- Devel - libvirt List Archives

[libvirt] [PATCH] storage_util: Prefer generic FICLONE over btrfs/xfs defines
by Michal Privoznik 10 Jul '18

10 Jul '18

After my change to the original patch that resulted in commit 8ed874b39b3 it was brought to my attention that all three defines are the same: FICLONE = BTRFS_IOC_CLONE = XFS_IOC_CLONE. Therefore we should prefer generic FICLONE over 'specific' defines for btrfs/xfs. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/storage/storage_util.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/src/storage/storage_util.c b/src/storage/storage_util.c index da99043e0a..715d5c2f88 100644 --- a/src/storage/storage_util.c +++ b/src/storage/storage_util.c @@ -46,14 +46,14 @@ # include <selinux/selinux.h> #endif -#if HAVE_LINUX_BTRFS_H +#ifdef FICLONE +# define REFLINK_IOC_CLONE FICLONE +#elif HAVE_LINUX_BTRFS_H # include <linux/btrfs.h> # define REFLINK_IOC_CLONE BTRFS_IOC_CLONE #elif HAVE_XFS_XFS_H # include <xfs/xfs.h> # define REFLINK_IOC_CLONE XFS_IOC_CLONE -#elif defined(FICLONE) -# define REFLINK_IOC_CLONE FICLONE #endif #include "datatypes.h" -- 2.16.4

2 1

[libvirt] [dbus PATCH] test_connect: fix methods to follow naming conventions
by Anya Harter 10 Jul '18

10 Jul '18

Signed-off-by: Anya Harter <aharter(a)redhat.com> --- tests/test_connect.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/test_connect.py b/tests/test_connect.py index 7084397..2299b8a 100755 --- a/tests/test_connect.py +++ b/tests/test_connect.py @@ -22,7 +22,7 @@ class TestConnect(libvirttest.BaseTestClass): self.main_loop() - def test_comnect_domain_define_xml(self): + def test_connect_domain_define_xml(self): def domain_defined(path, event, detail): if event != libvirttest.DomainEvent.DEFINED: return @@ -84,7 +84,7 @@ class TestConnect(libvirttest.BaseTestClass): sysinfo = self.connect.GetSysinfo(0) assert isinstance(sysinfo, dbus.String) - def test_list_networks(self): + def test_connect_list_networks(self): networks = self.connect.ListNetworks(0) assert isinstance(networks, dbus.Array) assert len(networks) == 1 -- 2.17.1

2 1

[libvirt] [dbus PATCH] connect: fix gchar declaration in connect.h
by Anya Harter 10 Jul '18

10 Jul '18

so that gchar *nodeDevPath; line appears in alphabetical order with the rest of the lines Signed-off-by: Anya Harter <aharter(a)redhat.com> --- src/connect.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/connect.h b/src/connect.h index b81b6a8..0b9ae10 100644 --- a/src/connect.h +++ b/src/connect.h @@ -12,9 +12,9 @@ struct virtDBusConnect { GDBusConnection *bus; const gchar *uri; const gchar *connectPath; - gchar *nodeDevPath; gchar *domainPath; gchar *networkPath; + gchar *nodeDevPath; gchar *nwfilterPath; gchar *secretPath; gchar *storagePoolPath; -- 2.17.1

2 1

[libvirt] [PATCH] news: Update for 4.5.0 release
by Andrea Bolognani 10 Jul '18

10 Jul '18

Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- I've CC'd all those who introduced the changes mentioned below, both to give them a chance to point out any mistake I might have made and to possibly annoy them into updating the release notes themselves next time :) docs/news.xml | 83 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 83 insertions(+) diff --git a/docs/news.xml b/docs/news.xml index 468d34093a..65c23e51b6 100644 --- a/docs/news.xml +++ b/docs/news.xml @@ -61,6 +61,16 @@ Support specifying extended TSEG size for SMM in QEMU. </description> </change> + <change> + <summary> + qemu: Add support for SEV guests + </summary> + <description> + SEV (Secure Encrypted Virtualization) is a feature available on AMD + CPUs that encrypts the guest memory and makes it inaccessible even + to the host OS. + </description> + </change> </section> <section title="Removed features"> <change> @@ -76,6 +86,24 @@ secret objects, but that support was never added to libvirt. </description> </change> + <change> + <summary> + Make GnuTLS mandatory + </summary> + <description> + Building without GnuTLS is no longer possible. + </description> + </change> + <change> + <summary> + qemu: Remove allow_disk_format_probing configuration option + </summary> + <description> + The option represented a security risk when used with malicious + disk images, so users were recommended against enabling it; with + this release, it's been removed altogether. + </description> + </change> </section> <section title="Improvements"> <change> @@ -130,6 +158,41 @@ or virStorageVolCreateXMLFrom. </description> </change> + <change> + <summary> + qemu: Add support for vsock hot (un)plug and cold (un)plug + </summary> + </change> + <change> + <summary> + qemu: Add support for NBD over TLS + </summary> + <description> + Securely accessing NBD volumes no longer requires tunnelling over + SSH or another secure protocol: the native TLS support can now be + used instead. + </description> + </change> + <change> + <summary> + qemu: Implement FD passing for Unix sockets + </summary> + <description> + Instead of having QEMU open the socket and then connecting to it, + which is inherently racy, starting with QEMU 2.12 we can open the + socket ourselves and pass it to QEMU, avoiding race conditions. + </description> + </change> + <change> + <summary> + virsh: Introduce --nowait option for domstat command + </summary> + <description> + When this option is specified, virsh will try to fetch the guest + stats but abort instead of stalling if they can't be retrieved right + away. + </description> + </change> </section> <section title="Bug fixes"> <change> @@ -143,6 +206,26 @@ us getting to the worker pool initialization. </description> </change> + <change> + <summary> + qemu: Fix domain resume after failed migration + </summary> + <description> + Recent versions of QEMU activate block devices before the guest CPU + has been started, which makes it impossible to roll back a failed + migration. Use the <code>late-block-activate</code> migration + capability if supported to avoid the issue. + </description> + </change> + <change> + <summary> + vmx: Permit guests to have an odd number of vCPUs + </summary> + <description> + This was forbidden in the past, but current versions of ESX can + deal with a guest configured as such. + </description> + </change> </section> </release> <release version="v4.4.0" date="2018-06-04"> -- 2.17.1

5 8

[libvirt] [PATCH 0/5] Making libvirt aware of a server need
by Marcos Paulo de Souza 10 Jul '18

10 Jul '18

Hi guys, this is basically a second version of [1], but I rewrote the patch following Matthias ideas [2]. It really seems a cleaner change, but, feel free to suggest any other change that could make the code better. I did a lot of tests using virsh tool, so when using the ESX driver without a proper server, it returns a error. I could not test hyperv and phyp. Thanks, [1]: https://www.redhat.com/archives/libvir-list/2018-July/msg00393.html [2]: https://www.redhat.com/archives/libvir-list/2018-July/msg00400.html Marcos Paulo de Souza (5): driver.h: Add remoteOnly member to virConnectDriver struct libvirt.c: Return error when remoteOnly is set but server is empty esx_driver: Set remoteOnly member of virConnectDriver hyperv_driver: Set remoteOnly member of virConnectDriver phyp_driver: Set remoteOnly member of virConnectDriver src/driver.h | 2 ++ src/esx/esx_driver.c | 8 +------- src/hyperv/hyperv_driver.c | 8 +------- src/libvirt.c | 5 +++++ src/phyp/phyp_driver.c | 7 +------ 5 files changed, 10 insertions(+), 20 deletions(-) -- 2.17.1

1 6

[libvirt] [PATCH 0/5] Introduce new video model type 'none'
by Erik Skultety 09 Jul '18

09 Jul '18

Historically, we've always been adding a default video device onto the cmdline whenever a graphical framebuffer was requested but a video device was missing. With the appearance of mdev vgpus, having an emulated video device is suboptimal, especially with spice where the streaming client will by default open multiple windows, one for each video device (or 'head' in this case). Therefore, we should have a mechanism to disable the 'default video device addition' and this series does that by introducing a new video model 'none'. This can be applied and tried independently, however, it truly only makes sense on top of [1]. [1] https://www.redhat.com/archives/libvir-list/2018-June/msg01740.html Erik Skultety (5): qemu: address: Handle all the video devices within a single loop conf: Introduce virDomainVideoDefClear helper conf: Introduce virDomainDefPostParseVideo helper qemu: validate: Enforce compile time switch type checking for videos conf: Introduce new video type 'none' docs/formatdomain.html.in | 10 ++- docs/schemas/domaincommon.rng | 1 + src/conf/domain_conf.c | 86 +++++++++++++++++----- src/conf/domain_conf.h | 4 +- src/libvirt_private.syms | 1 + src/qemu/qemu_command.c | 13 +++- src/qemu/qemu_domain.c | 6 +- src/qemu/qemu_domain_address.c | 20 +++-- tests/domaincapsschemadata/full.xml | 1 + .../video-invalid-multiple-devices.xml | 33 +++++++++ tests/qemuxml2argvdata/video-none-device.args | 27 +++++++ tests/qemuxml2argvdata/video-none-device.xml | 39 ++++++++++ tests/qemuxml2argvtest.c | 4 +- tests/qemuxml2xmloutdata/video-none-device.xml | 42 +++++++++++ tests/qemuxml2xmltest.c | 1 + 15 files changed, 255 insertions(+), 33 deletions(-) create mode 100644 tests/qemuxml2argvdata/video-invalid-multiple-devices.xml create mode 100644 tests/qemuxml2argvdata/video-none-device.args create mode 100644 tests/qemuxml2argvdata/video-none-device.xml create mode 100644 tests/qemuxml2xmloutdata/video-none-device.xml -- 2.14.4

2 11

[libvirt] [PATCH v1 00/11] Enable vfio-pci 'property' for mediated device
by Erik Skultety 09 Jul '18

09 Jul '18

See the RFC here: https://www.redhat.com/archives/libvir-list/2018-May/msg02218.html Since RFC: - split graphics 'gl' to a standalone structure since it's now SPICE, SDL, *and* VNC that will support it - egl-headless support for VNC, since VNC doesn't support OpenGL natively like SPICE does - added a new attribute 'native' for spice which will instruct libvirt to use '-display egl-headless' instead of libvirt trying to figure this on out by itself, since egl-headless might have other uses besides mdev with VNC - dropped formatting of the 'display' attribute to the domain XML if it had the default value, so it will only be formatted if user explicitly set it (previously patch 6) - minor adjustments like splitting the second patch into 2 (now 2 and 3) - more cleanup Erik Skultety (11): conf: Remove a redundant model/address-type check in mdev post parse qemu: command: Move graphics iteration to its own function qemu: command: Add virReportEnumRangeError to BuildHostdevCommandline conf: Replace error label with cleanup in virDomainGraphicsDefParseVNCXML qemu: command: Fix building of the SDL display command line conf: Make graphics's GL a standalone structure conf: Allow usage of the <gl> element with VNC graphics conf: Introduce new <gl> attribute 'native' for SPICE qemu: caps: Add vfio-pci.display capability conf: Introduce new <hostdev> attribute 'display' qemu: command: Enable formatting vfio-pci.display option onto cmdline docs/formatdomain.html.in | 33 ++- docs/schemas/domaincommon.rng | 28 ++- src/conf/domain_conf.c | 230 ++++++++++++++------- src/conf/domain_conf.h | 14 +- src/qemu/qemu_capabilities.c | 2 + src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_cgroup.c | 10 +- src/qemu/qemu_command.c | 189 +++++++++++------ src/qemu/qemu_domain.c | 79 ++++++- src/security/security_dac.c | 7 +- tests/qemucapabilitiesdata/caps_2.12.0.aarch64.xml | 1 + tests/qemucapabilitiesdata/caps_2.12.0.ppc64.xml | 1 + tests/qemucapabilitiesdata/caps_2.12.0.s390x.xml | 1 + tests/qemucapabilitiesdata/caps_2.12.0.x86_64.xml | 1 + tests/qemucapabilitiesdata/caps_3.0.0.ppc64.xml | 1 + .../qemuxml2argvdata/graphics-sdl-fullscreen.args | 2 +- tests/qemuxml2argvdata/graphics-sdl.args | 2 +- .../qemuxml2argvdata/graphics-spice-gl-native.args | 26 +++ .../qemuxml2argvdata/graphics-spice-gl-native.xml | 25 +++ .../graphics-spice-gl-non-native.args | 27 +++ .../graphics-spice-gl-non-native.xml | 24 +++ tests/qemuxml2argvdata/graphics-vnc-gl-invalid.xml | 37 ++++ tests/qemuxml2argvdata/graphics-vnc-gl.args | 28 +++ tests/qemuxml2argvdata/graphics-vnc-gl.xml | 37 ++++ .../hostdev-mdev-display-missing-graphics.xml | 35 ++++ .../hostdev-mdev-display-spice-egl-headless.args | 32 +++ .../hostdev-mdev-display-spice-egl-headless.xml | 41 ++++ .../hostdev-mdev-display-spice-opengl.args | 31 +++ .../hostdev-mdev-display-spice-opengl.xml | 41 ++++ .../hostdev-mdev-display-vnc-egl-headless.args | 32 +++ .../hostdev-mdev-display-vnc-egl-headless.xml | 41 ++++ .../qemuxml2argvdata/hostdev-mdev-display-vnc.args | 31 +++ .../qemuxml2argvdata/hostdev-mdev-display-vnc.xml | 39 ++++ tests/qemuxml2argvdata/hostdev-mdev-display.xml | 39 ++++ .../qemuxml2argvdata/video-virtio-gpu-sdl-gl.args | 2 +- tests/qemuxml2argvtest.c | 38 ++++ .../hostdev-mdev-display-active.xml | 47 +++++ .../hostdev-mdev-display-inactive.xml | 47 +++++ .../video-virtio-gpu-spice-gl.xml | 2 +- tests/qemuxml2xmltest.c | 2 + 40 files changed, 1143 insertions(+), 163 deletions(-) create mode 100644 tests/qemuxml2argvdata/graphics-spice-gl-native.args create mode 100644 tests/qemuxml2argvdata/graphics-spice-gl-native.xml create mode 100644 tests/qemuxml2argvdata/graphics-spice-gl-non-native.args create mode 100644 tests/qemuxml2argvdata/graphics-spice-gl-non-native.xml create mode 100644 tests/qemuxml2argvdata/graphics-vnc-gl-invalid.xml create mode 100644 tests/qemuxml2argvdata/graphics-vnc-gl.args create mode 100644 tests/qemuxml2argvdata/graphics-vnc-gl.xml create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-missing-graphics.xml create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-spice-egl-headless.args create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-spice-egl-headless.xml create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-spice-opengl.args create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-spice-opengl.xml create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-vnc-egl-headless.args create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-vnc-egl-headless.xml create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-vnc.args create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display-vnc.xml create mode 100644 tests/qemuxml2argvdata/hostdev-mdev-display.xml create mode 100644 tests/qemuxml2xmloutdata/hostdev-mdev-display-active.xml create mode 100644 tests/qemuxml2xmloutdata/hostdev-mdev-display-inactive.xml -- 2.14.4

4 30

[libvirt] [PATCH] spec: Explicitly require matching libvirt-libs
by Jiri Denemark 09 Jul '18

09 Jul '18

All drivers now link directly to libvirt.so rather than getting the symbols from the daemon. Let's explicitly mention this dependency in the spec file. Theoretically, the automatic libvirt.so.0(LIBVIRT_PRIVATE_4.5.0) dependency should take care of everything, but rpmdiff complains it would like to see an explicit dependency on the exact libvirt-libs version too. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- Notes: The rpmdiff check may also be wrong, but I figured the explicit requires statement makes some sense and can't harm in any case. libvirt.spec.in | 20 ++++++++++++++++++++ 1 file changed, 20 insertions(+) diff --git a/libvirt.spec.in b/libvirt.spec.in index 720870e2b1..6f0d399064 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -544,6 +544,7 @@ Network filter configuration files for cleaning guest traffic %package daemon-driver-network Summary: Network driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Requires: dnsmasq >= 2.41 Requires: radvd Requires: iptables @@ -560,6 +561,7 @@ bridge capabilities. %package daemon-driver-nwfilter Summary: Nwfilter driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Requires: iptables %if 0%{?rhel} && 0%{?rhel} < 7 Requires: iptables-ipv6 @@ -575,6 +577,7 @@ iptables and ip6tables capabilities %package daemon-driver-nodedev Summary: Nodedev driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} # needed for device enumeration %if 0%{?fedora} || 0%{?rhel} >= 7 Requires: systemd >= 185 @@ -591,6 +594,7 @@ capabilities. %package daemon-driver-interface Summary: Interface driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} %if (0%{?fedora} || 0%{?rhel} >= 7) Requires: netcf-libs >= 0.2.2 %endif @@ -604,6 +608,7 @@ netcf library %package daemon-driver-secret Summary: Secret driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} %description daemon-driver-secret The secret driver plugin for the libvirtd daemon, providing @@ -612,6 +617,7 @@ an implementation of the secret key APIs. %package daemon-driver-storage-core Summary: Storage driver plugin including base backends for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Requires: nfs-utils # For mkfs Requires: util-linux @@ -628,6 +634,7 @@ iSCSI, and multipath storage. %package daemon-driver-storage-logical Summary: Storage driver plugin for lvm volumes Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Requires: lvm2 %description daemon-driver-storage-logical @@ -638,6 +645,7 @@ volumes using lvm. %package daemon-driver-storage-disk Summary: Storage driver plugin for disk Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Requires: parted Requires: device-mapper @@ -649,6 +657,7 @@ volumes using the host disks. %package daemon-driver-storage-scsi Summary: Storage driver plugin for local scsi devices Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} %description daemon-driver-storage-scsi The storage driver backend adding implementation of the storage APIs for scsi @@ -658,6 +667,7 @@ host devices. %package daemon-driver-storage-iscsi Summary: Storage driver plugin for iscsi Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Requires: iscsi-initiator-utils %description daemon-driver-storage-iscsi @@ -668,6 +678,7 @@ volumes using the host iscsi stack. %package daemon-driver-storage-mpath Summary: Storage driver plugin for multipath volumes Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Requires: device-mapper %description daemon-driver-storage-mpath @@ -679,6 +690,7 @@ multipath storage using device mapper. %package daemon-driver-storage-gluster Summary: Storage driver plugin for gluster Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} %if 0%{?fedora} Requires: glusterfs-client >= 2.0.1 %endif @@ -696,6 +708,7 @@ volumes using libgfapi. %package daemon-driver-storage-rbd Summary: Storage driver plugin for rbd Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} %description daemon-driver-storage-rbd The storage driver backend adding implementation of the storage APIs for rbd @@ -707,6 +720,7 @@ volumes using the ceph protocol. %package daemon-driver-storage-sheepdog Summary: Storage driver plugin for sheepdog Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Requires: sheepdog %description daemon-driver-storage-sheepdog @@ -719,6 +733,7 @@ sheepdog volumes using. %package daemon-driver-storage-zfs Summary: Storage driver plugin for ZFS Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} # Support any conforming implementation of zfs Requires: /sbin/zfs Requires: /sbin/zpool @@ -760,6 +775,7 @@ parted and more. %package daemon-driver-qemu Summary: QEMU driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} # There really is a hard cross-driver dependency here Requires: libvirt-daemon-driver-network = %{version}-%{release} Requires: libvirt-daemon-driver-storage-core = %{version}-%{release} @@ -784,6 +800,7 @@ QEMU %package daemon-driver-lxc Summary: LXC driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} # There really is a hard cross-driver dependency here Requires: libvirt-daemon-driver-network = %{version}-%{release} %if 0%{?fedora} || 0%{?rhel} > 7 @@ -801,6 +818,7 @@ the Linux kernel %package daemon-driver-uml Summary: Uml driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} %description daemon-driver-uml The UML driver plugin for the libvirtd daemon, providing @@ -813,6 +831,7 @@ User Mode Linux %package daemon-driver-vbox Summary: VirtualBox driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} %description daemon-driver-vbox The vbox driver plugin for the libvirtd daemon, providing @@ -825,6 +844,7 @@ VirtualBox %package daemon-driver-libxl Summary: Libxl driver plugin for the libvirtd daemon Requires: libvirt-daemon = %{version}-%{release} +Requires: libvirt-libs = %{version}-%{release} Obsoletes: libvirt-daemon-driver-xen < 4.3.0 %description daemon-driver-libxl -- 2.18.0

2 1

[libvirt] [PATCH 0/2] Usb and sata for virsh attach-disk --address
by Han Han 09 Jul '18

09 Jul '18

Add usb and sata support for attach-disk of address parameter in virsh. Han Han (2): virsh: usb support for virsh attach-disk --address virsh: sata support for virsh attach-disk --address tools/virsh-domain.c | 82 +++++++++++++++++++++++++++++++++++++++++++- tools/virsh.pod | 2 +- 2 files changed, 82 insertions(+), 2 deletions(-) -- 2.17.1

2 6

[libvirt] [RFC PATCH 0/4] Add new macro to check for existent server
by Marcos Paulo de Souza 09 Jul '18

09 Jul '18

Hello, I am trying to add a generic way to check for server, specially in drivers that needs the server to be declared. If there is better way to put the check, or a better way to check this setting, let me know. Thanks in advance! Marcos Paulo de Souza (4): driver.h: Add macro VIR_DRV_CONN_CHECK_SERVER esx_driver: Use VIR_DRV_CONN_CHECK_SERVER macro hyperv_driver: Use VIR_DRV_CONN_CHECK_SERVER macro phyp_driver: Use VIR_DRV_CONN_CHECK_SERVER macro src/driver.h | 9 +++++++++ src/esx/esx_driver.c | 7 +------ src/hyperv/hyperv_driver.c | 7 +------ src/phyp/phyp_driver.c | 6 +----- 4 files changed, 12 insertions(+), 17 deletions(-) -- 2.17.1

3 7

[libvirt] [PATCH] qemu: Fix typo in qemu_command.c
by Boris Fiuczynski 09 Jul '18

09 Jul '18

Rename qemuBulildFloppyCommandLineOptions into qemuBuildFloppyCommandLineOptions. Signed-off-by: Boris Fiuczynski <fiuczy(a)linux.ibm.com> --- src/qemu/qemu_command.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 32eb59b6ab..ea567262a9 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -2138,10 +2138,10 @@ qemuBuildDriveDevStr(const virDomainDef *def, static int -qemuBulildFloppyCommandLineOptions(virCommandPtr cmd, - const virDomainDef *def, - virDomainDiskDefPtr disk, - unsigned int bootindex) +qemuBuildFloppyCommandLineOptions(virCommandPtr cmd, + const virDomainDef *def, + virDomainDiskDefPtr disk, + unsigned int bootindex) { virBuffer fdc_opts = VIR_BUFFER_INITIALIZER; @@ -2252,8 +2252,8 @@ qemuBuildDiskCommandLine(virCommandPtr cmd, if (qemuDiskBusNeedsDeviceArg(disk->bus)) { if (disk->bus == VIR_DOMAIN_DISK_BUS_FDC) { - if (qemuBulildFloppyCommandLineOptions(cmd, def, disk, - bootindex) < 0) + if (qemuBuildFloppyCommandLineOptions(cmd, def, disk, + bootindex) < 0) return -1; } else { virCommandAddArg(cmd, "-device"); -- 2.17.0

2 1

[libvirt] [PATCH 0/4] Clean up QMP schema testing and add support for RBD auth for -blockdev
by Peter Krempa 09 Jul '18

09 Jul '18

QEMU recently added support for passing RBD credentials also for -blockdev/blockdev-add. First two patches remove the need for a separate stash for the QMP schema data, patch 3 adds new capabilities and finally the last patch adds the RBD auth support. Peter Krempa (4): tests: qemu: Use qmp schema data from the qemucapabilities test tests: remove qemuqapischema.json tests: qemucaps: Add test data for upcoming qemu 3.0.0 qemu: block: Add support for RBD authentication for blockdev src/qemu/qemu_block.c | 26 +- tests/Makefile.am | 1 - .../network-qcow2-backing-chain-cache-unsafe.json | 5 + ...etwork-qcow2-backing-chain-encryption_auth.json | 5 + .../qemucapabilitiesdata/caps_3.0.0.x86_64.replies | 22718 +++++++++++++++++++ tests/qemucapabilitiesdata/caps_3.0.0.x86_64.xml | 1213 + tests/qemucapabilitiestest.c | 1 + tests/qemumonitorjsontest.c | 13 +- tests/qemuqapischema.json | 11364 ---------- tests/testutilsqemuschema.c | 65 +- tests/testutilsqemuschema.h | 3 + 11 files changed, 24040 insertions(+), 11374 deletions(-) create mode 100644 tests/qemucapabilitiesdata/caps_3.0.0.x86_64.replies create mode 100644 tests/qemucapabilitiesdata/caps_3.0.0.x86_64.xml delete mode 100644 tests/qemuqapischema.json -- 2.16.2

3 8

Re: [libvirt] [dbus PATCH] gitignore: add tests/.pytest_cache
by Andrea Bolognani 09 Jul '18

09 Jul '18

[re-adding libvirt-list] On Mon, 2018-07-09 at 09:23 -0400, Anya Harter wrote: > On 07/09/2018 04:34 AM, Andrea Bolognani wrote: > > On Thu, 2018-07-05 at 13:48 -0400, Anya Harter wrote: > > > +/tests/.pytest_cache > > > > When does the directory (or file?) get created? I just ran 'make > > test' but it's not there for me. > > The tests/.pytest_cache/ directory is created when make check is run and > the test suite fails. > > Additionally, a .pytest_cache/ directory is created in the home > directory if ./run tests/<testfile> fails. s/home directory/top-level directory of the project/ ;) > If you want I can give you a patch that breaks the test suite and > creates the directories in the two ways above. No need: I verified the above on my own, and in that light... > > Either way, this looks like it should be a more generic exclude > > pattern like the one we already have for __pycache__ directories > > instead of being limited to tests/. ... this suggestion was spot-on, as adding the pattern you suggested would take care of the failed 'make check' case but not of the failed './run tests/...' case. Looking forward to v2 :) -- Andrea Bolognani / Red Hat / Virtualization

1 0

[libvirt] [PATCH] domain_nwfilter: Return early if net has no name in virDomainConfNWFilterTeardownImpl
by Michal Privoznik 09 Jul '18

09 Jul '18

This function is called from various clean up paths (e.g. from qemuBuildInterfaceCommandLine). However, depending on the stage the interface creation process failed at, net->ifname might still be not filled in when control jumps to cleanup label. If that is the case return early (avoiding useless error message produced in virNWFilterBindingLookupByPortDev) as there is no NWFilter to tear down anyway. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/conf/domain_nwfilter.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/conf/domain_nwfilter.c b/src/conf/domain_nwfilter.c index 24b5f42ddd..f39c8a1f9b 100644 --- a/src/conf/domain_nwfilter.c +++ b/src/conf/domain_nwfilter.c @@ -133,6 +133,9 @@ virDomainConfNWFilterTeardownImpl(virConnectPtr conn, { virNWFilterBindingPtr binding; + if (!net->ifname) + return; + binding = virNWFilterBindingLookupByPortDev(conn, net->ifname); if (!binding) return; -- 2.16.4

2 1

[libvirt] [PATCH v2 0/2] qemu: monitor: Use 'target' instead of 'arch' in reply of 'query-cpus-fast'
by Peter Krempa 09 Jul '18

09 Jul '18

Peter Krempa (2): qemu: monitor: Remove qemuMonitorJSONExtractCPUArchInfo wrapper qemu: monitor: Use 'target' instead of 'arch' in reply of 'query-cpus-fast' src/qemu/qemu_monitor_json.c | 31 +++++++++---------------------- 1 file changed, 9 insertions(+), 22 deletions(-) -- 2.16.2

2 4

[libvirt] [PATCH 0/6] Couple of important fixes
by Michal Privoznik 09 Jul '18

09 Jul '18

I think the most important fix is 2/6 as it might touch live user data. Also, I haven't run 'make check-access' in a while. If I did I would have noticed new paths being accessed. I'll try to update white list (in a separate patch set) and hopefully we can then enable by default and threat its errors as fatal. Michal Prívozník (6): qemuBuildSCSIiSCSIHostdevDrvStr: Don't leak @netsource and @srcprops qemuxml2argvtest: Don't spawn session daemon Forget last daemon/ dir artefacts virtestmock: Track connect() too check-file-access: Allow specifying action virtestmock: Track action Makefile.am | 2 +- run.in | 2 +- src/driver.h | 2 +- src/qemu/qemu_command.c | 6 ++--- tests/check-file-access.pl | 32 ++++++++++++++++++---- tests/file_access_whitelist.txt | 15 +++++++---- tests/libvirtd-fail | 4 +-- tests/qemuxml2argvmock.c | 7 +++++ tests/virtestmock.c | 59 ++++++++++++++++++++++++++++++----------- 9 files changed, 95 insertions(+), 34 deletions(-) -- 2.16.4

3 12

[libvirt] [PATCH] qemu: monitor: Note change in output of query-cpus-fast
by Peter Krempa 09 Jul '18

09 Jul '18

We did not use the field that has changed but note it in the code which extracts the data for future reference. Especially since there is an example of the output. https://bugzilla.redhat.com/show_bug.cgi?id=1598829 Signed-off-by: Peter Krempa <pkrempa(a)redhat.com> --- src/qemu/qemu_monitor_json.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 3e90279b71..41abb920ab 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -1645,6 +1645,9 @@ qemuMonitorJSONExtractCPUArchInfo(virJSONValuePtr jsoncpu, * ...}, * {...} * ] + * + * Note that since QEMU 2.13.0 the "arch" output member of the + * "query-cpus-fast" command is replaced by "target". */ static int qemuMonitorJSONExtractCPUInfo(virJSONValuePtr data, -- 2.16.2

2 2

[libvirt] [PATCH 0/2] Adding reflink support for XFS and others.
by Julio Faracco 09 Jul '18

09 Jul '18

This serie add XFS library to use the macro for reflink (clone). This is the case where BTRFS is not available but XFS is. In case of both is unavailable, the code will use FICLONE from filesystem library if it is defined. This implementation is similar to copy command. If you run a copy and trace the syscalls, you will see that ioctl uses BTRFS_IOC_CLONE or FICLONE to create a reflink. Julio Faracco (2): configure: Adding XFS library/headers check. storage: Rename btrfsCloneFile to support other filesystems. configure.ac | 7 +++++++ src/storage/storage_util.c | 20 ++++++++++++++------ 2 files changed, 21 insertions(+), 6 deletions(-) -- 2.17.1

4 6

[libvirt] [dbus PATCH] gitignore: Add trailing slashes to directory patterns
by Andrea Bolognani 09 Jul '18

09 Jul '18

This makes it explicit that they're meant to ignore entire directories rather than single files. Signed-off-by: Andrea Bolognani <abologna(a)redhat.com> --- .gitignore | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/.gitignore b/.gitignore index ca9e4e1..45a1eb4 100644 --- a/.gitignore +++ b/.gitignore @@ -7,9 +7,9 @@ *Makefile *Makefile.in *~ -.deps -.libs -__pycache__ +.deps/ +.libs/ +__pycache__/ vgcore.* /AUTHORS -- 2.17.1

1 0

[libvirt] [dbus PATCH] gitignore: add tests/.pytest_cache
by Anya Harter 09 Jul '18

09 Jul '18

Signed-off-by: Anya Harter <aharter(a)redhat.com> --- .gitignore | 1 + 1 file changed, 1 insertion(+) diff --git a/.gitignore b/.gitignore index ca9e4e1..c41c82c 100644 --- a/.gitignore +++ b/.gitignore @@ -44,3 +44,4 @@ vgcore.* /src/org.libvirt.service /tests/test_util +/tests/.pytest_cache -- 2.17.1

2 1

[libvirt] [PATCH python] fix build with older libvirt missing virNWFilterBindingPtr
by Daniel P. Berrangé 09 Jul '18

09 Jul '18

While we correctly disable all public API bindings that rely on virNWFilterBindingPtr, the generator is still creating a helper function that needs it - libvirt_virNWFilterBinding_pointer. Making this helper conditional is a little hairy, so just provide a trivial typedef to keep it happy. Signed-off-by: Daniel P. Berrangé <berrange(a)redhat.com> --- typewrappers.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/typewrappers.h b/typewrappers.h index 1570c6a..4423774 100644 --- a/typewrappers.h +++ b/typewrappers.h @@ -27,6 +27,10 @@ typedef ssize_t Py_ssize_t; #endif +#if !LIBVIR_CHECK_VERSION(4, 5, 0) +typedef struct _virNWFilterBinding *virNWFilterBindingPtr; +#endif + #define PyvirConnect_Get(v) (((v) == Py_None) ? NULL : \ (((PyvirConnect_Object *)(v))->obj)) -- 2.17.1

1 0

[libvirt] [PATCH] examples: Remove config.h from hellolibvirt and openauth
by Marcos Paulo de Souza 09 Jul '18

09 Jul '18

These two examples do not use any definition from config.h, and by removing it these examples can be compiled standalone. Signed-off-by: Marcos Paulo de Souza <marcos.souza.org(a)gmail.com> --- When trying to remove config.h from domtop and domsuspend, a later inclusion of unistd.h complains about the missing config.h. This unistd.h come from gnulib, so I need more time to check why this header would complain, so for now just these examples are a good start ... examples/hellolibvirt/hellolibvirt.c | 2 -- examples/openauth/openauth.c | 2 -- 2 files changed, 4 deletions(-) diff --git a/examples/hellolibvirt/hellolibvirt.c b/examples/hellolibvirt/hellolibvirt.c index 02c4401987..bfb12846e6 100644 --- a/examples/hellolibvirt/hellolibvirt.c +++ b/examples/hellolibvirt/hellolibvirt.c @@ -2,8 +2,6 @@ * hypervisor and gather a few bits of information about domains. * Similar API's exist for storage pools, networks, and interfaces. */ -#include <config.h> - #include <stdio.h> #include <stdlib.h> #include <libvirt/libvirt.h> diff --git a/examples/openauth/openauth.c b/examples/openauth/openauth.c index eef46d5f52..efd21c374f 100644 --- a/examples/openauth/openauth.c +++ b/examples/openauth/openauth.c @@ -1,8 +1,6 @@ /* This is a copy of the hellolibvirt example demonstaring how to use * virConnectOpenAuth with a custom auth callback */ -#include <config.h> - #include <stdio.h> #include <stdlib.h> #include <string.h> -- 2.17.1

2 1

[libvirt] [PATCH 0/1] Changed LPC slot from 1 to 31
by Ivan.Mishonov＠redhat.com 09 Jul '18

09 Jul '18

From: Ivan Mishonov <ivan(a)conquex.com> There was no keyboard input on Graphical UEFI guests running Windows 2016. Changed slot number according to the example given in the FreeBSD Handbook section related to bhyve and Graphical UEFI guests Ivan Mishonov (1): Changed LPC slot from 1 to 31 src/bhyve/bhyve_command.c | 2 +- src/bhyve/bhyve_device.c | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) -- 2.18.0

2 2

[libvirt] [PATCH 0/2] Minor fixes for virTypedParams(De)Serialize
by Marc Hartmayer 07 Jul '18

07 Jul '18

Some minor fixes and two questions: * Is the first method, which is described in the documentation for virTypedParamsDeserialize, in sync with the actual code? ("Older APIs do not rely on deserializer allocating memory for @params, ...") * Do we also have to set *nparams = 0 in case of an error and the user has allocated the memory? (virTypedParamsDeserialize) Marc Hartmayer (2): virTypedParamsSerialize: minor fixes virTypedParamsDeserialize: minor fixes src/util/virtypedparam.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) -- 2.13.4

4 15

[libvirt] [RFCv2 PATCH 0/5] Introduce RDT memory bandwidth allocation support
by bing.niu＠intel.com 07 Jul '18

07 Jul '18

From: Bing Niu <bing.niu(a)intel.com> This series is to introduce RDT memory bandwidth allocation support by extending current virresctrl implementation. The Memory Bandwidth Allocation (MBA) feature provides indirect and approximate control over memory bandwidth available per-core. This feature provides a method to control applications which may be over-utilizing bandwidth relative to their priority in environments such as the data-center. The details can be found in Intel's SDM 17.19.7. Kernel supports MBA through resctrl file system same as CAT. Each resctrl group have a MB parameter to control how much memory bandwidth it can utilize in unit of percentage. In this series, MBA is enabled by enhancing existing virresctrl implementation. The policy employed for MBA is similar with CAT: The sum of each MBA group's bandwidth dose not exceed 100%. The enhancement of virresctrl include two parts: Part 1: Add two new structures virResctrlInfoMB and virResctrlAllocMB for collecting host system MBA capability and domain memory bandwidth allocation. Those two structures are the extension of existing virResctrlInfo and virResctrlAlloc. With them, virresctrl framework can support MBA and CAT concurrently. Each virResctrlAlloc represent a resource allocation including CAT, or MBA, or CAT&MBA. The policy of MBA is that: total memory bandwidth of each resctrl group, created by virresctrl, does not exceed to 100%. Part 2: On XML part, add new elements to host capabilities query and domain allocation to support memory bandwidth allocation. ---------------------------------------------------------------------------------------------- For host capabilities XML, new XML format like below example, <host> ..... <memory> <node id='0' cpus='0-19'> <control granularity='10' min ='10' maxAllocs='8'/> </node> </memory> </host> granularity --- memory bandwidth granularity min --- minimum memory bandwidth allowed maxAllocs --- maximum concurrent memory bandwidth allocation allowed. ---------------------------------------------------------------------------------------------- For domain XML, new format as below example <domain type='kvm' id='2'> ...... <cputune> ...... <shares>1024</shares> <memorytune vcpus='0-1'> <node id='0' bandwidth='20'/> </memorytune> </cputune> ...... </domain> id --- node where memory bandwidth allocation will happen bandwidth --- bandwidth allocated in percentage ---------------------------------------------------------------------------------------------- Changelog: v1->v2: Jano's comment: 1. put renaming parts into separated patches. 2. set the initial return value as -1. 3. using full name in structure definition. 4. do not use VIR_CACHE_TYPE_LAST for memory bandwidth allocation formatting. Pavel's comment: 1. add host capabilities XML for memory bandwidth allocation. 2. do not mix use cachetune section in XML for memory bandwidth allocation in domain XML. define a dedicated one for memory bandwidth allocation. Bing Niu (5): util: Rename and packing parts of virresctrl util: Add memory bandwidth support to resctrl conf: rename cachetune to restune conf: Introduce cputune/memorytune to support memory bandwidth allocation conf: add memory bandwidth allocation capability of host src/conf/capabilities.c | 112 +++++++++++ src/conf/capabilities.h | 11 ++ src/conf/domain_conf.c | 304 ++++++++++++++++++++++++++--- src/conf/domain_conf.h | 10 +- src/libvirt_private.syms | 4 +- src/qemu/qemu_process.c | 18 +- src/util/virresctrl.c | 496 ++++++++++++++++++++++++++++++++++++++++++++--- src/util/virresctrl.h | 35 +++- 8 files changed, 918 insertions(+), 72 deletions(-) -- 2.7.4

3 16