March 2015 - Devel - libvirt List Archives

[libvirt] [PATCH 0/2] PowerPC : Miscellaneous fixes for 'ppc64le' architecture.
by Prerna Saxena 04 Mar '15

04 Mar '15

>From a28ef5a3e7b9cb023948cf97d9f472bb3a1e06d3 Mon Sep 17 00:00:00 2001 From: Prerna Saxena <prerna(a)linux.vnet.ibm.com> Date: Thu, 26 Feb 2015 22:31:05 +0530 This series adds few miscellaneous fixes for PowerPC 64-bit Little Endian Architecture. Changelog : ========== v1 of Patch 1/2 already posted and acked : https://www.redhat.com/archives/libvir-list/2015-February/msg00486.html Patch 2/2 adds a testcase to supplement patch 1. Prerna Saxena (2): PowerPC: Augment XML schema to include 'ppc64le' arch and newer pseries-2.* machine types. Tests : Add test for 'ppc64le' architecture. docs/schemas/basictypes.rng | 1 + docs/schemas/domaincommon.rng | 7 ++++- .../qemuxml2argv-pseries-cpu-le.args | 7 +++++ .../qemuxml2argv-pseries-cpu-le.xml | 17 ++++++++++++ tests/qemuxml2argvtest.c | 2 ++ tests/testutilsqemu.c | 30 ++++++++++++++++++++++ 6 files changed, 63 insertions(+), 1 deletion(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pseries-cpu-le.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-pseries-cpu-le.xml -- 1.9.3 -- Prerna Saxena Linux Technology Centre, IBM Systems and Technology Lab, Bangalore, India

3 7

[libvirt] [PATCHv3 0/4] Automaticaly fill <memory> element for NUMA enabled guests
by Peter Krempa 04 Mar '15

04 Mar '15

While there is still a discussion open upstream on a possible change of naming here I'm posting this so that comments can be held against this version. Peter Krempa (4): conf: Replace access to def->mem.max_balloon with accessor functions qemu: command: Add helper to align memory sizes conf: Automatically use NUMA memory size in case NUMA is enabled conf: Make specifying <memory> optional docs/schemas/domaincommon.rng | 18 ++--- src/conf/domain_conf.c | 81 +++++++++++++++++++--- src/conf/domain_conf.h | 4 ++ src/hyperv/hyperv_driver.c | 2 +- src/libvirt_private.syms | 3 + src/libxl/libxl_conf.c | 2 +- src/libxl/libxl_driver.c | 8 +-- src/lxc/lxc_cgroup.c | 2 +- src/lxc/lxc_driver.c | 12 ++-- src/lxc/lxc_fuse.c | 4 +- src/lxc/lxc_native.c | 4 +- src/openvz/openvz_driver.c | 2 +- src/parallels/parallels_driver.c | 2 +- src/parallels/parallels_sdk.c | 12 ++-- src/phyp/phyp_driver.c | 11 +-- src/qemu/qemu_command.c | 23 +++--- src/qemu/qemu_domain.c | 21 ++++++ src/qemu/qemu_domain.h | 2 + src/qemu/qemu_driver.c | 21 +++--- src/qemu/qemu_hotplug.c | 8 ++- src/qemu/qemu_process.c | 2 +- src/test/test_driver.c | 8 +-- src/uml/uml_driver.c | 8 +-- src/vbox/vbox_common.c | 4 +- src/vmware/vmware_driver.c | 2 +- src/vmx/vmx.c | 12 ++-- src/xen/xm_internal.c | 14 ++-- src/xenapi/xenapi_driver.c | 2 +- src/xenapi/xenapi_utils.c | 4 +- src/xenconfig/xen_common.c | 8 ++- src/xenconfig/xen_sxpr.c | 9 +-- .../qemuxml2argv-cpu-numa-no-memory-element.args | 7 ++ .../qemuxml2argv-cpu-numa-no-memory-element.xml | 24 +++++++ .../qemuxml2argv-minimal-no-memory.xml | 25 +++++++ .../qemuxml2argv-numatune-memnode.args | 2 +- tests/qemuxml2argvtest.c | 2 + .../qemuxml2xmlout-cpu-numa-no-memory-element.xml | 28 ++++++++ tests/qemuxml2xmltest.c | 1 + 38 files changed, 299 insertions(+), 105 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-cpu-numa-no-memory-element.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-cpu-numa-no-memory-element.xml create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-minimal-no-memory.xml create mode 100644 tests/qemuxml2xmloutdata/qemuxml2xmlout-cpu-numa-no-memory-element.xml -- 2.2.2

1 4

[libvirt] [PATCH] parallels: Cleanup partly opened drivers on connect open failure
by Michal Privoznik 04 Mar '15

04 Mar '15

Well, the parallelsConnectOpen() joins several sub-driver openings into one big if condition. If any of sub-driver fails to open, the whole API finishes immediately. The problem is, sub-drivers may have left some memory allocated. Fortunately, we have a free function for that: parallelsConnectClose(). This is, however, not prepared for partially allocated driver structure. So, prepare the free function for it and call it at the right place, in the if body. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- This commit assumes that virNetworkObjListFree() accepts NULL, and thus depends on this patch (which has been ACKed already): https://www.redhat.com/archives/libvir-list/2015-February/msg01127.html Should I push both during freeze or wait until after the release? src/parallels/parallels_driver.c | 7 ++++++- src/parallels/parallels_network.c | 4 ++++ src/parallels/parallels_storage.c | 6 ++++++ 3 files changed, 16 insertions(+), 1 deletion(-) diff --git a/src/parallels/parallels_driver.c b/src/parallels/parallels_driver.c index c9338b5..650b790 100644 --- a/src/parallels/parallels_driver.c +++ b/src/parallels/parallels_driver.c @@ -264,8 +264,10 @@ parallelsConnectOpen(virConnectPtr conn, if ((ret = parallelsOpenDefault(conn)) != VIR_DRV_OPEN_SUCCESS || (ret = parallelsStorageOpen(conn, flags)) != VIR_DRV_OPEN_SUCCESS || - (ret = parallelsNetworkOpen(conn, flags)) != VIR_DRV_OPEN_SUCCESS) + (ret = parallelsNetworkOpen(conn, flags)) != VIR_DRV_OPEN_SUCCESS) { + parallelsConnectClose(conn); return ret; + } return VIR_DRV_OPEN_SUCCESS; } @@ -275,6 +277,9 @@ parallelsConnectClose(virConnectPtr conn) { parallelsConnPtr privconn = conn->privateData; + if (!privconn) + return 0; + parallelsNetworkClose(conn); parallelsStorageClose(conn); diff --git a/src/parallels/parallels_network.c b/src/parallels/parallels_network.c index 3e7087d..c5834c5 100644 --- a/src/parallels/parallels_network.c +++ b/src/parallels/parallels_network.c @@ -336,6 +336,10 @@ parallelsNetworkOpen(virConnectPtr conn, int parallelsNetworkClose(virConnectPtr conn) { parallelsConnPtr privconn = conn->privateData; + + if (!privconn) + return 0; + parallelsDriverLock(privconn); virNetworkObjListFree(&privconn->networks); parallelsDriverUnlock(privconn); diff --git a/src/parallels/parallels_storage.c b/src/parallels/parallels_storage.c index d2c5bf2..56c1b42 100644 --- a/src/parallels/parallels_storage.c +++ b/src/parallels/parallels_storage.c @@ -72,9 +72,15 @@ parallelsStorageClose(virConnectPtr conn) { parallelsConnPtr privconn = conn->privateData; + if (!privconn) + return 0; + virStorageDriverStatePtr storageState = privconn->storageState; privconn->storageState = NULL; + if (!storageState) + return 0; + parallelsStorageLock(storageState); virStoragePoolObjListFree(&privconn->pools); VIR_FREE(storageState->configDir); -- 2.0.5

3 2

[libvirt] QEMU, libvirt, and KVM are participating in Outreachy May-August 2015
by Stefan Hajnoczi 04 Mar '15

04 Mar '15

QEMU, libvirt, and KVM are participating in the Outreachy (formerly known as Outreach Program for Women) May-August 2015 round. Outreachy is a 12-week full-time paid internship for working on open source projects. We are able to participate thanks to sponsors who will be listed on the Outreachy website in the future: http://outreachy.org/ >From the website: Outreachy is "open to women (cis and trans), trans men, genderqueer people, and all participants of the Ascend Project (http://ascendproject.org/) regardless of gender". Check out the project ideas list and details on how to apply: http://qemu-project.org/Outreachy_2015_MayAugust We encourage applicants to apply for both Outreachy and Google Summer of Code if they are eligible: http://qemu-project.org/Google_Summer_of_Code_2015 Stefan

1 0

[libvirt] [libvirt-designer][PATCH] Makefile: Build tests at the same time as the rest
by Michal Privoznik 04 Mar '15

04 Mar '15

Currently, since we were using check_PROGRAMS in the Makefile template, the test program build was postponed until 'make check' was run. That's not optimal. Lets build (not run!) test program in the 'make all' phase. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- >From my libvirt development experience, I'm used to just run make from my text editor. For instance, when changing some function header, I just run make and let the compiler catch all the places that need fixing. I think it's common practice. libvirt-designer/Makefile.am | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/libvirt-designer/Makefile.am b/libvirt-designer/Makefile.am index bda4240..fd7627d 100644 --- a/libvirt-designer/Makefile.am +++ b/libvirt-designer/Makefile.am @@ -82,9 +82,9 @@ libvirt-designer-enum-types.c: $(DESIGNER_HEADER_FILES) libvirt-designer-enum-ty $(AM_V_GEN) ( $(GLIB_MKENUMS) --template $(srcdir)/libvirt-designer-enum-types.c.template $(DESIGNER_HEADER_FILES:%=$(srcdir)/%) ) | \ sed -e "s/G_TYPE_VIR_CONFIG/GVIR_CONFIG_TYPE/" -e "s/g_vir/gvir/" > libvirt-designer-enum-types.c -check_PROGRAMS = test-designer-domain +noinst_PROGRAMS = test-designer-domain -TESTS = $(check_PROGRAMS) +TESTS = $(noinst_PROGRAMS) test_designer_domain_CFLAGS = \ -I$(top_srcdir) \ -- 2.0.5

2 2

[libvirt] [PATCH] storage: tweak condition to properly test lseek
by Erik Skultety 04 Mar '15

04 Mar '15

According to the POSIX standard, off_t (returned by lseek) is defined as signed integral type no shorter than int. Because our offset variable is defined as unsigned long long, the original check was passed successfully if UINT64_MAX had been used as offset value, due to implicit conversion. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1177219 --- src/fdstream.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/fdstream.c b/src/fdstream.c index 5d80fc2..b8ea86e 100644 --- a/src/fdstream.c +++ b/src/fdstream.c @@ -610,7 +610,7 @@ virFDStreamOpenFileInternal(virStreamPtr st, } if (offset && - lseek(fd, offset, SEEK_SET) != offset) { + lseek(fd, offset, SEEK_SET) < 0) { virReportSystemError(errno, _("Unable to seek %s to %llu"), path, offset); -- 1.9.3

3 2

[libvirt] [PATCH] virsh: tweak domif-getlink link state reporting message
by Erik Skultety 04 Mar '15

04 Mar '15

According to docs, we only support 2 link states for an interface up/down, 'up' being the default state if link state is unspecified in domain's XML, so the message when no link state is provided should be changed a little. Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1141119 --- tools/virsh-domain-monitor.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/virsh-domain-monitor.c b/tools/virsh-domain-monitor.c index 925eb1b..18ce9ae 100644 --- a/tools/virsh-domain-monitor.c +++ b/tools/virsh-domain-monitor.c @@ -756,7 +756,7 @@ cmdDomIfGetLink(vshControl *ctl, const vshCmd *cmd) if ((state = virXPathString("string(./link/@state)", ctxt))) vshPrint(ctl, "%s %s", iface, state); else - vshPrint(ctl, "%s default", iface); + vshPrint(ctl, "%s up", iface); ret = true; -- 1.9.3

2 2

[libvirt] [libvirt-designer 1/2] Fix gvir_designer_domain_get_fallback_disk_controller leak
by Christophe Fergeau 04 Mar '15

04 Mar '15

When the list of devices is empty (osinfo_list_get_length(devices) == 0), the 'devices' object would be leaked. --- libvirt-designer/libvirt-designer-domain.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/libvirt-designer/libvirt-designer-domain.c b/libvirt-designer/libvirt-designer-domain.c index c1a0e83..6629fb5 100644 --- a/libvirt-designer/libvirt-designer-domain.c +++ b/libvirt-designer/libvirt-designer-domain.c @@ -1517,7 +1517,7 @@ gvir_designer_domain_get_fallback_disk_controller(GVirDesignerDomain *design, GError **error) { OsinfoEntity *dev = NULL; - OsinfoDeviceList *devices; + OsinfoDeviceList *devices = NULL; OsinfoFilter *filter; int virt_type; @@ -1552,9 +1552,11 @@ gvir_designer_domain_get_fallback_disk_controller(GVirDesignerDomain *design, dev = osinfo_list_get_nth(OSINFO_LIST(devices), 0); g_object_ref(G_OBJECT(dev)); - g_object_unref(G_OBJECT(devices)); cleanup: + if (devices != NULL) + g_object_unref(G_OBJECT(devices)); + return OSINFO_DEVICE(dev); } -- 2.1.0

2 3

[libvirt] [libvirt-designer][PATCH] test: Add new SimpleDisk test
by Michal Privoznik 04 Mar '15

04 Mar '15

This test is meant for adding CDROM, floppy and disk to a domain. For now, each method has _file and _device variant, which are tested for now. Signed-off-by: Michal Privoznik <mprivozn(a)redhat.com> --- Funny, when reviewing Chritophe's patches, I've found one forgotten branch too. libvirt-designer/test-designer-domain.c | 101 ++++++++++++++++++++++++++++++++ 1 file changed, 101 insertions(+) diff --git a/libvirt-designer/test-designer-domain.c b/libvirt-designer/test-designer-domain.c index bbd5a28..90f7e68 100644 --- a/libvirt-designer/test-designer-domain.c +++ b/libvirt-designer/test-designer-domain.c @@ -125,6 +125,41 @@ static const gchar *capslxcxml = " </guest>" "</capabilities>"; +static const gchar *domain_machine_simple_iso_result = + "<domain>\n" + " <devices>\n" + " <disk type=\"file\">\n" + " <source file=\"/foo/bar1\"/>\n" + " <driver name=\"qemu\" type=\"raw\"/>\n" + " <target bus=\"ide\" dev=\"hda\"/>\n" + " </disk>\n" + " <disk type=\"block\">\n" + " <source dev=\"/foo/bar2\"/>\n" + " <driver name=\"qemu\" type=\"raw\"/>\n" + " <target bus=\"ide\" dev=\"hdb\"/>\n" + " </disk>\n" + " <disk type=\"file\">\n" + " <source file=\"/foo/bar3\"/>\n" + " <driver name=\"qemu\" type=\"qcow2\"/>\n" + " <target bus=\"ide\" dev=\"hdc\"/>\n" + " </disk>\n" + " <disk type=\"block\">\n" + " <source dev=\"/foo/bar4\"/>\n" + " <driver name=\"qemu\" type=\"raw\"/>\n" + " <target bus=\"ide\" dev=\"hdd\"/>\n" + " </disk>\n" + " <disk type=\"file\">\n" + " <source file=\"/foo/bar5\"/>\n" + " <driver name=\"qemu\" type=\"bochs\"/>\n" + " <target bus=\"ide\" dev=\"hde\"/>\n" + " </disk>\n" + " <disk type=\"block\">\n" + " <source dev=\"/foo/bar6\"/>\n" + " <driver name=\"qemu\" type=\"raw\"/>\n" + " <target bus=\"ide\" dev=\"hdf\"/>\n" + " </disk>\n" + " </devices>\n" + "</domain>"; static void test_domain_machine_setup(GVirDesignerDomain **design, gconstpointer opaque) { @@ -142,6 +177,49 @@ static void test_domain_machine_setup(GVirDesignerDomain **design, gconstpointer } +static void test_domain_machine_simple_disk_setup(GVirDesignerDomain **design, gconstpointer opaque) +{ + GError *error = NULL; + GVirConfigDomainDisk *disk; + OsinfoOs *os = osinfo_os_new("http://myoperatingsystem/amazing/4.2"); + OsinfoDb *db = osinfo_db_new(); + OsinfoPlatform *platform = osinfo_platform_new("http://myhypervisor.org/awesome/6.6.6"); + GVirConfigCapabilities *caps = gvir_config_capabilities_new_from_xml(capsqemuxml, NULL); + + *design = gvir_designer_domain_new(db, os, platform, caps); + g_assert(*design); + + disk = gvir_designer_domain_add_cdrom_file(*design, "/foo/bar1", "raw", &error); + g_assert(disk); + g_object_unref(disk); + + disk = gvir_designer_domain_add_cdrom_device(*design, "/foo/bar2", &error); + g_assert(disk); + g_object_unref(disk); + + disk = gvir_designer_domain_add_disk_file(*design, "/foo/bar3", "qcow2", &error); + g_assert(disk); + g_object_unref(disk); + + disk = gvir_designer_domain_add_disk_device(*design, "/foo/bar4", &error); + g_assert(disk); + g_object_unref(disk); + + disk = gvir_designer_domain_add_floppy_file(*design, "/foo/bar5", "bochs", &error); + g_assert(disk); + g_object_unref(disk); + + disk = gvir_designer_domain_add_disk_device(*design, "/foo/bar6", &error); + g_assert(disk); + g_object_unref(disk); + + g_object_unref(os); + g_object_unref(db); + g_object_unref(platform); + g_object_unref(caps); +} + + static void test_domain_container_setup(GVirDesignerDomain **design, gconstpointer opaque) { OsinfoOs *os = osinfo_os_new("http://myoperatingsystem/amazing/4.2"); @@ -243,6 +321,22 @@ static void test_domain_machine_alt_arch_run(GVirDesignerDomain **design, gconst g_object_unref(osconfig); } +static void test_domain_machine_simple_disk_run(GVirDesignerDomain **design, gconstpointer opaque) +{ + GError *error = NULL; + gboolean ret; + GVirConfigDomain *config; + gchar *xml; + + config = gvir_designer_domain_get_config(*design); + xml = gvir_config_object_to_xml(GVIR_CONFIG_OBJECT(config)); + g_test_message("XML %s", xml); + g_assert_cmpstr(xml, + ==, + domain_machine_simple_iso_result); + g_free(xml); +} + static void test_domain_container_host_arch_run(GVirDesignerDomain **design, gconstpointer opaque) { GError *error = NULL; @@ -362,5 +456,12 @@ int main(int argc, char **argv) test_domain_container_setup, test_domain_container_alt_arch_run, test_domain_teardown); + g_test_add("/TestDesignerDomain/SimpleDisk", + GVirDesignerDomain *, + &domain, + test_domain_machine_simple_disk_setup, + test_domain_machine_simple_disk_run, + test_domain_teardown); + return g_test_run(); } -- 2.0.5

2 2

[libvirt] overlayfs support to lxc driver
by Vasiliy Tolstov 04 Mar '15

04 Mar '15

Hello. I'm interesting in implementing overlayfs support to lxc libvirt driver. As i see i simply can utilize filesystem type='template' for this usage. In case os template fs i need to mount with lowerdir /var/lib/libvirt/filesystems/template to /var/lib/libvirt/filesystems/name But i have problem: overlayfs fs need upperdir that store differences and mountpoint that used by clients. In case of libvirt usage - upperdir need to be places in /var/lib/libvirt/filesystems, but where i can put mountpoint dir ? /var/run/libvirt or in /tmp/libvirt-XXXXX ? Also what is the preferred way to integrate overlayfs support to libvirt? I can addd mount type to filesystem type='mount' or use (like now) template type. What is the preferred way? Thanks -- Vasiliy Tolstov, e-mail: v.tolstov(a)selfip.ru jabber: vase(a)selfip.ru

2 4