- Devel - libvirt List Archives

[libvirt] [PATCH v2] qemu: fix domain id after domainCreateWithFlags()
by Sascha Silbe 01 Aug '16

01 Aug '16

Ever since virDomainCreateWithFlags() was introduced by de3aadaa [drivers: add virDomainCreateWithFlags if virDomainCreate exists], the domain ID retrieved with virDomainGetID() was incorrect for several drivers after virDomainCreateWithFlags() was called. The API consumer had to look up the domain anew to retrieve the correct ID. For the ESX driver, this was fixed in 6139b274 [esx: Update ID after starting a domain]. For the openvz driver, it was fixed in fd81a097 [openvzDomainCreateWithFlags: set domain id to the correct value]. The test driver, the OpenNebula driver (removed in the meantime) and the vbox driver were already updating the domain ID correctly in domainCreate(). Copy over the ID in qemuDomainCreateWithFlags() to fix this for the qemu driver, too. Fixes: de3aadaa ("drivers: add virDomainCreateWithFlags if virDomainCreate exists") Reported-by: Marc Hartmayer <mhartmay(a)linux.vnet.ibm.com> Signed-off-by: Sascha Silbe <silbe(a)linux.vnet.ibm.com> Tested-by: Marc Hartmayer <mhartmay(a)linux.vnet.ibm.com> Reviewed-by: Marc Hartmayer <mhartmay(a)linux.vnet.ibm.com> --- v1→v2: - Adjusted commit message. remoteDomainCreate() is actually fine since it explicitly queries the ID in a second RPC, only remoteDomainCreateWithFlags() is broken. Found and mentioned several other commits that fixed this for several of the other hypervisor drivers. src/qemu/qemu_driver.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 1820f85..43242e0 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -7100,6 +7100,7 @@ qemuDomainCreateWithFlags(virDomainPtr dom, unsigned int flags) QEMU_ASYNC_JOB_START) < 0) goto endjob; + dom->id = vm->def->id; ret = 0; endjob: -- 1.9.1

2 1

[libvirt] [PATCH] wireshark: Drop glib dependency
by Michal Privoznik 01 Aug '16

01 Aug '16

The only function that we currently use from glib is g_sprintf(). That's a very big gun for such small target. Not only that, but we've silently relied on wireshark dragging in the glib. Replace the g_sprintf() with plain sprinf() so that we can drop the glib dependency. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tools/wireshark/src/packet-libvirt.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/tools/wireshark/src/packet-libvirt.c b/tools/wireshark/src/packet-libvirt.c index 569a0a1..9eadf4f 100644 --- a/tools/wireshark/src/packet-libvirt.c +++ b/tools/wireshark/src/packet-libvirt.c @@ -27,8 +27,6 @@ #include <wireshark/epan/proto.h> #include <wireshark/epan/packet.h> #include <wireshark/epan/dissectors/packet-tcp.h> -#include <glib.h> -#include <glib/gprintf.h> #ifdef HAVE_RPC_TYPES_H # include <rpc/types.h> #endif @@ -113,7 +111,7 @@ format_xdr_bytes(guint8 *bytes, guint32 length) for (i = 0; i < length; i++) { /* We know that buf has enough size to contain 2 * length + '\0' characters. */ - g_sprintf(buf, "%02x", bytes[i]); + sprintf(buf, "%02x", bytes[i]); buf += 2; } return buf - length*2; -- 2.8.4

2 1

[libvirt] [PATCH] tests: qemu: Don't leak security manager object
by Peter Krempa 01 Aug '16

01 Aug '16

==2064442== 200 (88 direct, 112 indirect) bytes in 1 blocks are definitely lost in loss record 54 of 73 ==2064442== at 0x4C2E0F0: calloc (vg_replace_malloc.c:711) ==2064442== by 0x18E75B80: virAllocVar (viralloc.c:560) ==2064442== by 0x18EC43B0: virObjectNew (virobject.c:193) ==2064442== by 0x18EC476E: virObjectLockableNew (virobject.c:219) ==2064442== by 0x1906BC73: virSecurityManagerNewDriver (security_manager.c:93) ==2064442== by 0x1906C076: virSecurityManagerNewStack (security_manager.c:115) ==2064442== by 0x43CC39: qemuTestDriverInit (testutilsqemu.c:548) ==2064442== by 0x4337ED: mymain (qemumonitorjsontest.c:2440) ==2064442== by 0x43BABE: virTestMain (testutils.c:982) ==2064442== by 0x43A490: main (qemumonitorjsontest.c:2558) --- tests/testutilsqemu.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tests/testutilsqemu.c b/tests/testutilsqemu.c index 1cf72a9..133de29 100644 --- a/tests/testutilsqemu.c +++ b/tests/testutilsqemu.c @@ -475,6 +475,7 @@ void qemuTestDriverFree(virQEMUDriver *driver) virObjectUnref(driver->xmlopt); virObjectUnref(driver->caps); virObjectUnref(driver->config); + virObjectUnref(driver->securityManager); } int qemuTestCapsCacheInsert(virQEMUCapsCachePtr cache, const char *binary, -- 2.9.2

2 1

[libvirt] [PATCH] spec: Require perl-XML-XPath during build
by Martin Kletzander 01 Aug '16

01 Aug '16

Some time ago I went through the trouble to make support for perl module requirements in our bootstrap. Even though we require it in the bootstrap, it was not listed as a build requirement in the specfile. Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- libvirt.spec.in | 1 + 1 file changed, 1 insertion(+) diff --git a/libvirt.spec.in b/libvirt.spec.in index cc5d72209873..b182f4521453 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -269,6 +269,7 @@ BuildRequires: /usr/bin/pod2man %endif BuildRequires: git BuildRequires: perl +BuildRequires: perl-XML-XPath BuildRequires: python %if %{with_systemd} BuildRequires: systemd-units -- 2.9.2

3 4

[libvirt] [PATCH 0/2] qemu: caps: Sanitize storage of machine type related data
by Peter Krempa 01 Aug '16

01 Aug '16

Peter Krempa (2): qemu: capabilities: Drop unused function virQEMUCapsGetMachineTypes qemu: caps: Sanitize storage of machine type related data src/qemu/qemu_capabilities.c | 125 ++++++++++++++++--------------------------- src/qemu/qemu_capabilities.h | 2 - 2 files changed, 46 insertions(+), 81 deletions(-) -- 2.9.2

2 5

[libvirt] [PATCH] virsh: Report error when explicit connection fails
by Martin Kletzander 01 Aug '16

01 Aug '16

Commit 0c56d9431839 forgot to return false in the cmdConnect command after the clean up made there. Before (assuming you don't have uri alias for 'asdf'): $ virsh connect asdf error: failed to connect to the hypervisor $ echo $? 0 After (with the same assumption): $ virsh connect asdf error: failed to connect to the hypervisor error: no connection driver available for asdf $ echo $? 1 Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1356461 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- tools/virsh.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/tools/virsh.c b/tools/virsh.c index f74698fa5f9a..d3fe06f19032 100644 --- a/tools/virsh.c +++ b/tools/virsh.c @@ -210,7 +210,7 @@ virshConnect(vshControl *ctl, const char *uri, bool readonly) * Reconnect after a disconnect from libvirtd * */ -static void +static int virshReconnect(vshControl *ctl, const char *name, bool readonly, bool force) { bool connected = false; @@ -237,6 +237,7 @@ virshReconnect(vshControl *ctl, const char *name, bool readonly, bool force) vshError(ctl, "%s", _("Failed to reconnect to the hypervisor")); else vshError(ctl, "%s", _("failed to connect to the hypervisor")); + return -1; } else { if (name) { VIR_FREE(ctl->connname); @@ -253,6 +254,7 @@ virshReconnect(vshControl *ctl, const char *name, bool readonly, bool force) priv->useGetInfo = false; priv->useSnapshotOld = false; priv->blockJobNoBytes = false; + return 0; } int virshStreamSink(virStreamPtr st ATTRIBUTE_UNUSED, @@ -301,7 +303,8 @@ cmdConnect(vshControl *ctl, const vshCmd *cmd) if (vshCommandOptStringReq(ctl, cmd, "name", &name) < 0) return false; - virshReconnect(ctl, name, ro, true); + if (virshReconnect(ctl, name, ro, true) < 0) + return false; return true; } @@ -333,11 +336,13 @@ virshConnectionHandler(vshControl *ctl) { virshControlPtr priv = ctl->privData; - if (!priv->conn || disconnected) - virshReconnect(ctl, NULL, false, false); + if ((!priv->conn || disconnected) && + virshReconnect(ctl, NULL, false, false) < 0) + return NULL; if (virshConnectionUsability(ctl, priv->conn)) return priv->conn; + return NULL; } @@ -444,14 +449,13 @@ virshInit(vshControl *ctl) return false; if (ctl->connname) { - virshReconnect(ctl, NULL, false, false); /* Connecting to a named connection must succeed, but we delay * connecting to the default connection until we need it * (since the first command might be 'connect' which allows a * non-default connection, or might be 'help' which needs no * connection). */ - if (!priv->conn) { + if (virshReconnect(ctl, NULL, false, false) < 0) { vshReportError(ctl); return false; } -- 2.9.2

2 1

[libvirt] [PATCH] qemuDomainDetermineDiskChain: initialize ret as -1
by Chen Hanxiao 31 Jul '16

31 Jul '16

From: Chen Hanxiao <chenhanxiao(a)gmail.com> Signed-off-by: Chen Hanxiao <chenhanxiao(a)gmail.com> --- src/qemu/qemu_domain.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 9b439df..35d4365 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -4394,7 +4394,7 @@ qemuDomainDetermineDiskChain(virQEMUDriverPtr driver, bool report_broken) { virQEMUDriverConfigPtr cfg = virQEMUDriverGetConfig(driver); - int ret = 0; + int ret = -1; uid_t uid; gid_t gid; @@ -4416,6 +4416,8 @@ qemuDomainDetermineDiskChain(virQEMUDriverPtr driver, report_broken) < 0) ret = -1; + ret = 0; + cleanup: virObjectUnref(cfg); return ret; -- 1.8.3.1

2 1

[libvirt] [PATCH] qemu: Fix domain state after reset
by Martin Kletzander 30 Jul '16

30 Jul '16

When reset was called from a domain that crashed we didn't change the crashed state into a paused one which could confuse users. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1269575 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- src/qemu/qemu_driver.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index cda85f65fc98..866580705493 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -2139,6 +2139,7 @@ qemuDomainReset(virDomainPtr dom, unsigned int flags) virDomainObjPtr vm; int ret = -1; qemuDomainObjPrivatePtr priv; + virDomainState state; virCheckFlags(0, -1); @@ -2165,6 +2166,10 @@ qemuDomainReset(virDomainPtr dom, unsigned int flags) priv->fakeReboot = false; + state = virDomainObjGetState(vm, NULL); + if (state == VIR_DOMAIN_CRASHED) + virDomainObjSetState(vm, VIR_DOMAIN_PAUSED, VIR_DOMAIN_PAUSED_CRASHED); + endjob: qemuDomainObjEndJob(driver, vm); -- 2.9.0

2 1

[libvirt] [PATCH 0/5] qemu: report actual vcpu state in virVcpuInfo
by Viktor Mihajlovski 29 Jul '16

29 Jul '16

Currently, the virVcpuInfo returned by virDomainGetVcpus() will always report a state of VIR_VCPU_RUNNING for each defined domain vcpu even if the vcpu is currently in the halted state. As the monitor interface is in fact reporting the accurate state, it is rather easy to transport this information with the existing API. This is done by - adding a new state of VIR_VCPU_HALTED - adding a monitor function to pass back the halted state for the vcpus - adding a new field to the private domain vcpu object reflecting the halted state for the vcpu - modifying the driver code to report the vcpu state based on the halted indicator - extending virsh vcpuinfo to also display the halted state The vcpu state is however not recorded in the internal XML format, since the state can change asynchronously (without notification). Viktor Mihajlovski (5): domain: Add new VCPU state "halted" qemu: Add monitor APIs for the detection of halted VCPUs tests: Add testcase for qemuMonitorJSONGetCPUState qemu: Add domain support for VCPU halted state qemu: Ensure reported VCPU state is current in driver API include/libvirt/libvirt-domain.h | 1 + src/qemu/qemu_domain.c | 82 ++++++++++++++++++++++++++++++ src/qemu/qemu_domain.h | 4 ++ src/qemu/qemu_driver.c | 23 ++++++++- src/qemu/qemu_monitor.c | 22 ++++++++ src/qemu/qemu_monitor.h | 2 + src/qemu/qemu_monitor_json.c | 52 +++++++++++++++++-- src/qemu/qemu_monitor_json.h | 2 + src/qemu/qemu_monitor_text.c | 54 ++++++++++++++++---- src/qemu/qemu_monitor_text.h | 2 + tests/qemumonitorjsontest.c | 106 ++++++++++++++++++++++++--------------- tools/virsh-domain.c | 3 +- 12 files changed, 295 insertions(+), 58 deletions(-) -- 1.9.1

2 11

[libvirt] [PATCH 00/13] Move functions from qemu_domain_address.c to domain_addr.c
by Tomasz Flendrich 29 Jul '16

29 Jul '16

This series depends on another one: https://www.redhat.com/archives/libvir-list/2016-July/msg00696.html Because the address sets are no longer cached, it's possible to move functions that don't depend on qemu anymore from qemu_domain_address.c to domain_addr.c. I did that with virtioSerial and PCI. To make some functions not dependent on qemuCaps, their parameters were changed to booleans. These changes are in [PATCH 02/13]. If you are satisfied with this approach, I can do the same with the rest of the functions or just some of them. Tomasz Flendrich (13): Move and rename qemuDomainAssignVirtioSerialAddresses make pci address handling functions qemu-agnostic Move and rename qemuDomainCollectPCIAddress Move and rename qemuDomainMachineIsQ35 et al Move and rename qemuDomainValidateDevicePCISlotsPIIX3 Move and rename qemuDomainValidateDevicePCISlotsQ35 Move and rename qemuDomainValidateDevicePCISlotsChipsets Move and rename qemuDomainAddressFindNewBusNr et al Move and rename qemuDomainMachineIsVirt Move and rename qemuDomainSupportsPCI Move and rename qemuDomainPCIAddrSetCreateFromDomain Move and rename qemuDomainAssignPCIAddresses et al Make functions in qemu_domain_address static src/conf/domain_addr.c | 1315 ++++++++++++++++++++++++++++++++++++++++ src/conf/domain_addr.h | 28 + src/libvirt_private.syms | 6 + src/qemu/qemu_alias.c | 2 +- src/qemu/qemu_capabilities.c | 6 +- src/qemu/qemu_command.c | 20 +- src/qemu/qemu_domain.c | 55 +- src/qemu/qemu_domain.h | 3 - src/qemu/qemu_domain_address.c | 1284 +-------------------------------------- src/qemu/qemu_domain_address.h | 4 - src/qemu/qemu_hotplug.c | 33 +- 11 files changed, 1406 insertions(+), 1350 deletions(-) -- 2.7.4 (Apple Git-66)

3 15

[libvirt] [gconfig v2 0/2] Add host setters
by Visarion Alexandru 29 Jul '16

29 Jul '16

From: Visarion Alexandru <viorel.visarion(a)gmail.com> These new patches are a bit different than the last ones. I first delete the <listen address=""> child node and then I modify the <listen> attribute. Visarion Alexandru (2): config: Add vnc host setter config: Add spice host setter libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.c | 10 ++++++++++ libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.h | 3 +++ libvirt-gconfig/libvirt-gconfig-domain-graphics-vnc.c | 9 +++++++++ libvirt-gconfig/libvirt-gconfig-domain-graphics-vnc.h | 3 +++ libvirt-gconfig/libvirt-gconfig.sym | 2 ++ 5 files changed, 27 insertions(+) -- 2.7.4

2 4

[libvirt] [PATCH] conf: Catch invalid memory model earlier
by Michal Privoznik 29 Jul '16

29 Jul '16

Consider the following XML snippet: <memory model=''> <target> <size unit='KiB'>523264</size> <node>0</node> </target> </memory> Whats wrong you ask? The @model attribute. This should result in an error thrown into users faces during virDomainDefine phase. Except it doesn't. The XML validation catches this error, but if users chose to ignore that, they will end up with invalid XML. Well, they won't be able to start the machine - that's when error is produced currently. But it would be nice if we could catch the error like this earlier. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- src/conf/domain_conf.c | 2 +- src/qemu/qemu_command.c | 2 -- src/qemu/qemu_domain.c | 2 -- 3 files changed, 1 insertion(+), 5 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index c1ddb5a..a56e0f5 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -13215,7 +13215,7 @@ virDomainMemoryDefParseXML(xmlNodePtr memdevNode, goto error; } - if ((def->model = virDomainMemoryModelTypeFromString(tmp)) < 0) { + if ((def->model = virDomainMemoryModelTypeFromString(tmp)) <= 0) { virReportError(VIR_ERR_XML_ERROR, _("invalid memory model '%s'"), tmp); goto error; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index c4da8b5..5325f48 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -3397,8 +3397,6 @@ qemuBuildMemoryDeviceStr(virDomainMemoryDefPtr mem) case VIR_DOMAIN_MEMORY_MODEL_NONE: case VIR_DOMAIN_MEMORY_MODEL_LAST: - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("invalid memory device type")); break; } diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index b3ca993..9b439df 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -5183,8 +5183,6 @@ qemuDomainDefValidateMemoryHotplugDevice(const virDomainMemoryDef *mem, case VIR_DOMAIN_MEMORY_MODEL_NONE: case VIR_DOMAIN_MEMORY_MODEL_LAST: - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("invalid memory device type")); return -1; } -- 2.8.4

2 1

[libvirt] [PATCH python] allow pkg-config binary to be set by env
by Cole Robinson 28 Jul '16

28 Jul '16

From: Markus Rothe <markusr815(a)gmail.com> https://bugzilla.redhat.com/show_bug.cgi?id=1350523 --- setup.py | 2 ++ 1 file changed, 2 insertions(+) diff --git a/setup.py b/setup.py index 099b1e0..a4cfb88 100755 --- a/setup.py +++ b/setup.py @@ -28,6 +28,8 @@ _pkgcfg = -1 def get_pkgcfg(do_fail=True): global _pkgcfg if _pkgcfg == -1: + _pkgcfg = os.getenv('PKG_CONFIG') + if _pkgcfg is None: _pkgcfg = distutils.spawn.find_executable("pkg-config") if _pkgcfg is None and do_fail: raise Exception("pkg-config binary is required to compile libvirt-python") -- 2.7.4

2 3

[libvirt] [gconfig v2] config: Add spice host setter
by Visarion Alexandru 28 Jul '16

28 Jul '16

From: Visarion Alexandru <viorel.visarion(a)gmail.com> Learn to set the address that spice is listening on. Visarion Alexandru (1): config: Add spice host setter libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.c | 10 ++++++++++ libvirt-gconfig/libvirt-gconfig-domain-graphics-spice.h | 3 +++ libvirt-gconfig/libvirt-gconfig.sym | 1 + 3 files changed, 14 insertions(+) -- 2.7.4

1 0

[libvirt] [PATCH] qemu: Need to free fileprops in error path
by John Ferlan 28 Jul '16

28 Jul '16

The virJSONValueObjectCreate only consumes the object on success, so on failure we must free - from commit id 'f4441017' (found by Coverity). Signed-off-by: John Ferlan <jferlan(a)redhat.com> --- Coverity is also grumpy about the consumption of the object on the success path - probably because of the va_args manner in which the object gets consumed and later free'd. src/qemu/qemu_command.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 0094665..c4da8b5 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -1038,8 +1038,10 @@ qemuGetDriveSourceProps(virStorageSourcePtr src, } if (fileprops && - virJSONValueObjectCreate(props, "a:file", fileprops, NULL) < 0) + virJSONValueObjectCreate(props, "a:file", fileprops, NULL) < 0) { + virJSONValueFree(fileprops); return -1; + } return 0; } -- 2.5.5

2 1

[libvirt] [PATCH glib] Fix check for libvirt optional functions
by Daniel P. Berrange 28 Jul '16

28 Jul '16

The checks for virDomainOpenGraphicsFD and virNetworkGetDHCPLeases in configure are simply trying to compile and link a test program. This was assuming that libvirt was installed in /usr/lib & /usr/include, which is not always the case. Using a pkg-config version check avoids needing to compile and thus avoids the bad assumptions about install locations. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- Pushed as a broken build fix configure.ac | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/configure.ac b/configure.ac index f71e5cf..b2848f2 100644 --- a/configure.ac +++ b/configure.ac @@ -92,13 +92,19 @@ LIBVIRT_GLIB_COMPILE_WARNINGS PKG_CHECK_MODULES(LIBVIRT, libvirt >= $LIBVIRT_REQUIRED) # virDomainOpenGraphicsFD was introduced in libvirt 1.2.8 -AC_CHECK_LIB([virt], - [virDomainOpenGraphicsFD], - [AC_DEFINE([HAVE_VIR_DOMAIN_OPEN_GRAPHICS_FD], 1, [Have virDomainOpenGraphicsFD?])]) +AC_MSG_CHECKING([for virDomainOpenGraphicsFD]) +PKG_CHECK_EXISTS(libvirt >= 21.2.8, + [ + AC_DEFINE([HAVE_VIR_DOMAIN_OPEN_GRAPHICS_FD], 1, [Have virDomainOpenGraphicsFD?]) + AC_MSG_RESULT([yes]) + ],[AC_MSG_RESULT([no])]) # virNetworkGetDHCPLeases was introduced in libvirt 1.2.6 -AC_CHECK_LIB([virt], - [virNetworkGetDHCPLeases], - [AC_DEFINE([HAVE_VIR_NETWORK_GET_DHCP_LEASES], 1, [Have virNetworkGetDHCPLeases?])]) +AC_MSG_CHECKING([for virNetworkGetDHCPLeases]) +PKG_CHECK_EXISTS(libvirt >= 21.2.6, + [ + AC_DEFINE([HAVE_VIR_NETWORK_GET_DHCP_LEASES], 1, [Have virNetworkGetDHCPLeases?]) + AC_MSG_RESULT([yes]) + ],[AC_MSG_RESULT([no])]) enable_tests=no PKG_CHECK_MODULES(GLIB2, glib-2.0 >= $GLIB2_TEST_REQUIRED, [enable_tests=yes], -- 2.7.4

1 0

[libvirt] [PATCH glib] Avoid unused variable in DHCP lease finalize method
by Daniel P. Berrange 28 Jul '16

28 Jul '16

If the network DHCP lease support is not detected, the build will generate a warning / error: libvirt-gobject-network-dhcp-lease.c: In function 'gvir_network_dhcp_lease_finalize': libvirt-gobject-network-dhcp-lease.c:99:27: error: unused variable 'lease' [-Werror=unused-variable] GVirNetworkDHCPLease *lease = GVIR_NETWORK_DHCP_LEASE(object); ^~~~~ cc1: all warnings being treated as errors Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- Pushed as a broken build fix libvirt-gobject/libvirt-gobject-network-dhcp-lease.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/libvirt-gobject/libvirt-gobject-network-dhcp-lease.c b/libvirt-gobject/libvirt-gobject-network-dhcp-lease.c index 7d1c979..200778d 100644 --- a/libvirt-gobject/libvirt-gobject-network-dhcp-lease.c +++ b/libvirt-gobject/libvirt-gobject-network-dhcp-lease.c @@ -96,12 +96,10 @@ static void gvir_network_dhcp_lease_set_property(GObject *object, static void gvir_network_dhcp_lease_finalize(GObject *object) { - GVirNetworkDHCPLease *lease = GVIR_NETWORK_DHCP_LEASE(object); #ifdef HAVE_VIR_NETWORK_GET_DHCP_LEASES + GVirNetworkDHCPLease *lease = GVIR_NETWORK_DHCP_LEASE(object); GVirNetworkDHCPLeasePrivate *priv = lease->priv; -#endif /* HAVE_VIR_NETWORK_GET_DHCP_LEASES */ -#ifdef HAVE_VIR_NETWORK_GET_DHCP_LEASES virNetworkDHCPLeaseFree(priv->handle); #endif /* HAVE_VIR_NETWORK_GET_DHCP_LEASES */ -- 2.7.4

1 0

[libvirt] [PATCH] wireshark: Adapt to dissector function header change
by Michal Privoznik 28 Jul '16

28 Jul '16

In wireshark commit bbdd89b9 (contained in 2.1.0 release) they have changed prototype of dissector function. Now it returns number of bytes consumed by the dissector, and can get a pointer to user specified data (which we don't use). Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- tools/wireshark/src/packet-libvirt.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/tools/wireshark/src/packet-libvirt.c b/tools/wireshark/src/packet-libvirt.c index aa1c323..569a0a1 100644 --- a/tools/wireshark/src/packet-libvirt.c +++ b/tools/wireshark/src/packet-libvirt.c @@ -442,8 +442,14 @@ get_message_len(packet_info *pinfo ATTRIBUTE_UNUSED, tvbuff_t *tvb, int offset) return tvb_get_ntohl(tvb, offset); } +#if WIRESHARK_VERSION >= 2000001 +static int +dissect_libvirt(tvbuff_t *tvb, packet_info *pinfo, + proto_tree *tree, void *data ATTRIBUTE_UNUSED) +#else static void dissect_libvirt(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree) +#endif { /* Another magic const - 4; simply, how much bytes * is needed to tell the length of libvirt packet. */ @@ -454,6 +460,10 @@ dissect_libvirt(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree) tcp_dissect_pdus(tvb, pinfo, tree, TRUE, 4, get_message_len, dissect_libvirt_message, NULL); #endif + +#if WIRESHARK_VERSION >= 2000001 + return tvb_captured_length(tvb); +#endif } void -- 2.8.4

2 1

[libvirt] [PATCH 0/2] tools: Fix usage of incorrect environment variables
by Erik Skultety 28 Jul '16

28 Jul '16

According to https://bugzilla.redhat.com/show_bug.cgi?id=1357363, most of virt-admin's environment variables are not taken into effect when set. As it turned out, virsh suffers from the very same problem which resides in the generic vsh.c module. Erik Skultety (2): vsh: Make vshInitDebug return int instead of void tools: Make use of the correct environment variables tools/virsh.c | 1 + tools/virt-admin.c | 1 + tools/vsh.c | 34 +++++++++++++++++++--------------- tools/vsh.h | 1 + 4 files changed, 22 insertions(+), 15 deletions(-) -- 2.5.5

2 4

[libvirt] logging: Libvirtd could not create its own logfile, when the path /var/log/libvirt is not exist.
by weifuqiang 28 Jul '16

28 Jul '16

Hi all: Libvirtd could not create its own logfile, such as /var/log/libvirt/libvirtd.log, when the path /var/log/libvirt is not exist. It may be a problem,I think,so try to make patch when open file failed in virLogAddOutputToFile. or any better idea? ------------ Libvirtd just creates its logfile, such as /var/log/libvirt/libvirtd.log, DURING INSTALLATION. If the logfile(we mean its path here,that's /var/log/libvirt/) is deleted during its runtime, even if we restart libvirtd, such file would not be generated automatically, thus, the logs are all missing after then, it becomes impossible to debug problems after then as well. So, we try to mkdir the deleted path after restarting libvirtd. Or, any better idea? Thanks in advance. Wei Fuqiang

2 1

[libvirt] [PATCH 0/2] Couple of build fixes
by Michal Privoznik 28 Jul '16

28 Jul '16

Contradictory to my previous patch set [1], these are not pushed yet and thus require proper review. 1: https://www.redhat.com/archives/libvir-list/2016-July/msg01147.html Michal Privoznik (2): vshCmddefGetOption: Change type of opt_index vshReadlineParse: Drop some unused variables tools/vsh.c | 13 +++++-------- 1 file changed, 5 insertions(+), 8 deletions(-) -- 2.8.4

2 3

[libvirt] [PATCH] Fix RPM spec for wireshark on Fedora < 24
by Daniel P. Berrange 28 Jul '16

28 Jul '16

This previous commit commit cd9fcc8be7dcb9126d70f744ce54b0b742eeefb8 Author: Michal Privoznik <mprivozn(a)redhat.com> Date: Wed Jul 27 16:58:32 2016 +0200 libvirt.spec.in: Adapt to newest wireshark plugindir Adapted the libvirt spec for wireshark >= 2.1.0 but this ignored the fact that we enable wireshark from Fedora 21 and 2.1.0 was only added in Fedora 24 Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- libvirt.spec.in | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/libvirt.spec.in b/libvirt.spec.in index 35c7449..5841f45 100644 --- a/libvirt.spec.in +++ b/libvirt.spec.in @@ -406,7 +406,11 @@ BuildRequires: numad %endif %if %{with_wireshark} +%if 0%{fedora} >= 24 BuildRequires: wireshark-devel >= 2.1.0 +%else +BuildRequires: wireshark-devel >= 1.12.1 +%endif %endif Provides: bundled(gnulib) @@ -1212,7 +1216,13 @@ rm -f $RPM_BUILD_ROOT%{_libdir}/libvirt/lock-driver/*.a rm -f $RPM_BUILD_ROOT%{_libdir}/libvirt/connection-driver/*.la rm -f $RPM_BUILD_ROOT%{_libdir}/libvirt/connection-driver/*.a %if %{with_wireshark} +%if 0%{fedora} >= 24 rm -f $RPM_BUILD_ROOT%{_libdir}/wireshark/plugins/libvirt.la +%else +rm -f $RPM_BUILD_ROOT%{_libdir}/wireshark/plugins/*/libvirt.la +mv $RPM_BUILD_ROOT%{_libdir}/wireshark/plugins/*/libvirt.so \ + $RPM_BUILD_ROOT%{_libdir}/wireshark/plugins/libvirt.so +%endif %endif install -d -m 0755 $RPM_BUILD_ROOT%{_datadir}/lib/libvirt/dnsmasq/ -- 2.5.5

1 0

[libvirt] [PATCH 0/3] libxl hooks
by Cédric Bosdonnat 28 Jul '16

28 Jul '16

Hi there! Here is small patchset adding hooks support to the libxl driver. Cédric Bosdonnat (3): libxl: add a flag to mark guests as tainted by a hook libxl: fix segfault in libxlReconnectDomain libxl: add hooks support docs/hooks.html.in | 53 ++++++++++++++++++++++++++-- src/libxl/libxl_domain.c | 85 +++++++++++++++++++++++++++++++++++++++++++++ src/libxl/libxl_domain.h | 2 ++ src/libxl/libxl_driver.c | 52 ++++++++++++++++++++------- src/libxl/libxl_migration.c | 58 +++++++++++++++++++++++++++++++ src/util/virhook.c | 16 ++++++++- src/util/virhook.h | 13 +++++++ 7 files changed, 264 insertions(+), 15 deletions(-) -- 2.6.6

3 7

[libvirt] [PATCH] qemu: remove panic device with model s390 when migrating
by Boris Fiuczynski 28 Jul '16

28 Jul '16

The panic device with model s390 is autogenerated on domains with S390 architecture since libvirt version 1.3.5. For backwards compatibility reasons the device is to be removed when migrating. Signed-off-by: Boris Fiuczynski <fiuczy(a)linux.vnet.ibm.com> --- src/qemu/qemu_domain.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index ceac22f..0d97ea6 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c @@ -3310,6 +3310,28 @@ qemuDomainDefFormatBuf(virQEMUDriverPtr driver, virDomainControllerDefFree(usb); } + /* Remove the panic device with model s390 if present */ + for (i = 0; i < def->npanics; i++) { + if (def->panics[i]->model == VIR_DOMAIN_PANIC_MODEL_S390) { + virDomainPanicDefPtr *panics = def->panics; + int npanics = def->npanics; + size_t j; + + if (VIR_ALLOC_N(def->panics, npanics - 1) < 0) { + def->panics = panics; + goto cleanup; + } + + def->npanics = 0; + for (j = 0; j < npanics; j++) { + if (j != i) + def->panics[def->npanics++] = panics[j]; + } + VIR_FREE(panics); + break; /* only one panic dev with model s390 can exist */ + } + } + for (i = 0; i < def->nchannels; i++) qemuDomainChrDefDropDefaultPath(def->channels[i], driver); } -- 2.9.0

3 3

[libvirt] [libvirt RFC PATCH 00/10] Add support for qemu's JSON pseudo'protocol for backing store
by Peter Krempa 28 Jul '16

28 Jul '16

Libvirt didn't handle this for a long time and VMs with such config would not start we should implement it. Using JSON is basically the only option to specify advanced configuration for a backing file. Field names were taken from qemu's source since there isn't really documentation for this. CCing qemu-block for possible comments on patches 2-8,10. Peter Krempa (10): tests: Add testing of backing store string parser util: storage: Add parser for qemu's "json" backing pseudo-protocol util: storage: Add support for host device backing specified via JSON util: storage: Add support for URI based backing volumes in qemu's JSON pseudo-protocol util: storage: Add json pseudo protocol support for gluster volumes util: storage: Add json pseudo protocol support for iSCSI volumes Add JSON backing volume parser for 'nbd' protocol util: storage: Add JSON backing store parser for 'sheepdog' protocol util: storage: Add 'ssh' network storage protocol util: storage: Add JSON backing volume parser for 'ssh' protocol src/libvirt_private.syms | 1 + src/libxl/libxl_conf.c | 1 + src/qemu/qemu_command.c | 7 + src/qemu/qemu_driver.c | 3 + src/qemu/qemu_parse_command.c | 1 + src/util/virstoragefile.c | 288 ++++++++++++++++++++++++++++++++++++++++-- src/util/virstoragefile.h | 4 + src/xenconfig/xen_xl.c | 1 + tests/virstoragetest.c | 128 +++++++++++++++++++ 9 files changed, 425 insertions(+), 9 deletions(-) -- 2.8.2

3 18

[libvirt] [PATCH 0/2] Two wireshark fixes
by Michal Privoznik 27 Jul '16

27 Jul '16

While trying to do an RPM, I've ran into two problems. I've already pushed the fixes under 'build breaker rule'. I'm sending them for completeness. Michal Privoznik (2): virt-wireshark: Properly substract wireshark prefix libvirt.spec.in: Adapt to newest wireshark plugindir libvirt.spec.in | 6 ++---- m4/virt-wireshark.m4 | 2 +- 2 files changed, 3 insertions(+), 5 deletions(-) -- 2.8.4

2 3

[libvirt] [PATCH 0/2] Updates for LUKS support
by Daniel P. Berrange 27 Jul '16

27 Jul '16

Daniel P. Berrange (2): virstoragefile: refactor virStorageFileMatchesNNN methods storage: remove "luks" storage volume type src/qemu/qemu_command.c | 10 +- src/qemu/qemu_domain.c | 2 +- src/qemu/qemu_hotplug.c | 2 +- src/storage/storage_backend.c | 41 ++-- src/storage/storage_backend_fs.c | 17 +- src/storage/storage_backend_gluster.c | 5 - src/util/virstoragefile.c | 265 +++++++++++++++------ src/util/virstoragefile.h | 1 - tests/qemuxml2argvdata/qemuxml2argv-luks-disks.xml | 4 +- tests/storagevolxml2xmlin/vol-luks-cipher.xml | 2 +- tests/storagevolxml2xmlin/vol-luks.xml | 2 +- tests/storagevolxml2xmlout/vol-luks-cipher.xml | 2 +- tests/storagevolxml2xmlout/vol-luks.xml | 2 +- 13 files changed, 235 insertions(+), 120 deletions(-) -- 2.7.4

2 9

[libvirt] [PATCH libvirt-glib] docs: Document gvir_connection_get_{storage_pools, networks, domains}
by Guido Günther 27 Jul '16

27 Jul '16

In contrast to libvirt itself all get_* methods need to prefetch the corresponding information first so document this. --- libvirt-gobject/libvirt-gobject-connection.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/libvirt-gobject/libvirt-gobject-connection.c b/libvirt-gobject/libvirt-gobject-connection.c index 00d5eda..3f17265 100644 --- a/libvirt-gobject/libvirt-gobject-connection.c +++ b/libvirt-gobject/libvirt-gobject-connection.c @@ -700,6 +700,11 @@ void gvir_connection_close(GVirConnection *conn) * gvir_connection_fetch_domains: * @conn: a #GVirConnection * @cancellable: (allow-none)(transfer none): cancellation object + * + * Use this method to fetch all domains managed by connection + * @conn. Use e.g. #gvir_connection_find_domain_by_id or + * #gvir_connection_get_domain afterwards to query the fetched + * domains. */ gboolean gvir_connection_fetch_domains(GVirConnection *conn, GCancellable *cancellable, @@ -783,6 +788,11 @@ cleanup: * gvir_connection_fetch_storage_pools: * @conn: a #GVirConnection * @cancellable: (allow-none)(transfer none): cancellation object + * + * Use this method to fetch all storage pools managed by connection + * @conn. Use e.g. #gvir_connection_find_storage_pool_by_name or + * #gvir_connection_get_storage_pools afterwards to query the fetched + * pools. */ gboolean gvir_connection_fetch_storage_pools(GVirConnection *conn, GCancellable *cancellable, @@ -1722,6 +1732,11 @@ GVirInterface *gvir_connection_find_interface_by_mac(GVirConnection *conn, * gvir_connection_fetch_networks: * @conn: a #GVirConnection * @cancellable: (allow-none)(transfer none): cancellation object + * + * Use this method to fetch all networks managed by connection + * @conn. Use e.g. #gvir_connection_find_network_by_name or + * #gvir_connection_get_networks afterwards to query the fetched + * domains. */ gboolean gvir_connection_fetch_networks(GVirConnection *conn, GCancellable *cancellable, -- 2.8.1

2 2

[libvirt] Entering freeze for libvirt-2.1.0
by Daniel Veillard 27 Jul '16

27 Jul '16

So as planned I tagged the Release Candidate 1 version in git and pushed signed tarball and rpms to the usual place: ftp://libvirt.org/libvirt/ Please give it a try, I plan to push the rc2 on Friday and if everything goes fine have the final release on Monday or Tuesday. Please give it a good test :) thanks ! Daniel -- Daniel Veillard | Open Source and Standards, Red Hat veillard(a)redhat.com | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt] [PATCH] virconf: Fix config file path construction
by Erik Skultety 27 Jul '16

27 Jul '16

Since commit c4bdff19, the path to the configuration file has been constructed in the following manner: - if no config filename was passed to virConfLoadConfigPath, libvirt.conf was used as default - otherwise the filename was concatenated with "<config_dir>/libvirt/libvirt%s%s.conf" which in admin case resulted in "libvirt-libvirt-admin.conf.conf". Obviously, this non-existent config led to ignoring all user settings in libvirt-admin.conf. This patch requires the config filename to be always provided as an argument with the concatenation being simplified. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1357364 Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- src/libvirt.c | 2 +- src/util/virconf.c | 12 ++++-------- 2 files changed, 5 insertions(+), 9 deletions(-) diff --git a/src/libvirt.c b/src/libvirt.c index 68c8317..52462e3 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -969,7 +969,7 @@ virConnectOpenInternal(const char *name, if (ret == NULL) return NULL; - if (virConfLoadConfig(&conf, NULL) < 0) + if (virConfLoadConfig(&conf, "libvirt.conf") < 0) goto failed; if (name && name[0] == '\0') diff --git a/src/util/virconf.c b/src/util/virconf.c index ee54072..3e49f41 100644 --- a/src/util/virconf.c +++ b/src/util/virconf.c @@ -1566,20 +1566,16 @@ virConfLoadConfigPath(const char *name) { char *path; if (geteuid() == 0) { - if (virAsprintf(&path, "%s/libvirt/libvirt%s%s.conf", - SYSCONFDIR, - name ? "-" : "", - name ? name : "") < 0) + if (virAsprintf(&path, "%s/libvirt/%s", + SYSCONFDIR, name) < 0) return NULL; } else { char *userdir = virGetUserConfigDirectory(); if (!userdir) return NULL; - if (virAsprintf(&path, "%s/libvirt%s%s.conf", - userdir, - name ? "-" : "", - name ? name : "") < 0) { + if (virAsprintf(&path, "%s/%s", + userdir, name) < 0) { VIR_FREE(userdir); return NULL; } -- 2.5.5

3 3

[libvirt] [PATCH] admin: Fix default uri config option name s/admin_uri_default/uri_default
by Erik Skultety 27 Jul '16

27 Jul '16

The original name 'admin_uri_default' was introduced to our code by commit dbecb87f. However, at that time we already had a separate config file for admin library but the commit mentioned above didn't properly adjust the config's option name. The result is that when we're loading the config, we check a non-existent config option (there's not much to do with the URIs anyway, since we only allow local connection). Additionally, virt-admin's man page documents, that the default URI can be altered by setting admin_uri_default option. So the fix proposed by this patch leaves the libvirt-admin.conf as is and adjusts the naming in the code as well as in the virt-admin's man page. Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- src/libvirt-admin.c | 2 +- tools/virt-admin.pod | 7 ++++--- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/src/libvirt-admin.c b/src/libvirt-admin.c index 859f937..4bf29b1 100644 --- a/src/libvirt-admin.c +++ b/src/libvirt-admin.c @@ -167,7 +167,7 @@ virAdmGetDefaultURI(virConfPtr conf, char **uristr) return -1; VIR_DEBUG("Using LIBVIRT_ADMIN_DEFAULT_URI '%s'", *uristr); } else { - if (virConfGetValueString(conf, "admin_uri_default", uristr) < 0) + if (virConfGetValueString(conf, "uri_default", uristr) < 0) return -1; if (*uristr) { diff --git a/tools/virt-admin.pod b/tools/virt-admin.pod index 2765bc5..2199038 100644 --- a/tools/virt-admin.pod +++ b/tools/virt-admin.pod @@ -127,7 +127,7 @@ Will print the current directory. (Re)-Connect to a daemon's administrating server. The I<URI> parameter specifies how to connect to the administrating server. -If I<LIBVIRT_ADMIN_DEFAULT_URI> or I<admin_uri_default> (see below) were set, +If I<LIBVIRT_ADMIN_DEFAULT_URI> or I<uri_default> (see below) were set, I<connect> is automatically issued every time a command that requires an active connection is executed. Note that this only applies if there is no connection at all or there is an inactive one. @@ -137,8 +137,9 @@ To find the currently used URI, check the I<uri> command documented below. =item B<uri> Prints the administrating server canonical URI, can be useful in shell mode. If -no I<uri> was specified, neither I<LIBVIRT_ADMIN_DEFAULT_URI> or -I<admin_uri_default> were set, libvirtd:///system is used. +no I<uri> was specified, neither I<LIBVIRT_ADMIN_DEFAULT_URI> environment +variable nor I<uri_default> option (libvirt-admin.conf) were set, +libvirtd:///system is used. =back -- 2.5.5

2 1

[libvirt] [PATCH v7 0/7] qemu: expand domain memory statistics
by Derbyshev Dmitriy 27 Jul '16

27 Jul '16

From: Derbyshev Dmitry <dderbyshev(a)virtuozzo.com> QEMU reports timestamp and available along with other memory statistics. This information was not saved into domain statistics. 1st to 3rd patches add new stats to reports. Also, to collect all balloon statistics for all guests it was necessary to make several libvirt requests (one per VE).. 4th and 5th patch allows doing this via qemuConnectGetAllDomainStats in one request. 6th patch speed-ups DomainMemoryStats by caching so that requests would not gain much additional overhead from previous patches. 7th patch adds probbing and can be dropped. Changes since v1: * Enum numeration fixed * Macro getting "usage" field fixed Changes since v2: * Previous patches were on wrong branch * qemu's stat name was "stat-available-memory" Changes since v3: * domstats patch added Changes since v4: * Formatted and rephrased commit messages * Fixed libvirt crash, caused by simultaneous incorrect QUERY job execution Changes since v5: * Updated virsh.pod * Splitted patch about domstats into 2 patches * Do not report balloon.current as balloon.actual Changes since v6: * caching added for DomainMemoryStats Derbyshev Dmitry (5): virsh: Add balloon stats description to .pod qemu: expand domain memory statistics with 'usable' qemu: expand domain memory statistics with 'last-update' timestamp qemu: split qemuDomainMemoryStats into internal and external functions qemu: return balloon statistics when all domain statistics reported Igor Redko (2): qemu: add cache for DomainMemoryStats qemu-probes: add probes for cache MemoryStats cache include/libvirt/libvirt-domain.h | 11 ++- src/libvirt-domain.c | 5 ++ src/libvirt_qemu_probes.d | 4 + src/qemu/qemu_domain.h | 6 ++ src/qemu/qemu_driver.c | 187 +++++++++++++++++++++++++++++++++++---- src/qemu/qemu_monitor_json.c | 24 +++-- tools/virsh-domain-monitor.c | 4 + tools/virsh.pod | 29 +++++- 8 files changed, 242 insertions(+), 28 deletions(-) -- 1.9.5.msysgit.0

4 18

[libvirt] [PATCH] virsh: use virConnectGetDomainCapabilities with maxvcpus
by Shivaprasad G Bhat 27 Jul '16

27 Jul '16

The following patch fixes the maxvcpus output on PPC64. Earlier I tried fixing this in kvmGetMaxVcpus() which was later decided to be document not use the virConnectGetMaxVcpus() instead use the virConnectGetDomainCapabilities api. I have not implemented the suggestedcpus as mentioned in my previous patches in this yet. Previous series fixing the virConnectGetDomainCapabilities can be found here. https://www.redhat.com/archives/libvir-list/2016-June/msg01873.html --- Shivaprasad G Bhat (1): virsh: use virConnectGetDomainCapabilities with maxvcpus tools/virsh-host.c | 33 +++++++++++++++++++++++++++++---- 1 file changed, 29 insertions(+), 4 deletions(-) -- Signature

2 2

[libvirt] [PATCH sandbox] image: Rename 'create' and 'delete' to 'prepare' and 'purge'
by Daniel P. Berrange 27 Jul '16

27 Jul '16

Currently we have three virt-sandbox-image commands - 'create' - downloads a template and creates qcow2 - 'delete' - deletes template qcow2 files - 'run' - runs an instance of a template The 'run' command is generating a transient guest which disappears when it stops. We want to have the ability to create / delete persistent guests later, for which the command names "create" and "delete" are a natural fit. So to avoid clash, rename the existing "create" command to "prepare" and the "delete" command to "purge" --- libvirt-sandbox/image/cli.py | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/libvirt-sandbox/image/cli.py b/libvirt-sandbox/image/cli.py index 66854e4..3a5ccfa 100644 --- a/libvirt-sandbox/image/cli.py +++ b/libvirt-sandbox/image/cli.py @@ -68,13 +68,13 @@ def get_template_dir(args): tmpl = template.Template.from_uri(args.template) return "%s/%s" % (args.template_dir, tmpl.source) -def delete(args): +def purge(args): tmpl = template.Template.from_uri(args.template) source = tmpl.get_source_impl() source.delete_template(template=tmpl, templatedir=get_template_dir(args)) -def create(args): +def prepare(args): tmpl = template.Template.from_uri(args.template) source = tmpl.get_source_impl() source.create_template(template=tmpl, @@ -91,7 +91,7 @@ def run(args): # Create the template image if needed if not source.has_template(tmpl, template_dir): - create(args) + prepare(args) name = args.name if name is None: @@ -197,26 +197,26 @@ Example supported URI formats: """) return parser -def gen_delete_args(subparser): - parser = gen_command_parser(subparser, "delete", - _("Delete template data")) +def gen_purge_args(subparser): + parser = gen_command_parser(subparser, "purge", + _("Purge cached template")) requires_debug(parser) requires_template(parser) requires_template_dir(parser) - parser.set_defaults(func=delete) + parser.set_defaults(func=purge) -def gen_create_args(subparser): - parser = gen_command_parser(subparser, "create", - _("Create image from template data")) +def gen_prepare_args(subparser): + parser = gen_command_parser(subparser, "prepare", + _("Prepare local template")) requires_debug(parser) requires_template(parser) requires_connect(parser) requires_template_dir(parser) - parser.set_defaults(func=create) + parser.set_defaults(func=prepare) def gen_run_args(subparser): parser = gen_command_parser(subparser, "run", - _("Run an already built image")) + _("Run an instance of a template")) requires_debug(parser) requires_name(parser) requires_template(parser) @@ -248,8 +248,8 @@ def main(): parser = argparse.ArgumentParser(description="Sandbox Container Image Tool") subparser = parser.add_subparsers(help=_("commands")) - gen_delete_args(subparser) - gen_create_args(subparser) + gen_purge_args(subparser) + gen_prepare_args(subparser) gen_run_args(subparser) gen_list_args(subparser) -- 2.7.4

2 1

[libvirt] [PATCH] virt-admin.pod: Remove a statement about remote access to the daemon
by Erik Skultety 27 Jul '16

27 Jul '16

There's been a forgotten fragment (copy-paste error probably) in the virt-admin's man page referring the reader to our web page on how to construct URIs in case of remote access, which sort of implies that we support it which we don't at the moment, so better remove that. Signed-off-by: Erik Skultety <eskultet(a)redhat.com> --- Pushed under trivial rule. Erik tools/virt-admin.pod | 3 --- 1 file changed, 3 deletions(-) diff --git a/tools/virt-admin.pod b/tools/virt-admin.pod index afc4689..2765bc5 100644 --- a/tools/virt-admin.pod +++ b/tools/virt-admin.pod @@ -134,9 +134,6 @@ connection at all or there is an inactive one. To find the currently used URI, check the I<uri> command documented below. -For remote access see the documentation page at -L<http://libvirt.org/uri.html> on how to make URIs. - =item B<uri> Prints the administrating server canonical URI, can be useful in shell mode. If -- 2.5.5

1 0

[libvirt] [PATCH 00/25] Add support for multi-host gluster drives
by Peter Krempa 27 Jul '16

27 Jul '16

This is a updated take based on stuff I had laying around and parts from https://www.redhat.com/archives/libvir-list/2016-July/msg00872.html This addresses the backing store parser, adds and improves bits to the JSON->commandline generator prior to plugging in the gluster support. This series does not yet address block jobs (snapshot/block copy) using multi-host gluster volumes. Peter Krempa (24): tests: qemuxml2xml: Avoid crash when processing an XML that fails to parse tests: Add testing of backing store string parser util: storage: Add parser for qemu's "json" backing pseudo-protocol util: storage: Add support for host device backing specified via JSON util: storage: Add support for URI based backing volumes in qemu's JSON pseudo-protocol util: storage: Add json pseudo protocol support for gluster volumes util: storage: Add json pseudo protocol support for iSCSI volumes util: storage: Add JSON backing volume parser for 'nbd' protocol util: storage: Add JSON backing store parser for 'sheepdog' protocol util: storage: Add 'ssh' network storage protocol util: storage: Add JSON backing volume parser for 'ssh' protocol util: json: Make first argument of virJSONValueObjectForeachKeyValue const util: qemu: Add wrapper for JSON -> commandline conversion util: qemu: Add support for user-passed strings in JSON->commandline util: qemu: Allow nested objects in JSON -> commandline generator util: qemu: Allow for different approaches to format JSON arrays util: qemu: Don't generate any extra commas in virQEMUBuildCommandLineJSON qemu: command: Rename qemuBuildNetworkDriveURI to qemuBuildNetworkDriveStr qemu: command: Split out network disk URI building qemu: command: Extract drive source command line formatter qemu: command: Refactor code extracted to qemuBuildDriveSourceStr storage: gluster: Support multiple hosts in backend functions util: qemu: Add support for numbered array members qemu: command: Add infrastructure for object specified disk sources Prasanna Kumar Kalever (1): qemu: command: Add support for multi-host gluster disks docs/formatdomain.html.in | 2 +- src/libvirt_private.syms | 5 + src/libxl/libxl_conf.c | 1 + src/qemu/qemu_command.c | 428 +++++++++++++++------ src/qemu/qemu_driver.c | 3 + src/qemu/qemu_parse_command.c | 1 + src/storage/storage_backend_gluster.c | 82 ++-- src/util/virjson.c | 2 +- src/util/virjson.h | 2 +- src/util/virqemu.c | 219 +++++++++-- src/util/virqemu.h | 16 + src/util/virstoragefile.c | 375 +++++++++++++++++- src/util/virstoragefile.h | 4 + src/xenconfig/xen_xl.c | 1 + tests/qemucommandutiltest.c | 68 +++- .../qemuxml2argv-disk-drive-network-gluster.args | 9 +- .../qemuxml2argv-disk-drive-network-gluster.xml | 9 + .../qemuxml2xmlout-disk-drive-network-gluster.xml | 10 + tests/qemuxml2xmltest.c | 9 +- tests/virstoragetest.c | 150 ++++++++ 20 files changed, 1169 insertions(+), 227 deletions(-) -- 2.9.0

3 31

[libvirt] [PATCHv2 00/10] JSON pseudo-protocol support for backing chains
by Peter Krempa 27 Jul '16

27 Jul '16

This part is extracted from the gluster series since Eric's comment on the original RFC triggered a rework of certain parts. This version allows to handle both flattened and hierarchical syntaxes (see patch 1/10) for details. Peter Krempa (10): util: storage: Add parser for qemu's json backing pseudo-protocol util: json: Make first argument of virJSONValueCopy const util: storage: Add support for host device backing specified via JSON util: storage: Add support for URI based backing volumes in qemu's JSON pseudo-protocol util: storage: Add json pseudo protocol support for gluster volumes util: storage: Add json pseudo protocol support for iSCSI volumes util: storage: Add JSON backing volume parser for 'nbd' protocol util: storage: Add JSON backing store parser for 'sheepdog' protocol util: storage: Add 'ssh' network storage protocol util: storage: Add JSON backing volume parser for 'ssh' protocol src/libxl/libxl_conf.c | 1 + src/qemu/qemu_command.c | 7 + src/qemu/qemu_driver.c | 3 + src/qemu/qemu_parse_command.c | 1 + src/util/virjson.c | 2 +- src/util/virjson.h | 2 +- src/util/virstoragefile.c | 447 +++++++++++++++++++++++++++++++++++++++++- src/util/virstoragefile.h | 1 + src/xenconfig/xen_xl.c | 1 + tests/virstoragetest.c | 131 +++++++++++++ 10 files changed, 586 insertions(+), 10 deletions(-) -- 2.9.0

2 11

[libvirt] [PATCH] qemu: Adjust the cur_ballon on coldplug/unplug of dimms
by Shivaprasad G Bhat 27 Jul '16

27 Jul '16

The cur_balloon also increases/decreases with dimm hotplug/unplug. To be consistent, adjust the value for coldplug too. This was inconsistently taken care when cur_ballon != memory to begin with. The patch fixes it irrespective of that. Signed-off-by: Shivaprasad G Bhat <sbhat(a)linux.vnet.ibm.com> --- src/conf/domain_conf.c | 3 +-- src/qemu/qemu_driver.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 6dfcf81..8b0b790 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -14674,8 +14674,7 @@ virDomainMemoryRemove(virDomainDefPtr def, VIR_DELETE_ELEMENT(def->mems, idx, def->nmems); /* fix up balloon size */ - if (def->mem.cur_balloon > virDomainDefGetMemoryTotal(def)) - def->mem.cur_balloon = virDomainDefGetMemoryTotal(def); + def->mem.cur_balloon -= ret->size; /* fix total memory size of the domain */ virDomainDefSetMemoryTotal(def, memory - ret->size); diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index cda85f6..3824e1d 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -7769,8 +7769,7 @@ qemuDomainAttachDeviceConfig(virDomainDefPtr vmdef, return -1; } - if (vmdef->mem.cur_balloon == virDomainDefGetMemoryTotal(vmdef)) - vmdef->mem.cur_balloon += dev->data.memory->size; + vmdef->mem.cur_balloon += dev->data.memory->size; if (virDomainMemoryInsert(vmdef, dev->data.memory) < 0) return -1;

2 1

[libvirt] [PATCH 0/2] Use static configuration method for iSCSI target discovery
by John Ferlan 27 Jul '16

27 Jul '16

https://bugzilla.redhat.com/show_bug.cgi?id=1356436 Details in patches. John Ferlan (2): util: Introduce virISCSINodeNew iscsi: Establish connection to target via static target login src/libvirt_private.syms | 1 + src/storage/storage_backend_iscsi.c | 9 ++++---- src/util/viriscsi.c | 45 +++++++++++++++++++++++++++++++++++++ src/util/viriscsi.h | 6 +++++ 4 files changed, 56 insertions(+), 5 deletions(-) -- 2.5.5

2 8

[libvirt] [PATCH] docs: Add Kimchi as Web Application
by Ramon Medeiros 27 Jul '16

27 Jul '16

Kimchi is a open-source interface to kvm. It runs with HTML5, simple and easy to manage kvm guests. Signed-off-by: Ramon Medeiros <ramonn(a)linux.vnet.ibm.com> --- docs/apps.html.in | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/docs/apps.html.in b/docs/apps.html.in index b8337b9..1a6fd3f 100644 --- a/docs/apps.html.in +++ b/docs/apps.html.in @@ -404,6 +404,14 @@ infrastructure. You can deploy a new service just dragging and dropping a VM. </dd> + <dt><a href="https://kimchi-project.github.io/kimchi/">Kimchi</a></dt> + <dd> + Kimchi is an HTML5 based management tool for KVM. It is designed to + make it as easy as possible to get started with KVM and create your first guest. + + Kimchi manages KVM guests through libvirt. The management interface is accessed + over the web using a browser that supports HTML5. + </dd> <dt><a href="http://ovirt.org/">oVirt</a></dt> <dd> oVirt provides the ability to manage large numbers of virtual -- 2.5.5

2 1

[libvirt] [PATCH v3 0/2] qemu: expand domain memory statistics
by Derbyshev Dmitriy 27 Jul '16

27 Jul '16

From: Derbyshev Dmitry <dderbyshev(a)virtuozzo.com> QEMU reports timestamp and available along with other memory statistics. This information was not saved into domain statistics. Changes since v1: * Enum numeration fixed * Macro getting "usage" field fixed Changes since v2: * previous patches were on wrong branch * qemu's stat name was "stat-available-memory" Derbyshev Dmitry (2): qemu: expand domain memory statistics with 'usable' qemu: expand domain memory statistics with 'last-update' timestamp include/libvirt/libvirt-domain.h | 11 ++++++++++- src/libvirt-domain.c | 5 +++++ src/qemu/qemu_monitor_json.c | 22 +++++++++++++--------- tools/virsh-domain-monitor.c | 4 ++++ 4 files changed, 32 insertions(+), 10 deletions(-) -- 1.9.5.msysgit.0

3 6

[libvirt] [PATCH v2 00/10] Remove caching of address sets
by Tomasz Flendrich 27 Jul '16

27 Jul '16

From: Tomasz Flendrich <Darge(a)users.noreply.github.com> These patches delete the caching of pci, virtioSerial and ccw address sets. I am deleting them, because they can be recalculated from the domain definition, and there's no point in keeping redundant data, especially because handling a persistent cache of addresses required using functions that released addresses. These functions aren't useful anymore, so they are dropped too. I know that USB addresses were added and that they are cached now. I am sure that they can be calculated on demand too though. If this gets accepted, I will rebase the other patches that depend on this one, making functions less dependant on qemu, and removing the code duplication. Changes in v2: * I rebased it, because there were some changes that added a usb address set. * virDomainReleaseDeviceAddress is no longer deleted, because it's used for usb addresses. It also means that in qemu_hotplug.c, this function's calls are not deleted anymore. * In qemu_hotplug.c, in one place, virDomainPCIAddressSetFree() was previously called after the "error:" label. I moved it to the "cleanup:", so the address set is freed even if an error occurs. * Add one missing free of pci address set, in qemuDomainAttachNetDevice(). Tomasz Flendrich (10): add virDomainVirtioSerialAddrSetCreateFromDomain qemu_hotplug: generate vioserial address list on demand qemu: remove vioserialaddrs caching Add qemuDomainCCWAddrSetCreateFromDomain qemu_hotplug: generate ccw address list on demand qemu: remove ccwaddrs caching add qemuDomainPCIAddrSetCreateFromDomain qemu_hotplug: generate pci address list on demand qemu: remove pciaddrs caching Remove unused functions that release addresses src/conf/domain_addr.c | 131 +++++++++--------------------------- src/conf/domain_addr.h | 19 +----- src/libvirt_private.syms | 4 +- src/qemu/qemu_domain.c | 3 - src/qemu/qemu_domain.h | 3 - src/qemu/qemu_domain_address.c | 147 ++++++++++++++++++++--------------------- src/qemu/qemu_domain_address.h | 9 +++ src/qemu/qemu_hotplug.c | 111 ++++++++++++++++++++++++------- 8 files changed, 201 insertions(+), 226 deletions(-) -- 2.7.4 (Apple Git-66)

3 19

[libvirt] [PATCH v2 00/10] Test persistent device attachment
by Tomasz Flendrich 27 Jul '16

27 Jul '16

Changes in v2: * rebased This patch requires another one, otherwise a small conflict appears: https://www.redhat.com/archives/libvir-list/2016-July/msg00402.html In qemu_driver.c, two functions were split to have less responsibility. They were also modified so that they could be used in qemuhotplugtest. Now instead of running different functions to attach different device types, a generic function is run. Hopefully this will result in more device types tested in the future. Suggestions on how to name these two new functions are very welcome. In the future, if one uses the following command: virsh attach-device --live --config the new device will have the same address assigned in both live and config domain. To make sure that everything will work properly after making these changes, I reworked qemuhotplug.c to work with attaching devices to persistent domains. So far, there is only one test for persistent attachment. I will add more testcases as soon as I hear some feedback on these changes. qemuhotplugtest.c should first be modified anyway, so that the three xmls's filenames (basis domain xml, device xml, expected xml) are stated explicitly instead of being calculated. This will allow for more flexibility in testing and less xml files duplicates. Tomasz Tomasz Flendrich (10): Change parameters to qemuDomainAttachDeviceLive Remove an unnecessary variable Split qemuDomainAttachDeviceFlags in two Narrow down a parameter Split qemuDomainDetachDeviceFlags in two Make qemu attach/detach functions public Narrow down a parameter to function Use more generic functions in qemuhotplugtest Make qemuhotplugtest work with persistent domains Add a persistent attachment testcase src/qemu/qemu_driver.c | 221 +++++++++++---------- src/qemu/qemu_driver.h | 14 ++ tests/qemuhotplugtest.c | 213 ++++++++++++-------- .../qemuhotplug-base-config+qemu-agent+config.xml | 45 +++++ .../qemuhotplug-base-config.xml | 40 ++++ 5 files changed, 349 insertions(+), 184 deletions(-) create mode 100644 tests/qemuhotplugtestdomains/qemuhotplug-base-config+qemu-agent+config.xml create mode 100644 tests/qemuhotplugtestdomains/qemuhotplug-base-config.xml -- 2.7.4

2 22

[libvirt] [PATCH v5] libxl: implement virDomainBlockStats
by Joao Martins 26 Jul '16

26 Jul '16

Introduce initial support for domainBlockStats API call that allow us to query block device statistics. openstack nova uses this API call to query block statistics, alongside virDomainMemoryStats and virDomainInterfaceStats. Note that this patch only introduces it for VBD for starters. QDisk would come in a separate patch series. A new statistics data structure is introduced to fit common statistics among others specific to the underlying block backends. For the VBD statistics on linux these are exported via sysfs on the path: "/sys/bus/xen-backend/devices/vbd-<domid>-<devid>/statistics" To calculate the block devno libxlDiskPathToID is introduced. Each backend implements its own function to extract statistics and let there be handled the different platforms. VBD stats are exposed in reqs and number of sectors from blkback, and it's up to us to convert it to sector sizes. The sector size is gathered through xenstore in the device backend entry "physical-sector-size". BlockStatsFlags variant is also implemented which has the added benefit of getting the number of flush requests. Signed-off-by: Joao Martins <joao.m.martins(a)oracle.com> --- Changes since v4: - No need to handle virtpath as devid - return id = -1 instead - Rework compilation of BlockStatsVBD compilation on other platforms. - Use disk->dst as opposed to plain path to also cover the case when input virtpath is path to file. Before it was returning "cannot find device number" Changes since v3: - No need for error in libxlDiskSectorSize() - Fix an identation issue. - Rework cleanup paths on libxlDiskSectorSize() - Do not unlock vm if libxlDomainObjEndJob() returns false and adjust to the newly introduced virDomainObjEndAPI() - Bump version to 2.1.0 Changes since v1: - Fix identation issues - Set ret to LIBXL_VBD_SECTOR_SIZE - Reuse VIR_STRDUP error instead of doing virReportError when we fail to set stats->backend - Change virAsprintf(...) error checking - Change error to VIR_ERR_OPERATION_FAILED when xenstore path does not exist and when failing to read stat. - Resolve issues with 'make syntax-check' with cppi installed. - Remove libxlDiskPathMatches in favor of using virutil virDiskNameParse to fetch disk and partition index. - Rename libxlDiskPathParse to libxlDiskPathToID and rework function to just convert disk and partition index to devno. - Bump version to 1.2.22 --- src/libxl/libxl_driver.c | 367 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 367 insertions(+) diff --git a/src/libxl/libxl_driver.c b/src/libxl/libxl_driver.c index f153f69..b33c898 100644 --- a/src/libxl/libxl_driver.c +++ b/src/libxl/libxl_driver.c @@ -30,6 +30,7 @@ #include <math.h> #include <libxl.h> #include <libxl_utils.h> +#include <xenstore.h> #include <fcntl.h> #include <regex.h> @@ -72,6 +73,7 @@ VIR_LOG_INIT("libxl.libxl_driver"); #define LIBXL_DOM_REQ_HALT 4 #define LIBXL_NB_TOTAL_CPU_STAT_PARAM 1 +#define LIBXL_NB_TOTAL_BLK_STAT_PARAM 6 #define HYPERVISOR_CAPABILITIES "/proc/xen/capabilities" #define HYPERVISOR_XENSTORED "/dev/xen/xenstored" @@ -100,6 +102,25 @@ struct _libxlOSEventHookInfo { int id; }; +/* Object used to store disk statistics across multiple xen backends */ +typedef struct _libxlBlockStats libxlBlockStats; +typedef libxlBlockStats *libxlBlockStatsPtr; +struct _libxlBlockStats { + long long rd_req; + long long rd_bytes; + long long wr_req; + long long wr_bytes; + long long f_req; + + char *backend; + union { + struct { + long long ds_req; + long long oo_req; + } vbd; + } u; +}; + /* Function declarations */ static int libxlDomainManagedSaveLoad(virDomainObjPtr vm, @@ -5031,6 +5052,350 @@ libxlDomainGetJobStats(virDomainPtr dom, } static int +libxlDiskPathToID(const char *virtpath) +{ + static char const* drive_prefix[] = {"xvd", "hd", "sd"}; + int disk, partition, chrused; + int fmt, id; + size_t i; + + fmt = id = -1; + + /* Find any disk prefixes we know about */ + for (i = 0; i < ARRAY_CARDINALITY(drive_prefix); i++) { + if (STRPREFIX(virtpath, drive_prefix[i]) && + !virDiskNameParse(virtpath, &disk, &partition)) { + fmt = i; + break; + } + } + + /* Handle it same way as xvd */ + if (fmt < 0 && + (sscanf(virtpath, "d%ip%i%n", &disk, &partition, &chrused) >= 2 + && chrused == strlen(virtpath))) + fmt = 0; + + /* Test indexes ranges and calculate the device id */ + switch (fmt) { + case 0: /* xvd */ + if (disk <= 15 && partition <= 15) + id = (202 << 8) | (disk << 4) | partition; + else if ((disk <= ((1<<20)-1)) || partition <= 255) + id = (1 << 28) | (disk << 8) | partition; + break; + case 1: /* hd */ + if (disk <= 3 && partition <= 63) + id = ((disk < 2 ? 3 : 22) << 8) | ((disk & 1) << 6) | partition; + break; + case 2: /* sd */ + if (disk <= 15 && (partition <= 15)) + id = (8 << 8) | (disk << 4) | partition; + break; + default: /* invalid */ + break; + } + return id; +} + +#define LIBXL_VBD_SECTOR_SIZE 512 + +static int +libxlDiskSectorSize(int domid, int devno) +{ + char *path, *val; + struct xs_handle *handle; + int ret = LIBXL_VBD_SECTOR_SIZE; + unsigned int len; + + handle = xs_daemon_open_readonly(); + if (!handle) { + VIR_WARN("cannot read sector size"); + return ret; + } + + path = val = NULL; + if (virAsprintf(&path, "/local/domain/%d/device/vbd/%d/backend", + domid, devno) < 0) + goto cleanup; + + if ((val = xs_read(handle, XBT_NULL, path, &len)) == NULL) + goto cleanup; + + VIR_FREE(path); + if (virAsprintf(&path, "%s/physical-sector-size", val) < 0) + goto cleanup; + + VIR_FREE(val); + if ((val = xs_read(handle, XBT_NULL, path, &len)) == NULL) + goto cleanup; + + if (sscanf(val, "%d", &ret) != 1) + ret = LIBXL_VBD_SECTOR_SIZE; + + cleanup: + VIR_FREE(val); + VIR_FREE(path); + xs_daemon_close(handle); + return ret; +} + +#ifdef __linux__ +static int +libxlDomainBlockStatsVBD(virDomainObjPtr vm, + const char *dev, + libxlBlockStatsPtr stats) +{ + int ret = -1; + int devno = libxlDiskPathToID(dev); + int size = libxlDiskSectorSize(vm->def->id, devno); + char *path, *name, *val; + unsigned long long stat; + + path = name = val = NULL; + if (devno < 0) { + virReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("cannot find device number")); + return ret; + } + if (VIR_STRDUP(stats->backend, "vbd") < 0) + return ret; + + if (virAsprintf(&path, "/sys/bus/xen-backend/devices/vbd-%d-%d/statistics", + vm->def->id, devno) < 0) + return ret; + + if (!virFileExists(path)) { + virReportError(VIR_ERR_OPERATION_FAILED, + "%s", _("cannot open bus path")); + goto cleanup; + } + +# define LIBXL_SET_VBDSTAT(FIELD, VAR, MUL) \ + if ((virAsprintf(&name, "%s/"FIELD, path) < 0) || \ + (virFileReadAll(name, 256, &val) < 0) || \ + (sscanf(val, "%llu", &stat) != 1)) { \ + virReportError(VIR_ERR_OPERATION_FAILED, \ + _("cannot read %s"), name); \ + goto cleanup; \ + } \ + VAR += (stat * MUL); \ + VIR_FREE(name); \ + VIR_FREE(val); + + LIBXL_SET_VBDSTAT("f_req", stats->f_req, 1) + LIBXL_SET_VBDSTAT("wr_req", stats->wr_req, 1) + LIBXL_SET_VBDSTAT("rd_req", stats->rd_req, 1) + LIBXL_SET_VBDSTAT("wr_sect", stats->wr_bytes, size) + LIBXL_SET_VBDSTAT("rd_sect", stats->rd_bytes, size) + + LIBXL_SET_VBDSTAT("ds_req", stats->u.vbd.ds_req, size) + LIBXL_SET_VBDSTAT("oo_req", stats->u.vbd.oo_req, 1) + ret = 0; + + cleanup: + VIR_FREE(name); + VIR_FREE(path); + VIR_FREE(val); + +# undef LIBXL_SET_VBDSTAT + + return ret; +} +#else +static int +libxlDomainBlockStatsVBD(virDomainObjPtr vm ATTRIBUTE_UNUSED, + const char *dev ATTRIBUTE_UNUSED, + libxlBlockStatsPtr stats ATTRIBUTE_UNUSED) +{ + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, + "%s", _("platform unsupported")); + return -1; +} +#endif + +static int +libxlDomainBlockStatsGatherSingle(virDomainObjPtr vm, + const char *path, + libxlBlockStatsPtr stats) +{ + virDomainDiskDefPtr disk; + const char *disk_drv; + int ret = -1, disk_fmt; + + if (!(disk = virDomainDiskByName(vm->def, path, false))) { + virReportError(VIR_ERR_OPERATION_INVALID, + _("invalid path: %s"), path); + return ret; + } + + disk_fmt = virDomainDiskGetFormat(disk); + if (!(disk_drv = virDomainDiskGetDriver(disk))) + disk_drv = "qemu"; + + if (STREQ(disk_drv, "phy")) { + if (disk_fmt != VIR_STORAGE_FILE_RAW && + disk_fmt != VIR_STORAGE_FILE_NONE) { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, + _("unsupported format %s"), + virStorageFileFormatTypeToString(disk_fmt)); + return ret; + } + + ret = libxlDomainBlockStatsVBD(vm, disk->dst, stats); + } else { + virReportError(VIR_ERR_OPERATION_UNSUPPORTED, + _("unsupported disk driver %s"), + disk_drv); + } + return ret; +} + +static int +libxlDomainBlockStatsGather(virDomainObjPtr vm, + const char *path, + libxlBlockStatsPtr stats) +{ + int ret = -1; + + if (*path) { + if (libxlDomainBlockStatsGatherSingle(vm, path, stats) < 0) + return ret; + } else { + size_t i; + + for (i = 0; i < vm->def->ndisks; ++i) { + if (libxlDomainBlockStatsGatherSingle(vm, vm->def->disks[i]->dst, + stats) < 0) + return ret; + } + } + return 0; +} + +static int +libxlDomainBlockStats(virDomainPtr dom, + const char *path, + virDomainBlockStatsPtr stats) +{ + libxlDriverPrivatePtr driver = dom->conn->privateData; + virDomainObjPtr vm; + libxlBlockStats blkstats; + int ret = -1; + + if (!(vm = libxlDomObjFromDomain(dom))) + goto cleanup; + + if (virDomainBlockStatsEnsureACL(dom->conn, vm->def) < 0) + goto cleanup; + + if (libxlDomainObjBeginJob(driver, vm, LIBXL_JOB_QUERY) < 0) + goto cleanup; + + if (!virDomainObjIsActive(vm)) { + virReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + + memset(&blkstats, 0, sizeof(libxlBlockStats)); + if ((ret = libxlDomainBlockStatsGather(vm, path, &blkstats)) < 0) + goto endjob; + + stats->rd_req = blkstats.rd_req; + stats->rd_bytes = blkstats.rd_bytes; + stats->wr_req = blkstats.wr_req; + stats->wr_bytes = blkstats.wr_bytes; + if (STREQ_NULLABLE(blkstats.backend, "vbd")) + stats->errs = blkstats.u.vbd.oo_req; + else + stats->errs = -1; + + endjob: + libxlDomainObjEndJob(driver, vm); + + cleanup: + virDomainObjEndAPI(&vm); + return ret; +} + +static int +libxlDomainBlockStatsFlags(virDomainPtr dom, + const char *path, + virTypedParameterPtr params, + int *nparams, + unsigned int flags) +{ + libxlDriverPrivatePtr driver = dom->conn->privateData; + virDomainObjPtr vm; + libxlBlockStats blkstats; + int nstats; + int ret = -1; + + virCheckFlags(VIR_TYPED_PARAM_STRING_OKAY, -1); + + flags &= ~VIR_TYPED_PARAM_STRING_OKAY; + + if (!(vm = libxlDomObjFromDomain(dom))) + goto cleanup; + + if (virDomainBlockStatsFlagsEnsureACL(dom->conn, vm->def) < 0) + goto cleanup; + + if (libxlDomainObjBeginJob(driver, vm, LIBXL_JOB_QUERY) < 0) + goto cleanup; + + if (!virDomainObjIsActive(vm)) { + virReportError(VIR_ERR_OPERATION_INVALID, + "%s", _("domain is not running")); + goto endjob; + } + + /* return count of supported stats */ + if (*nparams == 0) { + *nparams = LIBXL_NB_TOTAL_BLK_STAT_PARAM; + ret = 0; + goto endjob; + } + + memset(&blkstats, 0, sizeof(libxlBlockStats)); + if ((ret = libxlDomainBlockStatsGather(vm, path, &blkstats)) < 0) + goto endjob; + + nstats = 0; + +#define LIBXL_BLKSTAT_ASSIGN_PARAM(VAR, NAME) \ + if (nstats < *nparams && (blkstats.VAR) != -1) { \ + if (virTypedParameterAssign(params + nstats, NAME, \ + VIR_TYPED_PARAM_LLONG, (blkstats.VAR)) < 0) \ + goto endjob; \ + nstats++; \ + } + + LIBXL_BLKSTAT_ASSIGN_PARAM(wr_bytes, VIR_DOMAIN_BLOCK_STATS_WRITE_BYTES); + LIBXL_BLKSTAT_ASSIGN_PARAM(wr_req, VIR_DOMAIN_BLOCK_STATS_WRITE_REQ); + + LIBXL_BLKSTAT_ASSIGN_PARAM(rd_bytes, VIR_DOMAIN_BLOCK_STATS_READ_BYTES); + LIBXL_BLKSTAT_ASSIGN_PARAM(rd_req, VIR_DOMAIN_BLOCK_STATS_READ_REQ); + + LIBXL_BLKSTAT_ASSIGN_PARAM(f_req, VIR_DOMAIN_BLOCK_STATS_FLUSH_REQ); + + if (STREQ_NULLABLE(blkstats.backend, "vbd")) + LIBXL_BLKSTAT_ASSIGN_PARAM(u.vbd.oo_req, VIR_DOMAIN_BLOCK_STATS_ERRS); + + *nparams = nstats; + +#undef LIBXL_BLKSTAT_ASSIGN_PARAM + + endjob: + libxlDomainObjEndJob(driver, vm); + + cleanup: + virDomainObjEndAPI(&vm); + return ret; +} + +static int libxlConnectDomainEventRegisterAny(virConnectPtr conn, virDomainPtr dom, int eventID, virConnectDomainEventGenericCallback callback, void *opaque, virFreeCallback freecb) @@ -5847,6 +6212,8 @@ static virHypervisorDriver libxlHypervisorDriver = { .domainMemoryStats = libxlDomainMemoryStats, /* 1.3.0 */ .domainGetCPUStats = libxlDomainGetCPUStats, /* 1.3.0 */ .domainInterfaceStats = libxlDomainInterfaceStats, /* 1.3.2 */ + .domainBlockStats = libxlDomainBlockStats, /* 2.1.0 */ + .domainBlockStatsFlags = libxlDomainBlockStatsFlags, /* 2.1.0 */ .connectDomainEventRegister = libxlConnectDomainEventRegister, /* 0.9.0 */ .connectDomainEventDeregister = libxlConnectDomainEventDeregister, /* 0.9.0 */ .domainManagedSave = libxlDomainManagedSave, /* 0.9.2 */ -- 2.1.4

2 2

[libvirt] [PATCH] qemu: rename QEMU_CAPS_MLOCK to QEMU_CAPS_REALTIME_MLOCK
by Henning Schild 26 Jul '16

26 Jul '16

Purely cosmetic change to be consistent with the other names. Signed-off-by: Henning Schild <henning.schild(a)siemens.com> --- Something i found in an old staging queue and that might be considered useful. src/qemu/qemu_capabilities.c | 2 +- src/qemu/qemu_capabilities.h | 2 +- src/qemu/qemu_command.c | 4 ++-- tests/qemuxml2argvtest.c | 4 ++-- 4 files changed, 6 insertions(+), 6 deletions(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index f600ce9..d5b73e6 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -2655,7 +2655,7 @@ static struct virQEMUCapsCommandLineProps virQEMUCapsCommandLine[] = { { "machine", "vmport", QEMU_CAPS_MACHINE_VMPORT_OPT }, { "drive", "discard", QEMU_CAPS_DRIVE_DISCARD }, { "drive", "detect-zeroes", QEMU_CAPS_DRIVE_DETECT_ZEROES }, - { "realtime", "mlock", QEMU_CAPS_MLOCK }, + { "realtime", "mlock", QEMU_CAPS_REALTIME_MLOCK }, { "boot-opts", "strict", QEMU_CAPS_BOOT_STRICT }, { "boot-opts", "reboot-timeout", QEMU_CAPS_REBOOT_TIMEOUT }, { "boot-opts", "splash-time", QEMU_CAPS_SPLASH_TIMEOUT }, diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index ca84f27..bd5c6d9 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -254,7 +254,7 @@ typedef enum { QEMU_CAPS_MEM_MERGE, /* -machine mem-merge */ QEMU_CAPS_VNC_WEBSOCKET, /* -vnc x:y,websocket */ QEMU_CAPS_DRIVE_DISCARD, /* -drive discard=off(ignore)|on(unmap) */ - QEMU_CAPS_MLOCK, /* -realtime mlock=on|off */ + QEMU_CAPS_REALTIME_MLOCK, /* -realtime mlock=on|off */ /* 150 */ QEMU_CAPS_VNC_SHARE_POLICY, /* set display sharing policy */ diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 4558b9f..3dc131b 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -7022,12 +7022,12 @@ qemuBuildMemCommandLine(virCommandPtr cmd, qemuBuildMemPathStr(cfg, def, qemuCaps, cmd) < 0) return -1; - if (def->mem.locked && !virQEMUCapsGet(qemuCaps, QEMU_CAPS_MLOCK)) { + if (def->mem.locked && !virQEMUCapsGet(qemuCaps, QEMU_CAPS_REALTIME_MLOCK)) { virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("memory locking not supported by QEMU binary")); return -1; } - if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_MLOCK)) { + if (virQEMUCapsGet(qemuCaps, QEMU_CAPS_REALTIME_MLOCK)) { virCommandAddArg(cmd, "-realtime"); virCommandAddArgFormat(cmd, "mlock=%s", def->mem.locked ? "on" : "off"); diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 356f843..a5d51a8 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -1750,9 +1750,9 @@ mymain(void) QEMU_CAPS_VIRTIO_SCSI, QEMU_CAPS_VIRTIO_SCSI, QEMU_CAPS_DEVICE_SCSI_GENERIC); - DO_TEST("mlock-on", QEMU_CAPS_MLOCK); + DO_TEST("mlock-on", QEMU_CAPS_REALTIME_MLOCK); DO_TEST_FAILURE("mlock-on", NONE); - DO_TEST("mlock-off", QEMU_CAPS_MLOCK); + DO_TEST("mlock-off", QEMU_CAPS_REALTIME_MLOCK); DO_TEST("mlock-unsupported", NONE); DO_TEST_PARSE_ERROR("pci-bridge-negative-index-invalid", -- 2.7.3

2 1

[libvirt] [PATCH 0/9] Introducing node device lifecycle event APIs
by Jovanka Gulicoska 26 Jul '16

26 Jul '16

Introducing implementation of node device event APIs. Code changes follow network/storage pool event APIs. Implemented functions: virNodeDeviceEventRegisterAny(), virNodeDeviceEventDeregisterAny(), virNodeDeviceLifeCycleEventNew(), introduced CREATED and DELETED. There are also test as well as unittests for the new functions and signals. This is part of a GSOC project: Asynchronous lifecycle events for storage objects Jovanka Gulicoska (9): Introduce node device lifecycle event APIs conf: add node_device_event handling conf: events: don't crash on NULL uuid test: implement node device lifecycle event APIs remote: implement node device lifecycle event APIs node_device: implement node device lifecycle event APIs node_device: Implement event queue in udev event-test: support node device lifecycle event APIs virsh: Introduce nodedev-event command daemon/libvirtd.h | 2 + daemon/remote.c | 206 ++++++++++++++++++++++++++++++ examples/object-events/event-test.c | 68 ++++++++++ include/libvirt/libvirt-nodedev.h | 90 ++++++++++++++ po/POTFILES.in | 1 + src/Makefile.am | 5 + src/conf/node_device_conf.h | 4 + src/conf/node_device_event.c | 234 +++++++++++++++++++++++++++++++++++ src/conf/node_device_event.h | 59 +++++++++ src/conf/object_event.c | 3 +- src/datatypes.h | 13 ++ src/driver-nodedev.h | 14 +++ src/libvirt-nodedev.c | 127 +++++++++++++++++++ src/libvirt_private.syms | 5 + src/libvirt_public.syms | 6 + src/node_device/node_device_driver.c | 42 +++++++ src/node_device/node_device_driver.h | 10 ++ src/node_device/node_device_udev.c | 24 ++++ src/remote/remote_driver.c | 139 +++++++++++++++++++++ src/remote/remote_protocol.x | 43 ++++++- src/remote_protocol-structs | 19 +++ src/test/test_driver.c | 49 ++++++++ tests/objecteventtest.c | 72 +++++++++++ tools/virsh-nodedev.c | 211 +++++++++++++++++++++++++++++++ tools/virsh-nodedev.h | 10 ++ tools/virsh.pod | 18 +++ 26 files changed, 1472 insertions(+), 2 deletions(-) create mode 100644 src/conf/node_device_event.c create mode 100644 src/conf/node_device_event.h -- 2.7.4

3 20

[libvirt] [PATCH] qemu: Use legacy USB on ppc64
by Andrea Bolognani 26 Jul '16

26 Jul '16

Commit 8156493d8db9 changed libvirt so that '-device pci-ohci' would be used instead of '-usb' on ppc64 when no specific USB controller model had been specified in the guest configuration. While the device that ends up being presented by the guest is exactly the same, '-usb' causes it to be assigned to PCI address 00:00.0 while '-device pci-ohci', being subject to the regular PCI address assignment logic, will be at a different address. This PCI address mismatch breaks migration of existing guests to new libvirt versions. Luckily, when QEMU has switched its default '-usb' controller from pci-ohci to nec-usb-xhci (QEMU commit 57040d451315), it has done so without affecting older machine types, which means we can keep using '-usb' without risking guest ABI breakage. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1357468 --- src/qemu/qemu_command.c | 8 +++++--- tests/qemuxml2argvdata/qemuxml2argv-ppc64-usb-controller.args | 2 +- tests/qemuxml2argvdata/qemuxml2argv-pseries-usb-default.args | 2 +- 3 files changed, 7 insertions(+), 5 deletions(-) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 2e04f57..9050e77 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -2827,9 +2827,11 @@ qemuBuildControllerDevCommandLine(virCommandPtr cmd, /* We're not using legacy usb controller for q35 */ if (ARCH_IS_PPC64(def->os.arch)) { - /* For ppc64 the legacy was OHCI */ - if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_PCI_OHCI)) - need_legacy = true; + /* ppc64 needs to use the -usb flag in order not to break + * migration of existing guests: the legacy USB controller + * uses a PCI address that we have no way of assigning + * using the usual -device machinery */ + need_legacy = true; } else { /* For anything else, we used PIIX3_USB_UHCI */ if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_PIIX3_USB_UHCI)) diff --git a/tests/qemuxml2argvdata/qemuxml2argv-ppc64-usb-controller.args b/tests/qemuxml2argvdata/qemuxml2argv-ppc64-usb-controller.args index 2ec2231..bed5045 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-ppc64-usb-controller.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-ppc64-usb-controller.args @@ -15,5 +15,5 @@ QEMU_AUDIO_DRV=none \ -nodefaults \ -monitor unix:/tmp/lib/domain--1-QEMUGuest1/monitor.sock,server,nowait \ -boot c \ --device pci-ohci,id=usb,bus=pci,addr=0x1 \ +-usb \ -device virtio-balloon-pci,id=balloon0,bus=pci,addr=0x6 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-pseries-usb-default.args b/tests/qemuxml2argvdata/qemuxml2argv-pseries-usb-default.args index 251e786..2839dea 100644 --- a/tests/qemuxml2argvdata/qemuxml2argv-pseries-usb-default.args +++ b/tests/qemuxml2argvdata/qemuxml2argv-pseries-usb-default.args @@ -19,6 +19,6 @@ server,nowait \ -mon chardev=charmonitor,id=monitor,mode=readline \ -no-acpi \ -boot c \ --device pci-ohci,id=usb,bus=pci,addr=0x1 \ +-usb \ -chardev pty,id=charserial0 \ -device spapr-vty,chardev=charserial0,reg=0x30000000 -- 2.7.4

2 6

[libvirt] [PATCHv3] network: Added hook for network modification event
by Khramov Anton 26 Jul '16

26 Jul '16

From: Anton Khramov <anton(a)endocode.com> Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1181539 Reworked previous patchs. Moved "networkRunHook" after "UPDATE_AFFECT_LIVE" as Cole suggested. Anton Khramov (1): network: Added hook for network modification event docs/hooks.html.in | 2 ++ src/network/bridge_driver.c | 6 ++++++ src/util/virhook.c | 3 ++- src/util/virhook.h | 1 + 4 files changed, 11 insertions(+), 1 deletion(-) -- 1.9.1

4 5

[libvirt] [PATCH 0/2] Remove calls that just clear alias
by John Ferlan 26 Jul '16

26 Jul '16

Based off review of : http://www.redhat.com/archives/libvir-list/2016-July/msg00672.html Rather than add the ignore_value, just remove the unused / unnecessary code Patch 1 has no callers any more and from inspection patch 2 is essentially redone again as each device is handled during domain def deletion. Ran make -C tests valgrind with no errors too John Ferlan (2): conf: Remove CCW,PCI clear address helpers Remove unnecessary virDomainDefClearDeviceAliases src/conf/domain_conf.c | 50 ------------------------------------------------ src/conf/domain_conf.h | 3 --- src/libvirt_private.syms | 3 --- src/qemu/qemu_process.c | 2 -- 4 files changed, 58 deletions(-) -- 2.5.5

2 3

[libvirt] [RFC] Switch to nftables from netfilter
by Michal Privoznik 26 Jul '16

26 Jul '16

Dear list, I've came across nftables [1]. They look very promising, more than old netfilter. It offers new features [2], from which I'd pick: - better performance under high traffic workloads - atomic filter/chain replacements - transactions I haven't investigated how much work will be required on our side if we try to implement the switch (well, for starters, we can have two subsystems living next to each other). I want to check what do you guys think before actually digging into the code. The nftables was merged into 3.13 Linux kernel and thus should be available on all major distros. Well, since we will have both subsystems available, we should be good to go. BTW: it's a bit shame that this nifty project hasn't received much more advertising. Looks cool so far. Michal 1: https://en.wikipedia.org/wiki/Nftables 2: http://people.netfilter.org/kaber/nfws2008/nftables.odp

2 1