On 09/17/2014 11:05 PM, John Ferlan wrote:
> On 09/17/2014 06:45 AM, Ján Tomko wrote:
>> Commit f36a94f introduced a double free on all success paths
>> in qemuSharedDeviceEntryInsert.
>>
>> Only call qemuSharedDeviceEntryFree on the error path and
>> set entry to NULL before jumping there if the e
This internal API can be used to allocate or free some pages in
the huge pages pool.
Signed-off-by: Michal Privoznik
---
src/libvirt_private.syms | 1 +
src/util/virnuma.c | 108 +++
src/util/virnuma.h | 4 ++
3 files changed, 113 inser
And add stubs to other drivers like: lxc, qemu, uml and vbox.
Signed-off-by: Michal Privoznik
---
src/libvirt_private.syms | 1 +
src/lxc/lxc_driver.c | 22 ++
src/nodeinfo.c | 29 +
src/nodeinfo.h | 7 +++
src/qemu/qe
This may seem like a cherry on top of the cake, but once we allow
guests to use huge pages we must allow admins to allocate ones.
Michal Privoznik (4):
Introduce virNodeAllocPages
virnuma: Introduce virNumaSetPagePoolSize
nodeinfo: Implement nodeAllocPages
virsh: Expose virNodeAllocPages
The new virsh command is named 'allocpages'.
Signed-off-by: Michal Privoznik
---
tools/virsh-host.c | 134 +
tools/virsh.pod| 12 +
2 files changed, 146 insertions(+)
diff --git a/tools/virsh-host.c b/tools/virsh-host.c
index 7fc2120.
A long time ago in a galaxy far, far away it has been decided
that libvirt will manage not only domains but host as well. And
with my latest work on qemu driver supporting huge pages, we miss
the cherry on top: an API to allocate huge pages on the run.
Currently users are forced to log into the hos
On Wed, Sep 17, 2014 at 04:24:07PM -0600, Eric Blake wrote:
> Any objections to retiring the v0.9.6-maint branch? After all, we have
> already retired the v0.9.11-maint branch
> (http://libvirt.org/git/?p=libvirt.git;a=commit;h=cd0d348ed), and the
> only activity on v0.9.6-maint since 0.9.6.4 was
Cleanup virDomanDef structure from other nested structure and create
separate type definition for them.
Fix a typo in virDomainHugePage.
Signed-off-by: Pavel Hrdina
---
src/conf/domain_conf.h | 102 +
1 file changed, 61 insertions(+), 41 deletions
This patch series introduce new tunable event to inform management
applications about changes of tunable values. With this universal
event we will be able to report updates for all different tunable
values like cpu tuning, block tinning, memory tinning, etc...
There is missing documentation for al
This new event will use typedParameters to expose what has been actually
updated and the reason is that we can in the future extend any tunable
values or add new tunable values. With typedParameters we don't have to
worry about creating some other events, we will just use this universal
event to in
Signed-off-by: Pavel Hrdina
---
examples/object-events/event-test.c | 52 -
1 file changed, 51 insertions(+), 1 deletion(-)
diff --git a/examples/object-events/event-test.c
b/examples/object-events/event-test.c
index d6cfe46..9e09736 100644
--- a/examples/obj
On Wed, Sep 17, 2014 at 03:07:47PM -0400, John Ferlan wrote:
> While doing some investigation for another bug I found that I could
> not qemu-attach to the process and got the following:
>
>error: Operation not supported: JSON monitor is required
>
> while running thru qemuProcessAttach. Sinc
Now we have universal tunable event so we can use it for reporting
changes to user. The cputune values will be prefixed with "cputune" to
distinguish it from other tunable events.
Signed-off-by: Pavel Hrdina
---
src/qemu/qemu_cgroup.c | 18 +++-
src/qemu/qemu_driver.c | 76 ++
On 18.09.2014 08:47, Roman Bogorodskiy wrote:
Commit b20d39a introduced a new argument for the
virNetDevTapCreateInBridgePort function, however, its mock
in bhyve tests wasn't updated, so the build failed.
Fix build by adding this new argument to the mock version.
---
tests/bhyvexml2argvmock.c
Michal Privoznik wrote:
> On 18.09.2014 08:47, Roman Bogorodskiy wrote:
> > Commit b20d39a introduced a new argument for the
> > virNetDevTapCreateInBridgePort function, however, its mock
> > in bhyve tests wasn't updated, so the build failed.
> >
> > Fix build by adding this new argument to the
Add a new parameter to virStorageFileGetMetadata that will break the
backing chain detection process and report useful error message rather
than having to use virStorageFileChainGetBroken.
This patch just introduces the option, usage will be provided
separately.
---
src/qemu/qemu_domain.c
New version incorporates some review feedback from John. The changes were
borderline-trivial so I've reposted the series.
Patch 2/4 is new.
Peter Krempa (4):
util: storage: Allow metadata crawler to report useful errors
qemu: Sanitize argument names and empty disk check in
qemuDomainDete
Request erroring out from the backing chain traveller and drop qemu's
internal backing chain integrity tester.
The backin chain traveller reports errors by itself with possibly more
detail than qemuDiskChainCheckBroken ever could.
We also need to make sure that we reconnect to existing qemu insta
Reuse virStorageSourceIsEmpty and rename "force" argument to
"force_probe".
---
src/qemu/qemu_domain.c | 8 +++-
src/qemu/qemu_domain.h | 2 +-
2 files changed, 4 insertions(+), 6 deletions(-)
diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c
index 19b935d..515bcac 100644
--- a/src
Report also the name of the parent file and uid/gid used to access it to
help debugging broken storage configurations.
---
src/storage/storage_driver.c | 20 +++-
1 file changed, 15 insertions(+), 5 deletions(-)
diff --git a/src/storage/storage_driver.c b/src/storage/storage_drive
Ján Tomko (4):
Fix leak in x86UpdateHostModel
Fixes for domains with no iothreads
audit: remove redundant NULL assignment
audit: fix memory leak without WITH_AUDIT
src/cpu/cpu_x86.c | 4 +++-
src/qemu/qemu_cgroup.c | 2 +-
src/qemu/qemu_process.c | 6 --
src/util/viraudit.c
Commit de0aeaf introduced a memory leak.
---
src/cpu/cpu_x86.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/src/cpu/cpu_x86.c b/src/cpu/cpu_x86.c
index 7571f16..a98a847 100644
--- a/src/cpu/cpu_x86.c
+++ b/src/cpu/cpu_x86.c
@@ -2094,8 +2094,10 @@ x86UpdateHostModel(virCPU
Plug a memory leak and silence a warning.
---
src/qemu/qemu_cgroup.c | 2 +-
src/qemu/qemu_process.c | 6 --
2 files changed, 5 insertions(+), 3 deletions(-)
diff --git a/src/qemu/qemu_cgroup.c b/src/qemu/qemu_cgroup.c
index 9d39370..7c6b2c1 100644
--- a/src/qemu/qemu_cgroup.c
+++ b/src/qemu
Free str unconditionally since we allocate it without WITH_AUDIT
too.
---
src/util/viraudit.c | 9 ++---
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/src/util/viraudit.c b/src/util/viraudit.c
index d0ad9b9..23928fd 100644
--- a/src/util/viraudit.c
+++ b/src/util/viraudit.c
@@
virVasprintf sets the output to NULL on failure.
---
src/util/viraudit.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/src/util/viraudit.c b/src/util/viraudit.c
index 8023c60..d0ad9b9 100644
--- a/src/util/viraudit.c
+++ b/src/util/viraudit.c
@@ -99,10 +99,8 @@ void virAud
On 09/18/2014 04:39 AM, Daniel P. Berrange wrote:
> On Wed, Sep 17, 2014 at 03:07:47PM -0400, John Ferlan wrote:
>> While doing some investigation for another bug I found that I could
>> not qemu-attach to the process and got the following:
>>
>>error: Operation not supported: JSON monitor is
On Thu, Sep 18, 2014 at 06:18:22AM -0400, John Ferlan wrote:
>
>
> On 09/18/2014 04:39 AM, Daniel P. Berrange wrote:
> > On Wed, Sep 17, 2014 at 03:07:47PM -0400, John Ferlan wrote:
> >> While doing some investigation for another bug I found that I could
> >> not qemu-attach to the process and go
On 09/18/14 12:14, Ján Tomko wrote:
> virVasprintf sets the output to NULL on failure.
> ---
> src/util/viraudit.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
ACK
Peter
signature.asc
Description: OpenPGP digital signature
--
libvir-list mailing list
libvir-list@redhat.com
http
On 09/18/14 12:14, Ján Tomko wrote:
> Commit de0aeaf introduced a memory leak.
> ---
> src/cpu/cpu_x86.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
ACK,
Peter
signature.asc
Description: OpenPGP digital signature
--
libvir-list mailing list
libvir-list@redhat.com
https://ww
On 09/18/14 12:14, Ján Tomko wrote:
> Plug a memory leak and silence a warning.
> ---
> src/qemu/qemu_cgroup.c | 2 +-
> src/qemu/qemu_process.c | 6 --
> 2 files changed, 5 insertions(+), 3 deletions(-)
>
ACK
signature.asc
Description: OpenPGP digital signature
--
libvir-list mailing li
On 09/18/14 12:14, Ján Tomko wrote:
> Free str unconditionally since we allocate it without WITH_AUDIT
> too.
> ---
> src/util/viraudit.c | 9 ++---
> 1 file changed, 2 insertions(+), 7 deletions(-)
>
ACK
signature.asc
Description: OpenPGP digital signature
--
libvir-list mailing list
li
When trying to remove nonexistent metadata from XML, libvirt daemon
crashes due to dereferencing NULL pointer.
Resolves https://bugzilla.redhat.com/show_bug.cgi?id=1143955
---
src/util/virxml.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/src/util/virxml.c b/src/util/vi
On 09/18/14 14:25, Erik Skultety wrote:
> When trying to remove nonexistent metadata from XML, libvirt daemon
> crashes due to dereferencing NULL pointer.
>
> Resolves https://bugzilla.redhat.com/show_bug.cgi?id=1143955
> ---
> src/util/virxml.c | 5 -
> 1 file changed, 4 insertions(+), 1 del
Yes, all of these patches below are from me.
On 2014/9/18 1:21, Eric Blake wrote:
> On 09/16/2014 07:19 AM, James wrote:
>> I clean up all _virDomainMemoryStat.
>>
>> Signed-off-by: James
>> Signed-off-by: Wang Rui
>> ---
>> daemon/remote.c | 2 +-
>> src/driver.h
If you have a bridge network in running domain and libvirtd is restarted
the information about host bridge interface is lost from live xml.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1140085
Signed-off-by: Pavel Hrdina
---
src/conf/domain_conf.c | 4
1 file changed, 4 insertions
Re-pushing this as the old thread got rather stale. Some of the
VFB setup went in a bug fix. Not sure I missed a detail in rebasing
bug the keyboard setting may be the only thing missing...
-Stefan
[v2: Check return code of VIR_STRDUP and fix indentation]
[v3: Split out VRAM fixup and return erro
Add options for tuning segment offloading:
which control the respective host_ and guest_ properties
of the virtio-net device.
---
docs/formatdomain.html.in | 24 ++-
docs/schemas/domaincommon.rng | 44 -
src/conf/domain_conf.c
v2: rework XML to avoid underscores
Ján Tomko (2):
conf: add options for disabling segment offloading
qemu: wire up virtio-net segment offloading options
docs/formatdomain.html.in | 24 ++-
docs/schemas/domaincommon.rng | 44 -
src/conf/dom
Format the segment offloading options specified by
on virtio-net command line.
---
src/qemu/qemu_command.c| 40 ++
.../qemuxml2argv-net-virtio-disable-offloads.args | 10 ++
tests/qemuxml2argvtest.c | 2 ++
3 f
Ján Tomko wrote:
> On 09/14/2014 05:46 AM, Roman Bogorodskiy wrote:
> > - Provide an implementation for buildPool and deletePool operations
> >for the ZFS storage backend.
> > - Add VIR_STORAGE_POOL_SOURCE_DEVICE flag to ZFS pool poolOptions
> >as now we can specify devices to build po
Ján Tomko wrote:
> On 09/14/2014 07:24 AM, Roman Bogorodskiy wrote:
> > - docs/formatstorage.html.in: document 'zfs' pool type, add it
> >to a list of pool types that could use source physical devices
> > - docs/storage.html.in: update a ZFS pool example XML with
> >source physical dev
https://bugzilla.redhat.com/show_bug.cgi?id=1141879
A long time ago I've implemented support for so called multiqueue
net. The idea was to let guest network traffic be processed by
multiple host CPUs and thus increasing performance. However, this
behavior is enabled by QEMU via special ioctl() it
Libvirt Security Notice: LSN-2014-0004
==
Summary: Querying blkiotune after disk hotplug can lead to
libvirtd crash
Reported on: 20140911
Published on: 20140917
Fixed on: 20140917
Reported by: Luyao Huang
On 09/18/2014 02:36 AM, Daniel P. Berrange wrote:
> On Wed, Sep 17, 2014 at 04:24:07PM -0600, Eric Blake wrote:
>> Any objections to retiring the v0.9.6-maint branch? After all, we have
>> already retired the v0.9.11-maint branch
>> (http://libvirt.org/git/?p=libvirt.git;a=commit;h=cd0d348ed), and
On Thu, Sep 18, 2014 at 09:15:10AM -0600, Eric Blake wrote:
> On 09/18/2014 02:36 AM, Daniel P. Berrange wrote:
> > On Wed, Sep 17, 2014 at 04:24:07PM -0600, Eric Blake wrote:
> >> Any objections to retiring the v0.9.6-maint branch? After all, we have
> >> already retired the v0.9.11-maint branch
On 09/15/2014 11:49 AM, Daniel P. Berrange wrote:
On Mon, Sep 15, 2014 at 11:43:10AM +0200, Pavel Hrdina wrote:
On 09/15/2014 11:24 AM, Daniel P. Berrange wrote:
On Fri, Sep 12, 2014 at 06:42:08PM +0200, Pavel Hrdina wrote:
On 09/12/2014 06:25 PM, Daniel P. Berrange wrote:
On Fri, Sep 12, 201
On 09/18/2014 09:22 AM, Daniel P. Berrange wrote:
> On Thu, Sep 18, 2014 at 09:15:10AM -0600, Eric Blake wrote:
>> On 09/18/2014 02:36 AM, Daniel P. Berrange wrote:
>>> On Wed, Sep 17, 2014 at 04:24:07PM -0600, Eric Blake wrote:
Any objections to retiring the v0.9.6-maint branch? After all, w
We are not detecting the presence of FIPS from QEMU, but from procfs and
that means it's not QEMU capability. It was decided that we will pass
this flag to QEMU even if it's not supported by old QEMU binaries.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1135431
Signed-off-by: Pavel Hrdi
[adding libvirt]
On 09/18/2014 06:28 AM, Cole Robinson wrote:
> - Say you are connecting from a new libvirt UNDEFINE_NVRAM support (say Fedora
> 21 GA), to an old libvirt without it, like F20 or RHEL7.0. If we specify the
> flag unconditionally, the undefineFlags call will fail, which also means
On 09/18/2014 09:52 AM, Pavel Hrdina wrote:
> We are not detecting the presence of FIPS from QEMU, but from procfs and
> that means it's not QEMU capability. It was decided that we will pass
> this flag to QEMU even if it's not supported by old QEMU binaries.
>
> Resolves: https://bugzilla.redhat.
On 09/18/2014 10:20 AM, Michal Privoznik wrote:
> https://bugzilla.redhat.com/show_bug.cgi?id=1141879
>
> A long time ago I've implemented support for so called multiqueue
> net. The idea was to let guest network traffic be processed by
> multiple host CPUs and thus increasing performance. Howeve
On 09/18/2014 10:20 AM, Michal Privoznik wrote:
> https://bugzilla.redhat.com/show_bug.cgi?id=1141879
>
> A long time ago I've implemented support for so called multiqueue
> net. The idea was to let guest network traffic be processed by
> multiple host CPUs and thus increasing performance. Howe
On 09/18/2014 06:29 PM, Eric Blake wrote:
On 09/18/2014 09:52 AM, Pavel Hrdina wrote:
We are not detecting the presence of FIPS from QEMU, but from procfs and
that means it's not QEMU capability. It was decided that we will pass
this flag to QEMU even if it's not supported by old QEMU binaries.
On 10.09.2014 01:40, John Ferlan wrote:
Add a "rawio" to the hostdev XML and process it mimicing the
disk XML for a lun which supports/requires rawio
Signed-off-by: John Ferlan
---
docs/formatdomain.html.in | 12 ++--
docs/schemas/domaincommon.rng
On 10.09.2014 01:40, John Ferlan wrote:
Mimic the "Disk" processing for 'rawio', but for a scsi_host hostdev
lun device.
Signed-off-by: John Ferlan
---
src/qemu/qemu_domain.c | 21 +
src/qemu/qemu_domain.h | 4
src/qemu/qemu_driver.c | 1 +
src/qemu/qemu_proce
On 10.09.2014 01:40, John Ferlan wrote:
Add the 'rawio' attribute to match _virDomainDiskDef
Signed-off-by: John Ferlan
---
src/conf/domain_conf.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h
index 1107fa8..b1d13ef 100644
--- a/src/con
On 09/18/2014 06:48 PM, Eric Blake wrote:
On 09/18/2014 10:44 AM, Pavel Hrdina wrote:
Ouch. This will make our testsuite differ based on whether it is run on
Linux in FIPS mode (where FIPS might exist) or on any other setup. I
think you need to hoist the check for virFileExists() to the calle
On 09/18/2014 10:44 AM, Pavel Hrdina wrote:
>> Ouch. This will make our testsuite differ based on whether it is run on
>> Linux in FIPS mode (where FIPS might exist) or on any other setup. I
>> think you need to hoist the check for virFileExists() to the caller, and
>> pass in the result as a ne
On 18.09.2014 18:36, John Ferlan wrote:
On 09/18/2014 10:20 AM, Michal Privoznik wrote:
https://bugzilla.redhat.com/show_bug.cgi?id=1141879
A long time ago I've implemented support for so called multiqueue
net. The idea was to let guest network traffic be processed by
multiple host CPUs and
On 09/18/2014 10:55 AM, Pavel Hrdina wrote:
> On 09/18/2014 06:48 PM, Eric Blake wrote:
>> On 09/18/2014 10:44 AM, Pavel Hrdina wrote:
>>
Ouch. This will make our testsuite differ based on whether it is
run on
Linux in FIPS mode (where FIPS might exist) or on any other setup. I
>>>
We are not detecting the presence of FIPS from QEMU, but from procfs and
that means it's not QEMU capability. It was decided that we will pass
this flag to QEMU even if it's not supported by old QEMU binaries.
This patch also reverts changes done by commit a21cfb0f to
qemucapabilitestest and imple
On 09/18/2014 12:50 PM, Michal Privoznik wrote:
> On 10.09.2014 01:40, John Ferlan wrote:
>> Mimic the "Disk" processing for 'rawio', but for a scsi_host hostdev
>> lun device.
<...snip...>
>> diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c
>> index b1d8a32..3544716 100644
>> ---
virDomainVirtTypeToString() returns 'qemu' and 'kvm' separately.
Don't require a separate apparmor profile for both, rather always
look for TEMPLATE.qemu.
Signed-off-by: Serge Hallyn
---
src/security/virt-aa-helper.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/se
On 09/18/2014 12:01 PM, Pavel Hrdina wrote:
> We are not detecting the presence of FIPS from QEMU, but from procfs and
> that means it's not QEMU capability. It was decided that we will pass
> this flag to QEMU even if it's not supported by old QEMU binaries.
>
> This patch also reverts changes do
On Wed, Sep 17, 2014 at 04:24:07PM -0600, Eric Blake wrote:
> Any objections to retiring the v0.9.6-maint branch? After all, we have
> already retired the v0.9.11-maint branch
> (http://libvirt.org/git/?p=libvirt.git;a=commit;h=cd0d348ed), and the
> only activity on v0.9.6-maint since 0.9.6.4 was
There is no need to acquire the driver-wide lock in
libxlDomainDefineXML. When switching to jobs in the libxl
driver, most driver-wide locks were removed. The locking here
was preserved since I mistakenly thought virDomainObjListAdd
needed protection. This is not the case, so remove the
unnecess
libc has another constant with the same name, which leads to
redefinition error when building against static libvirt
---
src/util/virtime.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/util/virtime.c b/src/util/virtime.c
index acbec41..7b3ec44 100644
--- a/src/util/v
On 09/18/2014 03:24 PM, Cédric Bosdonnat wrote:
> libc has another constant with the same name, which leads to
> redefinition error when building against static libvirt
> ---
> src/util/virtime.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/src/util/virtime.c b/src
commit
3020594ac57c5e06e79f3db8c765f6bb18c40802
add HOME environment variable.
Add a doc for this.
Signed-off-by: Chen Hanxiao
---
docs/drvlxc.html.in | 2 ++
1 file changed, 2 insertions(+)
diff --git a/docs/drvlxc.html.in b/docs/drvlxc.html.in
index 403ce24..31da37c 100644
--- a/docs/drvlxc.h
Stefan Bader wrote:
> Re-pushing this as the old thread got rather stale.
Thanks.
> Some of the
> VFB setup went in a bug fix. Not sure I missed a detail in rebasing
> bug the keyboard setting may be the only thing missing...
>
Yes, agreed.
> -Stefan
>
> [v2: Check return code of VIR_STRDUP
Any comments on this change?
Regards,
Jim
Jim Fehlig wrote:
> If an NTP server is configured on the host, it is possible for
> libvirt-guests to start before the NTP service, in which case
> guest clocks won't be synchronized to the host clock.
>
> Add ntp-wait.service to "After" in libvirt-guest
I was trying different video devices to correct the problem of having
1024x768 as the maximum resolution for Debian guests. I found out that
adding VMVGA gives the wide variety of choice of resolutions I was
looking for and adjusts the display properly, in contrast to the
deformed way xconf fil
libc has another constant with the same name, which leads to
redefinition error when building against static libvirt
---
src/util/virtime.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/util/virtime.c b/src/util/virtime.c
index acbec41..9d365d5 100644
--- a/src/util/v
73 matches
Mail list logo
