption secrets into a helper

Signed-off-by: John Ferlan <jferlan@redhat.com>
2016-06-06 11:41:45 -04:00
John Ferlan
1933b87855 util: Clean up code formatting in virstorageencryption
Bring style more in line with more recent code.

Signed-off-by: John Ferlan <jferlan@redhat.com>
2016-06-06 11:41:45 -04:00
Michal Privoznik
cc874efb5b docs: Document our event loop
I was asked the other day what's event loop and how libvirt uses
it. Well, I haven't found any good sources on the Internet so I
thought of writing the documentation on my own.

Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
2016-06-06 17:35:44 +02:00
Ján Tomko
8c40ede4a4 Do not check for domain liveness in virDomainObjSetDefTransient
Remove the live attribute and mark the definition as transient
whether the domain is runing or not.

There were only two callers left calling with live=false:
* testDomainStartState, where the domain already is active
  because we assigned vm->def->id just a few lines above the call
* virDomainObjGetPersistentDef, which now only calls
  virDomainObjSetDefTransient for an active domain
2016-06-06 08:34:22 +02:00
Ján Tomko
1dcb1dc0f9 Check if the domain is active in virDomainObjGetPersistentDef
Calling virDomainObjSetDefTransient with live=false is a no-op
on an inactive domain.

Only call it on an active domain, since this is the only place using
the live bool.
2016-06-06 08:34:22 +02:00
Ján Tomko
9b111048ad Clean up redundant usage of virDomainObjSetDefTransient
Commit 45ec297d from November 2010:
    Make state driver device hotplug/update actually transient
added virDomainObjSetDefTransient calls to the domain startup
function in several drivers.

In November 2011, commit 8866eed:
    Set aliases for LXC/UML console devices
added a call earlier in the startup function, without removing the
existing ones.

Also, in the UML driver it seems the function never did anything
useful - vm->def->id is set asynchronnously in umlNotifyEvent.
At the time of calling virDomainObjSetDefTransient with live=false,
vm->def->id was likely still -1, making the call a no-op.
2016-06-06 08:34:22 +02:00
Ján Tomko
2ccacdb3ec Post-release version bump to 1.3.6 2016-06-06 08:34:22 +02:00
Daniel Veillard
7324444bc7 Release of libvirt-1.3.5
* docs/news.html.in: update with 1.3.5 data
2016-06-04 23:29:59 +08:00