[PATCH 1/2] qemu_capabilities: Introduce QEMU_CAPS_ACCEL

From: Hyman The "-machine" options for accelerators are legacy, the "-accel" options is a better mechanism. The following are the details: https://lore.kernel.org/qemu-devel/3aa73987-40e8-3619-0723-9f17f7385...@redhat.com/ This patch introduce QEMU_CAPS_ACCEL capability to tell if we're dealing

[PATCH 0/2] specify the accelerator type using "-accel" parameter

From: Hyman This patchset aims to support dirty ring feature which has been introduced in kernel: https://lore.kernel.org/lkml/8b3f68dd-c61c-16a0-2077-0a5d3d57a...@redhat.com/ The QEMU patchset is also being applied for merge, the review may be under the way:

Re: [PATCH v3 3/3] qemu: suppor dirty ring feature

在 2021/1/13 21:32, Han Han 写道: On Wed, Jan 13, 2021 at 1:09 AM > wrote: From: Hyman mailto:huang...@chinatelecom.cn>> QEMU introduced a dirty ring feature, this patch add a new KVM feature 'dirty-ring' to set this feature for kvm guests. To

Re: [libvirt PATCH] qemu: Fix memstat for (non-)transitional memballoon

On 1/12/21 6:47 PM, Andrea Bolognani wrote: Depending on the memballoon model, the corresponding QOM node will have a different type and we need to account for this when searching for it in the QOM tree. https://bugzilla.redhat.com/show_bug.cgi?id=1911786 Signed-off-by: Andrea Bolognani ---

Re: [PATCH v3 3/3] qemu: suppor dirty ring feature

On Wed, Jan 13, 2021 at 1:09 AM wrote: > From: Hyman > > QEMU introduced a dirty ring feature, this patch add a new > KVM feature 'dirty-ring' to set this feature for kvm guests. > > To enable the feature, libvirt add "-accel dirty-gfn-count=xxx" > to QEMU command line, the following XML needs

Re: [PATCHv3 4/4] netlink: Introduce a helper function to simplify netlink functions

On 1/12/21 2:29 AM, Shi Lei wrote: On 2021-01-11 at 19:26, Michal Privoznik wrote: On 1/11/21 3:23 AM, Shi Lei wrote: Extract common code as helper function virNetlinkTalk, then simplify the functions virNetlink[DumpLink|NewLink|DelLink|GetNeighbor]. Signed-off-by: Shi Lei ---   

Re: [PATCH 2/2] qemu: Do not Use canonical path for system memory

On Wed, Jan 13, 2021 at 08:06:30 +0100, Peter Krempa wrote: > On Tue, Jan 12, 2021 at 21:13:54 +0100, Jiri Denemark wrote: > > On Tue, Jan 12, 2021 at 20:21:19 +0100, Igor Mammedov wrote: > > > On Tue, 12 Jan 2021 09:29:50 +0100 > > > Michal Privoznik wrote: > > [...] > > > > > We consume

[RFC v5 3/3] qemu: suppor dirty ring feature

From: Hyman QEMU introduced a dirty ring feature, this patch add a new KVM feature 'dirty-ring' to set this feature for kvm guests. To enable the feature, libvirt add "-accel dirty-gfn-count=xxx" to QEMU command line, the following XML needs to be added to the guest's domain description:

[RFC v5 1/3] qemu_capabilities: Introduce QEMU_CAPS_ACCEL

From: Hyman The "-machine" options for accelerators are legacy, the "-accel" options is a better mechanism. The following are the details: https://lore.kernel.org/qemu-devel/3aa73987-40e8-3619-0723-9f17f7385...@redhat.com/ This patch introduce QEMU_CAPS_ACCEL capability to tell if we're dealing

[RFC v5 0/3] Dirty Ring support (Libvirt)

From: Hyman The v5 of Dirty Ring correct the mistake introduced in v3 patch. Which using the shell script to modify the hundereds of test cases in qemuxml2argvdata directory, and do not take the symlink into account. The following is detail:

Re: [PATCH] NEWS: Mention stuff I've dealt with for 7.0.0

On 1/13/21 10:50 AM, Ján Tomko wrote: On a Wednesday in 2021, Michal Privoznik wrote: Signed-off-by: Michal Privoznik --- NEWS.rst | 47 +++ 1 file changed, 47 insertions(+) +  * conf: Add support for keeping TPM emulator state + Is this a new

Re: [PATCH] NEWS: Mention stuff I've dealt with for 7.0.0

On Wed, 2021-01-13 at 10:23 +0100, Michal Privoznik wrote: > + * virnetdevopenvswitch: Various improvements > + > +The code that handles was given various > +improvements. So far, libvirt assumed vhostuser interfaces are handled > +exclusively by OpenVSwitch and refused to start

Re: [PATCH] test-wrap-argv.py: Actually print filename of offending script

On Wed, Jan 13, 2021 at 06:42:40 -0300, Daniel Henrique Barboza wrote: > > > On 1/12/21 3:51 PM, Peter Krempa wrote: > > The error message doesn't actually print the filename of the offending > > file: > > > > Incorrect line wrapping in %file > > > > Signed-off-by: Peter Krempa > > --- > >

Re: [PATCH] NEWS: Mention stuff I've dealt with for 7.0.0

On a Wednesday in 2021, Michal Privoznik wrote: Signed-off-by: Michal Privoznik --- NEWS.rst | 47 +++ 1 file changed, 47 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index 07cec19c2b..b9a3d94eb2 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -48,6 +48,28

Re: [PATCH] test-wrap-argv.py: Actually print filename of offending script

On 1/12/21 3:51 PM, Peter Krempa wrote: The error message doesn't actually print the filename of the offending file: Incorrect line wrapping in %file Signed-off-by: Peter Krempa --- scripts/test-wrap-argv.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git

Re: [libvirt PATCH] qemu: Fix memstat for (non-)transitional memballoon

On Wed, Jan 13, 2021 at 10:18:35 +0100, Andrea Bolognani wrote: > On Tue, 2021-01-12 at 19:56 +0100, Peter Krempa wrote: > > On Tue, Jan 12, 2021 at 18:47:36 +0100, Andrea Bolognani wrote: > > > +switch (balloon->model) { [...] > > > +default: > > > +

Re: [RFC v4 1/4] tests: Add 'end-of-file' character to args file in qemuxml2argvdata

On Wed, Jan 13, 2021 at 09:19:19 +, huang...@chinatelecom.cn wrote: Firstly, please don't CC libvirt developers directly, since we are subscribed to the list, and we don't need a copy of your patches in our inbox. In this case it's namely, me, Daniel Berrange, and Jan Tomko. > From: Hyman >

[PATCH] NEWS: Mention stuff I've dealt with for 7.0.0

Signed-off-by: Michal Privoznik --- NEWS.rst | 47 +++ 1 file changed, 47 insertions(+) diff --git a/NEWS.rst b/NEWS.rst index 07cec19c2b..b9a3d94eb2 100644 --- a/NEWS.rst +++ b/NEWS.rst @@ -48,6 +48,28 @@ v7.0.0 (unreleased) an usage name which

Re: [libvirt PATCH] qemu: Fix memstat for (non-)transitional memballoon

On Tue, 2021-01-12 at 19:56 +0100, Peter Krempa wrote: > On Tue, Jan 12, 2021 at 18:47:36 +0100, Andrea Bolognani wrote: > > +switch (balloon->model) { > > This is an 'int'. Please typecast it appropriately ... > > > +case VIR_DOMAIN_MEMBALLOON_MODEL_VIRTIO: > > +

[RFC v4 4/4] qemu: suppor dirty ring feature

From: Hyman QEMU introduced a dirty ring feature, this patch add a new KVM feature 'dirty-ring' to set this feature for kvm guests. To enable the feature, libvirt add "-accel dirty-gfn-count=xxx" to QEMU command line, the following XML needs to be added to the guest's domain description:

[RFC v4 2/4] qemu_capabilities: Introduce QEMU_CAPS_ACCEL

From: Hyman The "-machine" options for accelerators are legacy, the "-accel" options is a better mechanism. The following are the details: https://lore.kernel.org/qemu-devel/3aa73987-40e8-3619-0723-9f17f7385...@redhat.com/ This patch introduce QEMU_CAPS_ACCEL capability to tell if we're dealing

[RFC v4 1/4] tests: Add 'end-of-file' character to args file in qemuxml2argvdata

From: Hyman The test file disk-backing-chains-noindex.x86_64-latest.args has 'No newline at end of file', this patch rewrite it in case unexpected behavior bewteen diffrent platform. See the details: https://www.redhat.com/archives/libvir-list/2021-January/msg00626.html Signed-off-by: Hyman

[RFC v4 0/4] Dirty Ring support (Libvirt)

From: Hyman The v4 of Dirty Ring just fix args file in testsuite which has 'No-newline-at-end-of-file' and mak it a stand-alone patch. The following is detail: https://www.redhat.com/archives/libvir-list/2021-January/msg00626.html The v3 of Dirty Ring do some modification based on the

libvirt-7.0.0 release candidate 2

I have just tagged v7.0.0-rc2 in the repository and pushed signed tarballs and source RPMs to https://libvirt.org/sources/ Please give the release candidate some testing and in case you find a serious issue which should have a fix in the upcoming release, feel free to reply to this thread to make