Re: [libvirt] [PATCH v5 1/3] nodedev: Fix usage of virNodeDeviceObjListGetParentHost

> >> > >> @@ -594,6 +600,11 @@ nodeDeviceDestroy(virNodeDevicePtr device) > >> int ret = -1; > >> virNodeDeviceObjPtr obj = NULL; > >> virNodeDeviceDefPtr def; > >> +char *name = NULL; > >> +char *parent = NULL; > >> +char *parent_wwnn = NULL; > >> +char *parent_wwpn

Re: [libvirt] [PATCH] virStorageNetHostDef: Turn @port into integer

On 07/19/2017 05:58 PM, Peter Krempa wrote: > On Wed, Jul 19, 2017 at 17:26:27 +0200, Michal Privoznik wrote: >> Currently, @port is type of string. Well, that's overkill and >> waste of memory. Port is always an integer. Use it as such. >> >> Signed-off-by: Michal Privoznik

Re: [libvirt] [PATCH] virStorageNetHostDef: Turn @port into integer

On Wed, Jul 19, 2017 at 17:58:15 +0200, Peter Krempa wrote: > On Wed, Jul 19, 2017 at 17:26:27 +0200, Michal Privoznik wrote: > > Currently, @port is type of string. Well, that's overkill and > > waste of memory. Port is always an integer. Use it as such. > > > > Signed-off-by: Michal Privoznik

Re: [libvirt] [PATCH 3/5] qemu: blockcopy: reuse storage driver APIs to pre-create copy target

On Wed, Jul 19, 2017 at 17:44:16 -0400, John Ferlan wrote: > > > On 07/11/2017 11:46 AM, Peter Krempa wrote: > > Rather than using the local-file only implementation 'qemuOpenFile' > > switch to the imagelabel aware storage driver implementation. > > --- > > src/qemu/qemu_driver.c | 21

Re: [libvirt] [PATCH] conf: Fix backwards migration of pSeries guests

On Wed, Jul 19, 2017 at 16:41:54 +0200, Andrea Bolognani wrote: > On Wed, 2017-07-19 at 15:33 +0200, Peter Krempa wrote: > > > Recent commits made it so that pci-root controllers for > >  > > Did we release this? > > Nope, it will be in the next release. > > > > +/* Don't format the

Re: [libvirt] [PATCH] LXC: set the right var to NULL

On Thu, 2017-07-20 at 10:02 +0800, Chen Hanxiao wrote: > From: Chen Hanxiao > >    For attaching hosdev, we should set dev->data.hostdev >    rather than dev->data.disk > > Signed-off-by: Chen Hanxiao > --- >  src/lxc/lxc_driver.c | 2 +- >  1 file

[libvirt] [PATCH] nodedev: fix an improper comment

From: Chen Hanxiao Actually we use virConnectNodeDeviceEventGenericCallback. Signed-off-by: Chen Hanxiao --- include/libvirt/libvirt-nodedev.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/libvirt/libvirt-nodedev.h

[libvirt] [PATCH 1/2] qemu: command: Remove qemuNetworkDriveGetPort

Now that the function is only parsing the string to a number, move it's contents to the only caller. --- src/qemu/qemu_command.c | 23 ++- 1 file changed, 6 insertions(+), 17 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index

Re: [libvirt] [PATCH] security: Don't overwrite error of GetProcessLabel

On 07/20/2017 01:22 AM, Cole Robinson wrote: > Security impls of this function already raise errors, don't > overwrite them. > > Signed-off-by: Cole Robinson > --- > src/lxc/lxc_driver.c | 6 ++ > src/qemu/qemu_driver.c | 7 +-- > 2 files changed, 3 insertions(+),

Re: [libvirt] [PATCH RESEND] qemu: shared disks with cache=directsync should be safe for migration

On Sat, Jul 15, 2017 at 11:01:25PM +0800, Peng Hao wrote: > From: Hao Peng > > At present shared disks can be migrated with either readonly or cache=none. > But > cache=directsync should be safe for migration, because both cache=directsync > and cache=none > don't use the

Re: [libvirt] [PATCH go-xml] Add support for DNS in network

On Wed, Jul 12, 2017 at 03:28:13PM +0200, Thomas Hipp wrote: > Add support for DNS in network, and add test code. > > Signed-off-by: Thomas Hipp > --- > network.go | 39 ++ > network_test.go | 65 >

Re: [libvirt] [PATCH go-xml] Add support for NAT in network forward

On Wed, Jul 12, 2017 at 01:26:03PM +0200, Thomas Hipp wrote: > Add support for NAT in network forward, and add test code. > > Signed-off-by: Thomas Hipp > --- > network.go | 20 ++-- > network_test.go | 31 +-- > 2 files changed,

[libvirt] [PATCH 2/2] util: uri: Convert port number to unsigned integer

Negative ports don't make sense so use a unsigned integer. --- src/qemu/qemu_command.c | 3 +-- src/util/viruri.h | 2 +- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index b994940a2..f7858302b 100644 ---

[libvirt] [PATCH 0/2] util: uri: use unsigned numbers for port

Peter Krempa (2): qemu: command: Remove qemuNetworkDriveGetPort util: uri: Convert port number to unsigned integer src/qemu/qemu_command.c | 22 +- src/util/viruri.h | 2 +- 2 files changed, 6 insertions(+), 18 deletions(-) -- 2.13.2 -- libvir-list mailing list

Re: [libvirt] [PATCH v2 5/6] qemu: don't pass qemuctime into virQEMUCapsIsValid

On Wed, Jul 19, 2017 at 17:09:50 +0200, Pavel Hrdina wrote: > It's not required and following patches will change the code. > > Signed-off-by: Pavel Hrdina > --- > src/qemu/qemu_capabilities.c | 27 +++ > 1 file changed, 11 insertions(+), 16

Re: [libvirt] [PATCH v3 00/16] qemu: migration: show disks stats for nbd migration

On 14.07.2017 12:20, Jiri Denemark wrote: > On Fri, Jul 14, 2017 at 09:51:48 +0300, Nikolay Shirokovskiy wrote: >> ping > > Oops, I completely forgot about this series. But since it is a few > months old, could you resend the series after rebasing it to current > libvirt? I promise to review

Re: [libvirt] [PATCH v1] numa: compute and set matching vcpus for numa domains

On Thu, Jul 20, 2017 at 11:29:26AM +0200, Wim Ten Have wrote: > From: Wim ten Have > > The QEMU driver can erroneously allocate more vpus to a domain > than there are cpus in the domain if the element is used > to describe element topology. Fix this by calculating >

Re: [libvirt] [PATCH 2/5] qemu: domain: Extract parsing and formatting of priv->autoNodeset

On Wed, 2017-07-12 at 15:44 +0200, Peter Krempa wrote: >  static int > +qemuDomainObjPrivateXMLParseAutomaticPlacement(virQEMUDriverPtr driver, > +   qemuDomainObjPrivatePtr priv, > +   xmlXPathContextPtr ctxt)

Re: [libvirt] [PATCH v1] numa: compute and set matching vcpus for numa domains

On Thu, Jul 20, 2017 at 03:03:25PM +0200, Wim ten Have wrote: > On Thu, 20 Jul 2017 11:10:31 +0100 > "Daniel P. Berrange" wrote: > > > On Thu, Jul 20, 2017 at 11:29:26AM +0200, Wim Ten Have wrote: > > > From: Wim ten Have > > > > > > The QEMU

Re: [libvirt] [PATCH 1/2] Refactored OVS VLAN configuration

On 07/17/2017 05:48 PM, Antoine Millet wrote: > Moved OVS VLAN configuration arguments construction into a separated > function to be reused. > --- > src/util/virnetdevopenvswitch.c | 104 > > src/util/virnetdevopenvswitch.h | 5 ++ > 2 files changed,

Re: [libvirt] [PATCH 2/2] Handle hotplug change on VLAN configuration using OVS

On 07/17/2017 05:49 PM, Antoine Millet wrote: > A new function virNetDevOpenvswitchUpdateVlan has been created to instruct > OVS of the changes. qemuDomainChangeNet has been modified to handle the > update of the VLAN configuration for a running guest and rely on > virNetDevOpenvswitchUpdateVlan

Re: [libvirt] [PATCH 0/2] Handle hotplug change on VLAN configuration using OVS

On 07/17/2017 05:48 PM, Antoine Millet wrote: > This patch set allow to change VLAN configuration of running guest using OVS > as networking backend. > > Use case here: > https://www.redhat.com/archives/libvirt-users/2017-July/msg00043.html > > "Refactored OVS VLAN configuration" moves the code

Re: [libvirt] [PATCH v2] qemu: Check for existence of provided *_tls_x509_cert_dir

On Wed, Jul 19, 2017 at 04:02:59PM -0400, John Ferlan wrote: On 07/19/2017 09:56 AM, Ján Tomko wrote: On Thu, Jun 29, 2017 at 10:32:35AM -0400, John Ferlan wrote: # # ca-cert.pem - the CA master certificate # server-cert.pem - the server certificate signed with ca-cert.pem @@ -13,6 +13,12 @@

Re: [libvirt] [PATCH 3/5] qemu: domain: Store and restore autoCpuset to status XML

On Thu, 2017-07-20 at 15:46 +0200, Peter Krempa wrote: > > > -virBufferAsprintf(buf, "\n", nodeset); > > > +if (priv->autoCpuset && > > > +!((cpuset = virBitmapFormat(priv->autoCpuset > > > +goto cleanup; > > > > The previous implementation didn't format the > >

Re: [libvirt] [PATCH v1] numa: compute and set matching vcpus for numa domains

On Thu, 20 Jul 2017 11:10:31 +0100 "Daniel P. Berrange" wrote: > On Thu, Jul 20, 2017 at 11:29:26AM +0200, Wim Ten Have wrote: > > From: Wim ten Have > > > > The QEMU driver can erroneously allocate more vpus to a domain > > than there are cpus in

Re: [libvirt] [PATCH 5/5] qemu: process: Don't put memoryless NUMA nodes into autoNodeset

On Wed, 2017-07-12 at 15:44 +0200, Peter Krempa wrote: > 'numad' may return a nodeset which contains NUMA nodes without memory > for certain configurations. Since cgroups code will not be happy using > nodes without memory we need to store only numa nodes with memory in > autoNodeset. >  > On the

Re: [libvirt] [PATCH v1] numa: compute and set matching vcpus for numa domains

On Thu, Jul 20, 2017 at 14:17:29 +0100, Daniel Berrange wrote: > On Thu, Jul 20, 2017 at 03:03:25PM +0200, Wim ten Have wrote: > > On Thu, 20 Jul 2017 11:10:31 +0100 > > "Daniel P. Berrange" wrote: > > > > > On Thu, Jul 20, 2017 at 11:29:26AM +0200, Wim Ten Have wrote: > > >

[libvirt] [PATCH v6 1/4] nodedev: Alter node device deletion logic

Alter the node device deletion logic to make use of the parent field from the obj->def rather than call virNodeDeviceObjListGetParentHost. As it turns out the saved @def won't have parent_wwnn/wwpn or parent_fabric_wwn, so the only logical path would be to call

[libvirt] [PATCH v6 2/4] nodedev: Remove @create from virNodeDeviceObjListGetParentHost

The only callers to this function are from CreateXML paths now, so let's just remove the unnecessary parameter. Signed-off-by: John Ferlan --- src/conf/virnodedeviceobj.c | 5 ++--- src/conf/virnodedeviceobj.h | 3 +-- src/node_device/node_device_driver.c |

[libvirt] [PATCH v6 3/4] nodedev: Convert virNodeDeviceObjListPtr to use hash tables

Rather than use a forward linked list of elements, it'll be much more efficient to use a hash table to reference the elements by unique name and to perform hash searches. This patch does all the heavy lifting of converting the list object to use a self locking list that contains the hash table.

[libvirt] [PATCH v6 0/4] Make virNodeDeviceObj and virNodeDeviceObjList private

NB: Like v5, keeping same title for consistency v5: https://www.redhat.com/archives/libvir-list/2017-July/msg00610.html Changes since v5: * Replace patch 1 w/ the aha moment from review. The deletion API's don't need to use virNodeDeviceObjListGetParentHost. Instead they can just use

Re: [libvirt] [PATCH 0/2] util: uri: use unsigned numbers for port

On Thu, Jul 20, 2017 at 10:00:22 +0200, Peter Krempa wrote: > Peter Krempa (2): > qemu: command: Remove qemuNetworkDriveGetPort > util: uri: Convert port number to unsigned integer Self NACK. I'll repost this later with Michal's cleanup patch, since I'll probably change the first patch along

Re: [libvirt] [PATCH v2 1/3] storage: Fix existing parent check for vHBA creation

On Wed, Jul 19, 2017 at 11:09:00AM -0400, John Ferlan wrote: > > > On 07/19/2017 10:21 AM, Erik Skultety wrote: > > On Wed, Jul 19, 2017 at 09:59:06AM -0400, John Ferlan wrote: > >> > >> > >> On 07/19/2017 07:38 AM, Erik Skultety wrote: > >>> On Tue, Jul 18, 2017 at 11:10:38AM -0400, John Ferlan

Re: [libvirt] [PATCH 4/5] qemu: process: Extract gathering of 'numad' placement into a function

On Wed, 2017-07-12 at 15:44 +0200, Peter Krempa wrote: > +static int > +qemuProcessPrepareDomainNUMAPlacement(virDomainObjPtr vm, > +  virCapsPtr caps) > +{ > +qemuDomainObjPrivatePtr priv = vm->privateData; > +char *nodesetstr = NULL; > +int ret =

Re: [libvirt] [PATCH v5 1/3] nodedev: Fix usage of virNodeDeviceObjListGetParentHost

On 07/20/2017 03:22 AM, Erik Skultety wrote: @@ -594,6 +600,11 @@ nodeDeviceDestroy(virNodeDevicePtr device) int ret = -1; virNodeDeviceObjPtr obj = NULL; virNodeDeviceDefPtr def; +char *name = NULL; +char *parent = NULL; +char

Re: [libvirt] [PATCH 3/5] qemu: domain: Store and restore autoCpuset to status XML

On Thu, Jul 20, 2017 at 14:12:44 +0200, Andrea Bolognani wrote: > On Wed, 2017-07-12 at 15:44 +0200, Peter Krempa wrote: > > @@ -1765,20 +1765,30 @@ > > qemuDomainObjPtrivateXMLFormatAutomaticPlacement(virBufferPtr buf, > >   qemuDomainObjPrivatePtr

[libvirt] [PATCH v6 4/4] nodedev: Remove driver locks around object list mgmt code

Since virnodedeviceobj now has a self-lockable hash table, there's no need to lock the table from the driver for processing. Thus remove the locks from the driver for NodeDeviceObjList mgmt. This includes the test driver as well. Signed-off-by: John Ferlan ---

Re: [libvirt] [PATCH 00/12] qemu: support chardev for all machvirt config

On Thu, 2017-07-20 at 13:17 +0200, Christoffer Dall wrote: > > Can't think of anything specific, but when I backported > > the fixes to libvirt 3.2.0 there were quite a few conflicts > > to take care of, so if Linaro is using a much older libvirt > > version I can imagine the backport would get

Re: [libvirt] [PATCH v2 6/6] qemu: separate virQEMUCapsInitCached out of virQEMUCapsNewForBinaryInternal

On Wed, Jul 19, 2017 at 17:09:51 +0200, Pavel Hrdina wrote: > Preparation for switching to virFileCache where there are two callbacks, > one to get a new data and second one to load a cached data. > > This also removes virQEMUCapsReset which is no longer required. > > Signed-off-by: Pavel Hrdina

Re: [libvirt] [PATCH 00/12] qemu: support chardev for all machvirt config

Hi Andrea, On Wed, Jul 19, 2017 at 03:07:25PM +0200, Andrea Bolognani wrote: > On Tue, 2017-07-04 at 11:25 +0200, Christoffer Dall wrote: > > I tried with two mainline QEMU versions (the default old one packaged > > with Debian and a recent upstream commit) and this series seems to work > > as

Re: [libvirt] [PATCH 00/12] qemu: support chardev for all machvirt config

On Thu, 2017-07-20 at 12:55 +0200, Christoffer Dall wrote: > > Christoffer, > >  > > now that the series has been merged, would you mind updating > >  > >   https://bugs.linaro.org/show_bug.cgi?id=2777 > >  > > with the relevant information? >  > I have updated the bug documenting that this works

[libvirt] [PATCH 2/2] tests: Free @fakerootdir in error path

Commit id 'dd9b29dad' added this new variable, but didn't free it in one instance where status was returned to the caller. Found by Coverity Signed-off-by: John Ferlan --- tests/qemumemlocktest.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git

[libvirt] [PATCH 1/2] daemon: Don't conditionally free @origErr in daemonStreamEvent

Commit id '0fe4aa149' added @origErr, but since it's assigned outside the if condition, the free should be outside as well. Found by Coverity Signed-off-by: John Ferlan --- daemon/stream.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/daemon/stream.c

Re: [libvirt] 答复: [PATCH] fix 1 << -1 at JOB_MASK macro

On Tue, Jul 18, 2017 at 02:04:27 +, Caoxinhua wrote: > When we start a vm, we call JOB_MASK(QEMU_JOB_NONE) at follow backtrace > > Breakpoint 1, qemuDomainObjPrivateXMLFormat (buf=0x7f8078fcd510, > vm=0x7f808c2c7d70) at qemu/qemu_domain.c:1779 > 1779if (!qemuDomainTrackJob(job)) >

Re: [libvirt] [PATCH 0/2] Fix a couple of resource leaks

On Thu, Jul 20, 2017 at 07:00:38AM -0400, John Ferlan wrote: > Resource a couple Coverity found resource leaks > > John Ferlan (2): > daemon: Don't conditionally free @origErr in daemonStreamEvent > tests: Free @fakerootdir in error path > > daemon/stream.c | 2 +- >

[libvirt] [PATCH v1] numa: compute and set matching vcpus for numa domains

From: Wim ten Have The QEMU driver can erroneously allocate more vpus to a domain than there are cpus in the domain if the element is used to describe element topology. Fix this by calculating the number of cpus described in the element of a element and comparing it

Re: [libvirt] [PATCH 3/5] qemu: domain: Store and restore autoCpuset to status XML

On Wed, 2017-07-12 at 15:44 +0200, Peter Krempa wrote: > @@ -1765,20 +1765,30 @@ > qemuDomainObjPtrivateXMLFormatAutomaticPlacement(virBufferPtr buf, >   qemuDomainObjPrivatePtr >priv) >  { [...] > -virBufferAsprintf(buf, "\n", nodeset); > +

[libvirt] [PATCH v2 6/6] virStorageNetHostDef: Turn @port into integer

Currently, @port is type of string. Well, that's overkill and waste of memory. Port is always an integer. Use it as such. Signed-off-by: Michal Privoznik --- src/conf/domain_conf.c| 20 +++--- src/libxl/libxl_conf.c| 2 +-

[libvirt] [PATCH v2 5/6] conf: domain: Split up virDomainStorageHostParse and rename it

Split out parsing of one host into a separate function and add a new function to loop through all the host XML nodes. This change removes multiple levels of nesting due to the old XML parsing approach used. --- src/conf/domain_conf.c | 117 +++-- 1

[libvirt] [PATCH v2 4/6] util: storage: fill in default ports when parsing backing chain

Similarly to when parsing XML we need to fill in default ports for the backing chain. This was missed in commit 5bda835466a8050625dd8bb10566e --- src/util/virstoragefile.c | 3 +++ tests/virstoragetest.c| 14 +++--- 2 files changed, 10 insertions(+), 7 deletions(-) diff --git

[libvirt] [PATCH v2 2/6] util: uri: Convert port number to unsigned integer

Negative ports don't make sense so use a unsigned integer. --- src/util/virstring.c | 2 +- src/util/virstring.h | 2 +- src/util/viruri.h| 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) diff --git a/src/util/virstring.c b/src/util/virstring.c index fe61a3516..69e0fd173 100644 ---

[libvirt] [PATCH v2 1/6] qemu: command: Rename and move qemuNetworkDriveGetPort

Move it to virstring.c and improve it to parse and validate ports. New name is virStringParsePort. --- src/libvirt_private.syms | 1 + src/qemu/qemu_command.c | 18 +- src/util/virstring.c | 37 + src/util/virstring.h | 4 4

[libvirt] [PATCH v2 3/6] qemu: command: Remove condition to use default sheepdog port

Since we now set the default ports when parsing disks, it's not necessary to have default port numbers encoded in the command line generator. --- src/qemu/qemu_command.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index

[libvirt] [PATCH v2 0/6] Convert port to unsigned integer

Since Michal did not want to finis this I did. I also found a few bugs in my previous patches which are addressed here. Peter Krempa (6): qemu: command: Rename and move qemuNetworkDriveGetPort util: uri: Convert port number to unsigned integer qemu: command: Remove condition to use default

[libvirt] [PATCH 0/2] Fix a couple of resource leaks

Resource a couple Coverity found resource leaks John Ferlan (2): daemon: Don't conditionally free @origErr in daemonStreamEvent tests: Free @fakerootdir in error path daemon/stream.c | 2 +- tests/qemumemlocktest.c | 4 +++- 2 files changed, 4 insertions(+), 2 deletions(-) --

Re: [libvirt] [PATCH 1/5] util: bitmap: Modify virBitmapSubtract to virBitmapIntersect

On Wed, 2017-07-12 at 15:44 +0200, Peter Krempa wrote: > Since virBitmapSubtract is unused modify it to perform bitmap > intersection. Normally I would ask you to remove the old function in one patch and introduce the new one in another, but I can see you manage to reduce the test suite churn by

Re: [libvirt] [PATCH 00/12] qemu: support chardev for all machvirt config

On Thu, Jul 20, 2017 at 1:16 PM, Andrea Bolognani wrote: > On Thu, 2017-07-20 at 12:55 +0200, Christoffer Dall wrote: >> > Christoffer, >> > >> > now that the series has been merged, would you mind updating >> > >> > https://bugs.linaro.org/show_bug.cgi?id=2777 >> > >> >

[libvirt] [PATCH v1] numa: compute and set matching vcpus for numa domains

From: Wim ten Have The QEMU driver can erroneously allocate more vpus to a domain than there are cpus in the domain if the element is used to describe element topology. Wim ten Have (1): numa: compute and set matching vcpus for numa domains

Re: [libvirt] [PATCH] nodedev: fix an improper comment

On Thu, Jul 20, 2017 at 05:01:57PM +0800, Chen Hanxiao wrote: > From: Chen Hanxiao > > Actually we use virConnectNodeDeviceEventGenericCallback. > > Signed-off-by: Chen Hanxiao Slightly adjusted the commit subject and pushed. Thanks, Erik --

Re: [libvirt] [PATCH] rpc : fix a access for null pointer

On 07/15/2017 05:00 PM, Peng Hao wrote: > virNetSocketRemoveIOCallback get sock's ObjectLock and will call > virNetSocketEventFree. virNetSocketEventFree may be free sock > object and virNetSocketRemoveIOCallback will access a null pointer > in release sock's ObjectLock. > > Signed-off-by: Liu

Re: [libvirt] [PATCH v2 6/6] qemu: separate virQEMUCapsInitCached out of virQEMUCapsNewForBinaryInternal

On Thu, Jul 20, 2017 at 01:32:36PM +0200, Jiri Denemark wrote: > On Wed, Jul 19, 2017 at 17:09:51 +0200, Pavel Hrdina wrote: > > Preparation for switching to virFileCache where there are two callbacks, > > one to get a new data and second one to load a cached data. > > > > This also removes

Re: [libvirt] [PATCH 00/12] qemu: support chardev for all machvirt config

On Thu, Jul 20, 2017 at 4:27 PM, Andrea Bolognani wrote: > On Thu, 2017-07-20 at 13:17 +0200, Christoffer Dall wrote: >> > Can't think of anything specific, but when I backported >> > the fixes to libvirt 3.2.0 there were quite a few conflicts >> > to take care of, so if

Re: [libvirt] [PATCH v2 3/3] qemu: Enable NUMA node tag in pci-root for PPC64

On Mon, 2017-07-17 at 21:29 +0530, Shivaprasad G Bhat wrote: > @@ -3786,6 +3786,11 @@ >  part of the specified NUMA node (it is up to the user of the >  libvirt API to attach host devices to the correct >  pci-expander-bus when assigning them to the domain). > +On

Re: [libvirt] [PATCH v5 2/3] nodedev: Convert virNodeDeviceObjListPtr to use hash tables

> @@ -39,13 +40,19 @@ struct _virNodeDeviceObj { > }; > > struct _virNodeDeviceObjList { > -size_t count; > -virNodeDeviceObjPtr *objs; > +virObjectLockable parent; > + > +/* name string -> virNodeDeviceObj mapping > + * for O(1), lockless lookup-by-uuid */ I think you

Re: [libvirt] [PATCHv2 0/3] exposing busy polling support for vhost-net

On 07/17/2017 07:39 AM, Peter Krempa wrote: > On Mon, Jul 17, 2017 at 07:27:24 -0400, sferd...@redhat.com wrote: >> From: Sahid Orentino Ferdjaoui >> >> In version 2.7.0, QEMU introduced support of busy polling for >> vhost-net [0]. To avoid paraphrasing original

Re: [libvirt] [PATCH v2] network: allow to specify buffer size for netlink socket

On 07/17/2017 08:00 AM, John Ferlan wrote: > > On 07/16/2017 05:04 PM, Laine Stump wrote: >> On 07/11/2017 09:01 PM, ZhiPeng Lu wrote: >>> This patchs allow to set the buffer size for netlink socket in >>> the libvirtd configuration file. The default buffer size remain >>> as before at 128k. >>

Re: [libvirt] [PATCH v2] Add support for virtio-net.tx_queue_size

On 07/18/2017 07:12 AM, Michal Privoznik wrote: > [Adding MST who wrote qemu patches] > > On 07/18/2017 12:21 PM, Peter Krempa wrote: >> On Tue, Jul 18, 2017 at 09:01:31 +0200, Michal Privoznik wrote: >>> On 07/18/2017 08:23 AM, Peter Krempa wrote: On Mon, Jul 17, 2017 at 15:39:56 +0200,

[libvirt] 答复: Re: [PATCH v2] network: allow to specify buffer size fornetlink socket

>On 07/17/2017 08:00 AM, John Ferlan wrote:>>> On 07/16/2017 05:04 PM, Laine >Stump wrote:>> On 07/11/2017 09:01 PM, ZhiPeng Lu wrote:>>> This patchs allow >to set the buffer size for netlink socket in>>> the libvirtd configuration >file. The default buffer size remain>>> as before at 128k.>>

[libvirt] [PATCH v3 2/2] virt-aa-helper-test: Add test for aarch32 UEFI image path

Signed-off-by: dann frazier --- tests/virt-aa-helper-test | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/virt-aa-helper-test b/tests/virt-aa-helper-test index 51072f622c..0599046408 100755 --- a/tests/virt-aa-helper-test +++ b/tests/virt-aa-helper-test @@

[libvirt] [PATCH v3 1/2] qemu: Add AAVMF32 to the list of known UEFIs

Add a path for UEFI VMs for AArch32 VMs, based on the path Debian is using. libvirt is the de facto canonical location for defining where distros should place these firmware images, so let's define this path here to try and minimize distro fragmentation. --- src/qemu/qemu.conf

[libvirt] [PATCH v3 0/2] qemu: Add AAVMF32 to the list of known UEFIs

Changes since v2: * Include updates to qemu.conf * Update tests for parity w/ AAVMF * Drop reference to 'experimental' in commit message Changes since v1: * Add a virt-aa-helper-test case. dann frazier (2): qemu: Add AAVMF32 to the list of known UEFIs virt-aa-helper-test: Add test for

Re: [libvirt] [PATCH v5 2/3] nodedev: Convert virNodeDeviceObjListPtr to use hash tables

On 07/20/2017 10:48 AM, Erik Skultety wrote: >> @@ -39,13 +40,19 @@ struct _virNodeDeviceObj { >> }; >> >> struct _virNodeDeviceObjList { >> -size_t count; >> -virNodeDeviceObjPtr *objs; >> +virObjectLockable parent; >> + >> +/* name string -> virNodeDeviceObj mapping >> +

Re: [libvirt] [PATCH v3 1/2] qemu: Add AAVMF32 to the list of known UEFIs

On 07/20/2017 03:56 PM, dann frazier wrote: > Add a path for UEFI VMs for AArch32 VMs, based on the path Debian is using. > libvirt is the de facto canonical location for defining where distros > should place these firmware images, so let's define this path here to try > and minimize distro

[libvirt] [PATCH] tools: virsh: Adding unix socket support to 'domdisplay' command.

This commit adds the unix socket URL support to 'domdisplay' command. Before, even if an user was using unix socket to define a spice graphics, the command 'domdisplay' showed that the settings were not supported. Now, the command shows the proper URL: spice+unix://foo/bar.sock. Resolves:

Re: [libvirt] [PATCH 08/19] storage: Use consistent variable names for driver

On 07/14/2017 11:07 AM, Pavel Hrdina wrote: > On Tue, May 09, 2017 at 11:30:15AM -0400, John Ferlan wrote: >> A virStoragePoolObjPtr will be an 'obj'. >> >> A virStoragePoolPtr will be a 'pool'. >> >> A virStorageVolPtr will be a 'vol'. >> >> A virStorageVolDefPtr will be a 'voldef'. >> >>

[libvirt] [PATCH] docs: fix documentation of enum constants

The documentation string has to follow the definition of a constant in the enum. Otherwise, the HTML documentation will be generated incorrectly. Signed-off-by: Tomáš Golembiovský --- include/libvirt/libvirt-domain.h | 62 1 file

Re: [libvirt] [PATCH v2 1/3] storage: Fix existing parent check for vHBA creation

[...] >> >> Because it hasn't really been characterized as a misconfiguration >> previously. I doubt anyone outside of QE has ever done something like >> this as there's no reason to do so. If they want to use the HBA they'd >> use the 'scsi_host' format. >> >> IMO: something with 'fc_host' what

[libvirt] [PATCH v3 1/4] storage: Fix existing parent check for vHBA creation

https://bugzilla.redhat.com/show_bug.cgi?id=1472277 Commit id '106930aaa' altered the order of checking for an existing vHBA (e.g something created via nodedev-create functionality outside of the storage pool logic) which inadvertantly broke the code to decide whether to alter/force the

[libvirt] [PATCH v3 4/4] storage: Disallow usage of the HBA for a fc_host backing

Disallow providing the wwnn/wwpn of the HBA in the adapter XML: This should be considered a configuration error since a vHBA would not be created. In order to use the HBA as the backing the following XML should be used: This also alters the caller such that the @parent_name param into

[libvirt] [PATCH v3 3/4] storage: Check if provided parent is vHBA capable

https://bugzilla.redhat.com/show_bug.cgi?id=1458708 If the parent provided for the storage pool adapter is not vHBA capable, then issue a configuration error even though the provided wwnn/wwpn were found. It is a configuration error to provide a mismatched parent to the wwnn/wwpn. The @parent is

[libvirt] [PATCH v3 0/4] Fix a couple issues found w/ vHBA logic

v2: https://www.redhat.com/archives/libvir-list/2017-July/msg00661.html Changes since v2: * Patch 1 - make the requested logic adjustment * Patch 2 - no change * Patch 3 - make this *just* the check that the provided @parent_name is a valid fc_host (whether it's vHBA or HBA).

[libvirt] [PATCH v3 2/4] storage: Remove @conn from virNodeDeviceCreateVport

It's no longer needed since the checkParent code moved back to storage_backend_scsi.c Signed-off-by: John Ferlan --- src/conf/node_device_conf.c| 8 +++- src/conf/node_device_conf.h| 3 +-- src/storage/storage_backend_scsi.c | 2 +- 3 files changed, 5

Re: [libvirt] [PATCH 10/19] storage: Create accessor API's for virStoragePoolObj

On 07/14/2017 11:37 AM, Pavel Hrdina wrote: > On Tue, May 09, 2017 at 11:30:17AM -0400, John Ferlan wrote: >> In preparation for making a private object, create accessor API's for >> consumer storage functions to use: >> >> virStoragePoolObjGetDef >> virStoragePoolObjGetNewDef >>

Re: [libvirt] [PATCH 10/19] storage: Create accessor API's for virStoragePoolObj

On 07/14/2017 09:55 AM, Pavel Hrdina wrote: > On Tue, May 09, 2017 at 11:30:17AM -0400, John Ferlan wrote: >> In preparation for making a private object, create accessor API's for >> consumer storage functions to use: >> >> virStoragePoolObjGetDef >> virStoragePoolObjGetNewDef >>

Re: [libvirt] [PATCH 11/19] storage: Introduce virStoragePoolObjNew

On 07/11/2017 11:17 AM, Pavel Hrdina wrote: > On Tue, May 09, 2017 at 11:30:18AM -0400, John Ferlan wrote: >> Create/use a helper to perform object allocation. >> >> Signed-off-by: John Ferlan >> --- >> src/conf/virstorageobj.c | 34 +++--- >> 1