On 2/20/19 9:53 AM, Eric Blake wrote:
> Continue the work of the previous patch in making it possible
> to copy the state of a transient domain with snapshots from one
> host to another, by allowing the destination to perform bulk
> redefines. Note that the destination still has to do separate
> c
There have been a few requests [1][2] to support Xen's max_grant_frames setting
in libvirt domXML, but I'm not quite sure how to model it. The documentation [3]
on this setting states:
Specify the maximum number of grant frames the domain is allowed to have. This
value controls how many pages
On 2/10/19 6:09 AM, Roman Bogorodskiy wrote:
> Describe bhyve's ignoring unknown MSRs writes feature
> introduced by commit 525918ac5c.
>
> Signed-off-by: Roman Bogorodskiy
> ---
> docs/news.xml | 10 ++
> 1 file changed, 10 insertions(+)
>
> diff --git a/docs/news.xml b/docs/news.xml
>
On 2/10/19 6:09 AM, Roman Bogorodskiy wrote:
> Implement the MSRs ignore unknown reads and writes feature
> that's specified using:
>
>
> ...
>
> ...
>
>
> in the domain XML.
>
> In bhyve, it's just passing '-w' command line argument to the bhyve(8)
> executable.
>
> Signed-o
On 2/10/19 6:09 AM, Roman Bogorodskiy wrote:
> Introduce the 'msrs' feature element that controls Model Specific
> Registers related behaviour. At this moment it allows only
> single tunable attribute "unknown":
>
>
>
> Which tells hypervisor to ignore accesses to unimplemented
> Model Specific
On 2/14/19 4:39 AM, Andrea Bolognani wrote:
> It hasn't been needed since dbe55924f90d, and flake8 rightfully
> complains about that.
>
> Signed-off-by: Andrea Bolognani
> ---
> guests/lcitool | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/guests/lcitool b/guests/lcitool
> index 759eff
Recently I've played with libvirt coverage/gcov/lcov support. The
libvirt specific way to enable it is:
./configure --enable-test-coverage
make check
make cov
firefox coverage/index.html
However it looks like gnulib provides an automagic way to get similar
output. The process is:
make
On 2/18/19 6:21 PM, Laine Stump wrote:
> dnsmasq documentation says that the *IPv4* prefix/network
> address/broadcast address sent to dhcp clients will be automatically
> determined by dnsmasq by looking at the interface it's listening on,
> so the original libvirt code did not add a netmask to
On 2/13/19 3:31 AM, Andrea Bolognani wrote:
> On Fri, 2019-02-08 at 17:11 -0500, Cole Robinson wrote:
>> v2 libvirt patches:
>> https://www.redhat.com/archives/libvir-list/2019-January/msg00877.html
>> v1 libvirt patches:
>> https://www.redhat.com/archives/libvir-list/2019-January/msg00593.html
>>
Now that we're using VIR_AUTOPTR(virBitmap) there's a couple of methods
that we can clean up some now unnecessary goto paths.
Signed-off-by: John Ferlan
---
src/conf/domain_conf.c | 26 +-
1 file changed, 9 insertions(+), 17 deletions(-)
diff --git a/src/conf/domain_conf
Let's make use of the auto __cleanup capabilities for virBitmapPtr.
Signed-off-by: John Ferlan
---
src/conf/domain_conf.c | 21 +++--
1 file changed, 7 insertions(+), 14 deletions(-)
diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
index 542d53e709..f27af65d80 100644
In preparation for using auto free mechanism, change to using the
VIR_STEAL_PTR on @def to @ret and of course be sure to properly clean
up @def in cleanup.
Signed-off-by: John Ferlan
---
src/conf/domain_conf.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/src/conf/d
In preparation for VIR_AUTOFREE usage, let's remove a couple
of unused variables so that clang compilations won't fail.
Signed-off-by: John Ferlan
---
src/conf/domain_conf.c | 7 +--
1 file changed, 1 insertion(+), 6 deletions(-)
diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c
Let's make use of the auto __cleanup capabilities for VIR_FREE consumers.
In some cases adding or removing blank lines for readability.
Signed-off-by: John Ferlan
---
src/conf/domain_conf.c | 1080 +---
1 file changed, 347 insertions(+), 733 deletions(-)
diff
Now that we're using VIR_AUTOFREE there's quite a bit of clean up
possible for now unnecessary goto paths.
Signed-off-by: John Ferlan
---
src/conf/domain_conf.c | 801 -
1 file changed, 310 insertions(+), 491 deletions(-)
diff --git a/src/conf/domain_conf
v1: https://www.redhat.com/archives/libvir-list/2019-February/msg01160.html
Changes since v1:
* Push patch 1
* Split patch 2 to follow code review guidance:
* (Patch1) Pull out virDomainEmulatorPinDefParseXML changes to
use VIR_STEAL_PTR
* (Patch2) Use VIR_AUTOPTR(virBit
Signed-off-by: Daniel P. Berrangé
---
po/id.mini.po| 20 +
po/id.po | 153 ---
po/ilo.mini.po | 20 +
po/ilo.po| 153 ---
po/is.mini.po| 20 +
p
Similar to the libvirt-glib.pot, .po files contain line numbers and file
names identifying where in the source a translatable string comes from.
The source locations in the .po files are thrown away and replaced with
content from the libvirt-glib.pot whenever msgmerge is run, so this is not
preciou
Signed-off-by: Daniel P. Berrangé
---
po/{uk.po => uk.mini.po} | 152 +++---
po/ur.mini.po| 20 +
po/ur.po | 153 ---
po/vi.mini.po| 20 +
po/vi.po | 153 -
Historically we have relied on intltool to install a standard
po/Makefile.in.in which has very limited scope for customization.
intltool is deprecated in favour of standard gettextize tools,
but these share the same disadvantages.
Writing make rules for po file management is no more difficult
than
Add rules to handle pushing libvirt-glib.pot to zanata, and refreshing .po
files with new content from zanata.
Signed-off-by: Daniel P. Berrangé
---
po/Makefile.am | 8
1 file changed, 8 insertions(+)
diff --git a/po/Makefile.am b/po/Makefile.am
index f2818c7..d66aec7 100644
--- a/po/M
This applies the same improvements previously done in libvirt:
https://www.redhat.com/archives/libvir-list/2018-April/msg01004.html
https://www.berrange.com/posts/2018/11/29/improved-translation-po-file-handling-by-ditching-gettext-autotools-integration/
The key problems with our current app
Signed-off-by: Daniel P. Berrangé
---
po/bn.mini.po| 20 +
po/bn.po | 153 -
po/bo.mini.po| 20 +
po/bo.po | 153 -
po/br.mini.po| 20 +
po/br
Signed-off-by: Daniel P. Berrangé
---
po/{fr.po => fr.mini.po} | 156 +++
po/gl.mini.po| 20 +
po/gl.po | 153 --
po/gu.mini.po| 20 +
po/gu.po | 153 -
On 2/20/19 10:47 AM, Ján Tomko wrote:
> On Wed, Feb 20, 2019 at 09:46:59AM -0500, John Ferlan wrote:
>> Let's make use of the auto __cleanup capabilities cleaning up any
>> now unnecessary goto paths.
>>
>
> At this scale, separating the AUTOFREE conversion and the return values
> would really
Continue the work of the previous patch in making it possible
to copy the state of a transient domain with snapshots from one
host to another, by allowing the destination to perform bulk
redefines. Note that the destination still has to do separate
calls for creating/defining the domain first, and
Make it possible to grab all snapshot XMLs via a single API
call, by adding a new internal flag. All callers are adjusted,
for now passing NULL and not using the new flag. A new wrapper
virDomainDefFormatFull() is added to make it easier for the
next patch to add a few callers without having to re
virDomainSnapshotDefFormat currently takes two sets of knobs:
an 'unsigned int flags' argument that can currently just be
VIR_DOMAIN_DEF_FORMAT_SECURE, and an 'int internal' argument used as
a bool to determine whether to output an additional element. It
then reuses the 'flags' knob to call into v
Split out an internal helper that produces format into a
virBuffer, similar to what domain_conf.c does, and making
the next patch easier to write.
Signed-off-by: Eric Blake
---
src/conf/snapshot_conf.c | 103 ++-
1 file changed, 59 insertions(+), 44 deletions(
Right now, copying the state of a transient domain with snapshots
from one host to another requires multiple API calls on both
machines - on the host: get the domain XML, get a list of the
snapshots, and then for each snapshot get the snapshot's XML;
then on the destination: create the domain, then
While looking at my work on incremental backups, Nir raised a good
point: if we want to recreate a set of known checkpoints on one
machine that will be taking over a domain from another machine,
my initial proposal required making multiple API calls to list the
XML for each checkpoint on the source
Commit 28f8dfdc (1.0.0) added a flag to virDomainGetXMLDesc, but
failed to document its effects. And considering that the
MIGRATABLE flag has been the source of past bugs (CVE-2014-7823,
fixed in commit b1674ad5 (1.2.11), or even cf2d4c60 (1.2.13) where
flag mismatch broke virsh edit), make the wo
On 2/20/19 10:42 AM, Ján Tomko wrote:
> On Wed, Feb 20, 2019 at 09:46:58AM -0500, John Ferlan wrote:
>> Alter the logic to return @tmpnode to ctxt->node in both the error
>> and non-error path when calling virDomainNetIPInfoParseXML.
>>
>
> What is the motivation here?
>
> On error, ctxt->node
On Wed, Feb 20, 2019 at 09:46:59AM -0500, John Ferlan wrote:
Let's make use of the auto __cleanup capabilities cleaning up any
now unnecessary goto paths.
At this scale, separating the AUTOFREE conversion and the return values
would really make this easier to read.
Also, this fails compilatio
Signed-off-by: Pavel Hrdina
---
src/util/vircgroupv2.c | 24
1 file changed, 24 insertions(+)
diff --git a/src/util/vircgroupv2.c b/src/util/vircgroupv2.c
index f3aa6ebc48..25afab1cad 100644
--- a/src/util/vircgroupv2.c
+++ b/src/util/vircgroupv2.c
@@ -1600,6 +1600,28 @@
Cgroups v2 don't have memory_migrate interface and the migration is
enabled by default.
Signed-off-by: Pavel Hrdina
---
src/util/vircgroupv2.c | 19 +++
1 file changed, 19 insertions(+)
diff --git a/src/util/vircgroupv2.c b/src/util/vircgroupv2.c
index 4cfbd52f2d..f3aa6ebc48 100
Signed-off-by: Pavel Hrdina
---
src/util/vircgroupv2.c | 25 +
1 file changed, 25 insertions(+)
diff --git a/src/util/vircgroupv2.c b/src/util/vircgroupv2.c
index e0fa8e1cc0..4cfbd52f2d 100644
--- a/src/util/vircgroupv2.c
+++ b/src/util/vircgroupv2.c
@@ -1561,6 +1561,28 @
Pavel Hrdina (4):
util: implement virCgroupV2(Set|Get)CpusetMems
util: implement virCgroupV2(Set|Get)CpusetMemoryMigrate
util: implement virCgroupV2(Set|Get)CpusetCpus
util: enable cgroups v2 cpuset controller for threads
src/util/vircgroupv2.c | 74 +++
Alter the logic to return @tmpnode to ctxt->node in both the error
and non-error path when calling virDomainNetIPInfoParseXML.
Signed-off-by: John Ferlan
---
src/conf/domain_conf.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/src/conf/domain_conf.c b/src/conf/domain
Let's make use of the auto __cleanup capabilities cleaning up any
now unnecessary goto paths.
Signed-off-by: John Ferlan
---
src/conf/domain_conf.c | 53 --
1 file changed, 20 insertions(+), 33 deletions(-)
diff --git a/src/conf/domain_conf.c b/src/conf/d
Adding to the recent hype for autofree... It's more than just saving
a few lines - in some cases the code is easier to read too. The last
patch is really long - I suppose it could be broken up as could perhaps
domain_conf into more easily process-able units.
John Ferlan (4):
conf: Use VIR_STEAL
virFileWrapperFdFree(), like all free functions, is supposed
to only release allocated resources, so error reporting is
better suited for virFileWrapperFdClose().
This reverts commit b0c3e931804a86cd7146db0164ab4843039c410b.
Signed-off-by: Andrea Bolognani
---
src/util/virfile.c | 8 +++-
1
We'll want to use this function in the cleanup path soon,
and in order to be able to do that we need to make sure we
can call it multiple times on the same virFileWrapperFd
without side effects.
Signed-off-by: Andrea Bolognani
---
src/util/virfile.c | 8 +++-
1 file changed, 7 insertions(+),
Changes from [v3]:
* Make sure errors from virFileWrapperFdClose() result in an
overall failure being bubbled up;
* don't call virReportError() unless the function is going to
report the failure through its return code as well;
* revert b0c3e931804a more thoroughly.
Changes from [v2]:
* Move
On Wed, Feb 20, 2019 at 02:20:10PM +0100, Michal Privoznik wrote:
Currently, some arguments are called strcontent and strsrc, or
content and src or some other combination. This makes it
impossible to see at the first glance what argument is supposed
to represent 'expected' value and which one rep
libvirt_iohelper is used internally by the virFileWrapperFd APIs;
more specifically, in the QEMU driver we have the doCoreDump() and
qemuDomainSaveMemory() helper functions as users, and those in turn
end up being called by the implementation of several driver APIs.
By calling virReportError() if
Right now we're reporting errors in virFileWrapperFdFree(),
but that's hardly the appropriate place to do so, as free
functions are supposed to do nothing more than release
allocated resources.
We want to move that code back into virFileWrapperFdClose(),
but before we can do that we need to make s
On Wed, Feb 20, 2019 at 02:20:09PM +0100, Michal Privoznik wrote:
The current naming makes it hard for me to see which holds the
expected value and which holds the actual value. Rename them to
make it obvious.
Signed-off-by: Michal Privoznik
---
tests/testutils.c | 10 +-
1 file changed,
On Wed, Feb 20, 2019 at 02:20:08PM +0100, Michal Privoznik wrote:
In order to save a few lines of code, and also since it's hype
let's use VIR_AUTOFREE() for the two strings we allocate there.
Signed-off-by: Michal Privoznik
---
tests/testutils.c | 17 +
1 file changed, 5 inserti
On Wed, Feb 20, 2019 at 02:20:09PM +0100, Michal Privoznik wrote:
The current naming makes it hard for me to see which holds the
expected value and which holds the actual value. Rename them to
make it obvious.
Well, the naming was confusing because of the function parameters, so I would
mainly
On Wed, 2019-02-20 at 14:20 +0100, Michal Privoznik wrote:
[...]
> /*
> - * @param strcontent: String input content
> - * @param filename: File to compare strcontent against
> + * @param actual: String input content
> + * @param filename: File to compare @actual against
> *
> - * If @strcontent
On Wed, 2019-02-20 at 14:06 +0100, Michal Privoznik wrote:
> On 2/20/19 1:56 PM, Andrea Bolognani wrote:
> > There is barely ever a reason *not* to split changes into smaller,
> > independent units; having to write "at the same time" in the commit
> > message should be your hint that you're doing i
In order to save a few lines of code, and also since it's hype
let's use VIR_AUTOFREE() for the two strings we allocate there.
Signed-off-by: Michal Privoznik
---
tests/testutils.c | 17 +
1 file changed, 5 insertions(+), 12 deletions(-)
diff --git a/tests/testutils.c b/tests/te
This is v2 of:
https://www.redhat.com/archives/libvir-list/2019-February/msg01077.html
diff to v1:
- Split one patch into smaller ones
- Made myself a new coffee meanwhile
Michal Prívozník (3):
virTestCompareToULL: Use VIR_AUTOFREE()
virTestCompareToULL: Rename local variables
testutils: E
The current naming makes it hard for me to see which holds the
expected value and which holds the actual value. Rename them to
make it obvious.
Signed-off-by: Michal Privoznik
---
tests/testutils.c | 10 +-
1 file changed, 5 insertions(+), 5 deletions(-)
diff --git a/tests/testutils.c b
Currently, some arguments are called strcontent and strsrc, or
content and src or some other combination. This makes it
impossible to see at the first glance what argument is supposed
to represent 'expected' value and which one represents 'actual'
value. Rename the arguments to make it obvious.
Si
On Wed, Feb 20, 2019 at 01:34:15PM +0100, Andrea Bolognani wrote:
> On Tue, 2019-02-19 at 16:58 +, Daniel P. Berrangé wrote:
> > On Tue, Feb 19, 2019 at 05:52:31PM +0100, Andrea Bolognani wrote:
> [...]
> > > @@ -337,8 +337,12 @@ virFileWrapperFdClose(virFileWrapperFdPtr wfd)
> > >
> > >
On Tue, 2019-02-19 at 16:58 +, Daniel P. Berrangé wrote:
> On Tue, Feb 19, 2019 at 05:52:31PM +0100, Andrea Bolognani wrote:
[...]
> > @@ -337,8 +337,12 @@ virFileWrapperFdClose(virFileWrapperFdPtr wfd)
> >
> > ret = virCommandWait(wfd->cmd, NULL);
> >
> > +/* If the command used t
On 2/20/19 2:37 AM, Michal Privoznik wrote:
> On 2/19/19 9:19 PM, John Ferlan wrote:
>> Assuming extraction (sigh) of the VIR_AUTOFREE,
>>
>> Reviewed-by: John Ferlan
>>
>
> While I'd definitely want this to be split into two patches if it was
> fixing something under src/, but this is under te
On Tue, 2019-02-19 at 16:59 +, Daniel P. Berrangé wrote:
> On Tue, Feb 19, 2019 at 05:52:30PM +0100, Andrea Bolognani wrote:
> > virFileWrapperFdFree(), like all free functions, is supposed
> > to only release allocated resources, so error reporting is
> > better suited for virFileWrapperFdClos
On Wed, Feb 20, 2019 at 01:16:34PM +0100, Andrea Bolognani wrote:
> On Tue, 2019-02-19 at 17:01 +, Daniel P. Berrangé wrote:
> > On Tue, Feb 19, 2019 at 05:52:29PM +0100, Andrea Bolognani wrote:
> [...]
> > > @@ -3231,6 +3231,7 @@ qemuDomainSaveMemory(virQEMUDriverPtr driver,
> > >
> > > cl
On Tue, 2019-02-19 at 17:01 +, Daniel P. Berrangé wrote:
> On Tue, Feb 19, 2019 at 05:52:29PM +0100, Andrea Bolognani wrote:
[...]
> > @@ -3231,6 +3231,7 @@ qemuDomainSaveMemory(virQEMUDriverPtr driver,
> >
> > cleanup:
> > VIR_FORCE_CLOSE(fd);
> > +qemuFileWrapperFDClose(vm, wrapp
Signed-off-by: Daniel P. Berrangé
---
po/id.mini.po| 20 +
po/id.po | 153 ---
po/ilo.mini.po | 20 +
po/ilo.po| 153 ---
po/is.mini.po| 20 +
p
The element in zanata.xml is no longer relevant as this info
is recorded server side.
Signed-off-by: Daniel P. Berrangé
---
po/zanata.xml | 99 ---
1 file changed, 99 deletions(-)
diff --git a/po/zanata.xml b/po/zanata.xml
index 70cc22a..042c695
Signed-off-by: Daniel P. Berrangé
---
po/af.mini.po| 20 ++
po/af.po | 153 --
po/am.mini.po| 20 ++
po/am.po | 153 --
po/anp.mini.po | 19 ++
po/anp.po| 152 -
Signed-off-by: Daniel P. Berrangé
---
po/sq.mini.po | 20 ++
po/sq.po| 153 ---
po/sr.mini.po | 21 ++
po/sr.po| 154
po/s...@latin.mini.po | 21 ++
po/s...@lati
This applies the same improvements previously done in libvirt:
https://www.redhat.com/archives/libvir-list/2018-April/msg01004.html
https://www.berrange.com/posts/2018/11/29/improved-translation-po-file-handling-by-ditching-gettext-autotools-integration/
The key problems with our current app
Add rules to handle pushing libvirt-glib.pot to zanata, and refreshing .po
files with new content from zanata.
Signed-off-by: Daniel P. Berrangé
---
po/Makefile.am | 8
1 file changed, 8 insertions(+)
diff --git a/po/Makefile.am b/po/Makefile.am
index 279fc47..d37014b 100644
--- a/po/M
Signed-off-by: Daniel P. Berrangé
---
po/{uk.po => uk.mini.po} | 152 +++---
po/ur.mini.po| 20 +
po/ur.po | 153 ---
po/vi.mini.po| 20 +
po/vi.po | 153 -
Signed-off-by: Daniel P. Berrangé
---
po/{fr.po => fr.mini.po} | 156 +++
po/gl.mini.po| 20 +
po/gl.po | 153 --
po/gu.mini.po| 20 +
po/gu.po | 153 -
Signed-off-by: Daniel P. Berrangé
---
po/mn.mini.po | 20 +++
po/mn.po | 153 -
po/mr.mini.po | 20 +++
po/mr.po | 153 -
po/ms.mini.po | 20 +++
po/ms.po | 153
Historically we have relied on intltool to install a standard
po/Makefile.in.in which has very limited scope for customization.
intltool is deprecated in favour of standard gettextize tools,
but these share the same disadvantages.
Writing make rules for po file management is no more difficult
than
Signed-off-by: Daniel P. Berrangé
---
po/de.mini.po | 20 +
po/de.po | 153
po/el.mini.po | 20 +
po/el.po | 153
po/{en_GB.po => en_GB.mini.po
Signed-off-by: Daniel P. Berrangé
---
po/or.mini.po | 20 +
po/or.po | 153
po/pa.mini.po | 20 +
po/pa.po | 153
po/{pl.po => pl.mini.po}
Signed-off-by: Daniel P. Berrangé
---
po/bn.mini.po| 20 +
po/bn.po | 153 -
po/bo.mini.po| 20 +
po/bo.po | 153 -
po/br.mini.po| 20 +
po/br
Hi,
I created a patch for libvirt-php to support virDomainUndefineFlags and
create snapshot from XML string.
The test case file is
"examples/domain_undefine_flags_and_snapshot_create_xml_test_case.php".
Hoping this is useful for libvirt-php.
>From e753d0014e8ce8cbcafb0fbd92159a6cc9f32168 Mon Se
Changelog:
+ RFC -> v1:
- Remove the event handle in the error path of nodeStateInitializeEnumerate
- Added patch 2: wake up udev thread in case of an error
Marc Hartmayer (2):
udev: nodeStateInitializeEnumerate: remove watch handle in case of an
error
udev: wake up the udev thread fo
If the udev thread is stopped, it must be ensured that the watch
handle is also removed from the main loop.
Signed-off-by: Marc Hartmayer
---
src/node_device/node_device_udev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/node_device/node_device_udev.c
b/src/node_device/node_device
Signal the udev thread the change of `priv->threadQuit` by using the
thread condition.
Signed-off-by: Marc Hartmayer
---
src/node_device/node_device_udev.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/node_device/node_device_udev.c
b/src/node_device/node_device_udev.c
index 4dd2e9004
On Tue, Feb 12, 2019 at 09:46 PM +0100, John Ferlan wrote:
> On 2/7/19 11:08 AM, Marc Hartmayer wrote:
>> Commit "nodedev: Move device enumumeration out of nodeStateInitialize"
>> (9f0ae0b18e3e620) has moved the heavy task of device enumeration into
>> a separate thread. The problem with this comm
Replace virDomainChrSourceDefFree with virObjectUnref.
Signed-off-by: Marc Hartmayer
Reviewed-by: Boris Fiuczynski
---
cfg.mk| 1 -
src/conf/domain_conf.c| 17 +
src/conf/domain_conf.h| 1 -
src/libvirt_private.syms | 1 -
src/qemu
Marc Hartmayer (2):
qemu: Use refcounting for priv->monConfig
qemu: Replace virDomainChrSourceDefFree with virObjectUnref
cfg.mk| 1 -
src/conf/domain_conf.c| 17 +
src/conf/domain_conf.h| 1 -
src/libvirt_private.syms | 1 -
src
82 matches
Mail list logo
