March 2014 - Devel - libvirt List Archives

[libvirt] [PATCH] Fix missing char dev lock path case in configure
by Roman Bogorodskiy 05 Mar '14

05 Mar '14

configure check for character devices lock path calls AC_DEFINE_UNQUOTED for VIR_CHRDEV_LOCK_FILE_PATH even if $with_chrdev_lock_files = "no". So the locking code in conf/virchrdev.c: #ifdef VIR_CHRDEV_LOCK_FILE_PATH is compiled in even if it shouldn't, because VIR_CHRDEV_LOCK_FILE_PATH is defined as "no", so it tries to create lock files with strange lock path like 'no/LCK..'. Fix that by calling AC_DEFINE_UNQUOTED only if $with_chrdev_lock_files is not 'no'. --- configure.ac | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/configure.ac b/configure.ac index a9339ce..62b74c5 100644 --- a/configure.ac +++ b/configure.ac @@ -1401,8 +1401,10 @@ if test "$with_chrdev_lock_files" != "no"; then AC_MSG_ERROR([You must specify path for the lock files on this platform]) fi - AC_DEFINE_UNQUOTED([VIR_CHRDEV_LOCK_FILE_PATH], "$with_chrdev_lock_files", - [path to directory containing UUCP device lock files]) + if test "$with_chrdev_lock_files" != "no"; then + AC_DEFINE_UNQUOTED([VIR_CHRDEV_LOCK_FILE_PATH], "$with_chrdev_lock_files", + [path to directory containing UUCP device lock files]) + fi fi AM_CONDITIONAL([VIR_CHRDEV_LOCK_FILE_PATH], [test "$with_chrdev_lock_files" != "no"]) -- 1.8.4.2

2 2

[libvirt] [PATCH 00/13] libxl: add basic support for migration
by Jim Fehlig 05 Mar '14

05 Mar '14

Based on an earlier patch from Chunyan Liu https://www.redhat.com/archives/libvir-list/2013-September/msg00667.html This patch series adds basic migration support to the libxl driver. Follow-up patches can improve pre-migration checks and add support for additional migration flags. Patches 1-12 are almost exclusively code motion, moving functions from the main driver module into the libxl_domain and libxl_conf modules. Patch13 is rather large and provides the migration implementation. If it is preferred to split patch13 further, I'm open to suggestions on ways to do that. I've done some basic migration testing of this series using NFS shared storage. Error handling only minimally tested though. I attempted a migration to host with a different bridge than the source, verifying the migration failed and the vm was resumed on the source. Jim Fehlig (13): libxl: move libxlDomainEventQueue to libxl_domain libxl: move libxlDomainManagedSavePath to libxl_domain libxl: move libxlSaveImageOpen to libxl_domain libxl: move libxlVmCleanup{,Job} to libxl_domain libxl: move libxlDomEventsRegister to libxl_domain libxl: move libxlDomainAutoCoreDump to libxl_domain libxl: move libxlDoNodeGetInfo to libxl_conf libxl: move libxlDomainSetVcpuAffinities to libxl_domain libxl: move libxlFreeMem to libxl_domain libxl: move libxlVmStart to libxl_domain libxl: include a pointer to the driver in libxlDomainObjPrivate libxl: move domain event handler to libxl_domain libxl: add migration support po/POTFILES.in | 1 + src/Makefile.am | 3 +- src/libxl/libxl_conf.c | 36 ++ src/libxl/libxl_conf.h | 10 + src/libxl/libxl_domain.c | 696 +++++++++++++++++++++++++++++++ src/libxl/libxl_domain.h | 51 ++- src/libxl/libxl_driver.c | 972 +++++++++++--------------------------------- src/libxl/libxl_migration.c | 579 ++++++++++++++++++++++++++ src/libxl/libxl_migration.h | 78 ++++ 9 files changed, 1681 insertions(+), 745 deletions(-) create mode 100644 src/libxl/libxl_migration.c create mode 100644 src/libxl/libxl_migration.h -- 1.8.1.4

1 14

[libvirt] [PATCH] qemu: monitor: Provide more information in generic block job error
by Peter Krempa 05 Mar '14

05 Mar '14

The qemuMonitorJSONBlockJob handles a few errors internally. If qemu returns a different error we would report a rather unhelpful message: $ virsh blockpull gluster-job vda --base /dev/null error: internal error: Unexpected error As the actual message from qemu contains a bit more info, let's use it to report something a little more useful: $ virsh blockpull gluster-job vda --base /dev/null error: internal error: Unexpected error: (GenericError) 'Base '/dev/null' not found' --- src/qemu/qemu_monitor_json.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 9a5b812..ee3ae15 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -3678,8 +3678,12 @@ qemuMonitorJSONBlockJob(qemuMonitorPtr mon, virReportError(VIR_ERR_OPERATION_INVALID, _("Command '%s' is not found"), cmd_name); } else { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("Unexpected error")); + virJSONValuePtr error = virJSONValueObjectGet(reply, "error"); + + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Unexpected error: (%s) '%s'"), + NULLSTR(virJSONValueObjectGetString(error, "class")), + NULLSTR(virJSONValueObjectGetString(error, "desc"))); } } -- 1.9.0

2 2

[libvirt] [PATCHv8 0/2] Implement RBD storage pool support
by Adam Walters 05 Mar '14

05 Mar '14

This is an updated version of my RBD storage pool support patch. Relevant changes since v7 are: - Fixed bug where I compared disk->srcpool->mode to VIR_DOMAIN_DISK_TYPE_NETWORK In response to a couple of items Ján mentioned, I decided not to pursue adding anything to qemuxml2argvtest. The reason I decided not to pursue this is there is not really any need. In comparison with other volume types, RBD already has the same test coverage. Namely, a base set of test-cases without using storage pools, plus a generic storage pool test. In addition, I also attempted to implement a proper check for expected vs actual secret usage, but at that particular point in the code, Ján was correct that the pool definition has not yet been parsed. Thus, I had to keep my bypass. Adam Walters (2): qemu: conf Implement domain RBD storage pool support domain: conf: Fix secret type checking for network volumes src/conf/domain_conf.c | 9 ++++++-- src/qemu/qemu_conf.c | 62 +++++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 68 insertions(+), 3 deletions(-) -- 1.8.5.3

2 4

[libvirt] [PATCH v13 00/49] write separate module for hostdev passthrough
by Chunyan Liu 05 Mar '14

05 Mar '14

These patches implements a separate module for hostdev passthrough so that it could be shared by different drivers and can maintain a global state of a host device. Patches 1~6 are to switch existing qemu and lxc driver to use common library lists, so that to maintain a global state of every host device. Patches 7~45 are to extract general code from qemu_hostdev.c piece by piece, make them reusable common APIs. Patches 46: unit test for the virhostdev common library Patches 47: add a hostdev backend type for xen Patches 48: add pci passthrough to libxl driver based on the common library Patches 49: change lxc to use common library APIs --- changes to v12: * split "add hostdev passthrough common library" patch into small patches for easier review. * fix v12 comments * rebase to libxl changes Chunyan Liu (49): add 'driver' info to used_by qemu: reuse hostdev interfaces to avoid duplicate qemu: remove functions now used internally only from qemu_hostdev.h add virhostdev files to maintain global state of host devices qemu: use general virhostdev lists instead of its own lxc: use general virhostdev lists instead of its own qemu_hostdev: move cfg->relaxedACS as a flag qemu_hostdev: move ColdBoot as a flag qemu_hostdev: move netconfig file location to virhostdev stateDir extract general code from qemuPrepareHostdevPCIDevices rename qemu*NetConfigRestore/Replace to virHostdevNetConfigRestore/Replace rename qemuGet*PciHostDeviceList to virHostdevGet*PciHostDeviceList pass driver name as a parameter to virHostdevPrepareHostdevPCIDevices extract general code from qemuDomainReAttachHostdevDevices pass driver name as a parameter to virHostdevReAttachPCIDevices rename qemuReAttachPciDevice to virHostdevReAttachPciDevice move virHostdevPrepare(ReAttach)PCIDevices to virhostdev.c extract general code from qemuUpdateActivePciHostdevs extract general code from qemuUpdateActiveUsbHostdevs extract general code from qemuUpdateActiveScsiHostdevs pass driver_name as parameter of virHostdevUpdate*Hostdevs functions move virHostdevUpdate* functions to virhostdev.c qemuPrepareUSBDevices: code adjustment for extracting general code extract general code from qemuPrepareHostUSBDevices rename qemu*USBDevices to virHostdev*USBDevices pass driver name to virHostdevPrepareUSBDevices move virHostdevPrepareHostUSBDevices to virhostdev.c extract general code from qemuPrepareHostSCSIDevices pass driver name as parameter to virHostdevPrepareSCSIDevices move virHostdevPrepareHostSCSIDevices to virhostdev.c extract general code from qemuDomainReAttachHostUsbDevices pass driver name as paramter to virHostdevReAttachUsbHostdevs move virHostdevDomainReAttachHostUsbDevices to virhostdev.c extract general code from qemuDomainReAttachHostScsiDevices pass driver name as parameter to virHostdevReAttachScciHostdevs move virHostdevReAttachHostScsiDevices to virhostdev.c extract general code of NodeDeviceDetach extract general code of NodeDeviceReAttach extract general code of NodeDeviceReset move virHostdevNodeDevice* to virhostdev.c improve parameter name to let it more meaningful rename some function names to keep consistency improve virHostdevUpdate* parameters to make it more widely used add 3 wrapper functions for prepare/reattach/update domain hostdevs add parameter checks to common interfaces add unit test for new virhostdev common library change lxc_hostdev.c to use virhostdev common library APIs add hostdev pci backend type for xen add pci passthrough to libxl driver .gitignore | 1 + docs/schemas/domaincommon.rng | 1 + po/POTFILES.in | 1 + src/Makefile.am | 1 + src/conf/domain_conf.c | 3 +- src/conf/domain_conf.h | 1 + src/libvirt_private.syms | 19 + src/libxl/libxl_conf.c | 63 ++ src/libxl/libxl_conf.h | 4 + src/libxl/libxl_domain.c | 9 + src/libxl/libxl_driver.c | 447 +++++++++++- src/lxc/lxc_conf.h | 4 - src/lxc/lxc_driver.c | 17 +- src/lxc/lxc_hostdev.c | 315 +-------- src/qemu/qemu_command.c | 3 +- src/qemu/qemu_conf.h | 10 +- src/qemu/qemu_driver.c | 88 +-- src/qemu/qemu_hostdev.c | 1220 ++----------------------------- src/qemu/qemu_hostdev.h | 27 +- src/qemu/qemu_hotplug.c | 77 +-- src/qemu/qemu_process.c | 8 +- src/util/virhostdev.c | 1621 +++++++++++++++++++++++++++++++++++++++++ src/util/virhostdev.h | 140 ++++ src/util/virpci.c | 31 +- src/util/virpci.h | 9 +- src/util/virscsi.c | 32 +- src/util/virscsi.h | 7 +- src/util/virusb.c | 31 +- src/util/virusb.h | 8 +- tests/Makefile.am | 5 + tests/virhostdevtest.c | 507 +++++++++++++ tests/virscsitest.c | 6 +- 32 files changed, 3081 insertions(+), 1635 deletions(-) create mode 100644 src/util/virhostdev.c create mode 100644 src/util/virhostdev.h create mode 100644 tests/virhostdevtest.c

5 69

[libvirt] [PATCH v4 0/3] support dumping guest memory in compressed format
by qiaonuohan 05 Mar '14

05 Mar '14

dumping guest's memroy is introduced without compression supported, and this is a freature regression of 'virsh dump --memory-only'. This patchset is used to add support in libvirt side to make qemu dump guest's memory in kdump-compressed format and please refer the following address to see implementation of the qemu side, the lastest version of qemu side is v9(ready for being queued). http://lists.nongnu.org/archive/html/qemu-devel/2014-02/msg03016.html ChangLog: Changes from v3 to v4: 1. dropping patch "add dump_memory_format in qemu.conf" 2. fix to follow some conventions Changes from v2 to v3: 1. address Jiri Denemark's comment about adding a new public API instead of changing an old one. Changes from v1 to v2: 1. address Daniel P. Berrange's comment about using a new parameter to replace flags like VIR_DUMP_COMPRESS_ZLIB. qiaonuohan (3): add new virDomainCoreDumpWithFormat API add qemu support to virDomainCoreDumpWithFormat API allow "virsh dump --memory-only" specify dump format include/libvirt/libvirt.h.in | 31 ++++++++++++++ src/driver.h | 7 +++ src/libvirt.c | 100 +++++++++++++++++++++++++++++++++++++++++++ src/libvirt_public.syms | 5 +++ src/qemu/qemu_driver.c | 45 +++++++++++++++---- src/qemu/qemu_monitor.c | 7 +-- src/qemu/qemu_monitor.h | 3 +- src/qemu/qemu_monitor_json.c | 4 +- src/qemu/qemu_monitor_json.h | 3 +- src/remote/remote_driver.c | 1 + src/remote/remote_protocol.x | 15 ++++++- src/remote_protocol-structs | 7 +++ src/test/test_driver.c | 19 ++++++-- tests/qemumonitorjsontest.c | 2 +- tools/virsh-domain.c | 45 ++++++++++++++++++- 15 files changed, 273 insertions(+), 21 deletions(-) -- 1.8.5.3

3 12

[libvirt] [PATCHv2 0/7] various fixes related to gluster volumes
by Peter Krempa 05 Mar '14

05 Mar '14

Peter Krempa (7): virsh: volume: Fix lookup of volumes to provide better error messages storage: Use cleanup label instead of out storage: Error out when attempting to vol-upload into a remote pool storage: Avoid mangling paths of non-local filesystems when looking up storage: Don't lie about path used to look up in error message doc: storage: Explicitly state that it's possible to have non-unique key gluster: Fix "key" attribute for gluster volumes docs/formatstorage.html.in | 6 +- src/storage/storage_backend_gluster.c | 10 +- src/storage/storage_driver.c | 215 +++++++++++++++++++++------------- tools/virsh-volume.c | 12 +- 4 files changed, 154 insertions(+), 89 deletions(-) -- 1.9.0

2 15

[libvirt] libvirt xml can't specify USB physical port when passthrough USB device to VM
by Wangrui (K) 05 Mar '14

05 Mar '14

Hello, I passthrough a host usb device to a guest with a bus number and device number in the xml to identify the usb device. When I unplug and then plug back into the same USB physical port, the guest identifies the usb device failed. Because on the host ,the usb device number increases. So , I want to know that whether Libvirt will provide an interface that identifies a host usb device with bus number and physical port . As far as I know ,Qemu upstream had added a hostport property which allows to specify the host usb with bus number and physical port. the patch is : https://lists.gnu.org/archive/html/qemu-devel/2011-05/msg02341.html Best Regards

1 0

[libvirt] [Question] Will we support to set boot order online?
by Wangyufei (James) 05 Mar '14

05 Mar '14

Hello, I find that QEMU supports to set boot(QMP command "boot_set") order online, namely the virtual machine can be reboot in appointed order, but libvirt had no corresponding interface. So, I want to know if libvirt intend to support to set boot order online. If the answer is yes, I'll try to apply the patch. If no, why? Best Regards, -WangYufei

3 4

[libvirt] [PATCH] qemu: add PCI-multibus support for ppc
by Olivia Yin 05 Mar '14

05 Mar '14

Signed-off-by: Olivia Yin <hong-hua.yin(a)freescale.com> --- src/qemu/qemu_capabilities.c | 10 ++++++++++ 1 files changed, 10 insertions(+), 0 deletions(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 7bc1ebc..7d7791d 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -2209,6 +2209,11 @@ virQEMUCapsInitHelp(virQEMUCapsPtr qemuCaps, uid_t runUid, gid_t runGid) virQEMUCapsClear(qemuCaps, QEMU_CAPS_NO_ACPI); } + /* ppc support PCI-multibus */ + if (qemuCaps->arch == VIR_ARCH_PPC) { + virQEMUCapsSet(qemuCaps, QEMU_CAPS_PCI_MULTIBUS); + } + /* virQEMUCapsExtractDeviceStr will only set additional caps if qemu * understands the 0.13.0+ notion of "-device driver,". */ if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_DEVICE) && @@ -2450,6 +2455,11 @@ virQEMUCapsInitQMP(virQEMUCapsPtr qemuCaps, virQEMUCapsSet(qemuCaps, QEMU_CAPS_NO_ACPI); } + /* ppc support PCI-multibus */ + if (qemuCaps->arch == VIR_ARCH_PPC) { + virQEMUCapsSet(qemuCaps, QEMU_CAPS_PCI_MULTIBUS); + } + if (virQEMUCapsProbeQMPCommands(qemuCaps, mon) < 0) goto cleanup; if (virQEMUCapsProbeQMPEvents(qemuCaps, mon) < 0) -- 1.6.4

5 7