On 03/12/2014 02:42 PM, Peter Krempa wrote:
A backing chain of 3 files (base - mid - top) in the local file system:
disk type='file' device='disk'
driver name='qemu' type='qcow2'/
source file='/var/lib/libvirt/images/top.qcow2'/
backingStore type='file'
driver
Commit 6b306d66 converted virHostdevManager to a virObject, but
missed adding a virObject field to the virHostdevManager struct.
Result is memory corruption when taking a reference on an instance
of the object, where atomic inc is done on the stateDir field.
Later use of stateDir crashes libvirtd.
Stefan Bader wrote:
On 12.03.2014 13:08, Ian Campbell wrote:
On Wed, 2014-03-12 at 13:03 +0100, Stefan Bader wrote:
I have been looking into a bug report (see BugLink) which reported
libvirt to fail starting inside a Xen guest. Upon further investigation
I found that some tools that
2014-03-12 23:12 GMT+08:00 Daniel P. Berrange berra...@redhat.com:
On Fri, Mar 07, 2014 at 06:53:09PM +0800, Chunyan Liu wrote:
Signed-off-by: Chunyan Liu cy...@suse.com
---
src/libvirt_private.syms | 8
src/qemu/qemu_hostdev.c | 16 +++
src/util/virhostdev.c
2014-03-12 23:11 GMT+08:00 Daniel P. Berrange berra...@redhat.com:
On Fri, Mar 07, 2014 at 06:53:12PM +0800, Chunyan Liu wrote:
Check NULL parameter inputs
Signed-off-by: Chunyan Liu cy...@suse.com
---
src/util/virhostdev.c | 57
+++
This is a followup to
https://www.redhat.com/archives/libvir-list/2014-March/msg00350.html
Eric had ACKed all of those patches, but pointed out some
omissions. The first patch of this series should be squashed into 1/9
of the previous series (which I haven't yet pushed), and the others
are
This is also never seen by a human.
---
src/qemu/qemu_migration.c | 49 ---
1 file changed, 29 insertions(+), 20 deletions(-)
diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c
index 54c6fec..0bbec2e 100644
---
These are never seen externally, only passed into libvirt APIs, so in
practice this makes no real difference, but it's good to be
consistent.
---
tools/virsh-domain.c | 43 ---
tools/virsh-pool.c | 40 +---
This should be squashed into the first patch of the previous
eliminate hardcoded indent series (which was ACKed but hasn't yet
been pushed):
https://www.redhat.com/archives/libvir-list/2014-March/msg00361.html
The initial patch neglected to change the functions that format
private metadata in
This same set of functions is used by the qemu, xen, and lxc drivers'
connectSysInfo function.
---
src/util/virsysinfo.c | 72 ---
1 file changed, 40 insertions(+), 32 deletions(-)
diff --git a/src/util/virsysinfo.c b/src/util/virsysinfo.c
index
---
src/qemu/qemu_capabilities.c | 20 +++-
1 file changed, 11 insertions(+), 9 deletions(-)
diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c
index c589a57..47f8bcb 100644
--- a/src/qemu/qemu_capabilities.c
+++ b/src/qemu/qemu_capabilities.c
@@ -2539,38
This rule wouldn't be able to find any case of a hardcoded indent that
was in the middle of a string, but then virBuffer doesn't add
indentation in the middle of a string either.
---
cfg.mk | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/cfg.mk b/cfg.mk
index
On 03/12/2014 05:26 PM, Jim Fehlig wrote:
Commit 6b306d66 converted virHostdevManager to a virObject, but
missed adding a virObject field to the virHostdevManager struct.
Result is memory corruption when taking a reference on an instance
of the object, where atomic inc is done on the stateDir
101 - 113 of 113 matches
Mail list logo