An error will be reported even if loading partially worked (for
example when the system DB could be loaded, but the user DB failed
to load). Since the code will still work correctly even if the DB
could not be loaded at all, we can ignore the error when DB loading
failed. (code still works but
All memory was not cleaned up properly upon exit, and a string
could be leaked from guess_os_from_disk when we issue a
continue/break early during the loop iteration.
---
examples/virtxml.c | 17 ++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/examples/virtxml.c
virtxml is using osinfo_db_guess_os_from_media for that which
is deprecated.
---
examples/virtxml.c | 11 ---
1 file changed, 4 insertions(+), 7 deletions(-)
diff --git a/examples/virtxml.c b/examples/virtxml.c
index 18b0456..d719e5f 100644
--- a/examples/virtxml.c
+++
OsinfoLoader is declared as a global variable in virtxml.c but
is only used in one function, so it can be made local.
---
examples/virtxml.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/examples/virtxml.c b/examples/virtxml.c
index 2918ee0..18b0456 100644
---
On Tue, Dec 18, 2012 at 02:26:53PM +0100, al...@redhat.com wrote:
From: Alexander Larsson al...@redhat.com
diff --git
a/libvirt-gconfig/libvirt-gconfig-domain-smartcard-host-certificates.h
b/libvirt-gconfig/libvirt-gconfig-domain-smartcard-host-certificates.h
new file mode 100644
index
Ping?
Christophe
On Fri, Dec 14, 2012 at 05:03:37PM +0100, Christophe Fergeau wrote:
This guarantees that the string will be nul-terminated. Coverity
warned about this issue.
---
libvirt-gobject/libvirt-gobject-connection.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
I've fixed the space-before-parens nit and pushed this, thanks for the
patch!
Christophe
On Wed, Dec 19, 2012 at 12:21:44PM +0100, Christophe Fergeau wrote:
On Tue, Dec 18, 2012 at 02:26:53PM +0100, al...@redhat.com wrote:
From: Alexander Larsson al...@redhat.com
diff --git
a/libvirt
fetch_list implementations in libvirt-gobject-connection.c and
libvirt-gobject-storage-pool.c can misbehave in error situations
or when the call is cancelled:
- when the call is cancelled, 'lst' will be NULL and 'n' non-0 so
we'll try to iterate over 'lst', which will cause a crash
- when
This guarantees that the string will be nul-terminated. Coverity
warned about this issue.
---
libvirt-gobject/libvirt-gobject-connection.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libvirt-gobject/libvirt-gobject-connection.c
On Tue, Dec 11, 2012 at 09:30:53PM +0100, Peter Krempa wrote:
This patch gets rid of the undeterministic error reporting code done on
return values of get(pw|gr)nam_r. With this patch, if the group record
is not returned by the corresponding function this error is not
considered fatal even if
On Fri, Dec 07, 2012 at 01:40:20AM +0800, Osier Yang wrote:
How about make the logic a bit more clear like:
if (!pw) {
if ((rc == 0) || (rc == EINTR) || (rc == EIO) ||
(rc == EMFILE) || (rc == ENFILE) || (rc == ENOMEM)) {
This does not look right, we want to run the
On Thu, Dec 06, 2012 at 03:07:32PM +0100, Peter Krempa wrote:
On 12/05/12 13:03, Christophe Fergeau wrote:
virGetUserIDByName is documented as returning 1 if the username
cannot be found. getpwnam_r is documented as returning:
« 0 or ENOENT or ESRCH or EBADF or EPERM or ... The given name
virGetUserIDByName is documented as returning 1 if the username
cannot be found. getpwnam_r is documented as returning:
« 0 or ENOENT or ESRCH or EBADF or EPERM or ... The given name
or uid was not found. »
and that:
« The formulation given above under RETURN VALUE is from POSIX.1-2001.
It does
virGetGroupIDByName is documented as returning 1 if the groupname
cannot be found. getgrnam_r is documented as returning:
« 0 or ENOENT or ESRCH or EBADF or EPERM or ... The given name
or gid was not found. »
and that:
« The formulation given above under RETURN VALUE is from POSIX.1-2001.
It
Here is the v2 of this patch series.
Changes since v1:
- return an error only for documented error values returnef from
getpwnam_r/getgrnam_r instead of trying to guess what error values
mean 'entry not found'
- this addresses the indenting comment raised by Osier
--
libvir-list mailing list
On Thu, Dec 06, 2012 at 05:01:22PM +0100, Peter Krempa wrote:
On 12/06/12 15:37, Christophe Fergeau wrote:
virGetUserIDByName is documented as returning 1 if the username
cannot be found. getpwnam_r is documented as returning:
« 0 or ENOENT or ESRCH or EBADF or EPERM or ... The given name
virGetGroupIDByName is documented as returning 1 if the groupname
cannot be found. getgrnam_r is documented as returning:
« 0 or ENOENT or ESRCH or EBADF or EPERM or ... The given name
or gid was not found. »
and that:
« The formulation given above under RETURN VALUE is from POSIX.1-2001.
It
Hey,
I've been having issues recently after upgrading to a newer libvirt on a
RHEL6 box. 'virsh start vmname' fails with
'error :internal error Failed to get user record for name '107': No such
file or directory '
Looking deeper into this, this comes from 'user not found' errors from
virGetUserIDByName is documented as returning 1 if the username
cannot be found. getpwnam_r is documented as returning:
« 0 or ENOENT or ESRCH or EBADF or EPERM or ... The given name
or uid was not found. »
and that:
« The formulation given above under RETURN VALUE is from POSIX.1-2001.
It does
On Thu, Nov 22, 2012 at 11:32:38AM +0100, Ján Tomko wrote:
On 11/22/12 11:03, Martin Kletzander wrote:
I'll rework it to make it as URI as possible then. Just one question,
though (for anyone, I guess); why do we append port as a parameter for
spice scheme?
Martin
I think
On Wed, Nov 21, 2012 at 08:04:36PM +0200, Zeeshan Ali (Khattak) wrote:
From: Zeeshan Ali (Khattak) zeesha...@gnome.org
Without this patch, storage pool still lists the volume even after it is
deleted.
Related Boxes bug: https://bugzilla.gnome.org/show_bug.cgi?id=688724
---
On Wed, Nov 21, 2012 at 09:30:02AM +0100, Michal Privoznik wrote:
On 20.11.2012 19:51, Zeeshan Ali (Khattak) wrote:
The latter one has been already checked as can be seen in context of the
last chunk. The first one should not fail since it is us who inserted
the value into volume-priv-vol;
On Tue, Nov 20, 2012 at 08:51:23PM +0200, Zeeshan Ali (Khattak) wrote:
From: Zeeshan Ali (Khattak) zeesha...@gnome.org
Without this patch, storage pool still lists the volume even after it is
deleted.
Related Boxes bug: https://bugzilla.gnome.org/show_bug.cgi?id=688724
---
virConnectGetVersion can return 0 'if the version can't be extracted
by lack of capabilities'. I've also seen it return 0 when libvirtd
1.0 is running but virsh 0.10.2 is used to get the version.
gvir_connection_get_version is documented as returning 0 on error.
However, currently it only sets the
On Thu, Nov 15, 2012 at 11:37:32AM +0100, Ján Tomko wrote:
If the domain can't be looked up, name is used unitialized after the
cleanup label.
Found by coverity.
---
tools/virsh-snapshot.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/virsh-snapshot.c
On Fri, Nov 16, 2012 at 12:10:10PM +0100, Ján Tomko wrote:
On 11/16/12 11:58, Christophe Fergeau wrote:
Shouldn't we call vshError with a different message in the !ret !name
case?
Christophe
vshCommandOptDomain and vshLookupSnapshot already give reasonable error
messages in that case
Hey,
On Wed, Nov 14, 2012 at 03:45:49PM +0100, Michal Privoznik wrote:
On 14.11.2012 15:16, Christophe Fergeau wrote:
This method returns a version number, which can be 0, so we cannot
check the return value for 0 to know if an error happened. Test if
the GError is set instead to detect
On Wed, Nov 14, 2012 at 01:12:39AM +0100, Zeeshan Ali (Khattak) wrote:
On Tue, Nov 13, 2012 at 8:06 PM, Christophe Fergeau cferg...@redhat.com
wrote:
On Tue, Nov 13, 2012 at 07:22:35PM +0100, Zeeshan Ali (Khattak) wrote:
On Tue, Nov 13, 2012 at 7:10 PM, Alexander Larsson al...@redhat.com
On Tue, Nov 13, 2012 at 06:45:56PM +0100, Zeeshan Ali (Khattak) wrote:
On Tue, Nov 6, 2012 at 1:45 PM, Christophe Fergeau cferg...@redhat.com
wrote:
gvir_storage_pool_refresh must be called and must be successful before
trying to use gvir_storage_pool_get_volume
On Tue, Nov 13, 2012 at 06:30:26PM +0100, Christophe Fergeau wrote:
This series does misc leak fixes in libvirt-glib, as well as an API doc
fix, and a deprecation warning fix.
Pushed all of this except 4/5, I've split the g_type_init() one into 2
separate commits.
Christophe
On Tue, Nov 13, 2012 at 06:50:11PM +0100, Zeeshan Ali (Khattak) wrote:
On Tue, Nov 6, 2012 at 1:41 PM, Christophe Fergeau cferg...@redhat.com
wrote:
They are similar to g_warning/g_critical, but also log the last
libvirt error. They are meant to be called right after a libvirt
call failed
On Wed, Nov 14, 2012 at 02:05:11PM +0100, Zeeshan Ali (Khattak) wrote:
On Wed, Nov 14, 2012 at 10:31 AM, Christophe Fergeau
cferg...@redhat.com wrote:
On Tue, Nov 13, 2012 at 06:45:56PM +0100, Zeeshan Ali (Khattak) wrote:
On Tue, Nov 6, 2012 at 1:45 PM, Christophe Fergeau cferg...@redhat.com
g_type_init has been deprecated in glib 2.35, not 2.34. With versions
older than 2.35, we have to call it or we'll get a runtime failure.
---
libvirt-gconfig/libvirt-gconfig-compat.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libvirt-gconfig/libvirt-gconfig-compat.h
This method returns a version number, which can be 0, so we cannot
check the return value for 0 to know if an error happened. Test if
the GError is set instead to detect errors.
---
examples/conn-test.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/examples/conn-test.c
When GLib deprecated g_mutex_new/g_mutex_free, we introduced a
compatibility wrapper to implement these using non-deprecated
functions. This was done by allocating 0-filled memory by the
mutex, and then letting GLib initialize the structure when
needed. However, we must call g_mutex_clear when
Hey,
On Wed, Nov 14, 2012 at 03:45:00PM +0100, Michal Privoznik wrote:
On 14.11.2012 15:16, Christophe Fergeau wrote:
g_type_init has been deprecated in glib 2.35, not 2.34. With versions
older than 2.35, we have to call it or we'll get a runtime failure.
---
libvirt-gconfig/libvirt
Ping?
Christophe
On Tue, Nov 06, 2012 at 01:41:01PM +0100, Christophe Fergeau wrote:
They are similar to g_warning/g_critical, but also log the last
libvirt error. They are meant to be called right after a libvirt
call failed, in other cases g_warning/g_critical are still to be
used
---
This goes on top of the [libvirt-glib 1/4] gobject: don't try to use
pool's volumes before a successful refresh series.
Christophe
libvirt-gobject/libvirt-gobject-storage-pool.c | 15 +++
libvirt-gobject/libvirt-gobject-storage-pool.h | 2 ++
Ping for these 4 patches?
On Tue, Nov 06, 2012 at 01:45:12PM +0100, Christophe Fergeau wrote:
gvir_storage_pool_refresh must be called and must be successful before
trying to use gvir_storage_pool_get_volume,
gvir_storage_pool_get_volumes and gvir_storage_pool_create_volume.
As the storage
Without this change, gvir_storage_pool_get_volume segfaults when
trying to do the hash table lookup with a NULL 'name' key.
---
libvirt-gobject/libvirt-gobject-storage-pool.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libvirt-gobject/libvirt-gobject-storage-pool.c
, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Author: Christophe Fergeau cferg...@redhat.com
+ */
+
+#ifndef __LIBVIRT_GCONFIG_COMPAT_H__
+#define __LIBVIRT_GCONFIG_COMPAT_H__
+
+#include glib-object.h
+
+#if GLIB_CHECK_VERSION(2, 34, 0)
+#define g_type_init()
+#endif
+
+#endif
This series does misc leak fixes in libvirt-glib, as well as an API doc
fix, and a deprecation warning fix.
Christophe
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
When creating an object wrapping a libvirt object (GVirDomain,
GVirStoragePool, GVirStorageVol), libvirt-gobject gets a reference
to a libvirt object to be used as a handle, and then creates the wrapper
object by calling g_object_new(..., handle, handle, NULL);
However, the underlying libvirt
When GLib deprecated g_mutex_new/g_mutex_free, we introduced a
compatibility wrapper to implement these using non-deprecated
functions. This was done by allocating 0-filled memory by the
mutex, and then letting GLib initialize the structure when
needed. However, we must call g_mutex_clear when
---
libvirt-gobject/libvirt-gobject-stream.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/libvirt-gobject/libvirt-gobject-stream.c
b/libvirt-gobject/libvirt-gobject-stream.c
index f0e43d0..2cb8967 100644
--- a/libvirt-gobject/libvirt-gobject-stream.c
+++
On Tue, Nov 13, 2012 at 06:53:48PM +0100, Zeeshan Ali (Khattak) wrote:
On Tue, Nov 13, 2012 at 6:30 PM, Christophe Fergeau cferg...@redhat.com
wrote:
diff --git a/libvirt-gconfig/tests/test-domain-parse.c
b/libvirt-gconfig/tests/test-domain-parse.c
index 11880de..33fc7be 100644
On Tue, Nov 13, 2012 at 07:22:35PM +0100, Zeeshan Ali (Khattak) wrote:
On Tue, Nov 13, 2012 at 7:10 PM, Alexander Larsson al...@redhat.com wrote:
This is a wrapper for virDomainUpdateDeviceFlags.
Looks good. ACK.
diff --git a/libvirt-gobject/libvirt-gobject.sym
On Mon, Nov 05, 2012 at 02:29:27PM +0100, Michal Privoznik wrote:
On 05.11.2012 14:21, Christophe Fergeau wrote:
GVirObjectStream::finalize() logs a critical() if virStreamFinish
fails. This commit adds the libvirt error message to this critical
to ease debugging.
ACK
Actually I'm going
They are similar to g_warning/g_critical, but also log the last
libvirt error. They are meant to be called right after a libvirt
call failed, in other cases g_warning/g_critical are still to be
used.
---
libvirt-glib/libvirt-glib-error.c | 34 +++
gvir_storage_pool_refresh must be called and must be successful before
trying to use gvir_storage_pool_get_volume,
gvir_storage_pool_get_volumes and gvir_storage_pool_create_volume.
As the storage pool refresh can fail for reasons external to
libvirt/libvirt-gobject, the library user should check
---
libvirt-gobject/libvirt-gobject-storage-pool.c | 90 ++
libvirt-gobject/libvirt-gobject-storage-pool.h | 10 +++
libvirt-gobject/libvirt-gobject.sym| 6 ++
3 files changed, 106 insertions(+)
diff --git a/libvirt-gobject/libvirt-gobject-storage-pool.c
---
libvirt-gobject/libvirt-gobject-storage-pool.c | 90 ++
libvirt-gobject/libvirt-gobject-storage-pool.h | 11
libvirt-gobject/libvirt-gobject.sym| 3 +
3 files changed, 104 insertions(+)
diff --git a/libvirt-gobject/libvirt-gobject-storage-pool.c
---
libvirt-gobject/libvirt-gobject-storage-pool.c | 105 +
libvirt-gobject/libvirt-gobject-storage-pool.h | 11 +++
libvirt-gobject/libvirt-gobject.sym| 3 +
3 files changed, 119 insertions(+)
diff --git a/libvirt-gobject/libvirt-gobject-storage-pool.c
Looks good, ACK
Christophe
On Wed, Oct 31, 2012 at 11:51:54PM +0200, Zeeshan Ali (Khattak) wrote:
From: Zeeshan Ali (Khattak) zeesha...@gnome.org
Add flags to be used with gvir_domain_reboot().
---
libvirt-gobject/libvirt-gobject-domain.c | 2 +-
libvirt-gobject/libvirt-gobject-domain.h
GVirObjectStream::finalize() logs a critical() if virStreamFinish
fails. This commit adds the libvirt error message to this critical
to ease debugging.
---
libvirt-gobject/libvirt-gobject-stream.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git
After upgrading to libvirt 1.0.0, I've started getting warnings
from libvirt:
(gnome-boxes:5965): Libvirt.GObject-CRITICAL **: cannot finish stream:
internal error libvirt_iohelper: unable to fsync stdout: Invalid argument
On further investigation, in iohelper.c:runIO(), fdatasync can get called
On Mon, Oct 22, 2012 at 02:59:14PM -0600, Eric Blake wrote:
On 10/22/2012 03:49 AM, Christophe Fergeau wrote:
When using --without-$name --without-secdriver-$name with $name being
selinux or apparmor, configure will fail saying that AppArmor/SELinux
development package must be installed
When using --without-$name --without-secdriver-$name with $name being
selinux or apparmor, configure will fail saying that AppArmor/SELinux
development package must be installed.
This is caused by a small bug in --with-secdriver-$name handling in
configure.ac which treats --without-secdriver-$name
When using --without-apparmor --without-secdriver-apparmor, configure
will fail saying that AppArmor development package must be installed.
This is caused by a small bug in --with-secdriver-apparmor handling in
configure.ac which treats --without-secdriver-apparmor as if the user
had requested to
self NACK, a similar issue seems to be present a bit further in
configure.ac, I'll have to look some more and send a v2
Christophe
On Sun, Oct 21, 2012 at 08:38:49PM +0200, Christophe Fergeau wrote:
When using --without-apparmor --without-secdriver-apparmor, configure
will fail saying
Thanks for figuring out why it was not working for you initially!
Especially as this was hiding 2 bugs (one in libvirt documentation and one
in this patch).
Looks good, as for the very long function prototype lines, I sometimes
split them right after the return value:
void
very_long_foo(Foo
with this library; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Authors:
+ *
+ * Zeeshan Ali (Khattak) zeesha...@gnome.org
+ * Christophe Fergeau cferg...@redhat.com
+ */
+
+#include config.h
+
+#include libvirt
On Wed, Oct 03, 2012 at 03:45:53PM +0200, Michal Privoznik wrote:
On 03.10.2012 14:23, Alexander Larsson wrote:
+gboolean gvir_domain_wakeup(GVirDomain *dom,
+guint flags,
+GError **err)
Maybe we could use gvir_domain_pmwakeup here;
On Fri, Sep 28, 2012 at 07:01:23PM +0200, Christophe Fergeau wrote:
Emit a signal when switching to the PMSUSPENDED state, and add
an enum entry to describe this state. This avoids runtime warnings
with newer libvirt.
This also breaks compilation with older libvirt, so we need to either bump
Looks good, ACK.
Christophe
On Wed, Oct 03, 2012 at 02:23:01PM +0200, Alexander Larsson wrote:
---
libvirt-gobject/libvirt-gobject-domain.c | 87
libvirt-gobject/libvirt-gobject-domain.h | 11
libvirt-gobject/libvirt-gobject.sym | 4 ++
3 files
The recent PMSuspend changes need libvirt 0.10.2 or newer.
---
configure.ac | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configure.ac b/configure.ac
index 369b368..c117106 100644
--- a/configure.ac
+++ b/configure.ac
@@ -9,7 +9,7 @@ AC_CANONICAL_HOST
The minimum libvirt version we need is duplicated in configure.ac
and libvirt.spec.in. Since libvirt.spec.in already gets some variables
substituted, we can also substitute LIBVIRT_REQUIRED to avoid the
duplication.
---
configure.ac | 1 +
libvirt-glib.spec.in | 2 +-
2 files changed, 2
Emit a signal when switching to the PMSUSPENDED state, and add
an enum entry to describe this state. This avoids runtime warnings
with newer libvirt.
---
libvirt-gobject/libvirt-gobject-connection.c | 7 +++
libvirt-gobject/libvirt-gobject-domain.c | 11 +++
This avoids a runtime warning about this kind of event not
being handled.
---
libvirt-gobject/libvirt-gobject-connection.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libvirt-gobject/libvirt-gobject-connection.c
b/libvirt-gobject/libvirt-gobject-connection.c
index ad7aa07..9628989
---
libvirt-gconfig/libvirt-gconfig-domain-graphics-vnc.c | 10 ++
libvirt-gconfig/libvirt-gconfig-domain-graphics-vnc.h | 3 +++
libvirt-gconfig/libvirt-gconfig.sym | 2 ++
3 files changed, 15 insertions(+)
diff --git
---
libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.c | 8
libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.h | 1 +
libvirt-gconfig/libvirt-gconfig.sym | 2 ++
libvirt-gconfig/tests/test-domain-create.c | 1 +
4 files changed, 12 insertions(+)
Ping?
On Wed, Sep 12, 2012 at 04:22:04PM +0200, Christophe Fergeau wrote:
From: Jovanka Gulicoska jovanka.gulico...@gmail.com
The format string passed to g_set_error cannot be NULL or a runtime
warning will be printed.
---
libvirt-gobject/libvirt-gobject-stream.c | 6 --
1 file
Ping?
On Wed, Sep 12, 2012 at 03:55:25PM +0200, Christophe Fergeau wrote:
Hey,
This patch series implements a few more getters/setters for
GVirConfigDomainGraphicsVnc and GVirConfigDomainGraphicsSpice.
They will be useful in GNOME Boxes as the same is currently
achieved through XPath use
Ping?
Can this patch go as is, or do we want to go all the way to switching to
GInitable as Marc-André suggested in
http://www.mail-archive.com/libvir-list@redhat.com/msg59538.html ?
Christophe
On Tue, Jul 24, 2012 at 05:49:21PM +0200, Christophe Fergeau wrote:
For objects with a subtype 'type
On Tue, Sep 25, 2012 at 02:07:39PM +0200, Michal Privoznik wrote:
Hey,
libvirt supports listen on IP address or a network and I think we need
to distinguish these.
As I understand it, to listen on an IP address or a network, you'd use a
listen child node to the graphics node. This patch sets
On Tue, Sep 25, 2012 at 03:04:47PM +0200, Michal Privoznik wrote:
On 25.09.2012 14:47, Christophe Fergeau wrote:
It seems it's not necessarily an IP address but this can be a hostname if
the .rng is to be trusted. I had in the back of my mind the possibility of
adding _set_listen_address
On Tue, Sep 25, 2012 at 03:32:10PM +0100, Daniel P. Berrange wrote:
I wonder if we should simply not expose an API for the listen= attribute
at all. Only have APIs for reading/writing the listen elements [...]
IOW, I don't think applications should need to care about the listen=
attribute at
Ping ?
Christophe
On Wed, Sep 19, 2012 at 05:08:08PM +0200, Christophe Fergeau wrote:
Commit 9298bfbcb introduced code to detect if netcf is linked with
libnl1, and to prefer libnl1 over libnl3 when this is the case.
This behaviour can be disabled by setting LIBNL_CFLAGS to any value
ACK
Christophe
On Thu, Sep 20, 2012 at 04:18:34PM +0200, Michal Privoznik wrote:
---
diff to v1:
-move content from virtxml.pod to virtxml.c
-wording and spelling as pointed out by Christophe
.gitignore |1 +
examples/Makefile.am |9 +++
examples/virtxml.c
Hey,
On Thu, Sep 13, 2012 at 06:14:05AM -0600, Eric Blake wrote:
On 09/12/2012 10:46 PM, Osier Yang wrote:
On 2012年09月11日 06:14, Eric Blake wrote:
Recent spec file changes ensure that in distro situations, netcf
and libvirt will link against the same libnl in order to avoid
dumping core.
Commit 9298bfbcb introduced code to detect if netcf is linked with
libnl1, and to prefer libnl1 over libnl3 when this is the case.
This behaviour can be disabled by setting LIBNL_CFLAGS to any value,
including the empty string.
However, configure.ac sets LIBNL_CFLAGS to before attempting
libnl
On Tue, Sep 18, 2012 at 05:52:25PM +0200, Michal Privoznik wrote:
In fact, it's only bare skeleton for gtkdoc
This seems to be missing a call to GTK_DOC_CHECK in configure.ac, and
potentially DISTCHECK_CONFIGURE_FLAGS = --enable-gtk-doc=yes
in Makefile.am
Christophe
pgpoIIPtjwEJm.pgp
On Tue, Sep 18, 2012 at 05:52:28PM +0200, Michal Privoznik wrote:
---
.gitignore |1 +
examples/Makefile.am | 11
examples/virtxml.pod | 121
++
It's weird to have it in examples/ while one of the commit adds a
On Tue, Sep 18, 2012 at 05:52:23PM +0200, Michal Privoznik wrote:
and couple of fixes for clean RPM build.
ACK except for 2/6 and 5/6 (see my comments in the corresponding patches)
Christophe
pgpVgNqGFEI4a.pgp
Description: PGP signature
--
libvir-list mailing list
libvir-list@redhat.com
On Thu, Sep 13, 2012 at 04:04:28PM +0200, Michal Privoznik wrote:
to avoid their uninitialized usage.
ACK, I've sent a similar patch but forgot --subject-prefix for some
reason...
https://www.redhat.com/archives/libvir-list/2012-September/msg00804.html
Let's use yours.
Christophe
---
Hey,
Looks good, a few cosmetic comments below,
On Thu, Sep 13, 2012 at 04:04:29PM +0200, Michal Privoznik wrote:
In some cases telling OS version is redundant as ISO image
with specified OS is passed some arguments later as disk.
Don't require --os then.
---
examples/virtxml.c | 110
On Thu, Sep 13, 2012 at 04:04:31PM +0200, Michal Privoznik wrote:
Users can choose between minimum and recommended values
for VCPU count and amount of RAM. Moreover, recommended
values should inherit defaults from the minimum.
---
examples/virtxml.c | 12
Looks good, ACK (would have been easier to review with the load_osinfo
changes in a separate patch, this would have halved the size of the patch
doing real code changes I think).
Christophe
On Thu, Sep 13, 2012 at 04:04:30PM +0200, Michal Privoznik wrote:
as in nearly all cases users will
On Fri, Sep 14, 2012 at 11:05:05AM -0600, Eric Blake wrote:
On 09/12/2012 09:42 AM, Christophe Fergeau wrote:
Fix copied from libvirt, commit by Eric Blake.
glibc 2.15 (on Fedora 17) coupled with explicit disabling of
optimization during development dies a painful death:
/usr
Hey,
On Wed, Sep 12, 2012 at 05:09:32PM +0200, Michal Privoznik wrote:
to address Chritophe's improvements hints.
Ah thanks for polishing these small nits, series looks good to me too.
Christophe
pgpWJyI3DbVfX.pgp
Description: PGP signature
--
libvir-list mailing list
libvir-list@redhat.com
Hey,
Series looks good to me, ACK to both patches.
Christophe
On Wed, Sep 12, 2012 at 07:42:56PM +0200, Michal Privoznik wrote:
which is basically a wrapper for virConnectGetVersion().
---
examples/conn-test.c | 13 +
Hey,
I should have looked more in depth at these commits earlier, sorry for the
late feedback ;)
On Mon, Sep 10, 2012 at 03:58:25PM +0200, Michal Privoznik wrote:
diff --git a/libvirt-designer/libvirt-designer-internal.h
b/libvirt-designer/libvirt-designer-internal.h
new file mode 100644
On Mon, Sep 10, 2012 at 03:58:28PM +0200, Michal Privoznik wrote:
Let users add NICs to domains.
---
examples/virtxml.c | 78 ++
libvirt-designer/libvirt-designer-domain.c | 83
On Mon, Sep 10, 2012 at 03:58:26PM +0200, Michal Privoznik wrote:
+static GList *
+gvir_designer_domain_get_supported_disk_bus_types(GVirDesignerDomain *design)
+{
+GVirDesignerDomainPrivate *priv = design-priv;
+OsinfoDeviceList *dev_list;
+GHashTable *bus_hash =
On Wed, Sep 12, 2012 at 11:48:31AM +0200, Michal Privoznik wrote:
+osinfo_loader = osinfo_loader_new();
+osinfo_loader_process_default_path(osinfo_loader, err);
+if (err)
+return FALSE;
I'm not sure we want to error out there, see
On Tue, Sep 11, 2012 at 07:42:41PM +0200, Michal Privoznik wrote:
which is basically a wrapper for virConnectGetType().
---
examples/conn-test.c | 12 ++
libvirt-gobject/libvirt-gobject-connection.c | 53
++
On Tue, Sep 11, 2012 at 07:42:42PM +0200, Michal Privoznik wrote:
+gboolean
+gvir_connection_get_version(GVirConnection *conn,
+gulong *version,
+GError **err)
gulong gvir_connection_get_version(GVirConnection *conn, GError **error)
Hey,
I've pushed your 2 patches and added your name to the AUTHORS file, thanks
for the patches!
Christophe
On Mon, Sep 10, 2012 at 04:12:43PM +0300, Timo Juhani Lindfors wrote:
---
examples/config-demo.py |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git
Implement gvir_config_domain_graphics_vnc_set_listen and
gvir_config_domain_graphics_vnc_set_socket.
---
libvirt-gconfig/libvirt-gconfig-domain-graphics-vnc.c | 19 +++
libvirt-gconfig/libvirt-gconfig-domain-graphics-vnc.h | 6 ++
libvirt-gconfig/libvirt-gconfig.sym
---
libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.c | 16
libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.h | 2 ++
libvirt-gconfig/libvirt-gconfig.sym | 2 ++
libvirt-gconfig/tests/test-domain-create.c | 2 ++
4 files changed, 22
701 - 800 of 1510 matches
Mail list logo
