Currently, only LXC has hostdev mode 'capabilities' support,
so the other drivers should forbid to define it in XML.
The hostdev mode check is added to devicesPostParseCallback()
for each hypervisor driver.
But there are some drivers lack function devicesPostParseCallback(),
so only add check for
Since qemu has supported memory-backend-ram and memory-backend-file object,
so we should add a new 'memdev' device type in XML to introduce the
memory element, Its definition like the following:
ram0
1000
then we enable to support -numa memdev=ram0 command line for binding gue
Signed-off-by: Chen Fan
---
docs/formatdomain.html.in | 71 +++-
docs/schemas/domaincommon.rng | 76 +-
tests/qemuxml2argvdata/qemuxml2argv-cpu-numa3.args | 9 +++
tests/qemuxml2argvdata/qemuxml2argv-cpu-numa3.xml
Since qemu has supported numa option -memdev config:
-object
memory-backend-ram,size=1024M,policy=membind,host-nodes=0,id=ram-node0 \
-numa node,nodeid=0,cpus=0,memdev=ram-node0 \
for binding guest numa nodes to host numa nodes.
So we introduce this capability in libvirt by configuration
Since qemu has supported '-numa memdev=ram0' command option, so libvirt
should add numa element to support specified memdev attrubute in XML.
Signed-off-by: Chen Fan
---
src/conf/cpu_conf.c | 73 +++--
src/conf/cpu_conf.h | 13 -
src/qe
On 24.06.2014 21:34, Eric Blake wrote:
On 06/24/2014 03:39 AM, Daniel P. Berrange wrote:
We also have a subelement of in the
XML which is used for a similar thing although it doesn't support a
per-machine-type output, only per-binary capabilities. Should we add
this more granular approach a
On 06/24/14 21:34, Eric Blake wrote:
> On 06/24/2014 03:39 AM, Daniel P. Berrange wrote:
>
>>>
>>> We also have a subelement of in the
>>> XML which is used for a similar thing although it doesn't support a
>>> per-machine-type output, only per-binary capabilities. Should we add
>>> this more g
On Wed, Jun 25, 2014 at 09:57:32AM +0800, Chen Hanxiao wrote:
Signed-off-by: Chen Hanxiao
---
src/lxc/lxc_cgroup.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/lxc/lxc_cgroup.c b/src/lxc/lxc_cgroup.c
index 8dfdc60..39e30ad 100644
--- a/src/lxc/lxc_cgroup.c
+++ b/
Jim Fehlig wrote:
> Ian Campbell wrote:
>
>> Jim,
>>
>> What would be the right thing to do for this? The intention is for
>> osstest to skip migration tests when that feature is unimplemented by
>> the toolstack/xen/arch combination.
>>
>> It looks like the output of "virsh capabilities" indica
Signed-off-by: Chen Hanxiao
---
src/lxc/lxc_cgroup.c | 7 +--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/lxc/lxc_cgroup.c b/src/lxc/lxc_cgroup.c
index 8dfdc60..39e30ad 100644
--- a/src/lxc/lxc_cgroup.c
+++ b/src/lxc/lxc_cgroup.c
@@ -288,8 +288,11 @@ int virLXCCgroupGetM
> -Original Message-
> From: Michal Privoznik [mailto:mpriv...@redhat.com]
> Sent: Tuesday, June 24, 2014 9:38 PM
> To: Wangrui (K); libvir-list@redhat.com
> Subject: Re: [libvirt] [PATCH] LXC: update comments of
> lxcDomainCreateXMLWithFiles() and lxcDomainCreateXML()
>
>
> One does
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> To allow changing the name that is recorded in the overlay of the TOP
> image used in a block commit operation, we need to specify the backing
> name to qemu. This is done via the "backing-file" attribute to the
> block-commit command.
Yeah, I can see
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> The parent directory doesn't necessarily need to be stored after we
> don't mangle the path stored in the image. Remove it and tweak the code
> to avoid using it.
> ---
> src/storage/storage_driver.c | 11 ++-
> src/util/virstoragefile.c| 68
>
Ian Campbell wrote:
> On Fri, 2014-06-20 at 15:07 -0600, Jim Fehlig wrote:
>
>> The libxl driver currently sets the disk backend to
>> LIBXL_DISK_BACKEND_TAP when is specified
>> in the config. qdisk should be prefered with this
>> configuration, otherwise existing configuration such as the
>
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> Store backing chain paths as non-canonical. The canonicalization step
> will be already taken. This will allow to avoid storing unnecessary
> amounts of data.
> ---
> src/util/virstoragefile.c | 33 ++---
> tests/virstoraget
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> After we don't test relative paths, remove even more unnecessary cruft
> from the test code.
> ---
> tests/virstoragetest.c | 61
> +++---
> 1 file changed, 18 insertions(+), 43 deletions(-)
ACK.
--
Eric
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> libvirt always uses an absolute path to address the top image of an
> image chain. Our storage test tests also the relative path which won't
> ever be used. Additionally it makes the test more complicated.
It covers some interesting corner cases that m
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> Now that we store only relative names in virStorageSource's member
> relPath the backingRelative member is obsolete. Remove it and adapt the
> code to the removal.
> ---
> src/util/virstoragefile.c | 4 +---
> src/util/virstoragefile.h | 2 --
> tests/
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> Separately remove the now unused variable.
> ---
> tests/virstoragetest.c | 15 ---
> 1 file changed, 15 deletions(-)
ACK.
--
Eric Blake eblake redhat com+1-919-301-3266
Libvirt virtualization library http://libvirt.org
signatur
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> Due to various refactors and compatibility with the virstoragetest the
> relPath field of the virStorageSource structure was always filled either
> with the relative name or the full path in case of abslutely backed
s/abslutely/absolutely/
> storage.
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> Now that we changed ordering of the stored metadata so that the backing
> store is described by the child element the test should reflect this
> change too.
>
> Remove the expected backing store field as it's actually described by
> the next element in
On 06/19/2014 07:59 AM, Peter Krempa wrote:
> This patch introduces a function that will allow us to resolve a
> relative difference between two elements of a disk backing chain. This
> fucntion will be used to allow relative block commit and block pull
s/fucntion/function/
> where we need to spe
On 06/23/2014 09:42 PM, coperd wrote:
[can you convince your mailer to wrap long lines?]
> Hi, all. I’m current doing monitoring jobs(CPU, memory and disk
utilisation) on some KVM-based VMs using libvirt, but i came to some
confusions.
> For the disk data(blkstat), I find that the results acqui
On 06/24/2014 03:20 AM, Ian Campbell wrote:
>
> I see VIR_STRDUP actually does handle NULL input correctly, which was
> what I was worried about.
Yes, it is by design that VIR_STRDUP(NULL) works, and gives a different
return value (0) than when dup'ing a string (positive) or on failure
(negative)
On 06/24/2014 03:39 AM, Daniel P. Berrange wrote:
>>
>> We also have a subelement of in the
>> XML which is used for a similar thing although it doesn't support a
>> per-machine-type output, only per-binary capabilities. Should we add
>> this more granular approach and abandon the old one?
>
>
On 06/24/2014 08:07 AM, Jason J. Herne wrote:
> During a recent managed save operation I received the following error
> message:
>
> error: operation failed: domain save job: unexpectedly failed.
>
> It turns out that I had run out of disk space. After a brief
> investigation I
> discovered
On 06/24/2014 01:02 PM, Eric Blake wrote:
> On 06/15/2014 10:32 AM, Julio Faracco wrote:
>> As we are doing with the enum structures, a cleanup in
>> "src/nwfilter/" directory was done now. All the enums that were
>> defined in the header files were converted to typedefs in this
>> directory. This
On 06/15/2014 10:32 AM, Julio Faracco wrote:
> As we are doing with the enum structures, a cleanup in
> "src/nwfilter/" directory was done now. All the enums that were
> defined in the header files were converted to typedefs in this
> directory. This patch includes all the adjustments to remove
> c
On 06/15/2014 10:32 AM, Julio Faracco wrote:
> As we are doing with the enum structures, a cleanup in "src/qemu/"
> directory was done now. All the enums that were defined in the
> header files were converted to typedefs in this directory. This
> patch includes all the adjustments to remove conflic
Commit 9b8d6e1e missed updating the libvirt version which
introduced support for domainMigrate*3Params functions.
Signed-off-by: Jim Fehlig
---
This patch only changes the comments describing when these functions
were introduced. Pushing under the trivial rule.
src/libxl/libxl_driver.c | 10 +
ping...
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
ping...
--
libvir-list mailing list
libvir-list@redhat.com
https://www.redhat.com/mailman/listinfo/libvir-list
On 06/24/2014 07:10 AM, Puneet Bakshi wrote:
[please don't top-post on technical lists]
> 1. In that case, I hope it allow sending binary data (non-ASCII) also.
Yes, the reason guest-file-write takes base64 encoded data is PRECISELY
so that it can cope with binary data while still using only asc
On 06/24/2014 06:50 AM, Puneet Bakshi wrote:
> Hi,
>
>>From host, I wrote 26 alphabets in guest file (/tmp/testqga) using
> guest-file-write guest agent command (logs pasted below). I faced 2 issues
> when doing that.
>
> 1a. It could wrote only 18bytes! Why could it not write all 26 characters?
During a recent managed save operation I received the following error
message:
error: operation failed: domain save job: unexpectedly failed.
It turns out that I had run out of disk space. After a brief investigation I
discovered that libvirt_iohelper is exec'ed and is used to handle all
I
On 24.06.2014 11:45, Wangrui (K) wrote:
update comments of lxcDomainCreateXMLWithFiles() and lxcDomainCreateXML()
Signed-off-by: Yue wenyuan
Signed-off-by: Wang Rui
---
src/lxc/lxc_driver.c | 17 ++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/src/lxc/lxc_driver
Hi, all. I’m current doing monitoring jobs(CPU, memory and disk utilisation)
on some KVM-based VMs using libvirt, but i came to some confusions.
For the disk data(blkstat), I find that the results acquired by
virDomainBlockStats is not consistent with the value read from
/proc/[vm_pid]/io(i on
On 23.06.2014 06:25, Taowei wrote:
Version specified codes for vboxUniformedAPI is implemented here,
almost the same as PATCHv2.
Common codes are moved to vbox_common.h and vbox_common.c.
---
src/vbox/vbox_tmpl.c | 142 ++
1 file changed, 52 in
On 23.06.2014 06:25, Taowei wrote:
The vboxGlobalData and the vboxUniformedAPI is defined here. It defines
the middle layer interface. When using this file, it needs a context
to explain how *.c files treat vbox structs in vboxGlobalData(vboxObj,
vboxSession and so on). See annotate in this file.
On 23.06.2014 06:25, Taowei wrote:
Changes in makefile and make new files to be compiled.
---
po/POTFILES.in |1 +
src/Makefile.am |4 +++-
2 files changed, 4 insertions(+), 1 deletion(-)
diff --git a/po/POTFILES.in b/po/POTFILES.in
index 31a8381..d20107d 100644
--- a/po/POTFILES.i
On 23.06.2014 06:25, Taowei wrote:
This is a example of how common codes work.
The file vbox_commondef.h tells the common codes to treat vbox structs
defined in vbox_common.h as a void* type.
---
src/vbox/vbox_common.c| 89 +
src/vbox/vbox_com
On 23.06.2014 06:25, Taowei wrote:
Michal asked me to try working out a sample of vbox driver's common code. Here
it is. As the PATCHv2 is pending, I merged it.
Taowei (5):
define interface with vbox_common
a sample of vbox_common function : vboxInitialize
implement vboxUniformedAPI in
On 23.06.2014 06:25, Taowei wrote:
Register pVBoxAPI in vbox_common.c with the current vbox
version.
---
src/vbox/vbox_driver.c | 35 +++
1 file changed, 31 insertions(+), 4 deletions(-)
diff --git a/src/vbox/vbox_driver.c b/src/vbox/vbox_driver.c
index 7d00
On 06/24/14 14:23, Daniel P. Berrange wrote:
> On Tue, Jun 24, 2014 at 01:54:42PM +0200, Peter Krempa wrote:
>> Don't free individual JSON array members as the array will be freed at
>> the end. This may potentially lead to a crash although it didn't crash
>> on my setup.
>> ---
>> src/network/bri
1. In that case, I hope it allow sending binary data (non-ASCII) also.
2. If I encode my data in base-64 form, is there a limit on size of the
data I can pass (or I can pass arbitrarily large amount of data).
Regards,
~Puneet
On Tue, Jun 24, 2014 at 6:27 PM, Daniel P. Berrange
wrote:
> On Tu
On Tue, Jun 24, 2014 at 06:20:16PM +0530, Puneet Bakshi wrote:
> Hi,
>
> >From host, I wrote 26 alphabets in guest file (/tmp/testqga) using
> guest-file-write guest agent command (logs pasted below). I faced 2 issues
> when doing that.
>
> 1a. It could wrote only 18bytes! Why could it not write
On Tue, Jun 24, 2014 at 02:45:34PM +0200, Ján Tomko wrote:
> Introduced by commit ba51398
> ---
> src/network/bridge_driver.c | 1 +
> 1 file changed, 1 insertion(+)
ACK
Regards,
Daniel
--
|: http://berrange.com -o-http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org
On Tue, Jun 24, 2014 at 02:45:32PM +0200, Ján Tomko wrote:
> Introduced by commit 2bdb8b9
> ---
> src/storage/storage_driver.c | 1 +
> 1 file changed, 1 insertion(+)
ACK
Regards,
Daniel
--
|: http://berrange.com -o-http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org
Hi,
>From host, I wrote 26 alphabets in guest file (/tmp/testqga) using
guest-file-write guest agent command (logs pasted below). I faced 2 issues
when doing that.
1a. It could wrote only 18bytes! Why could it not write all 26 characters?
Are we supposed to track how much data is written and need
On Tue, Jun 24, 2014 at 02:45:33PM +0200, Ján Tomko wrote:
> Don't leak the temporary variables on success if NULL is returned
> for that field.
>
> Don't dereference NULL on failure to allocate some of the temporaries.
>
> Introduced by commit 990c3b6
> ---
> daemon/remote.c | 53 ++
On 24.06.2014 14:45, Ján Tomko wrote:
Ján Tomko (3):
Free file header in virStorageFileGetMetadataRecurse
Rework remoteSerializeDHCPLease
Free DHCP leases file in networkGetDHCPLeasesHelper
daemon/remote.c | 53
src/network/br
Introduced by commit ba51398
---
src/network/bridge_driver.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/network/bridge_driver.c b/src/network/bridge_driver.c
index a407c6e..69c66f3 100644
--- a/src/network/bridge_driver.c
+++ b/src/network/bridge_driver.c
@@ -3529,6 +3529,7 @@ network
Ján Tomko (3):
Free file header in virStorageFileGetMetadataRecurse
Rework remoteSerializeDHCPLease
Free DHCP leases file in networkGetDHCPLeasesHelper
daemon/remote.c | 53
src/network/bridge_driver.c | 1 +
src/storage/storage_dr
Don't leak the temporary variables on success if NULL is returned
for that field.
Don't dereference NULL on failure to allocate some of the temporaries.
Introduced by commit 990c3b6
---
daemon/remote.c | 53 ++---
1 file changed, 34 insertions(+),
Introduced by commit 2bdb8b9
---
src/storage/storage_driver.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/storage/storage_driver.c b/src/storage/storage_driver.c
index 7116185..a5adc63 100644
--- a/src/storage/storage_driver.c
+++ b/src/storage/storage_driver.c
@@ -2906,6 +2906,7 @@ vi
Variables/fields named 'interface' clash with system
header symbols on some platforms.
Signed-off-by: Daniel P. Berrange
---
daemon/remote.c | 4 ++--
include/libvirt/libvirt.h.in | 2 +-
src/libvirt.c| 2 +-
src/network/bridge_driver.c | 2 +-
src/remote/remote_dri
On Tue, Jun 24, 2014 at 02:18:32PM +0200, Peter Krempa wrote:
> On 06/23/14 23:01, Nehal J Wani wrote:
> > Introduce 3 new APIs, virNetworkGetDHCPLeases, virNetworkGetDHCPLeasesForMAC
> > and virNetworkDHCPLeaseFree.
> >
> > * virNetworkGetDHCPLeases: returns the dhcp leases information for a give
On Tue, Jun 24, 2014 at 01:54:42PM +0200, Peter Krempa wrote:
> Don't free individual JSON array members as the array will be freed at
> the end. This may potentially lead to a crash although it didn't crash
> on my setup.
> ---
> src/network/bridge_driver.c | 4 +---
> 1 file changed, 1 insertion
On 24.06.2014 14:04, Giuseppe Scrivano wrote:
Michal Privoznik writes:
On 24.06.2014 13:34, Giuseppe Scrivano wrote:
virPortAllocatorSetUsed permits to set a port as already used and
prevent the port allocator to use it without any attempt to bind it.
Signed-off-by: Giuseppe Scrivano
---
On 06/23/14 23:01, Nehal J Wani wrote:
> This API returns the leases information stored in the DHCP leases
> file of dnsmasq for a given virtual network. It contacts the
> bridge network driver, which parses a custom leases file created
> by libvirt.
>
> It supports two methods:
>
> 1. Return inf
On 06/23/14 23:01, Nehal J Wani wrote:
> Introduce 3 new APIs, virNetworkGetDHCPLeases, virNetworkGetDHCPLeasesForMAC
> and virNetworkDHCPLeaseFree.
>
> * virNetworkGetDHCPLeases: returns the dhcp leases information for a given
> virtual network.
>
> For DHCPv4, the information returned:
>
On 06/24/14 13:54, Peter Krempa wrote:
> Don't free individual JSON array members as the array will be freed at
> the end. This may potentially lead to a crash although it didn't crash
> on my setup.
> ---
> src/network/bridge_driver.c | 4 +---
> 1 file changed, 1 insertion(+), 3 deletions(-)
>
Michal Privoznik writes:
> On 24.06.2014 13:34, Giuseppe Scrivano wrote:
>> virPortAllocatorSetUsed permits to set a port as already used and
>> prevent the port allocator to use it without any attempt to bind it.
>>
>> Signed-off-by: Giuseppe Scrivano
>> ---
>> src/libvirt_private.syms|
Don't free individual JSON array members as the array will be freed at
the end. This may potentially lead to a crash although it didn't crash
on my setup.
---
src/network/bridge_driver.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/src/network/bridge_driver.c b/src/networ
On 24.06.2014 13:34, Giuseppe Scrivano wrote:
When looking for a port to allocate, the port allocator didn't take in
consideration ports that are statically set by the user. Defining
these two graphics elements in the XML would cause an error, as the
port allocator would try to use the same port
On 24.06.2014 13:34, Giuseppe Scrivano wrote:
virPortAllocatorSetUsed permits to set a port as already used and
prevent the port allocator to use it without any attempt to bind it.
Signed-off-by: Giuseppe Scrivano
---
src/libvirt_private.syms| 1 +
src/util/virportallocator.c | 44 +
When looking for a port to allocate, the port allocator didn't take in
consideration ports that are statically set by the user. Defining
these two graphics elements in the XML would cause an error, as the
port allocator would try to use the same port for the spice graphics
element:
The
virPortAllocatorSetUsed permits to set a port as already used and
prevent the port allocator to use it without any attempt to bind it.
Signed-off-by: Giuseppe Scrivano
---
src/libvirt_private.syms| 1 +
src/util/virportallocator.c | 44 +++-
src/util/
Changes in v2:
* virPortAllocatorSetUsed returns an error if the port is already used.
* Changed *[pP]ortAllocated to *[pP]ortReserved that keep track only of ports
that
were marked used but not bound.
* Handle separately port and tlsPort for SPICE.
Fix a conflict when both autoport graphics a
On Tue, Jun 24, 2014 at 02:31:49AM +0530, Nehal J Wani wrote:
> Introduce 3 new APIs, virNetworkGetDHCPLeases, virNetworkGetDHCPLeasesForMAC
> and virNetworkDHCPLeaseFree.
>
> * virNetworkGetDHCPLeases: returns the dhcp leases information for a given
> virtual network.
>
> For DHCPv4, the
On Tue, Jun 24, 2014 at 02:31:50AM +0530, Nehal J Wani wrote:
> diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c
> index bef9fd7..d6eb477 100644
> --- a/src/remote/remote_driver.c
> +++ b/src/remote/remote_driver.c
> @@ -7549,6 +7549,190 @@ remoteNodeGetFreePages(virConnectPtr c
Commit 2cff94c fixed the shadowed 'link' added by commit 975f0e2,
but forgot the 'link' added by commit 08aa22e.
---
Pushed as a build-breaker fix.
tests/virstoragetest.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/tests/virstoragetest.c b/tests/virstoragetest.c
inde
On Fri, 2014-06-20 at 15:07 -0600, Jim Fehlig wrote:
> The libxl driver currently sets the disk backend to
> LIBXL_DISK_BACKEND_TAP when is specified
> in the config. qdisk should be prefered with this
> configuration, otherwise existing configuration such as the
> following, which worked with t
On 06/24/2014 11:44 AM, Chen Hanxiao wrote:
> fix:
> virsh -c lxc:/// memtune DOMAIN
> error: Unable to get number of memory parameters
> error: unsupported flags (0x4) in function lxcDomainGetMemoryParameters
>
> Signed-off-by: Chen Hanxiao
> ---
> v2: also fix a similar issue in lxcDomainGetSch
On 06/24/2014 11:56 AM, Michal Privoznik wrote:
> On 24.06.2014 10:15, Ján Tomko wrote:
>> On 06/23/2014 03:59 PM, Michal Privoznik wrote:
>>> + * kernel limitation. The problem is, if there are some huge
>>> + * pages allocated, they are accounted under the 'MemUsed' field
>>> + * in the meminfo f
On 19.6.2014 19:59, Jincheng Miao wrote:
> hostdev has mode "capabilities" for LXC, from formatdomain.html:
> "
> Block / character devices from the host can be passed through to
> the guest using the hostdev element.
> This is only possible with container based virtualization.
> since after 1.0.1
On 24.06.2014 10:15, Ján Tomko wrote:
On 06/23/2014 03:59 PM, Michal Privoznik wrote:
On the Linux kernel, if huge pages are allocated the size they cut off
from memory is accounted under the 'MemUsed' in the meminfo file.
However, we want the sum to be subtracted from 'MemTotal'. This patch
imp
update comments of lxcDomainCreateXMLWithFiles() and lxcDomainCreateXML()
Signed-off-by: Yue wenyuan
Signed-off-by: Wang Rui
---
src/lxc/lxc_driver.c | 17 ++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c
index 06f3e18..1
fix:
virsh -c lxc:/// memtune DOMAIN
error: Unable to get number of memory parameters
error: unsupported flags (0x4) in function lxcDomainGetMemoryParameters
Signed-off-by: Chen Hanxiao
---
v2: also fix a similar issue in lxcDomainGetSchedulerParametersFlags
v3: fix improper comments
src/lxc/lx
On Tue, Jun 24, 2014 at 09:03:51AM +0200, Peter Krempa wrote:
> On 06/20/14 16:19, Michal Privoznik wrote:
> > The virEmulatorCapabilities is going to hold emulator capabilities,
> > surprisingly. It's intended to be able to cover qemuCaps, lxcCaps
> > (once we invent them, if ever) and so on. Amon
> -Original Message-
> From: Ján Tomko [mailto:jto...@redhat.com]
> Sent: Tuesday, June 24, 2014 4:57 PM
> To: Chen, Hanxiao/陈 晗霄; libvir-list@redhat.com
> Subject: Re: [libvirt] [PATCH v2] LXC: trivially support flag
> VIR_DRV_FEATURE_TYPED_PARAM_STRING
>
> On 06/24/2014 09:24 AM, Chen
s/link/linkpath/g in virStorageFileBackendGlusterReadlinkCallback as
older gcc complains.
---
Notes:
Pushed as trivial.
src/storage/storage_backend_gluster.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/src/storage/storage_backend_gluster.c
b/src/storage/storage
On Thu, 2014-06-19 at 00:15 -0600, Jim Fehlig wrote:
> Ian Campbell wrote:
> > On Tue, 2014-06-17 at 16:24 +0100, Ian Campbell wrote:
> >
> >> +if (xenXMConfigGetString(conf, "extra", &extra, NULL) < 0)
> >>
> >
> > This was subtly broken. The default needs to be "".
> >
>
> Tur
ping any ideas?
- Original Message -
> hostdev has mode "capabilities" for LXC, from formatdomain.html:
> "
> Block / character devices from the host can be passed through to
> the guest using the hostdev element.
> This is only possible with container based virtualization.
> since after 1
On 06/24/2014 09:24 AM, Chen Hanxiao wrote:
> fix:
> virsh -c lxc:/// memtune DOMAIN
> error: Unable to get number of memory parameters
> error: unsupported flags (0x4) in function lxcDomainGetMemoryParameters
>
> Signed-off-by: Chen Hanxiao
> ---
> v2: also fix a similar issue in lxcDomainGetSch
Ján Tomko writes:
> On 06/23/2014 08:15 PM, Giuseppe Scrivano wrote:
>> When looking for a port to allocate, the port allocator didn't take in
>> consideration ports that are statically set by the user. Defining
>> these two graphics elements in the XML would cause an error, as the
>> port alloc
On 06/24/14 05:35, Eric Blake wrote:
> On 06/19/2014 07:59 AM, Peter Krempa wrote:
>> Use virStorageFileSimplifyPathInternal to canonicalize gluster paths
>> via a callback and use it for the unique volume path retrieval API.
>> ---
>> src/storage/storage_backend_gluster.c | 80
>> +++
On 06/24/14 01:30, Eric Blake wrote:
> With this in place, I can (finally!) now do:
>
> virsh blockcommit $dom vda --shallow --verbose --pivot
>
> and watch qemu shorten the backing chain by one, followed by
> libvirt automatically updating the dumpxml output, effectively
> undoing the work of vi
On 06/24/14 01:30, Eric Blake wrote:
> A future patch is going to wire up qemu active block commit jobs;
> but as they have similar events and are canceled/pivoted in the
> same way as block copy jobs, it is easiest to track all bookkeeping
> for the commit job by reusing the element. This patch
On 06/23/2014 03:59 PM, Michal Privoznik wrote:
> On the Linux kernel, if huge pages are allocated the size they cut off
> from memory is accounted under the 'MemUsed' in the meminfo file.
> However, we want the sum to be subtracted from 'MemTotal'. This patch
> implements this feature. After this
On 06/23/2014 03:59 PM, Michal Privoznik wrote:
> If we are running on a system that is not capable of huge pages (e.g.
> because the kernel is not configured that way) we still try to open
> "/sys/kernel/mm/hugepages/" which however does not exist. We should
> be tolerant to this specific use case
On 06/24/14 01:30, Eric Blake wrote:
> Add an element to QEMU's capability XML, to show if the underlying
> qemu binary supports active commit. This allows the client to know
> ahead of time if they can rely on this method, or must fall back
> to older techniques such as blockpull. Without this i
On 06/23/2014 08:15 PM, Giuseppe Scrivano wrote:
> When looking for a port to allocate, the port allocator didn't take in
> consideration ports that are statically set by the user. Defining
> these two graphics elements in the XML would cause an error, as the
> port allocator would try to use the
fix:
virsh -c lxc:/// memtune DOMAIN
error: Unable to get number of memory parameters
error: unsupported flags (0x4) in function lxcDomainGetMemoryParameters
Signed-off-by: Chen Hanxiao
---
v2: also fix a similar issue in lxcDomainGetSchedulerParametersFlags
src/lxc/lxc_driver.c | 16 ++
On 06/24/14 01:30, Eric Blake wrote:
> Use the probing functionality added in the last patch to turn on
> a capability bit when active commit is present, and gate active
> commit on that capability.
>
> For my own reference: the difference between BLOCKJOB_SYNC and
> BLOCKJOB_ASYNC is whether qemu
On 06/24/14 01:30, Eric Blake wrote:
> We are about to turn on support for active block commit. Although
> qemu 2.0 was the first version to mostly support it, that version
> mis-handles 0-length files, and doesn't have anything available for
> easy probing. But qemu 2.1 fixed bugs, and made life
On 06/20/14 16:19, Michal Privoznik wrote:
> The virEmulatorCapabilities is going to hold emulator capabilities,
> surprisingly. It's intended to be able to cover qemuCaps, lxcCaps
> (once we invent them, if ever) and so on. Among with adding the code
> itself, both some documentation and basic tes
97 matches
Mail list logo
