- Devel - libvirt List Archives

[libvirt] [PATCH] conf: fix crash when parse a invalid vcpus in virDomainThreadSchedParse()
by Luyao Huang 10 Apr '15

10 Apr '15

When we set a invalid vcpus in /domain/cputune/vcpusched, like this: <vcpusched vcpus='0,^0' scheduler='fifo' priority='1'/> libvirtd will get segfault, because we will free bitmap (here is sp->ids) we passed in virBitmapParse() then return -1, then call virBitmapIsAllClear() with a null pointer bitmap will crash libvirtd. Split this check in two part to report parse error when cannot parse vcpus, and report invalid argument error when the cpu number is not correct. virBitmapIsAllClear (bitmap=0x0) at util/virbitmap.c:649 649 for (i = 0; i < bitmap->map_len; i++) (gdb) bt #0 virBitmapIsAllClear (bitmap=0x0) at util/virbitmap.c:649 #1 0x00007f28322f1245 in virDomainThreadSchedParse at conf/domain_conf.c:13464 #2 0x00007f283238aec8 in virDomainDefParseXML at conf/domain_conf.c:14059 #3 0x00007f2832390ddb in virDomainDefParseNode at conf/domain_conf.c:15588 #4 0x00007f2832390f02 in virDomainDefParse at conf/domain_conf.c:15530 #5 0x00007f2832390f53 in virDomainDefParseString at conf/domain_conf.c:15546 #6 0x00007f2819d8e8de in qemuDomainDefineXMLFlags at qemu/qemu_driver.c:7385 Signed-off-by: Luyao Huang <lhuang(a)redhat.com> --- src/conf/domain_conf.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 1763305..46dfea1 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -13473,12 +13473,11 @@ virDomainThreadSchedParse(xmlNodePtr node, goto error; } - if (!virBitmapParse(tmp, 0, &sp->ids, - VIR_DOMAIN_CPUMASK_LEN) || - virBitmapIsAllClear(sp->ids) || - virBitmapNextSetBit(sp->ids, -1) < minid || - virBitmapLastSetBit(sp->ids) > maxid) { + if (virBitmapParse(tmp, 0, &sp->ids, VIR_DOMAIN_CPUMASK_LEN) < 0) + goto error; + if (virBitmapNextSetBit(sp->ids, -1) < minid || + virBitmapLastSetBit(sp->ids) > maxid) { virReportError(VIR_ERR_XML_ERROR, _("Invalid value of '%s': %s"), name, tmp); -- 1.8.3.1

2 2

[libvirt] [PATCH v3 00/14] parallels: implement managed save
by Dmitry Guryanov 10 Apr '15

10 Apr '15

This patch series is intended to implement all needed code, so that suspend/resume in openstack nova will work. It implements .domainHasManagedSaveImage, .domainManagedSave and .domainManagedSaveRemove functions. It's not possible to change VM configuration, if it is in suspended state in PCS. It's possible for containers, but it has different meaning from the same action in libvirt: libvirt stores domain config in a save file and restores it if a user decided to start domain. But if the user discarded state file - domain will have new configuration. PCS doesn't store domain config for containers, so it will have new configuration either if the user will start it from, save file or discard state file and start after that. So the idea is to forbid changing domain configuration, if a domain is in managed saved state. We can't forbid calling this function in this state, because openstack/nova calls virDomainDefineXML in suspended state. So I compare current virDomainDef with the new one. And if there are no changes - function does nothing and returns success. Dmitry Guryanov (14): parallels: fix headers in parallels_sdk.h parallels: split prlsdkDomainChangeState function parallels: implement virDomainManagedSave parallels: report, that cdroms are readonly parallels: add controllers in prlsdkLoadDomain parallels: fill adapter model in virDomainNetDef parallels: don't fill net adapter model for containers conf: add VIR_DOMAIN_VIDEO_TYPE_PARALLELS video type conf: return proper default video type for parallels conf: add input device type for parallels containers parallels: add implicit input devices conf: fix virDomainDefParseXML for parallels conf: fix virDomainDefFormatInternal for parallels parallels: fix virDomainDefineXML for domain in saved state src/conf/domain_conf.c | 70 +++++++++++++++++++--- src/conf/domain_conf.h | 2 + src/parallels/parallels_driver.c | 104 ++++++++++++++++++++++++++++++-- src/parallels/parallels_sdk.c | 124 ++++++++++++++++++++++++++++++++++----- src/parallels/parallels_sdk.h | 17 ++++-- src/qemu/qemu_command.c | 6 +- 6 files changed, 285 insertions(+), 38 deletions(-) -- 2.1.0

2 17

Re: [libvirt] [Qemu-devel] can't bot from scsi http cdrom
by Stefan Hajnoczi 10 Apr '15

10 Apr '15

On Mon, Apr 06, 2015 at 01:15:24PM +0300, Vasiliy Tolstov wrote: > Hi. I'm try to boot from http cdrom and can't do that in case of scsi: > <disk type='network' device='cdrom'> > <driver name='qemu' type='raw' cache='unsafe'/> > <source protocol='http' name='/vps/rescue-4.5.2'> > <host name='xx.xx' port='80'/> > </source> > <backingStore/> > <target dev='sdb' bus='scsi'/> > <readonly/> > <alias name='scsi0-0-0-1'/> > <address type='drive' controller='0' bus='0' target='0' unit='1'/> > </disk> > > but in case of ide: > <disk type='network' device='cdrom'> > <driver name='qemu' type='raw' cache='unsafe'/> > <source protocol='http' name='/vps/rescue-4.5.2'> > <host name='xx.xx' port='80'/> > </source> > <backingStore/> > <target dev='hda' bus='ide'/> > <readonly/> > <alias name='ide0-0-0'/> > <address type='drive' controller='0' bus='0' target='0' unit='0'/> > </disk> > > all works fine. > > libvirt - 1.2.10 > qemu - 2.0.0 > seabios - 1.7.5 This is a libvirt question. I have CCed the libvirt mailing list. Stefan

1 0

[libvirt] possible bug regarding cpu flag 'pclmulqdq' in cpu_map.xml
by Sven Kieske 10 Apr '15

10 Apr '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, laine on IRC was so kind to redirect me to this list, so I didn't file a BZ at: https://bugzilla.redhat.com/enter_bug.cgi?product=Virtualization%20Tools &component=libvirt yet. Please keep me CC'ed as I'm not subscribed to the list. So here is my bug report: Description of problem: I believe the flag "pclmulqdq" is wrongly spelled in cpu_map.xml leading to wrong cpu being detected: Version-Release number of selected component (if applicable): current master: https://libvirt.org/git/?p=libvirt.git;a=blob;f=src/cpu/cpu_map.xml;h=21 10c0b9b2a2a70f0334b2ff63416bd1b704b669;hb=HEAD How reproducible: always Steps to Reproduce: 1.install e.g. ovirt, or some other libvirt consumer on a system with a cpu like intel xeon E3-1230 2. it will get reported as nehalem 3. Actual results: cpu gets reported as "nehalem" Expected results: cpu gets reported as "sandy bridge" Additional info: I found no real info on this flag: grep --color pclmuldq /usr/share/libvirt/cpu_map.xml <feature name='pclmuldq'> <feature name='pclmuldq'/> <feature name='pclmuldq'/> wikipedia (not always right, I know) and Intel just seem to know the flag 'pclmulqdq' See: https://en.wikipedia.org/wiki/CLMUL_instruction_set https://en.wikipedia.org/wiki/CPUID#EAX.3D1:_Processor_Info_and_Feature_ Bits https://software.intel.com/file/36945 But the Intel documentation is vast and I might have missed information on their website. So please check if this is the correct cpu flag indeed. Thanks! PS: If this is a real bug, it goes way back into el6 released versions of libvirt, and I would like to see a backport, if possible! Here is the BZ which introduced the flag: https://bugzilla.redhat.com/show_bug.cgi?id=761005 I guess this got wrongly typed, when you look at this comment, which has the correct notation: https://bugzilla.redhat.com/show_bug.cgi?id=761005#c16 - -- Mit freundlichen Grüßen / Regards Sven Kieske Systemadministrator Mittwald CM Service GmbH & Co. KG Königsberger Straße 6 32339 Espelkamp T: +49-5772-293-100 F: +49-5772-293-333 https://www.mittwald.de Geschäftsführer: Robert Meyer St.Nr.: 331/5721/1033, USt-IdNr.: DE814773217, HRA 6640, AG Bad Oeynhaus en Komplementärin: Robert Meyer Verwaltungs GmbH, HRB 13260, AG Bad Oeynhausen -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQIcBAEBAgAGBQJVJpqoAAoJEMby9TMDAbQRFlcP/1xajVdbv3fNfQPzJ3wIRa6M szkdbqMkI7HZTBjGvgSFXcf/ufb2RNfCU7AXpMi+FsE+oMB6qafke/J5VM6H/GUW TWuOy1wXz6QLxkP/p0VeU6ZX28XmwPOUMPcCQVhexIAbrmpgYCY42iaiEZUStcNp QQIqapOR6kL57WazTjY0U5GlGkV479oYlStF3E4maaJz+viNUqD6pE3WhXTbpcsu 1U1hPYANkfaMiXglm4MqpLklHw2PqgSfRvPzH58JStByuizPiWXZ9h42xE5ykVkW B7sw4XBSVk0mH99LNwnXC/VIw4pPz7uecTSbxdubZ/p4vvlG8JdszOekMfZVhy1+ ic93VPB+jSbiRGjWx8jkyYAsVOREaPBbmFF3ioLkAo9eVl8v8MXcBNppQQASVv7R vidSBu4qAVeu3ghmY5L+zEqTT3cG8pDcqzrFcELOtY1b6qsOOksgUxiylFv25g6F QPoDZpkpTvJ8YjOZx4H3ZjlfPQBhvBqd3QESNR8fKBUcxgxfGRwBVVU+14otvElE buEhBux0DvEfYrLbbiJtlGFtBmuz28RToPM1bmj52ujnGS1+0xYYYrHL35GhwnPc pvJ8mxAIRMYTdhn5eMsIh50N3uxz9Zky0CU8DJZ5St2Fsds3n7fR7Oao+PXOsOX1 mtJPsI1bYnxJ4RYVaIzk =ergB -----END PGP SIGNATURE-----

2 2

[libvirt] [PATCH] doc: Add info (where necessary) that paths should be specified as absolute
by Erik Skultety 09 Apr '15

09 Apr '15

We documented this almost everywhere, but missed it on several places. https://bugzilla.redhat.com/show_bug.cgi?id=1208763 --- Hopefully I didn't miss any place where it should be fixed as well. I also thought about adding a check for this issue in the code, but we probably do not want to do that during parsing, as the existing domains might as well disappear. The other thing is, that if the path is invalid, "No such file or directory" is returned which in my opinion describes the problem quite well. docs/formatdomain.html.in | 35 +++++++++++++++++++---------------- docs/formatstorage.html.in | 7 ++++--- 2 files changed, 23 insertions(+), 19 deletions(-) diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 1b496c3..7ceb1fa 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -126,7 +126,8 @@ provides details on allowed values for these. Since 0.0.1</dd> <dt><code>loader</code></dt> - <dd>The optional <code>loader</code> tag refers to a firmware blob + <dd>The optional <code>loader</code> tag refers to a firmware blob, + which is specified by absolute path, used to assist the domain creation process. It is used by Xen fully virtualized domains as well as setting the QEMU BIOS file path for QEMU/KVM domains. Xen since 0.1.0, @@ -142,10 +143,10 @@ <code>pflash</code>.</dd> <dt><code>nvram</code></dt> <dd>Some UEFI firmwares may want to use a non-volatile memory to store - some variables. In the host, this is represented as a file and the path - to the file is stored in this element. Moreover, when the domain is - started up libvirt copies so called master NVRAM store file defined - in <code>qemu.conf</code>. If needed, the <code>template</code> + some variables. In the host, this is represented as a file and the + absolute path to the file is stored in this element. Moreover, when the + domain is started up libvirt copies so called master NVRAM store file + defined in <code>qemu.conf</code>. If needed, the <code>template</code> attribute can be used to per domain override map of master NVRAM stores from the config file. Note, that for transient domains if the NVRAM file has been created by libvirt it is left behind and it is management @@ -971,7 +972,7 @@ resource partitions, potentially with nesting of said partitions. The <code>resource</code> element groups together configuration related to resource partitioning. It currently supports a child - element <code>partition</code> whose content defines the path + element <code>partition</code> whose content defines the absolute path of the resource partition in which to place the domain. If no partition is listed, then the domain will be placed in a default partition. It is the responsibility of the app/admin to ensure @@ -1954,8 +1955,8 @@ <dt><code>type='block'</code> since 0.0.3</dt> <dd> - The <code>dev</code> attribute specifies the path to the - host device to serve as the disk. + The <code>dev</code> attribute specifies the fully-qualified path + to the host device to serve as the disk. </dd> <dt><code>type='dir'</code> since 0.7.5</dt> @@ -3226,7 +3227,7 @@ versions of qemu used a default of "off", while newer qemus have a default of "on"). Since 0.9.7 (QEMU and KVM only). The optional - <code>file</code> attribute is used to point to a binary file + <code>file</code> attribute contains an absolute path to a binary file to be presented to the guest as the device's ROM BIOS. This can be useful, for example, to provide a PXE boot ROM for a virtual function of an sr-iov capable ethernet device (which @@ -3285,7 +3286,8 @@ Block / character devices from the host can be passed through to the guest using the <code>hostdev</code> element. This is - only possible with container based virtualization. + only possible with container based virtualization. Devices are specified + by a fully qualified path. since after 1.0.1 for LXC: @@ -5486,11 +5488,12 @@ qemu-kvm -net nic,model=? /dev/null <dl> <dt><code>master</code></dt> - <dd>Master device of the pair, that is passed to the hypervisor.</dd> + <dd>Master device of the pair, that is passed to the hypervisor. + Device is specified by a fully qualified path.</dd> <dt><code>slave</code></dt> <dd>Slave device of the pair, that is passed to the clients for connection - to the guest console.</dd> + to the guest console. Device is specified by a fully qualified path.</dd> </dl> <h4><a name="elementsSound">Sound devices</a></h4> @@ -5834,8 +5837,8 @@ qemu-kvm -net nic,model=? /dev/null This backend type requires exclusive access to a TPM device on the host. - An example for such a device is /dev/tpm0. The filename is - specified as path attribute of the <code>source</code> element. + An example for such a device is /dev/tpm0. The fully qualified file + name is specified by path attribute of the <code>source</code> element. If no file name is specified then /dev/tpm0 is automatically used. </dd> @@ -5941,8 +5944,8 @@ qemu-kvm -net nic,model=? /dev/null <dd> The optional <code>server</code> element can be used to configure a server socket the device is supposed to connect to. The optional - <code>path</code> attribute specifies the path to the unix socket and - defaults to <code>/var/lib/libvirt/shmem/$shmem-$name-sock</code>. + <code>path</code> attribute specifies the absolute path to the unix socket + and defaults to <code>/var/lib/libvirt/shmem/$shmem-$name-sock</code>. </dd> <dt><code>msi</code></dt> <dd> diff --git a/docs/formatstorage.html.in b/docs/formatstorage.html.in index 9c7b1bd..479e73c 100644 --- a/docs/formatstorage.html.in +++ b/docs/formatstorage.html.in @@ -379,9 +379,10 @@ <dl> <dt><code>path</code></dt> <dd>Provides the location at which the pool will be mapped into - the local filesystem namespace. For a filesystem/directory based - pool it will be the name of the directory in which volumes will - be created. For device based pools it will be the name of the directory in which + the local filesystem namespace, as an absolute path. For a + filesystem/directory based pool it will be a fully qualified name of + the directory in which volumes will be created. For device based pools + it will be a fully qualified name of the directory in which devices nodes exist. For the latter <code>/dev/</code> may seem like the logical choice, however, devices nodes there are not guaranteed stable across reboots, since they are allocated on -- 1.9.3

2 2

[libvirt] [PATCH 00/11] qemu: Refactor the block job code
by Peter Krempa 09 Apr '15

09 Apr '15

At the expense of adding 113 lines of code, kill the ugly qemuBlockJobImpl method and spread it's guts into separate functions. This series additionally fixes a issue with failed drive pivot and the abort function now returns errors if the returned event contained failure. Peter Krempa (11): qemu: monitor: Extract handling of JSON block job error codes qemu: domain: Introduce helper to retrieve domain monitor object qemu: domain: Add helper to check block job support qemu: blockjob: Use the new helpers in qemuDomainGetBlockJobInfo qemu: blockjob: Split qemuDomainBlockJobSetSpeed from qemuDomainBlockJobImpl qemu: Clean up old leftovers in qemuMonitorDrivePivot qemu: blockPivot: Don't pause the VM any more since we don't use drive-reopen qemu: blockjob: Separate qemuDomainBlockJobAbort from qemuDomainBlockJobImpl qemu: blockPull: Refactor the rest of qemuDomainBlockJobImpl qemu: drivePivot: Fix assumption when 'block-job-complete' fails qemu: Refactor qemuDomainBlockJobAbort() src/qemu/qemu_domain.c | 43 +++++ src/qemu/qemu_domain.h | 4 + src/qemu/qemu_driver.c | 437 +++++++++++++++++++++---------------------- src/qemu/qemu_migration.c | 8 +- src/qemu/qemu_monitor.c | 88 ++++++--- src/qemu/qemu_monitor.h | 35 ++-- src/qemu/qemu_monitor_json.c | 188 +++++++++++-------- src/qemu/qemu_monitor_json.h | 28 ++- tests/qemumonitorjsontest.c | 2 +- 9 files changed, 473 insertions(+), 360 deletions(-) -- 2.2.2

3 32

[libvirt] [PATCH] qemu: Avoid shadow of 'sync' symbol
by Peter Krempa 09 Apr '15

09 Apr '15

Old compilers whine that 'sync' is being shadowed in the function introduced in 1eccac1d2da7bbe97e1df25fd0ddac6e71b0794a. --- Pushed under the build-breaker rule. src/qemu/qemu_domain.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index c18b864..ae632c5 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -3063,17 +3063,17 @@ qemuDomainSupportsBlockJobs(virDomainObjPtr vm, bool *modern) { qemuDomainObjPrivatePtr priv = vm->privateData; - bool async = virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_BLOCKJOB_ASYNC); - bool sync = virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_BLOCKJOB_SYNC); + bool asynchronous = virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_BLOCKJOB_ASYNC); + bool synchronous = virQEMUCapsGet(priv->qemuCaps, QEMU_CAPS_BLOCKJOB_SYNC); - if (!sync && !async) { + if (!synchronous && !asynchronous) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("block jobs not supported with this QEMU binary")); return -1; } if (modern) - *modern = async; + *modern = asynchronous; return 0; } -- 2.2.2

1 0

[libvirt] [PATCHv2.5 00/10] Add support for memory hotplug
by Peter Krempa 09 Apr '15

09 Apr '15

Rebased version after changes to the prequel series. This applies on top of http://www.redhat.com/archives/libvir-list/2015-March/msg00134.html Convenience branch to pull from: git fetch git://pipo.sk/pipo/libvirt.git memory-hotplug-2.5 Peter Krempa (10): conf: Add support for parsing and formatting max memory and slot count qemu: Implement setup of memory hotplug parameters conf: Add device address type for dimm devices conf: Add interface to parse and format memory device information qemu: memdev: Add infrastructure to load memory device information qemu: migration: Forbid migration with memory modules lacking info qemu: add support for memory devices qemu: conf: Add support for memory device cold(un)plug qemu: Implement memory device hotplug qemu: Implement memory device hotunplug docs/formatdomain.html.in | 98 ++++ docs/schemas/domaincommon.rng | 76 +++ src/bhyve/bhyve_domain.c | 9 +- src/conf/domain_conf.c | 564 ++++++++++++++++++++- src/conf/domain_conf.h | 60 +++ src/libvirt_private.syms | 7 + src/libxl/libxl_domain.c | 8 + src/lxc/lxc_domain.c | 8 + src/openvz/openvz_driver.c | 14 +- src/parallels/parallels_driver.c | 6 +- src/phyp/phyp_driver.c | 6 +- src/qemu/qemu_command.c | 166 +++++- src/qemu/qemu_command.h | 15 + src/qemu/qemu_domain.c | 80 +++ src/qemu/qemu_domain.h | 5 + src/qemu/qemu_driver.c | 29 ++ src/qemu/qemu_hotplug.c | 177 +++++++ src/qemu/qemu_hotplug.h | 6 + src/qemu/qemu_migration.c | 14 + src/qemu/qemu_monitor.c | 42 ++ src/qemu/qemu_monitor.h | 14 + src/qemu/qemu_monitor_json.c | 122 +++++ src/qemu/qemu_monitor_json.h | 5 + src/qemu/qemu_process.c | 4 + src/uml/uml_driver.c | 9 +- src/vbox/vbox_common.c | 6 +- src/vmware/vmware_driver.c | 6 +- src/vmx/vmx.c | 6 +- src/xen/xen_driver.c | 7 + src/xenapi/xenapi_driver.c | 9 +- tests/domainschemadata/maxMemory.xml | 19 + .../qemuxml2argv-memory-hotplug-dimm.args | 11 + .../qemuxml2argv-memory-hotplug-dimm.xml | 50 ++ .../qemuxml2argv-memory-hotplug-nonuma.xml | 22 + .../qemuxml2argv-memory-hotplug.args | 6 + .../qemuxml2argv-memory-hotplug.xml | 34 ++ tests/qemuxml2argvtest.c | 6 + tests/qemuxml2xmltest.c | 4 + 38 files changed, 1707 insertions(+), 23 deletions(-) create mode 100644 tests/domainschemadata/maxMemory.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-dimm.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-dimm.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug-nonuma.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-memory-hotplug.xml -- 2.2.2

3 27

[libvirt] [PATCH 0/3] Fix some storage issues
by John Ferlan 09 Apr '15

09 Apr '15

Patch 1 & 3 are bz based, while Patch 2 was determined while working on Patch 3. Details in each commit message John Ferlan (3): storage: Fix issues in storageVolResize storage: Need to update freeExtent at delete primary partition storage: Don't duplicate efforts of backend driver src/storage/storage_backend_disk.c | 9 +++-- src/storage/storage_driver.c | 67 ++++++++++++++++++++++++++++++-------- 2 files changed, 59 insertions(+), 17 deletions(-) -- 2.1.0

2 5

[libvirt] [PATCH 0/3] Honour sparse files on migration
by Michal Privoznik 09 Apr '15

09 Apr '15

When the storage is pre-created on the destination, it might have been allocated as sparse on the source. Let's keep it that way if we can. Michal Privoznik (3): qemuMigrationPrecreateStorage: Fix debug message qemuMigrationCookieNBD: Turn some items into their own struct qemuMigrationPrecreateDisk: Preserve sparse files src/qemu/qemu_migration.c | 49 +++++++++++++++++++++++++++++++++++------------ 1 file changed, 37 insertions(+), 12 deletions(-) -- 2.0.5

2 4