- Devel - libvirt List Archives

[libvirt] [PATCH 0/7] Remove overhead of disabled logging calls
by Daniel P. Berrange 10 Mar '14

10 Mar '14

Looking at the performance of the libvirtd event loop when used with libvirt sandbox and apps which do lots of I/O I discovered that a huge amount of time was being spent in our logging APIs making the event loop quite poor at dealing with frequently firing poll events for small data sizes. To track this down I created the following short demo program which simulates a libvirtd event loop thread, watching 500 FDs (eg 500 QEMU or LXC console or monitor FDs). In the test no logging outputs are enabled, so ideally the log overhead would be near zero. Running the demo for 51200 iterations initially took 1 minute 40 seconds. oprofile traces show that in this test run the logging APIs were totally dominant in CPU time, over the work done in the event loop. Note the demo program's FD handler doesn't actually do any work, so this is showing the worst case behaviour, but it is none-the-less a good illustration of the loggin overhead. After refactoring the logging code, the overhead of logging calls when no outputs are enabled is reduced to a single function call which does 2 integer comparisons without any mutex locks. The demo program now completes in 3.3 seconds. IOW, 97% of the original CPU time for the demo was consumed by the logging APIs ! The main feature we loose in this series is the global logging buffer. This ring buffer currently collects all log mesages regardless of log filter settings, and dumps them to stderr upon crash of libvirtd. As the amount of logging code in libvirt has grown, the signal/noise ratio of data in the global log buffer has gone down significantly. Whenever libvirtd crashes, my first reaction to ask the reporter to reproduce with a specifically tailored set of log filters set. Thus, IMHO, the global log buffer is no longer a compelling enough feature to justify its significant performance overhead. Rather than remove it entirely though, I've changed it in two parts. The first patch removes the overhead of it, but leaves a more tailored verson of it still present. This removes the contribution it makes to performance overheads. The last patch, which is not performance relevant, simply deletes the rest of the global log buffer entirely. I'm mostly ambivalent on whether to apply patch 7 or not. IMHO the global log buffer is mostly useless once patch 1 is applied, but if people think it is worth keeping it, patch 7 can be dropped and I won't mind. The demo program used is $ cat demo.c #include <libvirt/libvirt.h> #include <unistd.h> #include <stdio.h> static void ignore(int handle, int fd, int events, void *opaque) { // nada } #define STEPS (1024 * 5) #define ITERATIONS (STEPS * 10) #define NFILES 500 int main(int argc, char **argv) { int fds[NFILES][2]; int i; virEventRegisterDefaultImpl(); for (i = 0 ; i < NFILES ; i++) { char c = '0'; if (pipe(fds[i]) < 0) { perror("pipe"); return -1; } if (write(fds[i][1], &c, 1) < 0) return -1; virEventAddHandle(fds[i][0], VIR_EVENT_HANDLE_READABLE, ignore, NULL, NULL); } for (i = 0 ; i < ITERATIONS ; i++) { if (!(i % STEPS)) fprintf(stderr, "."); virEventRunDefaultImpl(); } return 0; } Simply compile & run it with no args $ gcc -Wall `pkg-config --cflags --libs libvirt` -o demo demo.c $ time ./demo .......... real 1m39.981s user 1m34.894s sys 0m4.851s Now make it use the newly compiled libvirt version $ export LD_LIBRARY_PATH=`pwd`/src/.libs $ time ./demo .......... real 0m3.316s user 0m2.896s sys 0m0.412s 97% execution time removed ! Daniel P. Berrange (7): Reduce performance overhead of the global log buffer Refactor code that skips logging of error messages Move dtrace probe macros into separate header file Turn virLogSource into a struct instead of an enum Add virLogSource variables to all source files Switch to filtering based on log source name instead of filename Remove global log buffer feature entirely daemon/libvirtd-config.c | 5 +- daemon/libvirtd-config.h | 1 - daemon/libvirtd.aug | 1 - daemon/libvirtd.c | 5 +- daemon/libvirtd.conf | 7 - daemon/libvirtd.h | 1 - daemon/remote.c | 3 + daemon/stream.c | 2 + daemon/test_libvirtd.aug.in | 1 - docs/apibuild.py | 30 +++ docs/logging.html.in | 10 - src/access/viraccessdriverpolkit.c | 3 + src/access/viraccessmanager.c | 3 + src/bhyve/bhyve_command.c | 2 + src/bhyve/bhyve_driver.c | 2 + src/bhyve/bhyve_process.c | 2 + src/conf/domain_audit.c | 2 + src/conf/domain_conf.c | 2 + src/conf/domain_event.c | 1 + src/conf/network_event.c | 2 + src/conf/nwfilter_params.c | 2 + src/conf/object_event.c | 2 + src/conf/secret_conf.c | 2 + src/conf/snapshot_conf.c | 2 + src/conf/storage_conf.c | 1 - src/conf/virchrdev.c | 2 + src/cpu/cpu.c | 2 + src/cpu/cpu_powerpc.c | 2 + src/cpu/cpu_x86.c | 2 + src/datatypes.c | 2 + src/driver.c | 2 + src/esx/esx_device_monitor.c | 1 - src/esx/esx_driver.c | 2 + src/esx/esx_interface_driver.c | 2 - src/esx/esx_network_driver.c | 2 - src/esx/esx_nwfilter_driver.c | 2 - src/esx/esx_secret_driver.c | 2 - src/esx/esx_storage_backend_iscsi.c | 2 - src/esx/esx_storage_backend_vmfs.c | 2 + src/esx/esx_util.c | 2 +- src/esx/esx_vi.c | 2 +- src/esx/esx_vi_methods.c | 1 - src/esx/esx_vi_types.c | 2 +- src/fdstream.c | 2 + src/hyperv/hyperv_device_monitor.c | 2 - src/hyperv/hyperv_driver.c | 3 +- src/hyperv/hyperv_interface_driver.c | 2 - src/hyperv/hyperv_network_driver.c | 2 - src/hyperv/hyperv_nwfilter_driver.c | 2 - src/hyperv/hyperv_secret_driver.c | 2 - src/hyperv/hyperv_storage_driver.c | 2 - src/hyperv/hyperv_util.c | 2 +- src/hyperv/hyperv_wmi.c | 2 - src/interface/interface_backend_netcf.c | 2 + src/internal.h | 74 ------ src/libvirt-lxc.c | 2 + src/libvirt-qemu.c | 2 + src/libvirt.c | 2 + src/libxl/libxl_conf.c | 2 + src/libxl/libxl_domain.c | 1 + src/libxl/libxl_driver.c | 2 + src/locking/domain_lock.c | 2 + src/locking/lock_daemon.c | 4 +- src/locking/lock_daemon_config.c | 4 +- src/locking/lock_daemon_config.h | 1 - src/locking/lock_daemon_dispatch.c | 5 +- src/locking/lock_driver_lockd.c | 2 + src/locking/lock_driver_nop.c | 2 + src/locking/lock_driver_sanlock.c | 2 + src/locking/lock_manager.c | 2 + src/locking/test_virtlockd.aug.in | 2 - src/locking/virtlockd.aug | 1 - src/locking/virtlockd.conf | 7 - src/lxc/lxc_cgroup.c | 2 + src/lxc/lxc_conf.c | 2 + src/lxc/lxc_container.c | 2 + src/lxc/lxc_controller.c | 2 + src/lxc/lxc_domain.c | 2 + src/lxc/lxc_driver.c | 1 + src/lxc/lxc_fuse.c | 1 - src/lxc/lxc_hostdev.c | 2 + src/lxc/lxc_monitor.c | 2 + src/lxc/lxc_native.c | 1 + src/lxc/lxc_process.c | 2 + src/network/bridge_driver.c | 2 + src/network/bridge_driver_linux.c | 3 + src/network/bridge_driver_platform.h | 1 - src/node_device/node_device_driver.c | 2 - src/node_device/node_device_hal.c | 2 + src/node_device/node_device_linux_sysfs.c | 2 + src/node_device/node_device_udev.c | 4 +- src/nodeinfo.c | 1 - src/nwfilter/nwfilter_dhcpsnoop.c | 2 + src/nwfilter/nwfilter_driver.c | 2 + src/nwfilter/nwfilter_ebiptables_driver.c | 2 + src/nwfilter/nwfilter_gentech_driver.c | 1 + src/nwfilter/nwfilter_learnipaddr.c | 2 + src/openvz/openvz_driver.c | 2 + src/parallels/parallels_driver.c | 2 + src/phyp/phyp_driver.c | 3 + src/qemu/qemu_agent.c | 2 + src/qemu/qemu_bridge_filter.c | 1 - src/qemu/qemu_capabilities.c | 4 +- src/qemu/qemu_cgroup.c | 2 + src/qemu/qemu_command.c | 2 + src/qemu/qemu_conf.c | 2 + src/qemu/qemu_domain.c | 2 + src/qemu/qemu_driver.c | 2 + src/qemu/qemu_hostdev.c | 2 + src/qemu/qemu_hotplug.c | 3 + src/qemu/qemu_migration.c | 2 + src/qemu/qemu_monitor.c | 3 + src/qemu/qemu_monitor_json.c | 3 + src/qemu/qemu_monitor_text.c | 3 + src/qemu/qemu_process.c | 2 + src/remote/remote_driver.c | 2 + src/rpc/virkeepalive.c | 3 + src/rpc/virnetclient.c | 3 + src/rpc/virnetclientprogram.c | 2 + src/rpc/virnetclientstream.c | 2 + src/rpc/virnetmessage.c | 2 + src/rpc/virnetsaslcontext.c | 2 + src/rpc/virnetserver.c | 47 +--- src/rpc/virnetserverclient.c | 3 + src/rpc/virnetservermdns.c | 2 + src/rpc/virnetserverprogram.c | 2 + src/rpc/virnetsocket.c | 2 + src/rpc/virnetsshsession.c | 2 + src/rpc/virnettlscontext.c | 3 + src/secret/secret_driver.c | 2 + src/security/security_apparmor.c | 3 + src/security/security_dac.c | 3 + src/security/security_driver.c | 2 + src/security/security_manager.c | 1 + src/security/security_selinux.c | 2 + src/storage/storage_backend.c | 2 + src/storage/storage_backend_disk.c | 2 + src/storage/storage_backend_fs.c | 2 + src/storage/storage_backend_gluster.c | 2 + src/storage/storage_backend_iscsi.c | 2 + src/storage/storage_backend_logical.c | 2 + src/storage/storage_backend_mpath.c | 2 + src/storage/storage_backend_rbd.c | 2 + src/storage/storage_backend_scsi.c | 2 + src/storage/storage_backend_sheepdog.c | 1 - src/storage/storage_driver.c | 2 + src/test/test_driver.c | 2 + src/uml/uml_conf.c | 1 + src/uml/uml_driver.c | 2 + src/util/viralloc.c | 2 + src/util/virarch.c | 2 + src/util/viraudit.c | 9 +- src/util/viraudit.h | 10 +- src/util/virauth.c | 2 + src/util/virauthconfig.c | 1 + src/util/vircgroup.c | 2 + src/util/virclosecallbacks.c | 2 + src/util/vircommand.c | 2 + src/util/virconf.c | 2 + src/util/virdbus.c | 2 + src/util/virdnsmasq.c | 3 + src/util/virebtables.c | 2 + src/util/virerror.c | 16 +- src/util/virevent.c | 2 + src/util/vireventpoll.c | 3 + src/util/virfile.c | 2 + src/util/virhash.c | 2 + src/util/virhook.c | 2 + src/util/viridentity.c | 1 + src/util/viriptables.c | 2 + src/util/virjson.c | 1 + src/util/virkeyfile.c | 2 + src/util/virlockspace.c | 2 + src/util/virlog.c | 379 ++++-------------------------- src/util/virlog.h | 49 ++-- src/util/virnetdev.c | 2 + src/util/virnetdevmacvlan.c | 1 + src/util/virnetdevtap.c | 2 + src/util/virnetdevveth.c | 2 + src/util/virnetdevvportprofile.c | 2 + src/util/virnetlink.c | 2 + src/util/virnodesuspend.c | 2 + src/util/virnuma.c | 2 + src/util/virobject.c | 3 + src/util/virpci.c | 2 + src/util/virpidfile.c | 2 + src/util/virprobe.h | 100 ++++++++ src/util/virprocess.c | 2 + src/util/virrandom.c | 2 + src/util/virscsi.c | 1 - src/util/virstoragefile.c | 2 + src/util/virstring.c | 2 + src/util/virsysinfo.c | 1 - src/util/virsystemd.c | 1 + src/util/virusb.c | 2 + src/util/virutil.c | 2 + src/util/viruuid.c | 2 + src/vbox/vbox_MSCOMGlue.c | 2 + src/vbox/vbox_XPCOMCGlue.c | 1 + src/vbox/vbox_driver.c | 1 + src/vbox/vbox_tmpl.c | 3 + src/vmx/vmx.c | 2 + src/xen/xen_driver.c | 3 + src/xen/xen_hypervisor.c | 2 + src/xen/xen_inotify.c | 2 + src/xen/xend_internal.c | 2 + src/xen/xm_internal.c | 2 + src/xen/xs_internal.c | 2 + src/xenapi/xenapi_utils.c | 2 + src/xenxs/xen_sxpr.c | 2 + tests/domainconftest.c | 2 + tests/eventtest.c | 2 + tests/fdstreamtest.c | 2 + tests/libvirtdconftest.c | 2 + tests/qemumonitortestutils.c | 2 + tests/securityselinuxlabeltest.c | 2 + tests/securityselinuxtest.c | 2 + tests/sockettest.c | 2 + tests/testutils.c | 4 +- tests/virauthconfigtest.c | 2 + tests/vircgrouptest.c | 2 + tests/virdbustest.c | 2 + tests/virdrivermoduletest.c | 2 + tests/virhashtest.c | 2 + tests/viridentitytest.c | 1 + tests/virkeycodetest.c | 1 + tests/virkeyfiletest.c | 1 + tests/virlockspacetest.c | 2 + tests/virnetmessagetest.c | 2 + tests/virnetsockettest.c | 2 + tests/virnettlscontexttest.c | 2 + tests/virnettlshelpers.c | 2 + tests/virnettlssessiontest.c | 2 + tests/virportallocatortest.c | 1 + tests/virstoragetest.c | 2 + tests/virstringtest.c | 2 + tests/virsystemdtest.c | 2 + tests/virtimetest.c | 2 + tests/viruritest.c | 2 + tools/virsh-console.c | 2 + 240 files changed, 637 insertions(+), 563 deletions(-) create mode 100644 src/util/virprobe.h -- 1.8.5.3

4 26

[libvirt] limit downtime during life migration from xl/virsh
by Olaf Hering 10 Mar '14

10 Mar '14

During live migration of VMs from one host to another the VM is suspended for an unpredictable amount of time. The actual downtime depends on how many new pages will be dirty and the band width to the destination host. Since VM memory size grows faster than transfer rates the currently available tuneables will cause troubles for workloads within the VM which can not handle large timejumps. I have already written code to tweak the inner loop doing the actual migration work in libxc. But the patchset exposes the details of the loop to the cmdline, as such it is not portable nor is it a friendly UI for the hostadmin. Here is my proposal for a new option for virsh and 2 new options for xl: [xl | virsh --live] --max-suspend-time N --timeout N VM host --max-suspend-time N: as the name suggests, the VM downtime must not be longer than specified. The code doing the migration has to estimate the transfer speed. If the VM is about to be suspended, it has to check if the remaining dirty pages can be transfered within the required timeframe. If not, the migration is aborted, the VM continues to run on the src host, the new VM on the dst host is destroyed and an error is returned. --timeout N: if a VM is busy and its workload causes many new dirty pages the migrate command would take forever. This option is supposed to stop the migration attempt if the number of new dirty pages is too high. It would change the semantics of "virsh migrate --timeout n", which currently forces a suspend (according to the help text). I'm not sure if its acceptable to add this option just for the libxl (and maybe xend) target in libvirt, until someone steps up to do also the kvm part. For Xen it would be added for xl only, obviously. Olaf

3 4

[libvirt] QEMU participating in Outreach Program for Women
by Stefan Hajnoczi 10 Mar '14

10 Mar '14

I am delighted to announce that we are participating in Outreach Program for Women! The same QEMU, KVM, and libvirt project ideas are shared with Google Summer of Code. Outreach Program for Women was started by GNOME specifically for women who want to start contributing to open source. It has since grown to include other organizations and we are able to take part this summer! If you would like to apply or know someone who might be interested in a 12-week, paid, open source gig from May 19 to August 18, take a look at our wiki: http://qemu-project.org/Outreach_Program_for_Women_Summer_2014 The application period starts today (March 10th) and the deadline is March 19th. Join the #qemu-gsoc IRC channel on irc.oftc.net to chat with mentors or ask questions. Also see the official OPW page for an overview of how it works: https://wiki.gnome.org/OutreachProgramForWomen/ Stefan

1 0

[libvirt] [PATCH 0/8] Misc ebtables cleanup patches
by Daniel P. Berrange 10 Mar '14

10 Mar '14

In doing some work on the firewall code in libvirt I came across lots of cruft which could usefully be cleaned up.. Daniel P. Berrange (8): Remove many decls from bridge driver platform header Remove decl of method which doesn't exist in virebtables.h Make ebtablesForwardPolicyReject static Remove unused variables from ebtablesContext Remove data structure holding list of ebtables rules Remove worthless ebtRules data structure Remove unused ebtablesRemoveForwardPolicyReject method Remove broken error reporting in QEMU mac filtering src/Makefile.am | 4 +- src/network/bridge_driver_linux.c | 42 ++++--- src/network/bridge_driver_nop.c | 42 ------- src/network/bridge_driver_platform.h | 22 ---- src/qemu/qemu_bridge_filter.c | 104 ---------------- src/qemu/qemu_bridge_filter.h | 37 ------ src/qemu/qemu_command.c | 11 +- src/qemu/qemu_conf.c | 1 - src/qemu/qemu_driver.c | 7 +- src/qemu/qemu_hotplug.c | 11 +- src/qemu/qemu_process.c | 10 +- src/util/virebtables.c | 233 +++-------------------------------- src/util/virebtables.h | 24 ---- 13 files changed, 60 insertions(+), 488 deletions(-) delete mode 100644 src/qemu/qemu_bridge_filter.c delete mode 100644 src/qemu/qemu_bridge_filter.h -- 1.8.5.3

2 9

[libvirt] [PATCH] libvirt-tck: Update hook syntax for libvirt 0.9.0+
by Mike Latimer 10 Mar '14

10 Mar '14

Starting with libvirt 0.9.0+, hook scripts can be called from several new locations. These locations must also be reflected in the expected logs of the hook tests. The final test in 052-domain-hook.t intentionally produces a failed start, which should show the first stage in the startup hook process, followed by all the stages in the destroy process. In addition to the above changes, libvirtd init scripts can return daemon status in several different ways (due to distribution differences, sysvinit vs. systemd, etc). This patch allows for 'running|active', and 'stopped|unused|inactive' responses which the hook tests rely on. --- lib/Sys/Virt/TCK/Hooks.pm | 17 ++++++++++++----- scripts/hooks/051-daemon-hook.t | 2 +- scripts/hooks/052-domain-hook.t | 6 +++--- 3 files changed, 16 insertions(+), 9 deletions(-) diff --git a/lib/Sys/Virt/TCK/Hooks.pm b/lib/Sys/Virt/TCK/Hooks.pm index 75f98e9..2ae2c2a 100644 --- a/lib/Sys/Virt/TCK/Hooks.pm +++ b/lib/Sys/Virt/TCK/Hooks.pm @@ -73,10 +73,10 @@ sub libvirtd_status { my $status = `service libvirtd status`; my $_ = $status; - if (/running/) { - $self->{libvirtd_status} = 'running'; - } elsif (/stopped/) { + if (/stopped|unused|inactive/) { $self->{libvirtd_status} = 'stopped'; + } elsif (/running|active/) { + $self->{libvirtd_status} = 'running'; } return $self; @@ -146,13 +146,20 @@ sub expect_log { } elsif ($self->{type} eq 'qemu' or $self->{type} eq 'lxc') { if ($domain_state eq &Sys::Virt::Domain::STATE_RUNNING) { if ($action eq 'destroy') { - $expect_log = "$hook $domain_name stopped end -"; + $expect_log = "$hook $domain_name stopped end -\n". + "$hook $domain_name release end -"; } else { die "hooks testing doesn't support $action running domain"; } } elsif ($domain_state eq &Sys::Virt::Domain::STATE_SHUTOFF) { if ($action eq 'start') { - $expect_log = "$hook $domain_name start begin -"; + $expect_log = "$hook $domain_name prepare begin -\n". + "$hook $domain_name start begin -\n". + "$hook $domain_name started begin -"; + } elsif ($action eq 'failstart') { + $expect_log = "$hook $domain_name prepare begin -\n". + "$hook $domain_name stopped end -\n". + "$hook $domain_name release end -"; } else { die "hooks testing doesn't support $action shutoff domain"; } diff --git a/scripts/hooks/051-daemon-hook.t b/scripts/hooks/051-daemon-hook.t index d9cfb3a..165cf4e 100644 --- a/scripts/hooks/051-daemon-hook.t +++ b/scripts/hooks/051-daemon-hook.t @@ -85,7 +85,7 @@ SKIP: { ok($hook->compare_log(), "$hook->{name} is invoked correctly while $hook->{action} libvirtd"); diag "check if libvirtd is stopped"; - ok(`service libvirtd status` =~ /stopped/, "libvirtd is stopped"); + ok(`service libvirtd status` =~ /stopped|unused|inactive/, "libvirtd is stopped"); # start libvirtd $hook->action('start'); diff --git a/scripts/hooks/052-domain-hook.t b/scripts/hooks/052-domain-hook.t index e3b55ec..90b8a48 100644 --- a/scripts/hooks/052-domain-hook.t +++ b/scripts/hooks/052-domain-hook.t @@ -90,7 +90,7 @@ SKIP: { ok(-f "$hook->{name}", "$hook->{name} is invoked"); my $actual_log_data = slurp($hook->{log_name}); - diag "acutal log: $hook->{log_name} '$actual_log_data'"; + diag "actual log: $hook->{log_name} '$actual_log_data'"; diag "expect log:\n $hook->{expect_log}"; @@ -147,7 +147,7 @@ SKIP: { $hook->domain_name($domain_name); $hook->domain_state($domain_state); - $hook->action('start'); + $hook->action('failstart'); $hook->expect_log(); diag "start $domain_name"; @@ -170,7 +170,7 @@ SKIP: { diag "expect log:\n $hook->{expect_log}"; diag "check if the actual log is same with expected log"; - ok($hook->compare_log, "$hook->{name} is invoked correctly while start $domain_name"); + ok($hook->compare_log, "$hook->{name} is invoked correctly while failing to start $domain_name"); # undefine domain diag "undefine $domain_name"; -- 1.8.4.5

3 4

[libvirt] [PATCH v2] qemu: Advertise <drivediscard/> in capabilities if 'discard' is supported.
by Richard W.M. Jones 10 Mar '14

10 Mar '14

The capabilities will have a <drivediscard/> feature if you are allowed to use the qemu attribute: <driver name='qemu' ... discard='unmap|ignore' /> The new capabilities XML looks like this: <capabilities> ... <guest> <os_type>hvm</os_type> ... <features> <drivediscard/> </features> </guest> </capabilities> --- src/conf/capabilities.c | 3 ++- src/qemu/qemu_capabilities.c | 4 ++++ 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/src/conf/capabilities.c b/src/conf/capabilities.c index c1c4ab8..ff5912a 100644 --- a/src/conf/capabilities.c +++ b/src/conf/capabilities.c @@ -947,7 +947,8 @@ virCapabilitiesFormatXML(virCapsPtr caps) STREQ(caps->guests[i]->features[j]->name, "nonpae") || STREQ(caps->guests[i]->features[j]->name, "ia64_be") || STREQ(caps->guests[i]->features[j]->name, "cpuselection") || - STREQ(caps->guests[i]->features[j]->name, "deviceboot")) { + STREQ(caps->guests[i]->features[j]->name, "deviceboot") || + STREQ(caps->guests[i]->features[j]->name, "drivediscard")) { virBufferAsprintf(&xml, " <%s/>\n", caps->guests[i]->features[j]->name); } else { diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index cae25e0..b755fb2 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -778,6 +778,10 @@ virQEMUCapsInitGuest(virCapsPtr caps, !virCapabilitiesAddGuestFeature(guest, "deviceboot", 1, 0)) goto error; + if (virQEMUCapsGet(qemubinCaps, QEMU_CAPS_DRIVE_DISCARD) && + !virCapabilitiesAddGuestFeature(guest, "drivediscard", 0, 0)) + goto error; + if (virCapabilitiesAddGuestDomain(guest, "qemu", NULL, -- 1.8.5.3

1 1

[libvirt] [PATCH] qemu: Advertise <qcow2discard/> in capabilities if discard works for qcow2 backing files.
by Richard W.M. Jones 10 Mar '14

10 Mar '14

Qemu since 1.5 supports discard. However until qemu 1.6 it didn't work for qcow2 files. Ref: http://wiki.qemu.org/ChangeLog/1.5 http://wiki.qemu.org/ChangeLog/1.6 If qemu >= 1.6, add this to guest capabilities: <features> <drivediscard/> <qcow2discard/> <--- </features> --- src/conf/capabilities.c | 3 ++- src/qemu/qemu_capabilities.c | 13 +++++++++++++ src/qemu/qemu_capabilities.h | 1 + tests/qemucapabilitiesdata/caps_1.6.0-1.caps | 1 + tests/qemucapabilitiesdata/caps_1.6.50-1.caps | 1 + 5 files changed, 18 insertions(+), 1 deletion(-) diff --git a/src/conf/capabilities.c b/src/conf/capabilities.c index ff5912a..08c9b73 100644 --- a/src/conf/capabilities.c +++ b/src/conf/capabilities.c @@ -948,7 +948,8 @@ virCapabilitiesFormatXML(virCapsPtr caps) STREQ(caps->guests[i]->features[j]->name, "ia64_be") || STREQ(caps->guests[i]->features[j]->name, "cpuselection") || STREQ(caps->guests[i]->features[j]->name, "deviceboot") || - STREQ(caps->guests[i]->features[j]->name, "drivediscard")) { + STREQ(caps->guests[i]->features[j]->name, "drivediscard") || + STREQ(caps->guests[i]->features[j]->name, "qcow2discard")) { virBufferAsprintf(&xml, " <%s/>\n", caps->guests[i]->features[j]->name); } else { diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index b755fb2..b527690 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -251,6 +251,7 @@ VIR_ENUM_IMPL(virQEMUCaps, QEMU_CAPS_LAST, "spiceport", "usb-kbd", /* 165 */ + "qcow2-discard", ); struct _virQEMUCaps { @@ -782,6 +783,10 @@ virQEMUCapsInitGuest(virCapsPtr caps, !virCapabilitiesAddGuestFeature(guest, "drivediscard", 0, 0)) goto error; + if (virQEMUCapsGet(qemubinCaps, QEMU_CAPS_QCOW2_DISCARD) && + !virCapabilitiesAddGuestFeature(guest, "qcow2discard", 0, 0)) + goto error; + if (virCapabilitiesAddGuestDomain(guest, "qemu", NULL, @@ -1113,6 +1118,10 @@ virQEMUCapsComputeCmdFlags(const char *help, if (strstr(help, "-machine")) virQEMUCapsSet(qemuCaps, QEMU_CAPS_MACHINE_OPT); + /* discard option works for qcow2 backing files since 1.6 */ + if (qemuCaps->version >= 1006000) + virQEMUCapsSet(qemuCaps, QEMU_CAPS_QCOW2_DISCARD); + /* USB option is supported v1.3.0 onwards */ if (qemuCaps->version >= 1003000) virQEMUCapsSet(qemuCaps, QEMU_CAPS_MACHINE_USB_OPT); @@ -2569,6 +2578,10 @@ virQEMUCapsInitQMPMonitor(virQEMUCapsPtr qemuCaps, if (virQEMUCapsInitArchQMPBasic(qemuCaps, mon) < 0) goto cleanup; + /* discard option works for qcow2 backing files since 1.6 */ + if (qemuCaps->version >= 1006000) + virQEMUCapsSet(qemuCaps, QEMU_CAPS_QCOW2_DISCARD); + /* USB option is supported v1.3.0 onwards */ if (qemuCaps->version >= 1003000) virQEMUCapsSet(qemuCaps, QEMU_CAPS_MACHINE_USB_OPT); diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index b5445e7..a0ce506 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -204,6 +204,7 @@ enum virQEMUCapsFlags { QEMU_CAPS_SPICE_FILE_XFER_DISABLE = 163, /* -spice disable-agent-file-xfer */ QEMU_CAPS_CHARDEV_SPICEPORT = 164, /* -chardev spiceport */ QEMU_CAPS_DEVICE_USB_KBD = 165, /* -device usb-kbd */ + QEMU_CAPS_QCOW2_DISCARD = 166, /* discard works for qcow2 backing files */ QEMU_CAPS_LAST, /* this must always be the last item */ }; diff --git a/tests/qemucapabilitiesdata/caps_1.6.0-1.caps b/tests/qemucapabilitiesdata/caps_1.6.0-1.caps index 597f873..0ab5e22 100644 --- a/tests/qemucapabilitiesdata/caps_1.6.0-1.caps +++ b/tests/qemucapabilitiesdata/caps_1.6.0-1.caps @@ -142,4 +142,5 @@ <flag name='spice-file-xfer-disable'/> <flag name='spiceport'/> <flag name='usb-kbd'/> + <flag name='qcow2-discard'/> </qemuCaps> diff --git a/tests/qemucapabilitiesdata/caps_1.6.50-1.caps b/tests/qemucapabilitiesdata/caps_1.6.50-1.caps index 0c1dd87..d5cca41 100644 --- a/tests/qemucapabilitiesdata/caps_1.6.50-1.caps +++ b/tests/qemucapabilitiesdata/caps_1.6.50-1.caps @@ -140,4 +140,5 @@ <flag name='spice-file-xfer-disable'/> <flag name='spiceport'/> <flag name='usb-kbd'/> + <flag name='qcow2-discard'/> </qemuCaps> -- 1.8.5.3

1 1

[libvirt] [PATCH] Change file names in comments to match the files they are in
by Ján Tomko 10 Mar '14

10 Mar '14

Some of these are leftovers from renaming the files, others are just typos. Also introduce an ugly awk script to enforce this. --- cfg.mk | 25 +++++++++++++++++++++++++ daemon/libvirtd-config.c | 2 +- daemon/libvirtd-config.h | 2 +- src/bhyve/bhyve_command.c | 2 +- src/bhyve/bhyve_command.h | 2 +- src/conf/domain_audit.h | 2 +- src/datatypes.c | 2 +- src/hyperv/hyperv_wmi.c | 2 +- src/interface/interface_backend_netcf.c | 4 ++-- src/locking/domain_lock.h | 2 +- src/locking/lock_daemon_config.c | 2 +- src/lxc/lxc_hostdev.c | 2 +- src/lxc/lxc_hostdev.h | 2 +- src/network/bridge_driver.h | 2 +- src/network/bridge_driver_platform.h | 2 +- src/node_device/node_device_driver.c | 2 +- src/node_device/node_device_driver.h | 2 +- src/nodeinfo.h | 2 +- src/openvz/openvz_conf.h | 2 +- src/openvz/openvz_util.c | 2 +- src/openvz/openvz_util.h | 2 +- src/parallels/parallels_driver.h | 2 +- src/parallels/parallels_network.c | 2 +- src/qemu/qemu_agent.c | 2 +- src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_driver.h | 2 +- src/qemu/qemu_hotplug.c | 2 +- src/remote/remote_driver.h | 2 +- src/rpc/virnetservermdns.h | 2 +- src/storage/storage_backend_mpath.h | 2 +- src/storage/storage_backend_sheepdog.h | 2 +- src/test/test_driver.c | 2 +- src/test/test_driver.h | 2 +- src/uml/uml_conf.h | 2 +- src/util/viratomic.c | 2 +- src/util/virbitmap.c | 2 +- src/util/virebtables.h | 2 +- src/util/virerror.h | 2 +- src/util/vireventpoll.h | 2 +- src/util/viruuid.c | 2 +- src/util/virxml.h | 2 +- src/xen/xen_driver.h | 2 +- src/xen/xen_hypervisor.c | 2 +- src/xen/xen_hypervisor.h | 2 +- src/xen/xen_inotify.c | 2 +- src/xen/xen_inotify.h | 2 +- src/xen/xm_internal.c | 2 +- src/xenapi/xenapi_driver.h | 2 +- tests/testutils.h | 2 +- tools/virsh-interface.h | 2 +- tools/virsh-network.h | 2 +- tools/virt-host-validate.c | 2 +- 52 files changed, 77 insertions(+), 52 deletions(-) diff --git a/cfg.mk b/cfg.mk index 2a8957a..c5f99aa 100644 --- a/cfg.mk +++ b/cfg.mk @@ -863,6 +863,31 @@ sc_prohibit_atoi: halt='Use virStrToLong* instead of atoi, atol, atof, atoq, atoll' \ $(_sc_search_regexp) +# Don't +sc_prohibit_wrong_filename_in_comment: + @fail=0; \ + awk 'BEGIN { \ + fail=0; \ + } FNR < 3 { \ + n=match($$0, /[[:space:]][^[:space:]]*[.][ch][[:space:]:]/); \ + if (n > 0) { \ + A=substr($$0, RSTART+1, RLENGTH-2); \ + n=split(FILENAME, arr, "/"); \ + if (A != arr[n]) { \ + print "in " FILENAME ": " A " mentioned in comments "; \ + fail=1; \ + } \ + } \ + } END { \ + if (fail == 1) { \ + exit 1; \ + } \ + }' $$($(VC_LIST_EXCEPT) | grep '\.[ch]$$') || fail=1; \ + if test $$fail -eq 1; then \ + { echo '$(ME): The file name in comments must match the actual ' \ + 'file name' 1>&2; exit 1; } \ + fi; + # We don't use this feature of maint.mk. prev_version_file = /dev/null diff --git a/daemon/libvirtd-config.c b/daemon/libvirtd-config.c index c816fda..856a038 100644 --- a/daemon/libvirtd-config.c +++ b/daemon/libvirtd-config.c @@ -1,5 +1,5 @@ /* - * libvirtd.c: daemon start of day, guest process & i/o management + * libvirtd-config.c: daemon start of day, guest process & i/o management * * Copyright (C) 2006-2012 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/daemon/libvirtd-config.h b/daemon/libvirtd-config.h index a24d5d2..0829273 100644 --- a/daemon/libvirtd-config.h +++ b/daemon/libvirtd-config.h @@ -1,5 +1,5 @@ /* - * libvirtd.c: daemon start of day, guest process & i/o management + * libvirtd-config.h: daemon start of day, guest process & i/o management * * Copyright (C) 2006-2012 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/bhyve/bhyve_command.c b/src/bhyve/bhyve_command.c index 1fcc0f2..a19daac 100644 --- a/src/bhyve/bhyve_command.c +++ b/src/bhyve/bhyve_command.c @@ -1,5 +1,5 @@ /* - * bhyve_process.c: bhyve command generation + * bhyve_command.c: bhyve command generation * * Copyright (C) 2014 Roman Bogorodskiy * diff --git a/src/bhyve/bhyve_command.h b/src/bhyve/bhyve_command.h index 8326971..66d934d 100644 --- a/src/bhyve/bhyve_command.h +++ b/src/bhyve/bhyve_command.h @@ -1,5 +1,5 @@ /* - * bhyve_process.c: bhyve command generation + * bhyve_command.h: bhyve command generation * * Copyright (C) 2014 Roman Bogorodskiy * diff --git a/src/conf/domain_audit.h b/src/conf/domain_audit.h index 9486216..70b09e5 100644 --- a/src/conf/domain_audit.h +++ b/src/conf/domain_audit.h @@ -1,5 +1,5 @@ /* - * domain_audit.c: Domain audit management + * domain_audit.h: Domain audit management * * Copyright (C) 2006-2011 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/datatypes.c b/src/datatypes.c index 33d941d..20752cd 100644 --- a/src/datatypes.c +++ b/src/datatypes.c @@ -1,5 +1,5 @@ /* - * datatypes.h: management of structs for public data types + * datatypes.c: management of structs for public data types * * Copyright (C) 2006-2014 Red Hat, Inc. * diff --git a/src/hyperv/hyperv_wmi.c b/src/hyperv/hyperv_wmi.c index 39eed0c..a603bad 100644 --- a/src/hyperv/hyperv_wmi.c +++ b/src/hyperv/hyperv_wmi.c @@ -1,6 +1,6 @@ /* - * hyperv_wmi.h: general WMI over WSMAN related functions and structures for + * hyperv_wmi.c: general WMI over WSMAN related functions and structures for * managing Microsoft Hyper-V hosts * * Copyright (C) 2011 Matthias Bolte <matthias.bolte(a)googlemail.com> diff --git a/src/interface/interface_backend_netcf.c b/src/interface/interface_backend_netcf.c index c525ca9..b4c1fe9 100644 --- a/src/interface/interface_backend_netcf.c +++ b/src/interface/interface_backend_netcf.c @@ -1,6 +1,6 @@ /* - * interface_driver.c: backend driver methods to handle physical - * interface configuration using the netcf library. + * interface_backend_netcf.c: backend driver methods to handle physical + * interface configuration using the netcf library. * * Copyright (C) 2006-2013 Red Hat, Inc. * diff --git a/src/locking/domain_lock.h b/src/locking/domain_lock.h index eefe6cd..a9b19c8 100644 --- a/src/locking/domain_lock.h +++ b/src/locking/domain_lock.h @@ -1,5 +1,5 @@ /* - * domain_lock.c: Locking for domain lifecycle operations + * domain_lock.h: Locking for domain lifecycle operations * * Copyright (C) 2010-2011 Red Hat, Inc. * diff --git a/src/locking/lock_daemon_config.c b/src/locking/lock_daemon_config.c index 8e632f5..68cd9e9 100644 --- a/src/locking/lock_daemon_config.c +++ b/src/locking/lock_daemon_config.c @@ -1,5 +1,5 @@ /* - * lock_daemon_config.h: virtlockd config file handling + * lock_daemon_config.c: virtlockd config file handling * * Copyright (C) 2006-2012 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/lxc/lxc_hostdev.c b/src/lxc/lxc_hostdev.c index b7248df..c3ecc48 100644 --- a/src/lxc/lxc_hostdev.c +++ b/src/lxc/lxc_hostdev.c @@ -1,5 +1,5 @@ /* - * virLXC_hostdev.c: VIRLXC hostdev management + * lxc_hostdev.c: VIRLXC hostdev management * * Copyright (C) 2006-2007, 2009-2012 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/lxc/lxc_hostdev.h b/src/lxc/lxc_hostdev.h index 41bb178..cf89a99 100644 --- a/src/lxc/lxc_hostdev.h +++ b/src/lxc/lxc_hostdev.h @@ -1,5 +1,5 @@ /* - * virLXC_hostdev.h: VIRLXC hostdev management + * lxc_hostdev.h: VIRLXC hostdev management * * Copyright (C) 2006-2007, 2009-2010 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/network/bridge_driver.h b/src/network/bridge_driver.h index 1c2a4a7..decc08f 100644 --- a/src/network/bridge_driver.h +++ b/src/network/bridge_driver.h @@ -1,5 +1,5 @@ /* - * network_driver.h: core driver methods for managing networks + * bridge_driver.h: core driver methods for managing networks * * Copyright (C) 2006-2013 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/network/bridge_driver_platform.h b/src/network/bridge_driver_platform.h index 82d96f6..22f3aae 100644 --- a/src/network/bridge_driver_platform.h +++ b/src/network/bridge_driver_platform.h @@ -1,5 +1,5 @@ /* - * bridge_driver.h: platform specific routines for bridge driver + * bridge_driver_platform.h: platform specific routines for bridge driver * * Copyright (C) 2006-2013 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/node_device/node_device_driver.c b/src/node_device/node_device_driver.c index d5a92cd..55b3eb3 100644 --- a/src/node_device/node_device_driver.c +++ b/src/node_device/node_device_driver.c @@ -1,5 +1,5 @@ /* - * node_device.c: node device enumeration + * node_device_driver.c: node device enumeration * * Copyright (C) 2010-2013 Red Hat, Inc. * Copyright (C) 2008 Virtual Iron Software, Inc. diff --git a/src/node_device/node_device_driver.h b/src/node_device/node_device_driver.h index 0f62c83..e238ff8 100644 --- a/src/node_device/node_device_driver.h +++ b/src/node_device/node_device_driver.h @@ -1,5 +1,5 @@ /* - * node_device.h: node device enumeration + * node_device_driver.h: node device enumeration * * Copyright (C) 2008 Virtual Iron Software, Inc. * Copyright (C) 2008 David F. Lively diff --git a/src/nodeinfo.h b/src/nodeinfo.h index 413fddd..c81fcbb 100644 --- a/src/nodeinfo.h +++ b/src/nodeinfo.h @@ -1,5 +1,5 @@ /* - * nodeinfo.c: Helper routines for OS specific node information + * nodeinfo.h: Helper routines for OS specific node information * * Copyright (C) 2006-2008, 2011-2012 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/openvz/openvz_conf.h b/src/openvz/openvz_conf.h index 47b5123..a7de7d2 100644 --- a/src/openvz/openvz_conf.h +++ b/src/openvz/openvz_conf.h @@ -1,5 +1,5 @@ /* - * openvz_config.h: config information for OpenVZ VPSs + * openvz_conf.h: config information for OpenVZ VPSs * * Copyright (C) 2010 Red Hat, Inc. * Copyright (C) 2006, 2007 Binary Karma. diff --git a/src/openvz/openvz_util.c b/src/openvz/openvz_util.c index 8920d14..3113751 100644 --- a/src/openvz/openvz_util.c +++ b/src/openvz/openvz_util.c @@ -1,5 +1,5 @@ /* - * openvz_driver.c: core driver methods for managing OpenVZ VEs + * openvz_util.c: core driver methods for managing OpenVZ VEs * * Copyright (C) 2013 Red Hat, Inc. * Copyright (C) 2012 Guido Günther diff --git a/src/openvz/openvz_util.h b/src/openvz/openvz_util.h index 680e507..0c8d55a 100644 --- a/src/openvz/openvz_util.h +++ b/src/openvz/openvz_util.h @@ -1,5 +1,5 @@ /* - * openvz_driver.h: common util functions for managing openvz VPEs + * openvz_util.h: common util functions for managing openvz VPEs * * Copyright (C) 2012 Guido Günther * diff --git a/src/parallels/parallels_driver.h b/src/parallels/parallels_driver.h index a503abe..babe5c0 100644 --- a/src/parallels/parallels_driver.h +++ b/src/parallels/parallels_driver.h @@ -1,5 +1,5 @@ /* - * parallels_driver.c: core driver functions for managing + * parallels_driver.h: core driver functions for managing * Parallels Cloud Server hosts * * Copyright (C) 2012 Parallels, Inc. diff --git a/src/parallels/parallels_network.c b/src/parallels/parallels_network.c index 63daf53..432f0cc 100644 --- a/src/parallels/parallels_network.c +++ b/src/parallels/parallels_network.c @@ -1,5 +1,5 @@ /* - * parallels_storage.c: core privconn functions for managing + * parallels_network.c: core privconn functions for managing * Parallels Cloud Server hosts * * Copyright (C) 2013 Red Hat, Inc. diff --git a/src/qemu/qemu_agent.c b/src/qemu/qemu_agent.c index 4a3820c..553bff9 100644 --- a/src/qemu/qemu_agent.c +++ b/src/qemu/qemu_agent.c @@ -1,5 +1,5 @@ /* - * qemu_agent.h: interaction with QEMU guest agent + * qemu_agent.c: interaction with QEMU guest agent * * Copyright (C) 2006-2013 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index bad63c5..c321eda 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -1,5 +1,5 @@ /* - * qemu_domain.h: QEMU domain private state + * qemu_domain.c: QEMU domain private state * * Copyright (C) 2006-2013 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/qemu/qemu_driver.h b/src/qemu/qemu_driver.h index 84f74c4..df7533a 100644 --- a/src/qemu/qemu_driver.h +++ b/src/qemu/qemu_driver.h @@ -1,5 +1,5 @@ /* - * driver.h: core driver methods for managing qemu guests + * qemu_driver.h: core driver methods for managing qemu guests * * Copyright (C) 2006, 2007 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/qemu/qemu_hotplug.c b/src/qemu/qemu_hotplug.c index 6703c92..dd72a79 100644 --- a/src/qemu/qemu_hotplug.c +++ b/src/qemu/qemu_hotplug.c @@ -1,5 +1,5 @@ /* - * qemu_hotplug.h: QEMU device hotplug management + * qemu_hotplug.c: QEMU device hotplug management * * Copyright (C) 2006-2014 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/remote/remote_driver.h b/src/remote/remote_driver.h index 912f0f8..4033a3c 100644 --- a/src/remote/remote_driver.h +++ b/src/remote/remote_driver.h @@ -1,5 +1,5 @@ /* - * remote_internal.h: driver to provide access to libvirtd running + * remote_driver.h: driver to provide access to libvirtd running * on a remote machine * * Copyright (C) 2006-2007, 2010 Red Hat, Inc. diff --git a/src/rpc/virnetservermdns.h b/src/rpc/virnetservermdns.h index a1c19c1..e91b264 100644 --- a/src/rpc/virnetservermdns.h +++ b/src/rpc/virnetservermdns.h @@ -1,5 +1,5 @@ /* - * virnetservermdns.c: advertise server sockets + * virnetservermdns.h: advertise server sockets * * Copyright (C) 2011 Red Hat, Inc. * Copyright (C) 2007 Daniel P. Berrange diff --git a/src/storage/storage_backend_mpath.h b/src/storage/storage_backend_mpath.h index d7f20f3..b666645 100644 --- a/src/storage/storage_backend_mpath.h +++ b/src/storage/storage_backend_mpath.h @@ -1,5 +1,5 @@ /* - * storage_backend_scsi.h: storage backend for SCSI handling + * storage_backend_mpath.h: storage backend for multipath handling * * Copyright (C) 2009-2009 Red Hat, Inc. * Copyright (C) 2009-2008 Dave Allan diff --git a/src/storage/storage_backend_sheepdog.h b/src/storage/storage_backend_sheepdog.h index bc114ff..b0d8440 100644 --- a/src/storage/storage_backend_sheepdog.h +++ b/src/storage/storage_backend_sheepdog.h @@ -1,5 +1,5 @@ /* - * storage_backend_sheepog.h: storage backend for Sheepdog handling + * storage_backend_sheepdog.h: storage backend for Sheepdog handling * * Copyright (C) 2012 Wido den Hollander * Copyright (C) 2012 Frank Spijkerman diff --git a/src/test/test_driver.c b/src/test/test_driver.c index 6806ffd..27295db 100644 --- a/src/test/test_driver.c +++ b/src/test/test_driver.c @@ -1,5 +1,5 @@ /* - * test.c: A "mock" hypervisor for use by application unit tests + * test_driver.c: A "mock" hypervisor for use by application unit tests * * Copyright (C) 2006-2014 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/test/test_driver.h b/src/test/test_driver.h index b9bf566..df74433 100644 --- a/src/test/test_driver.h +++ b/src/test/test_driver.h @@ -1,5 +1,5 @@ /* - * test.h: A "mock" hypervisor for use by application unit tests + * test_driver.h: A "mock" hypervisor for use by application unit tests * * Copyright (C) 2006-2006 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/uml/uml_conf.h b/src/uml/uml_conf.h index a914be0..05e19ff 100644 --- a/src/uml/uml_conf.h +++ b/src/uml/uml_conf.h @@ -1,5 +1,5 @@ /* - * config.h: VM configuration management + * uml_conf.h: VM configuration management * * Copyright (C) 2006, 2007, 2010 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/util/viratomic.c b/src/util/viratomic.c index 160a20b..278a749 100644 --- a/src/util/viratomic.c +++ b/src/util/viratomic.c @@ -1,5 +1,5 @@ /* - * viratomic.h: atomic integer operations + * viratomic.c: atomic integer operations * * Copyright (C) 2012 Red Hat, Inc. * diff --git a/src/util/virbitmap.c b/src/util/virbitmap.c index 870c8fe..775262d 100644 --- a/src/util/virbitmap.c +++ b/src/util/virbitmap.c @@ -1,5 +1,5 @@ /* - * virbitmap.h: Simple bitmap operations + * virbitmap.c: Simple bitmap operations * * Copyright (C) 2010-2013 Red Hat, Inc. * Copyright (C) 2010 Novell, Inc. diff --git a/src/util/virebtables.h b/src/util/virebtables.h index c0d443d..2b6f1dd 100644 --- a/src/util/virebtables.h +++ b/src/util/virebtables.h @@ -1,5 +1,5 @@ /* - * virebtables.c: Helper APIs for managing ebtables + * virebtables.h: Helper APIs for managing ebtables * * Copyright (C) 2007-2008, 2013 Red Hat, Inc. * Copyright (C) 2009 IBM Corp. diff --git a/src/util/virerror.h b/src/util/virerror.h index 5ae4eb5..2de04f4 100644 --- a/src/util/virerror.h +++ b/src/util/virerror.h @@ -1,5 +1,5 @@ /* - * virerror.c: error handling and reporting code for libvirt + * virerror.h: error handling and reporting code for libvirt * * Copyright (C) 2006-2014 Red Hat, Inc. * diff --git a/src/util/vireventpoll.h b/src/util/vireventpoll.h index c46ab80..8844c9c 100644 --- a/src/util/vireventpoll.h +++ b/src/util/vireventpoll.h @@ -1,5 +1,5 @@ /* - * vireventpool.h: Poll based event loop for monitoring file handles + * vireventpoll.h: Poll based event loop for monitoring file handles * * Copyright (C) 2007 Daniel P. Berrange * Copyright (C) 2007 Red Hat, Inc. diff --git a/src/util/viruuid.c b/src/util/viruuid.c index c5fa9a8..48f5e2e 100644 --- a/src/util/viruuid.c +++ b/src/util/viruuid.c @@ -1,5 +1,5 @@ /* - * viruuid.h: helper APIs for dealing with UUIDs + * viruuid.c: helper APIs for dealing with UUIDs * * Copyright (C) 2007-2013 Red Hat, Inc. * diff --git a/src/util/virxml.h b/src/util/virxml.h index d967a2e..781b3bf 100644 --- a/src/util/virxml.h +++ b/src/util/virxml.h @@ -1,5 +1,5 @@ /* - * virxml.c: helper APIs for dealing with XML documents + * virxml.h: helper APIs for dealing with XML documents * * Copyright (C) 2005, 2007-2012 Red Hat, Inc. * diff --git a/src/xen/xen_driver.h b/src/xen/xen_driver.h index b8c1c27..1ed870d 100644 --- a/src/xen/xen_driver.h +++ b/src/xen/xen_driver.h @@ -1,5 +1,5 @@ /* - * xen_unified.c: Unified Xen driver. + * xen_driver.h: Unified Xen driver. * * Copyright (C) 2007, 2010-2011 Red Hat, Inc. * diff --git a/src/xen/xen_hypervisor.c b/src/xen/xen_hypervisor.c index f9c7b40..d24db06 100644 --- a/src/xen/xen_hypervisor.c +++ b/src/xen/xen_hypervisor.c @@ -1,5 +1,5 @@ /* - * xen_internal.c: direct access to Xen hypervisor level + * xen_hypervisor.c: direct access to Xen hypervisor level * * Copyright (C) 2005-2013 Red Hat, Inc. * diff --git a/src/xen/xen_hypervisor.h b/src/xen/xen_hypervisor.h index 6aeab79..2990e51 100644 --- a/src/xen/xen_hypervisor.h +++ b/src/xen/xen_hypervisor.h @@ -1,5 +1,5 @@ /* - * xen_internal.h: internal API for direct access to Xen hypervisor level + * xen_hypervisor.h: internal API for direct access to Xen hypervisor level * * Copyright (C) 2005, 2010-2011 Red Hat, Inc. * diff --git a/src/xen/xen_inotify.c b/src/xen/xen_inotify.c index 2e9787f..c9718bd 100644 --- a/src/xen/xen_inotify.c +++ b/src/xen/xen_inotify.c @@ -1,5 +1,5 @@ /* - * xen_inofify.c: Xen notification of xml file activity in the + * xen_inotify.c: Xen notification of xml file activity in the * following dirs: * /etc/xen * /var/lib/xend/domains diff --git a/src/xen/xen_inotify.h b/src/xen/xen_inotify.h index 6055c88..8a9370d 100644 --- a/src/xen/xen_inotify.h +++ b/src/xen/xen_inotify.h @@ -1,5 +1,5 @@ /* - * xen_inofify.h: Xen notification of xml files + * xen_inotify.h: Xen notification of xml files * * Copyright (C) 2011 Red Hat, Inc. * Copyright (C) 2008 VirtualIron diff --git a/src/xen/xm_internal.c b/src/xen/xm_internal.c index d9a3502..5d078e7 100644 --- a/src/xen/xm_internal.c +++ b/src/xen/xm_internal.c @@ -1,5 +1,5 @@ /* - * xm_internal.h: helper routines for dealing with inactive domains + * xm_internal.c: helper routines for dealing with inactive domains * * Copyright (C) 2006-2007, 2009-2013 Red Hat, Inc. * Copyright (C) 2006 Daniel P. Berrange diff --git a/src/xenapi/xenapi_driver.h b/src/xenapi/xenapi_driver.h index 30a8f1e..0925731 100644 --- a/src/xenapi/xenapi_driver.h +++ b/src/xenapi/xenapi_driver.h @@ -1,5 +1,5 @@ /* - * xenapi_driver.h.c: Xen API driver header file to be included in libvirt.c. + * xenapi_driver.h: Xen API driver header file to be included in libvirt.c. * Copyright (C) 2009, 2010 Citrix Ltd. * * This library is free software; you can redistribute it and/or diff --git a/tests/testutils.h b/tests/testutils.h index fa37246..e89492b 100644 --- a/tests/testutils.h +++ b/tests/testutils.h @@ -1,5 +1,5 @@ /* - * utils.c: test utils + * testutils.h: test utils * * Copyright (C) 2005, 2008-2013 Red Hat, Inc. * diff --git a/tools/virsh-interface.h b/tools/virsh-interface.h index 2544f7e..6272aab 100644 --- a/tools/virsh-interface.h +++ b/tools/virsh-interface.h @@ -1,5 +1,5 @@ /* - * virsh-interface.c: Commands to manage host interface + * virsh-interface.h: Commands to manage host interface * * Copyright (C) 2005, 2007-2012 Red Hat, Inc. * diff --git a/tools/virsh-network.h b/tools/virsh-network.h index 7178fde..3beeeb9 100644 --- a/tools/virsh-network.h +++ b/tools/virsh-network.h @@ -1,5 +1,5 @@ /* - * virsh-network.c: Commands to manage network + * virsh-network.h: Commands to manage network * * Copyright (C) 2005, 2007-2012 Red Hat, Inc. * diff --git a/tools/virt-host-validate.c b/tools/virt-host-validate.c index d3cf19c..a8c2075 100644 --- a/tools/virt-host-validate.c +++ b/tools/virt-host-validate.c @@ -1,5 +1,5 @@ /* - * virt-host-check.c: Sanity check a hypervisor host + * virt-host-validate.c: Sanity check a hypervisor host * * Copyright (C) 2012 Red Hat, Inc. * -- 1.8.3.2

3 3

[libvirt] Seeking for advice on 'virsh console' problems debugging
by Roman Bogorodskiy 10 Mar '14

10 Mar '14

Hi, I've been trying to understand why 'virsh console' doesn't work on FreeBSD for some time and now I feel like I got stuck. Background of the problem. Some time ago I started implementing console support for the bhyve driver. I initially implemented it using nmdm(4) device: http://www.freebsd.org/cgi/man.cgi?query=nmdm&sektion=4&manpath=FreeBSD+10.… But 'virsh console' didn't work on it and exited immediately. I though that it could be nmdm compatibility issue and decided to go with the plan /dev/pts devices. But I got the same problem with it. So I decided to try use console with qemu driver. And... I got the same problem. After some investigation, I spotted this block in tools/virsh-console.c:376 while (!con->quit) { if (virCondWait(&con->cond, &con->lock) < 0) { VIR_ERROR(_("unable to wait on console condition")); goto cleanup; } } The strange thing is that it doesn't look con->lock mutex before waiting. FreeBSD man page for pthread_cond_wait() says that it errors in that situation: [EPERM] The specified mutex was not locked by the calling thread. I replaced VIR_ERROR with virReportSystemError and it returns EPERM indeed. So I've added virMutexLock() before calling virCondWait() and it partially solved the problem for me -- 'virsh console' stopped exiting immediately. Actually, it started output for guest, but it's badly formatted: it includes lots and lots of whitespaces. Input doesn't work properly (cannot type in sensible stuff). Here's a sample output, even libvirt messages are badly formatted (I added some debug prints along the way). http://people.freebsd.org/~novel/misc/ttyweirdness.png So, e.g. dmesg output for the guest looks almost the same. Looks like there's something wrong with the terminal settings, but how could that be debugged? PS Worth to mention that 'cu' worked all the time for me on these devices (both created by bhyve and qemu). Roman Bogorodskiy

2 8

[libvirt] [PATCH] Wildcard ignore tests/*test instead of listing every one
by Daniel P. Berrange 10 Mar '14

10 Mar '14

There is a forever growing list of test cases. It is just not worth listing each one individually when a wildcard can do the job. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- .gitignore | 73 +------------------------------------------------------------- 1 file changed, 1 insertion(+), 72 deletions(-) diff --git a/.gitignore b/.gitignore index cb60734..efbce50 100644 --- a/.gitignore +++ b/.gitignore @@ -145,85 +145,14 @@ /tests/*.log /tests/*.pid /tests/*.trs -/tests/*xml2*test /tests/commandhelper -/tests/commandtest -/tests/conftest -/tests/cputest -/tests/domainconftest -/tests/domainsnapshotxml2xmltest -/tests/esxutilstest -/tests/eventtest -/tests/fchosttest -/tests/fdstreamtest -/tests/hashtest -/tests/jsontest -/tests/libvirtdconftest -/tests/lxcconf2xmltest -/tests/metadatatest -/tests/networkxml2argvtest -/tests/nodeinfotest -/tests/nwfilterxml2xmltest -/tests/objecteventtest +/tests/*test /tests/object-locking /tests/object-locking-files.txt /tests/object-locking.cm[ix] -/tests/openvzutilstest -/tests/qemuagenttest -/tests/qemuargv2xmltest -/tests/qemucapabilitiestest -/tests/qemuhelptest -/tests/qemuhotplugtest -/tests/qemumonitorjsontest -/tests/qemumonitortest -/tests/qemuxmlnstest -/tests/qparamtest /tests/reconnect -/tests/secaatest -/tests/seclabeltest -/tests/securityselinuxlabeltest -/tests/securityselinuxtest -/tests/sexpr2xmltest -/tests/shunloadtest -/tests/sockettest /tests/ssh -/tests/statstest -/tests/storagebackendsheepdogtest -/tests/sysinfotest /tests/test_conf -/tests/utiltest -/tests/viratomictest -/tests/virauthconfigtest -/tests/virbitmaptest -/tests/virbuftest -/tests/vircapstest -/tests/vircgrouptest -/tests/virdbustest -/tests/virdrivermoduletest -/tests/virendiantest -/tests/virfiletest -/tests/virhashtest -/tests/viridentitytest -/tests/virkeycodetest -/tests/virkeyfiletest -/tests/virkmodtest -/tests/virlockspacetest -/tests/virlogtest -/tests/virnet*test -/tests/virpcitest -/tests/virportallocatortest -/tests/virscsitest -/tests/virshtest -/tests/virstoragetest -/tests/virstringtest -/tests/virsystemdtest -/tests/virtimetest -/tests/viruritest -/tests/virusbtest -/tests/vmwarevertest -/tests/vmx2xmltest -/tests/xencapstest -/tests/xmconfigtest /tools/*.[18] /tools/libvirt-guests.init /tools/libvirt-guests.service -- 1.8.5.3

4 4

[libvirt] [PATCH] Add a mutex to serialize updates to firewall
by Daniel P. Berrange 10 Mar '14

10 Mar '14

For https://bugzilla.redhat.com/show_bug.cgi?id=1066801 The nwfilter conf update mutex previously serialized updates to the internal data structures for firewall rules, and updates to the firewall itself. The latter was recently turned into a read/write lock, and filter instantiation allowed to proceed in parallel. It was believed that this was ok, since each filter is created on a seperate iptables/ebtables chain. It turns out that there is a sutle lock ordering problem on virNWFilterObjPtr instances. __virNWFilterInstantiateFilter will hold a lock on the virNWFilterObjPtr it is instantiating. This in turn invokes virNWFilterInstantiate which then invokes virNWFilterDetermineMissingVarsRec which then invokes virNWFilterObjFindByName. This iterates over every single virNWFilterObjPtr in the list, locking them and checking their name. So if 2 or more threads try to instantiate a filter in parallel, they'll all hold 1 lock at the top level in the __virNWFilterInstantiateFilter method which will cause the other thread to deadlock in virNWFilterObjFindByName. The fix is to add an exclusive mutex to serialize the execution of __virNWFilterInstantiateFilter. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/nwfilter/nwfilter_driver.c | 6 ++++-- src/nwfilter/nwfilter_gentech_driver.c | 34 ++++++++++++++++++++++++++++++++-- src/nwfilter/nwfilter_gentech_driver.h | 2 +- 3 files changed, 37 insertions(+), 5 deletions(-) diff --git a/src/nwfilter/nwfilter_driver.c b/src/nwfilter/nwfilter_driver.c index 5908df7..2e89d07 100644 --- a/src/nwfilter/nwfilter_driver.c +++ b/src/nwfilter/nwfilter_driver.c @@ -200,7 +200,8 @@ nwfilterStateInitialize(bool privileged, if (virNWFilterDHCPSnoopInit() < 0) goto err_exit_learnshutdown; - virNWFilterTechDriversInit(privileged); + if (virNWFilterTechDriversInit(privileged) < 0) + goto err_dhcpsnoop_shutdown; if (virNWFilterConfLayerInit(virNWFilterDomainFWUpdateCB, driverState) < 0) @@ -251,6 +252,7 @@ error: err_techdrivers_shutdown: virNWFilterTechDriversShutdown(); +err_dhcpsnoop_shutdown: virNWFilterDHCPSnoopShutdown(); err_exit_learnshutdown: virNWFilterLearnShutdown(); @@ -327,10 +329,10 @@ nwfilterStateCleanup(void) { if (driverState->privileged) { virNWFilterConfLayerShutdown(); - virNWFilterTechDriversShutdown(); virNWFilterDHCPSnoopShutdown(); virNWFilterLearnShutdown(); virNWFilterIPAddrMapShutdown(); + virNWFilterTechDriversShutdown(); nwfilterDriverLock(driverState); diff --git a/src/nwfilter/nwfilter_gentech_driver.c b/src/nwfilter/nwfilter_gentech_driver.c index 8c5cd57..5144dce 100644 --- a/src/nwfilter/nwfilter_gentech_driver.c +++ b/src/nwfilter/nwfilter_gentech_driver.c @@ -55,15 +55,34 @@ static virNWFilterTechDriverPtr filter_tech_drivers[] = { NULL }; +/* Serializes instantiation of filters. This is neccessary + * to avoid lock ordering deadlocks. eg __virNWFilterInstantiateFilter + * will hold a lock on a virNWFilterObjPtr. This in turn invokes + * virNWFilterInstantiate which invokes virNWFilterDetermineMissingVarsRec + * which invokes virNWFilterObjFindByName. This iterates over every single + * virNWFilterObjPtr in the list. So if 2 threads try to instantiate a + * filter in parallel, they'll both hold 1 lock at the top level in + * __virNWFilterInstantiateFilter which will cause the other thread + * to dead lock in virNWFilterObjFindByName. + * + * XXX better long term solution is to make virNWFilterObjList use a + * hash table as is done for virDomainObjList. You can then get + * lockless lookup of objects by name. + */ +static virMutex updateMutex; -void virNWFilterTechDriversInit(bool privileged) { +int virNWFilterTechDriversInit(bool privileged) { size_t i = 0; VIR_DEBUG("Initializing NWFilter technology drivers"); + if (virMutexInitRecursive(&updateMutex) < 0) + return -1; + while (filter_tech_drivers[i]) { if (!(filter_tech_drivers[i]->flags & TECHDRV_FLAG_INITIALIZED)) filter_tech_drivers[i]->init(privileged); i++; } + return 0; } @@ -74,6 +93,7 @@ void virNWFilterTechDriversShutdown(void) { filter_tech_drivers[i]->shutdown(); i++; } + virMutexDestroy(&updateMutex); } @@ -935,6 +955,8 @@ _virNWFilterInstantiateFilter(virNWFilterDriverStatePtr driver, int ifindex; int rc; + virMutexLock(&updateMutex); + /* after grabbing the filter update lock check for the interface; if it's not there anymore its filters will be or are being removed (while holding the lock) and we don't want to build new ones */ @@ -962,6 +984,8 @@ _virNWFilterInstantiateFilter(virNWFilterDriverStatePtr driver, foundNewFilter); cleanup: + virMutexUnlock(&updateMutex); + return rc; } @@ -981,6 +1005,7 @@ virNWFilterInstantiateFilterLate(virNWFilterDriverStatePtr driver, bool foundNewFilter = false; virNWFilterReadLockFilterUpdates(); + virMutexLock(&updateMutex); rc = __virNWFilterInstantiateFilter(driver, vmuuid, @@ -1006,6 +1031,7 @@ virNWFilterInstantiateFilterLate(virNWFilterDriverStatePtr driver, } virNWFilterUnlockFilterUpdates(); + virMutexUnlock(&updateMutex); return rc; } @@ -1129,7 +1155,11 @@ _virNWFilterTeardownFilter(const char *ifname) int virNWFilterTeardownFilter(const virDomainNetDef *net) { - return _virNWFilterTeardownFilter(net->ifname); + int ret; + virMutexLock(&updateMutex); + ret = _virNWFilterTeardownFilter(net->ifname); + virMutexUnlock(&updateMutex); + return ret; } diff --git a/src/nwfilter/nwfilter_gentech_driver.h b/src/nwfilter/nwfilter_gentech_driver.h index f4789e1..d72e040 100644 --- a/src/nwfilter/nwfilter_gentech_driver.h +++ b/src/nwfilter/nwfilter_gentech_driver.h @@ -31,7 +31,7 @@ virNWFilterTechDriverPtr virNWFilterTechDriverForName(const char *name); int virNWFilterRuleInstAddData(virNWFilterRuleInstPtr res, void *data); -void virNWFilterTechDriversInit(bool privileged); +int virNWFilterTechDriversInit(bool privileged); void virNWFilterTechDriversShutdown(void); enum instCase { -- 1.8.5.3

2 2

[libvirt] [PATCH] Fix locking in virsh console
by Roman Bogorodskiy 10 Mar '14

10 Mar '14

While I'm still fighting with console on FreeBSD, I decided to sumbit that fix as it seems to be useful on its own. With that change I'm able to run virsh on FreeBSD to connect to libvirtd on Linux and get console working properly there. Roman Bogorodskiy (1): Fix locking in virsh console tools/virsh-console.c | 5 +++++ 1 file changed, 5 insertions(+) -- 1.8.4.2

2 3

[libvirt] How to detect if qemu supports discard?
by Richard W.M. Jones 10 Mar '14

10 Mar '14

from src/qemu/qemu_command.c: if (disk->discard) { if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_DRIVE_DISCARD)) { virBufferAsprintf(&opt, ",discard=%s", virDomainDiskDiscardTypeToString(disk->discard)); } else { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("discard is not supported by this QEMU binary")); goto error; } } However there appears to be no way to tell programmatically if the qemu binary supports discard (ie. via libvirt). Therefore libguestfs can never know when it is safe to use this option in the XML. Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones libguestfs lets you edit virtual machines. Supports shell scripting, bindings from many languages. http://libguestfs.org

3 2

[libvirt] [bug-report] libvirtd can't be connected by using the master branch with lots of VMs
by Wangpan 10 Mar '14

10 Mar '14

bug address: https://bugzilla.redhat.com/show_bug.cgi?id=1066801 I can reproduce it at master/v1.2.1-maint/v1.1.4-maint/v1.1.3.4 release branches everytime, but cann't reproduce it at v1.2.1，so I believe this bug is imported by the maintain patches after v1.2.1 release. Steps to Reproduce: 1.create many VMs(40 in my env) on the host 2.kill -9 `pid of libvirtd` 3.start libvirtd by using service libvirt-bin start Actual results: virsh version/list command is hung there without any responses Expected results: virsh version/list return the correct things Additional info: if there are few VMs(1 vm in my env), the virsh version command is OK Thanks in advance! 2014-03-07 Wangpan

2 4

[libvirt] [PATCH 0/4] Add keepalive support in virsh
by Martin Kletzander 10 Mar '14

10 Mar '14

*** BLURB *** Martin Kletzander (4): virsh: Sort options alphabetically virsh: Add keepalive in new vshConnect function virsh: Prohibit virConnectOpen* functions in virsh [DO_NOT_APPLY_UPSTREAM] virsh: add connection monitoring into vshWatchJob cfg.mk | 8 ++- tests/virsh-optparse | 3 +- tools/virsh-domain.c | 21 +++++--- tools/virsh.c | 141 +++++++++++++++++++++++++++++++++++++-------------- tools/virsh.h | 7 +++ tools/virsh.pod | 38 +++++++------- 6 files changed, 152 insertions(+), 66 deletions(-) -- 1.9.0

2 9

[libvirt] [PATCH] tests: Fix SELinux tests in VPATH build
by Jiri Denemark 10 Mar '14

10 Mar '14

Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- Pushed as a build-breaker. Notes: abs_builddir is used in more places in tests/securityselinuxhelper.c but since the path they are used in does not exist in abs_srcdir either, I guess it doesn't really matter what prefix is used there. tests/securityselinuxhelper.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tests/securityselinuxhelper.c b/tests/securityselinuxhelper.c index 703381c..dbc4c29 100644 --- a/tests/securityselinuxhelper.c +++ b/tests/securityselinuxhelper.c @@ -278,7 +278,7 @@ const char *selinux_virtual_domain_context_path(void) if (realis_selinux_enabled()) return realselinux_virtual_domain_context_path(); - return abs_builddir "/securityselinuxhelperdata/virtual_domain_context"; + return abs_srcdir "/securityselinuxhelperdata/virtual_domain_context"; } const char *selinux_virtual_image_context_path(void) @@ -288,7 +288,7 @@ const char *selinux_virtual_image_context_path(void) if (realis_selinux_enabled()) return realselinux_virtual_image_context_path(); - return abs_builddir "/securityselinuxhelperdata/virtual_image_context"; + return abs_srcdir "/securityselinuxhelperdata/virtual_image_context"; } #ifdef HAVE_SELINUX_LXC_CONTEXTS_PATH @@ -299,7 +299,7 @@ const char *selinux_lxc_contexts_path(void) if (realis_selinux_enabled()) return realselinux_lxc_contexts_path(); - return abs_builddir "/securityselinuxhelperdata/lxc_contexts"; + return abs_srcdir "/securityselinuxhelperdata/lxc_contexts"; } #endif -- 1.9.0

1 0

[libvirt] virDomainGetMaxVcpus does not work as expected
by Claudio Bley 10 Mar '14

10 Mar '14

Hi. When calling virDomainGetMaxVcpus (http://libvirt.org/html/libvirt-libvirt.html#virDomainGetMaxVcpus) on an inactive domain, I receive this error: scala> res2.getMaxVcpus() libvirt: Domain Config error : Requested operation is not valid: domain is not running org.libvirt.LibvirtException: Requested operation is not valid: domain is not running at org.libvirt.ErrorHandler.processError(ErrorHandler.java:31) at org.libvirt.ErrorHandler.processError(ErrorHandler.java:46) at org.libvirt.Domain.getMaxVcpus(Domain.java:571) at .<init>(<console>:13) ... (this is from Java, but that doesn't matter) The docs say: > If the guest is inactive, this is basically the same as > virConnectGetMaxVcpus(). If the guest is running this will reflect > the maximum number of virtual CPUs the guest was booted with. But, apparently, all the driver implementations for virDomainGetMaxVcpus forward to <driver>DomainGetVcpusFlags(.., VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_VCPU_MAXIMUM). _______________________________,~~~~~~~~~~~~~~~~~~~~~~ $ git grep --show-function 'GetVcpusFlags.*AFFECT_LIVE' src/esx/esx_driver.c=esxDomainGetMaxVcpus(virDomainPtr domain) src/esx/esx_driver.c: return esxDomainGetVcpusFlags(domain, (VIR_DOMAIN_AFFECT_LIVE | src/openvz/openvz_driver.c=static int openvzDomainGetMaxVcpus(virDomainPtr dom) src/openvz/openvz_driver.c: return openvzDomainGetVcpusFlags(dom, (VIR_DOMAIN_AFFECT_LIVE | src/qemu/qemu_driver.c=qemuDomainGetMaxVcpus(virDomainPtr dom) src/qemu/qemu_driver.c: return qemuDomainGetVcpusFlags(dom, (VIR_DOMAIN_AFFECT_LIVE | src/test/test_driver.c=testDomainGetMaxVcpus(virDomainPtr domain) src/test/test_driver.c: return testDomainGetVcpusFlags(domain, (VIR_DOMAIN_AFFECT_LIVE | src/vbox/vbox_tmpl.c=vboxDomainGetMaxVcpus(virDomainPtr dom) src/vbox/vbox_tmpl.c: return vboxDomainGetVcpusFlags(dom, (VIR_DOMAIN_AFFECT_LIVE | AFAICS, this was introduced with commit 50c51f13e2af04afac46e181c4ed62581545a488 Author: Eric Blake <eblake(a)redhat.com> Date: Mon Sep 27 16:37:53 2010 -0600 vcpu: make old API trivially wrap to new API Whereas the function's contract was documented earlier by commit b412cfadb502c76df095c2c4548c27abf7c4873f Author: Daniel Veillard <veillard(a)redhat.com> Date: Thu Mar 8 08:31:07 2007 +0000 To be honest, I'm not sure whether this worked as described at some time in the past _at all_. How to fix this? Change the documentation or the flag? Claudio

5 11

[libvirt] [PATCH] qemu: Advertize <drivediscard/> in capabilities if 'discard' is supported.
by Richard W.M. Jones 09 Mar '14

09 Mar '14

The capabilities will have a <drivediscard/> feature if you are allowed to use the qemu attribute: <driver name='qemu' ... discard='unmap|ignore' /> The new capabilities XML looks like this: <capabiltiies> ... <guest> <os_type>hvm</os_type> ... <features> <drivediscard/> </features> </guest> </capabiltiies> --- src/conf/capabilities.c | 3 ++- src/qemu/qemu_capabilities.c | 4 ++++ 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/src/conf/capabilities.c b/src/conf/capabilities.c index c1c4ab8..ff5912a 100644 --- a/src/conf/capabilities.c +++ b/src/conf/capabilities.c @@ -947,7 +947,8 @@ virCapabilitiesFormatXML(virCapsPtr caps) STREQ(caps->guests[i]->features[j]->name, "nonpae") || STREQ(caps->guests[i]->features[j]->name, "ia64_be") || STREQ(caps->guests[i]->features[j]->name, "cpuselection") || - STREQ(caps->guests[i]->features[j]->name, "deviceboot")) { + STREQ(caps->guests[i]->features[j]->name, "deviceboot") || + STREQ(caps->guests[i]->features[j]->name, "drivediscard")) { virBufferAsprintf(&xml, " <%s/>\n", caps->guests[i]->features[j]->name); } else { diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index cae25e0..b755fb2 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -778,6 +778,10 @@ virQEMUCapsInitGuest(virCapsPtr caps, !virCapabilitiesAddGuestFeature(guest, "deviceboot", 1, 0)) goto error; + if (virQEMUCapsGet(qemubinCaps, QEMU_CAPS_DRIVE_DISCARD) && + !virCapabilitiesAddGuestFeature(guest, "drivediscard", 0, 0)) + goto error; + if (virCapabilitiesAddGuestDomain(guest, "qemu", NULL, -- 1.8.5.3

1 1

[libvirt] [PATCH 00/13] Utilize more of VIR_(APPEND|INSERT|DELETE)_ELEMENT
by Michal Privoznik 07 Mar '14

07 Mar '14

2 26

[libvirt] Suspending access to opened/active /dev/nodes during application runtime
by Lukasz Pawelczyk 07 Mar '14

07 Mar '14

Problem: Has anyone thought about a mechanism to limit/remove an access to a device during an application runtime? Meaning we have an application that has an open file descriptor to some /dev/node and depending on *something* it gains or looses the access to it gracefully (with or without a notification, but without any fatal consequences). Example: LXC. Imagine we have 2 separate containers. Both running full operating systems. Specifically with 2 X servers. Both running concurrently of course. Both need the same input devices (e.g. we have just one mouse). This creates a security problem when we want to have completely separate environments. One container is active (being displayed on a monitor and controlled with a mouse) while the other container runs evtest /dev/input/something and grabs the secret password user typed in the other. Solutions: The complete solution would comprise of 2 parts: - a mechanism that would allow to temporally "hide" a device from an open file descriptor. - a mechanism for deciding whether application/process/namespace should have an access to a specific device at a specific moment Let's focus on the first problem only, as it would need to be solved first anyway. I haven't found anything that would allow me to do it. There are a lot mechanisms that make it possible to restrict an access during open(): - DAC - ACL (controlled by hand or with uaccess) - LSM (in general) - device cgroups But all of those can't do a thing when the device is already opened and an application has a file descriptor. I don't see such mechanism in kernel sources either. I do imagine that it would not be possible for every device to handle such a thing (dri comes to mind) without breaking something (graphics card state in dri example). But there is class of simple input/output devices that would handle this without problems. I did implement some proof-of-concept solution for an evdev driver by allowing or disallowing events that go to evdev_client structure using some arbitrary condition. But this is far from a generic solution. My proof-of-concept is somewhat similar to this (I just found it): http://www.spinics.net/lists/linux-input/msg25547.html Though a little bit wider in scope. But neither is flawless nor generic. Has anyone had any thoughts about a similar problem? -- Regards Havner

2 3

[libvirt] Release of libvirt-1.2.2
by Daniel Veillard 07 Mar '14

07 Mar '14

As planned I have tagged libvirt-1.2.2 in git and it seem the git tree finally updated despite the horrible network I get in china ATM, the tarballs may finally arrive at some time too at the usual place after working around the local infrastructure flaws... : ftp://libvirt.org/libvirt/ I am also pushing a libvirt-python-1.2.2 update at: ftp://libvirt.org/libvirt/libvirt/ since fedpkg upload libvirt-1.2.2.tar.gz failed to progress further than 7% using eithe Cable or ADSL, i assume someone lese will push those to Fedora build system. This is a rather large release with a few new features including beginning of support for the BSD hypervisor, and a lot of significant improvements in various place like linux containers, network storage, Xen drivers, internals etc ... Also of note the fixes for the LXC security bug CVE-2013-6456: Features: - bhyve: add a basic driver (Roman Bogorodskiy) - add LXC from native conversion tool (Cédric Bosdonnat) - vbox: add support for v4.2.20+ and v4.3.4+ (Jean-Baptiste Rouault) - Introduce Libvirt Wireshark dissector (Yuto KAWAMURA) Security: - CVE-2013-6456: Avoid unsafe use of /proc/$PID/root in LXC hotunplug code (Daniel P. Berrange) - CVE-2013-6456: Avoid unsafe use of /proc/$PID/root in LXC chardev hostdev hotplug (Daniel P. Berrange) - CVE-2013-6456: Avoid unsafe use of /proc/$PID/root in LXC block hostdev hotplug (Daniel P. Berrange) - CVE-2013-6456: Avoid unsafe use of /proc/$PID/root in LXC USB hotplug (Daniel P. Berrange) - CVE-2013-6456: Avoid unsafe use of /proc/$PID/root in LXC disk hotplug (Daniel P. Berrange) - CVE-2013-6456: Avoid unsafe use of /proc/$PID/root in LXC shutdown/reboot code (Eric Blake) Documentation: - bhyve: add basic documentation (Roman Bogorodskiy) - Add docs about use of systemd journal for logging (Daniel P. Berrange) - Auto-generate the table of contents in logging doc (Daniel P. Berrange) - Fix heading level in logging docs (Daniel P. Berrange) - Document the keyboard as a valid input type (Ján Tomko) - bandwidth: Adjust documentation (John Ferlan) - remove <auth> from secret XML format (Ján Tomko) - LXC: added some doc on domxml-from-native with mention of limitations (Cédric Bosdonnat) - libxl: fix libxlDoDomainSave documentation (Jim Fehlig) - grammar fixes in formatdomain (Chen Hanxiao) - maint: fix grammar in conf file (Eric Blake) - Fix minor typo in governance doc (Justin Clift) - Write up the project governance process (Daniel P. Berrange) - man: shm-merge-across-nodes is optional (Ján Tomko) - docs/page.xls: remove unnecessary namespace attribute (Pavel Hrdina) - add a permalink to html headers (Dan Kenigsberg) - storage: Add document for possible problem on volume detection (Osier Yang) - Add "note" for node-memory-tune (Osier Yang) - Improve the document for nodesuspend (Osier Yang) - refer to the correct event ID for DomainEventIOErrorReasonCallback (Claudio Bley) - Add sample output of Wireshark dissector (Yuto KAWAMURA) - Add missing space in <clock> documentation (Christophe Fergeau) Portability: - build: skip virportallocatortest on cygwin (Eric Blake) - build: ignore cygwin toolchain droppings (Eric Blake) - build: avoid ld_preload tests on mingw (Eric Blake) - build: fix cgroups on non-Linux (Eric Blake) - build: fix build on 32-bit hosts (Eric Blake) - maint: update to latest gnulib, for older autoconf (Eric Blake) - Fix build of portallocator on mingw (Ján Tomko) - lxc: Don't shadow global symbol "link" (Peter Krempa) - storage: Fix build with older compilers afeter gluster snapshot series (Peter Krempa) - spec: add missing dep of libvirt-daemon-config-network (Thierry Parmentelat) - spec: require libvirt-daemon-driver-interface only when built (Thierry Parmentelat) - Fixed build with clang. (Cédric Bosdonnat) - Rename 'index' in virCapabilitiesGetCpusForNode (Ján Tomko) - qemuxml2argvmock: Mock time() on non-linux platforms too (Michal Privoznik) - build: Fix 'make rpm' in VPATH with wireshark (Martin Kletzander) - Remove windows thread implementation in favour of pthreads (Daniel P. Berrange) - Fix pthread_sigmask check for mingw32 without winpthreads (Daniel P. Berrange) - Skip check-augeas-lockd when QEMU is disabled (Daniel P. Berrange) - maint: add configure checks for BSD CPU affinity (Roman Bogorodskiy) - BSD: implement virProcess{Get,Set}Affinity (Roman Bogorodskiy) - tests: Fix PCI test data filenames for Windows (Matthias Bolte) - wireshark: Fix VPATH build (Jiri Denemark) - Honour prefix in wireshark install dir (Daniel P. Berrange) - Use AC_PATH_PROG to search for dmidecode (Roman Bogorodskiy) Bug fixes: - bhyve: defined domains should be persistent (Roman Bogorodskiy) - Fix journald PRIORITY values (Daniel P. Berrange) - spec: Fix braces around macros (Peter Krempa) - spec: Use correct versions of libgfapi in RHEL builds (Peter Krempa) - network: unplug bandwidth and call networkRunHook only when appropriate (Laine Stump) - network: don't even call networkRunHook if there is no network (Laine Stump) - Fix memory leak in virSCSIDeviceListDel() (Nehal J Wani) - libxl: queue domain event earlier in shutdown handler (Jim Fehlig) - virsh: mark CPU usage field names as translatable (Ján Tomko) - virsh: initialize str to NULL to solve a build issue (Chen Hanxiao) - virsh: Don't leak buffer if GetFDs fails in cmdCreate (Ján Tomko) - libvirt-guests: Wait for libvirtd to initialize (Michal Privoznik) - virSystemdCreateMachine: Set dependencies for slices (Michal Privoznik) - Ignore additional fields in iscsiadm output (Ján Tomko) - libxl: always use libxlVmCleanupJob in shutdown thread (Jim Fehlig) - qemu: adjust maxmem/maxvcpu computation (Eric Blake) - Fix multiple bugs in LXC domainMemoryStats driver (Daniel P. Berrange) - Fix misspelled cpuacct.usage_percpu in cgroup mock. (Thorsten Behrens) - virsh: fix memleak when starting a guest with invalid fd (Jincheng Miao) - networkRunHook: Run hook only if possible (Michal Privoznik) - bridge_driver.h: Fix build --without-network (Michal Privoznik) - Fix conflicting types of virInitctlSetRunLevel (Ján Tomko) - Fix reset of cgroup when detaching USB device from LXC guests (Daniel P. Berrange) - Fix path used for USB device attach with LXC (Daniel P. Berrange) - Don't block use of USB with containers (Daniel P. Berrange) - storage: gluster: Don't leak private data when storage file init fails (Peter Krempa) - storage: handle NULL return from virGetStorageVol (Michael Chapman) - Fix leaks in vircapstest (Ján Tomko) - AppArmor: Fix the place where the template should be installed (Cédric Bosdonnat) - Libvirt lose sheepdogs volumes on pool refresh or restart. When restarting sheepdog pool, all volumes are missing. This patch add automatically all volume from the added pool. (Joel SIMOES) - maint: fix line numbers in check-aclrules reports (Yuri Myasoedov) - qemu: Use correct permissions when determining the image chain (Peter Krempa) - virpci: Resolve coverity issues (John Ferlan) - qemu: keep pre-migration domain state after failed migration (Martin Kletzander) - qemu: Fix crash in virDomainMemoryStats with old qemu (Jiri Denemark) - network: disallow <bandwidth>/<mac> for bridged/macvtap/hostdev networks (Laine Stump) - Honor blacklist for modprobe command (John Ferlan) - qemu: be sure we're using the updated value of backend during hotplug (Laine Stump) - Resolve Coverity dead_error_begin (John Ferlan) - qemu: Fix the error message for scsi host device's shareable checking (Osier Yang) - xen: fix parsing xend http response (Jim Fehlig) - storage: Fix the memory leak (Osier Yang) - Fix buffer size in linuxNodeGetCPUstats (Bing Bu Cao) - util: Correct the NUMA node range checking (Osier Yang) - qemu: Avoid crash in qemuDiskGetActualType (Peter Krempa) - linuxNodeGetCPUStats: Correctly handle cpu prefix (Bing Bu Cao) - qemu: snapshot: Forbid snapshots when backing is a scsi passthrough disk (Peter Krempa) - qemu: snapshot: Avoid libvirtd crash when qemu crashes while snapshotting (Peter Krempa) - lxc: Fix coverity (Martin Kletzander) - qemu: Don't detach devices if passthrough doesn't work (Jincheng Miao) - pci: Fix failure paths in detach (Jiri Denemark) - virSecuritySELinuxSetFileconHelper: Don't fail on read-only NFS (Michal Privoznik) - Fix possible memory leak in virsh-domain-monitor.c in cmdDomblklist (Pavel Hrdina) Improvements: - virsh: add --all flag to 'event' command (Eric Blake) - virsh: support remaining domain events (Eric Blake) - bhyve: support domain undefine (Roman Bogorodskiy) - Add comments describing the different log sources (Daniel P. Berrange) - Include error domain and code in log messages from errors (Daniel P. Berrange) - Send virLogMetadata fields onto the journal (Daniel P. Berrange) - qemu: Enable 'host-passthrough' cpu mode for arm (Oleg Strikov) - domblkstat: Produce error message that at least sounds like English (Michal Privoznik) - virDomainBlockStats(Flags): Produce saner error message on empty disk path (Michal Privoznik) - tests: avoid littering /tmp (Eric Blake) - sanlock: Truncate domain names longer than SANLK_NAME_LEN (Jiri Denemark) - maint: Fix minor typo (unkown) (Yuri Chornoivan) - libxl: Recognise ARM architectures (Ian Campbell) - virsh: Honour -q in domblklist, vcpupin and emulatorpin (Michal Privoznik) - spec: make systemd_daemon usage configurable (Eric Blake) - spec: require device-mapper-devel for storage-disk (Eric Blake) - spec: explicitly avoid bhyve on Linux (Eric Blake) - build: use --with-systemd-daemon as configure option (Eric Blake) - virNetDevVethCreate: Serialize callers (Michal Privoznik) - network: include plugged interface XML in "plugged" network hook (Laine Stump) - conf: output actual netdev status in <interface> XML (Laine Stump) - conf: new function virDomainActualNetDefContentsFormat (Laine Stump) - conf: re-situate <bandwidth> element in <interface> (Laine Stump) - conf: make virDomainNetDefFormat a public function (Laine Stump) - conf: handle null pointer in virNetDevVlanFormat (Laine Stump) - conf: clarify what is returned for actual bandwidth and vlan (Laine Stump) - rbd: Set timeout options for librados (Wido den Hollander) - rbd: Include return statuses from librados/librbd in logging (Wido den Hollander) - virsh: kill over-engineered asprintf failure recovery (Eric Blake) - virsh: use more compact VIR_ENUM_IMPL (Eric Blake) - libxl: handle on_crash coredump actions (Jim Fehlig) - libxl: add dump dir to libxlDriverConfig object (Jim Fehlig) - libxl: honor domain lifecycle event configuration (Jim Fehlig) - Ensure systemd cgroup ownership is delegated to container with userns (Richard Weinberger) - bhyve: implement node information reporting (Roman Bogorodskiy) - Add virStringReplace method for substring replacement (Daniel P. Berrange) - Add virStringSearch method for regex matching (Manuel VIVES) - virNetServerRun: Notify systemd that we're accepting clients (Michal Privoznik) - Add a stub for virCgroupGetDomainTotalCpuStats (Ján Tomko) - maint: update to latest gnulib (Eric Blake) - virsh: add net-event command (Eric Blake) - virsh: add event command, for lifecycle events (Eric Blake) - virsh: common code for waiting for an event (Eric Blake) - virsh: common code for parsing --seconds (Eric Blake) - libxl: queue shutdown event on domain shutdown (Jim Fehlig) - Rename virDomainGetRootFilesystem to virDomainGetFilesystemForTarget (Daniel P. Berrange) - Introduce new OOM testing support (Daniel P. Berrange) - Add unit test for virCgroupGetPercpuStats. (Thorsten Behrens) - Add unit test for virCgroupGetMemoryUsage. (Thorsten Behrens) - Add unit test for virCgroupGetBlkioIo*Serviced (Thorsten Behrens) - Widening API change - accept empty path for virDomainBlockStats (Thorsten Behrens) - Implement lxcDomainBlockStats* for lxc driver (Thorsten Behrens) - Implement domainGetCPUStats for lxc driver. (Thorsten Behrens) - Make qemuGetDomainTotalCPUStats a virCgroup function. (Thorsten Behrens) - Implement domainMemoryStats API slot for LXC driver. (Thorsten Behrens) - Add util virCgroupGetBlkioIo*Serviced methods. (Thorsten Behrens) - lxc: Add destroy support for suspended domains (Richard Weinberger) - libxl: use job functions in libxlDomainSetSchedulerParametersFlags (Jim Fehlig) - libxl: use job functions in libxlDomainSetAutostart (Jim Fehlig) - libxl: use job functions in device attach and detach functions (Jim Fehlig) - libxl: use job functions in vcpu set and pin functions (Jim Fehlig) - libxl: use job functions in libxlDomainCoreDump (Jim Fehlig) - libxl: use job functions in domain save operations (Jim Fehlig) - libxl: use job functions when cleaning up a domain (Jim Fehlig) - libxl: use job functions in libxlDomain{Suspend,Resume} (Jim Fehlig) - libxl: use job functions in libxlDomainSetMemoryFlags (Jim Fehlig) - libxl: use job functions in libxlVmStart (Jim Fehlig) - libxl: Add job support to libxl driver (Jim Fehlig) - libxl: remove libxlVmReap function (Jim Fehlig) - libxl: always set vm id to -1 on shutdown (Jim Fehlig) - qemu: Use virtio network device for aarch64/virt (Oleg Strikov) - Add a default USB keyboard and USB mouse for PPC64 (Li Zhang) - xen: format xen config for USB keyboard (Li Zhang) - qemu: format qemu command line for USB keyboard (Li Zhang) - qemu: Add USB keyboard capability (Li Zhang) - conf: Remove the implicit PS2 devices for non-X86 platforms (Li Zhang) - conf: Add keyboard input device type (Li Zhang) - conf: Add one interface to add default input devices (Li Zhang) - network: Taint networks that are using hook script (Michal Privoznik) - network: Introduce network hooks (Michal Privoznik) - network_conf: Expose virNetworkDefFormatInternal (Michal Privoznik) - Add helper for running code in separate namespaces (Daniel P. Berrange) - Add virFileMakeParentPath helper function (Daniel P. Berrange) - Move check for cgroup devices ACL upfront in LXC hotplug (Daniel P. Berrange) - Disks are always block devices, never character devices (Daniel P. Berrange) - Record hotplugged USB device in LXC live guest config (Daniel P. Berrange) - qemu: Implement VIR_DOMAIN_TAINT_HOOK (Michal Privoznik) - virDomainTaintFlags: Introduce VIR_DOMAIN_TAINT_HOOK (Michal Privoznik) - Add tests for secret XML parsing (Ján Tomko) - Forgot to add lxcconf2xmldata to dist. (Cédric Bosdonnat) - Support IPv6 in port allocator (Ján Tomko) - Split out bind() from virPortAllocatorAcquire (Ján Tomko) - qemu: snapshot: Add support for external active snapshots on gluster (Peter Krempa) - qemu: snapshot: Use new APIs to detect presence of existing storage files (Peter Krempa) - qemu: Switch snapshot deletion to the new API functions (Peter Krempa) - storage: Add storage file backends for gluster (Peter Krempa) - storage: add file functions for local and block files (Peter Krempa) - storage: Add file storage APIs in the default storage driver (Peter Krempa) - conf: Move qemuSnapshotDiskGetActualType to virDomainSnapshotDiskGetActualType (Peter Krempa) - conf: Move qemuDiskGetActualType to virDomainDiskGetActualType (Peter Krempa) - spec: add missing dep of libvirt-daemon-config-nwfilter (Eric Blake) - lxc from native: removed now remaining useless line (Cédric Bosdonnat) - Fix stream related spelling mistakes (Philipp Hahn) - LXC from native: convert blkio throttle config (Cédric Bosdonnat) - LXC from native: map vlan network type (Cédric Bosdonnat) - LXC from native: map block filesystems (Cédric Bosdonnat) - LXC from native: map lxc.arch to /domain/os/type@arch (Cédric Bosdonnat) - LXC from native: add lxc.cgroup.blkio.* mapping (Cédric Bosdonnat) - LXC from native: map lxc.cgroup.cpuset.* (Cédric Bosdonnat) - LXC from native: map lxc.cgroup.cpu.* (Cédric Bosdonnat) - LXC from native: migrate memory tuning (Cédric Bosdonnat) - LXC from native: convert lxc.id_map into <idmap> (Cédric Bosdonnat) - LXC from native: convert macvlan network configuration (Cédric Bosdonnat) - LXC from native: convert lxc.tty to console devices (Cédric Bosdonnat) - LXC from native: convert phys network types to net hostdev devices (Cédric Bosdonnat) - LXC from native: migrate veth network configuration (Cédric Bosdonnat) - LXC from native: implement no network conversion (Cédric Bosdonnat) - LXC from native: migrate fstab and lxc.mount.entry (Cédric Bosdonnat) - LXC from native: import rootfs (Cédric Bosdonnat) - LXC driver: started implementing connectDomainXMLFromNative (Cédric Bosdonnat) - Improve virConf parse to handle LXC config format (Cédric Bosdonnat) - event: pass reason for PM events (Eric Blake) - event: convert remaining domain events to new style (Eric Blake) - event: client RPC protocol tweaks for domain lifecycle events (Eric Blake) - event: prepare client to track domain callbackID (Eric Blake) - event: server RPC protocol tweaks for domain lifecycle events (Eric Blake) - event: dynamically manage server-side RPC domain events (Eric Blake) - qemu: Implement a stub cpuArchDriver.baseline() handler for aarch64 (Oleg Strikov) - libxl: register for domain events immediately after creation (Jim Fehlig) - libxl: rename libxlCreateDomEvents to libxlDomEventsRegister (Jim Fehlig) - vircapstest: Introduce virCapabilitiesGetCpusForNodemask test (Pradipta Kr. Banerjee) - Handle non-sequential NUMA node numbers (Pradipta Kr. Banerjee) - storage: gluster: Set volume metadata in a separate function (Peter Krempa) - qemu: introduce spiceport chardev backend (Martin Kletzander) - qemu: remove pointless condition (Martin Kletzander) - qemu: rework '-serial none' (Martin Kletzander) - conf: introduce spiceport chardev backend (Martin Kletzander) - rbd: Use rbd_create3 to create RBD format 2 images by default (Wido den Hollander) - build: correctly check for SOICGIFVLAN GET_VLAN_VID_CMD command (Laine Stump) - virNetworkLoadState: Disallow mangled 'floor' element (Michal Privoznik) - networkStartNetwork: Be more verbose (Michal Privoznik) - qemu: hyperv: Add support for timer enlightenments (Peter Krempa) - conf: Enforce supported options for certain timers (Peter Krempa) - schema: Fix guest timer specification schema according to the docs (Peter Krempa) - apparmor: Improve profiles (Felix Geyer) - Add glusterfs to VIR_CONNECT_LIST_STORAGE_POOLS_FILTERS_POOL_TYPE (Christophe Fergeau) - libxl: remove unneeded locking of driver when restoring (Jim Fehlig) - libxl: improve subprocess handling (Jim Fehlig) - libxl: handle domain shutdown events in a thread (Jim Fehlig) - libxl: remove list of timer registrations from libxlDomainObjPrivate (Jim Fehlig) - libxl: fix leaking libxlDomainObjPrivate (Jim Fehlig) - qemu_driver: Introduce <filesystem/> support in device attach/detach (Matthieu Coudron) - virDomainHostdev{Insert,Delete}: Replace VIR_REALLOC_N by VIR_{APPEND,DELETE}_ELEMENT (Matthieu Coudron) - qemuxml2argvtest: Set timezone (Michal Privoznik) - virsh: only report filled values in nodecpustats (Ján Tomko) - BSD: implement nodeGetCPUStats (Roman Bogorodskiy) - qemu: blockjob: Print correct file name in error message (Peter Krempa) - maint: Change the text of the NULLSTR() macro to "<null>" (Peter Krempa) - qemuxml2argvtest: Test localtime clock basis (Michal Privoznik) - qemuBuildClockArgStr: Allow localtime clock basis (Michal Privoznik) - Generate a valid imagelabel even for type 'none' (Ján Tomko) - event: move event filtering to daemon (regression fix) (Eric Blake) - rpm: create libvirt-wireshark sub-package (Eric Blake) - tests: Add test for new virkmod functions (John Ferlan) - utils: Introduce functions for kernel module manipulation (John Ferlan) - network: change default of forwardPlainNames to 'yes' (Laine Stump) - network: only prevent forwarding of DNS requests for unqualified names (Laine Stump) - virnetdevbandwidthtest: Link with libxml2 (Michal Privoznik) - spice: don't force user to specify spicevmc channel (Martin Kletzander) - virnetdevbandwidthtest: Introduce some more tests (Michal Privoznik) - virnetdevbandwidthtest: fix hard coded /sbin/tc (Cédric Bosdonnat) - Push nwfilter update locking up to top level (Daniel P. Berrange) - Add a read/write lock implementation (Daniel P. Berrange) - tests: Modify the scsi util tests (Osier Yang) - util: Accept test data path for scsi device's sg_path (Osier Yang) - tests: Add tests for scsi utils (Osier Yang) - util: Add one argument for several scsi utils (Osier Yang) - qemu: Don't fail if the SCSI host device is shareable between domains (Osier Yang) - virnetdevbandwidthtest: Introduce testVirNetDevBandwidthSet (Michal Privoznik) - virCommand: Introduce virCommandSetDryRun (Michal Privoznik) - snapshot: Add support for specifying snapshot disk backing type (Peter Krempa) - tests: Add more tests for virConnectBaselineCPU (Jiri Denemark) - cpu: Try to use source CPU model in virConnectBaselineCPU (Jiri Denemark) - cpu: Fix VIR_CONNECT_BASELINE_CPU_EXPAND_FEATURES (Jiri Denemark) - tests: Better support for VIR_CONNECT_BASELINE_CPU_EXPAND_FEATURES (Jiri Denemark) - Reword error message for oversized cpu time fields (Ján Tomko) - Simplify linuxNodeGetCPUStats (Ján Tomko) - Add hw random number generator (/dev/hwrng) to cgroup ACL (Pradipta Kr. Banerjee) - tests: Introduce virnetdevbandwidthtest (Michal Privoznik) - Add test for linuxNodeGetCPUStats (Ján Tomko) - Move test-local declarations to nodeinfopriv.h (Ján Tomko) - qemu: Enable 'host-passthrough' cpu mode for aarch64 (Oleg Strikov) - Block info query: Add check for transient domain (John Ferlan) - maint: update to latest gnulib, for mingw improvements (Eric Blake) - util: Add "shareable" field for virSCSIDevice struct (Osier Yang) - storage: Fix autostart of pool with "fc_host" type adapter (Osier Yang) - api: require write permission for guest agent interaction (Eric Blake) - virtlockd: make re-exec more robust (Michael Chapman) - build: add $(prefix) to SYSTEMD_UNIT_DIR (Laine Stump) - spice: expose the QEMU disable file transfer option (Francesco Romani) - spice: detect if qemu can disable file transfer (Francesco Romani) - lxc: allow to setup throttle blkio cgroup through virsh (Gao feng) - Add test for transient disk support in VMX files (Wout Mertens) - Make syntax check notice assignments w/o surrounding spaces. (Thorsten Behrens) - maint: align whitespaces with project conventions. (Thorsten Behrens) - virpcitest: Test virPCIDeviceDetach failure (Jiri Denemark) - virpcimock: Add PCI driver which always fails (Jiri Denemark) - virpcitest: More tests for device detach and reattach (Jiri Denemark) - virpcimock: Mock /sys/bus/pci/drivers_probe (Jiri Denemark) - pci: Publish some internal code for virpcitest (Jiri Denemark) - virpcitest: Show PCI device tested by each test (Jiri Denemark) - pci: Make reattach work for unbound devices (Jiri Denemark) - qemu: allow to setup throttle blkio cgroup through virsh (Gao feng) - virsh: add setting throttle blkio cgroup option to blkiotune (Gao feng) - blkio: Setting throttle blkio cgroup for domain (Gao feng) - domain: introduce xml elements for throttle blkio cgroup (Gao feng) - maint: replace remaining virLib*Error with better names (Eric Blake) - maint: simplify driver registration at startup (Eric Blake) - maint: clean up error reporting in migration (Eric Blake) - maint: don't lose error on canceled migration (Eric Blake) - maint: avoid nested use of virConnect{Ref,Close} (Eric Blake) - maint: don't leave garbage on early API exit (Eric Blake) - qemu: Change the default unix monitor timeout (Martin Kletzander) - storage: Sheepdog: Separate creating of the volume from building (Peter Krempa) - storage: RBD: Separate creating of the volume from building (Peter Krempa) - storage: Support deletion of volumes on gluster pools (Peter Krempa) - conf: Always use VIR_ERR_CONFIG_UNSUPPORTED on enumFromString() failures (Christophe Fergeau) Cleanups: - build-sys: Removed unused variable from configure.ac (Christophe Fergeau) - qemu: remove memset params array to zero in qemuDomainGetPercpuStats (Gao feng) - util: Fix the indention (Osier Yang) - virsh: Fix the string breaking style (Osier Yang) thanks everybody for the help with those release, be it with patches, bug reports, reviews, documentation, localizations, etc... Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

4 6

[libvirt] [PATCH] virscsi: Introduce virSCSIDeviceUsedByInfoFree
by John Ferlan 07 Mar '14

07 Mar '14

This resolves a Coverity RESOURCE_LEAK issue introduced by commit id 'de6fa535' where the virSCSIDeviceSetUsedBy() didn't VIR_FREE the 'copy' or possibly VIR_STRDUP()'d values. Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- src/util/virscsi.c | 26 +++++++++++++++----------- 1 file changed, 15 insertions(+), 11 deletions(-) diff --git a/src/util/virscsi.c b/src/util/virscsi.c index 69eae24..66e3161 100644 --- a/src/util/virscsi.c +++ b/src/util/virscsi.c @@ -268,6 +268,14 @@ cleanup: return ret; } +static void +virSCSIDeviceUsedByInfoFree(virUsedByInfoPtr used_by) +{ + VIR_FREE(used_by->drvname); + VIR_FREE(used_by->domname); + VIR_FREE(used_by); +} + void virSCSIDeviceFree(virSCSIDevicePtr dev) { @@ -279,11 +287,8 @@ virSCSIDeviceFree(virSCSIDevicePtr dev) VIR_FREE(dev->id); VIR_FREE(dev->name); VIR_FREE(dev->sg_path); - for (i = 0; i < dev->n_used_by; i++) { - VIR_FREE(dev->used_by[i]->drvname); - VIR_FREE(dev->used_by[i]->domname); - VIR_FREE(dev->used_by[i]); - } + for (i = 0; i < dev->n_used_by; i++) + virSCSIDeviceUsedByInfoFree(dev->used_by[i]); VIR_FREE(dev->used_by); VIR_FREE(dev); } @@ -296,10 +301,11 @@ virSCSIDeviceSetUsedBy(virSCSIDevicePtr dev, virUsedByInfoPtr copy; if (VIR_ALLOC(copy) < 0) return -1; - if (VIR_STRDUP(copy->drvname, drvname) < 0) - return -1; - if (VIR_STRDUP(copy->domname, domname) < 0) + if (VIR_STRDUP(copy->drvname, drvname) < 0 || + VIR_STRDUP(copy->domname, domname) < 0) { + virSCSIDeviceUsedByInfoFree(copy); return -1; + } return VIR_APPEND_ELEMENT(dev->used_by, dev->n_used_by, copy); } @@ -449,9 +455,7 @@ virSCSIDeviceListDel(virSCSIDeviceListPtr list, if (STREQ_NULLABLE(dev->used_by[i]->drvname, drvname) && STREQ_NULLABLE(dev->used_by[i]->domname, domname)) { if (dev->n_used_by > 1) { - VIR_FREE(dev->used_by[i]->drvname); - VIR_FREE(dev->used_by[i]->domname); - VIR_FREE(dev->used_by[i]); + virSCSIDeviceUsedByInfoFree(dev->used_by[i]); VIR_DELETE_ELEMENT(dev->used_by, i, dev->n_used_by); } else { tmp = virSCSIDeviceListSteal(list, dev); -- 1.8.5.3

3 3

[libvirt] [PATCH] add option to enforce minimal pagesize for hugetlbfs backed guests
by Marcelo Tosatti 07 Mar '14

07 Mar '14

Require a minimal pagesize for hugetlbfs backed guests. Fail guest initialization if hugetlbfs mount is configured with smaller page size. Signed-off-by: Marcelo Tosatti <mtosatti(a)redhat.com> diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index fd02864..e28d182 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -632,6 +632,10 @@ <dt><code>hugepages</code></dt> <dd>This tells the hypervisor that the guest should have its memory allocated using hugepages instead of the normal native page size.</dd> + <dt><code>pagesize</code></dt> + <dd>This tells the hypervisor that the guest should refuse to start + in case of failure to allocate guest memory with hugepages equal + to or larger than the specified size</dd> <dt><code>nosharepages</code></dt> <dd>Instructs hypervisor to disable shared pages (memory merge, KSM) for this domain. <span class="since">Since 1.0.6</span></dd> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 28e24f9..babb745 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -11274,6 +11274,10 @@ virDomainDefParseXML(xmlDocPtr xml, &def->mem.swap_hard_limit, false) < 0) goto error; + if (virDomainParseMemory("./memoryBacking/hugepages/pagesize[1]", ctxt, + &def->mem.page_size, false) < 0) + goto error; + n = virXPathULong("string(./vcpu[1])", ctxt, &count); if (n == -2) { virReportError(VIR_ERR_XML_ERROR, "%s", diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index d8f2e49..03a900d 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -1984,6 +1984,7 @@ struct _virDomainDef { unsigned long long soft_limit; /* in kibibytes */ unsigned long long min_guarantee; /* in kibibytes */ unsigned long long swap_hard_limit; /* in kibibytes */ + unsigned long long page_size; /* in kibibytes */ } mem; unsigned short vcpus; unsigned short maxvcpus; diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 8bcd98e..cd5e1c8 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -3570,6 +3570,33 @@ cleanup: return ret; } +#ifdef __linux__ + +#include <sys/vfs.h> + +#define HUGETLBFS_MAGIC 0x958458f6 + +static long gethugepagesize(const char *path) +{ + struct statfs fs; + int ret; + + do { + ret = statfs(path, &fs); + } while (ret != 0 && errno == EINTR); + + if (ret != 0) { + perror(path); + return 0; + } + + if (fs.f_type != HUGETLBFS_MAGIC) + return 0; + + return fs.f_bsize; +} +#endif + int qemuProcessStart(virConnectPtr conn, virQEMUDriverPtr driver, @@ -3712,6 +3739,31 @@ int qemuProcessStart(virConnectPtr conn, "%s", _("Unable to set huge path in security driver")); goto cleanup; } + + if (vm->def->mem.page_size) { +#ifdef __linux__ + unsigned long hpagesize = gethugepagesize(cfg->hugepagePath); + + if (!hpagesize) { + virReportError(VIR_ERR_INTERNAL_ERROR, + "%s", _("Unable to stat hugepage path")); + goto cleanup; + } + + hpagesize /= 1024; + + if (hpagesize < vm->def->mem.page_size) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Error: hugetlbfs page size=%ld < pagesize=%lld"), + hpagesize, vm->def->mem.page_size); + goto cleanup; + } +#else + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + "%s", _("pagesize option unsupported")); + goto cleanup; +#endif + } } /* Ensure no historical cgroup for this VM is lying around bogus

3 2

[libvirt] [PATCH] libvirt-tck: prefer kvm if multiple domain types exist
by Mike Latimer 07 Mar '14

07 Mar '14

When matching capabilities of a guest, if multiple domain types exist (for example, 'qemu' and 'kvm') the order in which they are returned can change. To avoid unpredictable test results, this patch prefers kvm if that domain type exists. If not, the behavior matches what existed before, and the first domain type is returned. --- lib/Sys/Virt/TCK.pm | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/lib/Sys/Virt/TCK.pm b/lib/Sys/Virt/TCK.pm index b2c16e7..56547c6 100644 --- a/lib/Sys/Virt/TCK.pm +++ b/lib/Sys/Virt/TCK.pm @@ -502,7 +502,17 @@ sub match_kernel { my @domains = $caps->guest_domain_types($i); next unless int(@domains); - return ($domains[0], + # Prefer kvm if multiple domain types are returned + my $domain; + if (int(@domains) gt 1) { + for (my $j = 0 ; $j < int(@domains) ; $j++) { + $domain = "kvm" if ($domains[$j] eq "kvm"); + } + } + # If kvm was not found, default to the first one + $domain = $domains[0] if (!defined($domain)); + + return ($domain, $caps->guest_domain_emulator($i, $domains[0]), $caps->guest_domain_loader($i, $domains[0])); } -- 1.8.4.5

2 2

[libvirt] [PATCH 0/4] Fix for virIdentityGetSystem when SELinux is disabled
by Michael Chapman 07 Mar '14

07 Mar '14

If SELinux is compiled into libvirt but it is disabled on the host, libvirtd logs: error : virIdentityGetSystem:173 : Unable to lookup SELinux process context: Invalid argument on each and every client connection. This patch series adds a runtime check for SELinux to this function. I've added security_disable() to securityselinuxhelper so virIdentityGetSystem can be tested twice, once with SELinux enabled and once with it disabled. A few other libselinux functions have also been added, so now securityselinuxlabeltest and securityselinuxtest do not need to be skipped even when SELinux isn't enabled on the test system. Michael Chapman (4): tests: Flesh out securityselinuxhelper tests: SELinux tests do not need to be skipped virIdentityGetSystem: don't fail if SELinux is disabled tests: Test virIdentityGetSystem src/util/viridentity.c | 18 ++- tests/Makefile.am | 4 + tests/securityselinuxhelper.c | 162 ++++++++++++++++++++- tests/securityselinuxhelperdata/lxc_contexts | 5 + .../virtual_domain_context | 2 + .../virtual_image_context | 2 + tests/securityselinuxlabeltest.c | 3 - tests/securityselinuxtest.c | 3 - tests/viridentitytest.c | 75 +++++++++- 9 files changed, 254 insertions(+), 20 deletions(-) create mode 100644 tests/securityselinuxhelperdata/lxc_contexts create mode 100644 tests/securityselinuxhelperdata/virtual_domain_context create mode 100644 tests/securityselinuxhelperdata/virtual_image_context -- 1.8.5.3

2 5

[libvirt] Looking for project ideas and mentors for Google Summer of Code 2014
by Stefan Hajnoczi 07 Mar '14

07 Mar '14

KVM & libvirt: you are welcome to join the QEMU umbrella organization like last year. What is GSoC? Google Summer of Code 2014 (GSoC) provides funding for students to work on open source projects for 12-weeks over the summer. Open source organizations apply to participate and those accepted receive funding for one or more students. How we are participating QEMU has participated, together with KVM and libvirt, in past years and I would like to continue that tradition this year. We now need to collect a list of project ideas on our wiki. We also need mentors to volunteer. http://qemu-project.org/Google_Summer_of_Code_2014 Project ideas Please post project ideas on the wiki page below. Project ideas should be suitable as a 12-week project that a student fluent in C/Python/etc can complete. No prior knowledge of QEMU/KVM/libvirt internals can be assumed. http://qemu-project.org/Google_Summer_of_Code_2014 Mentors Please add your name to project ideas you are willing to mentor. In order to mentor you must be an established contributor (regularly contribute patches). You must be willing to spend about 5 hours per week from May 19 to August 18. I have CCed the 8 most active committers since QEMU 1.5.0 but everyone is invited. Official timeline: https://www.google-melange.com/gsoc/events/google/gsoc2014 Stefan

8 16

[libvirt] [PATCH 1/9] conf: eliminate hardcoded indent from all xml
by Laine Stump 07 Mar '14

07 Mar '14

Many of the domain xml format functions (including all of the device format functions) had hard-coded spaces, which made for incorrect indentation when those functions were called in a different context (for example, commit 2122cf39 added <interface> XML into the document provided to a network hook script, and in this case it should have been indented by 2 spaces, but was instead indented by 6 spaces). In that patch I mentioned doing a followup patch to make the device xml formatters more consistent. After doing that patch, it felt incomplete to not give the same treatment to the entire directory. The one downside to this series is that it may create merge conflicts during backports, but fortunately the conflicts should all be fairly easy to resolve. Laine Stump (9): conf: eliminate hardcoded indent from domain xml conf: eliminate hardcoded indent from domain snapshot xml conf: eliminate hardcoded indent from network xml conf: eliminate outmoded/odd indent method from interface xml conf: eliminate hardcoded indentation in nwfilter xml conf: eliminate hardcoded indentation in capabilities xml conf: eliminate hardcoded indentation in node device xml conf: eliminate hardcoded indent in volume/pool xml conf: eliminate hardcoded indentation in all remaining xml src/conf/capabilities.c | 183 ++++++----- src/conf/cpu_conf.c | 11 +- src/conf/domain_conf.c | 599 +++++++++++++++++++---------------- src/conf/interface_conf.c | 137 ++++---- src/conf/netdev_bandwidth_conf.c | 6 +- src/conf/netdev_vlan_conf.c | 6 +- src/conf/netdev_vport_profile_conf.c | 6 +- src/conf/network_conf.c | 8 +- src/conf/node_device_conf.c | 207 ++++++------ src/conf/nwfilter_conf.c | 94 ++---- src/conf/nwfilter_params.c | 6 +- src/conf/secret_conf.c | 20 +- src/conf/snapshot_conf.c | 48 +-- src/conf/storage_conf.c | 174 +++++----- src/conf/storage_encryption_conf.c | 6 +- 15 files changed, 811 insertions(+), 700 deletions(-) -- 1.8.5.3

2 22

[libvirt] Python-libvirt in a virtual environment.
by Sijo Jose 07 Mar '14

07 Mar '14

Hi, Could you guide me to install libvirt in a virtual environment..? OR How python-libvirt could be used within a virtual environment after installing it, in the system. Regards Sijo

3 4

[libvirt] [PATCH] spec: Let translations be properly updated
by Jiri Denemark 06 Mar '14

06 Mar '14

Libvirt tarball contains po/stamp-po file which prevents any po/*.gmo file to be regenerated even if a corresponding po/*.po file is newer. By removing the stamp-po file, all *.gmo files are properly updated if required. This allows downstreams to provide patches that update translations. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- libvirt.spec.in | 1 + 1 file changed, 1 insertion(+) diff --git a/libvirt.spec.in b/libvirt.spec.in index 014fe5d..23374e7 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -1421,6 +1421,7 @@ driver autoreconf -if %endif +rm -f po/stamp-po %configure %{?_without_xen} \ %{?_without_qemu} \ %{?_without_openvz} \ -- 1.9.0

2 4

[libvirt] [PATCH v2 1/5] Introduce virDomainFSFreeze() and virDomainFSThaw() public API
by Tomoki Sekiyama 06 Mar '14

06 Mar '14

These will freeze and thaw filesystems within guest. The APIs take flags arguments which are currently not used, for future extensions. Signed-off-by: Tomoki Sekiyama <tomoki.sekiyama(a)hds.com> --- include/libvirt/libvirt.h.in | 6 ++++ src/driver.h | 10 ++++++ src/libvirt.c | 70 ++++++++++++++++++++++++++++++++++++++++++ src/libvirt_public.syms | 5 +++ 4 files changed, 91 insertions(+) diff --git a/include/libvirt/libvirt.h.in b/include/libvirt/libvirt.h.in index 295d551..22f373b 100644 --- a/include/libvirt/libvirt.h.in +++ b/include/libvirt/libvirt.h.in @@ -5245,6 +5245,12 @@ int virDomainFSTrim(virDomainPtr dom, unsigned long long minimum, unsigned int flags); +int virDomainFSFreeze(virDomainPtr dom, + unsigned int flags); + +int virDomainFSThaw(virDomainPtr dom, + unsigned int flags); + /** * virSchedParameterType: * diff --git a/src/driver.h b/src/driver.h index fbfaac4..7e6aec0 100644 --- a/src/driver.h +++ b/src/driver.h @@ -1130,6 +1130,14 @@ typedef int unsigned int flags, int cancelled); +typedef int +(*virDrvDomainFSFreeze)(virDomainPtr dom, + unsigned int flags); + +typedef int +(*virDrvDomainFSThaw)(virDomainPtr dom, + unsigned int flags); + typedef struct _virDriver virDriver; typedef virDriver *virDriverPtr; @@ -1341,6 +1349,8 @@ struct _virDriver { virDrvDomainMigrateFinish3Params domainMigrateFinish3Params; virDrvDomainMigrateConfirm3Params domainMigrateConfirm3Params; virDrvConnectGetCPUModelNames connectGetCPUModelNames; + virDrvDomainFSFreeze domainFSFreeze; + virDrvDomainFSThaw domainFSThaw; }; diff --git a/src/libvirt.c b/src/libvirt.c index a385935..84c9783 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -20552,3 +20552,73 @@ error: virDispatchError(dom->conn); return -1; } + +/** + * virDomainFSFreeze: + * @dom: a domain object + * @flags: extra flags, not used yet, so callers should always pass 0 + * + * Freeze filesystems within the guest (hence guest agent may be + * required depending on hypervisor used). + * + * Returns 0 on success, -1 otherwise. + */ +int +virDomainFSFreeze(virDomainPtr dom, + unsigned int flags) +{ + VIR_DOMAIN_DEBUG(dom, "flags=%x", flags); + + virResetLastError(); + + virCheckDomainReturn(dom, -1); + virCheckReadOnlyGoto(dom->conn->flags, error); + + if (dom->conn->driver->domainFSFreeze) { + int ret = dom->conn->driver->domainFSFreeze(dom, flags); + if (ret < 0) + goto error; + return ret; + } + + virReportUnsupportedError(); + +error: + virDispatchError(dom->conn); + return -1; +} + +/** + * virDomainFSThaw: + * @dom: a domain object + * @flags: extra flags, not used yet, so callers should always pass 0 + * + * Thaw the frozen filesystems within the guest (hence guest agent + * may be required depending on hypervisor used). + * + * Returns 0 on success, -1 otherwise. + */ +int +virDomainFSThaw(virDomainPtr dom, + unsigned int flags) +{ + VIR_DOMAIN_DEBUG(dom, "flags=%x", flags); + + virResetLastError(); + + virCheckDomainReturn(dom, -1); + virCheckReadOnlyGoto(dom->conn->flags, error); + + if (dom->conn->driver->domainFSThaw) { + int ret = dom->conn->driver->domainFSThaw(dom, flags); + if (ret < 0) + goto error; + return ret; + } + + virReportUnsupportedError(); + +error: + virDispatchError(dom->conn); + return -1; +} diff --git a/src/libvirt_public.syms b/src/libvirt_public.syms index 6ed6ce6..9e49a9d 100644 --- a/src/libvirt_public.syms +++ b/src/libvirt_public.syms @@ -645,5 +645,10 @@ LIBVIRT_1.2.1 { virConnectNetworkEventDeregisterAny; } LIBVIRT_1.1.3; +LIBVIRT_1.2.3 { + global: + virDomainFSFreeze; + virDomainFSThaw; +} LIBVIRT_1.2.1; # .... define new API here using predicted next version number ....

1 0

[libvirt] [RFC][PATCH 0/2] Fix zero cpu shares handling
by Ján Tomko 06 Mar '14

06 Mar '14

Hello, This series tries to fix the behavior of <cputune><shares>0</shares></cputune> (means default shares of 1024) and virsh schedinfo --set cpu_shares=0 (means minimum of 2) If 0 was explicitly specified, treat it as a valid value (which will get converted to 2 by the kernel) Re-read the shares value to reflect the cpu shares in live XML. In v1, I've tried to change the schedinfo behavior instead, but I have not found a way to reset the value to the kernel default (which is currently 1024 - see ROOT_TASK_GROUP_LOAD in kernel/sched/sched.h in Linux source) other than writing it to the cgroup fs. Partially fixes https://bugzilla.redhat.com/show_bug.cgi?id=998431 Ján Tomko (2): Treat zero cpu shares as a valid value Show the real cpu shares value in live xml src/conf/domain_conf.c | 12 +++++++----- src/conf/domain_conf.h | 1 + src/lxc/lxc_cgroup.c | 13 ++++++++++--- src/lxc/lxc_driver.c | 8 +++++++- src/lxc/lxc_native.c | 8 +++++--- src/qemu/qemu_cgroup.c | 12 +++++++++--- src/qemu/qemu_driver.c | 13 +++++++++++-- src/vmx/vmx.c | 1 + 8 files changed, 51 insertions(+), 17 deletions(-) -- 1.8.3.2

2 7

[libvirt] [PATCH v2] qemu: Reject unsupported tuning in session mode
by Martin Kletzander 06 Mar '14

06 Mar '14

When domain is started with setting that cannot be done, i.e. those that require cgroups, there is no error reported and it succeeds without any message whatsoever. When setting with API, virsh, an error is reported, but only due to the fact that no cgroups are mounted (priv->cgroup == NULL). Given the above it seems reasonable to reject such unsupported settings. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1023366 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- Notes: v2: - Allow CPU pinning since setting affinity should still work. - Remove bogus 'cfg = ...' in SetBlockIoTune. src/qemu/qemu_command.c | 26 ++++++++++++++++++ src/qemu/qemu_driver.c | 73 ++++++++++++++++++++++++++++++++++++++++++++++++- 2 files changed, 98 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 18d0a64..611d21d 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -7762,6 +7762,32 @@ qemuBuildCommandLine(virConnectPtr conn, emulator = def->emulator; + if (!cfg->privileged) { + /* If we have no cgroups than we can have no tunings that + * require them */ + + if (def->mem.hard_limit || def->mem.soft_limit || + def->mem.min_guarantee || def->mem.swap_hard_limit) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Memory tuning is not available in session mode")); + goto error; + } + + if (def->blkio.weight || def->blkio.ndevices) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto error; + } + + if (def->cputune.shares || def->cputune.period || + def->cputune.quota || def->cputune.emulator_period || + def->cputune.emulator_quota) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto error; + } + } + for (i = 0; i < def->ngraphics; ++i) { switch (def->graphics[i]->type) { case VIR_DOMAIN_GRAPHICS_TYPE_SDL: diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 4fbcb27..1e50572 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -7434,6 +7434,8 @@ static char *qemuDomainGetSchedulerType(virDomainPtr dom, char *ret = NULL; virDomainObjPtr vm = NULL; qemuDomainObjPrivatePtr priv; + virQEMUDriverPtr driver = dom->conn->privateData; + virQEMUDriverConfigPtr cfg = NULL; if (!(vm = qemuDomObjFromDomain(dom))) goto cleanup; @@ -7443,6 +7445,13 @@ static char *qemuDomainGetSchedulerType(virDomainPtr dom, if (virDomainGetSchedulerTypeEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto cleanup; + } + /* Domain not running, thus no cgroups - return defaults */ if (!virDomainObjIsActive(vm)) { if (nparams) @@ -7469,6 +7478,7 @@ static char *qemuDomainGetSchedulerType(virDomainPtr dom, cleanup: if (vm) virObjectUnlock(vm); + virObjectUnref(cfg); return ret; } @@ -7683,6 +7693,12 @@ qemuDomainSetBlkioParameters(virDomainPtr dom, if (virDomainSetBlkioParametersEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -7863,6 +7879,7 @@ qemuDomainGetBlkioParameters(virDomainPtr dom, int ret = -1; virCapsPtr caps = NULL; qemuDomainObjPrivatePtr priv; + virQEMUDriverConfigPtr cfg = NULL; virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG | @@ -7881,6 +7898,13 @@ qemuDomainGetBlkioParameters(virDomainPtr dom, if (virDomainGetBlkioParametersEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -8269,6 +8293,7 @@ cleanup: if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } @@ -8316,6 +8341,12 @@ qemuDomainSetMemoryParameters(virDomainPtr dom, if (virDomainSetMemoryParametersEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("Memory tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -8423,6 +8454,7 @@ qemuDomainGetMemoryParameters(virDomainPtr dom, int ret = -1; virCapsPtr caps = NULL; qemuDomainObjPrivatePtr priv; + virQEMUDriverConfigPtr cfg = NULL; virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG | @@ -8439,6 +8471,13 @@ qemuDomainGetMemoryParameters(virDomainPtr dom, if (virDomainGetMemoryParametersEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, "%s", + _("Memory tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -8550,6 +8589,7 @@ cleanup: if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } @@ -8975,6 +9015,13 @@ qemuDomainSetSchedulerParametersFlags(virDomainPtr dom, if (virDomainSetSchedulerParametersFlagsEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -9210,6 +9257,7 @@ qemuDomainGetSchedulerParametersFlags(virDomainPtr dom, virDomainDefPtr persistentDef; virCapsPtr caps = NULL; qemuDomainObjPrivatePtr priv; + virQEMUDriverConfigPtr cfg = NULL; virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG | @@ -9226,6 +9274,13 @@ qemuDomainGetSchedulerParametersFlags(virDomainPtr dom, if (virDomainGetSchedulerParametersFlagsEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("CPU tuning is not available in session mode")); + goto cleanup; + } + if (*nparams > 1) cpu_bw_status = virCgroupSupportsCpuBW(priv->cgroup); @@ -9320,6 +9375,7 @@ cleanup: if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } @@ -15521,11 +15577,17 @@ qemuDomainSetBlockIoTune(virDomainPtr dom, if (virDomainSetBlockIoTuneEnsureACL(dom->conn, vm->def, flags) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto cleanup; + } + if (qemuDomainObjBeginJob(driver, vm, QEMU_JOB_MODIFY) < 0) goto cleanup; priv = vm->privateData; - cfg = virQEMUDriverGetConfig(driver); if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto endjob; @@ -15668,6 +15730,7 @@ qemuDomainGetBlockIoTune(virDomainPtr dom, int ret = -1; size_t i; virCapsPtr caps = NULL; + virQEMUDriverConfigPtr cfg = NULL; virCheckFlags(VIR_DOMAIN_AFFECT_LIVE | VIR_DOMAIN_AFFECT_CONFIG | @@ -15682,6 +15745,13 @@ qemuDomainGetBlockIoTune(virDomainPtr dom, if (virDomainGetBlockIoTuneEnsureACL(dom->conn, vm->def) < 0) goto cleanup; + cfg = virQEMUDriverGetConfig(driver); + if (!cfg->privileged) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Block I/O tuning is not available in session mode")); + goto cleanup; + } + if (!(caps = virQEMUDriverGetCapabilities(driver, false))) goto cleanup; @@ -15784,6 +15854,7 @@ cleanup: if (vm) virObjectUnlock(vm); virObjectUnref(caps); + virObjectUnref(cfg); return ret; } -- 1.9.0

2 4

[libvirt] virDomainShutdown() API failed with error '-1'
by Panday Ritesh Sharma (rpanday) 06 Mar '14

06 Mar '14

Hi Team, We are calling virDomainShutdown() API for shutting the VM. Sometime we see, this API is failing because of 'unknown' reason. It took almost 15 secs to bringdown the VM and at last failed. And the domain goes into shut off state. You can find the time stamp in below log. Could you please let us know in which case this API can fail, how to fix/debug this issue ? We are running 1.0.2 libvirtd version. =========== 20.53.48.532818176: Shuting down domain... 20.54.03.539286528: Failure of libvirt library call: Code: internal error Context: None Message: An error occurred, but the cause is unknown Level: ERROR 20.54.03.539291392: virDomainShutdown() call returned error -1 for CP—3 ============ =========== [host:/var/log/libvirt/qemu]$ virsh list --all Id Name State ---------------------------------------------------- - CP--3 shut off =========== ============ [host:/var/log/libvirt/qemu]$ [host:/var/log/libvirt/qemu]$ libvirtd --version libvirtd (libvirt) 1.0.2 [host:/var/log/libvirt/qemu]$ ============ Regards Ritesh Sharma

1 0

[libvirt] [PATCH 0/4] Add caching of QEMU probed capabilities
by Daniel P. Berrange 06 Mar '14

06 Mar '14

Probing capabilities takes 200-300ms per binary and we have as many as 26 binaries. This noticably slows down libvirtd startup. It does not look like performance of probing QEMU can be improved, so this series introduces caching of the capabilities information. So the first time libvirtd starts it'll be slow, but thereafter it is fast. The cache is invalidated any time the QEMU binary timestamp changes or the libvirtd binary or driver module timestamp changes. Daniel P. Berrange (4): Add helper APIs for generating cryptographic hashes Convert lock driver plugins to use new crypto APIs Cache result of QEMU capabilities extraction Refresh qemu capabilities if libvirtd binary changes .gitignore | 1 + daemon/libvirtd.c | 2 + include/libvirt/virterror.h | 1 + po/POTFILES.in | 1 + src/Makefile.am | 1 + src/driver.c | 2 + src/libvirt_private.syms | 6 + src/locking/lock_driver_lockd.c | 32 +-- src/locking/lock_driver_sanlock.c | 42 +--- src/qemu/qemu_capabilities.c | 418 +++++++++++++++++++++++++++++++++++++- src/qemu/qemu_capabilities.h | 2 + src/qemu/qemu_driver.c | 1 + src/util/vircrypto.c | 80 ++++++++ src/util/vircrypto.h | 40 ++++ src/util/virerror.c | 1 + src/util/virutil.c | 23 +++ src/util/virutil.h | 4 + tests/Makefile.am | 5 + tests/vircryptotest.c | 90 ++++++++ 19 files changed, 682 insertions(+), 70 deletions(-) create mode 100644 src/util/vircrypto.c create mode 100644 src/util/vircrypto.h create mode 100644 tests/vircryptotest.c -- 1.8.5.3

3 12

[libvirt] [RFC] support memory reserved feature and optimize mlock guest memory propose
by Zhanghailiang 06 Mar '14

06 Mar '14

Hi all： Currently, we use cgroup(memory) to support memory QoS on KVM platform, and use "mlock" on qemu to support "memory reserved". The "mlock" seems to be not appropriate. Now qemu "mlock" memory in the main thread, which would lock iothread (qemu_mutex_lock_iothread), if the memory size is large, that will consume lots of time. It means whenever we want to set a new 'mlock', the VM would be blocked for a while. Here is my optimization: 1. Add a global variable (lock_ram_size) to save the value of "memory reserved"; 2. Add a qmp commond "set_ram_minguarantee" to change lock_ram_size; 3. Create a new thread to mlock(lock_ram_zie), while is waked up by the "set_ram_minguarantee" qmp command. Flow chart: main funciton qmp command "set_ram_minguarantee" | | | | create "mlock" thread change value of lock_ram_size | | | | |------>thread wait<-------------wake up "mlock" thread | | | | | | |-------mlock(lock_ram_zie) We have tested this demo a few days, it seems to be worked well. But we are not sure is there any other problems , if the main thread and mlock thread access one memory zone at one time without a mutex lock. Is it workable?Or Is there any other idea to support "memory reserved" ? Thanks zhanghailiang

2 3

[libvirt] [PATCH v3 0/2] fix query-command-line-options
by Amos Kong 06 Mar '14

06 Mar '14

This patchset fixed some issues of query-command-line-options: * some new options that haven't argument can't be queried. (eg: -enable-fips) * some legacy options that have argument can't be queried. (eg: -vnc display) More discussion: http://marc.info/?l=qemu-devel&m=139081830416684&w=2 V2: remove duplicate option tables, update schema V3: fix typo in commitlog and export qemu_options talbe Amos Kong (2): qmp: rename query_option_descs() to get_param_infolist() query-command-line-options: query all the options in qemu-options.hx qapi-schema.json | 8 ++++++-- qemu-options.h | 18 ++++++++++++++++++ util/qemu-config.c | 41 ++++++++++++++++++++++++++++++++--------- vl.c | 17 ----------------- 4 files changed, 56 insertions(+), 28 deletions(-) -- 1.8.5.3

2 8

[libvirt] [PATCH 00/10] virFork cleanups
by Eric Blake 05 Mar '14

05 Mar '14

Some of these patches were written while working on CVE-2013-6456; we decided to reorder things and fix that problem first. While rebasing these patches to the latest tree, I found other things worth fixing. Eric Blake (10): nwfilter: don't ignore child process failures virFork: give specific status on failure prior to exec util: make it easier to reflect child exit status util: preserve exit status from mount namespace callback util: make it easier to grab only regular process exit util: make it easier to grab only regular command exit virFork: simplify semantics virt-login-shell: use single instead of double fork virt-login-shell: saner exit value virsh: report exit status of failed lxc-enter-namespace daemon/libvirtd.c | 4 +- daemon/remote.c | 7 +- docs/internals/command.html.in | 17 ++- src/access/viraccessdriverpolkit.c | 9 +- src/bhyve/bhyve_process.c | 19 +--- src/fdstream.c | 3 +- src/internal.h | 7 ++ src/libvirt.c | 4 +- src/libvirt_private.syms | 2 + src/lxc/lxc_container.c | 6 +- src/lxc/lxc_process.c | 11 +- src/nwfilter/nwfilter_ebiptables_driver.c | 89 ++++++--------- src/openvz/openvz_driver.c | 18 +--- src/qemu/qemu_capabilities.c | 1 + src/qemu/qemu_command.c | 3 +- src/storage/storage_backend_iscsi.c | 7 +- src/util/vircommand.c | 173 +++++++++++++++--------------- src/util/vircommand.h | 4 +- src/util/virebtables.c | 5 +- src/util/virfile.c | 35 ++---- src/util/viriptables.c | 7 +- src/util/virnetdevveth.c | 4 +- src/util/virprocess.c | 121 +++++++++++++++------ src/util/virprocess.h | 8 +- src/xen/xen_driver.c | 9 +- tests/commandtest.c | 126 +++++++++++++++++++++- tests/reconnect.c | 3 +- tests/statstest.c | 3 +- tests/testutils.c | 4 +- tools/virsh-domain.c | 30 +++--- tools/virsh.pod | 5 +- tools/virt-login-shell.c | 141 ++++++++++-------------- tools/virt-login-shell.pod | 25 ++++- 33 files changed, 525 insertions(+), 385 deletions(-) -- 1.8.5.3

5 21

[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

[libvirt] [PATCH 0/4] Expose FSFreeze/FSThaw within the guest as commands
by Tomoki Sekiyama 04 Mar '14

04 Mar '14

Currently FSFreeze and FSThaw are supported by qemu guest agent and they are used internally in snapshot-create command with --quiesce option. However, when users want to utilize the native snapshot feature of storage devices (such as LVM over iSCSI, various enterprise storage systems, etc.), they need to issue fsfreeze command separately from libvirt-driven snapshots. (OpenStack cinder provides these storages' snapshot feature, but it cannot quiesce the guest filesystems automatically for now.) Although virDomainQemuGuestAgent() API could be used for this purpose, it depends too much on specific hypervisor implementation. This patchset adds virDomainFSFreeze()/virDomainFSThaw() APIs and virsh domfsfreeze/domfsthaw commands to enable the users to freeze and thaw domain's filesystems cleanly. The APIs has mountPoint and flags option currently unsupported for future extension, as virDomainFSTrim() API. Duplicated FSFreeze results in error caused by qemu guest agent. --- Tomoki Sekiyama (4): Introduce virDomainFSFreeze() public API remote: Implement virDomainFSFreeze and virDomainFSThaw qemu: Implement virDomainFSFreeze virsh: Expose new virDomainFSFreeze and virDomainFSThaw API include/libvirt/libvirt.h.in | 8 ++ src/access/viraccessperm.c | 2 - src/access/viraccessperm.h | 6 ++ src/driver.h | 12 ++++ src/libvirt.c | 92 +++++++++++++++++++++++++++ src/libvirt_public.syms | 6 ++ src/qemu/qemu_driver.c | 142 ++++++++++++++++++++++++++++++++++++++++++ src/remote/remote_driver.c | 2 + src/remote/remote_protocol.x | 26 +++++++- src/remote_protocol-structs | 12 ++++ src/rpc/gendispatch.pl | 2 + tools/virsh-domain.c | 108 ++++++++++++++++++++++++++++++++ tools/virsh.pod | 17 +++++ 13 files changed, 433 insertions(+), 2 deletions(-)

3 10

[libvirt] [PATCH 0/2] Properly error out on unsupported settings
by Martin Kletzander 04 Mar '14

04 Mar '14

<BLURB/> Martin Kletzander (2): tests: Create privileged driver config in qemuxml2argvtest qemu: Reject unsupported tuning in session mode src/qemu/qemu_command.c | 32 +++++++++++++++ src/qemu/qemu_driver.c | 103 ++++++++++++++++++++++++++++++++++++++++++++++- tests/qemuxml2argvtest.c | 2 +- 3 files changed, 135 insertions(+), 2 deletions(-) -- 1.9.0

2 7

[libvirt] [PATCH v2 0/2] Introduce max_anonymous_clients
by Michal Privoznik 04 Mar '14

04 Mar '14

https://bugzilla.redhat.com/show_bug.cgi?id=981729 So far we can limit how many clients are connected, how many are waiting in accept() line but we could not control the count of accepted but not authenticated yet. Michal Privoznik (2): virNetServer: Introduce unauth clients counter daemon: Introduce max_anonymous_clients daemon/libvirtd-config.c | 1 + daemon/libvirtd-config.h | 1 + daemon/libvirtd.aug | 1 + daemon/libvirtd.c | 1 + daemon/libvirtd.conf | 4 +++ daemon/remote.c | 21 ++++++----- daemon/test_libvirtd.aug.in | 1 + src/locking/lock_daemon.c | 3 +- src/lxc/lxc_controller.c | 2 +- src/rpc/virnetserver.c | 88 ++++++++++++++++++++++++++++++++++++++++++--- src/rpc/virnetserver.h | 3 ++ 11 files changed, 110 insertions(+), 16 deletions(-) -- 1.9.0

2 8