- Devel - libvirt List Archives

[libvirt] [libvirt-users] Release of libvirt-0.9.2
by Daniel Veillard 06 Jun '11

06 Jun '11

As planned the new release is available at: ftp://libvirt.org/libvirt/ It is a rather large release with near 400 commits included. From an user point of view the main improvement is likely to be when using migration as various work has been done to extend the protocol and to avoid having the migration command stop other concurrent operations (like virsh list). See below for a number of new features, improvement and bug fixes: Features: - Framework for lock manager plugins (Daniel P. Berrange) - API for network config change transactions (Michal Privoznik) - flags for setting memory parameters (Hu Tao) - virDomainGetState public API (Jiri Denemark) - qemu: allow blkstat/blkinfo calls during migration (Federico Simoncelli) - Introduce migration v3 API (Daniel P. Berrange) - Defining the Screenshot public API (Michal Privoznik) - public API for NMI injection (Lai Jiangshan) Documentation: - esx: Fix driver method version annotations (Matthias Bolte) - vbox: Fix typo in error message (Matthias Bolte) - virsh: Document nodeinfo output (Jiri Denemark) - Correct 'cputune' documentation example. (Neil Wilson) - document iface-* commands (Eric Blake) - Add initial docs about the lock managers (Daniel P. Berrange) - maint: use consistent file name for threading notes (Eric Blake) - Fix virTypedParameter alias comments (Matthias Bolte) - Fix error syntax of vcpupin example XML (Osier Yang) - Automatically generate the hvsupport.html.in file from source files (Daniel P. Berrange) - Fix documentation for cputune parameters (Osier Yang) - avoid double 'the' (Eric Blake) - Add documentation about test suites (Daniel Veillard) - updates to CA cert and client cert/key info (Doug Goldstein) - node filesystem attr is 'accessmode', not 'mode' (Cole Robinson) Portability: - vbox: Fix version extraction on Windows for newer VirtualBox versions (Matthias Bolte) - Make dlopen usage in lock manager conditional (Matthias Bolte) - build: fix VPATH build break from previous patch (Eric Blake) - Remove call to deprecated gnutls_certificate_type_set_priority (Daniel P. Berrange) - Fix sysinfo/virsh build problems on Win32 (Daniel P. Berrange) - Fix build with --with-driver-modules enabled (Matthias Bolte) - build: fix 'make dist' for ./configure --without-remote (Eric Blake) - virsh: time_t is not a long on FreeBSD (Matthias Bolte) - time_t is not a long on FreeBSD, switch internal type to long long (Matthias Bolte) - build: avoid strerror_r pitfalls (Eric Blake) - build: fix VIR_DEBUG on mingw (Eric Blake) - build: generate files when building without libvirtd (Wen Congyang) - build: drop unused sys/syslimits.h header (Eric Blake) - Fix prototype of virRun for Win32 targets (Daniel P. Berrange) - build: tolerate unlimited group size (Eric Blake) - Disable virCommandExec on Win32 (Daniel P. Berrange) - command: Fix compilation on FreeBSD (Matthias Bolte) - apparmor: Fix compilation by removing remains from virCommand conversion (Matthias Bolte) - configure: Fix mpath check on non-Linux systems (Matthias Bolte) - fix xdr detection and use with recent glibc (Christophe Fergeau) - build: avoid gcc preprocessor extensions (Eric Blake) - build: fix VPATH build with distributed generated files (Eric Blake) - build: fix VPATH builds (Eric Blake) Bug fixes: - Fix QEMU p2p v2 migration when run from a v3 client (Daniel P. Berrange) - Don't raise an error if the migration cookie is NULL (Daniel P. Berrange) - Fix check of virKillProcess return status (Daniel P. Berrange) - qemu: Avoid use after free in qemuCaps parsing (Jiri Denemark) - security driver: ignore EINVAL when chowning an image file (Laine Stump) - migrate: detect xml incompatibility (Eric Blake) - event: avoid memory leak on cleanup (Eric Blake) - qemu: avoid memory leak on vcpupin (Eric Blake) - remote: avoid leak on failure (Eric Blake) - lock: avoid leak on failure (Eric Blake) - storage: avoid memory leak on stat failure (Eric Blake) - storage: avoid memory leak (Eric Blake) - libvirtd: avoid leak on failure (Eric Blake) - command: avoid leak on failure (Eric Blake) - Explicitly set VM state to paused after migration completes (Daniel P. Berrange) - Fix regressions BlockStats/Info APIs in QEMU driver (Daniel P. Berrange) - Fix minor issues in libxenlight managed save (Jim Fehlig) - Fix auditing of disk hotunplug operations (Daniel P. Berrange) - Avoid crash on NULL pointer in lock driver impls during hotplug (Daniel P. Berrange) - Fix return value in lock manager hotplug methods (Daniel P. Berrange) - Make sure virDomainSave/virDomainManagedSave reset id to -1 (Daniel P. Berrange) - Fix handling of VIR_EVENT_HANDLE_ERROR in QEMU monitor (Daniel P. Berrange) - security: plug regression introduced in disk probe logic (Eric Blake) - Fix order of disks and controllers (Jiri Denemark) - util: Fix incorrect error in PCI functions (Osier Yang) - openvz: Restore original EOF handling in openvzGetProcessInfo (Matthias Bolte) - openvz: Handle getline failures in openvzReadConfigParam properly (Matthias Bolte) - conf: Fix incorrect spice graphic XML format on compression options (Michal Privoznik) - nwfilter: reorder locks (Stefan Berger) - openvz: Fix regression in openvzGetVPSUUID (Matthias Bolte) - esx: Fix regression in absolute file name handling (Matthias Bolte) - openvz: Fix regression in config file parsing (Taisuke Yamada) - audit: fix minor off-by-one (Eric Blake) - Fix libxl vm def handling on domU cleanup (Markus Groß) - Fix modifying disk devices in qemu driver (Markus Groß) - json: Avoid passing large positive 64 bit integers to QMP. (Richard W.M. Jones) - Fix initialization of current vcpus in libxl driver (Jim Fehlig) - qemu: Don't change ownership of file when appending to it (Michal Privoznik) - qemu: fix typo in name - should be VHOST_NET, not VNET_HOST (Laine Stump) - Fix QEMU -vnc arg generation with raw IPv6 addresses (Daniel P. Berrange) - Fix peer2peer migration with transient VMs (Daniel P. Berrange) - python: Don't free must-not-free variables (Michal Privoznik) - python: Fix typo in bindings (Michal Privoznik) - libvirt.spec: /var/cache/libvirt should be 0711. (Richard W.M. Jones) - remote: remove bogus virDomainFree. (Richard W.M. Jones) - qemudDomainMemoryPeek: change ownership/selinux label on temporary file. (Richard W.M. Jones) - Fix QEMU migration cookie crash for guests with no graphics (Daniel P. Berrange) - Blank out the 'listenAddr' parameter if empty string (Daniel P. Berrange) - Improve invalid argument checks for the public API (Matthias Bolte) - qemu: fix typo in spice migration code (Michal Privoznik) - openvz: fix logic bug in previous patch (Eric Blake) - Fix error reporting in stream creation code (Matthias Bolte) - virsh: Correctly initialize libvirt (Jiri Denemark) - Fix leak of mime type string in screenshot dispatcher (Daniel P. Berrange) - Fix remote dispatcher for screenshot command (Daniel P. Berrange) - Add missing initialization to 'ret' in qemu migration (Daniel P. Berrange) - Close all sockets before cancelling QEMU migration (Daniel P. Berrange) - Ensure we always read a full buffer in tunnelled migration (Daniel P. Berrange) - phyp: avoid a crash (Eric Blake) - stream: avoid use-after-free (Eric Blake) - virsh: Fix uninitialized variable warning (Matthias Bolte) - libxl: fix typos in previous patch (Eric Blake) - node_device: avoid null dereference on error (Eric Blake) - tests: avoid crash when run under gcov (Eric Blake) - Fix two uninitialized variable warnings (Matthias Bolte) - apparmor: Fix uninitalized variable warning in virt-aa-helper (Matthias Bolte) - Fix qemuMigrationToFile nonull annotation (Daniel P. Berrange) Improvements: - API: consolidate common unreleased enums (Eric Blake) - uml: correct command line networking parameters (Heath Petersen) - Add call to sanlock_restrict() in QEMU lock driver (Daniel P. Berrange) - screenshot: Expose the new API in virsh (Michal Privoznik) - libxl: adds managed save and restore support (Markus Groß) - libxl: get maximum memory of running domain (Markus Groß) - Add a plugin for the 'sanlock' project (Daniel P. Berrange) - Allow leases to be hotpluged with QEMU guests (Daniel P. Berrange) - Support passing QEMU lock state to dest during migration (Daniel P. Berrange) - Integrate the QEMU driver with the lock manager infrastructure (Daniel P. Berrange) - Add higher level lock API for domain objects (Daniel P. Berrange) - Add a 'nop' lock driver implementation. (Daniel P. Berrange) - Support leases in guest XML and lock manager (Daniel P. Berrange) - Allow handshake with child process during startup (Daniel P. Berrange) - tests: really fix QEMU XML-2-ARGV graphics-spice-timeout test (Eric Blake) - build: avoid corrupting / in RHEL 5 (Eric Blake) - tests: Add a test for correct disk device ordering (Jiri Denemark) - Fix QEMU XML-2-ARGV graphics-spice-timeout test (Matthias Bolte) - remote generator: Handle some virTypedParameterPtr using functions (Matthias Bolte) - apibuild: Restrict long usage to existing functions and structs (Matthias Bolte) - remote generator: Legacy support for hyper to long mappings (Matthias Bolte) - Use common parameter order for remote(De)SerializeTypedParameters (Matthias Bolte) - openvz: Add simple test for openvzReadNetworkConf (Matthias Bolte) - openvz: Set virtType to openvz in openvzLoadDomains (Matthias Bolte) - storage: List directory volumes for dir/fs/netfs pools (Cole Robinson) - tests: Add more complex domain scheme test data (Michal Privoznik) - tests: Test for SPICE compression options (Michal Privoznik) - Ignore backward compatibility macros in apibuild.py (Matthias Bolte) - openvz: fix bridge devices parsing in openvzReadNetworkConf() (Jean-Baptiste Rouault) - Allow custom XML to be passed in during migration (Daniel P. Berrange) - Allow virsh to pass in a custom XML document for migration (Daniel P. Berrange) - Add an API for comparing the ABI of two guest configurations (Daniel P. Berrange) - Remove unused 'target' field in virDomainHostdevDef (Daniel P. Berrange) - lxc: Seperate domain config loading (Osier Yang) - openvz: Automatically disable on non-Linux systems (Matthias Bolte) - Support maximum and current memory flags in libxl driver (Markus Groß) - Add domainSave/Restore to libxl driver (Markus Groß) - Add domainCoreDump to libxl driver (Markus Groß) - Populate domid field of devices for libxenlight (Markus Groß) - Issue full error messages when processing QEMU monitor I/O (Daniel P. Berrange) - Don't kill QEMU process when a monitor I/O parsing error occurs (Daniel P. Berrange) - Introduce a new event emitted when a virtualization failure occurs (Daniel P. Berrange) - sched: provide new API shims for remaining drivers (Eric Blake) - virsh: improve schedinfo querying ability (Eric Blake) - remote: introduce remoteGetSchedulerParametersFlags (Eric Blake) - qemu: introduce qemuGetSchedulerParametersFlags (Eric Blake) - sched: introduce virDomainGetSchedulerParametersFlags (Eric Blake) - remote: consolidate typed parameter handling (Eric Blake) - maint: prefer newer API names internally (Eric Blake) - libvirt.h: avoid regression, and document preferred name (Eric Blake) - Fix virExecWithHook Prototype (Richard Laager) - interface: reformat error logs (Laine Stump) - interface: implement a test driver for network config transaction API. (Michal Privoznik) - interface: Implement driver methods for network config transaction API (Michal Privoznik) - interface: expose network config transaction API to virsh (Michal Privoznik) - interface: implement remote protocol for network config transaction API (Michal Privoznik) - interface: implement public APIs for libvirt transactional network changes (Michal Privoznik) - interface: define internal driver API for network config transactions (Michal Privoznik) - sexpr: Improve serialization error reporting (Matthias Bolte) - qemu: implement the new flags for setting memory parameters (Hu Tao) - qemu: implement the new flags for getting memory parameters (Hu Tao) - qemu : support persistent add/delete network interface (KAMEZAWA Hiroyuki) - openvz: Add simple testcase for config file parsing function (Matthias Bolte) - Add disk attach/detach support to libxl driver (Markus Groß) - Refactored libxl datastructure instantiation (Markus Groß) - schema: Add graphics element passwdValidTo attribute to schema (Michal Privoznik) - Fix sign mismatches between public API, driver API and XDR protocol (Matthias Bolte) - remote generator: Fix XDR sign mismatch for virNodeGet(Cells)FreeMemory (Matthias Bolte) - remote generator: Handle stream-using functions (Matthias Bolte) - remote generator: Make call-by-reference handling stricter (Matthias Bolte) - remote generator: Rename virNodeGetCellsFreeMemory parameters to common pattern (Matthias Bolte) - remote generator: Handle virDomainCreateWithFlags (Matthias Bolte) - Allow to explicitly disable the secrets a directory storage driver (Matthias Bolte) - Fix the signature of virDomainMigrateFinish3 for error reporting (Daniel P. Berrange) - Fix preservation of errors across migration steps (Daniel P. Berrange) - Improve debug logging of migration APIs (Daniel P. Berrange) - Fix error propagation in finish method for v3 migration (Daniel P. Berrange) - Fix SPICE seamless migration hostname (Daniel P. Berrange) - Fix resume on destination when doing non-live tunnelled migration (Daniel P. Berrange) - Introduce two method migration APIs (Daniel P. Berrange) - Add a second URI parameter to virDomainMigratePerform3 method (Daniel P. Berrange) - Extend v3 migration protocol to allow app supplied XML for target (Daniel P. Berrange) - esx: Report an error when auth pointer is missing instead of declining (Matthias Bolte) - Perform feature flag compat checking in QEMU migration cookies (Daniel P. Berrange) - Replace all remaining setgid/setuid calls with virSetUIDGID (Jiri Denemark) - util: Keep errno set to the root error after when returning from virSetUIDGID (Jiri Denemark) - storage: Add comment to picking return value of qemu-img (Michal Privoznik) - spice: support streaming-video parameter (Alon Levy) - nwfilter: enable filtering of gratuitous ARP packets (Stefan Berger) - qemu: don't require is_kvm for vhost-net support (Laine Stump) - Use per-user TLS certificates when possible (Doug Goldstein) - storage: pick return value of qemu-img (Michal Privoznik) - remote generator: Add special case for virConnectGetType (Matthias Bolte) - remote generator: Don't rely on $_ being stable over a large function (Matthias Bolte) - build: require newer gettext (Eric Blake) - virsh: Use Env variables for debug level and logfile (Supriya Kannery) - Don't generate cookies with v2 migration protocol. (Daniel P. Berrange) - Ensure p2p and direct migration use the new v3 protocol if available (Daniel P. Berrange) - remote: remove special case for getting version (Eric Blake) - virsh: optimize creation of default connection (Eric Blake) - Clarify that virDomainGet(Memory|Blkio)Parameters doesn't support subsets (Matthias Bolte) - Clarify that virDomainSet(Memory|Blkio)Parameters can take parameter subsets (Matthias Bolte) - Clarify that virDomainSetSchedulerParameters(Flags) can take subsets (Matthias Bolte) - Clarify the semantic of virDomainGetSchedulerParameters arguments (Matthias Bolte) - Clarify semantic of nparams argument of virDomainGetSchedulerType (Matthias Bolte) - Fix messages using VIR_ERR_XML_ERROR (Cole Robinson) - libvirt.h: consolidate typed parameter handling (Eric Blake) - build: update to latest gnulib (Eric Blake) - Add support for 'passthru' mode for direct network interfaces (Dirk Herrendorefer) - virsh: add parameters --live, --config and --current to cmd schedinfo (Hu Tao) - remote: introduce remoteSetSchedulerParametersFlags (Hu Tao) - qemu: introduce qemuSetSchedulerParametersFlags (Hu Tao) - sched: adjust parameter values to make current = 0 (Eric Blake) - introduce virDomainSetSchedulerParametersFlags (Hu Tao) - storage_backend: Convert virRunWithHook usage to virCommand (Cole Robinson) - storage_backend: Fix error reporting with regex helper (Cole Robinson) - storage: Covert regex helpers to virCommand (Cole Robinson) - openvz: Convert virExec usage to virCommand (Cole Robinson) - storage: iscsi: Convert virExec to virCommand (Cole Robinson) - remote generator, client: Add more special case handling (Matthias Bolte) - remote generator, client: Handle functions that return an optional string (Matthias Bolte) - esx: Simplify some esxVI_Boolean to bool (Matthias Bolte) - esx: Convert autoAnswer from esxVI_Boolean to a simple bool (Matthias Bolte) - esx: Fix race condition in esxVI_EnsureSession (Matthias Bolte) - esx: Whitespace cleanup in the generator (Matthias Bolte) - esx: Change some alignments in the input file (Matthias Bolte) - esx: Change generated method parameter autobinding (Matthias Bolte) - virsh: Report an error when virGetUserDirectory fails (Matthias Bolte) - maint: mark more perl scripts executable (Eric Blake) - maint: mark perl script executable (Eric Blake) - maint: ignore generated file (Eric Blake) - Improve error message when XDR marshalling fails (Daniel P. Berrange) - Run tunnelled migration IO in separate thread (Daniel P. Berrange) - Don't overwrite error when stream send fails (Daniel P. Berrange) - Make tunnelled migration honour resource restriction (Daniel P. Berrange) - Refactor migration completion loop to allow code reuse (Daniel P. Berrange) - Implement migration v3 protocol in QEMU driver (Daniel P. Berrange) - Merge tunnel and non-tunnel migration impl into one (Daniel P. Berrange) - Fix VM teardown if prepare returns invalid URI in v2 migration (Daniel P. Berrange) - Refactor tunnelled migration methods (Daniel P. Berrange) - Wire up SPICE client relocation with QEMU migration (Daniel P. Berrange) - Pass graphics setup from dst back to src via migration cookies (Daniel P. Berrange) - Introduce migration cookies to QEMU driver (Daniel P. Berrange) - Fix locking with qemuMigrationPrepareTunnel method (Daniel P. Berrange) - Remote driver implementation of new migration API (Daniel P. Berrange) - Add many version number annotations to drivers (Daniel P. Berrange) - Convert all driver struct intializers to C99 style (Daniel P. Berrange) - Tweak driver naming for consistency with public API (Daniel P. Berrange) - qemu: Update domain state when reconnecting monitor (Jiri Denemark) - Implement domain state reason (Jiri Denemark) - Implement basic virDomainGetState in all drivers (Jiri Denemark) - virsh: Prefer virDomainGetState over virDomainGetInfo (Jiri Denemark) - Wire protocol format and remote driver for virDomainGetState (Jiri Denemark) - virDomainGetState public API implementation (Jiri Denemark) - Internal driver API for virDomainGetState (Jiri Denemark) - esx: Refactor common code in the generator (Matthias Bolte) - esx: Improve dynamic cast detection in the generator (Matthias Bolte) - esx: Improve list usage detection in the generator (Matthias Bolte) - esx: Remove 1000 lines of generated but unused code (Matthias Bolte) - esx: Generate implicit _this macros (Matthias Bolte) - esx: Move the Event type from the VI generator to manually written code (Matthias Bolte) - test: all test_scripts should be part of tarball when building without libvirtd (Wen Congyang) - build: probes.d and libvirtd.stp should be part of tarball (Wen Congyang) - remote: Convert SSH tunnel to virCommand (Cole Robinson) - storage: Convert qemu-img -help parsing to virCommand (Cole Robinson) - apparmor: Convert virExec usage to virCommand (Cole Robinson) - util: Combine __virExec and virExecWithHook (Cole Robinson) - remote_driver: Convert virExecDaemonize usage to virCommand (Cole Robinson) - remote: Use virDomainEventState helpers (Cole Robinson) - libxl: Convert to virDomainEventState (Cole Robinson) - test: Use virDomainEventState helpers (Cole Robinson) - lxc: Use virDomainEventState helpers (Cole Robinson) - qemu: Use virDomainEventState helpers (Cole Robinson) - domain_event: Add common domain event queue/flush helpers (Cole Robinson) - domain_event: Add virDomainEventState structure (Cole Robinson) - xml: Use virXMLParse* helpers everywhere (Cole Robinson) - domain: Require init node for container guests (Cole Robinson) - xml: Make sure virXpathNodeSet always sets an error (Cole Robinson) - vbox: Implement the Screenshot driver methods (Michal Privoznik) - qemu: Implement the Screenshot driver methods (Michal Privoznik) - virFDStream: Add option for delete file after it's opening (Michal Privoznik) - Implementing the Screenshot remote protocol (Michal Privoznik) - Implementing the Screenshot public API (Michal Privoznik) - Defining the Screenshot internal API (Michal Privoznik) - remote generator: Add skipgen/autogen flags to .x files (Matthias Bolte) - nwfilter: drop unused flag argument (Eric Blake) - maint: omit translation for all VIR_INFO (Eric Blake) - build: wrap macro body in one-shot do-while loop (Eric Blake) - build: drop need for VIR_DOMAIN_DEBUG0() (Eric Blake) - libvirt,logging: cleanup VIR_XXX0() (Lai Jiangshan) - maint: avoid a couple of gnulib regressions (Eric Blake) - build: remove some dead assignments (Eric Blake) - Allow destroying QEMU VM even if a job is active (Daniel P. Berrange) - Add support for YAJL version 2 API/ABI (Daniel P. Berrange) - libxl: support enabling the HPET (Paolo Bonzini) - libxl: support enabling the HPET (Paolo Bonzini) - xen: parse and generate hpet item in sxpr (Paolo Bonzini) - virsh: flexibility in CA cert and user cert/key (Doug Goldstein) - remote generator: Make parsing stricter (Matthias Bolte) - remote generator: Unify unsigned int notation in .x files (Matthias Bolte) - remote generator: Fix comment ignoring regex (Matthias Bolte) - Rename internal DumpXML functions to GetXMLDesc (Matthias Bolte) - qemu,inject-nmi: Implement the driver methods (Lai Jiangshan) - inject-nmi: Expose the new API in virsh (Lai Jiangshan) - inject-nmi: Implementing the remote protocol (Lai Jiangshan) - inject-nmi: Implementing the public API (Lai Jiangshan) - inject-nmi: Defining the internal API (Lai Jiangshan) - lxc: container: Build init cmd before we close stdout (Cole Robinson) - lxc: container: Convert init node exec to virCommand (Cole Robinson) - command: Add virCommandExec helper (Cole Robinson) - lxc: driver: Improve logging when launching emulator (Cole Robinson) - lxc: driver: Convert emulator launching to virCommand (Cole Robinson) - command: Add virCommandEnvAddFormat (Cole Robinson) - build: translate generated strings (Eric Blake) - Pull in gnulib fnmatch module (Daniel P. Berrange) - json: Fix *GetBoolean functions (Jiri Denemark) - build: rebuild generated files after rpcgen_fix.pl tweak (Eric Blake) - Re-add libvirt.spec to tarball to allwo "make rpm" (Daniel Veillard) - build: Remove all generated RPC files from GIT (Daniel P. Berrange) - build: refactor generated RPC files (Daniel P. Berrange) - build: rename generated files to .h, for automake's sake (Eric Blake) - remote generator, client: Handle arguments with limited length (Matthias Bolte) - remote generator, client: Handle functions that return lists (Matthias Bolte) - remote: Refactor remote*Open and remote*Close functions (Matthias Bolte) - remote generator, client: Handle functions with optional return values (Matthias Bolte) - remote generator, client: Handle functions with multiple return values (Matthias Bolte) - remote generator, client: Handle functions with wrapped return values (Matthias Bolte) - remote generator, client: Handle functions with simple return values (Matthias Bolte) - remote generator, client: Handle simple remote driver functions (Matthias Bolte) - remote generator: Reorder into prepare and print section (Matthias Bolte) - remote generator, daemon: Handle functions with multiple return values (Matthias Bolte) - remote generator, daemon: Combine multiple ifs in the generated code (Matthias Bolte) - remote generator, daemon: Handle the CPU functions (Matthias Bolte) - remote generator: Refactor duplicate code (Matthias Bolte) - remote generator, daemon: Handle functions that return lists (Matthias Bolte) - remote: Rename 'nameslen' to the common 'maxnames' (Matthias Bolte) - remote generator, daemon: Handle functions with optional arguments (Matthias Bolte) - remote generator, daemon: Handle more functions (Matthias Bolte) - remote generator, daemon: Handle simple-return-value functions (Matthias Bolte) - remote generator, daemon: Change capitalization of some words (Matthias Bolte) - remote: Replace 'domain' with 'dom' in the XDR protocol (Matthias Bolte) - remote generator, daemon: Output function bodies too (Matthias Bolte) - remote generator: Replace tabs with spaces and rename file (Matthias Bolte) - Balloon dom0 in libxl driver (Markus Groß) - remote: check actual access to the cert (Doug Goldstein) - maint: avoid comparisons to bool constants (Eric Blake) - Add warning message to XML definition files stored on disk (Michal Privoznik) - Fix security driver handling of FIFOs with QEMU (Daniel P. Berrange) - Make taint warnings also go into the domain logfile (Daniel P. Berrange) - Add a qemuDomainAppendLog method for writing to the domain logfile (Daniel P. Berrange) - Move qemuProcessLogReadFD and qemuProcessLogFD methods (Daniel P. Berrange) - Log taint warnings in QEMU driver (Daniel P. Berrange) - Add field to virDomainObjPtr to track "tainting" (Daniel P. Berrange) - virsh: improve TLS certificate error messages (Doug Goldstein) - virsh: use new method for easier log to file (Eric Blake) - buf: add virBufferVasprintf (Eric Blake) - maint: rename virBufferVSprintf to virBufferAsprintf (Eric Blake) - Make QEMU migration use cached qemu capabilities data (Daniel P. Berrange) - Make QEMU hotplug use cached qemu capabilities data (Daniel P. Berrange) - Persist qemu capabilities in the domain status file (Daniel P. Berrange) Cleanups: - build: silence coverity false positive (Eric Blake) - build: silence coverity false positive (Eric Blake) - .gitignore: Exempt a new test binary. (Eric Blake) - Ensure hvsupport.html.in is built before HTML validation (Daniel P. Berrange) - Fix a make check error (Daniel Veillard) - Add sexpr.c to the list of file with translatable messages (Matthias Bolte) - don't check flags in virDomainSetSchedulerParametersFlags (Hu Tao) - build: silence clang false positive (Eric Blake) - Remove obsolete remoteDispatchOOMError method (Daniel P. Berrange) - remote: keep 'make check' happy (Eric Blake) - build: avoid compiler warning during configure (Wen Congyang) - Remove bogus assert() from migration code (Daniel P. Berrange) - build: drop files generated by config.status from tarball (Eric Blake) Thanks everybody for contributing to the release, with documentation, ideas, feedback or patches ! Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH] esx: Fix driver method version annotations
by Matthias Bolte 06 Jun '11

06 Jun '11

--- src/esx/esx_interface_driver.c | 6 +++--- src/esx/esx_network_driver.c | 6 +++--- src/esx/esx_nwfilter_driver.c | 6 +++--- src/esx/esx_secret_driver.c | 6 +++--- src/esx/esx_storage_driver.c | 2 +- 5 files changed, 13 insertions(+), 13 deletions(-) diff --git a/src/esx/esx_interface_driver.c b/src/esx/esx_interface_driver.c index 3f9c175..a468976 100644 --- a/src/esx/esx_interface_driver.c +++ b/src/esx/esx_interface_driver.c @@ -66,9 +66,9 @@ esxInterfaceClose(virConnectPtr conn) static virInterfaceDriver esxInterfaceDriver = { - "ESX", /* name */ - .open = esxInterfaceOpen, /* 0.7.0 */ - .close = esxInterfaceClose, /* 0.7.0 */ + .name = "ESX", + .open = esxInterfaceOpen, /* 0.7.6 */ + .close = esxInterfaceClose, /* 0.7.6 */ }; diff --git a/src/esx/esx_network_driver.c b/src/esx/esx_network_driver.c index 9385a26..3c76fae 100644 --- a/src/esx/esx_network_driver.c +++ b/src/esx/esx_network_driver.c @@ -66,9 +66,9 @@ esxNetworkClose(virConnectPtr conn) static virNetworkDriver esxNetworkDriver = { - "ESX", /* name */ - .open = esxNetworkOpen, /* 0.7.0 */ - .close = esxNetworkClose, /* 0.7.0 */ + .name = "ESX", + .open = esxNetworkOpen, /* 0.7.6 */ + .close = esxNetworkClose, /* 0.7.6 */ }; diff --git a/src/esx/esx_nwfilter_driver.c b/src/esx/esx_nwfilter_driver.c index ee85abc..13cacd4 100644 --- a/src/esx/esx_nwfilter_driver.c +++ b/src/esx/esx_nwfilter_driver.c @@ -65,9 +65,9 @@ esxNWFilterClose(virConnectPtr conn) static virNWFilterDriver esxNWFilterDriver = { - "ESX", /* name */ - .open = esxNWFilterOpen, /* 0.8.0 */ - .close = esxNWFilterClose, /* 0.8.0 */ + .name = "ESX", + .open = esxNWFilterOpen, /* 0.8.1 */ + .close = esxNWFilterClose, /* 0.8.1 */ }; diff --git a/src/esx/esx_secret_driver.c b/src/esx/esx_secret_driver.c index dff2290..656224e 100644 --- a/src/esx/esx_secret_driver.c +++ b/src/esx/esx_secret_driver.c @@ -64,9 +64,9 @@ esxSecretClose(virConnectPtr conn) static virSecretDriver esxSecretDriver = { - "ESX", /* name */ - .open = esxSecretOpen, /* 0.7.1 */ - .close = esxSecretClose, /* 0.7.1 */ + .name = "ESX", + .open = esxSecretOpen, /* 0.7.6 */ + .close = esxSecretClose, /* 0.7.6 */ }; diff --git a/src/esx/esx_storage_driver.c b/src/esx/esx_storage_driver.c index 5ee0cf4..6461917 100644 --- a/src/esx/esx_storage_driver.c +++ b/src/esx/esx_storage_driver.c @@ -1645,7 +1645,7 @@ esxStoragePoolIsPersistent(virStoragePoolPtr pool ATTRIBUTE_UNUSED) static virStorageDriver esxStorageDriver = { - "ESX", /* name */ + .name = "ESX", .open = esxStorageOpen, /* 0.7.6 */ .close = esxStorageClose, /* 0.7.6 */ .numOfPools = esxNumberOfStoragePools, /* 0.8.2 */ -- 1.7.0.4

2 2

[libvirt] [PATCH] Fix QEMU p2p v2 migration when run from a v3 client
by Daniel P. Berrange 06 Jun '11

06 Jun '11

When peer-2-peer migration was invoked by a client supporting v3, but where the target server only supported v2, we'd not correctly shutdown the guest. * src/qemu/qemu_migration.c: Ensure guest is shutdown in v2 peer 2 peer migration --- src/qemu/qemu_migration.c | 17 +++++++++++------ 1 files changed, 11 insertions(+), 6 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index fa506e2..f7eaa1c 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -2199,12 +2199,12 @@ static int doPeer2PeerMigrate(struct qemud_driver *driver, const char *uri, unsigned long flags, const char *dname, - unsigned long resource) + unsigned long resource, + bool *v3proto) { int ret = -1; virConnectPtr dconn = NULL; bool p2p; - bool v3; VIR_DEBUG("driver=%p, sconn=%p, vm=%p, xmlin=%s, dconnuri=%s, " "uri=%s, flags=%lu, dname=%s, resource=%lu", driver, sconn, vm, NULLSTR(xmlin), NULLSTR(dconnuri), @@ -2226,8 +2226,12 @@ static int doPeer2PeerMigrate(struct qemud_driver *driver, qemuDomainObjEnterRemoteWithDriver(driver, vm); p2p = VIR_DRV_SUPPORTS_FEATURE(dconn->driver, dconn, VIR_DRV_FEATURE_MIGRATION_P2P); - v3 = VIR_DRV_SUPPORTS_FEATURE(dconn->driver, dconn, - VIR_DRV_FEATURE_MIGRATION_V3); + /* v3proto reflects whether the caller used Perform3, but with + * p2p migrate, regardless of whether Perform3 or Perform3 + * were used, we decide protocol based on what target supports + */ + *v3proto = VIR_DRV_SUPPORTS_FEATURE(dconn->driver, dconn, + VIR_DRV_FEATURE_MIGRATION_V3); qemuDomainObjExitRemoteWithDriver(driver, vm); if (!p2p) { @@ -2243,7 +2247,7 @@ static int doPeer2PeerMigrate(struct qemud_driver *driver, goto cleanup; } - if (v3) + if (*v3proto) ret = doPeer2PeerMigrate3(driver, sconn, dconn, vm, xmlin, dconnuri, uri, flags, dname, resource); else @@ -2310,7 +2314,8 @@ int qemuMigrationPerform(struct qemud_driver *driver, } if (doPeer2PeerMigrate(driver, conn, vm, xmlin, - dconnuri, uri, flags, dname, resource) < 0) + dconnuri, uri, flags, dname, + resource, &v3proto) < 0) /* doPeer2PeerMigrate already set the error, so just get out */ goto endjob; } else { -- 1.7.4.4

2 2

[libvirt] [PATCH] qemu: Parse current balloon value returned by query_balloon.
by Osier Yang 06 Jun '11

06 Jun '11

Qemu once supported following memory stats which will returned by "query_balloon": stat_put(dict, "actual", actual); stat_put(dict, "mem_swapped_in", dev->stats[VIRTIO_BALLOON_S_SWAP_IN]); stat_put(dict, "mem_swapped_out", dev->stats[VIRTIO_BALLOON_S_SWAP_OUT]); stat_put(dict, "major_page_faults", dev->stats[VIRTIO_BALLOON_S_MAJFLT]); stat_put(dict, "minor_page_faults", dev->stats[VIRTIO_BALLOON_S_MINFLT]); stat_put(dict, "free_mem", dev->stats[VIRTIO_BALLOON_S_MEMFREE]); stat_put(dict, "total_mem", dev->stats[VIRTIO_BALLOON_S_MEMTOT]); But it later disabled all the stats except "actual" by commit 07b0403dfc2b2ac179ae5b48105096cc2d03375a. libvirt doesn't parse "actual", so user will always see a empty result with "virsh dommemstat $domain". Even qemu haven't disabled the stats, we should support parsing "actual". --- include/libvirt/libvirt.h.in | 4 +++- src/libvirt.c | 2 ++ src/qemu/qemu_monitor_json.c | 12 ++++++++++++ tools/virsh.c | 2 ++ 4 files changed, 19 insertions(+), 1 deletions(-) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index 8058229..864852b 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -456,11 +456,13 @@ typedef enum { */ VIR_DOMAIN_MEMORY_STAT_AVAILABLE = 5, + /* Current balloon value (in KB). */ + VIR_DOMAIN_MEMORY_STAT_ACTUAL = 6, /* * The number of statistics supported by this version of the interface. * To add new statistics, add them to the enum and increase this value. */ - VIR_DOMAIN_MEMORY_STAT_NR = 6, + VIR_DOMAIN_MEMORY_STAT_NR = 7, } virDomainMemoryStatTags; typedef struct _virDomainMemoryStat virDomainMemoryStatStruct; diff --git a/src/libvirt.c b/src/libvirt.c index eaae0ec..c635ff5 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -5802,6 +5802,8 @@ error: * The amount of memory which is not being used for any purpose (in kb). * VIR_DOMAIN_MEMORY_STAT_AVAILABLE: * The total amount of memory available to the domain's OS (in kb). + * VIR_DOMAIN_MEMORY_STAT_ACTUAL: + * Current balloon value (in kb). * * Returns: The number of stats provided or -1 in case of failure. */ diff --git a/src/qemu/qemu_monitor_json.c b/src/qemu/qemu_monitor_json.c index 75adf66..9f16403 100644 --- a/src/qemu/qemu_monitor_json.c +++ b/src/qemu/qemu_monitor_json.c @@ -1119,6 +1119,18 @@ int qemuMonitorJSONGetMemoryStats(qemuMonitorPtr mon, goto cleanup; } + if (virJSONValueObjectHasKey(data, "actual") && (got < nr_stats)) { + if (virJSONValueObjectGetNumberUlong(data, "actual", &mem) < 0) { + qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("info balloon reply was missing balloon actual")); + ret = -1; + goto cleanup; + } + stats[got].tag = VIR_DOMAIN_MEMORY_STAT_ACTUAL; + stats[got].val = (mem/1024); + got++; + } + if (virJSONValueObjectHasKey(data, "mem_swapped_in") && (got < nr_stats)) { if (virJSONValueObjectGetNumberUlong(data, "mem_swapped_in", &mem) < 0) { qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", diff --git a/tools/virsh.c b/tools/virsh.c index 5679a2d..a358b73 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -1147,6 +1147,8 @@ cmdDomMemStat(vshControl *ctl, const vshCmd *cmd) vshPrint (ctl, "unused %llu\n", stats[i].val); if (stats[i].tag == VIR_DOMAIN_MEMORY_STAT_AVAILABLE) vshPrint (ctl, "available %llu\n", stats[i].val); + if (stats[i].tag == VIR_DOMAIN_MEMORY_STAT_ACTUAL) + vshPrint (ctl, "actual %llu\n", stats[i].val); } virDomainFree(dom); -- 1.7.4

3 3

[libvirt] Building on Solaris 11 Express
by Richard Laager 05 Jun '11

05 Jun '11

I would really like to use ZFS's zvols as storage for my guests. I really like ZFS's snapshots, send/receive, and checksumming features. (Compression and deduplication might be useful as well.) I'm roughly about half-way done writing a ZFS storage backend for libvirt. It behaves like the LVM backend. I've been writing it on Linux. However, ZFS on Linux is not yet production-ready. So I tried building libvirt on Solaris 11 Express. The following outlines the trouble (and successes) I've had so far. First, I had to prune down the features being built: I added --without-xen to deal with the following. I wasn't intending on using Xen, so this doesn't bother me personally. configure: error: You must install the Xen development package to compile Xen driver with -lxenstore I added --without-openvz to deal with the following. Likewise, I'm not using it. CC libvirt_driver_openvz_la-openvz_driver.lo openvz/openvz_driver.c:45:19: paths.h: No such file or directory In file included from ../src/conf/domain_conf.h:40, from openvz/openvz_conf.h:33, from openvz/openvz_driver.c:56: ../src/conf/nwfilter_conf.h:521: warning: parameter has incomplete type I added --without-network to deal with the following. I may need to actually fix this at some point. CC libvirt_util_la-bridge.lo util/bridge.c:39:20: paths.h: No such file or directory util/bridge.c:42:55: linux/param.h: No such file or directory util/bridge.c:43:55: linux/sockios.h: No such file or directory util/bridge.c:44:55: linux/if_bridge.h: No such file or directory util/bridge.c:45:55: linux/if_tun.h: No such file or directory util/bridge.c: In function `ifSetInterfaceMac': util/bridge.c:313: error: `SIOCGIFHWADDR' undeclared (first use in this function) util/bridge.c:313: error: (Each undeclared identifier is reported only once util/bridge.c:313: error: for each function it appears in.) util/bridge.c:316: error: structure has no member named `ifr_hwaddr' util/bridge.c:318: error: `SIOCSIFHWADDR' undeclared (first use in this function) util/bridge.c: In function `ifGetMtu': util/bridge.c:347: error: `SIOCGIFMTU' undeclared (first use in this function) util/bridge.c: In function `ifSetMtu': util/bridge.c:378: error: `SIOCSIFMTU' undeclared (first use in this function) util/bridge.c: In function `brAddTap': util/bridge.c:501: error: `IFF_TAP' undeclared (first use in this function) util/bridge.c:501: error: `IFF_NO_PI' undeclared (first use in this function) util/bridge.c:515: error: `TUNSETIFF' undeclared (first use in this function) util/bridge.c:537: error: `TUNSETPERSIST' undeclared (first use in this function) util/bridge.c: In function `brDeleteTap': util/bridge.c:567: error: `IFF_TAP' undeclared (first use in this function) util/bridge.c:567: error: `IFF_NO_PI' undeclared (first use in this function) util/bridge.c:574: error: `TUNSETIFF' undeclared (first use in this function) util/bridge.c:575: error: `TUNSETPERSIST' undeclared (first use in this function) util/bridge.c: In function `brSetInterfaceUp': util/bridge.c:612: error: `SIOCGIFFLAGS' undeclared (first use in this function) util/bridge.c:620: error: `SIOCSIFFLAGS' undeclared (first use in this function) util/bridge.c: In function `brGetInterfaceUp': util/bridge.c:652: error: `SIOCGIFFLAGS' undeclared (first use in this function) Then, I had to fix a trivial code error. I've mailed the patch to this list with a subject of "[PATCH] Fix virExecWithHook Prototype". I also had linker trouble. Solaris 11's gcc package uses Solaris ld, not GNU ld. I had to change the following: VERSION_SCRIPT_FLAGS=-Wl,--version-script= `$LD --help 2>&1 | grep -- --version-script >/dev/null` || \ - VERSION_SCRIPT_FLAGS="-Wl,-M -Wl," + VERSION_SCRIPT_FLAGS="" Interestingly, this code seems to have been added to support Solaris ld according to git blame; see git log c2fb8bfe. I'm not sure what is correct here. At this point, libvirt builds and installs. I tested virsh against a remote system using: virsh -c qemu+ssh://REMOTE/system This worked except for the fact that it was trying to attach to /usr/local/var/run/libvirt-sock on the remote system due to /usr/local being the --prefix used with the *local* virsh was built. I'm not sure how that architectural issue should be addressed. For testing, I just made a /usr/local/var -> /var symlink on the remote system. However, libvirtd will not start locally (on Solaris) which has pretty much stopped my progress. I'm sure this must be something trivial. Is it getting a bad prefix? Is something other than libvirtd supposed to make the socket file? I did try making a /.libvirt directory for testing. # rm /usr/local/var/run/libvirtd.pid ; /usr/local/sbin/libvirtd -v 20:13:33.663: 1: info : libvirt version: 0.9.1 20:13:33.663: 1: error : qemudListenUnix:605 : Failed to bind socket to '@//.libvirt/libvirt-sock': No such file or directory Thanks, Richard

4 25

[libvirt] [PATCH] vbox: Fix typo in error message
by Matthias Bolte 04 Jun '11

04 Jun '11

--- I'm pushing this under the trivial rule. src/vbox/vbox_XPCOMCGlue.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/vbox/vbox_XPCOMCGlue.c b/src/vbox/vbox_XPCOMCGlue.c index 0caeef9..e7e9c37 100644 --- a/src/vbox/vbox_XPCOMCGlue.c +++ b/src/vbox/vbox_XPCOMCGlue.c @@ -97,7 +97,7 @@ tryLoadOne(const char *dir, bool setAppHome, bool ignoreMissing, if (!virFileExists(name)) { if (!ignoreMissing) { - VIR_ERROR(_("Libaray '%s' doesn't exist"), name); + VIR_ERROR(_("Library '%s' doesn't exist"), name); } VIR_FREE(name); -- 1.7.0.4

1 0

[libvirt] [PATCH] Don't raise an error if the migration cookie is NULL
by Daniel P. Berrange 04 Jun '11

04 Jun '11

The v2 migration protocol doesn't use cookies, so we should not be raising an error if the cookie parameters are NULL. * src/qemu/qemu_migration.c: Don't raise error if cookie is NULL --- src/qemu/qemu_migration.c | 7 ++----- 1 files changed, 2 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_migration.c b/src/qemu/qemu_migration.c index aa74d86..fa506e2 100644 --- a/src/qemu/qemu_migration.c +++ b/src/qemu/qemu_migration.c @@ -618,11 +618,8 @@ qemuMigrationBakeCookie(qemuMigrationCookiePtr mig, int *cookieoutlen, int flags) { - if (!cookieout || !cookieoutlen) { - qemuReportError(VIR_ERR_INVALID_ARG, "%s", - _("missing migration cookie data")); - return -1; - } + if (!cookieout || !cookieoutlen) + return 0; *cookieoutlen = 0; -- 1.7.4.4

2 1

[libvirt] [PATCH] Fix check of virKillProcess return status
by Daniel P. Berrange 04 Jun '11

04 Jun '11

The error code for virKillProcess is returned in the errno variable not the return value. THis mistake caused the logs to be filled with errors when shutting down QEMU processes * src/qemu/qemu_process.c: Fix process kill check. --- src/qemu/qemu_process.c | 6 ++---- 1 files changed, 2 insertions(+), 4 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 116253e..ce05133 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -2513,7 +2513,6 @@ cleanup: void qemuProcessKill(virDomainObjPtr vm) { int i; - int rc; VIR_DEBUG("vm=%s pid=%d", vm->def->name, vm->pid); if (!virDomainObjIsActive(vm)) { @@ -2535,9 +2534,8 @@ void qemuProcessKill(virDomainObjPtr vm) else signum = 0; /* Just check for existence */ - rc = virKillProcess(vm->pid, signum); - if (rc < 0) { - if (rc != -ESRCH) { + if (virKillProcess(vm->pid, signum) < 0) { + if (errno != -ESRCH) { char ebuf[1024]; VIR_WARN("Failed to kill process %d %s", vm->pid, virStrerror(errno, ebuf, sizeof ebuf)); -- 1.7.4.4

2 1

[libvirt] [PATCH] vbox: Fix version extraction on Windows for newer VirtualBox versions
by Matthias Bolte 04 Jun '11

04 Jun '11

VirtualBox 4.0.8 changed the registry key layout. Before the version number was in a Version key. Now the Version key contains %VER% and the actual version number is in VersionExt now. Move value lookup code into its own function: vboxLookupRegistryValue. --- src/vbox/vbox_MSCOMGlue.c | 87 +++++++++++++++++++++++++++++++-------------- 1 files changed, 60 insertions(+), 27 deletions(-) diff --git a/src/vbox/vbox_MSCOMGlue.c b/src/vbox/vbox_MSCOMGlue.c index e31a763..8aef266 100644 --- a/src/vbox/vbox_MSCOMGlue.c +++ b/src/vbox/vbox_MSCOMGlue.c @@ -2,7 +2,7 @@ /* * vbox_MSCOMGlue.c: glue to the MSCOM based VirtualBox API * - * Copyright (C) 2010 Matthias Bolte <matthias.bolte(a)googlemail.com> + * Copyright (C) 2010-2011 Matthias Bolte <matthias.bolte(a)googlemail.com> * * This library is free software; you can redistribute it and/or * modify it under the terms of the GNU Lesser General Public @@ -338,45 +338,31 @@ static nsIEventQueue vboxEventQueue = { -static int -vboxLookupVersionInRegistry(void) +static char * +vboxLookupRegistryValue(HKEY key, const char *keyName, const char *valueName) { - int result = -1; - const char *keyName = VBOX_REGKEY_ORACLE; LONG status; - HKEY key; DWORD type; DWORD length; char *value = NULL; - status = RegOpenKeyEx(HKEY_LOCAL_MACHINE, keyName, 0, KEY_READ, &key); + status = RegQueryValueEx(key, valueName, NULL, &type, NULL, &length); if (status != ERROR_SUCCESS) { - keyName = VBOX_REGKEY_SUN; - status = RegOpenKeyEx(HKEY_LOCAL_MACHINE, keyName, 0, KEY_READ, &key); - - if (status != ERROR_SUCCESS) { - /* Both keys aren't there, or we cannot open them. In general this - * indicates that VirtualBox is not installed, so we just silently - * fail here making vboxRegister() register the dummy driver. */ - return -1; - } - } - - status = RegQueryValueEx(key, "Version", NULL, &type, NULL, &length); - - if (status != ERROR_SUCCESS) { - VIR_ERROR(_("Could not query registry value '%s\\Version'"), keyName); + VIR_ERROR(_("Could not query registry value '%s\\%s'"), + keyName, valueName); goto cleanup; } if (type != REG_SZ) { - VIR_ERROR(_("Registry value '%s\\Version' has unexpected type"), keyName); + VIR_ERROR(_("Registry value '%s\\%s' has unexpected type"), + keyName, valueName); goto cleanup; } if (length < 2) { - VIR_ERROR(_("Registry value '%s\\Version' is too short"), keyName); + VIR_ERROR(_("Registry value '%s\\%s' is too short"), + keyName, valueName); goto cleanup; } @@ -386,10 +372,12 @@ vboxLookupVersionInRegistry(void) goto cleanup; } - status = RegQueryValueEx(key, "Version", NULL, NULL, (LPBYTE)value, &length); + status = RegQueryValueEx(key, valueName, NULL, NULL, (LPBYTE)value, &length); if (status != ERROR_SUCCESS) { - VIR_ERROR(_("Could not query registry value '%s\\Version'"), keyName); + VIR_FREE(value); + VIR_ERROR(_("Could not query registry value '%s\\%s'"), + keyName, valueName); goto cleanup; } @@ -397,7 +385,52 @@ vboxLookupVersionInRegistry(void) value[length] = '\0'; } - if (virParseVersionString(value, &vboxVersion)) { + cleanup: + return value; +} + +static int +vboxLookupVersionInRegistry(void) +{ + int result = -1; + const char *keyName = VBOX_REGKEY_ORACLE; + LONG status; + HKEY key; + char *value = NULL; + + status = RegOpenKeyEx(HKEY_LOCAL_MACHINE, keyName, 0, KEY_READ, &key); + + if (status != ERROR_SUCCESS) { + keyName = VBOX_REGKEY_SUN; + status = RegOpenKeyEx(HKEY_LOCAL_MACHINE, keyName, 0, KEY_READ, &key); + + if (status != ERROR_SUCCESS) { + /* Both keys aren't there, or we cannot open them. In general this + * indicates that VirtualBox is not installed, so we just silently + * fail here making vboxRegister() register the dummy driver. */ + return -1; + } + } + + /* The registry key layout changed around version 4.0.8. Before the version + * number was in the Version key, now the Version key can contain %VER% and + * the actual version number is in the VersionExt key then. */ + value = vboxLookupRegistryValue(key, keyName, "Version"); + + if (value == NULL) { + goto cleanup; + } + + if (STREQ(value, "%VER%")) { + VIR_FREE(value); + value = vboxLookupRegistryValue(key, keyName, "VersionExt"); + + if (value == NULL) { + goto cleanup; + } + } + + if (virParseVersionString(value, &vboxVersion) < 0) { VIR_ERROR(_("Could not parse version number from '%s'"), value); goto cleanup; } -- 1.7.0.4

2 2

[libvirt] [PATCH] API: consolidate common unreleased enums
by Eric Blake 04 Jun '11

04 Jun '11

This commit is safe precisely because there has been no release for any of the enum values being deleted (they were added post-0.9.1). After the 0.9.2 release, we can then take advantage of virDomainModificationImpact in more places. * include/libvirt/libvirt.h.in (virDomainModificationImpact): New enum. (virDomainSchedParameterFlags, virMemoryParamFlags): Delete, since these were never released, and the new enum works fine here. * src/libvirt.c (virDomainGetMemoryParameters) (virDomainSetMemoryParameters) (virDomainGetSchedulerParametersFlags) (virDomainSetSchedulerParametersFlags): Update documentation. * src/qemu/qemu_driver.c (qemuDomainSetMemoryParameters) (qemuDomainGetMemoryParameters, qemuSetSchedulerParametersFlags) (qemuSetSchedulerParameters, qemuGetSchedulerParametersFlags) (qemuGetSchedulerParameters): Adjust clients. * tools/virsh.c (cmdSchedinfo, cmdMemtune): Likewise. Based on ideas by Daniel Veillard and Hu Tao. --- This has to be applied before 0.9.2 if we like the idea. See https://www.redhat.com/archives/libvir-list/2011-June/msg00177.html for more details about why this is safe. I am intentionally not touching any enums that existed in 0.9.1 at this point in the release cycle. include/libvirt/libvirt.h.in | 30 ++++++++------ src/libvirt.c | 18 ++++---- src/qemu/qemu_driver.c | 84 +++++++++++++++++++++--------------------- tools/virsh.c | 12 +++--- 4 files changed, 74 insertions(+), 70 deletions(-) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index 8058229..df213f1 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -142,6 +142,23 @@ typedef enum { } virDomainCrashedReason; /** + * virDomainModificationImpact: + * + * Several APIs take flags to determine whether a change to the domain + * affects just the running instance, just the persistent definition, + * or both. The use of VIR_DOMAIN_AFFECT_CURRENT will resolve to + * either VIR_DOMAIN_AFFECT_LIVE or VIR_DOMAIN_AFFECT_CONFIG according + * to current domain state. VIR_DOMAIN_AFFECT_LIVE requires a running + * domain, and VIR_DOMAIN_AFFECT_CONFIG requires a persistent domain + * (whether or not it is running). + */ +typedef enum { + VIR_DOMAIN_AFFECT_CURRENT = 0, /* Affect current domain state. */ + VIR_DOMAIN_AFFECT_LIVE = 1 << 0, /* Affect running domain state. */ + VIR_DOMAIN_AFFECT_CONFIG = 1 << 1, /* Affect persistent domain state. */ +} virDomainModificationImpact; + +/** * virDomainInfoPtr: * * a virDomainInfo is a structure filled by virDomainGetInfo() and extracting @@ -338,12 +355,6 @@ typedef virTypedParameter *virTypedParameterPtr; /* Management of scheduler parameters */ -typedef enum { - VIR_DOMAIN_SCHEDPARAM_CURRENT = 0, /* affect current domain state */ - VIR_DOMAIN_SCHEDPARAM_LIVE = (1 << 0), /* Affect active domain */ - VIR_DOMAIN_SCHEDPARAM_CONFIG = (1 << 1), /* Affect next boot */ -} virDomainSchedParameterFlags; - /* * Fetch scheduler parameters, caller allocates 'params' field of size 'nparams' */ @@ -799,13 +810,6 @@ int virDomainGetBlkioParameters(virDomainPtr domain, /* Manage memory parameters. */ -/* flags for setting memory parameters */ -typedef enum { - VIR_DOMAIN_MEMORY_PARAM_CURRENT = 0, /* affect current domain state */ - VIR_DOMAIN_MEMORY_PARAM_LIVE = (1 << 0), /* affect active domain */ - VIR_DOMAIN_MEMORY_PARAM_CONFIG = (1 << 1) /* affect next boot */ -} virMemoryParamFlags; - /** * VIR_DOMAIN_MEMORY_PARAM_UNLIMITED: * diff --git a/src/libvirt.c b/src/libvirt.c index eaae0ec..cbe1926 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -2994,7 +2994,7 @@ error: * @params: pointer to memory parameter objects * @nparams: number of memory parameter (this value can be the same or * less than the number of parameters supported) - * @flags: currently unused, for future extension + * @flags: bitwise-OR of virDomainModificationImpact * * Change all or a subset of the memory tunables. * This function requires privileged access to the hypervisor. @@ -3049,7 +3049,7 @@ error: * @params: pointer to memory parameter object * (return value, allocated by the caller) * @nparams: pointer to number of memory parameters - * @flags: currently unused, for future extension + * @flags: one of virDomainModificationImpact * * Get all memory parameters, the @params array will be filled with the values * equal to the number of parameters suggested by @nparams @@ -5480,13 +5480,13 @@ error: * @nparams: pointer to number of scheduler parameter * (this value should be same than the returned value * nparams of virDomainGetSchedulerType) - * @flags: virDomainSchedParameterFlags + * @flags: one of virDomainModificationImpact * * Get the scheduler parameters, the @params array will be filled with the * values. * - * The value of @flags can be exactly VIR_DOMAIN_SCHEDPARAM_CURRENT, - * VIR_DOMAIN_SCHEDPARAM_LIVE, or VIR_DOMAIN_SCHEDPARAM_CONFIG. + * The value of @flags can be exactly VIR_DOMAIN_AFFECT_CURRENT, + * VIR_DOMAIN_AFFECT_LIVE, or VIR_DOMAIN_AFFECT_CONFIG. * * Returns -1 in case of error, 0 in case of success. */ @@ -5596,12 +5596,12 @@ error: * @nparams: number of scheduler parameter objects * (this value can be the same or less than the returned value * nparams of virDomainGetSchedulerType) - * @flags: virDomainSchedParameterFlags + * @flags: bitwise-OR of virDomainModificationImpact * * Change a subset or all scheduler parameters. The value of @flags - * should be either VIR_DOMAIN_SCHEDPARAM_CURRENT, or a bitwise-or of - * values from VIR_DOMAIN_SCHEDPARAM_LIVE and - * VIR_DOMAIN_SCHEDPARAM_CURRENT, although hypervisors vary in which + * should be either VIR_DOMAIN_AFFECT_CURRENT, or a bitwise-or of + * values from VIR_DOMAIN_AFFECT_LIVE and + * VIR_DOMAIN_AFFECT_CURRENT, although hypervisors vary in which * flags are supported. * * Returns -1 in case of error, 0 in case of success. diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index ec23bb3..2957467 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -4921,8 +4921,8 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, int ret = -1; bool isActive; - virCheckFlags(VIR_DOMAIN_MEMORY_PARAM_LIVE | - VIR_DOMAIN_MEMORY_PARAM_CONFIG, -1); + virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | + VIR_DOMAIN_AFFECT_CONFIG, -1); qemuDriverLock(driver); @@ -4936,14 +4936,14 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, isActive = virDomainObjIsActive(vm); - if (flags == VIR_DOMAIN_MEMORY_PARAM_CURRENT) { + if (flags == VIR_DOMAIN_AFFECT_CURRENT) { if (isActive) - flags = VIR_DOMAIN_MEMORY_PARAM_LIVE; + flags = VIR_DOMAIN_AFFECT_LIVE; else - flags = VIR_DOMAIN_MEMORY_PARAM_CONFIG; + flags = VIR_DOMAIN_AFFECT_CONFIG; } - if (flags & VIR_DOMAIN_MEMORY_PARAM_LIVE) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) { if (!isActive) { qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); @@ -4963,7 +4963,7 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, } } - if (flags & VIR_DOMAIN_MEMORY_PARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { if (!vm->persistent) { qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", _("cannot change persistent config of a transient domain")); @@ -4986,7 +4986,7 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, continue; } - if (flags & VIR_DOMAIN_MEMORY_PARAM_LIVE) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) { rc = virCgroupSetMemoryHardLimit(group, params[i].value.ul); if (rc != 0) { virReportSystemError(-rc, "%s", @@ -4995,7 +4995,7 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, } } - if (flags & VIR_DOMAIN_MEMORY_PARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { persistentDef->mem.hard_limit = params[i].value.ul; } } else if (STREQ(param->field, VIR_DOMAIN_MEMORY_SOFT_LIMIT)) { @@ -5007,7 +5007,7 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, continue; } - if (flags & VIR_DOMAIN_MEMORY_PARAM_LIVE) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) { rc = virCgroupSetMemorySoftLimit(group, params[i].value.ul); if (rc != 0) { virReportSystemError(-rc, "%s", @@ -5016,7 +5016,7 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, } } - if (flags & VIR_DOMAIN_MEMORY_PARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { persistentDef->mem.soft_limit = params[i].value.ul; } } else if (STREQ(param->field, VIR_DOMAIN_MEMORY_SWAP_HARD_LIMIT)) { @@ -5028,7 +5028,7 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, continue; } - if (flags & VIR_DOMAIN_MEMORY_PARAM_LIVE) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) { rc = virCgroupSetMemSwapHardLimit(group, params[i].value.ul); if (rc != 0) { virReportSystemError(-rc, "%s", @@ -5036,7 +5036,7 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, ret = -1; } } - if (flags & VIR_DOMAIN_MEMORY_PARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { persistentDef->mem.swap_hard_limit = params[i].value.ul; } } else if (STREQ(param->field, VIR_DOMAIN_MEMORY_MIN_GUARANTEE)) { @@ -5050,7 +5050,7 @@ static int qemuDomainSetMemoryParameters(virDomainPtr dom, } } - if (flags & VIR_DOMAIN_MEMORY_PARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { ret = virDomainSaveConfig(driver->configDir, persistentDef); } @@ -5077,8 +5077,8 @@ static int qemuDomainGetMemoryParameters(virDomainPtr dom, int rc; bool isActive; - virCheckFlags(VIR_DOMAIN_MEMORY_PARAM_LIVE | - VIR_DOMAIN_MEMORY_PARAM_CONFIG, -1); + virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | + VIR_DOMAIN_AFFECT_CONFIG, -1); qemuDriverLock(driver); @@ -5092,14 +5092,14 @@ static int qemuDomainGetMemoryParameters(virDomainPtr dom, isActive = virDomainObjIsActive(vm); - if (flags == VIR_DOMAIN_MEMORY_PARAM_CURRENT) { + if (flags == VIR_DOMAIN_AFFECT_CURRENT) { if (isActive) - flags = VIR_DOMAIN_MEMORY_PARAM_LIVE; + flags = VIR_DOMAIN_AFFECT_LIVE; else - flags = VIR_DOMAIN_MEMORY_PARAM_CONFIG; + flags = VIR_DOMAIN_AFFECT_CONFIG; } - if (flags & VIR_DOMAIN_MEMORY_PARAM_LIVE) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) { if (!isActive) { qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); @@ -5119,7 +5119,7 @@ static int qemuDomainGetMemoryParameters(virDomainPtr dom, } } - if (flags & VIR_DOMAIN_MEMORY_PARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { if (!vm->persistent) { qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", _("cannot change persistent config of a transient domain")); @@ -5142,12 +5142,12 @@ static int qemuDomainGetMemoryParameters(virDomainPtr dom, goto cleanup; } - if (flags & VIR_DOMAIN_MEMORY_PARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { for (i = 0; i < *nparams; i++) { virMemoryParameterPtr param = &params[i]; val = 0; param->value.ul = 0; - param->type = VIR_DOMAIN_MEMORY_PARAM_ULLONG; + param->type = VIR_TYPED_PARAM_ULLONG; switch (i) { case 0: /* fill memory hard limit here */ @@ -5269,8 +5269,8 @@ static int qemuSetSchedulerParametersFlags(virDomainPtr dom, int ret = -1; bool isActive; - virCheckFlags(VIR_DOMAIN_SCHEDPARAM_LIVE | - VIR_DOMAIN_SCHEDPARAM_CONFIG, -1); + virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | + VIR_DOMAIN_AFFECT_CONFIG, -1); qemuDriverLock(driver); @@ -5284,20 +5284,20 @@ static int qemuSetSchedulerParametersFlags(virDomainPtr dom, isActive = virDomainObjIsActive(vm); - if (flags == VIR_DOMAIN_SCHEDPARAM_CURRENT) { + if (flags == VIR_DOMAIN_AFFECT_CURRENT) { if (isActive) - flags = VIR_DOMAIN_SCHEDPARAM_LIVE; + flags = VIR_DOMAIN_AFFECT_LIVE; else - flags = VIR_DOMAIN_SCHEDPARAM_CONFIG; + flags = VIR_DOMAIN_AFFECT_CONFIG; } - if ((flags & VIR_DOMAIN_SCHEDPARAM_CONFIG) && !vm->persistent) { + if ((flags & VIR_DOMAIN_AFFECT_CONFIG) && !vm->persistent) { qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", _("cannot change persistent config of a transient domain")); goto cleanup; } - if (flags & VIR_DOMAIN_SCHEDPARAM_LIVE) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) { if (!isActive) { qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", _("domain is not running")); @@ -5328,7 +5328,7 @@ static int qemuSetSchedulerParametersFlags(virDomainPtr dom, goto cleanup; } - if (flags & VIR_DOMAIN_SCHEDPARAM_LIVE) { + if (flags & VIR_DOMAIN_AFFECT_LIVE) { rc = virCgroupSetCpuShares(group, params[i].value.ul); if (rc != 0) { virReportSystemError(-rc, "%s", @@ -5339,7 +5339,7 @@ static int qemuSetSchedulerParametersFlags(virDomainPtr dom, vm->def->cputune.shares = params[i].value.ul; } - if (flags & VIR_DOMAIN_SCHEDPARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { persistentDef = virDomainObjGetPersistentDef(driver->caps, vm); if (!persistentDef) { qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", @@ -5378,7 +5378,7 @@ static int qemuSetSchedulerParameters(virDomainPtr dom, return qemuSetSchedulerParametersFlags(dom, params, nparams, - VIR_DOMAIN_SCHEDPARAM_LIVE); + VIR_DOMAIN_AFFECT_LIVE); } static int @@ -5395,11 +5395,11 @@ qemuGetSchedulerParametersFlags(virDomainPtr dom, int rc; bool isActive; - virCheckFlags(VIR_DOMAIN_SCHEDPARAM_LIVE | - VIR_DOMAIN_SCHEDPARAM_CONFIG, -1); + virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | + VIR_DOMAIN_AFFECT_CONFIG, -1); - if ((flags & (VIR_DOMAIN_SCHEDPARAM_LIVE | VIR_DOMAIN_SCHEDPARAM_CONFIG)) == - (VIR_DOMAIN_SCHEDPARAM_LIVE | VIR_DOMAIN_SCHEDPARAM_CONFIG)) { + if ((flags & (VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG)) == + (VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG)) { qemuReportError(VIR_ERR_INVALID_ARG, "%s", _("cannot query live and config together")); goto cleanup; @@ -5422,14 +5422,14 @@ qemuGetSchedulerParametersFlags(virDomainPtr dom, isActive = virDomainObjIsActive(vm); - if (flags == VIR_DOMAIN_SCHEDPARAM_CURRENT) { + if (flags == VIR_DOMAIN_AFFECT_CURRENT) { if (isActive) - flags = VIR_DOMAIN_SCHEDPARAM_LIVE; + flags = VIR_DOMAIN_AFFECT_LIVE; else - flags = VIR_DOMAIN_SCHEDPARAM_CONFIG; + flags = VIR_DOMAIN_AFFECT_CONFIG; } - if (flags & VIR_DOMAIN_SCHEDPARAM_CONFIG) { + if (flags & VIR_DOMAIN_AFFECT_CONFIG) { if (!vm->persistent) { qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", _("cannot query persistent config of a transient domain")); @@ -5502,7 +5502,7 @@ qemuGetSchedulerParameters(virDomainPtr dom, int *nparams) { return qemuGetSchedulerParametersFlags(dom, params, nparams, - VIR_DOMAIN_SCHEDPARAM_CURRENT); + VIR_DOMAIN_AFFECT_CURRENT); } /* This uses the 'info blockstats' monitor command which was diff --git a/tools/virsh.c b/tools/virsh.c index 5679a2d..d98be1c 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -1732,12 +1732,12 @@ cmdSchedinfo(vshControl *ctl, const vshCmd *cmd) vshError(ctl, "%s", _("--current must be specified exclusively")); return false; } - flags = VIR_DOMAIN_SCHEDPARAM_CURRENT; + flags = VIR_DOMAIN_AFFECT_CURRENT; } else { if (config) - flags |= VIR_DOMAIN_SCHEDPARAM_CONFIG; + flags |= VIR_DOMAIN_AFFECT_CONFIG; if (live) - flags |= VIR_DOMAIN_SCHEDPARAM_LIVE; + flags |= VIR_DOMAIN_AFFECT_LIVE; } if (!vshConnectionUsability(ctl, ctl->conn)) @@ -3505,12 +3505,12 @@ cmdMemtune(vshControl * ctl, const vshCmd * cmd) vshError(ctl, "%s", _("--current must be specified exclusively")); return false; } - flags = VIR_DOMAIN_MEMORY_PARAM_CURRENT; + flags = VIR_DOMAIN_AFFECT_CURRENT; } else { if (config) - flags |= VIR_DOMAIN_MEMORY_PARAM_CONFIG; + flags |= VIR_DOMAIN_AFFECT_CONFIG; if (live) - flags |= VIR_DOMAIN_MEMORY_PARAM_LIVE; + flags |= VIR_DOMAIN_AFFECT_LIVE; } if (!vshConnectionUsability(ctl, ctl->conn)) -- 1.7.4.4

3 3