On Tue, 01/28 06:58, Eric Blake wrote:
On 01/28/2014 04:14 AM, Paolo Bonzini wrote:
Let's see the feedback of Eric.
Eric's feedback is certainly useful, but I think we need to look at it
from the QEMU perspective more than the libvirt perspective.
Passing the raw schema and
On 01/28/2014 07:37 PM, Michal Privoznik wrote:
There are some units within libvirt that utilize virCommand API to run
some commands and deserve own unit testing. These units are, however,
not desired to be rewritten to dig virCommand API usage out. As a great
example virNetDevBandwidth could
On 01/28/2014 07:37 PM, Michal Privoznik wrote:
The test tries to set some QoS limits and check if the commands
that are actually executed are the expected ones.
Signed-off-by: Michal Privoznik mpriv...@redhat.com
---
tests/virnetdevbandwidthtest.c | 76
on 2014/01/28 03:27, Thorsten Behrens wrote:
Hi,
Hi,
Sorry, I'm out of office now, If nobody gives you review or comments,
I will pick this up when I am back on Feb 10.
Thanks!
sorry for the noise - but I don't see the patch series pushed yet?
Gao feng wrote:
[...]
ACK with these
On Tue, Jan 28, 2014 at 07:00:16PM -0500, John Ferlan wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=1045124
When loading modules, libvirt does not honor the modprobe blacklist.
By adding a -b to the modprobe command libvirt will fail to load a
module if it's on the blacklist
Check if
On 01/29/2014 05:40 AM, Daniel P. Berrange wrote:
On Tue, Jan 28, 2014 at 07:00:16PM -0500, John Ferlan wrote:
This is getting complex enough that I think we ought introduce
a src/util/virkmod.{c,h} and have some helper APIs in there
for loading and unload modules. eg virKModLoad,
On Wed, Jan 29, 2014 at 06:32:42AM -0500, John Ferlan wrote:
On 01/29/2014 05:40 AM, Daniel P. Berrange wrote:
On Tue, Jan 28, 2014 at 07:00:16PM -0500, John Ferlan wrote:
This is getting complex enough that I think we ought introduce
a src/util/virkmod.{c,h} and have some helper
On Wed, Jan 29, 2014 at 10:15:36AM +0800, Qiao Nuohan wrote:
This patch is used to add dump_memory_format to qemu.conf and make the
specified format as the default dump format of 'virsh dump --memory-only'.
But when --compress is specified with dump command, the format configured
by
On Tue, Jan 28, 2014 at 02:08:26PM +0100, Christophe Fergeau wrote:
It's possible to create a handle to watch for file events which do not
watch for any file event. Such a handle can be enabled later with
gvir_event_handle_update() by setting some conditions to watch for.
When a handle is
On Tue, Jan 28, 2014 at 02:08:27PM +0100, Christophe Fergeau wrote:
Trying to remove a disabled timer or handle will cause
gvir_{event,timer}_handle_remove() to return an error
rather than removing it.
---
libvirt-glib/libvirt-glib-event.c | 18 +-
1 file changed, 9
On Tue, Jan 28, 2014 at 02:08:28PM +0100, Christophe Fergeau wrote:
gvir_event_handle_add() creates a GIOChannel in order to watch the fd it
was given for changes.
gvir_event_handle_remove() is freeing all the resources allocated by
gvir_event_handle_add() except for this GIOChannel. This
On Tue, Jan 28, 2014 at 02:13:18PM +0100, Christophe Fergeau wrote:
diff --git a/tests/test-gconfig.c b/tests/test-gconfig.c
new file mode 100644
index 000..bc9aac0
--- /dev/null
+++ b/tests/test-gconfig.c
@@ -0,0 +1,674 @@
+/*
+ * test-gconfig.c: unit tests for libvirt-gconfig
+ *
+
On Tue, Jan 28, 2014 at 02:13:18PM +0100, Christophe Fergeau wrote:
Rather than the ever growing standalone
libvirt-gconfig/tests/test-domain-create
program, this commit adds support for glib GTester framework as well as 2
smaller tests cases for GVirConfigDomain and GVirConfigDomainClock.
On Tue, Jan 28, 2014 at 02:13:20PM +0100, Christophe Fergeau wrote:
---
.gitignore | 8
1 file changed, 8 insertions(+)
diff --git a/.gitignore b/.gitignore
index 712778f..3eabf64 100644
--- a/.gitignore
+++ b/.gitignore
@@ -7,6 +7,7 @@ build-aux/depcomp
build-aux/install-sh
On Tue, Jan 28, 2014 at 12:07 PM, Daniel P. Berrange berra...@redhat.comwrote:
On Thu, Jan 23, 2014 at 03:14:19PM -0500, Adam Walters wrote:
This patchset adds a driver named 'config' which provides access to
configuration data, such as secret and storage definitions, without
requiring a
On Tue, Jan 28, 2014 at 02:13:19PM +0100, Christophe Fergeau wrote:
It's currently only testing removal of disabled timer/watches
---
tests/Makefile.am | 6 +-
tests/test-events.c | 172
2 files changed, 176 insertions(+), 2
On 01/27/14 23:12, Eric Blake wrote:
On 01/16/2014 06:14 AM, Peter Krempa wrote:
Add support for specifying various types when doing snapshots. This will
later allow to do snapshots on network backed volumes. Disks of type
'volume' are not supported by snapshots (yet).
---
Notes:
On Wed, Jan 29, 2014 at 07:01:31AM -0500, Adam Walters wrote:
On Tue, Jan 28, 2014 at 12:07 PM, Daniel P. Berrange
berra...@redhat.comwrote:
On Thu, Jan 23, 2014 at 03:14:19PM -0500, Adam Walters wrote:
This patchset adds a driver named 'config' which provides access to
configuration
Hey,
On Wed, Jan 29, 2014 at 12:00:22PM +, Daniel P. Berrange wrote:
On Tue, Jan 28, 2014 at 02:13:18PM +0100, Christophe Fergeau wrote:
diff --git a/tests/test-gconfig.c b/tests/test-gconfig.c
new file mode 100644
index 000..bc9aac0
--- /dev/null
+++ b/tests/test-gconfig.c
On Wed, Jan 29, 2014 at 01:43:50PM +0100, Christophe Fergeau wrote:
Hey,
On Wed, Jan 29, 2014 at 12:00:22PM +, Daniel P. Berrange wrote:
On Tue, Jan 28, 2014 at 02:13:18PM +0100, Christophe Fergeau wrote:
diff --git a/tests/test-gconfig.c b/tests/test-gconfig.c
new file mode 100644
On Mon, Jan 27, 2014 at 04:53:53PM -0500, Stefan Berger wrote:
On 01/27/2014 12:18 PM, Daniel P. Berrange wrote:
The NWFilter code has as a deadlock race condition between
the virNWFilter{Define,Undefine} APIs and starting of guest
VMs due to mis-matched lock ordering.
In the
On 01/28/2014 11:49 AM, Michal Privoznik wrote:
On 26.01.2014 19:24, Cole Robinson wrote:
I've just pushed a new tool to virt-manager.git called virt-xml. virt-xml
uses
virt-install's command line options to allow building and editing domain XML.
Best way to describe what it can do is with a
On Wed, Jan 29, 2014 at 01:25:56PM +, Daniel P. Berrange wrote:
Yes, I'd rather prefer it to be LGPLv2+ licensed because IME it is not
uncommon to copy code from the library into the test suite and vica-verca.
So you want a consistent license that allows this bi-directional copying
IMHO
On 01/23/2014 03:44 PM, Michal Privoznik wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=1055484
Currently, libvirt's XML schema of network allows QoS to be defined for
every network even though it has no bridge. For instance:
network
namevdsm-no-bridge/name
forward
On 01/27/2014 12:10 PM, Dan Kenigsberg wrote:
On Fri, Jan 24, 2014 at 02:18:27PM +0200, Laine Stump wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=1057321 pointed out that
we weren't honoring the bandwidth element in libvirt networks using
forward mode='bridge'/. In fact, these networks
Since we introduced fine grained locking into the QEMU driver so
that VM start can run in parallel, we appear to have caused a
race with the nwfilter code. In particular since we no longer
hold the global QEMU driver lock for the duration of VM startup
we have a lock ordering flaw. This results in
Add virRWLock backed up by a POSIX rwlock primitive
Signed-off-by: Daniel P. Berrange berra...@redhat.com
---
src/libvirt_private.syms| 5 +
src/util/virthread.h| 10 ++
src/util/virthreadpthread.c | 33 +
src/util/virthreadpthread.h | 4
On Fedora 19 and older the pthreads impl provided with
mingw does not have any pthread_sigmask impl at all. The
configure.ac check was not distinguishing this scenario
from that of a broken pthread_sigmask impl, so was
mistakenly enabling the libvirt workaround even when it
was not needed. This in
The NWFilter code has as a deadlock race condition between
the virNWFilter{Define,Undefine} APIs and starting of guest
VMs due to mis-matched lock ordering.
In the virNWFilter{Define,Undefine} codepaths the lock ordering
is
1. nwfilter driver lock
2. virt driver lock
3. nwfilter update
There are a number of pthreads impls available on Win32
these days, in particular the mingw64 project has a good
impl. Delete the native windows thread implementation and
rely on using pthreads everywhere.
Signed-off-by: Daniel P. Berrange berra...@redhat.com
---
configure.ac|
Chegu Vinod wrote:
Busy enterprise workloads hosted on large sized VM's tend to dirty
memory faster than the transfer rate achieved via live guest migration.
Despite some good recent improvements ( using dedicated 10Gig NICs
between hosts) the live migration may NOT converge.
Recently
On 1/29/2014 8:27 AM, Jim Fehlig wrote:
Chegu Vinod wrote:
Busy enterprise workloads hosted on large sized VM's tend to dirty
memory faster than the transfer rate achieved via live guest migration.
Despite some good recent improvements ( using dedicated 10Gig NICs
between hosts) the live
It doesn't make sense to fail if the SCSI host device is specified
as shareable explicitly between domains (NB, it works if and only
if the device is specified as shareable for *all* domains,
otherwise it fails).
To fix the problem, this patch introduces an array for virSCSIDevice
struct, which
To support to pass the path of the test data to the utils, one
more argument is added to virSCSIDeviceGetSgName,
virSCSIDeviceGetDevName, and virSCSIDeviceNew, and the related
code is changed accordingly.
Signed-off-by: Osier Yang jy...@redhat.com
---
src/qemu/qemu_cgroup.c | 3 +-
*** BLURB HERE ***
Osier Yang (2):
qemu: Don't fail if the SCSI host device is shareable between domains
tests: Add tests for scsi utils
docs/formatdomain.html.in| 5 ++
src/libvirt_private.syms | 2 +-
src/qemu/qemu_cgroup.c | 3 +-
src/qemu/qemu_command.c
This reverts commit 2996e6be19a13199ded7c2aa21039cca97318e01
and some parts of 2636dc8c4de83cd37bc0680a6fbc3f6d25023bd7.
Conflicts:
tests/virnetdevbandwidthtest.c: New test has been introduced since
then.
Signed-off-by: Michal Privoznik mpriv...@redhat.com
---
On 29.01.2014 15:26, Laine Stump wrote:
On 01/23/2014 03:44 PM, Michal Privoznik wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=1055484
Currently, libvirt's XML schema of network allows QoS to be defined for
every network even though it has no bridge. For instance:
network
On Wed, Jan 29, 2014 at 06:44:22PM +0100, Michal Privoznik wrote:
This reverts commit 2996e6be19a13199ded7c2aa21039cca97318e01
and some parts of 2636dc8c4de83cd37bc0680a6fbc3f6d25023bd7.
Conflicts:
tests/virnetdevbandwidthtest.c: New test has been introduced since
then.
Be
This patch adds some new RPC call numbers, but for ease of review,
they sit idle until a later patch adds the client counterpart to
drive the new RPCs. Also for ease of review, I limited this patch
to just the lifecycle event; although converting the remaining
15 domain events will be quite
This patch continues the earlier conversion made for network
events, with a goal of introducing server-side event filtering
in a later patch. Actual behavior is unchanged without
further RPC changes.
* daemon/libvirtd.h (daemonClientPrivate): Alter the tracking of
domain events.
*
Commit 57ddcc23 (v0.9.11) introduced the pmwakeup event, with
an optional 'reason' field reserved for possible future expansion.
But it failed to wire the field through RPC, so even if we do
add a reason in the future, we will be unable to get it back
to the user.
Worse, commit 7ba5defb (v1.0.0)
v1 was here:
https://www.redhat.com/archives/libvir-list/2014-January/msg00689.html
Since then, I've rebased it on top of this patch which still
needs review:
https://www.redhat.com/archives/libvir-list/2014-January/msg01406.html
as well as added patch 6 so that we don't have to repeat the
We want to convert over to server-side events, even for older
APIs. To do that, the client side of the remote driver wants
to distinguish between legacy virConnectDomainEventRegister and
normal virConnectDomainEventRegisterAny, while knowing the
client callbackID and the server's serverID for
The counterpart to the server RPC additions; here, a single
function can serve both old and new calls, while incoming
events must be serviced by two different functions. Again,
some wise choices in our XDR makes event sharing easier.
While this only supports lifecycle events, it covers the
On 29.01.2014 18:48, Daniel P. Berrange wrote:
On Wed, Jan 29, 2014 at 06:44:22PM +0100, Michal Privoznik wrote:
This reverts commit 2996e6be19a13199ded7c2aa21039cca97318e01
and some parts of 2636dc8c4de83cd37bc0680a6fbc3f6d25023bd7.
Conflicts:
tests/virnetdevbandwidthtest.c: New test
On Wed, Jan 29, 2014 at 07:03:10PM +0100, Michal Privoznik wrote:
On 29.01.2014 18:48, Daniel P. Berrange wrote:
On Wed, Jan 29, 2014 at 06:44:22PM +0100, Michal Privoznik wrote:
This reverts commit 2996e6be19a13199ded7c2aa21039cca97318e01
and some parts of
Check for presence of sys/cpuset.h header and cpuset_getaffinity()
in configure instead of just using #ifdef __FreeBSD__ for that code.
---
configure.ac | 10 ++
src/util/virprocess.c | 7 +--
2 files changed, 15 insertions(+), 2 deletions(-)
diff --git a/configure.ac
https://bugzilla.redhat.com/show_bug.cgi?id=1038363
If a domain has a different maximum for persistent and live maxmem
or max vcpus, then it is possible to hit cases where libvirt
refuses to adjust the current values or gets halfway through
the adjustment before failing. Better is to determine
On 01/29/2014 11:31 AM, Roman Bogorodskiy wrote:
Check for presence of sys/cpuset.h header and cpuset_getaffinity()
in configure instead of just using #ifdef __FreeBSD__ for that code.
---
configure.ac | 10 ++
src/util/virprocess.c | 7 +--
2 files changed, 15
On 01/29/2014 10:22 AM, Osier Yang wrote:
It doesn't make sense to fail if the SCSI host device is specified
as shareable explicitly between domains (NB, it works if and only
if the device is specified as shareable for *all* domains,
otherwise it fails).
To fix the problem, this patch
On 01/29/2014 10:22 AM, Osier Yang wrote:
To support to pass the path of the test data to the utils, one
s/to pass/passing/
more argument is added to virSCSIDeviceGetSgName,
virSCSIDeviceGetDevName, and virSCSIDeviceNew, and the related
code is changed accordingly.
Signed-off-by: Osier
On 01/29/2014 09:53 AM, Daniel P. Berrange wrote:
The NWFilter code has as a deadlock race condition between
the virNWFilter{Define,Undefine} APIs and starting of guest
VMs due to mis-matched lock ordering.
In the virNWFilter{Define,Undefine} codepaths the lock ordering
is
1. nwfilter
On 01/29/2014 07:53 AM, Daniel P. Berrange wrote:
On Fedora 19 and older the pthreads impl provided with
mingw does not have any pthread_sigmask impl at all. The
configure.ac check was not distinguishing this scenario
from that of a broken pthread_sigmask impl, so was
mistakenly enabling the
Adding tests for new virModprobe{Config|Load|Unload|UseBlacklist} API's.
Other than the virModprobeConfig() calls, the tests require some special
set up and to be run from root.
Signed-off-by: John Ferlan jfer...@redhat.com
---
.gitignore | 2 +
tests/Makefile.am | 5 ++
Original patch here:
http://www.redhat.com/archives/libvir-list/2014-January/msg01408.html
Changes since v1:
* Add virkmod.{c,h} to perform the various modprobe commands
virModprobeConfig() - modprobe -c
virModprobeLoad() - modprobe driver
virModprobeUnload() -
This patch adds functions for various usages of modprobe
Signed-off-by: John Ferlan jfer...@redhat.com
---
src/Makefile.am | 1 +
src/libvirt_private.syms | 7 +++
src/util/virkmod.c | 138 +++
src/util/virkmod.h | 35
https://bugzilla.redhat.com/show_bug.cgi?id=1045124
When loading modules, libvirt does not honor the modprobe blacklist.
By adding a -b to the modprobe command libvirt will fail to load a
module if it's on the blacklist
Check if the failure to load a driver was due to it being on the blacklist
On 30/01/14 03:31, Eric Blake wrote:
On 01/29/2014 10:22 AM, Osier Yang wrote:
It doesn't make sense to fail if the SCSI host device is specified
as shareable explicitly between domains (NB, it works if and only
if the device is specified as shareable for *all* domains,
otherwise it fails).
To
On 30/01/14 03:35, Eric Blake wrote:
On 01/29/2014 10:22 AM, Osier Yang wrote:
To support to pass the path of the test data to the utils, one
s/to pass/passing/
Changed.
more argument is added to virSCSIDeviceGetSgName,
virSCSIDeviceGetDevName, and virSCSIDeviceNew, and the related
code
59 matches
Mail list logo
