On 6/25/20 4:18 PM, Jonathon Jongsma wrote:
Changes in this version:
- Add the test case input file
- modify the test itself to properly fail when an input file is missing.
Jonathon Jongsma (2):
qemu: ramfb video device doesn't support PCI address
tests: ensure failure if input file do
On 6/25/20 7:34 PM, Ján Tomko wrote:
On a Wednesday in 2020, Laine Stump wrote:
OOPS!!
I meant to squash this into patch 10 before posting. If you want to
just review it separately I can squash it in before push. Or if you
want to be pedantic I can squash it in and resend :-)
To me, these
On 6/25/20 7:08 PM, Ján Tomko wrote:
On a Wednesday in 2020, Laine Stump wrote:
The only reason for the error label in this function is to call
virBufferFreeAndReset(). It's actually more common for a failed format
function to just leave the virBuffer alone and let the caller free it
when there
On 6/25/20 7:17 PM, Ján Tomko wrote:
The cleanup function was already added by:
commit 2ad0284627ea3d6c123e0a266b9c7bb00aea4576
CommitDate: 2018-07-27 17:21:04 +0200
util: firewall: define cleanup function using VIR_DEFINE_AUTOPTR_FUNC
On a Wednesday in 2020, Laine Stump wrote:
Put in a se
On a Wednesday in 2020, Laine Stump wrote:
OOPS!!
I meant to squash this into patch 10 before posting. If you want to
just review it separately I can squash it in before push. Or if you
want to be pedantic I can squash it in and resend :-)
To me, these seem like changes unrelated to patch
On a Wednesday in 2020, Laine Stump wrote:
This includes those that use plain VIR_FREE() as well as those that
have a cleanup function defined for use via g_auto/g_autoptr
(virCommand, virFirewall, virBuffer, virJSONValue etc).
Signed-off-by: Laine Stump
---
src/network/bridge_driver.c |
On a Wednesday in 2020, Laine Stump wrote:
The 2nd arg to this function is a bool, not an int.
Signed-off-by: Laine Stump
---
src/network/bridge_driver.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
Reviewed-by: Ján Tomko
Jano
signature.asc
Description: PGP signature
On a Wednesday in 2020, Laine Stump wrote:
This struct isn't used anywhere else.
Signed-off-by: Laine Stump
---
src/network/bridge_driver.c | 10 ++
src/network/bridge_driver.h | 9 -
2 files changed, 10 insertions(+), 9 deletions(-)
Reviewed-by: Ján Tomko
Jano
signature.a
The cleanup function was already added by:
commit 2ad0284627ea3d6c123e0a266b9c7bb00aea4576
CommitDate: 2018-07-27 17:21:04 +0200
util: firewall: define cleanup function using VIR_DEFINE_AUTOPTR_FUNC
On a Wednesday in 2020, Laine Stump wrote:
Put in a separate patch so that two future patche
On a Wednesday in 2020, Laine Stump wrote:
hostsfilestr was not being freed. This will be turned into g_autofree
in an upcoming patch converting a lot more of the same file to using
g_auto*, but I wanted to make a separate patch for this first so the
other patch is simpler to review.
It also m
On a Wednesday in 2020, Laine Stump wrote:
The only reason for the error label in this function is to call
virBufferFreeAndReset(). It's actually more common for a failed format
function to just leave the virBuffer alone and let the caller free it
when there is a failure, and in fact the only cal
On a Wednesday in 2020, Laine Stump wrote:
The error: label in this function just does "return -1", so replace
all the "goto error" in the function with "return -1".
I split this out from virDomainDefParse quickly as a build breaker fix
and forgot to follow up with this cleanup.
Signed-off-b
On a Wednesday in 2020, Laine Stump wrote:
The strings allocated in virGetHostnameImpl() are all allocated via
g_strdup(), which will exit on OOM anyway, so the call to
virReportOOMError() is redundant, and removing it allows slight
modification to the code, in particular the cleanup label can be
On a Wednesday in 2020, Laine Stump wrote:
There were a few uses of xmlNodeGetContent() that didn't check for
NULL before using the result.
A NULL return from xmlNodeGetContent() *could* (probably does) mean
that there was an Out of Memory condition, but it is unclear from the
documentation if t
On a Wednesday in 2020, Laine Stump wrote:
virDomainBlkioDeviceParseXML() has multiple cases of sending the
return from xmlNodeGetContent() directly to virStrToLong_xx() without
checking for NULL. Although it is *very* rare for xmlNodeGetContent()
to return NULL (possibly it only happens in an OO
On a Wednesday in 2020, Laine Stump wrote:
AUTOPTR_CLEANUP_FUNC is set to xmlBufferFree() in util/virxml.h (This
is actually new - added accidentally (but fortunately harmlessly!) in
commit 257aba2dafe. I had added it along with the hunks in this patch,
then decided to remove it and submit separa
On a Wednesday in 2020, Laine Stump wrote:
Although libvirt itself uses g_malloc0() and friends, which exit when
there isn't enouogh memory, libxml2 uses standard malloc(), which just
returns NULL on OOM - this means we must check for NULL on return from
any libxml2 functions that allocate memory
When executing the hypervisor-cpu-compare/baseline commands and
the XML file contains a CPU definition using host-passthrough
and no model name, the commands will fail and return an error
message from the QMP response.
Let's fix this by checking for host-passthrough and a missing
model name when c
On Mon, 2020-06-22 at 18:18 +0200, Andrea Bolognani wrote:
> On Mon, 2020-06-22 at 09:43 -0500, Jonathon Jongsma wrote:
> > CentOS does not have the cppi package, so some code style checks
> > are
> > skipped. Switch to a openSUSE image to do the code style checks.
> >
> > Signed-off-by: Jonathon
Changes in this version:
- Add the test case input file
- modify the test itself to properly fail when an input file is missing.
Jonathon Jongsma (2):
qemu: ramfb video device doesn't support PCI address
tests: ensure failure if input file doesn't exist
src/qemu/qemu_validate.c
Although a ramfb video device is not a PCI device, we don't currently
report an error for ramfb device definitions containing a PCI address.
However, a guest configured with such a device will fail to start:
# virsh start test1
error: Failed to start domain test1
error: internal error:
When using the DO_TEST_PARSE_ERROR() macro, a failure to parse the input
file is considered a successful test. However, if the input file is
totally missing, that should be distinguished from a parsing error and
not be treated as a test success.
The function virDomainDefParseFile() simply returns
On Thu, 2020-06-18 at 10:25 +0200, Shalini Chellathurai Saroja wrote:
> Shalini Chellathurai Saroja (5):
> conf: use g_autofree to ensure automatic cleanup
> conf: fix zPCI address auto-generation on s390
> qemu: move ZPCI uid validation into device validation
> tests: qemu: add more tests
On Thu, 2020-06-18 at 10:25 +0200, Shalini Chellathurai Saroja wrote:
> Add test with a ZPCI host device and a CCW memballoon device to ensure
> that CCW address remains the default address assigned.
>
> Signed-off-by: Boris Fiuczynski
> Signed-off-by: Shalini Chellathurai Saroja
> Reviewed-by:
On Thu, 2020-06-18 at 10:25 +0200, Shalini Chellathurai Saroja wrote:
> 1. Test for auto-generating uids while specifying valid fids
> 2. Test for auto-generating fids while specifying valid uids
> 3. Test for parse error while specifying a valid fid and an invalid
>uid
> 4. Test for parse erro
On Thu, 2020-06-18 at 10:25 +0200, Shalini Chellathurai Saroja wrote:
> qemu: move ZPCI uid validation into device validation
>
> The ZPCI device validation is specific to qemu. So, let us move the
> ZPCI uid validation out of domain xml parsing into qemu domain device
> validation.
We can just t
On Thu, 2020-06-18 at 10:25 +0200, Shalini Chellathurai Saroja wrote:
> @@ -129,7 +129,8 @@ static void
> virDomainZPCIAddressReleaseUid(virHashTablePtr set,
> virZPCIDeviceAddressPtr addr)
> {
> -virDomainZPCIAddressReleaseId(set, &addr->uid, "uid");
> +if
On Thu, 2020-06-18 at 10:25 +0200, Shalini Chellathurai Saroja wrote:
> Signed-off-by: Bjoern Walk
> Signed-off-by: Shalini Chellathurai Saroja
> Reviewed-by: Boris Fiuczynski
> ---
> src/conf/device_conf.c | 17 ++---
> 1 file changed, 6 insertions(+), 11 deletions(-)
Reviewed-by:
On Thu, 2020-06-25 at 17:14 +0100, Daniel P. Berrangé wrote:
> +.git_native_build_job_template: &git_native_build_job_definition
> + image: $CI_REGISTRY_IMAGE/ci-$NAME:latest
> + stage: builds
> + cache:
> +paths:
> + - ccache/
> +key: "$CI_JOB_NAME"
> + before_script:
> +- *sc
On 6/25/20 12:30 PM, Jonathon Jongsma wrote:
On Thu, 2020-06-25 at 12:20 -0400, Laine Stump wrote:
On 6/25/20 10:34 AM, Jonathon Jongsma wrote:
Although a ramfb video device is not a PCI device, we don't
currently
report an error for ramfb device definitions containing a PCI
address.
However, a
On Thu, 2020-06-25 at 12:20 -0400, Laine Stump wrote:
> On 6/25/20 10:34 AM, Jonathon Jongsma wrote:
> > Although a ramfb video device is not a PCI device, we don't
> > currently
> > report an error for ramfb device definitions containing a PCI
> > address.
> > However, a guest configured with such
On 6/25/20 10:34 AM, Jonathon Jongsma wrote:
Although a ramfb video device is not a PCI device, we don't currently
report an error for ramfb device definitions containing a PCI address.
However, a guest configured with such a device will fail to start:
# virsh start test1
error: Failed
With the introduction of automated CI pipelines, we are now ready to switch
to using merge requests for the project. With this switch we longer wish
to have patches sent to the mailing list, and thus the git-publish
config is removed.
Reviewed-by: Andrea Bolognani
Signed-off-by: Daniel P. Berrang
The dbus build needs to validate one axis
- A variety of libvirt versions
We test a variety of libvirt versions by running a build against the
distro provided libvirt packages. All that is then missing is a build
against the latest libvirt git master, which only needs to be run on
a single dist
The virDomainSnapshotGetDomain() method does NOT increment the
refcount on the returned virDomain, so it must not be unrefed.
This double free is responsible for random failures of the
test_snapshot.py tests.
Reviewed-by: Andrea Bolognani
Signed-off-by: Daniel P. Berrangé
---
src/domainsnapsho
This introduces CI build coverage and then documents the switch to
use merge requests.
Daniel P. Berrangé (3):
src: fix double free of virDomain object
gitlab: introduce CI jobs testing git master & distro libvirt
gitlab: add CONTRIBUTING.rst file to indicate use of merge requests
.gitlab-
On Thu, Jun 25, 2020 at 11:01:48AM -0400, Laine Stump wrote:
> On 6/25/20 3:55 AM, Peter Krempa wrote:
> > On Wed, Jun 24, 2020 at 23:34:00 -0400, Laine Stump wrote:
> > > Signed-off-by: Laine Stump
> > > ---
> > > src/network/bridge_driver.c | 59 +
> > > 1
On 6/25/20 3:55 AM, Peter Krempa wrote:
On Wed, Jun 24, 2020 at 23:34:00 -0400, Laine Stump wrote:
Signed-off-by: Laine Stump
---
src/network/bridge_driver.c | 59 +
1 file changed, 33 insertions(+), 26 deletions(-)
diff --git a/src/network/bridge_driver.
On Thu, Jun 25, 2020 at 04:37:59PM +0200, Michal Privoznik wrote:
> When parsing domain XML post parse callbacks are run and one of
> them might try and call API from a non-hypervisor driver (e.g.
> just like qemuDomainDeviceNetDefPostParse() is doing - it calls a
> network API). To avoid this in t
On Thu, Jun 25, 2020 at 04:34:51PM +0200, Andrea Bolognani wrote:
> On Thu, 2020-06-25 at 15:20 +0100, Daniel P. Berrangé wrote:
> > On Thu, Jun 25, 2020 at 04:16:35PM +0200, Andrea Bolognani wrote:
> > > We should really try to figure out why the packages are absent from
> > > the repos, however.
When parsing domain XML post parse callbacks are run and one of
them might try and call API from a non-hypervisor driver (e.g.
just like qemuDomainDeviceNetDefPostParse() is doing - it calls a
network API). To avoid this in the test suite, set dummy drivers,
which renders all non-hypervisor APIs re
On Thu, 2020-06-25 at 15:20 +0100, Daniel P. Berrangé wrote:
> On Thu, Jun 25, 2020 at 04:16:35PM +0200, Andrea Bolognani wrote:
> > We should really try to figure out why the packages are absent from
> > the repos, however. Wasn't the PowerTools repo supposed to take care
> > of that? I'm pretty s
Although a ramfb video device is not a PCI device, we don't currently
report an error for ramfb device definitions containing a PCI address.
However, a guest configured with such a device will fail to start:
# virsh start test1
error: Failed to start domain test1
error: internal error:
On Thu, 2020-06-25 at 15:25 +0100, Daniel P. Berrangé wrote:
> On Thu, Jun 25, 2020 at 04:19:16PM +0200, Andrea Bolognani wrote:
> > All targets get pip3, which is now part of the base system, and
> > the number of native packages included in the MinGW containers is
> > significantly reduced.
>
>
On Thu, Jun 25, 2020 at 04:19:16PM +0200, Andrea Bolognani wrote:
> All targets get pip3, which is now part of the base system, and
> the number of native packages included in the MinGW containers is
> significantly reduced.
FWIW, I thinkit would be benefit to reduce and even eliminate
the base se
On Thu, Jun 25, 2020 at 04:16:35PM +0200, Andrea Bolognani wrote:
> On Thu, 2020-06-25 at 13:32 +0100, Daniel P. Berrangé wrote:
> > On Wed, Jun 24, 2020 at 07:16:16PM +0200, Andrea Bolognani wrote:
> > > What's the rationale for building libvirt and libvirt-glib from git
> > > on CentOS Stream in
On Thu, 2020-06-25 at 13:32 +0100, Daniel P. Berrangé wrote:
> On Wed, Jun 24, 2020 at 07:16:16PM +0200, Andrea Bolognani wrote:
> > What's the rationale for building libvirt and libvirt-glib from git
> > on CentOS Stream in addition to CentOS 8?
>
> There's something odd with the repos, probably
All targets get pip3, which is now part of the base system, and
the number of native packages included in the MinGW containers is
significantly reduced.
The corresponding libvirt-ci commit is 4ff697ba0b5d.
Signed-off-by: Andrea Bolognani
---
Pushed under the Dockerfile refresh rule.
ci/contain
On Wed, 2020-06-24 at 23:21 -0400, Laine Stump wrote:
> On 6/24/20 6:06 PM, Jonathon Jongsma wrote:
> > Although a ramfb video device is not a PCI device, we don't
> > currently
> > report an error for ramfb device definitions containing a PCI
> > address.
> > However, a guest configured with such
On Thu, Jun 25, 2020 at 03:54:18PM +0200, Michal Privoznik wrote:
> On 6/25/20 10:38 AM, Daniel P. Berrangé wrote:
> > On Thu, Jun 25, 2020 at 09:48:56AM +0200, Michal Privoznik wrote:
> > > A few commits back (in v6.4.0-131-gbdb8f2e418) the post parse
> > > function for domain interface was change
On 6/25/20 10:38 AM, Daniel P. Berrangé wrote:
On Thu, Jun 25, 2020 at 09:48:56AM +0200, Michal Privoznik wrote:
A few commits back (in v6.4.0-131-gbdb8f2e418) the post parse
function for domain interface was changed so that it doesn't fill
in model for hostdev types of interfaces (including net
On Thu, 2020-06-25 at 13:42 +0100, Daniel P. Berrangé wrote:
> Daniel P. Berrangé (2):
> gitlab: introduce CI jobs testing git master & distro libvirt
> gitlab: add CONTRIBUTING.rst file to indicate use of merge requests
>
> .gitlab-ci.yml| 114
On 6/25/20 3:48 AM, Michal Privoznik wrote:
A few commits back (in v6.4.0-131-gbdb8f2e418) the post parse
function for domain interface was changed so that it doesn't fill
in model for hostdev types of interfaces (including network type
interfaces which would end up hostdevs).
While the idea is
Switch to the new format for easier extension.
Signed-off-by: Peter Krempa
---
This was a surprisingly easy conversion done using:
pandoc --from html --to rst --toc --columns 80 --standalone
formatbackup.html.in -o formatbackup.rst
The file generated by pandoc required following manual tweak
The sandbox build needs to validate two axis
- A variety of distro versions
- A variety of libvirt versions
We test a variety of libvirt versions by running a build against the
distro provided libvirt packages. All that is then missing is a build
against the latest libvirt git master, which o
Daniel P. Berrangé (2):
gitlab: introduce CI jobs testing git master & distro libvirt
gitlab: add CONTRIBUTING.rst file to indicate use of merge requests
.gitlab-ci.yml| 114 ++
.gitpublish | 4 -
CONTRIBUTING
With the introduction of automated CI pipelines, we are now ready to switch
to using merge requests for the project. With this switch we longer wish
to have patches sent to the mailing list, and thus the git-publish
config is removed.
Reviewed-by: Andrea Bolognani
Signed-off-by: Daniel P. Berrang
On Wed, Jun 24, 2020 at 07:16:16PM +0200, Andrea Bolognani wrote:
> On Wed, 2020-06-24 at 17:26 +0100, Daniel P. Berrangé wrote:
> > The dbus build needs to validate one axis
> >
> > - A variety of libvirt versions
> >
> > We test a variety of libvirt versions by running a build against the
> >
On Wed, Jun 24, 2020 at 08:06:08PM +0200, Andrea Bolognani wrote:
> On Wed, 2020-06-24 at 17:20 +0100, Daniel P. Berrangé wrote:
> > diff --git a/.gitlab-ci.yml b/.gitlab-ci.yml
> > new file mode 100644
> > index 000..75d34c3
> > --- /dev/null
> > +++ b/.gitlab-ci.yml
>
> Oh and this doesn't a
On Wed, Jun 24, 2020 at 07:39:21PM +0200, Andrea Bolognani wrote:
> On Wed, 2020-06-24 at 17:20 +0100, Daniel P. Berrangé wrote:
> > The sandbox build needs to validate two axis
> >
> > - A variety of distro versions
> > - A variety of libvirt versions
> >
> > We test a variety of libvirt ver
On Wed, Jun 24, 2020 at 09:35:22 -0500, Eric Blake wrote:
> On 6/24/20 9:07 AM, Peter Krempa wrote:
> > oVirt does merge images out of libvirt in some cases. Add docs outlining
> > how it's done from a high level.
> >
> > Signed-off-by: Peter Krempa
> > ---
> > docs/kbase/incrementalbackupinter
On Wed, Jun 24, 2020 at 09:23:00 -0500, Eric Blake wrote:
> On 6/24/20 9:07 AM, Peter Krempa wrote:
> > Make it obvious what's meant by 'overlay' and 'backing image' for sake
> > of extension of the document.
> >
> > Signed-off-by: Peter Krempa
> > ---
> > docs/kbase/incrementalbackupinternals.
On Thu, 2020-06-25 at 10:25 +0200, Peter Krempa wrote:
> There were two upstream issues filed for the problem so it's worth
> mentioning.
>
> Signed-off-by: Peter Krempa
> ---
> NEWS.rst | 5 +
> 1 file changed, 5 insertions(+)
Reviewed-by: Andrea Bolognani
--
Andrea Bolognani / Red Hat
On Thu, Jun 25, 2020 at 09:48:56AM +0200, Michal Privoznik wrote:
> A few commits back (in v6.4.0-131-gbdb8f2e418) the post parse
> function for domain interface was changed so that it doesn't fill
> in model for hostdev types of interfaces (including network type
> interfaces which would end up ho
There were two upstream issues filed for the problem so it's worth
mentioning.
Signed-off-by: Peter Krempa
---
NEWS.rst | 5 +
1 file changed, 5 insertions(+)
diff --git a/NEWS.rst b/NEWS.rst
index 2fc43c31b9..54b1e4a992 100644
--- a/NEWS.rst
+++ b/NEWS.rst
@@ -51,6 +51,11 @@ v6.5.0 (unrele
On Wed, Jun 24, 2020 at 23:34:00 -0400, Laine Stump wrote:
> Signed-off-by: Laine Stump
> ---
> src/network/bridge_driver.c | 59 +
> 1 file changed, 33 insertions(+), 26 deletions(-)
>
> diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
>
A few commits back (in v6.4.0-131-gbdb8f2e418) the post parse
function for domain interface was changed so that it doesn't fill
in model for hostdev types of interfaces (including network type
interfaces which would end up hostdevs).
While the idea is sound, the execution can be a bit better:
virD
67 matches
Mail list logo
