April 2013 - Devel - libvirt List Archives

[libvirt] [PATCH] Fix build breaker with ATTRIBUTE_NONNULL defs
by John Ferlan 16 Apr '13

16 Apr '13

Using "./autogen.sh --system lv_cv_static_analysis=yes" for my daily Coverity builds resulted in the following error when building: In file included from util/vircgrouppriv.h:32:0, from util/vircgroup.c:44: util/vircgroup.h:59:5: error: nonnull argument with out-of-range operand number (argument 1, operand 5) util/vircgroup.h:74:5: error: nonnull argument references non-pointer operand (argument 1, operand 4) make[3]: *** [libvirt_util_la-vircgroup.lo] Error 1 make[3]: Leaving directory `/home/jferlan/libvirt.cov.curr/src' make[2]: *** [all] Error 2 make[2]: Leaving directory `/home/jferlan/libvirt.cov.curr/src' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/home/jferlan/libvirt.cov.curr' make: *** [all] Error 2 --- src/util/vircgroup.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/util/vircgroup.h b/src/util/vircgroup.h index f324fa3..b030e4a 100644 --- a/src/util/vircgroup.h +++ b/src/util/vircgroup.h @@ -54,11 +54,11 @@ int virCgroupNewPartition(const char *path, int virCgroupNewDriver(const char *name, bool create, int controllers, virCgroupPtr *group) - ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(5); + ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(4); int virCgroupNewSelf(virCgroupPtr *group) ATTRIBUTE_NONNULL(1); int virCgroupNewDomainDriver(virCgroupPtr driver, @@ -69,11 +69,11 @@ int virCgroupNewDomainDriver(virCgroupPtr driver, int virCgroupNewDomainPartition(virCgroupPtr partition, const char *driver, const char *name, bool create, virCgroupPtr *group) - ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(4); + ATTRIBUTE_NONNULL(1) ATTRIBUTE_NONNULL(2) ATTRIBUTE_NONNULL(5); int virCgroupNewVcpu(virCgroupPtr domain, int vcpuid, bool create, virCgroupPtr *group) -- 1.8.1.4

2 1

[libvirt] [PATCH] Resolve valgrind error
by John Ferlan 16 Apr '13

16 Apr '13

Commit id '1acfc171' resulted in the following valgrind failure: ==25317== 136 (24 direct, 112 indirect) bytes in 1 blocks are definitely lost in loss record 4 of 4 ==25317== at 0x4A06B6F: calloc (vg_replace_malloc.c:593) ==25317== by 0x4C6F851: virAlloc (viralloc.c:124) ==25317== by 0x4C71493: virBitmapNew (virbitmap.c:74) ==25317== by 0x4C71B79: virBitmapNewData (virbitmap.c:434) ==25317== by 0x402EF2: test8 (virbitmaptest.c:436) ==25317== by 0x40499F: virtTestRun (testutils.c:157) ==25317== by 0x402E8D: mymain (virbitmaptest.c:474) ==25317== by 0x404FDA: virtTestMain (testutils.c:719) ==25317== by 0x39D0821A04: (below main) (in /usr/lib64/libc-2.16.so) --- tests/virbitmaptest.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/tests/virbitmaptest.c b/tests/virbitmaptest.c index 90be0e2..2464c33 100644 --- a/tests/virbitmaptest.c +++ b/tests/virbitmaptest.c @@ -432,24 +432,25 @@ static int test8(const void *v ATTRIBUTE_UNUSED) { virBitmapPtr bitmap = NULL; char data[108] = {0x00,}; + int ret = -1; bitmap = virBitmapNewData(data, sizeof(data)); if (!bitmap) - goto error; + goto cleanup; if (!virBitmapIsAllClear(bitmap)) - goto error; + goto cleanup; if (virBitmapSetBit(bitmap, 11) < 0) - goto error; + goto cleanup; if (virBitmapIsAllClear(bitmap)) - goto error; + goto cleanup; - return 0; -error: + ret = 0; +cleanup: virBitmapFree(bitmap); - return -1; + return ret; } static int -- 1.8.1.4

2 2

[libvirt] [PATCH] Fix compilation error in util/vircgroup.c
by Stefan Berger 16 Apr '13

16 Apr '13

Fix the error util/vircgroup.c: In function 'virCgroupNewDomainPartition': util/vircgroup.c:1299:11: error: declaration of 'dirname' shadows a global declaration [-Werror=shadow] Signed-off-by: Stefan Berger <stefanb(a)linux.vnet.ibm.com> --- src/util/vircgroup.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) Index: libvirt-acl/src/util/vircgroup.c =================================================================== --- libvirt-acl.orig/src/util/vircgroup.c +++ libvirt-acl/src/util/vircgroup.c @@ -1296,13 +1296,13 @@ int virCgroupNewDomainPartition(virCgrou virCgroupPtr *group) { int rc; - char *dirname = NULL; + char *directoryname = NULL; - if (virAsprintf(&dirname, "%s.%s.libvirt", + if (virAsprintf(&directoryname, "%s.%s.libvirt", name, driver) < 0) return -ENOMEM; - rc = virCgroupNew(dirname, partition, -1, group); + rc = virCgroupNew(directoryname, partition, -1, group); if (rc == 0) { /* @@ -1322,7 +1322,7 @@ int virCgroupNewDomainPartition(virCgrou } } - VIR_FREE(dirname); + VIR_FREE(directoryname); return rc; } #else

2 1

[libvirt] [PATCH 0/2] Suppress log message when using unsupported TPM QMP commands
by Stefan Berger 16 Apr '13

16 Apr '13

Properly suppress log messages when using unsupported TPM QMP commands. Stefan

2 4

[libvirt] [libvirt-glib] Fix year in NEWS
by Christophe Fergeau 16 Apr '13

16 Apr '13

We switched from 2012 to 2013!! --- I've pushed this under the trivial rule. Christophe NEWS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/NEWS b/NEWS index b5f0c84..b1f4a80 100644 --- a/NEWS +++ b/NEWS @@ -1,7 +1,7 @@ libvirt-glib News ================= -0.1.6 - Mar 18, 2012 +0.1.6 - Mar 18, 2013 ==================== - Prefer g_strlcpy over strncpy @@ -14,7 +14,7 @@ - Replace FSF address with URL in copyright headers - Add API for changing Spice compression parameters. -0.1.5 - Jan 14, 2012 +0.1.5 - Jan 14, 2013 ==================== - Add support for bridge interface types in guest XML -- 1.8.1.4

1 0

[libvirt] [PATCH 0/7] Add support for guest-based CPU-hotplug and unplug
by Peter Krempa 16 Apr '13

16 Apr '13

This is a initial support for guest-agent based cpu hotplug that was commited in upstream qemu and should be released in qemu 1.5.0. This patchset relies on the existing qemuDomainSetVcpusFlags API that is not really ideal for this purpose but is usable. I'm planing on adding a specific-cpu hotplug API later that will allow more control. Peter Krempa (7): lib: Fix docs about return value of virDomainGetVcpusFlags() virsh-domain: Refactor cmdVcpucount and fix output on inactive domains qemu_agent: Add support for appending arrays to commands qemu_agent: Introduce helpers for agent based CPU hot(un)plug Introduce flag VIR_DOMAIN_VCPU agent, for agent based CPU unlpug(shutdown). qemu: Implement request of vCPU state using the guest agent qemu: Implement support for VIR_DOMAIN_VCPU in qemuDomainSetVcpusFlags include/libvirt/libvirt.h.in | 1 + src/libvirt.c | 9 +- src/qemu/qemu_agent.c | 141 +++++++++++++++++++++ src/qemu/qemu_agent.h | 12 ++ src/qemu/qemu_driver.c | 181 ++++++++++++++++++++++++--- tools/virsh-domain.c | 283 +++++++++++++++++++++---------------------- tools/virsh.pod | 13 +- 7 files changed, 474 insertions(+), 166 deletions(-) -- 1.8.1.5

2 15

[libvirt] [PATCH v4 00/11] Add support for guests with TPM passthrough device
by Stefan Berger 16 Apr '13

16 Apr '13

Hello! The following set of patches adds support to libvirt for adding a TPM passthrough device to a QEMU guest. Support for this was recently accepted into QEMU. This set of patches borrows a lot from the recently added support for rng's. Regards, Stefan --- v3->v4: - followed tree to ce65b435 - followed comments on v3 v2->v3: - followed tree to 03122d7b - some nits I found fixed v1->v2: - Followed Daniel Berrange's comments (except for the one on de-consolidating the JSON monitor code)

2 28

[libvirt] [PATCH 1/2] qemu: Support discard for disk
by Osier Yang 16 Apr '13

16 Apr '13

QEMU introduced "discard" option for drive since commit a9384aff53, <...> @var{discard} is one of "ignore" (or "off") or "unmap" (or "on") and controls whether @dfn{discard} (also known as @dfn{trim} or @dfn{unmap}) requests are ignored or passed to the filesystem. Some machine types may not support discard requests. </...> This patch exposes the support in libvirt. --- docs/formatdomain.html.in | 8 +++++ docs/schemas/domaincommon.rng | 11 +++++++ src/conf/domain_conf.c | 18 +++++++++++ src/conf/domain_conf.h | 10 ++++++ src/libvirt_private.syms | 1 + src/qemu/qemu_capabilities.c | 3 ++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 11 +++++++ .../qemuxml2argv-disk-discard.args | 8 +++++ .../qemuxml2argvdata/qemuxml2argv-disk-discard.xml | 36 ++++++++++++++++++++++ tests/qemuxml2argvtest.c | 3 ++ tests/qemuxml2xmltest.c | 2 ++ 12 files changed, 112 insertions(+) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-discard.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-discard.xml diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index 0cc56d9..f1cee14 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -1676,6 +1676,14 @@ network. By default copy-on-read is off. <span class='since'>Since 0.9.10 (QEMU and KVM only)</span> </li> + <li> + The optional <code>discard</code> attribute controls whether + to discard (also known as "trim" or "unmap") requests are + ignored or passed to the filesystem. The value can be either + "on" (allow the discard request to be passed) or "off" (ingore + the discard request). + <span class='since'>Since 1.0.5 (QEMU and KVM only)</span> + </li> </ul> </dd> <dt><code>boot</code></dt> diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 468c49c..4b71e16 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -1304,6 +1304,9 @@ <optional> <ref name="copy_on_read"/> </optional> + <optional> + <ref name="discard"/> + </optional> <empty/> </element> </define> @@ -1399,6 +1402,14 @@ <value>off</value> </choice> </attribute> + </define> + <define name="discard"> + <attribute name='discard'> + <choice> + <value>on</value> + <value>off</value> + </choice> + </attribute> </define> <define name="controller"> <element name="controller"> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 548368e..4925a80 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -728,6 +728,10 @@ VIR_ENUM_IMPL(virDomainTPMModel, VIR_DOMAIN_TPM_MODEL_LAST, VIR_ENUM_IMPL(virDomainTPMBackend, VIR_DOMAIN_TPM_TYPE_LAST, "passthrough") +VIR_ENUM_IMPL(virDomainDiskDiscard, VIR_DOMAIN_DISK_DISCARD_LAST, + "default", + "on", + "off") #define VIR_DOMAIN_XML_WRITE_FLAGS VIR_DOMAIN_XML_SECURE #define VIR_DOMAIN_XML_READ_FLAGS VIR_DOMAIN_XML_INACTIVE @@ -4324,6 +4328,7 @@ virDomainDiskDefParseXML(virDomainXMLOptionPtr xmlopt, char *wwn = NULL; char *vendor = NULL; char *product = NULL; + char *discard = NULL; int expected_secret_usage = -1; int auth_secret_usage = -1; @@ -4551,6 +4556,7 @@ virDomainDiskDefParseXML(virDomainXMLOptionPtr xmlopt, ioeventfd = virXMLPropString(cur, "ioeventfd"); event_idx = virXMLPropString(cur, "event_idx"); copy_on_read = virXMLPropString(cur, "copy_on_read"); + discard = virXMLPropString(cur, "discard"); } else if (!mirror && xmlStrEqual(cur->name, BAD_CAST "mirror") && !(flags & VIR_DOMAIN_XML_INACTIVE)) { char *ready; @@ -4997,6 +5003,14 @@ virDomainDiskDefParseXML(virDomainXMLOptionPtr xmlopt, def->copy_on_read = cor; } + if (discard) { + if ((def->discard = virDomainDiskDiscardTypeFromString(discard)) <= 0) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("unknown disk discard mode '%s'"), discard); + goto error; + } + } + if (devaddr) { if (virDomainParseLegacyDeviceAddress(devaddr, &def->info.addr.pci) < 0) { @@ -5116,6 +5130,7 @@ cleanup: VIR_FREE(ioeventfd); VIR_FREE(event_idx); VIR_FREE(copy_on_read); + VIR_FREE(discard); VIR_FREE(devaddr); VIR_FREE(serial); virStorageEncryptionFree(encryption); @@ -13291,6 +13306,7 @@ virDomainDiskDefFormat(virBufferPtr buf, const char *event_idx = virDomainVirtioEventIdxTypeToString(def->event_idx); const char *copy_on_read = virDomainVirtioEventIdxTypeToString(def->copy_on_read); const char *sgio = virDomainDiskSGIOTypeToString(def->sgio); + const char *discard = virDomainDiskDiscardTypeToString(def->discard); char uuidstr[VIR_UUID_STRING_BUFLEN]; @@ -13367,6 +13383,8 @@ virDomainDiskDefFormat(virBufferPtr buf, virBufferAsprintf(buf, " event_idx='%s'", event_idx); if (def->copy_on_read) virBufferAsprintf(buf, " copy_on_read='%s'", copy_on_read); + if (def->discard) + virBufferAsprintf(buf, " discard='%s'", discard); virBufferAddLit(buf, "/>\n"); } diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index f1f01fa..fdf5d93 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -600,6 +600,14 @@ enum virDomainDiskSGIO { VIR_DOMAIN_DISK_SGIO_LAST }; +enum virDomainDiskDiscard { + VIR_DOMAIN_DISK_DISCARD_DEFAULT = 0, + VIR_DOMAIN_DISK_DISCARD_ON, + VIR_DOMAIN_DISK_DISCARD_OFF, + + VIR_DOMAIN_DISK_DISCARD_LAST +}; + typedef struct _virDomainBlockIoTuneInfo virDomainBlockIoTuneInfo; struct _virDomainBlockIoTuneInfo { unsigned long long total_bytes_sec; @@ -682,6 +690,7 @@ struct _virDomainDiskDef { bool rawio_specified; int rawio; /* no = 0, yes = 1 */ int sgio; /* enum virDomainDiskSGIO */ + int discard; /* enum virDomainDiskDiscard */ size_t nseclabels; virSecurityDeviceLabelDefPtr *seclabels; @@ -2416,6 +2425,7 @@ VIR_ENUM_DECL(virDomainDiskIo) VIR_ENUM_DECL(virDomainDiskSecretType) VIR_ENUM_DECL(virDomainDiskSGIO) VIR_ENUM_DECL(virDomainDiskTray) +VIR_ENUM_DECL(virDomainDiskDiscard) VIR_ENUM_DECL(virDomainIoEventFd) VIR_ENUM_DECL(virDomainVirtioEventIdx) VIR_ENUM_DECL(virDomainDiskCopyOnRead) diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index 30fdcd7..a24bf5c 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -140,6 +140,7 @@ virDomainDiskDefFree; virDomainDiskDefGenSecurityLabelDef; virDomainDiskDefGetSecurityLabelDef; virDomainDiskDeviceTypeToString; +virDomainDiskDiscardTypeToString; virDomainDiskErrorPolicyTypeFromString; virDomainDiskErrorPolicyTypeToString; virDomainDiskFindByBusAndDst; diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index b174672..278012c 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -222,6 +222,7 @@ VIR_ENUM_IMPL(virQEMUCaps, QEMU_CAPS_LAST, "tpm-tis", "query-tpm-models", /* 140 */ + "drive-discard", ); struct _virQEMUCaps { @@ -963,6 +964,8 @@ virQEMUCapsComputeCmdFlags(const char *help, virQEMUCapsSet(qemuCaps, QEMU_CAPS_DRIVE_COPY_ON_READ); if (strstr(help, "bps=")) virQEMUCapsSet(qemuCaps, QEMU_CAPS_DRIVE_IOTUNE); + if (strstr(help, "discard")) + virQEMUCapsSet(qemuCaps, QEMU_CAPS_DRIVE_DISCARD); } if ((p = strstr(help, "-vga")) && !strstr(help, "-std-vga")) { const char *nl = strstr(p, "\n"); diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index 7a0f191..17c5335 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -180,6 +180,7 @@ enum virQEMUCapsFlags { QEMU_CAPS_DEVICE_TPM_PASSTHROUGH = 138, /* -tpmdev passthrough */ QEMU_CAPS_DEVICE_TPM_TIS = 139, /* -device tpm_tis */ QEMU_CAPS_TPM_MODELS = 140, /* query-tpm-models QMP command */ + QEMU_CAPS_DRIVE_DISCARD = 141, /* -drive discard=off(ignore)|on(unmap) */ QEMU_CAPS_LAST, /* this must always be the last item */ }; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 63b9350..36f20eb 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -2987,6 +2987,17 @@ qemuBuildDriveStr(virConnectPtr conn ATTRIBUTE_UNUSED, } } + 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; + } + } + if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_MONITOR_JSON)) { const char *wpolicy = NULL, *rpolicy = NULL; diff --git a/tests/qemuxml2argvdata/qemuxml2argv-disk-discard.args b/tests/qemuxml2argvdata/qemuxml2argv-disk-discard.args new file mode 100644 index 0000000..02e2ddb --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-discard.args @@ -0,0 +1,8 @@ +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test \ +/usr/bin/qemu -S -M pc-0.13 -m 1024 -smp 1 -nographic -nodefaults \ +-monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot dc -usb \ +-drive file=/var/lib/libvirt/images/f14.img,if=none,id=drive-virtio-disk0,discard=on \ +-device virtio-blk-pci,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0 \ +-drive file=/var/lib/libvirt/Fedora-14-x86_64-Live-KDE.iso,if=none,media=cdrom,id=drive-ide0-1-0,discard=off \ +-device ide-drive,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 \ +-device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x3 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-disk-discard.xml b/tests/qemuxml2argvdata/qemuxml2argv-disk-discard.xml new file mode 100644 index 0000000..fa38a2d --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-discard.xml @@ -0,0 +1,36 @@ +<domain type='qemu'> + <name>test</name> + <uuid>92d7a226-cfae-425b-a6d3-00bbf9ec5c9e</uuid> + <memory unit='KiB'>1048576</memory> + <currentMemory unit='KiB'>1048576</currentMemory> + <vcpu placement='static'>1</vcpu> + <os> + <type arch='x86_64' machine='pc-0.13'>hvm</type> + <boot dev='cdrom'/> + <boot dev='hd'/> + <bootmenu enable='yes'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>restart</on_crash> + <devices> + <emulator>/usr/bin/qemu</emulator> + <disk type='file' device='disk'> + <driver name='qemu' type='qcow2' discard='on'/> + <source file='/var/lib/libvirt/images/f14.img'/> + <target dev='vda' bus='virtio'/> + <address type='pci' domain='0x0000' bus='0x00' slot='0x04' function='0x0'/> + </disk> + <disk type='file' device='cdrom'> + <driver name='qemu' type='raw' discard='off'/> + <source file='/var/lib/libvirt/Fedora-14-x86_64-Live-KDE.iso'/> + <target dev='hdc' bus='ide'/> + <readonly/> + <address type='drive' controller='0' bus='1' target='0' unit='0'/> + </disk> + <controller type='usb' index='0'/> + <controller type='ide' index='0'/> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index be6c759..e09e070 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -575,6 +575,9 @@ mymain(void) QEMU_CAPS_DRIVE, QEMU_CAPS_DRIVE_COPY_ON_READ, QEMU_CAPS_VIRTIO_TX_ALG, QEMU_CAPS_DEVICE, QEMU_CAPS_VIRTIO_BLK_SCSI, QEMU_CAPS_VIRTIO_BLK_SG_IO); + DO_TEST("disk-discard", + QEMU_CAPS_DRIVE, QEMU_CAPS_DRIVE_DISCARD, + QEMU_CAPS_DEVICE); DO_TEST("disk-snapshot", QEMU_CAPS_DRIVE, QEMU_CAPS_DRIVE_CACHE_V2, QEMU_CAPS_DRIVE_FORMAT); DO_TEST("event_idx", diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index 3a1683f..7f32661 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -255,6 +255,8 @@ mymain(void) DO_TEST("disk-scsi-disk-vpd"); DO_TEST("disk-source-pool"); + DO_TEST("disk-discard"); + DO_TEST("virtio-rng-random"); DO_TEST("virtio-rng-egd"); -- 1.8.1.4

1 1

[libvirt] [Concern] Whether some clock element stability need to be check before migration?
by Xuesong Zhang 15 Apr '13

15 Apr '13

hi, This is xuesong from libvirt-QE, which is the feature owner of stable guest ABI in rhel7. I met one issue, but didn't know whether it is one bug or not, since in Guest ABI , some aspects belong to "a host-side configuration parameter". So, would you please kindly help to check it? Thanks. Description While check the guest clock ABI stability, some scenario fails. version kernel-3.8.0-0.43.el7.x86_64 qemu-kvm-1.3.0-6.el7.x86_64 libvirt-1.0.3-1.el7.x86_64 reproduced 100% Steps 1. prepare migration env on two hosts. 2. prepare one guest with the parallel. ...... <clock offset='localtime'> <timer name='rtc' tickpolicy='catchup' track='guest'> <catchup threshold='123' slew='120' limit='10000'/> </timer> <timer name='pit' tickpolicy='delay'/> </clock> .... 3. start and dumpxml this gust # virsh start mig Domain mig started # virsh dumpxml mig > mig.xml 4. Change "localtime" to "utc" in the generated xml "mig.xml". ...... <clock offset='utc'> <timer name='rtc' tickpolicy='catchup' track='guest'> <catchup threshold='123' slew='120' limit='10000'/> </timer> <timer name='pit' tickpolicy='delay'/> </clock> .... 5. migration with the modified xml is successfully.------------------it is issue, should prompt error message. # virsh migrate --live --xml mig.xml mig qemu+ssh://{$target host IP}/system 6. run step3 again, then change timer name "pit" to "platform". Note, maker sure only one change point in the xml. then migration with the modified xml. ----------------this result is as expect # virsh migrate mig --live --xml mig.xml qemu+ssh://10.66.83.38/system error: unsupported configuration: Target timer platform does not match source pit 7. run step3 again, then delete timer name "pit". Note, maker sure only one change point in the xml. then migration with the modified xml.-----------------the result is as expect #virsh migrate mig --live --xml mig.xml qemu+ssh://10.66.83.38/system error: unsupported configuration: Target domain timers do not match source 8. run step3 again, then change timer track from "guest" to "wall". Note, maker sure only one change point in the xml. then migration with the modified xml. The migration is succssful.------------------------------------------------------------------------------------------------------------------it is issue, should prompt error message. # virsh migrate --live --xml mig.xml mig qemu+ssh://{$target host IP}/system 9. run step3 again, then change timer tickpolicy from "catchup" to "delay". Note, maker sure only one change point in the xml. then migration with the modified xml. The migration is succssful.------------------------------------------------------------------------------------------------------------------it is issue, should prompt error message. # virsh migrate --live --xml mig.xml mig qemu+ssh://{$target host IP}/system Actual result The result in step 6 and 7 is as expected, the result of step 5, 8 and 9 is wrong. Expect result The clock element stability in step 5, 8 and 9 need to be check before migration. Addtional info: Below is all checks about clock in upstream 1.0.4 maint branch: In timer check Function virDomainTimerDefCheckABIStability only check 1.NAME :src->name != dst->name 2.PRESENT :src->present != dst->present) 3.FREQUENCY :src->name == VIR_DOMAIN_TIMER_NAME_TSC) { if (src->frequency != dst->frequency) } 4.MODE src->mode != dst->mode And in another function , check timer's numbers 5.NUMBERS : src->clock.ntimers != dst->clock.ntimers Best Regards, Zhang Xuesong IRC: xuzhang internal phone: 88393

1 0

[libvirt] [PATCH 0/2] Check domain status before virsh suspend or resume
by John Ferlan 15 Apr '13

15 Apr '13

https://bugzilla.redhat.com/show_bug.cgi?id=906644 Added checks to both virsh suspend and virsh resume for the domain to be in a the right state before trying the suspend/resume. Similar checks to examples/domsuspend/suspend.c. Although not stated in case, any thoughts on adding a --force for both? John Ferlan (2): virsh: Add check for domain state before attempting resume virsh: Add check for domain state before attempting suspend tools/virsh-domain.c | 40 +++++++++++++++++++++++++++++++--------- 1 file changed, 31 insertions(+), 9 deletions(-) -- 1.8.1.4

3 5