July 2016 - Devel - libvirt List Archives

[libvirt] [PATCH v4 0/4] perf: add more perf events support
by Qiaowei Ren 04 Aug '16

04 Aug '16

With current perf framework, this patchset refactor virPerfEventEnable() for general purpose and add more perf events support based on this change, including cache misses, cache references, cpu cycles, instrctions, etc.. Changes since v3: * separate the patch into 4 patches. * update virsh.pod for new perf events. * introduce a static attr table that would be able to convert the VIR_PERF_EVENT_* into their respective "attr.type" and "attr.config". Qiaowei Ren (4): perf: rename qemuDomainGetStatsPerfRdt() perf: introduce a static attr table and refactor virPerfEventEnable() for general purpose perf: add more perf events support perf: add description for new events in virsh.pod docs/formatdomain.html.in | 24 ++++ docs/schemas/domaincommon.rng | 4 + include/libvirt/libvirt-domain.h | 39 ++++++ src/libvirt-domain.c | 8 ++ src/qemu/qemu_driver.c | 23 ++-- src/util/virperf.c | 176 +++++++++++++++++----------- src/util/virperf.h | 11 ++ tests/genericxml2xmlindata/generic-perf.xml | 4 + tools/virsh.pod | 4 + 9 files changed, 211 insertions(+), 82 deletions(-) -- 1.9.1

3 9

[libvirt] [PATCH 0/2] Add ability to set specific IOThread scheduler quota/period
by John Ferlan 02 Aug '16

02 Aug '16

https://bugzilla.redhat.com/show_bug.cgi?id=1356937 John Ferlan (2): conf: Add IOThread quota and period scheduler/cputune defs qemu: Add support to get/set IOThread period and quota cgroup values docs/formatdomain.html.in | 25 ++++ docs/schemas/domaincommon.rng | 10 ++ include/libvirt/libvirt-domain.h | 32 ++++++ src/conf/domain_conf.c | 42 +++++++ src/conf/domain_conf.h | 2 + src/qemu/qemu_command.c | 3 +- src/qemu/qemu_driver.c | 127 ++++++++++++++++++++- src/qemu/qemu_process.c | 4 +- tests/qemuxml2argvdata/qemuxml2argv-cputune.xml | 2 + .../qemuxml2xmloutdata/qemuxml2xmlout-cputune.xml | 2 + tools/virsh.pod | 9 +- 11 files changed, 249 insertions(+), 9 deletions(-) -- 2.5.5

2 4

[libvirt] [PATCH python 0/2] Bindings for node device lifecycle events
by Jovanka Gulicoska 02 Aug '16

02 Aug '16

Python buindings and tests for node device lifecycle events API. Jovanka Gulicoska (2): Python binding for node poll lifecycle events API event-test: Add node device lifecycle event tests examples/event-test.py | 16 +++++ generator.py | 2 + libvirt-override-virConnect.py | 45 ++++++++++++ libvirt-override.c | 153 +++++++++++++++++++++++++++++++++++++++++ sanitytest.py | 3 + 5 files changed, 219 insertions(+) -- 2.7.4

3 5

[libvirt] [PATCH v2 0/9] Various hotplug cleanups
by John Ferlan 02 Aug '16

02 Aug '16

v1: http://www.redhat.com/archives/libvir-list/2016-June/msg02207.html Changes since v1... - Use the bz noted by Cole from the v1 series - Merged in recent hotplug cleanup changes - Patches 6 - 9 are new Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1336225 and probably: https://bugzilla.redhat.com/show_bug.cgi?id=1289391 Patches 1-5 address the USB hotplug cleanup Patch 6 addresses the SCSI hotplug cleanup Patch 7-9 address some SCSI host device cleanups found while looking at the other cleanups... John Ferlan (9): qemu: Reorder qemuDomainAttachUSBMassStorageDevice failure path qemu: Move drive alias processing to qemu_alias qemu: Use qemuAssignDeviceDiskDriveAlias qemu: Make QEMU_DRIVE_HOST_PREFIX more private qemu: Add attempt to call qemuMonitorDriveDel for USB failure path qemu: Add attempt to call qemuMonitorDriveDel for AttachSCSI failure path qemu: Introduce qemuAssignSCSIHostDeviceDriveAlias qemu: Use qemuAssignSCSIHostDeviceDriveAlias qemu: Use the hostdev alias in qemuDomainAttachHostSCSIDevice error path src/qemu/qemu_alias.c | 52 +++++++++++++++++++++++++ src/qemu/qemu_alias.h | 6 +++ src/qemu/qemu_command.c | 35 ++++++++--------- src/qemu/qemu_command.h | 1 - src/qemu/qemu_domain.c | 3 +- src/qemu/qemu_driver.c | 3 +- src/qemu/qemu_hotplug.c | 92 +++++++++++++++++++++++++++++--------------- src/qemu/qemu_migration.c | 11 ++---- src/qemu/qemu_monitor_json.c | 10 ++--- src/qemu/qemu_monitor_text.c | 18 +++++---- src/qemu/qemu_process.c | 3 +- 11 files changed, 156 insertions(+), 78 deletions(-) -- 2.5.5

2 24

[libvirt] [PATCH v2 0/3] libxl hooks
by Cédric Bosdonnat 02 Aug '16

02 Aug '16

Hey there! Diffs to v1: * move the ret = 0 to the proper patch * fix typos as pointed by Joao * use ignore_value where needed Cédric Bosdonnat (3): libxl: add a flag to mark guests as tainted by a hook libxl: fix segfault in libxlReconnectDomain libxl: add hooks support docs/hooks.html.in | 53 ++++++++++++++++++++++++++-- src/libxl/libxl_domain.c | 84 +++++++++++++++++++++++++++++++++++++++++++++ src/libxl/libxl_domain.h | 2 ++ src/libxl/libxl_driver.c | 52 +++++++++++++++++++++------- src/libxl/libxl_migration.c | 57 ++++++++++++++++++++++++++++++ src/util/virhook.c | 16 ++++++++- src/util/virhook.h | 13 +++++++ 7 files changed, 262 insertions(+), 15 deletions(-) -- 2.6.6

5 9

[libvirt] [PATCH v2] qemu: set fake reboot flag only in acpi mode
by Nikolay Shirokovskiy 02 Aug '16

02 Aug '16

First this fixes the same issue as e2b86f580. Only difference is that reboot is done via shutdown function with reboot semantics. Second 8be502fd tells us that we need to always set fake reboot flag if shutdown/reboot will trigger shutdown event. To put it simple we need to set it false if agent is used. References: e2b86f580 - fixes guest shutdown becames reboot after API reboot thru agent 8be502fd - fixes API shutdown thru agent becames reboot after API reboot thru acpi (more on https://www.redhat.com/archives/libvir-list/2015-April/msg00715.html) --- src/qemu/qemu_driver.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 2089359..8855387 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -1997,9 +1997,9 @@ static int qemuDomainShutdownFlags(virDomainPtr dom, unsigned int flags) useAgent = false; } - qemuDomainSetFakeReboot(driver, vm, isReboot); if (useAgent) { + qemuDomainSetFakeReboot(driver, vm, false); qemuDomainObjEnterAgent(vm); ret = qemuAgentShutdown(priv->agent, agentFlag); qemuDomainObjExitAgent(vm); @@ -2018,6 +2018,7 @@ static int qemuDomainShutdownFlags(virDomainPtr dom, unsigned int flags) goto endjob; } + qemuDomainSetFakeReboot(driver, vm, isReboot); qemuDomainObjEnterMonitor(driver, vm); ret = qemuMonitorSystemPowerdown(priv->mon); if (qemuDomainObjExitMonitor(driver, vm) < 0) @@ -2091,6 +2092,7 @@ qemuDomainReboot(virDomainPtr dom, unsigned int flags) } if (useAgent) { + qemuDomainSetFakeReboot(driver, vm, false); qemuDomainObjEnterAgent(vm); ret = qemuAgentShutdown(priv->agent, agentFlag); qemuDomainObjExitAgent(vm); -- 1.8.3.1

2 2

[libvirt] [PATCH V2 0/7] libxl: add support of pvusb controller
by Chunyan Liu 02 Aug '16

02 Aug '16

This patch series is to add pvusb controller support in libxl driver. It should be applied on previous pvusb device support patch series. --- Changes: * drop pvusb1 and pvusb2 model * add check in qemu device post-parse to report error of unsupported 'qusb1' and 'qusb2' model Chunyan Liu (7): extend usb controller model to support xen pvusb libxl: support USB controllers in creation time libxl: support usb controller hotplug libxl: check available controller and port when hotplugging USB device xenconfig: add conversion of usb controller config to and from xml xlconfigtest: add test for usb controller conversion qemuDomainDeviceDefPostParse: add USB controller model check docs/formatdomain.html.in | 4 +- docs/schemas/domaincommon.rng | 2 + src/conf/domain_conf.c | 2 + src/conf/domain_conf.h | 2 + src/libxl/libxl_conf.c | 84 ++++++++++++++++ src/libxl/libxl_conf.h | 4 + src/libxl/libxl_driver.c | 176 +++++++++++++++++++++++++++++++++ src/qemu/qemu_command.c | 2 + src/qemu/qemu_domain.c | 13 +++ src/xenconfig/xen_xl.c | 190 ++++++++++++++++++++++++++++++++++++ tests/xlconfigdata/test-usbctrl.cfg | 13 +++ tests/xlconfigdata/test-usbctrl.xml | 31 ++++++ tests/xlconfigtest.c | 1 + 13 files changed, 523 insertions(+), 1 deletion(-) create mode 100644 tests/xlconfigdata/test-usbctrl.cfg create mode 100644 tests/xlconfigdata/test-usbctrl.xml -- 2.1.4

3 15

[libvirt] [PATCH 0/5] admin: Fix updating of the client limits
by Erik Skultety 02 Aug '16

02 Aug '16

This series slightly refactors the code in virnetserver.c in order to resolve https://bugzilla.redhat.com/show_bug.cgi?id=1357776. The problem resides in updating of the client limits which even though updated, thus "permitting" new connections to be established, do not re-enable polling for the socket file descriptor which results in all new connections still queueing on the socket. Erik Skultety (5): rpc: virnetserver: Rename ClientSetProcessingControls to ClientSetLimits rpc: virnetserver: Move virNetServerCheckLimits which is static up in the file rpc: virnetserver: Add code to CheckLimits to handle suspending of services admin: rpc: virnetserver: Fix updating of the client limits rpc: virnetserver: Remove dead code checking the client limits daemon/admin_server.c | 4 +-- src/rpc/virnetserver.c | 97 +++++++++++++++++++++++--------------------------- src/rpc/virnetserver.h | 6 ++-- 3 files changed, 50 insertions(+), 57 deletions(-) -- 2.5.5

2 8

[libvirt] [PATCH 0/2] virpci: support driver_override sysfs interface
by Jim Fehlig 02 Aug '16

02 Aug '16

This series fixes a regression using Xen's pciback with kernels >= 3.16 containing https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/driv… It makes use of the driver_override sysfs interface introduced in kernel 3.12 https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/driv… This approach was suggested by Laine, falling back to using the new_id interface in the absence of driver_override https://www.redhat.com/archives/libvir-list/2016-June/msg02129.html The first patch simplifies the existing logic in virPCIDeviceBindToStub, making the second patch (which actually fixes the bug) a bit easier to code (and hopefully review). Jim Fehlig (2): virpci: simplify virPCIDeviceBindToStub virpci: support driver_override sysfs interface src/util/virpci.c | 253 +++++++++++++++++++++++++++++++++++++----------------- 1 file changed, 173 insertions(+), 80 deletions(-) -- 2.1.4

2 7

[libvirt] Question about LSN-2016-0001
by Jim Fehlig 01 Aug '16

01 Aug '16

I've noticed the behavior described by this LSN with libvirt+Xen. Config containing <graphics type='vnc' passwd=''/> allows any client to connect with no authentication check. I asked about this on the Xen security list and was told that "libxl interprets an empty password in the caller's configuration to mean that passwordless access should be permitted". The libvirt domXML docs are not clear on semantics of empty vnc password, only stating "The passwd attribute provides a VNC password in clear text". Should the libvirt domXML vnc passwd documentation be amended to define the semantics of an empty string in the passwd attribute? Is the behavior hypervisor-dependent as the documentation in qemu.conf suggests? Regards, Jim

2 4