September 2013 - Devel - libvirt List Archives

[libvirt] [PATCH 0/5] Change preference of default PCI passthrough type
by Peter Krempa 23 Sep '13

23 Sep '13

Change the default type to VFIO if available. Please note that the majority of the code here is only compile tested as I don't have appropriate hardware to test it. Peter Krempa (5): qemu: hostdev: Refactor PCI passhrough handling qemu: hostdev: Add checks if PCI passthrough is availabe in the host qemu: hotplug: Verify that the selected pci passhtrough option is present qemu: Check if PCI passhtrough is available in the host when starting qemu: Prefer VFIO for PCI device passthrough docs/formatdomain.html.in | 9 ++- src/conf/domain_conf.h | 2 +- src/qemu/qemu_command.c | 27 +++++--- src/qemu/qemu_hostdev.c | 76 ++++++++++++++++++++++ src/qemu/qemu_hostdev.h | 3 + src/qemu/qemu_hotplug.c | 63 +++++++++++++++--- src/qemu/qemu_process.c | 57 ++++++++++++++++ .../qemuxml2argv-hostdev-pci-address-device.args | 2 +- .../qemuxml2argvdata/qemuxml2argv-net-hostdev.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-pci-rom.args | 4 +- 10 files changed, 216 insertions(+), 29 deletions(-) -- 1.8.3.2

3 11

[libvirt] [PATCH] virsh: Add missing "async" option in virsh-domain.c opts_block_commit.
by Simone Gotti 23 Sep '13

23 Sep '13

After commit 8aecd351266a66efa59b7f7be77bf66693d99ce0 it'll detect that a required option is not defined and it will assert and exit with: virsh.c:1364: vshCommandOpt: Assertion `valid->name' failed. --- tools/virsh-domain.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 49cd154..856e888 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -1544,6 +1544,10 @@ static const vshCmdOptDef opts_block_commit[] = { .type = VSH_OT_INT, .help = N_("with --wait, abort if copy exceeds timeout (in seconds)") }, + {.name = "async", + .type = VSH_OT_BOOL, + .help = N_("with --wait, don't wait for cancel to finish") + }, {.name = NULL} }; -- 1.8.3.1

2 3

[libvirt] [PATCH] conf: Do better job when comparing features ABI compatibility
by Peter Krempa 23 Sep '13

23 Sep '13

The ABI compatibility check for domain features didn't check the expanded HyperV and APIC EOI values, thus possibly allowing change in guest ABI. Add the check and use typecasted switch statement to warn developers when adding a new HyperV feature. --- src/conf/domain_conf.c | 72 ++++++++++++++++++++++++++++++++++++++++++++++---- 1 file changed, 67 insertions(+), 5 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index c4b6464..8953579 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -13354,6 +13354,72 @@ virDomainRedirFilterDefCheckABIStability(virDomainRedirFilterDefPtr src, return true; } + +static bool +virDomainDefFeaturesCheckABIStability(virDomainDefPtr src, + virDomainDefPtr dst) +{ + size_t i; + + /* basic check */ + if (src->features != dst->features) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("Target domain features %d does not match source %d"), + dst->features, src->features); + return false; + } + + /* APIC EOI */ + if (src->apic_eoi != dst->apic_eoi) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("State of APIC EOI differs: " + "source: '%s', destination: '%s'"), + virDomainFeatureStateTypeToString(src->apic_eoi), + virDomainFeatureStateTypeToString(dst->apic_eoi)); + return false; + } + + /* hyperv */ + if (src->features & (1 << VIR_DOMAIN_FEATURE_HYPERV)) { + for (i = 0; i < VIR_DOMAIN_HYPERV_LAST; i++) { + switch ((enum virDomainHyperv) i) { + case VIR_DOMAIN_HYPERV_RELAXED: + case VIR_DOMAIN_HYPERV_VAPIC: + if (src->hyperv_features[i] != dst->hyperv_features[i]) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("State of HyperV enlightenment " + "feature '%s' differs: " + "source: '%s', destination: '%s'"), + virDomainHypervTypeToString(i), + virDomainFeatureStateTypeToString(src->hyperv_features[i]), + virDomainFeatureStateTypeToString(dst->hyperv_features[i])); + return false; + } + + break; + + case VIR_DOMAIN_HYPERV_SPINLOCKS: + /* spinlock count matters! */ + if (src->hyperv_spinlocks != dst->hyperv_spinlocks) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("HyperV spinlock retry count differs: " + "source: '%u', destination: '%u'"), + src->hyperv_spinlocks, + dst->hyperv_spinlocks); + return false; + } + break; + + case VIR_DOMAIN_HYPERV_LAST: + break; + } + } + } + + return true; +} + + /* This compares two configurations and looks for any differences * which will affect the guest ABI. This is primarily to allow * validation of custom XML config passed in during migration @@ -13455,12 +13521,8 @@ virDomainDefCheckABIStability(virDomainDefPtr src, return false; } - if (src->features != dst->features) { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, - _("Target domain features %d does not match source %d"), - dst->features, src->features); + if (!virDomainDefFeaturesCheckABIStability(src, dst)) return false; - } if (src->clock.ntimers != dst->clock.ntimers) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", -- 1.8.3.2

2 2

[libvirt] [v4 0/5] virsh: More intelligent auto-completion
by Tomas Meszaros 22 Sep '13

22 Sep '13

This series implements advanced auto-completion for virsh. Using custom option completers (examples in 3/5, 4/5 and 5/5), it is possible to further complete any option with defined completer function. Option completer function should return list of completed names for each desired option. Using C99 initialization, it is easy to enable specific completer for any option. For example, when we want domain completer for "start --domain" option, we write vshDomainCompleter() and along with desired flags, initialize targeted opts struct. static const vshCmdOptDef opts_start[] = { {.name = "domain", .type = VSH_OT_DATA, .flags = VSH_OFLAG_REQ, .help = N_("name of the inactive domain"), .completer = vshDomainCompleter, .completer_flags = VIR_CONNECT_LIST_DOMAINS_INACTIVE | VIR_CONNECT_LIST_DOMAINS_PERSISTENT }, Auto-completion itself then works like this: virsh # start --domain <TAB> domain1 domain2 domainN virsh # vol-key <TAB> --help --pool --vol vol1 vol2 --- v4: * rewritten to use only option completers v3: https://www.redhat.com/archives/libvir-list/2013-August/msg01294.html v2: https://www.redhat.com/archives/libvir-list/2013-August/msg00992.html v1: https://www.redhat.com/archives/libvir-list/2013-August/msg00371.html Tomas Meszaros (5): virsh: Add vshCmdCompleter and vshOptCompleter virsh: Improve readline generators and readline completion virsh: Add vshDomainCompleter virsh: Add vshSuspendTargetCompleter virsh: Add vshRebootShutdownModeCompleter tools/virsh-domain-monitor.c | 52 +++++-- tools/virsh-domain.c | 310 +++++++++++++++++++++++++++++--------- tools/virsh-snapshot.c | 50 +++++-- tools/virsh.c | 350 +++++++++++++++++++++++++++++++++++++++---- tools/virsh.h | 8 + 5 files changed, 648 insertions(+), 122 deletions(-) -- 1.8.3.1

2 8

Re: [libvirt] [Users] Starting VM Error
by Dan Kenigsberg 22 Sep '13

22 Sep '13

On Sun, Sep 15, 2013 at 01:59:29PM +0200, Gianluca Cecchi wrote: > On Sun, Sep 15, 2013 at 1:50 PM, Gianluca Cecchi wrote: > > and > > [root@tekkaman ~]# chgrp kvm /var/lib/libvirt/images/testvm.img > [root@tekkaman ~]# ll /var/lib/libvirt/images/testvm.img > -rw-rw----. 1 vdsm kvm 197120 Sep 15 12:24 /var/lib/libvirt/images/testvm.img > > [root@tekkaman ~]# virsh undefine testvm2 > Please enter your authentication name: virshuser > Please enter your password: > Domain testvm2 has been undefined > > [root@tekkaman ~]# virsh define /tmp/testvm2.xml > Please enter your authentication name: virshuser > Please enter your password: > Domain testvm2 defined from /tmp/testvm2.xml > > [root@tekkaman ~]# virsh start testvm2 > Please enter your authentication name: virshuser > Please enter your password: > error: Failed to start domain testvm2 > error: internal error Process exited while reading console log output: > char device redirected to /dev/pts/2 > qemu-kvm: -drive > file=/var/lib/libvirt/images//testvm.img,if=none,id=drive-virtio-disk0,format=qcow2: > could not open disk image /var/lib/libvirt/images//testvm.img: > Permission denied Gianluca, due to all the twists of this thread, I no longer recall which libvirt version you are using. Would you remind us? All signs show that it starts up qemu without setting its auxiliary group properly (Assuming that your # groups qemu qemu : qemu kvm is just as mine) I've added libvir-list so the fine folks there could check why qemu cannot read/write files group-owned by kvm. Dan.

3 5

[libvirt] ANNOUNCE: libvirt 0.10.2.8 maintenance release
by Cole Robinson 20 Sep '13

20 Sep '13

libvirt 0.10.2.8 maintenance release is now available. This is libvirt 0.10.2 with additional bugfixes that have accumulated upstream since the initial release. This release can be downloaded at: http://libvirt.org/sources/stable_updates/libvirt-0.10.2.8.tar.gz Changes in this version: * virsh: fix change-media bug on disk block type * libvirt: lxc: don't mkdir when selinux is disabled * Fix crash in remoteDispatchDomainMemoryStats (CVE-2013-4296) * Add support for using 3-arg pkcheck syntax for process (CVE-2013-4311) * Include process start time when doing polkit checks * win32: Pretend that close-on-exec works * virDomainDefParseXML: set the argument of virBitmapFree to NULL after calling virBitmapFree * security: provide supplemental groups even when parsing label (CVE-2013-4291) * virbitmap: Refactor virBitmapParse to avoid access beyond bounds of array * bitmap: add virBitmapCountBits For info about past maintenance releases, see: http://wiki.libvirt.org/page/Maintenance_Releases Thanks, Cole

1 0

[libvirt] [PATCH] qemu: use "ide" as device name for implicit SATA controller on Q35
by Laine Stump 20 Sep '13

20 Sep '13

This resolves https://bugzilla.redhat.com/show_bug.cgi?id=1008903 The Q35 machinetype has an implicit SATA controller at 00:1F.2 which isn't given the "expected" id of ahci0 by qemu when it's created. The original suggested solution to this problem was to not specify any controller for the disks that use the default controller and just specify "unit=n" instead; qemu should then use the first ide or sata controller for the disk. Unfortunately, this "solution" is ignorant of the fact that in the case of SATA disks, the "unit" attribute in the disk XML is actually *not* being used for the unit, but is instead used to specify the "bus" number; each SATA controller has 6 buses, and each bus only allows a single unit. This makes it nonsensical to specify unit='n' where n is anything other than 0. It also means that the only way to connect more than a single device to the implicit SATA controller is to explicitly give the bus names, which happen to be "ide.$n", where $n can be replaced by the disk's "unit" number. --- src/qemu/qemu_command.c | 15 ++++++--------- tests/qemuxml2argvdata/qemuxml2argv-pcihole64-q35.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-q35.args | 2 +- 3 files changed, 8 insertions(+), 11 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 4628dac..e6239c9 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -4383,18 +4383,15 @@ qemuBuildDriveDevStr(virDomainDefPtr def, if (qemuDomainMachineIsQ35(def) && disk->info.addr.drive.controller == 0) { /* Q35 machines have an implicit ahci (sata) controller at - * 00:1F.2 which has no "id" associated with it. For this - * reason, we can't refer to it as "ahci0". Instead, we - * don't give an id, which qemu interprets as "use the - * first ahci controller". We then need to specify the - * unit with "unit=%d" rather than adding it onto the bus - * arg. + * 00:1F.2 which for some reason is hardcoded with the id + * "ide" instead of the seemingly more reasonable "ahci0" + * or "sata0". */ - virBufferAsprintf(&opt, ",unit=%d", disk->info.addr.drive.unit); + virBufferAsprintf(&opt, ",bus=ide.%d", disk->info.addr.drive.unit); } else { /* All other ahci controllers have been created by - * libvirt, so they *do* have an id, and we can identify - * them that way. + * libvirt, and we gave them the id "ahci${n}" where ${n} + * is the controller number. So we identify them that way. */ virBufferAsprintf(&opt, ",bus=ahci%d.%d", disk->info.addr.drive.controller, diff --git a/tests/qemuxml2argvdata/qemuxml2argv-pcihole64-q35.args b/tests/qemuxml2argvdata/qemuxml2argv-pcihole64-q35.args index e10ccb1..175c0dc 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-pcihole64-q35.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-pcihole64-q35.args @@ -5,5 +5,5 @@ LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test QEMU_AUDIO_DRV=none \ -device i82801b11-bridge,id=pci.1,bus=pcie.0,addr=0x2 \ -device pci-bridge,chassis_nr=2,id=pci.2,bus=pci.1,addr=0x1 \ -drive file=/dev/HostVG/QEMUGuest1,if=none,id=drive-sata0-0-0 \ --device ide-drive,unit=0,drive=drive-sata0-0-0,id=sata0-0-0 \ +-device ide-drive,bus=ide.0,drive=drive-sata0-0-0,id=sata0-0-0 \ -vga qxl -global qxl.ram_size=67108864 -global qxl.vram_size=18874368 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-q35.args b/tests/qemuxml2argvdata/qemuxml2argv-q35.args index 4461eba..6bd5f4c 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-q35.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-q35.args @@ -4,5 +4,5 @@ LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test QEMU_AUDIO_DRV=none \ -device i82801b11-bridge,id=pci.1,bus=pcie.0,addr=0x2 \ -device pci-bridge,chassis_nr=2,id=pci.2,bus=pci.1,addr=0x1 \ -drive file=/dev/HostVG/QEMUGuest1,if=none,id=drive-sata0-0-0 \ --device ide-drive,unit=0,drive=drive-sata0-0-0,id=sata0-0-0 \ +-device ide-drive,bus=ide.0,drive=drive-sata0-0-0,id=sata0-0-0 \ -vga qxl -global qxl.ram_size=67108864 -global qxl.vram_size=18874368 -- 1.8.3.1

5 7

[libvirt] [PATCH] build: Generate libvirt_access*.xml in source dir
by Jiri Denemark 20 Sep '13

20 Sep '13

Technically, these files should be generated in build dir but since docs/newapi.xsl style sheet is expecting them to be in source dir, it's much easier to put them there than trying to convince newapi.xsl to look elsewhere. Signed-off-by: Jiri Denemark <jdenemar(a)redhat.com> --- src/Makefile.am | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/src/Makefile.am b/src/Makefile.am index 1388c5f..c456d48 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -863,9 +863,9 @@ ACCESS_DRIVER_SYM_FILES = \ libvirt_access_lxc.syms ACCESS_DRIVER_API_FILES = \ - libvirt_access.xml \ - libvirt_access_qemu.xml \ - libvirt_access_lxc.xml + $(srcdir)/libvirt_access.xml \ + $(srcdir)/libvirt_access_qemu.xml \ + $(srcdir)/libvirt_access_lxc.xml ACCESS_DRIVER_SOURCES = \ access/viraccessperm.h access/viraccessperm.c \ @@ -1554,15 +1554,15 @@ libvirt_access_lxc.syms: $(srcdir)/rpc/gendispatch.pl \ $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl --mode=aclsym \ lxc LXC $(LXC_PROTOCOL) > $@ -libvirt_access.xml: $(srcdir)/rpc/gendispatch.pl \ +$(srcdir)/libvirt_access.xml: $(srcdir)/rpc/gendispatch.pl \ $(REMOTE_PROTOCOL) Makefile.am $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl --mode=aclapi \ remote REMOTE $(REMOTE_PROTOCOL) > $@ -libvirt_access_qemu.xml: $(srcdir)/rpc/gendispatch.pl \ +$(srcdir)/libvirt_access_qemu.xml: $(srcdir)/rpc/gendispatch.pl \ $(QEMU_PROTOCOL) Makefile.am $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl --mode=aclapi \ qemu QEMU $(QEMU_PROTOCOL) > $@ -libvirt_access_lxc.xml: $(srcdir)/rpc/gendispatch.pl \ +$(srcdir)/libvirt_access_lxc.xml: $(srcdir)/rpc/gendispatch.pl \ $(LXC_PROTOCOL) Makefile.am $(AM_V_GEN)$(PERL) -w $(srcdir)/rpc/gendispatch.pl --mode=aclapi \ lxc LXC $(LXC_PROTOCOL) > $@ -- 1.8.3.2

3 4

[libvirt] storage pool that contains thin LVs
by Dusty Mabe 20 Sep '13

20 Sep '13

Hey guys, Just looking at https://bugzilla.redhat.com/show_bug.cgi?id=924672 and looks like all we need to do is ignore thin pools and thin pool data devices. With some trivial testing this seems to work fine (i.e. failed before and works now). diff --git a/src/storage/storage_backend_logical.c b/src/storage/storage_backend_logical.c index a1a37a1..0154256 100644 --- a/src/storage/storage_backend_logical.c +++ b/src/storage/storage_backend_logical.c @@ -85,6 +85,10 @@ virStorageBackendLogicalMakeVol(virStoragePoolObjPtr pool, if (attrs[4] != 'a') return 0; + /* Skip thin pools(t) and thin pool data(T) */ + if (attrs[0] == 't' || attrs[0] == 'T') + return 0; + /* See if we're only looking for a specific volume */ if (data != NULL) { vol = data; I'm sure the fix isn't this trivial but i can finish it up and submit an official patch (through git) if this is close. Dusty

3 3

[libvirt] [PATCH] Don't dereference NULL in qemumonitorjsontest
by Ján Tomko 20 Sep '13

20 Sep '13

In case of an error, qemuMonitorTestNewSimple returns NULL. Error out instead of dereferencing it. Found by Coverity, reported by John Ferlan. --- Pushed as trivial. tests/qemumonitorjsontest.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tests/qemumonitorjsontest.c b/tests/qemumonitorjsontest.c index b451e8e..0fb8d65 100644 --- a/tests/qemumonitorjsontest.c +++ b/tests/qemumonitorjsontest.c @@ -960,6 +960,9 @@ testQemuMonitorJSONCPU(const void *data) bool running = false; virDomainPausedReason reason = 0; + if (!test) + return -1; + if (qemuMonitorTestAddItem(test, "stop", "{\"return\": {}}") < 0 || qemuMonitorTestAddItem(test, "query-status", "{\"return\": {" @@ -1016,6 +1019,9 @@ testQemuMonitorJSONSimpleFunc(const void *opaque) const char *reply = data->reply; int ret = -1; + if (!test) + return -1; + if (!reply) reply = "{\"return\":{}}"; -- 1.8.1.5

2 1