If virSecuritySELinuxRestoreFileLabel returns 0 or -1 too soon, then
the @newpath will be leaked.
Suggested-by: Michal Privoznik
Signed-off-by: John Ferlan
---
src/security/security_selinux.c | 10 ++
1 file changed, 6 insertions(+), 4 deletions(-)
diff --git
The @con type security_context_t is actually a "char *", so the
correct check should be to dereference one more level; otherwise,
we could return/use the NULL pointer later in a subsequent
virSecuritySELinuxSetFileconImpl call (using @fcon).
Suggested-by: Michal Prívozník
Signed-off-by: John
Both found by Coverity...
Please, be kind if this send is messed up - I have a new laptop and
I'm slowly finding out all the things that I don't have on it that I
used to have on the old one.
John Ferlan (2):
security: Resolve possible memory leak
security: Fix comparison for
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> Signed-off-by: Nikolay Shirokovskiy
> ---
> docs/news.xml | 11 +++
> 1 file changed, 11 insertions(+)
>
Reviewed-by: John Ferlan
John
BTW: the explanation you give here matches my thoughts on this although
what I read and
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> This patch just adds basic checks for persistent domain config
> on snapshot metadata redefine. It also lets use previous version
> of config if it exists in previous version of metadata and
> not explicitly specified in new one just as in case
Please accept our apologies if you receive multiple copies of this Call for
Papers
CALL FOR PAPERS
14th Workshop on Virtualization in High-Performance Cloud Computing (VHPC
'19)
held in conjunction with the International
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> Renaming domain which has snapshots is prohibited. Also reverting
> to ABI compatible active domain with a different name can have
> issues later I guess. So let's prohibit changing domain name on snapshot
> metadata redefine as well.
>
>
On 12/20/18 12:39 AM, John Ferlan wrote:
>
>
> On 12/12/18 7:40 AM, Michal Privoznik wrote:
>> Similarly to what I did in DAC driver, this also requires the
>> same SELinux label to be used for shared paths. If a path is
>> already in use by a domain (or domains) then and the domain we
>> are
On 12/20/18 12:39 AM, John Ferlan wrote:
>
>
> On 12/12/18 7:40 AM, Michal Privoznik wrote:
>> Similarly to what I did in DAC driver, this also requires the
>> same SELinux label to be used for shared paths. If a path is
>> already in use by a domain (or domains) then and the domain we
>> are
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> Signed-off-by: Nikolay Shirokovskiy
> ---
> docs/formatsnapshot.html.in | 7 +++
> 1 file changed, 7 insertions(+)
>
This probably should be merged with patch5 - nice to separate for review
though. Although, like patch4 it's not that
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> Persistent in this case is a persistent config of active
of the active
> persistent domain. We need it for metadata to be complete
> as explained in previous patch.
Hence why I think the two should be combined.
>
> Config is saved like
Maybe a few words here about the option being used to hold a persistent
config that could be different than the snapshot's running domain def.
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> Signed-off-by: Nikolay Shirokovskiy
> ---
> docs/schemas/domainsnapshot.rng | 10 ++
> 1
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> In case of active persistent domain snapshot metadata is
> not complete. We save only active configuration and on
> restore use it both for active and inactive configuration.
> Let's fix it and save and restore both in this case.
>
> In case
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> Commit [1] changed cleanup code to first remove inactive domain
> and then release job so we can reuse common cleanup code.
>
> [1] 9115dcd83 qemu: Introduce and use qemuDomainRemoveInactiveJob
>
> Signed-off-by: Nikolay Shirokovskiy
> ---
>
On 12/13/18 3:03 AM, Nikolay Shirokovskiy wrote:
> Introduced-by: 6a6f6b91e qemu: "Fix CPU model broken by older libvirt"
Add a blank line here...
> Signed-off-by: Nikolay Shirokovskiy
> ---
> src/qemu/qemu_driver.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
Reviewed-by:
On 12/18/18 12:18 PM, Michal Privoznik wrote:
> Otherwise looking good. I agree with Martin that this might not be the
> best solution but it's the best we have (also, the problem is not on our
> side).
>
> ACK (will wait a day or two for others to chime in before pushing)
No objections were
On 12/20/18 12:48 PM, Marc Hartmayer wrote:
> On Wed, Dec 19, 2018 at 03:37 PM +0100, Michal Privoznik
> wrote:
>> On 12/19/18 2:54 PM, Ján Tomko wrote:
>>>
>>> Reviewed-by: Ján Tomko
>>
>> Thanks to you and Dan. I've pushed these.
>
> I tried out the current master (e05d8e570b) and I got the
From: Thomas Huth
They've been deprecated for two releases and nobody complained that they
are still required anymore, so it's time to remove these now.
And while we're at it, mark the other remaining old 0.x machine types
as deprecated (since they can not properly be used for live-migration
$SUBJ:
docs: Add news for new nvdimm options
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> add more configure options for NVDIMM
>
> Signed-off-by: Luyao Zhong
> ---
> docs/news.xml | 9 +
> 1 file changed, 9 insertions(+)
>
> diff --git a/docs/news.xml b/docs/news.xml
> index
$SUBJ:
qemu: Add qemu command-line to generate the nvdimm unarmed property
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> According to the result parsing from xml, add unarmed property
> into QEMU command line:
>
> -device nvdimm,...[,unarmed=on/off]
unarmed=on
(off is never added)
>
>
$SUBJ:
qemu: Add command-line to generate the nvdimm pmem property
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> According to the result parsing from xml, add pmem property
> into QEMU command line:
>
> -object memory-backend-file,...[,pmem=on/off]
"pmem=on"
(off is never added).
>
>
$SUBJ:
qemu: Add command-line to generate the nvdimm align property
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> According to the result parsing from xml, add align property
> into QEMU command line:
>
> -object memory-backend-file,...[,align=xxx]
>
> Signed-off-by: Luyao Zhong
> ---
>
$SUBJ:
tests: Use DO_TEST_CAPS_LATEST for nvdimm qemuxml2argv
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> Depracate DO_TEST to do nvdimm qemuxml2argvdata tests, because
Deprecate
> DO_TEST_CAPS_LATEST is a better choice. the DO_TEST need
needs
> to specify all qemu capabilities and is not easy
$SUBJ:
qemu: Introduce ... capability
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> This capability tracks if memory-backend-file has align
has the
> attribute or not.
>
> Signed-off-by: Luyao Zhong
> ---
> src/qemu/qemu_capabilities.c | 4 +++-
>
$SUBJ:
qemu: Introduce ... capability
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> This capability tracks if memory-backend-file has pmem
s/has/has the/
> attribute or not.
>
> Signed-off-by: Luyao Zhong
> ---
> src/qemu/qemu_capabilities.c | 4
>
$SUBJ:
qemu: Introduce ... capability
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> This capability tracks if nvdimm has unarmed attribute or not.
s/has/has the/
s/not./not for the nvdimm readonly xml attribute./
>
> Signed-off-by: Luyao Zhong
> ---
> src/qemu/qemu_capabilities.c
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> The 'readonly' option allows users to mark vNVDIMM read-only:
>
>
> ...
>
>
> /dev/dax0.0
>
>
> 4094
> 0
>
> 2
>
>
>
>
> ...
>
>
>
$SUBJ:
conf: Introduce
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> The 'pmem' option allows users to specify whether the backend
> storage of memory-backend-file is a real persistent memory:
>
>
> ...
>
>
> /dev/dax0.0
>
>
>
> 4094
>
$SUBJ:
s/nvdimm: introduce/conf: Introduce/
On 12/20/18 4:14 AM, Luyao Zhong wrote:
> NVDIMM emulation will mmap the backend file, it uses host pagesize
> as the alignment of mapping address before, but some backends may
> require alignments different from the pagesize. So the 'alignsize'
>
On Thu, Dec 20, 2018 at 02:30:03PM +0100, Andrea Bolognani wrote:
> The feature is deprecated, and modern Ansible versions
> complain about it. Pass the list of packages to install
> to the corresponding module directly instead.
>
> Signed-off-by: Andrea Bolognani
> ---
> guests/ansible.cfg
The feature is deprecated, and modern Ansible versions
complain about it. Pass the list of packages to install
to the corresponding module directly instead.
Signed-off-by: Andrea Bolognani
---
guests/ansible.cfg | 1 -
guests/playbooks/update/tasks/packages.yml | 4 +---
On Tue, Dec 18, 2018 at 02:31:27PM -0500, John Ferlan wrote:
On 12/18/18 11:16 AM, Daniel P. Berrangé wrote:
On Tue, Dec 18, 2018 at 11:10:34AM -0500, John Ferlan wrote:
On 12/18/18 10:47 AM, Daniel P. Berrangé wrote:
On Tue, Dec 18, 2018 at 10:39:20AM -0500, John Ferlan wrote:
Commit
On Wed, Dec 19, 2018 at 03:37 PM +0100, Michal Privoznik
wrote:
> On 12/19/18 2:54 PM, Ján Tomko wrote:
>>
>> Reviewed-by: Ján Tomko
>
> Thanks to you and Dan. I've pushed these.
I tried out the current master (e05d8e570b) and I got the following
error message regularly:
2018-12-20
On Wed, 2018-12-19 at 15:49 +0100, Michal Privoznik wrote:
> +
> +
> + Remember original owners and SELinux labels of files
> +
> +
> + When a domain is starting up libvirt changes DAC and
> + SELinux labels so that domain can access it.
On Thu, Dec 20, 2018 at 10:23:46AM +0100, Pino Toscano wrote:
> Including making it build again.
Would be good to get it added to the CI system too if you're
intereted in doing that
>
> Pino Toscano (6):
> Fix typo in autogen.sh
> Bump libvirt-config & libvirt-gobject requirements to 0.1.9
On Thu, Dec 20, 2018 at 10:23:52AM +0100, Pino Toscano wrote:
> While osinfo_entity_get_param_value is certainly not deprecated, make
> use of the direct methods to get the needed properties.
>
> Signed-off-by: Pino Toscano
> ---
> examples/virt-designer.c | 15 +--
> 1 file
On Thu, Dec 20, 2018 at 10:23:51AM +0100, Pino Toscano wrote:
> - move general glob patterns to the top, adding entries for generated
> files no matter their location
> - tie path-specific entries with a leading '/'
> - add mssing entries
> - sort all the sections
>
> Signed-off-by: Pino
On Thu, Dec 20, 2018 at 10:23:50AM +0100, Pino Toscano wrote:
> Use g_list_free_full instead of g_list_foreach + g_list_free, so the
> lists are cleared with a single call. This also avoids
> cast-function-type warnings because of the different signature of GFunc
> and g_object_unref.
>
>
On Thu, Dec 20, 2018 at 10:23:47AM +0100, Pino Toscano wrote:
> Mention the right project name.
>
> Signed-off-by: Pino Toscano
> ---
> autogen.sh | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
Reviewed-by: Daniel P. Berrangé
Regards,
Daniel
--
|: https://berrange.com -o-
On Thu, Dec 20, 2018 at 10:23:49AM +0100, Pino Toscano wrote:
> For a GVIR_DESIGNER_DOMAIN_GRAPHICS_DESKTOP graphics type, do not
> fallback on the code handling GVIR_DESIGNER_DOMAIN_GRAPHICS_RDP.
Presumably new GCC warned about this one ?
>
> Signed-off-by: Pino Toscano
> ---
>
On Thu, Dec 20, 2018 at 10:23:48AM +0100, Pino Toscano wrote:
> This version bumps the glib version required to >= 2.36.0, so using
> these more recent versions we can assume a more recent glib too.
>
> Signed-off-by: Pino Toscano
> ---
> configure.ac | 4 ++--
> 1 file changed, 2
On 2018/12/20 下午5:14, Luyao Zhong wrote:
According to the result parsing from xml, add unarmed property
into QEMU command line:
-device nvdimm,...[,unarmed=on/off]
Signed-off-by: Luyao Zhong
---
src/qemu/qemu_command.c| 16 --
src/qemu/qemu_command.h
While osinfo_entity_get_param_value is certainly not deprecated, make
use of the direct methods to get the needed properties.
Signed-off-by: Pino Toscano
---
examples/virt-designer.c | 15 +--
1 file changed, 5 insertions(+), 10 deletions(-)
diff --git a/examples/virt-designer.c
Use g_list_free_full instead of g_list_foreach + g_list_free, so the
lists are cleared with a single call. This also avoids
cast-function-type warnings because of the different signature of GFunc
and g_object_unref.
Signed-off-by: Pino Toscano
---
libvirt-designer/libvirt-designer-domain.c | 9
Mention the right project name.
Signed-off-by: Pino Toscano
---
autogen.sh | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/autogen.sh b/autogen.sh
index 72ce2c0..59fa0c7 100755
--- a/autogen.sh
+++ b/autogen.sh
@@ -12,7 +12,7 @@ DIE=0
(autoconf --version) < /dev/null
- move general glob patterns to the top, adding entries for generated
files no matter their location
- tie path-specific entries with a leading '/'
- add mssing entries
- sort all the sections
Signed-off-by: Pino Toscano
---
.gitignore | 64
For a GVIR_DESIGNER_DOMAIN_GRAPHICS_DESKTOP graphics type, do not
fallback on the code handling GVIR_DESIGNER_DOMAIN_GRAPHICS_RDP.
Signed-off-by: Pino Toscano
---
libvirt-designer/libvirt-designer-domain.c | 2 ++
1 file changed, 2 insertions(+)
diff --git
This version bumps the glib version required to >= 2.36.0, so using
these more recent versions we can assume a more recent glib too.
Signed-off-by: Pino Toscano
---
configure.ac | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/configure.ac b/configure.ac
index
Including making it build again.
Pino Toscano (6):
Fix typo in autogen.sh
Bump libvirt-config & libvirt-gobject requirements to 0.1.9
Add missing break in gvir_designer_domain_add_graphics
Use g_list_free_full
Update .gitignore
Switch away from osinfo_entity_get_param_value
According to the result parsing from xml, add unarmed property
into QEMU command line:
-device nvdimm,...[,unarmed=on/off]
Signed-off-by: Luyao Zhong
---
src/qemu/qemu_command.c| 16 --
src/qemu/qemu_command.h| 3 +-
According to the result parsing from xml, add pmem property
into QEMU command line:
-object memory-backend-file,...[,pmem=on/off]
Signed-off-by: Luyao Zhong
---
src/qemu/qemu_command.c| 14 +
.../memory-hotplug-nvdimm-pmem.x86_64-latest.args | 36
add more configure options for NVDIMM
Signed-off-by: Luyao Zhong
---
docs/news.xml | 9 +
1 file changed, 9 insertions(+)
diff --git a/docs/news.xml b/docs/news.xml
index e9b6bb7..2076cd5 100644
--- a/docs/news.xml
+++ b/docs/news.xml
@@ -79,6 +79,15 @@
qemu: Add support for
According to the result parsing from xml, add align property
into QEMU command line:
-object memory-backend-file,...[,align=xxx]
Signed-off-by: Luyao Zhong
---
src/qemu/qemu_command.c| 14 +
.../memory-hotplug-nvdimm-align.x86_64-latest.args | 36
Depracate DO_TEST to do nvdimm qemuxml2argvdata tests, because
DO_TEST_CAPS_LATEST is a better choice. the DO_TEST need
to specify all qemu capabilities and is not easy for scaling.
Signed-off-by: Luyao Zhong
---
.../memory-hotplug-nvdimm-access.args | 31 ---
This capability tracks if nvdimm has unarmed attribute or not.
Signed-off-by: Luyao Zhong
---
src/qemu/qemu_capabilities.c | 8 ++
src/qemu/qemu_capabilities.h | 1 +
.../caps_2.10.0.x86_64.replies | 62 -
NVDIMM emulation will mmap the backend file, it uses host pagesize
as the alignment of mapping address before, but some backends may
require alignments different from the pagesize. So the 'alignsize'
option will be introduced to allow users specifying the proper
alignment:
...
This capability tracks if memory-backend-file has pmem
attribute or not.
Signed-off-by: Luyao Zhong
---
src/qemu/qemu_capabilities.c | 4
src/qemu/qemu_capabilities.h | 3 +++
tests/qemucapabilitiesdata/caps_3.1.0.ppc64.xml | 1 +
This capability tracks if memory-backend-file has align
attribute or not.
Signed-off-by: Luyao Zhong
---
src/qemu/qemu_capabilities.c | 4 +++-
src/qemu/qemu_capabilities.h | 1 +
tests/qemucapabilitiesdata/caps_2.12.0.aarch64.xml | 1 +
The 'pmem' option allows users to specify whether the backend
storage of memory-backend-file is a real persistent memory:
...
/dev/dax0.0
4094
0
2
...
Signed-off-by: Luyao Zhong
The 'readonly' option allows users to mark vNVDIMM read-only:
...
/dev/dax0.0
4094
0
2
...
Signed-off-by: Luyao Zhong
---
docs/formatdomain.html.in | 45
Hi libvirt experts,
There are some gaps between qemu and libvirt, libvirt has not
supported several config options about NVDIMM memory while
qemu has now, including 'align', 'pmem', 'unarmed'.
So these three options is introduced in this patch set.
The 'align' and 'unarmed' was renamed to
From: Li XueLei
During making disk snapshot in an active domain, sometimes we
should update the domain's persistent definition.We must check if
the domain is defined, before we update the persistent definition
file.
First,we create a vm.Then,define the vm and undefine the vm.Last create a
62 matches
Mail list logo