- Devel - libvirt List Archives

[libvirt] [PATCH] tests: Unify style of test skipping code
by Matthias Bolte 29 Jul '11

29 Jul '11

Prefer 'return EXIT_AM_SKIP' over 'exit(EXIT_AM_SKIP)'. Prefer 'int main(void)' over 'int main(int argc, char **argv)'. Fix mymain signature in commandtest and nodeinfotest. --- tests/commandtest.c | 10 +++++----- tests/nodeinfotest.c | 10 +++++----- tests/qemuargv2xmltest.c | 6 +++++- tests/qemuxml2xmltest.c | 6 +++++- tests/virnettlscontexttest.c | 6 ++++-- tests/virshtest.c | 26 +++++++++++++++++--------- 6 files changed, 41 insertions(+), 23 deletions(-) diff --git a/tests/commandtest.c b/tests/commandtest.c index ef2850d..2e800ae 100644 --- a/tests/commandtest.c +++ b/tests/commandtest.c @@ -38,10 +38,10 @@ #ifdef WIN32 -static int -mymain(int argc ATTRIBUTE_UNUSED, char **argv ATTRIBUTE_UNUSED) +int +main(void) { - exit (EXIT_AM_SKIP); + return EXIT_AM_SKIP; } #else @@ -814,6 +814,6 @@ mymain(void) return(ret==0 ? EXIT_SUCCESS : EXIT_FAILURE); } -#endif /* !WIN32 */ - VIRT_TEST_MAIN(mymain) + +#endif /* !WIN32 */ diff --git a/tests/nodeinfotest.c b/tests/nodeinfotest.c index 5abee92..448e072 100644 --- a/tests/nodeinfotest.c +++ b/tests/nodeinfotest.c @@ -15,10 +15,10 @@ defined(__amd64__) || \ defined(__i386__))) -static int -mymain(int argc ATTRIBUTE_UNUSED, char **argv ATTRIBUTE_UNUSED) +int +main(void) { - exit (EXIT_AM_SKIP); + return EXIT_AM_SKIP; } #else @@ -130,6 +130,6 @@ mymain(void) return(ret==0 ? EXIT_SUCCESS : EXIT_FAILURE); } -#endif /* __linux__ */ - VIRT_TEST_MAIN(mymain) + +#endif /* __linux__ */ diff --git a/tests/qemuargv2xmltest.c b/tests/qemuargv2xmltest.c index bade95d..c2b6cf2 100644 --- a/tests/qemuargv2xmltest.c +++ b/tests/qemuargv2xmltest.c @@ -239,6 +239,10 @@ VIRT_TEST_MAIN(mymain) #else -int main (void) { return (EXIT_AM_SKIP); } +int +main(void) +{ + return EXIT_AM_SKIP; +} #endif /* WITH_QEMU */ diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index f1900c5..461fd0d 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -204,6 +204,10 @@ VIRT_TEST_MAIN(mymain) #else -int main (void) { exit (EXIT_AM_SKIP); } +int +main(void) +{ + return EXIT_AM_SKIP; +} #endif /* WITH_QEMU */ diff --git a/tests/virnettlscontexttest.c b/tests/virnettlscontexttest.c index 12ecf1e..520b006 100644 --- a/tests/virnettlscontexttest.c +++ b/tests/virnettlscontexttest.c @@ -1249,9 +1249,11 @@ mymain(void) VIRT_TEST_MAIN(mymain) #else + int -main(int argc ATTRIBUTE_UNUSED, char **argv ATTRIBUTE_UNUSED) +main(void) { - exit (EXIT_AM_SKIP); + return EXIT_AM_SKIP; } + #endif diff --git a/tests/virshtest.c b/tests/virshtest.c index e22e582..de5138c 100644 --- a/tests/virshtest.c +++ b/tests/virshtest.c @@ -9,7 +9,17 @@ #include "util.h" #include "testutils.h" -#define DOM_UUID "ef861801-45b9-11cb-88e3-afbfe5370493" +#ifdef WIN32 + +int +main(void) +{ + return EXIT_AM_SKIP; +} + +#else + +# define DOM_UUID "ef861801-45b9-11cb-88e3-afbfe5370493" static const char *dominfo_fc4 = "\ Id: 2\n\ @@ -71,13 +81,13 @@ cleanup: return result; } -#define VIRSH_DEFAULT "../tools/virsh", \ +# define VIRSH_DEFAULT "../tools/virsh", \ "--connect", \ "test:///default" static char *custom_uri; -#define VIRSH_CUSTOM "../tools/virsh", \ +# define VIRSH_CUSTOM "../tools/virsh", \ "--connect", \ custom_uri @@ -224,10 +234,6 @@ mymain(void) { int ret = 0; -#ifdef WIN32 - exit (EXIT_AM_SKIP); -#endif - if (virAsprintf(&custom_uri, "test://%s/../examples/xml/test/testnode.xml", abs_srcdir) < 0) return EXIT_FAILURE; @@ -298,7 +304,7 @@ mymain(void) /* It's a bit awkward listing result before argument, but that's a * limitation of C99 vararg macros. */ -#define DO_TEST(i, result, ...) \ +# define DO_TEST(i, result, ...) \ do { \ const char *myargv[] = { VIRSH_DEFAULT, __VA_ARGS__, NULL }; \ const struct testInfo info = { myargv, result }; \ @@ -380,10 +386,12 @@ mymain(void) DO_TEST(30, "--shell a\n", "echo \t '-'\"-\" \t --shell \t a"); -#undef DO_TEST +# undef DO_TEST free(custom_uri); return(ret==0 ? EXIT_SUCCESS : EXIT_FAILURE); } VIRT_TEST_MAIN(mymain) + +#endif /* WIN32 */ -- 1.7.4.1

2 2

[libvirt] [PATCH 0/4] xen callback cleanups
by Eric Blake 28 Jul '11

28 Jul '11

Noticed because of the recent xen break in commit 1edf5cc5 (fixed in 934fdcb3), which broke because it touched too many files, rather than focusing on just xen_driver.c. Renaming all xen callback structs makes it easier to grep for use of the callback structs, which in turn makes it easier to see what still remains in the goal of getting rid of the callback struct. And using C99 initialization syntax makes it less likely that future patches will needlessly try to patch too many files. Eric Blake (4): xen: cleanup callback struct xen: reduce callback special cases xen: make direct call when there is only one subdriver xen: drop unused callbacks src/xen/xen_driver.c | 186 ++++++++++++++++++---------------------------- src/xen/xen_driver.h | 63 ++++++---------- src/xen/xen_hypervisor.c | 52 ++++---------- src/xen/xen_inotify.c | 39 +---------- src/xen/xend_internal.c | 68 +++++++---------- src/xen/xend_internal.h | 6 ++ src/xen/xm_internal.c | 51 +++---------- src/xen/xs_internal.c | 45 ++---------- 8 files changed, 167 insertions(+), 343 deletions(-) -- 1.7.4.4

2 11

[libvirt] libvirt-0.9.3 and qemu-0.14.1
by Zdenek Styblik 28 Jul '11

28 Jul '11

Hello, I'm not sure if this has been reported or fixed, but ... ~~~ SNIP ~~~ 21:02:59.981: 13171: info : libvirt version: 0.9.3 21:02:59.981: 13171: error : virCommandWait:1957 : internal error Child process (LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/sbin:/usr/sbin:/bin:/usr/bin HOME=/root USER=root LOGNAME=root /usr/bin/qemu-kvm -help) status unexpected: exit status 1 21:02:59.986: 13171: error : virCommandWait:1957 : internal error Child process (LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/sbin:/usr/sbin:/bin:/usr/bin HOME=/root USER=root LOGNAME=root /usr/bin/qemu-kvm -help) status unexpected: exit status 1 21:02:59.988: 13171: error : virDomainDefParseXML:6069 : unknown OS type hvm ~~~ SNIP ~~~ qemu-0.14.1 and libvirt-0.9.3 What am I missing here? Thanks, Z. -- Zdenek Styblik email: stybla(a)turnovfree.net jabber: stybla(a)jabber.turnovfree.net

1 1

[libvirt] [PATCH] freebsd: Fix build problem due to picking up the wrong libvirt.h
by Matthias Bolte 28 Jul '11

28 Jul '11

AM_GNU_GETTEXT calls AM_ICONV_LINK. AM_ICONV_LINK saves and alters CPPFLAGS, but doesn't restore it when it finds libiconv. This results in /usr/local/include ending up in the gcc command line before the include path for the local include directory. This makes gcc pick a previous installed libvirt.h instead of the correct one from the source tree. Workaround this problem by saving and restoring CPPFLAGS around the AM_GNU_GETTEXT call. --- configure.ac | 8 ++++++++ 1 files changed, 8 insertions(+), 0 deletions(-) diff --git a/configure.ac b/configure.ac index b2ba930..8f46dbd 100644 --- a/configure.ac +++ b/configure.ac @@ -2011,8 +2011,16 @@ dnl Enable building libvirtd? AM_CONDITIONAL([WITH_LIBVIRTD],[test "x$with_libvirtd" = "xyes"]) dnl Check for gettext - don't go any newer than what RHEL 5 supports +dnl +dnl save and restore CPPFLAGS around gettext check as the internal iconv +dnl check might leave -I/usr/local/include in CPPFLAGS on FreeBSD resulting +dnl in the build picking up previously installed libvirt/libvirt.h instead +dnl of the correct one from the soucre tree + +save_CPPFLAGS="$CPPFLAGS" AM_GNU_GETTEXT_VERSION([0.17]) AM_GNU_GETTEXT([external]) +CPPFLAGS="$save_CPPFLAGS" ALL_LINGUAS=`cd "$srcdir/po" > /dev/null && ls *.po | sed 's+\.po$++'` -- 1.7.0.4

2 6

[libvirt] [PATCH v4] virsh: avoid missing zero value judgement in cmdBlkiotune
by ajia＠redhat.com 28 Jul '11

28 Jul '11

* tools/virsh.c: fix missing zero value judgement in cmdBlkiotune and correct vshError information. when weight is equal to 0, the cmdBlkiotune will not raise any error information when judge weight value first time, and execute else branch to judge weight value again, strncpy(temp->field, VIR_DOMAIN_BLKIO_WEIGHT, sizeof(temp->field)) will be not executed for ever. However, if and only if param->field is equal to VIR_DOMAIN_BLKIO_WEIGHT, underlying qemuDomainSetBlkioParameters function will check whether weight value is in range [100, 1000]. * how to reproduce? % virsh blkiotune ${guestname} --weight 0 Signed-off-by: Alex Jia <ajia(a)redhat.com> --- tools/virsh.c | 9 +++++---- 1 files changed, 5 insertions(+), 4 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 8bd22dc..feb45de 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -4004,6 +4004,7 @@ cmdBlkiotune(vshControl * ctl, const vshCmd * cmd) virDomainPtr dom; int weight = 0; int nparams = 0; + int rv = 0; unsigned int i = 0; virTypedParameterPtr params = NULL, temp = NULL; bool ret = false; @@ -4031,15 +4032,15 @@ cmdBlkiotune(vshControl * ctl, const vshCmd * cmd) if (!(dom = vshCommandOptDomain(ctl, cmd, NULL))) return false; - if (vshCommandOptInt(cmd, "weight", &weight) < 0) { + if ((rv = vshCommandOptInt(cmd, "weight", &weight)) < 0) { vshError(ctl, "%s", - _("Unable to parse integer parameter")); + _("Unable to parse non-integer parameter")); goto cleanup; } - if (weight) { + if (rv > 0) { nparams++; - if (weight < 0) { + if (weight <= 0) { vshError(ctl, _("Invalid value of %d for I/O weight"), weight); goto cleanup; } -- 1.7.1

4 4

[libvirt] [PATCH] qemu: fix nested job with driver lock held
by Eric Blake 28 Jul '11

28 Jul '11

qemuMigrationUpdateJobStatus (called in a loop by migration and save tasks) uses qemuDomainObjEnterMonitorWithDriver; however, that function ended up starting a nested job without releasing the driver. Since no one else is making nested calls, we can inline the internal functions to properly track driver_locked. * src/qemu/qemu_domain.h (qemuDomainObjBeginNestedJob) (qemuDomainObjBeginNestedJobWithDriver) (qemuDomainObjEndNestedJob): Drop unused prototypes. * src/qemu/qemu_domain.c (qemuDomainObjEnterMonitorInternal): Reflect driver lock to nested job. (qemuDomainObjBeginNestedJob) (qemuDomainObjBeginNestedJobWithDriver) (qemuDomainObjEndNestedJob): Drop unused functions. --- This does not solve the crash in 'virsh managedsave', but hopefully will make analysis easier in finding where we are freeing the monitor too early when probing to see if outgoing migration is completed. src/qemu/qemu_domain.c | 59 +++++++++++------------------------------------ src/qemu/qemu_domain.h | 9 ------- 2 files changed, 14 insertions(+), 54 deletions(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index fe88ce3..2eaaf3a 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -832,31 +832,6 @@ int qemuDomainObjBeginAsyncJobWithDriver(struct qemud_driver *driver, } /* - * Use this to protect monitor sections within active async job. - * - * The caller must call qemuDomainObjBeginAsyncJob{,WithDriver} before it can - * use this method. Never use this method if you only own non-async job, use - * qemuDomainObjBeginJob{,WithDriver} instead. - */ -int -qemuDomainObjBeginNestedJob(struct qemud_driver *driver, - virDomainObjPtr obj) -{ - return qemuDomainObjBeginJobInternal(driver, false, obj, - QEMU_JOB_ASYNC_NESTED, - QEMU_ASYNC_JOB_NONE); -} - -int -qemuDomainObjBeginNestedJobWithDriver(struct qemud_driver *driver, - virDomainObjPtr obj) -{ - return qemuDomainObjBeginJobInternal(driver, true, obj, - QEMU_JOB_ASYNC_NESTED, - QEMU_ASYNC_JOB_NONE); -} - -/* * obj must be locked before calling, qemud_driver does not matter * * To be called after completing the work associated with the @@ -888,21 +863,6 @@ qemuDomainObjEndAsyncJob(struct qemud_driver *driver, virDomainObjPtr obj) return virDomainObjUnref(obj); } -void -qemuDomainObjEndNestedJob(struct qemud_driver *driver, virDomainObjPtr obj) -{ - qemuDomainObjPrivatePtr priv = obj->privateData; - - qemuDomainObjResetJob(priv); - qemuDomainObjSaveJob(driver, obj); - virCondSignal(&priv->job.cond); - - /* safe to ignore since the surrounding async job increased the reference - * counter as well */ - ignore_value(virDomainObjUnref(obj)); -} - - static int ATTRIBUTE_NONNULL(1) qemuDomainObjEnterMonitorInternal(struct qemud_driver *driver, bool driver_locked, @@ -911,7 +871,9 @@ qemuDomainObjEnterMonitorInternal(struct qemud_driver *driver, qemuDomainObjPrivatePtr priv = obj->privateData; if (priv->job.active == QEMU_JOB_NONE && priv->job.asyncJob) { - if (qemuDomainObjBeginNestedJob(driver, obj) < 0) + if (qemuDomainObjBeginJobInternal(driver, driver_locked, obj, + QEMU_JOB_ASYNC_NESTED, + QEMU_ASYNC_JOB_NONE) < 0) return -1; if (!virDomainObjIsActive(obj)) { qemuReportError(VIR_ERR_OPERATION_FAILED, "%s", @@ -952,8 +914,15 @@ qemuDomainObjExitMonitorInternal(struct qemud_driver *driver, priv->mon = NULL; } - if (priv->job.active == QEMU_JOB_ASYNC_NESTED) - qemuDomainObjEndNestedJob(driver, obj); + if (priv->job.active == QEMU_JOB_ASYNC_NESTED) { + qemuDomainObjResetJob(priv); + qemuDomainObjSaveJob(driver, obj); + virCondSignal(&priv->job.cond); + + /* safe to ignore since the surrounding async job increased + * the reference counter as well */ + ignore_value(virDomainObjUnref(obj)); + } } /* @@ -962,7 +931,7 @@ qemuDomainObjExitMonitorInternal(struct qemud_driver *driver, * To be called immediately before any QEMU monitor API call * Must have already either called qemuDomainObjBeginJob() and checked * that the VM is still active or called qemuDomainObjBeginAsyncJob, in which - * case this will call qemuDomainObjBeginNestedJob. + * case this will start a nested job. * * To be followed with qemuDomainObjExitMonitor() once complete */ @@ -988,7 +957,7 @@ void qemuDomainObjExitMonitor(struct qemud_driver *driver, * To be called immediately before any QEMU monitor API call * Must have already either called qemuDomainObjBeginJobWithDriver() and * checked that the VM is still active or called qemuDomainObjBeginAsyncJob, - * in which case this will call qemuDomainObjBeginNestedJobWithDriver. + * in which case this will start a nested job. * * To be followed with qemuDomainObjExitMonitorWithDriver() once complete */ diff --git a/src/qemu/qemu_domain.h b/src/qemu/qemu_domain.h index 679259f..8bff8b0 100644 --- a/src/qemu/qemu_domain.h +++ b/src/qemu/qemu_domain.h @@ -143,9 +143,6 @@ int qemuDomainObjBeginAsyncJob(struct qemud_driver *driver, virDomainObjPtr obj, enum qemuDomainAsyncJob asyncJob) ATTRIBUTE_RETURN_CHECK; -int qemuDomainObjBeginNestedJob(struct qemud_driver *driver, - virDomainObjPtr obj) - ATTRIBUTE_RETURN_CHECK; int qemuDomainObjBeginJobWithDriver(struct qemud_driver *driver, virDomainObjPtr obj, enum qemuDomainJob job) @@ -154,9 +151,6 @@ int qemuDomainObjBeginAsyncJobWithDriver(struct qemud_driver *driver, virDomainObjPtr obj, enum qemuDomainAsyncJob asyncJob) ATTRIBUTE_RETURN_CHECK; -int qemuDomainObjBeginNestedJobWithDriver(struct qemud_driver *driver, - virDomainObjPtr obj) - ATTRIBUTE_RETURN_CHECK; int qemuDomainObjEndJob(struct qemud_driver *driver, virDomainObjPtr obj) @@ -164,9 +158,6 @@ int qemuDomainObjEndJob(struct qemud_driver *driver, int qemuDomainObjEndAsyncJob(struct qemud_driver *driver, virDomainObjPtr obj) ATTRIBUTE_RETURN_CHECK; -void qemuDomainObjEndNestedJob(struct qemud_driver *driver, - virDomainObjPtr obj); - void qemuDomainObjSetJobPhase(struct qemud_driver *driver, virDomainObjPtr obj, int phase); -- 1.7.4.4

2 2

[libvirt] [PATCH] build: avoid type-punning compiler warning
by Eric Blake 28 Jul '11

28 Jul '11

On RHEL 5, with gcc 4.1.2: rpc/virnetsaslcontext.c: In function 'virNetSASLSessionUpdateBufSize': rpc/virnetsaslcontext.c:396: warning: dereferncing type-punned pointer will break strict-aliasing rules [-Wstrict-aliasing] * src/rpc/virnetsaslcontext.c (virNetSASLSessionUpdateBufSize): Use a union to work around gcc warning. --- src/rpc/virnetsaslcontext.c | 11 +++++++---- 1 files changed, 7 insertions(+), 4 deletions(-) diff --git a/src/rpc/virnetsaslcontext.c b/src/rpc/virnetsaslcontext.c index 71796b9..a0752dd 100644 --- a/src/rpc/virnetsaslcontext.c +++ b/src/rpc/virnetsaslcontext.c @@ -390,10 +390,13 @@ cleanup: static int virNetSASLSessionUpdateBufSize(virNetSASLSessionPtr sasl) { - unsigned *maxbufsize; + union { + unsigned *maxbufsize; + const void *ptr; + } u; int err; - err = sasl_getprop(sasl->conn, SASL_MAXOUTBUF, (const void **)&maxbufsize); + err = sasl_getprop(sasl->conn, SASL_MAXOUTBUF, &u.ptr); if (err != SASL_OK) { virNetError(VIR_ERR_INTERNAL_ERROR, _("cannot get security props %d (%s)"), @@ -402,8 +405,8 @@ static int virNetSASLSessionUpdateBufSize(virNetSASLSessionPtr sasl) } VIR_DEBUG("Negotiated bufsize is %u vs requested size %zu", - *maxbufsize, sasl->maxbufsize); - sasl->maxbufsize = *maxbufsize; + *u.maxbufsize, sasl->maxbufsize); + sasl->maxbufsize = *u.maxbufsize; return 0; } -- 1.7.4.4

2 2

[libvirt] [PATCH] qemu: Fix memory leak on metadata fetching
by Michal Privoznik 28 Jul '11

28 Jul '11

As written in virStorageFileGetMetadataFromFD decription, caller must free metadata after use. Qemu driver miss this and therefore leak metadata which can grow to huge mem leak if somebody query for blockInfo a lot. --- src/qemu/qemu_driver.c | 14 ++++++++++---- 1 files changed, 10 insertions(+), 4 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 0f91910..d45c7c5 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -6949,7 +6949,7 @@ static int qemuDomainGetBlockInfo(virDomainPtr dom, int ret = -1; int fd = -1; off_t end; - virStorageFileMetadata meta; + virStorageFileMetadata *meta = NULL; virDomainDiskDefPtr disk = NULL; struct stat sb; int i; @@ -7017,9 +7017,14 @@ static int qemuDomainGetBlockInfo(virDomainPtr dom, } } + if (VIR_ALLOC(meta) < 0) { + virReportOOMError(); + goto cleanup; + } + if (virStorageFileGetMetadataFromFD(path, fd, format, - &meta) < 0) + meta) < 0) goto cleanup; /* Get info for normal formats */ @@ -7056,8 +7061,8 @@ static int qemuDomainGetBlockInfo(virDomainPtr dom, /* If the file we probed has a capacity set, then override * what we calculated from file/block extents */ - if (meta.capacity) - info->capacity = meta.capacity; + if (meta->capacity) + info->capacity = meta->capacity; /* Set default value .. */ info->allocation = info->physical; @@ -7091,6 +7096,7 @@ static int qemuDomainGetBlockInfo(virDomainPtr dom, } cleanup: + virStorageFileFreeMetadata(meta); VIR_FORCE_CLOSE(fd); if (vm) virDomainObjUnlock(vm); -- 1.7.5.rc3

2 2

Re: [libvirt] PCI devices passthough to LXC containers using libvirt
by Devendra K. Modium 28 Jul '11

28 Jul '11

Hi Thanks for the reply. I think the links that you provided show how to deal with pci devices in case the hypervisor is kvm. Please correct me if I am wrong. But I am using LXC containers.I have skimmed through the libvirt lxc driver code and found no functionality of allowing specified devices into a container exists other than currently where only hard coded devices are allowed which can be seen in the file lxc_controller.c struct cgroup_device_policy devices[] = { {'c', LXC_DEV_MAJ_MEMORY, LXC_DEV_MIN_NULL}, {'c', LXC_DEV_MAJ_MEMORY, LXC_DEV_MIN_ZERO}, {'c', LXC_DEV_MAJ_MEMORY, LXC_DEV_MIN_FULL}, {'c', LXC_DEV_MAJ_MEMORY, LXC_DEV_MIN_RANDOM}, {'c', LXC_DEV_MAJ_MEMORY, LXC_DEV_MIN_URANDOM}, {'c', LXC_DEV_MAJ_TTY, LXC_DEV_MIN_TTY}, {'c', LXC_DEV_MAJ_TTY, LXC_DEV_MIN_PTMX}, {0, 0, 0}}; Please confirm this or let me know if there is any other interface libvirt_lxc provides to allow specific pci/other devices into container. Thanks in advance Regards Devendra ----- Original Message ----- From: "Alex Jia" <ajia(a)redhat.com> To: "Devendra K. Modium" <dmodium(a)isi.edu> Sent: Thursday, July 28, 2011 3:22:36 AM Subject: Re: [libvirt] PCI devices passthough to LXC containers using libvirt On 07/28/2011 05:13 AM, Devendra K. Modium wrote: > Hi All > > Please let me know if anyone have given access to > PCI devices for a LXC container. > > I have tried getting the xml from > "virsh nodedev-dumpxml pci_device" and > added to the libvirt xml file as shown below > > <device> > <name>pci_0000_03_00_0</name> > <parent>pci_0000_00_03_0</parent> > <driver> > <name>nvidia</name> > </driver> > <capability type='pci'> > <domain>0</domain> > <bus>3</bus> > <slot>0</slot> > <function>0</function> > <product id='0x06fd' /> > <vendor id='0x10de'>nVidia Corporation</vendor> > </capability> > </device> You shouldn't directly add the above xml to guest xml configuration, here is a usage, it should be helpful for you: http://libvirt.org/formatdomain.html#elementsUSB http://fedoraproject.org/wiki/Category:Virtualization_KVM_PCI_Device_Assign… Good Luck! Alex > > But it didn't work. I see the logs and it says > couldn't get physical and virtual functions of these devices with error > > get_physical_function_linux:323 : Attempting to get SR IOV physical function for device with sysfs path '/sys/devices/pci0000:00/0000:00:00.0' > 16:48:34.033: 13802: debug : get_sriov_function:270 : Attempting to resolve device path from device link '/sys/devices/pci0000:00/0000:00:00.0/physfn' > 16:48:34.033: 13802: debug : get_sriov_function:274 : SR IOV function link '/sys/devices/pci0000:00/0000:00:00.0/physfn' does not exist > 16:48:34.033: 13802: debug : get_virtual_functions_linux:348 : Attempting to get SR IOV virtual functions for devicewith sysfs path '/sys/devices/pci0000:00/0000:00:00.0' > > > If anyone got some guidelines how to debug, please let me know. > > > Thanks in advance > > Regards > Devendra > > > > > > > > -- > libvir-list mailing list > libvir-list(a)redhat.com > https://www.redhat.com/mailman/listinfo/libvir-list

2 1

[libvirt] [PATCH v3] virsh: avoid missing zero value judgement in cmdBlkiotune
by ajia＠redhat.com 28 Jul '11

28 Jul '11

* tools/virsh.c: avoid missing zero value judgement in cmdBlkiotune, when weight is equal to 0, the cmdBlkiotune will not raise any error information when judge weight value first time, and execute else branch to judge weight value again, strncpy(temp->field, VIR_DOMAIN_BLKIO_WEIGHT, sizeof(temp->field)) will be not executed for ever. However, if and only if param->field is equal to VIR_DOMAIN_BLKIO_WEIGHT, underlying qemuDomainSetBlkioParameters function will check whether weight value is in range [100, 1000]. * how to reproduce? % virsh blkiotune ${guestname} --weight 0 Signed-off-by: Alex Jia <ajia(a)redhat.com> --- tools/virsh.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 8bd22dc..f24050d 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -4037,12 +4037,12 @@ cmdBlkiotune(vshControl * ctl, const vshCmd * cmd) goto cleanup; } - if (weight) { - nparams++; - if (weight < 0) { + if (vshCommandOptInt(cmd, "weight", &weight) > 0) { + if (weight <= 0) { vshError(ctl, _("Invalid value of %d for I/O weight"), weight); goto cleanup; } + nparams++; } if (nparams == 0) { -- 1.7.1

4 3

[libvirt] [PATCH v2] virsh: avoid missing zero value judgement in cmdBlkiotune
by ajia＠redhat.com 28 Jul '11

28 Jul '11

* tools/virsh.c: avoid missing zero value judgement in cmdBlkiotune, when weight is equal to 0, the cmdBlkiotune will not raise any error information when judge weight value first time, and execute else branch to judge weight value again, strncpy(temp->field, VIR_DOMAIN_BLKIO_WEIGHT, sizeof(temp->field)) will be not executed for ever. However, if and only if param->field is equal to VIR_DOMAIN_BLKIO_WEIGHT, underlying qemuDomainSetBlkioParameters function will check whether weight value is in range [100, 1000]. * how to reproduce? % virsh blkiotune ${guestname} --weight 0 Signed-off-by: Alex Jia <ajia(a)redhat.com> --- tools/virsh.c | 11 +++++------ 1 files changed, 5 insertions(+), 6 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 8bd22dc..512f2c6 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -4037,14 +4037,13 @@ cmdBlkiotune(vshControl * ctl, const vshCmd * cmd) goto cleanup; } - if (weight) { - nparams++; - if (weight < 0) { - vshError(ctl, _("Invalid value of %d for I/O weight"), weight); - goto cleanup; - } + if (weight <= 0) { + vshError(ctl, _("Invalid value of %d for I/O weight"), weight); + goto cleanup; } + nparams++; + if (nparams == 0) { /* get the number of blkio parameters */ if (virDomainGetBlkioParameters(dom, NULL, &nparams, flags) != 0) { -- 1.7.1

3 5

[libvirt] RFC: API additions for enhanced snapshot support
by Eric Blake 28 Jul '11

28 Jul '11

Right now, libvirt has a snapshot API via virDomainSnapshotCreateXML, but for qemu domains, it only works if all the guest disk images are qcow2, and qemu rather than libvirt does all the work. However, it has a couple of drawbacks: it is inherently tied to domains (there is no way to manage snapshots of storage volumes not tied to a domain, even though libvirt does that for qcow2 images associated with offline qemu domains by using the qemu-img application). And it necessarily operates on all of the images associated with a domain in parallel - if any disk image is not qcow2, the snapshot fails, and there is no way to select a subset of disks to save. However, it works on both active (disk and memory state) and inactive domains (just disk state). Upstream qemu is developing a 'live snapshot' feature, which allows the creation of a snapshot without the current downtime of several seconds required by the current 'savevm' monitor command, as well as means for controlling applications (libvirt) to request that qemu pause I/O to a particular disk, then externally perform a snapshot, then tell qemu to resume I/O (perhaps on a different file name or fd from the host, but with no change to the contents seen by the guest). Eventually, these changes will make it possible for libvirt to create fast snapshots of LVM partitions or btrfs files for guest disk images, as well as to select which disks are saved in a snapshot (that is, save a crash-consistent state of a subset of disks, without the corresponding RAM state, rather than making a full system restore point); the latter would work best with guest cooperation to quiesce disks before qemu pauses I/O to that disk, but that is an orthogonal enhancement. However, my first goal with API enhancements is to merely prove that libvirt can manage a live snapshot by using qemu-img on a qcow2 image rather than the current 'savevm' approach of qemu doing all the work. Additionally, libvirt provides the virDomainSave command, which saves just the state of the domain's memory, and stops the guest. A crude libvirt-only snapshot could thus already be done by using virDomainSave, then externally doing a snapshot of all disk images associated with the domain by using virStorageVol APIs, except that such APIs don't yet exist. Additionally, virDomainSave has no flags argument, so there is no way to request that the guest be resumed after the snapshot completes. Right now, I'm proposing the addition of virDomainSaveFlags, along with a series of virStorageVolSnapshot* APIs that mirror the virDomainSnapshot* APIs. This would mean adding: /* Opaque type to manage a snapshot of a single storage volume. */ typedef virStorageVolSnapshotPtr; /* Create a snapshot of a storage volume. XML is optional, if non-NULL, it would be a new top-level element <volsnapshot> which is similar to the top-level <domainsnapshot> for virDomainSnapshotCreateXML, to specify name and description. Flags is 0 for now. */ virStorageVolSnapshotPtr virDomainSnapshotCreateXML(virStorageVolPtr vol, const char *xml, unsigned int flags); [For qcow2, this would be implemented with 'qemu-img snapshot -c', similar to what virDomainSnapshotXML already does on inactive domains. Later, we can add LVM and btrfs support, or even allow full file copies of any file type. Also in the future, we could enhance XML to take a new element that describes a relationship between the name of the original and of the snapshot, in the case where a new filename has to be created to complete the snapshot process.] /* Probe if vol has snapshots. 1 if true, 0 if false, -1 on error. Flags is 0 for now. */ int virStorageVolHasCurrentSnapshot(virStorageVolPtr vol, unsigned int flags); [For qcow2 images, snapshots can be contained within the same file and managed with qemu-img -l, but for other formats, this may mean that libvirt has to start managing externally saved data associated with the storage pool that associates snapshots with filenames. In fact, even for qcow2 it might be useful to support creation of new files backed by the previous snapshot rather than cramming multiple snapshots in one file, so we may have a use for flags to filter out the presence of single-file vs. multiple-file snapshot setups.] /* Revert a volume back to the state of a snapshot, returning 0 on success. Flags is 0 for now. */ int virStorageVolRevertToSnapsot(virStorageVolSnapshotPtr snapshot, unsigned int flags); [For qcow2, this would involve qemu-img snapshot -a. Here, a useful flag might be whether to delete any changes made after the point of the snapshot; virDomainRevertToSnapshot should probably honor the same type of flag.] /* Return the most recent snapshot of a volume, if one exists, or NULL on failure. Flags is 0 for now. */ virStorageVolSnapshotPtr virStorageVolSnapshotCurrent(virStorageVolPtr vol, unsigned int flags); /* Delete the storage associated with a snapshot (although the opaque snapshot object must still be independently freed). If flags is 0, any child snapshots based off of this one are rebased onto the parent; if flags is VIR_STORAGE_VOL_SNAPSHOT_DELETE_CHILDREN , then any child snapshots based off of this one are also deleted. */ int virStorageVolSnapshotDelete(virStorageVolSnapshotPtr snapshot, unsigned int flags); [For qcow2, this would involve qemu-img snapshot -d. For multiple-file snapshots, this would also involve qemu-img commit.] /* Free the object returned by virStorageVolSnapshot{Current,CreateXML,LookupByName}. The storage snapshot associated with this object still exists, if it has not been deleted by virStorageVolSnapshotDelete. */ int virStorageVolSnapshotFree(virStorageVolSnapshotPtr snapshot); /* Return the <volsnapshot> XML details about this snapshot object. Flags is 0 for now. */ int virStorageVolSnapshotGetXMLDesc(virStorageVolSnapshotPtr snapshot, unsigned int flags); /* Return the names of all snapshots associated with this volume, using len from virStorageVolSnapshotLen. Flags is 0 for now. */ int virStorageVolSnapshotListNames(virStorageVolPtr vol, char **names, int nameslen, unsigned int flags); [For qcow2, this involves qemu-img -l. Additionally, if virStorageVolHasCurrentSnapshot learns to filter on in-file vs. multi-file snapshots, then the same flags would apply here.] /* Get the opaque object tied to a snapshot name. Flags is 0 for now. */ virStorageVolSnapshotPtr virStorageVolSnapshotLookupByName(virStorageVolPtr vol, const char *name, unsigned int flags); /* Determine how many snapshots are tied to a volume, or -1 on error. Flags is 0 for now. */ int virStorageVolSnapshotNum(virStorageVolPtr vol, unsigned int flags); [Same flags as for virStorageVolSnapshotListNames.] /* Save a domain into the file 'to' with additional actions. If flags is 0, then xml is ignored, and this is like virDomainSave. If flags includes VIR_DOMAIN_SAVE_DISKS, then all of the associated disk images are also snapshotted, as if by virStorageVolSnapshotCreateXML; the xml argument is optional, but if present, it should be a <domainsnapshot> element with <disk> sub-elements for directions on each disk that needs a non-empty xml argument for proper volume snapshot creation. If flags includes VIR_DOMAIN_SAVE_RESUME, then the guest is resumed after the offline snapshot is complete (note that VIR_DOMAIN_SAVE_RESUME without VIR_DOMAIN_SAVE_DISKS makes little sense, as a saved state file is rendered useless if the disk images are modified before it is resumed). If flags includes VIR_DOMAIN_SAVE_QUIESCE, this requests that a guest agent quiesce disk state before the saved state file is created. */ int virDomainSaveFlags(virDomainPtr domain, const char *to, const char *xml, unsigned int flags); Also, the existing virDomainSnapshotCreateXML can be made more powerful by adding new flags and enhancing the existing XML for <domainsnapshot>. When flags is 0, the current behavior of saving memory state alongside all disks (for running domains, via savevm) or just snapshotting all disks with default settings (for offline domains, via qemu-img) is kept. If flags includes VIR_DOMAIN_SNAPSHOT_LIVE, then the guest must be running, and the new monitor commands for live snapshots are used. If flags includes VIR_DOMAIN_SNAPSHOT_DISKS_ONLY, then only the disks are snapshotted (on a running guest, this generally means they will only be crash-consistent, and will need an fsck before that disk state can be remounted), but it will shave off time by not saving memory. If flags includes VIR_DOMAIN_SNAPSHOT_QUIESCE, then this will additionally request that a guest agent quiesce disk state before the live snapshot is taken (increasing the likelihood of a stable disk, rather than a crash-consistent disk; but it requires cooperation from the guest so it is no more reliable than memballoon changes). As for the XML changes, it makes sense to snapshot just a subset of disks when you only care about crash-consistent state or if you can rely on a guest agent to quiesce the subset of disk(s) you care about, so the existing <domainsnapshot> element needs a new optional subelement to control which disks are snapshotted; additionally, this subelement will be useful for disk image formats that require additional complexity (such as a secondary file name, rather than the inline snapshot feature of qcow2). I'm envisioning something like the following: <domainsnapshot> <name>whatever</name> <disk name='/path/to/image1' snapshot='no'/> <disk name='/path/to/image2'> <volsnapshot>...</volsnapshot> </disk> </domainsnapshot> where there can be up to as many <disk> elements as there are disk <devices> in the domain xml; any domain disk not listed is given default treatment. The name attribute of <disk> is mandatory, in order to match this disk element to one of the domain disks. The snapshot='yes|no' attribute is optional, defaulting to yes, in order to skip a particular disk. The <volsnapshot> subelement is optional, but if present, it would be the same XML as is provided to the virStorageVolSnapshotCreateXML. [And since my first phase of implementation will be focused on inline qcow2 snapshots, I don't yet know what that XML will need to contain for any other type of snapshots, such as mapping out how the snapshot backing file will be named in relation to the possibly new live file.] Any feedback on this approach? Any other APIs that would be useful to add? I'd like to get all the new APIs in place for 0.9.3 with minimal qcow2 functionality, then use the time before 0.9.4 to further enhance the APIs to cover more snapshot cases but without having to add any new APIs. -- Eric Blake eblake(a)redhat.com +1-801-349-2682 Libvirt virtualization library http://libvirt.org

9 33

[libvirt] [PATCH] virsh: avoid missing zero value judgement in cmdBlkiotune
by ajia＠redhat.com 28 Jul '11

28 Jul '11

* tools/virsh.c: avoid missing zero value judgement in cmdBlkiotune, when weight is equal to 0, the cmdBlkiotune will not raise any error information when judge weight value first time, and execute else branch to judge weight value again, strncpy(temp->field, VIR_DOMAIN_BLKIO_WEIGHT, sizeof(temp->field)) will be not executed for ever. However, if and only if param->field is equal to VIR_DOMAIN_BLKIO_WEIGHT, underlying qemuDomainSetBlkioParameters function will check whether weight value is in range [100, 1000]. * how to reproduce? % virsh blkiotune ${guestname} --weight 0 Signed-off-by: Alex Jia <ajia(a)redhat.com> --- tools/virsh.c | 10 ++++------ 1 files changed, 4 insertions(+), 6 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 8bd22dc..183d7c6 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -4037,12 +4037,10 @@ cmdBlkiotune(vshControl * ctl, const vshCmd * cmd) goto cleanup; } - if (weight) { - nparams++; - if (weight < 0) { - vshError(ctl, _("Invalid value of %d for I/O weight"), weight); - goto cleanup; - } + nparams++; + if (weight <= 0) { + vshError(ctl, _("Invalid value of %d for I/O weight"), weight); + goto cleanup; } if (nparams == 0) { -- 1.7.1

1 0

[libvirt] [PATCH 1/4] python: add python binding for virDomainGetBlkioParameters
by Hu Tao 28 Jul '11

28 Jul '11

--- python/libvirt-override-api.xml | 6 +- python/libvirt-override.c | 79 +++++++++++++++++++++++++++++++++++++- 2 files changed, 79 insertions(+), 6 deletions(-) diff --git a/python/libvirt-override-api.xml b/python/libvirt-override-api.xml index a8929b1..00f8e6a 100644 --- a/python/libvirt-override-api.xml +++ b/python/libvirt-override-api.xml @@ -208,10 +208,10 @@ <arg name='params' type='virBlkioParameterPtr' info='pointer to blkio tunable objects'/> </function> <function name='virDomainGetBlkioParameters' file='python'> - <info>Get the blkio parameters, the @params array will be filled with the values.</info> - <return type='int' info='-1 in case of error, 0 in case of success.'/> + <info>Get the blkio parameters</info> + <return type='virSchedParameterPtr' info='None in case of error, returns a dictionary of params'/> <arg name='domain' type='virDomainPtr' info='pointer to domain object'/> - <arg name='params' type='virBlkioParameterPtr' info='pointer to blkio tunable objects'/> + <arg name='flags' type='int' info='an OR'ed set of virDomainModificationImpact'/> </function> <function name='virDomainSetMemoryParameters' file='python'> <info>Change the memory tunables</info> diff --git a/python/libvirt-override.c b/python/libvirt-override.c index 70e0238..beb0969 100644 --- a/python/libvirt-override.c +++ b/python/libvirt-override.c @@ -572,11 +572,84 @@ libvirt_virDomainSetBlkioParameters(PyObject *self ATTRIBUTE_UNUSED, return VIR_PY_INT_FAIL; } -/* FIXME: This is a place holder for the implementation. */ static PyObject * libvirt_virDomainGetBlkioParameters(PyObject *self ATTRIBUTE_UNUSED, - PyObject *args ATTRIBUTE_UNUSED) { - return VIR_PY_INT_FAIL; + PyObject *args) { + virDomainPtr domain; + PyObject *pyobj_domain, *info; + int i_retval; + int nparams = 0, i; + unsigned int flags; + virTypedParameterPtr params; + + if (!PyArg_ParseTuple(args, (char *)"Oi:virDomainGetBlkioParameters", + &pyobj_domain, &flags)) + return(NULL); + domain = (virDomainPtr) PyvirDomain_Get(pyobj_domain); + + LIBVIRT_BEGIN_ALLOW_THREADS; + i_retval = virDomainGetBlkioParameters(domain, NULL, &nparams, flags); + LIBVIRT_END_ALLOW_THREADS; + + if (i_retval < 0) + return VIR_PY_NONE; + + if ((params = malloc(sizeof(*params)*nparams)) == NULL) + return VIR_PY_NONE; + + LIBVIRT_BEGIN_ALLOW_THREADS; + i_retval = virDomainGetBlkioParameters(domain, params, &nparams, flags); + LIBVIRT_END_ALLOW_THREADS; + + if (i_retval < 0) { + free(params); + return VIR_PY_NONE; + } + + /* convert to a Python tuple of long objects */ + if ((info = PyDict_New()) == NULL) { + free(params); + return VIR_PY_NONE; + } + for (i = 0 ; i < nparams ; i++) { + PyObject *key, *val; + + switch (params[i].type) { + case VIR_TYPED_PARAM_INT: + val = PyInt_FromLong((long)params[i].value.i); + break; + + case VIR_TYPED_PARAM_UINT: + val = PyInt_FromLong((long)params[i].value.ui); + break; + + case VIR_TYPED_PARAM_LLONG: + val = PyLong_FromLongLong((long long)params[i].value.l); + break; + + case VIR_TYPED_PARAM_ULLONG: + val = PyLong_FromLongLong((long long)params[i].value.ul); + break; + + case VIR_TYPED_PARAM_DOUBLE: + val = PyFloat_FromDouble((double)params[i].value.d); + break; + + case VIR_TYPED_PARAM_BOOLEAN: + val = PyBool_FromLong((long)params[i].value.b); + break; + + default: + free(params); + Py_DECREF(info); + return VIR_PY_NONE; + } + + key = libvirt_constcharPtrWrap(params[i].field); + PyDict_SetItem(info, key, val); + } + free(params); + return(info); } /* FIXME: This is a place holder for the implementation. */ -- 1.7.3.1

2 8

[libvirt] [PATCH 00/19] Rollback migration when libvirtd restarts
by Jiri Denemark 28 Jul '11

28 Jul '11

This series is also available at https://gitorious.org/~jirka/libvirt/jirka-staging/commits/migration-recove… The series does several things: - persists current job and its phase in status xml - allows safe monitor commands to be run during migration/save/dump jobs - implements recovery when libvirtd is restarted while a job is active - consolidates some code and fixes bugs I found when working in the area The series is not perfect and still needs some corner cases to be fixed but I think it's better to send the series for review now and add small additional fixes in the next version(s) instead of waiting for it to be perfect. Jiri Denemark (19): qemu: Separate job related data into a new object qemu: Consolidate BeginJob{,WithDriver} into a single method qemu: Consolidate {Enter,Exit}Monitor{,WithDriver} qemu: Allow all query commands to be run during long jobs qemu: Save job type in domain status XML qemu: Recover from interrupted jobs qemu: Add support for job phase qemu: Consolidate qemuMigrationPrepare{Direct,Tunnel} qemu: Implement migration job phases qemu: Migration job on destination daemon qemu: Migration job on source daemon qemu: Recover from interrupted migration qemu: Fix monitor unlocking in some error paths qemu: Remove special case for virDomainGetBlockInfo qemu: Remove special case for virDomainBlockStats qemu: Remove special case for virDomainMigrateSetMaxSpeed qemu: Remove special case for virDomainMigrateSetMaxDowntime qemu: Remove special case for virDomainSuspend qemu: Remove special case for virDomainAbortJob src/libvirt.c | 11 +- src/qemu/MIGRATION.txt | 55 +++ src/qemu/qemu_domain.c | 575 ++++++++++++++++++++++++------- src/qemu/qemu_domain.h | 130 +++++-- src/qemu/qemu_driver.c | 467 +++++++++++++------------ src/qemu/qemu_hotplug.c | 50 ++-- src/qemu/qemu_migration.c | 849 +++++++++++++++++++++++---------------------- src/qemu/qemu_migration.h | 36 ++ src/qemu/qemu_process.c | 244 ++++++++++++- 9 files changed, 1569 insertions(+), 848 deletions(-) create mode 100644 src/qemu/MIGRATION.txt -- 1.7.6

4 39

[libvirt] [PATCH] virsh: fix memory leak in cmdVolPath code
by ajia＠redhat.com 28 Jul '11

28 Jul '11

* tools/virsh.c: avoid memory leak in cmdVolPath. * how to reproduce? % dd if=/dev/zero of=/var/lib/libvirt/images/foo.img count=1 bs=10M % virsh pool-refresh default % valgrind -v --leak-check=full virsh vol-path --vol /var/lib/libvirt/images/foo.img * actual results: Detected in valgrind run: ==16436== 32 bytes in 1 blocks are definitely lost in loss record 7 of 22 ==16436== at 0x4A05FDE: malloc (vg_replace_malloc.c:236) ==16436== by 0x386A314B3D: xdr_string (in /lib64/libc-2.12.so) ==16436== by 0x3DF8CD770D: xdr_remote_nonnull_string (remote_protocol.c:30) ==16436== by 0x3DF8CD7EC8: xdr_remote_storage_vol_get_path_ret (remote_protocol.c:2952) ==16436== by 0x3DF8CDF161: virNetMessageDecodePayload (virnetmessage.c:286) ==16436== by 0x3DF8CDE9E5: virNetClientProgramCall (virnetclientprogram.c:318) ==16436== by 0x3DF8CC28A2: call (remote_driver.c:3929) ==16436== by 0x3DF8CC8412: remoteStorageVolGetPath (remote_client_bodies.h:5219) ==16436== by 0x3DF8C9BF14: virStorageVolGetPath (libvirt.c:11389) ==16436== by 0x418ECA: cmdVolPath (virsh.c:8754) ==16436== by 0x410CC2: vshCommandRun (virsh.c:12758) ==16436== by 0x41F286: main (virsh.c:14110) ==16436== ==16436== LEAK SUMMARY: ==16436== definitely lost: 32 bytes in 1 blocks Signed-off-by: Alex Jia <ajia(a)redhat.com> --- tools/virsh.c | 9 ++++++++- 1 files changed, 8 insertions(+), 1 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 841df61..d194a8b 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -9326,6 +9326,7 @@ cmdVolPath(vshControl *ctl, const vshCmd *cmd) { virStorageVolPtr vol; const char *name = NULL; + char * StorageVolPath; if (!vshConnectionUsability(ctl, ctl->conn)) return false; @@ -9334,7 +9335,13 @@ cmdVolPath(vshControl *ctl, const vshCmd *cmd) return false; } - vshPrint(ctl, "%s\n", virStorageVolGetPath(vol)); + if ((StorageVolPath = virStorageVolGetPath(vol)) == NULL) { + virStorageVolFree(vol); + return false; + } + + vshPrint(ctl, "%s\n", StorageVolPath); + VIR_FREE(StorageVolPath); virStorageVolFree(vol); return true; } -- 1.7.1

3 3

[libvirt] Memory leak of the remoteDomainSet* functions
by Osier Yang 28 Jul '11

28 Jul '11

Hello, there Per bug https://bugzilla.redhat.com/show_bug.cgi?id=725322, should we change the remoteDomainSet* functions into "skipgen", and fix the leaks like below? (NB, new VIR_FREE statements) static int remoteDomainSetBlkioParameters(virDomainPtr dom, virTypedParameterPtr params, int nparams, unsigned int flags) { int rv = -1; int i; struct private_data *priv = dom->conn->privateData; remote_domain_set_blkio_parameters_args args; remote_domain_set_blkio_parameters_args args; remoteDriverLock(priv); make_nonnull_domain(&args.dom, dom); args.flags = flags; if (remoteSerializeTypedParameters(params, nparams, &args.params.params_val, &args.params.params_len) < 0) { for (i = 0; i < nparams; i++) { VIR_FREE(args.params.params_val[i].field); } VIR_FREE(args.params.params_val); xdr_free((xdrproc_t)xdr_remote_domain_set_blkio_parameters_args, (char *)&args); goto done; } if (call(dom->conn, priv, 0, REMOTE_PROC_DOMAIN_SET_BLKIO_PARAMETERS, (xdrproc_t)xdr_remote_domain_set_blkio_parameters_args, (char *)&args, (xdrproc_t)xdr_void, (char *)NULL) == -1) { goto done; } rv = 0; done: if (args.params.params_val) { for (i = 0; i < nparams; i++) { VIR_FREE(args.params.params_val[i].field); } VIR_FREE(args.params.params_val); } remoteDriverUnlock(priv); return rv; } Regards, Osier

2 3

[libvirt] [PATCH] qemu: Remove bogus codes in function getCompressionType
by Osier Yang 28 Jul '11

28 Jul '11

The error will never be reported, remove the codes, and also improve the docs in qemu.conf to tell user the truth. --- src/qemu/qemu.conf | 4 ++++ src/qemu/qemu_driver.c | 16 +++++----------- 2 files changed, 9 insertions(+), 11 deletions(-) diff --git a/src/qemu/qemu.conf b/src/qemu/qemu.conf index 145062c..75d945b 100644 --- a/src/qemu/qemu.conf +++ b/src/qemu/qemu.conf @@ -199,6 +199,10 @@ # save_image_format is used when you use 'virsh save' at scheduled saving. # dump_image_format is used when you use 'virsh dump' at emergency crashdump. # +# If the specified format is not valid (the valid formats are "raw", "lzop", +# "gzip", "bzip2", and "xz"), or the compress program is not available, "raw" +# will be used as the format silently without error or warning. +# # save_image_format = "raw" # dump_image_format = "raw" diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index b928004..48fc46a 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -2712,19 +2712,13 @@ getCompressionType(struct qemud_driver *driver) */ if (driver->dumpImageFormat) { compress = qemudSaveCompressionTypeFromString(driver->dumpImageFormat); - if (compress < 0) { - qemuReportError(VIR_ERR_OPERATION_FAILED, "%s", - _("Invalid dump image format specified in " - "configuration file, using raw")); + /* Use "raw" as the format if the specified format is not valid, + * or the compress program is not available, + */ + if (compress < 0) return QEMUD_SAVE_FORMAT_RAW; - } - if (!qemudCompressProgramAvailable(compress)) { - qemuReportError(VIR_ERR_OPERATION_FAILED, - "%s", _("Compression program for dump image format " - "in configuration file isn't available, " - "using raw")); + if (!qemudCompressProgramAvailable(compress)) return QEMUD_SAVE_FORMAT_RAW; - } } return compress; } -- 1.7.6

2 4

[libvirt] [PATCH] qemu: improve thread documentation
by Eric Blake 28 Jul '11

28 Jul '11

* src/qemu/THREADS.txt: Fix problems with typos, grammar, and outdated examples. --- Doc only, so pushing under the trivial rule. src/qemu/THREADS.txt | 30 ++++++++++++++++-------------- 1 files changed, 16 insertions(+), 14 deletions(-) diff --git a/src/qemu/THREADS.txt b/src/qemu/THREADS.txt index 3a27a85..e73076c 100644 --- a/src/qemu/THREADS.txt +++ b/src/qemu/THREADS.txt @@ -6,7 +6,7 @@ the QEMU driver. The criteria for this model are: - Objects must never be exclusively locked for any prolonged time - Code which sleeps must be able to time out after suitable period - - Must be safe against dispatch asynchronous events from monitor + - Must be safe against dispatch of asynchronous events from monitor Basic locking primitives @@ -19,7 +19,7 @@ There are a number of locks on various objects This is the top level lock on the entire driver. Every API call in the QEMU driver is blocked while this is held, though some internal callbacks may still run asynchronously. This lock must never be held - for anything which sleeps/waits (ie monitor commands) + for anything which sleeps/waits (i.e. monitor commands) When obtaining the driver lock, under *NO* circumstances must any lock be held on a virDomainObjPtr. This *WILL* result in @@ -44,7 +44,7 @@ There are a number of locks on various objects to have the driver locked when re-acquiring the dropped locked, since the reference count prevents it being freed by another thread. - This lock must not be held for anything which sleeps/waits (ie monitor + This lock must not be held for anything which sleeps/waits (i.e. monitor commands). @@ -80,7 +80,7 @@ There are a number of locks on various objects whenever it hits a piece of code which may sleep/wait, and re-acquire it after the sleep/wait. Whenever an asynchronous job wants to talk to the monitor, it needs to acquire nested job (a - special kind of normla job) to obtain exclusive access to the + special kind of normal job) to obtain exclusive access to the monitor. Since the virDomainObjPtr lock was dropped while waiting for the @@ -139,7 +139,7 @@ To acquire the normal job condition - Increments ref count on virDomainObjPtr - Waits until the job is compatible with current async job or no async job is running - - Waits job.cond condition 'job.active != 0' using virDomainObjPtr + - Waits for job.cond condition 'job.active != 0' using virDomainObjPtr mutex - Rechecks if the job is still compatible and repeats waiting if it isn't @@ -150,7 +150,7 @@ To acquire the normal job condition - Unlocks driver - Waits until the job is compatible with current async job or no async job is running - - Waits job.cond condition 'job.active != 0' using virDomainObjPtr + - Waits for job.cond condition 'job.active != 0' using virDomainObjPtr mutex - Rechecks if the job is still compatible and repeats waiting if it isn't @@ -160,7 +160,7 @@ To acquire the normal job condition - Locks virDomainObjPtr NB: this variant is required in order to comply with lock ordering - rules for virDomainObjPtr vs driver + rules for virDomainObjPtr vs. driver qemuDomainObjEndJob() @@ -175,7 +175,7 @@ To acquire the asynchronous job condition qemuDomainObjBeginAsyncJob() (if driver is unlocked) - Increments ref count on virDomainObjPtr - Waits until no async job is running - - Waits job.cond condition 'job.active != 0' using virDomainObjPtr + - Waits for job.cond condition 'job.active != 0' using virDomainObjPtr mutex - Rechecks if any async job was started while waiting on job.cond and repeats waiting in that case @@ -185,7 +185,7 @@ To acquire the asynchronous job condition - Increments ref count on virDomainObjPtr - Unlocks driver - Waits until no async job is running - - Waits job.cond condition 'job.active != 0' using virDomainObjPtr + - Waits for job.cond condition 'job.active != 0' using virDomainObjPtr mutex - Rechecks if any async job was started while waiting on job.cond and repeats waiting in that case @@ -271,7 +271,7 @@ Design patterns - * Accessing something directly todo with a virDomainObjPtr + * Accessing something directly to do with a virDomainObjPtr virDomainObjPtr obj; @@ -285,7 +285,7 @@ Design patterns - * Accessing something directly todo with a virDomainObjPtr and driver + * Accessing something directly to do with a virDomainObjPtr and driver virDomainObjPtr obj; @@ -299,11 +299,11 @@ Design patterns - * Updating something directly todo with a virDomainObjPtr + * Updating something directly to do with a virDomainObjPtr virDomainObjPtr obj; - qemuDriverLockRO(driver); + qemuDriverLock(driver); obj = virDomainFindByUUID(driver->domains, dom->uuid); qemuDriverUnlock(driver); @@ -324,7 +324,7 @@ Design patterns virDomainObjPtr obj; qemuDomainObjPrivatePtr priv; - qemuDriverLockRO(driver); + qemuDriverLock(driver); obj = virDomainFindByUUID(driver->domains, dom->uuid); qemuDriverUnlock(driver); @@ -333,6 +333,7 @@ Design patterns ...do prep work... if (virDomainObjIsActive(vm)) { + /* using ignore_value is safe since vm is active */ ignore_value(qemuDomainObjEnterMonitor(obj)); qemuMonitorXXXX(priv->mon); qemuDomainObjExitMonitor(obj); @@ -360,6 +361,7 @@ Design patterns ...do prep work... if (virDomainObjIsActive(vm)) { + /* using ignore_value is safe since vm is active */ ignore_value(qemuDomainObjEnterMonitorWithDriver(driver, obj)); qemuMonitorXXXX(priv->mon); qemuDomainObjExitMonitorWithDriver(driver, obj); -- 1.7.4.4

1 0

[libvirt] [PATCH] virsh: expose change-protection during migration
by Eric Blake 27 Jul '11

27 Jul '11

* tools/virsh.c (doMigrate): Add --change-protection flag. * tools/virsh.pod (migrate): Document it. --- As promised here: https://www.redhat.com/archives/libvir-list/2011-July/msg01919.html tools/virsh.c | 4 ++++ tools/virsh.pod | 9 +++++++-- 2 files changed, 11 insertions(+), 2 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index 113124f..34036f9 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -4830,6 +4830,8 @@ static const vshCmdOptDef opts_migrate[] = { {"suspend", VSH_OT_BOOL, 0, N_("do not restart the domain on the destination host")}, {"copy-storage-all", VSH_OT_BOOL, 0, N_("migration with non-shared storage with full disk copy")}, {"copy-storage-inc", VSH_OT_BOOL, 0, N_("migration with non-shared storage with incremental copy (same base image shared between source and destination)")}, + {"change-protection", VSH_OT_BOOL, 0, + N_("prevent any configuration changes to source until migration ends)")}, {"verbose", VSH_OT_BOOL, 0, N_("display the progress of migration")}, {"domain", VSH_OT_DATA, VSH_OFLAG_REQ, N_("domain name, id or uuid")}, {"desturi", VSH_OT_DATA, VSH_OFLAG_REQ, N_("connection URI of the destination host as seen from the client(normal migration) or source(p2p migration)")}, @@ -4906,6 +4908,8 @@ doMigrate (void *opaque) if (vshCommandOptBool (cmd, "copy-storage-inc")) flags |= VIR_MIGRATE_NON_SHARED_INC; + if (vshCommandOptBool (cmd, "change-protection")) + flags |= VIR_MIGRATE_CHANGE_PROTECTION; if (xmlfile && virFileReadAll(xmlfile, 8192, &xml) < 0) diff --git a/tools/virsh.pod b/tools/virsh.pod index e9aaa80..6c3c960 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -564,8 +564,9 @@ type attribute for the <domain> element of XML. =item B<migrate> [I<--live>] [I<--direct>] [I<--p2p> [I<--tunnelled>]] [I<--persistent>] [I<--undefinesource>] [I<--suspend>] [I<--copy-storage-all>] -[I<--copy-storage-inc>] [I<--verbose>] I<domain-id> I<desturi> [I<migrateuri>] -[I<dname>] [I<--timeout> B<seconds>] [I<--xml> B<file>] +[I<--copy-storage-inc>] [I<--change-protection>] [I<--verbose>] +I<domain-id> I<desturi> [I<migrateuri>] [I<dname>] +[I<--timeout> B<seconds>] [I<--xml> B<file>] Migrate domain to another host. Add I<--live> for live migration; I<--p2p> for peer-2-peer migration; I<--direct> for direct migration; or I<--tunnelled> @@ -575,6 +576,10 @@ and I<--suspend> leaves the domain paused on the destination host. I<--copy-storage-all> indicates migration with non-shared storage with full disk copy, I<--copy-storage-inc> indicates migration with non-shared storage with incremental copy (same base image shared between source and destination). +I<--change-protection> enforces that no incompatible configuration changes +will be made to the domain while the migration is underway; this flag is +implicitly enabled when supported by the hypervisor, but can be used to +reject the migration if the hypervisor lacks change protection support. I<--verbose> displays the progress of migration. The I<desturi> is the connection URI of the destination host, and -- 1.7.4.4

2 2

Re: [libvirt] [virt-tools-list] virt-manager/libvirt backwards compatibility problem?
by Cole Robinson 27 Jul '11

27 Jul '11

On 07/27/2011 12:28 PM, Whit Blauvelt wrote: > Looks like my real problem may be not incorporating a Debian/Ubuntu patch > before building 0.9.x, since netcat differs: > > https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/517478 > Hmm, what was the error message you were getting? We've made efforts at the libvirt level to have useful error reporting here, so that the cause of failure is more obvious. Also, I patched virt-manager a while ago to use a shell script to handle both RH and debian 'nc' versions. Script is here: http://git.fedorahosted.org/git?p=virt-manager.git;a=blob;f=src/virtManager… Yeah, it's complete and utter crack, but this has been a constant frustration for users, so maybe worth revisiting at the libvirt level (and hey, launching nc over ssh is kinda crack to being with :) ). Even if we solved this in a clean way by providing our own /usr/libexec/libvirt_nc binary, we'd still probably have to play shell script games over SSH to handle connecting to older libvirt (or what if libvirt_nc was installed somewhere other than libexec?) - Cole

3 2

[libvirt] [PATCH] rpc: Fix memory leak in remoteDomainSet*Parameters functions
by Matthias Bolte 27 Jul '11

27 Jul '11

Add a new helper remoteFreeTypedParameters and teach the generator to add it to the cleanup section. https://bugzilla.redhat.com/show_bug.cgi?id=725322 --- src/remote/remote_driver.c | 22 +++++++++++++++++----- src/rpc/gendispatch.pl | 5 +++++ 2 files changed, 22 insertions(+), 5 deletions(-) diff --git a/src/remote/remote_driver.c b/src/remote/remote_driver.c index 0652e0d..e5bfa4b 100644 --- a/src/remote/remote_driver.c +++ b/src/remote/remote_driver.c @@ -1208,6 +1208,22 @@ done: return rv; } +/* Helper to free typed parameters. */ +static void +remoteFreeTypedParameters(remote_typed_param *args_params_val, + u_int args_params_len) +{ + u_int i; + + if (args_params_val == NULL) + return; + + for (i = 0; i < args_params_len; i++) + VIR_FREE(args_params_val[i].field); + + VIR_FREE(args_params_val); +} + /* Helper to serialize typed parameters. */ static int remoteSerializeTypedParameters(virTypedParameterPtr params, @@ -1264,11 +1280,7 @@ remoteSerializeTypedParameters(virTypedParameterPtr params, rv = 0; cleanup: - if (val) { - for (i = 0; i < nparams; i++) - VIR_FREE(val[i].field); - VIR_FREE(val); - } + remoteFreeTypedParameters(val, nparams); return rv; } diff --git a/src/rpc/gendispatch.pl b/src/rpc/gendispatch.pl index ceeb1f5..0d344e8 100755 --- a/src/rpc/gendispatch.pl +++ b/src/rpc/gendispatch.pl @@ -979,6 +979,7 @@ elsif ($opt_k) { my @args_check_list = (); my @setters_list = (); my @setters_list2 = (); + my @free_list = (); my $priv_src = "conn"; my $priv_name = "privateData"; my $call_args = "&args"; @@ -1105,6 +1106,7 @@ elsif ($opt_k) { " xdr_free((xdrproc_t)xdr_$call->{args}, (char *)&args);\n" . " goto done;\n" . " }"); + push(@free_list, " remoteFreeTypedParameters(args.params.params_val, args.params.params_len);\n"); } elsif ($args_member =~ m/^((?:unsigned )?int) (\S+);\s*\/\*\s*call-by-reference\s*\*\//) { my $type_name = "$1 *"; my $arg_name = $2; @@ -1500,6 +1502,9 @@ elsif ($opt_k) { print "\n"; print "done:\n"; + + print join("\n", @free_list); + print " remoteDriverUnlock(priv);\n"; print " return rv;\n"; print "}\n"; -- 1.7.4.1

2 2

[libvirt] Environmental Variables inside a container
by Devendra K. Modium 27 Jul '11

27 Jul '11

Hi All I am trying to pass an environmental flag to the LXC container when it is started.I found libvirt_lxc is already doing it in lxc_container.c as shown below static virCommandPtr lxcContainerBuildInitCmd(virDomainDefPtr vmDef) { char uuidstr[VIR_UUID_STRING_BUFLEN]; virCommandPtr cmd; virUUIDFormat(vmDef->uuid, uuidstr); cmd = virCommandNew(vmDef->os.init); virCommandAddEnvString(cmd, "PATH=/bin:/sbin"); virCommandAddEnvString(cmd, "TERM=linux-deva"); virCommandAddEnvPair(cmd, "LIBVIRT_LXC_UUID", uuidstr); virCommandAddEnvPair(cmd, "LIBVIRT_LXC_NAME", vmDef->name); return cmd; } But only TERM flag is getting visible inside the container not remaining. Does it depends on how /sbin/init is receiving them? Is /sbin/init only recognizing TERM flag. If so what is intention of sending remaining flags in current libvirt. Please let me know if there is any elegant way to send environmental variable to container. Thanks for your time. Regards Devendra

2 2

[libvirt] Add support for image write mode in virtualbox driver
by Boris FELD 27 Jul '11

27 Jul '11

Hi everyone, We currently using libvirt for PYTI GSOC Project (it works very well!) and we encounter a problem. We use VM with two disk images, one which must to be snapshoted/rollbacked and the other one which must be not. We use VirtualBox as hypervisor and to do so, VirtualBox define "Special image write modes" (http://www.virtualbox.org/manual/ch05.html#hdimagewrites) What we need is to set the type of one image to "writethrough". This information seems defined in VM config file. For instance: <HardDisk uuid="{402aef07-4df8-4518-831c-eeb9ae2a32d2}" location="/home/boris/VirtualBox VMs/PYTI.vdi" format="VDI" type="Normal"/> <HardDisk uuid="{63acf6d9-b7db-4d8d-8389-1276c3d500bb}" location="/home/boris/VirtualBox VMs/adhock.vdi" format="VDI" type="Writethrough"/> I found two pages mentioning the type of medium, this one: http://www.virtualbox.org/sdkref/_virtual_box_8idl.html#b4abb9d1b9e44b997d1… and this one: http://www.virtualbox.org/sdkref/interface_i_medium_attachment.html. Hope it will help. Cheers, FELD Boris

1 0

[libvirt] [PATCH 00/10] Rollback migration when libvirtd restarts
by Jiri Denemark 27 Jul '11

27 Jul '11

This is the rest of the original 19 patch series updated with some bugfixes and rebased on current master, which is also available at https://gitorious.org/~jirka/libvirt/jirka-staging/commits/migration-recove… I didn't manage to run this through the libvirt-tck migration test but I did some additional testing with the following combinations of libvirt (current contains this series): source destination client ------------------------------------------- current current current current current 0.9.2-1.el6 current 0.9.2-2.fc15 current 0.9.2-2.fc15 current 0.9.2-1.el6 In all combinations I did - normal migration from src to dst - normal migration back - p2p migration from src to dst - p2p migration back To test failure recovery, I aborted the migration or libvirtd deamons at various places. All this was done with a single running domain without restarting it. Jiri Denemark (10): qemu: Implement migration job phases qemu: Migration job on destination daemon qemu: Migration job on source daemon qemu: Recover from interrupted migration qemu: Remove special case for virDomainGetBlockInfo qemu: Remove special case for virDomainBlockStats qemu: Remove special case for virDomainMigrateSetMaxSpeed qemu: Remove special case for virDomainMigrateSetMaxDowntime qemu: Remove special case for virDomainSuspend qemu: Remove special case for virDomainAbortJob include/libvirt/libvirt.h.in | 3 + src/libvirt.c | 27 ++- src/libvirt_internal.h | 6 + src/qemu/MIGRATION.txt | 55 ++++ src/qemu/qemu_domain.c | 18 +- src/qemu/qemu_domain.h | 31 +-- src/qemu/qemu_driver.c | 285 ++++++++++++--------- src/qemu/qemu_migration.c | 574 ++++++++++++++++++++++++------------------ src/qemu/qemu_migration.h | 40 +++- src/qemu/qemu_process.c | 121 +++++++++- 10 files changed, 750 insertions(+), 410 deletions(-) create mode 100644 src/qemu/MIGRATION.txt -- 1.7.6

3 24