Re: [libvirt] [PATCH v2 04/15] qemu: Handle postcopy-active migration state

On Thu, Jan 21, 2016 at 11:20:49AM +0100, Jiri Denemark wrote: > Migration enters "postcopy-active" state after QEMU switches to > post-copy and pauses guest CPUs. From libvirt's point of view this state > is similar to "completed" because we need to transfer guest execution to > the destination

Re: [libvirt] [PATCH v2 05/15] qemu: Add QMP functions for post-copy migration

On Thu, Jan 21, 2016 at 11:20:50AM +0100, Jiri Denemark wrote: > From: Cristian Klein > > Signed-off-by: Cristian Klein > Signed-off-by: Jiri Denemark > --- > > Notes: > The migration capability is called x-postcopy-ram in

Re: [libvirt] [PATCH v2 02/15] Add public APIs for post-copy migration

On Fri, Jan 22, 2016 at 15:10:16 +, Daniel P. Berrange wrote: > On Thu, Jan 21, 2016 at 11:20:47AM +0100, Jiri Denemark wrote: > > From: Cristian Klein > > > > To use post-copy one has to start the migration with > > VIR_MIGRATE_POSTCOPY flag and, while migration is in

Re: [libvirt] [PATCH v2 01/15] Add event and state details for post-copy

On Fri, Jan 22, 2016 at 15:07:04 +, Daniel P. Berrange wrote: > On Thu, Jan 21, 2016 at 11:20:46AM +0100, Jiri Denemark wrote: > > VIR_DOMAIN_EVENT_SUSPENDED_POSTCOPY and VIR_DOMAIN_PAUSED_POSTCOPY are > > used on the source host once migration enters post-copy mode (which > > means the domain

Re: [libvirt] [PATCH v2 04/15] qemu: Handle postcopy-active migration state

On Fri, Jan 22, 2016 at 15:15:03 +, Daniel P. Berrange wrote: > On Thu, Jan 21, 2016 at 11:20:49AM +0100, Jiri Denemark wrote: > > Migration enters "postcopy-active" state after QEMU switches to > > post-copy and pauses guest CPUs. From libvirt's point of view this state > > is similar to

Re: [libvirt] [PATCH v2 01/15] Add event and state details for post-copy

On Fri, Jan 22, 2016 at 15:23:43 +, Daniel P. Berrange wrote: > On Fri, Jan 22, 2016 at 04:17:42PM +0100, Jiri Denemark wrote: > > On Fri, Jan 22, 2016 at 15:07:04 +, Daniel P. Berrange wrote: > > > On Thu, Jan 21, 2016 at 11:20:46AM +0100, Jiri Denemark wrote: > > > >

Re: [libvirt] [PATCH v2 05/15] qemu: Add QMP functions for post-copy migration

On Fri, Jan 22, 2016 at 15:16:20 +, Daniel P. Berrange wrote: > On Thu, Jan 21, 2016 at 11:20:50AM +0100, Jiri Denemark wrote: > > From: Cristian Klein > > > > Signed-off-by: Cristian Klein > > Signed-off-by: Jiri Denemark

Re: [libvirt] [PATCH] qemu: turn on virtlockd by default

On 01/22/2016 02:35 PM, Kashyap Chamarthy wrote: > On Fri, Jan 22, 2016 at 03:56:08PM +, Daniel P. Berrange wrote: >> We have had virtlockd available for a long time now but >> have always defaulted to the 'nop' lock driver which does >> no locking. This gives users an unsafe deployment by >>

Re: [libvirt] [PATCH 1/2] logical: Use correct syntax for thin/sparse pool creation

On 01/12/2016 11:03 AM, Joe Harvell wrote: > Le 11/01/2016 17:32, John Ferlan a écrit : >> >> On 01/11/2016 12:50 PM, Joe Harvell wrote: John/Jan: I actually made a patch similar to this and was about to submit it for review. Now I see there is a more thorough version here

[libvirt] [PATCH 07/17] qemu: command: drop QEMU_CAPS_DEVICE for PCI address assign

All the logic was in a QEMU_CAPS_DEVICE conditional. Unindent it --- src/qemu/qemu_command.c | 238 1 file changed, 117 insertions(+), 121 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 3ab53be..f2e9f6a 100644

[libvirt] [PATCH 03/17] qemu: command: Drop xen disk and vga handling

Yet more leftover bits from xenner days --- src/qemu/qemu_command.c| 180 + src/qemu/qemu_domain.c | 9 ++ .../qemuargv2xmldata/qemuargv2xml-disk-xenvbd.args | 25 --- .../qemuargv2xmldata/qemuargv2xml-disk-xenvbd.xml

[libvirt] [PATCH 05/17] qemu: Drop some simple uses of QEMU_CAPS_DEVICE

--- src/qemu/qemu_command.c | 82 + 1 file changed, 15 insertions(+), 67 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index dade9f5..594cee6 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@

[libvirt] [PATCH 09/17] qemu: command: drop QEMU_CAPS_DEVICE for net cli

-net none is never required anymore, even platforms that can't use -device abide -nodefaults. Additionally we conditionalized the cli building on nnets != 0, but that was redundant. --- src/qemu/qemu_command.c | 72 +++-- 1 file changed, 33

Re: [libvirt] [PATCH 03/17] qemu: command: Drop xen disk and vga handling

On 01/22/2016 02:30 PM, Cole Robinson wrote: > Yet more leftover bits from xenner days > --- > src/qemu/qemu_command.c| 180 > + > src/qemu/qemu_domain.c | 9 ++ > .../qemuargv2xmldata/qemuargv2xml-disk-xenvbd.args |

[libvirt] [PATCH v2 4/8] tests: qemuxml2argv: remove some QEMU_CAPS_DEVICE problem cases

When we unconditionally enable QEMU_CAPS_DEVICE, these tests need some massaging, so do it ahead of time to not mix it in with the big test refresh. - minimal-s390 is not a real world working config, so drop it - disk-usb was testing for an old code path that will be removed. instead use it to

[libvirt] [PATCH v2 7/8] qemu: hotplug: Drop !QEMU_CAPS_DEVICE code

Nowadays we only support qemu 0.12.0+ which provides QEMU_CAPS_DEVICE, so this is all dead code. --- src/qemu/qemu_hotplug.c | 480 +++- 1 file changed, 144 insertions(+), 336 deletions(-) diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c

[libvirt] [PATCH v2 6/8] qemu: process: Drop !QEMU_CAPS_DEVICE code

Nowadays we only support qemu 0.12.0+ which provides QEMU_CAPS_DEVICE, so this is all dead code. --- src/qemu/qemu_process.c | 471 +++- 1 file changed, 27 insertions(+), 444 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c

[libvirt] [PATCH v2 1/8] tests: Run test-wrap-argv with REGENERATE_OUTPUT

To get properly wrapped output --- tests/testutils.c | 38 +++--- 1 file changed, 31 insertions(+), 7 deletions(-) diff --git a/tests/testutils.c b/tests/testutils.c index b587f83..10c26648 100644 --- a/tests/testutils.c +++ b/tests/testutils.c @@ -433,6 +433,32

[libvirt] [PATCH v2 8/8] qemu: monitor: drop now unused hmp monitor commands

Basically all wrappers for these hmp commands: - usb_add - pci_add - pci_del - drive_add - info pci --- src/qemu/qemu_monitor.c | 161 src/qemu/qemu_monitor.h | 49 src/qemu/qemu_monitor_json.c | 88 --- src/qemu/qemu_monitor_json.h | 33 ---

Re: [libvirt] [PATCH v2 7/8] qemu: hotplug: Drop !QEMU_CAPS_DEVICE code

On 01/22/2016 02:09 PM, Cole Robinson wrote: > Nowadays we only support qemu 0.12.0+ which provides QEMU_CAPS_DEVICE, > so this is all dead code. > --- > src/qemu/qemu_hotplug.c | 480 > +++- > 1 file changed, 144 insertions(+), 336 deletions(-) >

Re: [libvirt] [PATCH] qemu: turn on virtlockd by default

On Fri, Jan 22, 2016 at 03:56:08PM +, Daniel P. Berrange wrote: > We have had virtlockd available for a long time now but > have always defaulted to the 'nop' lock driver which does > no locking. This gives users an unsafe deployment by > default unless they know to turn on lockd. virtlockd

[libvirt] [PATCH v2 0/8] qemu: Drop QEMU_CAPS_DEVICE part 1

libvirt only supports qemu 0.12.0+ nowadays, which means that the qemu binary always provides the -device/QEMU_CAPS_DEVICE option. This patch series is a step towards dropping support for that flag. Patch #1 is a test suite improvement, not strictly related Patch #2-5 adjust the test suite for

[libvirt] [PATCH 13/17] qemu: command: Drop QEMU_CAPS_DEVICE for USB

Even platforms that don't handle -device well can still use -device for USB, they will just have a hardcoded USB bus. May still need extra libvirt changes to make that work, but that's applies to the current code too. --- src/qemu/qemu_command.c | 60

[libvirt] [PATCH 06/17] qemu: command: Drop QEMU_CAPS_DEVICE for device aliases

--- src/qemu/qemu_command.c | 86 - 1 file changed, 7 insertions(+), 79 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 594cee6..3ab53be 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -772,74

[libvirt] [PATCH 17/17] qemu: zap QEMU_CAPS_DEVICE

Remove the last remnants and blacklist it --- src/qemu/qemu_capabilities.c | 11 +++-- src/qemu/qemu_capabilities.h | 2 +- tests/qemucapabilitiesdata/caps_1.2.2-1.caps | 1 - tests/qemucapabilitiesdata/caps_1.3.1-1.caps | 1 -

[libvirt] [PATCH 01/17] qemu: domain: Consistently use goto in PostParse

Isn't strictly required, but I have upcoming patches in here and it's just easier to be consistent. --- src/qemu/qemu_domain.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 0fa2dbe..9b456e8 100644 ---

[libvirt] [PATCH 02/17] qemu: Validate type != xen at parse time

There's a validation check at VM startup time that rejects type=xen or os type=xen|linux. The latter bit is already validated in generic domain_conf.c code, so it can be dropped. The former bit is can be done at XML parse time, so move it there. --- src/qemu/qemu_command.c | 8

[libvirt] [PATCH 11/17] qemu: command: drop QEMU_CAPS_DEVICE for watchdog cli

The -watchdog path will never be triggered. --- src/qemu/qemu_command.c | 22 -- 1 file changed, 4 insertions(+), 18 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index e3c6a51..516f351 100644 --- a/src/qemu/qemu_command.c +++

[libvirt] [PATCH 04/17] qemu: command: Don't unset QEMU_CAPS_DEVICE for -drive sd

Since -device can't be used with bus=sd disks, we would unset QEMU_CAPS_DEVICE temporarily for these disks. However since QEMU_CAPS_DEVICE is going away, we don't want code to key off that anymore. Push the bus=sd check down into BuildDriveStr, and use that to hit the different code paths. ---

[libvirt] [PATCH 10/17] qemu: command: drop QEMU_CAPS_DEVICE for serial/parallel cli

-serial none and -parallel none are never required anymore, even platforms that can't use -device abide -nodefaults. Additionally we conditionalized the cli building on nparallels != 0, but that was pointless. --- src/qemu/qemu_command.c | 50 - 1

[libvirt] [PATCH 00/17] qemu: Drop QEMU_CAPS_DEVICE part 2

This is the second series dropping QEMU_CAPS_DEVICE. It handles qemu_command.c and all remaining uses of QEMU_CAPS_DEVICE. Every qemu binary after v0.12.1 has the -device option, but many platforms cannot actually use it, because they are based around hardcoded machine models that cannot be

[libvirt] [PATCH 08/17] qemu: command: drop QEMU_CAPS_DEVICE for controller cli

All the logic was in a QEMU_CAPS_DEVICE conditional. Unindent it --- src/qemu/qemu_command.c | 112 1 file changed, 55 insertions(+), 57 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index f2e9f6a..a943d69 100644 ---

[libvirt] [PATCH 15/17] qemu: zap QEMU_CAPS_BALLOON

It's only usage is in a !QEMU_CAPS_DEVICE code path. Remove that code, and blacklist it --- src/qemu/qemu_capabilities.c | 3 --- src/qemu/qemu_capabilities.h | 2 +- src/qemu/qemu_command.c | 20 +---

[libvirt] [PATCH 12/17] qemu: command: drop QEMU_CAPS_DEVICE for video cli

The whole block was also conditionalized on nvideos >= 1, which the loop will handle fine after the QEMU_CAPS_DEVICE check is dropped. --- src/qemu/qemu_command.c | 35 +-- 1 file changed, 13 insertions(+), 22 deletions(-) diff --git a/src/qemu/qemu_command.c

[libvirt] [PATCH 14/17] qemu: command: Drop old style -soundhw usage

Even platforms that don't handle -device well cannot use this... they either have sound hardware baked in, or none at all. So this can be safely dropped. I changed the code a bit short circuit the loop for the pcspk case, which allows us to unindent the main logic some more ---

[libvirt] [PATCH 16/17] qemu: zap QEMU_CAPS_PCIDEVICE

All code paths that use it are !QEMU_CAPS_DEVICE code paths, so they will never be hit. This means we can drop it entirely. --- src/qemu/qemu_capabilities.c | 6 +--- src/qemu/qemu_capabilities.h | 2 +- src/qemu/qemu_command.c | 70 ++--

Re: [libvirt] [PATCH v2 6/8] qemu: process: Drop !QEMU_CAPS_DEVICE code

On 01/22/2016 02:09 PM, Cole Robinson wrote: > Nowadays we only support qemu 0.12.0+ which provides QEMU_CAPS_DEVICE, > so this is all dead code. > --- > src/qemu/qemu_process.c | 471 > +++- > 1 file changed, 27 insertions(+), 444 deletions(-) >

Re: [libvirt] [PATCH 08/17] qemu: command: drop QEMU_CAPS_DEVICE for controller cli

On 01/22/2016 02:30 PM, Cole Robinson wrote: > All the logic was in a QEMU_CAPS_DEVICE conditional. Unindent it > --- > src/qemu/qemu_command.c | 112 > > 1 file changed, 55 insertions(+), 57 deletions(-) > Whitespace-less diff attached - Cole

Re: [libvirt] [PATCH 07/17] qemu: command: drop QEMU_CAPS_DEVICE for PCI address assign

On 01/22/2016 02:30 PM, Cole Robinson wrote: > All the logic was in a QEMU_CAPS_DEVICE conditional. Unindent it > --- > src/qemu/qemu_command.c | 238 > > 1 file changed, 117 insertions(+), 121 deletions(-) > Whitespace-less diff attached -

Re: [libvirt] [PATCH 10/17] qemu: command: drop QEMU_CAPS_DEVICE for serial/parallel cli

On 01/22/2016 02:30 PM, Cole Robinson wrote: > -serial none and -parallel none are never required anymore, even > platforms that can't use -device abide -nodefaults. > > Additionally we conditionalized the cli building on nparallels != 0, > but that was pointless. Whitespace-less diff attached

Re: [libvirt] [PATCH 09/17] qemu: command: drop QEMU_CAPS_DEVICE for net cli

On 01/22/2016 02:30 PM, Cole Robinson wrote: > -net none is never required anymore, even platforms that can't use > -device abide -nodefaults. > > Additionally we conditionalized the cli building on nnets != 0, but that > was redundant. Whitespace-less diff attached - Cole

Re: [libvirt] [PATCH 14/17] qemu: command: Drop old style -soundhw usage

On 01/22/2016 02:30 PM, Cole Robinson wrote: > Even platforms that don't handle -device well cannot use this... they > either have sound hardware baked in, or none at all. So this can be > safely dropped. > > I changed the code a bit short circuit the loop for the pcspk case, > which allows us to

Re: [libvirt] parsing url with percent sign when user want to use iso disk via http

On Fri, Jan 22, 2016 at 09:59:53AM +0100, Michal Privoznik wrote: > On 19.01.2016 13:15, Vasiliy Tolstov wrote: > > I found a strange thing, when i specify url for iso in libvirt xml like: > > What does the XML look like? AFAIK libvirt does not support specifying the whole disk URL, just

Re: [libvirt] [PATCH 1/2] rbd: Add wiping RBD volumes by using rbd_discard() or rbd_write()

On Thu, Jan 21, 2016 at 02:10:02PM -0500, John Ferlan wrote: > > > On 12/23/2015 10:06 AM, Wido den Hollander wrote: > > This allows user to use the volume wiping functionality of the libvirt > > storage driver. > > > > This patch also adds a new wiping algorithm

Re: [libvirt] [PATCH 0/4] lxc: fix 'free' usage on fedora 23

On Thu, Jan 21, 2016 at 01:42:56PM -0500, Cole Robinson wrote: > We need to add a few bits to our /proc/meminfo virtualization to > make 'free' work correctly on Fedora 23. > > https://bugzilla.redhat.com/show_bug.cgi?id=1300781 ACK to all 4 Regards, Daniel -- |: http://berrange.com -o-

Re: [libvirt] [PATCH] qemu: turn on virtlockd by default

On Fri, Jan 22, 2016 at 03:35:33PM -0500, Cole Robinson wrote: > On 01/22/2016 02:35 PM, Kashyap Chamarthy wrote: > > On Fri, Jan 22, 2016 at 03:56:08PM +, Daniel P. Berrange wrote: > >> We have had virtlockd available for a long time now but > >> have always defaulted to the 'nop' lock driver

[libvirt] [PATCH 1/4] logical: Fix comment examples for virStorageBackendLogicalFindLVs

When commit id '82c1740a' made changes to the output format (changing from using a ',' separator to '#'), the examples in the lvs output from the comments weren't changed. Additionally, the two new fields added ('segtype' and 'stripes') were not included in the output, leaving it well confusing.

[libvirt] [PATCH 4/4] logical: Adjust regex for devices

From: Joe Harvell Since our 'devices' parsing logic now will use the 'nextents' (or lvs 'stripes' output) to decide whether or not to parse the field, use the regex of "(\\S*)" (e.g. zero or more) instead of "(\\S+)" (1 or more) when grabbing the 'groups[3]' or

[libvirt] [PATCH 3/4] logical: Remove need for 'segtype' field

From: Joe Harvell Modify the virStorageBackendLogicalMakeVol parsing to use only the 'stripes' output as the value for 'nextents' rather than assuming there is at least 1 extent and then adjusting "only if" the 'segtype' was 'striped'. This avoids the chance that

[libvirt] [PATCH 0/4] Some logical pool/volume changes

This series is a result of looking at Joe's comments : http://www.redhat.com/archives/libvir-list/2016-January/msg00403.html To an old series I posted and has been stuck on the bottom of a todo list regarding adding/using "thin" logical volumes:

[libvirt] [PATCH 2/4] logical: Create helper virStorageBackendLogicalParseVolDevice

Create a helper routine in order to parse the 'device' string contained within the generated 'lvs' output string. A future patch would then be able to avoid the code more cleanly Signed-off-by: John Ferlan --- src/storage/storage_backend_logical.c | 186

[libvirt] [PATCH v3 2/8] virConnectCloseCallbackData: fix connection object refcount

We have reference to connection object in virConnectCloseCallbackData object thus we have to refcount it. Obviously we have problems in dispose and call functions. Let's fix it. Signed-off-by: Nikolay Shirokovskiy --- src/datatypes.c | 4 1 file changed, 4

[libvirt] [PATCH v3 6/8] close callback: move it to driver

Signed-off-by: Nikolay Shirokovskiy --- src/datatypes.c| 31 ++ src/datatypes.h| 1 + src/driver-hypervisor.h| 12 src/libvirt-host.c | 7 --- src/remote/remote_driver.c | 47

[libvirt] [PATCH v3 7/8] daemon: add connection close rpc

remoteConnectUnregisterCloseCallback is not quite good. if it is given a callback function different from that was registered before then local part will fail silently. On the other hand we can not gracefully handle this fail as the remote part is already unregistered. There are a lot of options

[libvirt] [PATCH v3 1/8] factor out virConnectCloseCallbackDataPtr methods

Signed-off-by: Nikolay Shirokovskiy --- po/POTFILES.in | 1 + src/datatypes.c| 75 ++ src/datatypes.h| 10 +++ src/libvirt-host.c | 35 ++

[libvirt] [PATCH v3 4/8] close callback API: remove unnecessary locks

closeCallback pointer is immutable (set on connection object creation) and self-locking. --- src/libvirt-host.c | 10 -- 1 file changed, 10 deletions(-) diff --git a/src/libvirt-host.c b/src/libvirt-host.c index c492bd8..0bab5f2 100644 --- a/src/libvirt-host.c +++ b/src/libvirt-host.c @@

[libvirt] [PATCH v3 3/8] virConnectCloseCallbackData: factor out callback disarming

--- src/datatypes.c | 31 --- 1 file changed, 16 insertions(+), 15 deletions(-) diff --git a/src/datatypes.c b/src/datatypes.c index 73c7e7e..57162ec 100644 --- a/src/datatypes.c +++ b/src/datatypes.c @@ -161,6 +161,19 @@ virConnectDispose(void *obj) } +static

[libvirt] [PATCH v3 8/8] vz: implement connection close notification

Signed-off-by: Nikolay Shirokovskiy --- src/vz/vz_driver.c | 39 +++ src/vz/vz_sdk.c| 4 src/vz/vz_utils.h | 3 +++ 3 files changed, 46 insertions(+) diff --git a/src/vz/vz_driver.c b/src/vz/vz_driver.c index

[libvirt] [PATCH v3 0/8] add close callback for drivers with persistent connection

Currently close callback API can only inform us of closing the connection between remote driver and daemon. But what if a driver running in the daemon itself can have another persistent connection? In this case we want to be informed of that connection changes state too. This patch series extends

[libvirt] [PATCH v3 5/8] virConnectCloseCallbackDataDispose: remove unnecessary locks

We don't need locks in dispose functions as they can only be run in one thread for given object. --- src/datatypes.c | 8 +--- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/src/datatypes.c b/src/datatypes.c index 57162ec..29f94e8 100644 --- a/src/datatypes.c +++

Re: [libvirt] parsing url with percent sign when user want to use iso disk via http

On 19.01.2016 13:15, Vasiliy Tolstov wrote: > I found a strange thing, when i specify url for iso in libvirt xml like: > > qemu gets invalid url in cmdline > i think that problem is > qemu/qemu_command.c:3430:ret = virURIFormat(uri); > > test string looks like

Re: [libvirt] [PATCH 1/2] util: keep/use a bitmap of in-use macvtap devices

On Thu, Jan 21, 2016 at 02:54:10PM -0500, Laine Stump wrote: [...] > src/libvirt_private.syms| 2 + > src/qemu/qemu_process.c | 10 +- > src/util/virnetdevmacvlan.c | 438 > +--- > src/util/virnetdevmacvlan.h | 5 +- > 4 files changed, 390

Re: [libvirt] [PATCH v4] libvirtd: Increase NL buffer size for lots of interface

On 2016年01月22日 04:08, Laine Stump wrote: On 01/20/2016 08:51 AM, Leno Hou wrote: 1. When switching CPUs to offline/online in a system more than 128 cpus 2. When using virsh to destroy domain in a system with more interface All of above happens nl_recv returned with error: No buffer space

Re: [libvirt] Accessing libvirtd remotely as non-root user

I joined this mailing list after I googled and found this thread. https://www.redhat.com/archives/libvir-list/2015-June/msg00583.html Part of this thread read: >So, further digging led me to [2], which in essence was the section of the libvirtd wiki page that the page claimed >was obsolete.

Re: [libvirt] [PATCH 2/2] util: reset MAC address of macvtap passthrough physdev after disassociate

On Thu, Jan 21, 2016 at 02:54:11PM -0500, Laine Stump wrote: > libvirt always resets the MAC address of the physdev used for macvtap > passthrough when the guest is finished with it. This was happening > prior to the 802.1Qb[gh] DISASSOCIATE command, and was quite often > failing, presumably

Re: [libvirt] [PATCH 2/2] rbd: Implement buildVolFrom using RBD cloning

On 01/22/2016 08:14 AM, Wido den Hollander wrote: > > > On 21-01-16 21:55, John Ferlan wrote: >> >> >> On 12/23/2015 04:29 AM, Wido den Hollander wrote: >>> RBD supports cloning by creating a snapshot, protecting it and create >>> a child image based on that snapshot afterwards. >>> >>> The

Re: [libvirt] [PATCH 2/2] rbd: Implement buildVolFrom using RBD cloning

On 21-01-16 21:55, John Ferlan wrote: > > > On 12/23/2015 04:29 AM, Wido den Hollander wrote: >> RBD supports cloning by creating a snapshot, protecting it and create >> a child image based on that snapshot afterwards. >> >> The RBD storage driver will try to find a snapshot with zero deltas

Re: [libvirt] [PATCH 0/4] lxc: fix 'free' usage on fedora 23

On 01/22/2016 05:02 AM, Daniel P. Berrange wrote: > On Thu, Jan 21, 2016 at 01:42:56PM -0500, Cole Robinson wrote: >> We need to add a few bits to our /proc/meminfo virtualization to >> make 'free' work correctly on Fedora 23. >> >> https://bugzilla.redhat.com/show_bug.cgi?id=1300781 > > ACK to

Re: [libvirt] [PATCH 1/7] hostdev: Add reattachPCIDevices()

On Fri, 2016-01-22 at 15:00 +, Daniel P. Berrange wrote: >  > IMHO you should leave virHostdevReattachPCIDevice alone, and just make > this new method call that one. In later patches you are calling this > reattachPCIDevices() method with a single device, forcing you to put > it into a

Re: [libvirt] [PATCH 1/7] hostdev: Add reattachPCIDevices()

On Fri, Jan 22, 2016 at 05:39:33PM +0100, Andrea Bolognani wrote: > On Fri, 2016-01-22 at 15:00 +, Daniel P. Berrange wrote: > >  > > IMHO you should leave virHostdevReattachPCIDevice alone, and just make > > this new method call that one. In later patches you are calling this > >

Re: [libvirt] [PATCH 1/2] util: keep/use a bitmap of in-use macvtap devices

On 01/22/2016 07:47 AM, Pavel Hrdina wrote: On Thu, Jan 21, 2016 at 02:54:10PM -0500, Laine Stump wrote: +/** + * virNetDevMacVLanReserveNextFreeID: + * + * @id: id to start scanning at - return first free ID *after* this + * id (use -1 to start looking at the beginning) + *

[libvirt] [PATCH] qemu: turn on virtlockd by default

We have had virtlockd available for a long time now but have always defaulted to the 'nop' lock driver which does no locking. This gives users an unsafe deployment by default unless they know to turn on lockd. virtlockd will auto-activate via systemd when guests launch, so setting it on by default

Re: [libvirt] ARM KVM GICv3 Support

On Wed, Jan 06, 2016 at 01:30:16PM +, Peter Maydell wrote: > On 6 January 2016 at 12:49, Andrea Bolognani wrote: > > That's correct, having a QMP command that lists the values gic-version > > can have on the current host would be just great. > > > > If we had that, we

Re: [libvirt] [PATCH RFC 2/2] libxl: set net device prefix

On Wed, Jan 20, 2016 at 11:41:27PM +, Joao Martins wrote: > Use the newly added virCapabilitiesSetNetPrefix to set > the network prefix for the driver. This in return will > be use by NetDefFormat() and NetDefParseXML() routines > to free any interface name that start with the registered >

Re: [libvirt] [PATCH RFC 1/2] conf: add net device prefix to capabilities

On Wed, Jan 20, 2016 at 11:41:26PM +, Joao Martins wrote: > In the reverted commit d2e5538b1, the libxl driver was changed to copy > interface names autogenerated by libxl to the corresponding network def > in the domain's virDomainDef object. The copied name is freed when the > domain

Re: [libvirt] [PATCH 1/7] hostdev: Add reattachPCIDevices()

On Tue, Jan 19, 2016 at 04:36:03PM +0100, Andrea Bolognani wrote: > This function replaces virHostdevReattachPCIDevice() and handles several > PCI devices instead of requiring to be called once for every device. > > The handling of active and inactive devices is updated and made more > explicit,

Re: [libvirt] [PATCH v2 01/15] Add event and state details for post-copy

On Thu, Jan 21, 2016 at 11:20:46AM +0100, Jiri Denemark wrote: > VIR_DOMAIN_EVENT_SUSPENDED_POSTCOPY and VIR_DOMAIN_PAUSED_POSTCOPY are > used on the source host once migration enters post-copy mode (which > means the domain gets paused on the source. After the destination host > takes over the

Re: [libvirt] [PATCH 3/7] hostdev: Use common reattach code in virHostdevPCINodeDeviceReAttach()

On Tue, Jan 19, 2016 at 04:36:05PM +0100, Andrea Bolognani wrote: > This ensures the behavior for reattach is consistent, no matter how it > was triggered (eg. 'virsh nodedev-reattach', 'virsh detach-device' or > shutdown of a domain that is configured to use hostdevs). > --- >

Re: [libvirt] [PATCH 4/7] hostdev: Add detachPCIDevices()

On Tue, Jan 19, 2016 at 04:36:06PM +0100, Andrea Bolognani wrote: > This function mirrors reattachPCIDevices(). > > The handling of active and inactive devices is updated and made more > explicit, which means virHostdevPreparePCIDevices() has to be > updated as well. > --- >

Re: [libvirt] [PATCH v2 02/15] Add public APIs for post-copy migration

On Thu, Jan 21, 2016 at 11:20:47AM +0100, Jiri Denemark wrote: > From: Cristian Klein > > To use post-copy one has to start the migration with > VIR_MIGRATE_POSTCOPY flag and, while migration is in progress, call > virDomainMigrateStartPostCopy() to switch from pre-copy to

Re: [libvirt] [PATCH v2 01/15] Add event and state details for post-copy

On Fri, Jan 22, 2016 at 04:17:42PM +0100, Jiri Denemark wrote: > On Fri, Jan 22, 2016 at 15:07:04 +, Daniel P. Berrange wrote: > > On Thu, Jan 21, 2016 at 11:20:46AM +0100, Jiri Denemark wrote: > > > VIR_DOMAIN_EVENT_SUSPENDED_POSTCOPY and VIR_DOMAIN_PAUSED_POSTCOPY are > > > used on the

Re: [libvirt] [PATCH v2 05/15] qemu: Add QMP functions for post-copy migration

On Fri, Jan 22, 2016 at 04:30:44PM +0100, Jiri Denemark wrote: > On Fri, Jan 22, 2016 at 15:16:20 +, Daniel P. Berrange wrote: > > On Thu, Jan 21, 2016 at 11:20:50AM +0100, Jiri Denemark wrote: > > > From: Cristian Klein > > > > > > Signed-off-by: Cristian Klein

Re: [libvirt] [PATCH RFC 1/2] conf: add net device prefix to capabilities

On 01/22/2016 02:50 PM, Daniel P. Berrange wrote: > On Wed, Jan 20, 2016 at 11:41:26PM +, Joao Martins wrote: >> In the reverted commit d2e5538b1, the libxl driver was changed to copy >> interface names autogenerated by libxl to the corresponding network def >> in the domain's virDomainDef

Re: [libvirt] [PATCH 4/7] hostdev: Add detachPCIDevices()

On Fri, 2016-01-22 at 15:02 +, Daniel P. Berrange wrote: > On Tue, Jan 19, 2016 at 04:36:06PM +0100, Andrea Bolognani wrote: > > This function mirrors reattachPCIDevices(). > >  > > The handling of active and inactive devices is updated and made more > > explicit, which means

Re: [libvirt] [PATCH 4/7] hostdev: Add detachPCIDevices()

On Fri, Jan 22, 2016 at 06:30:42PM +0100, Andrea Bolognani wrote: > On Fri, 2016-01-22 at 15:02 +, Daniel P. Berrange wrote: > > On Tue, Jan 19, 2016 at 04:36:06PM +0100, Andrea Bolognani wrote: > > > This function mirrors reattachPCIDevices(). > > >  > > > The handling of active and inactive

[libvirt] [PATCHv2] util: keep/use a bitmap of in-use macvtap devices

This patch creates two bitmaps, one for macvlan devicenames and one for macvtap. The bitmap position is used to indicate that libvirt is currently using a device with the name macvtap%d/macvlan%d, where %d is the position in the bitmap. When requested to create a new macvtap/macvlan device,