- Devel - libvirt List Archives

[libvirt] [PATCH v3 1/2] Optimize machine option to set more options with it
by Li Zhang 07 Apr '13

07 Apr '13

From: Li Zhang <zhlcindy(a)linux.vnet.ibm.com> Currently, -machine option is used only when dump-guest-core is set. To use options defined in machine option for newer version of QEMU, it needs to use -machine xxx, and to be compatible with older version -M, this patch addes QEMU_CAPS_MACHINE_OPT capability for newer version which supports -machine option. Signed-off-by: Li Zhang <zhlcindy(a)linux.vnet.ibm.com> --- v3 -> v2: * Set QEMU_CAPS_MACHINE_OPT with help string * Set QEMU_CAPS_DUMP_GUEST_CORE with QMP v2 -> v1: * Split the patch to 2 parts suggested by Daniel P.Berrange * Rename QEMU_CAPS_MACH_OPT to QEMU_CAPS_MACHINE_OPT * Remove version 1.1 assertion for QEMU_CAPS_MACHINE_OPT src/qemu/qemu_capabilities.c | 11 +++++++++++ src/qemu/qemu_capabilities.h | 1 + src/qemu/qemu_command.c | 32 ++++++++++++++++++-------------- tests/qemuhelptest.c | 4 ++++ tests/qemuxml2argvtest.c | 6 +++--- 5 files changed, 37 insertions(+), 17 deletions(-) diff --git a/src/qemu/qemu_capabilities.c b/src/qemu/qemu_capabilities.c index 861d3c4..7d459db 100644 --- a/src/qemu/qemu_capabilities.c +++ b/src/qemu/qemu_capabilities.c @@ -213,6 +213,8 @@ VIR_ENUM_IMPL(virQEMUCaps, QEMU_CAPS_LAST, "virtio-ccw", "dtb", "megasas", + + "machine-opt", /* 135 */ ); struct _virQEMUCaps { @@ -1091,6 +1093,9 @@ virQEMUCapsComputeCmdFlags(const char *help, if (strstr(help, "-dtb")) virQEMUCapsSet(qemuCaps, QEMU_CAPS_DTB); + if (strstr(help, "-machine")) + virQEMUCapsSet(qemuCaps, QEMU_CAPS_MACHINE_OPT); + /* * Handling of -incoming arg with varying features * -incoming tcp (kvm >= 79, qemu >= 0.10.0) @@ -2442,6 +2447,12 @@ virQEMUCapsInitQMP(virQEMUCapsPtr qemuCaps, virQEMUCapsInitQMPBasic(qemuCaps); + /* machine option is supported for newer version */ + virQEMUCapsSet(qemuCaps, QEMU_CAPS_MACHINE_OPT); + + /* -dump-guest-core is supported for newer version*/ + virQEMUCapsSet(qemuCaps, QEMU_CAPS_DUMP_GUEST_CORE); + if (!(archstr = qemuMonitorGetTargetArch(mon))) goto cleanup; diff --git a/src/qemu/qemu_capabilities.h b/src/qemu/qemu_capabilities.h index 7101f67..2595868 100644 --- a/src/qemu/qemu_capabilities.h +++ b/src/qemu/qemu_capabilities.h @@ -174,6 +174,7 @@ enum virQEMUCapsFlags { QEMU_CAPS_VIRTIO_CCW = 132, /* -device virtio-*-ccw */ QEMU_CAPS_DTB = 133, /* -dtb file */ QEMU_CAPS_SCSI_MEGASAS = 134, /* -device megasas */ + QEMU_CAPS_MACHINE_OPT = 135, /* -machine xxxx*/ QEMU_CAPS_LAST, /* this must always be the last item */ }; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index a0c278f..0ee634b 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -5197,6 +5197,8 @@ qemuBuildMachineArgStr(virCommandPtr cmd, const virDomainDefPtr def, virQEMUCapsPtr qemuCaps) { + virBuffer buf = VIR_BUFFER_INITIALIZER; + /* This should *never* be NULL, since we always provide * a machine in the capabilities data for QEMU. So this * check is just here as a safety in case the unexpected @@ -5204,27 +5206,29 @@ qemuBuildMachineArgStr(virCommandPtr cmd, if (!def->os.machine) return 0; - if (!def->mem.dump_core) { + if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_MACHINE_OPT)) { /* if no parameter to the machine type is needed, we still use * '-M' to keep the most of the compatibility with older versions. */ virCommandAddArgList(cmd, "-M", def->os.machine, NULL); } else { - if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_DUMP_GUEST_CORE)) { - virReportError(VIR_ERR_CONFIG_UNSUPPORTED, - "%s", _("dump-guest-core is not available " - " with this QEMU binary")); - return -1; - } - /* However, in case there is a parameter to be added, we need to - * use the "-machine" parameter because qemu is not parsing the - * "-M" correctly */ virCommandAddArg(cmd, "-machine"); - virCommandAddArgFormat(cmd, - "%s,dump-guest-core=%s", - def->os.machine, - virDomainMemDumpTypeToString(def->mem.dump_core)); + virBufferAsprintf(&buf, "%s", def->os.machine); + + if (def->mem.dump_core) { + if (!virQEMUCapsGet(qemuCaps, QEMU_CAPS_DUMP_GUEST_CORE)) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + "%s", _("dump-guest-core is not available " + " with this QEMU binary")); + return -1; + } + + virBufferAsprintf(&buf, ",dump-guest-core=%s", + virDomainMemDumpTypeToString(def->mem.dump_core)); + } + + virCommandAddArg(cmd, virBufferContentAndReset(&buf)); } return 0; diff --git a/tests/qemuhelptest.c b/tests/qemuhelptest.c index a28109a..0c7e607 100644 --- a/tests/qemuhelptest.c +++ b/tests/qemuhelptest.c @@ -718,6 +718,7 @@ mymain(void) QEMU_CAPS_SCSI_LSI, QEMU_CAPS_BLOCKIO, QEMU_CAPS_VNC, + QEMU_CAPS_MACHINE_OPT, QEMU_CAPS_DEVICE_QXL, QEMU_CAPS_DEVICE_VGA, QEMU_CAPS_DEVICE_CIRRUS_VGA, @@ -806,6 +807,7 @@ mymain(void) QEMU_CAPS_VIRTIO_SCSI, QEMU_CAPS_BLOCKIO, QEMU_CAPS_VNC, + QEMU_CAPS_MACHINE_OPT, QEMU_CAPS_DEVICE_QXL, QEMU_CAPS_DEVICE_VGA, QEMU_CAPS_DEVICE_CIRRUS_VGA, @@ -903,6 +905,7 @@ mymain(void) QEMU_CAPS_SECCOMP_SANDBOX, QEMU_CAPS_DUMP_GUEST_CORE, QEMU_CAPS_VNC, + QEMU_CAPS_MACHINE_OPT, QEMU_CAPS_USB_REDIR_BOOTINDEX, QEMU_CAPS_USB_HOST_BOOTINDEX, QEMU_CAPS_DEVICE_QXL, @@ -1009,6 +1012,7 @@ mymain(void) QEMU_CAPS_SECCOMP_SANDBOX, QEMU_CAPS_DUMP_GUEST_CORE, QEMU_CAPS_VNC, + QEMU_CAPS_MACHINE_OPT, QEMU_CAPS_USB_REDIR_BOOTINDEX, QEMU_CAPS_USB_HOST_BOOTINDEX, QEMU_CAPS_DEVICE_QXL, diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 38787ac..4b45c55 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -363,9 +363,9 @@ mymain(void) DO_TEST("minimal-s390", QEMU_CAPS_NAME); DO_TEST("machine-aliases1", NONE); DO_TEST("machine-aliases2", QEMU_CAPS_KVM); - DO_TEST("machine-core-on", QEMU_CAPS_DUMP_GUEST_CORE); - DO_TEST("machine-core-off", QEMU_CAPS_DUMP_GUEST_CORE); - DO_TEST_FAILURE("machine-core-on", NONE); + DO_TEST("machine-core-on", QEMU_CAPS_MACHINE_OPT, QEMU_CAPS_DUMP_GUEST_CORE); + DO_TEST("machine-core-off", QEMU_CAPS_MACHINE_OPT, QEMU_CAPS_DUMP_GUEST_CORE); + DO_TEST_FAILURE("machine-core-on", QEMU_CAPS_MACHINE_OPT, NONE); DO_TEST("boot-cdrom", NONE); DO_TEST("boot-network", NONE); DO_TEST("boot-floppy", NONE); -- 1.7.10.1

4 10

[libvirt] RFC: dhcp <option> element revisited
by Laine Stump 07 Apr '13

07 Apr '13

Before 1.0.3, Pieter Hollants kindly implemented the XML I had earlier suggested to support configure-by-number dhcp options in libvirt networks. https://www.redhat.com/archives/libvir-list/2013-February/msg01251.html I also posted a followup patch for a "force" attribute to options: https://www.redhat.com/archives/libvir-list/2013-February/msg01349.html I actually pushed a slightly fixed version of Pieter's patch, and was looking for an ACK to mine when Eric questioned whether it was all really ready to push since we hadn't implemented *named* options (although there is no official RFC standard for the names of options, dnsmasq uses a pretty good set of lowercase-dash-separated names that make sense, and would certainly make it easier to decode the config). At the time I had just started trying to write such a patch, and was realizing that the XML we'd come up with didn't have any provision to easily deal with an option that took a list of values (multiple IP addresses, multiple domain names, etc); for numbered options we could just punt on this and say "put the multiple values in the single 'value' attribute, separated by commas", but we all know that's a cop-out, so I decided to temporarily pull Pieter's patch to avoid regrets over getting something suboptimal in an official release. That's all the history. Now on to discussing how we should modify these patches to be exactly what we want. I tried digging down into the data of the options to try and make an exhaustive XML representation, with the results below (these would all go directly inside the <dhcp> element of a libvirt network). But it's starting to seem like the deeper I go, the deeper it gets (the classless-static-route option is the current "deepest", and I haven't even attempted to do anything about vendor-specific options), so I'm starting to wonder where to draw the line - see the "STEPPING BACK" section at the bottom. 1) numeric option, single value: <option code='119'> <value type='domain' data='a.example.com'/> </option> (Note that I changed "number" to "code" after seeing that suggestion in [1] because that's what it's called in the RFC). (Also note that the actual encoding of option 119's data is more complicated than described here [2], but it's complicated enough that I think whatever dhcp server implementation is underneath libvirt will need to know the details anyway and decided what we should go for with the "type" attribute is enough for libvirt to validate the input, not necessarily to encode it into a DHCP response packet; any translation from the validated input will be handled internally to the backend driver (currently bridge_driver.c which uses dnsmasq) (I'm actually wondering if type is needed in the XML at all, since the type of any piece of data is always either implicit, or unknown (in which case we just accept any text and pass it through to the backend driver).) 2) numeric option, multiple values: <option code='119' force='yes'> <value type='domain' data='a.example.com'/> <value type='domain' data='b.example.com'/> <value type='domain' data='c.example.com'/> </option> 3) named option, single simple value: <option name='ip-forward-enable'> <value type='boolean' data='yes'/> </option> 4) named option, multiple simple values: <option name='dns-server'> <value type='ipv4Addr' data='10.1.1.1'/> <value type='ipv4Addr' data='10.1.1.2'/> </option> <option (for all named options and known numeric options, the "type" field would be optional, as libvirt will already have an internal table associating option names/code with the type of data, as well as whether or not multiples are allowed) 5) named option, multiple compound values: <option name='classless-static-route'> <value type='compound'> <value type='ipv4Addr' data='1.2.3.0'/> <value type='ipv4Prefix' data='24'/> <value type='ipv4Addr' data='192.168.122.5'/> </value> <value type='compound'> <value type='ipv4Addr' data='1.2.4.0'/> <value type='ipv4Prefix' data='24'/> <value type='ipv4Addr' data='192,.168.122.6'/> </value> </option> (I suppose this could also be done as a flat list, since every static route needs all three) The types I can see that we need (to describe everything in RFC2132 (except vendor-specific options) and the other options I know about) are: ipv4Addr ipv4Prefix text path domain boolean number ANOTHER POSSIBILITY =================== Another way of dealing with options that permit lists of multiple values - just specify the option multiple times, e.g.: <option name='dnsServer' type='ipv4Addr' data='10.1.1.1'/> <option name='dnsServer' type='ipv4Addr' data='10.1.1.2'/> ============================ BTW - another point; I had earlier said that we didn't need to worry about an option number blacklist/whitelist. After looking through the options in RFC 2132, I've changed my mind - many of the option codes (particularly those in the 50's) aren't really general purpose options, but are used by dhcp as integral parts of implementing the exchange sequence between the server and client. So I think what we need to do is have a whitelist that has an entry for every dhcp option we are willing to accept. It would contain: uint8 code; const char *name; /* could be "", in which case we don't recognize a name for it */ unsigned int flags; /* whether or not multiple values are allowed, name recognized by dnsmasq */ /* some representation of type here - what to do about the "compound" type? */ STEPPING BACK ------------- So are we going too far with this? Or should we dial it back to the original proposal in Pieter's patch and just say that 1) if <option> is used, the data is passed through unqualified to the dhcp implementation 2) only "simple", known option types with a single no-repeating value can be specified this way (and data type is implied)(this would be enforced with the "whitelist") 3) then implement separate elements for the complicated/repeating options, e.g.: <route destination='1.2.3.0' prefix='24' nextHop='192.168.122.5'/> <route destination='1.2.4.0' prefix='24' nextHop='192.168.122.6'/> ... <dnsServer address='10.1.1.1'/> <dnsServer address='10.1.1.2'/> Or maybe to simplify the parser this would work as: <option name='classless-static-route' destination='1.2.3.0' prefix='24' nextHop='192.168.122.5'/> <option name='classless-static-route' destination='1.2.4.0' prefix='24' nextHop='192.168.122.6'/> ... <option name='dns-server address='10.1.1.1'/> <option name='dns-server address='10.1.1.2'/> Either of these definitely seems more compact and easy to understand than all that ranting and raving at the top of this message... Any other suggestions? === [1] https://bugzilla.redhat.com/show_bug.cgi?id=666556 [2] https://tools.ietf.org/rfc/rfc3397.txt

2 2

[libvirt] Patch set to add InteractiveContainer support to virt-sandbox
by Dan Walsh 06 Apr '13

06 Apr '13

Resend of patches with lots of cleanups. This patch set adds InteractiveContainer support to virt-sandbox-service command. This is needed if we want to support the OpenShift model of containers. There are also some bug fix patches in the set. [sandbox PATCH 01/15] virt-sandbox-service-util needs to free [sandbox PATCH 02/15] Add support for InteractiveContainers to [sandbox PATCH 03/15] Move virt-sandbox-service bash completion [sandbox PATCH 04/15] Add -u UNITFILE option to virt-sandbox-service [sandbox PATCH 05/15] Change virt-sandbox-service-create.pod to use [sandbox PATCH 06/15] Internationalize all output strings in [sandbox PATCH 07/15] Remove distinction from Internal vs External [sandbox PATCH 08/15] Make CONFIG_PATH external to the Container [sandbox PATCH 09/15] Add exception handler GlibGerror to [sandbox PATCH 10/15] Change variable config to config_path to avoid [sandbox PATCH 11/15] Refactor Container class into Container and [sandbox PATCH 12/15] Add support for InteractiveContainer [sandbox PATCH 13/15] Use args.uri rather then hard coding lxc:/// [sandbox PATCH 14/15] Check for LXC if virt-sandbox-service execute [sandbox PATCH 15/15] Create new /etc/rc.d directory to bind mount

3 31

[libvirt] [PATCH v3] util: Add a helper to check if all bits of a bitmap are clear
by Osier Yang 06 Apr '13

06 Apr '13

--- src/libvirt_private.syms | 1 + src/util/virbitmap.c | 17 +++++++++++++++++ src/util/virbitmap.h | 3 +++ tests/virbitmaptest.c | 33 ++++++++++++++++++++++++++++++++- 4 files changed, 53 insertions(+), 1 deletion(-) diff --git a/src/libvirt_private.syms b/src/libvirt_private.syms index b7b9631..6b831b3 100644 --- a/src/libvirt_private.syms +++ b/src/libvirt_private.syms @@ -1046,6 +1046,7 @@ virBitmapEqual; virBitmapFormat; virBitmapFree; virBitmapGetBit; +virBitmapIsAllClear; virBitmapIsAllSet; virBitmapNew; virBitmapNewCopy; diff --git a/src/util/virbitmap.c b/src/util/virbitmap.c index 21509ac..998c302 100644 --- a/src/util/virbitmap.c +++ b/src/util/virbitmap.c @@ -591,6 +591,23 @@ bool virBitmapIsAllSet(virBitmapPtr bitmap) } /** + * virBitmapIsAllClear: + * @bitmap: the bitmap to check + * + * check if all bits in @bitmap are clear + */ +bool virBitmapIsAllClear(virBitmapPtr bitmap) +{ + int i; + + for (i = 0; i < bitmap->map_len; i++) + if (bitmap->map[i] != 0) + return false; + + return true; +} + +/** * virBitmapNextSetBit: * @bitmap: the bitmap * @pos: the position after which to search for a set bit diff --git a/src/util/virbitmap.h b/src/util/virbitmap.h index 044c7a6..b682523 100644 --- a/src/util/virbitmap.h +++ b/src/util/virbitmap.h @@ -100,6 +100,9 @@ void virBitmapClearAll(virBitmapPtr bitmap) bool virBitmapIsAllSet(virBitmapPtr bitmap) ATTRIBUTE_NONNULL(1); +bool virBitmapIsAllClear(virBitmapPtr bitmap) + ATTRIBUTE_NONNULL(1); + ssize_t virBitmapNextSetBit(virBitmapPtr bitmap, ssize_t pos) ATTRIBUTE_NONNULL(1); diff --git a/tests/virbitmaptest.c b/tests/virbitmaptest.c index 95d010a..3067f6e 100644 --- a/tests/virbitmaptest.c +++ b/tests/virbitmaptest.c @@ -125,6 +125,8 @@ static int test2(const void *data ATTRIBUTE_UNUSED) goto error; virBitmapClearAll(bitmap); + if (!virBitmapIsAllClear(bitmap)) + goto error; if (testBit(bitmap, 0, size - 1, false) < 0) goto error; if (virBitmapCountBits(bitmap) != 0) @@ -154,6 +156,9 @@ static int test3(const void *data ATTRIBUTE_UNUSED) if (!virBitmapIsAllSet(bitmap)) goto error; + virBitmapClearAll(bitmap); + if (!virBitmapIsAllClear(bitmap)) + goto error; ret = 0; error: @@ -196,6 +201,9 @@ static int test4(const void *data ATTRIBUTE_UNUSED) if (virBitmapNextClearBit(bitmap, i) != -1) goto error; + if (!virBitmapIsAllClear(bitmap)) + goto error; + virBitmapFree(bitmap); bitmap = NULL; @@ -406,6 +414,10 @@ static int test7(const void *v ATTRIBUTE_UNUSED) if (!virBitmapIsAllSet(bitmap)) goto error; + virBitmapClearAll(bitmap); + if (!virBitmapIsAllClear(bitmap)) + goto error; + virBitmapFree(bitmap); } @@ -416,6 +428,24 @@ error: return -1; } +static int test8(const void *v ATTRIBUTE_UNUSED) +{ + virBitmapPtr bitmap = NULL; + char data[108] = {0x00,}; + + bitmap = virBitmapNewData(data, sizeof(data)); + if (!bitmap) + goto error; + + if (!virBitmapIsAllClear(bitmap)) + goto error; + + return 0; +error: + virBitmapFree(bitmap); + return -1; +} + static int mymain(void) { @@ -435,7 +465,8 @@ mymain(void) ret = -1; if (virtTestRun("test7", 1, test7, NULL) < 0) ret = -1; - + if (virtTestRun("test8", 1, test8, NULL) < 0) + ret = -1; return ret; } -- 1.8.1.4

2 2

[libvirt] [PATCH v2] qemu: Support multiple queue virtio-scsi
by Osier Yang 06 Apr '13

06 Apr '13

This introduce a new attribute "num_queues" (same with the good name QEMU uses) for virtio-scsi controller. An example of the XML: <controller type='scsi' index='0' model='virtio-scsi' num_queues='8'/> The corresponding QEMU command line: -device virtio-scsi-pci,id=scsi0,num_queues=8,bus=pci.0,addr=0x3 \ --- docs/formatdomain.html.in | 18 +++++++++------ docs/schemas/domaincommon.rng | 5 +++++ src/conf/domain_conf.c | 13 +++++++++++ src/conf/domain_conf.h | 1 + src/qemu/qemu_command.c | 11 +++++++++ .../qemuxml2argv-disk-virtio-scsi-num_queues.args | 8 +++++++ .../qemuxml2argv-disk-virtio-scsi-num_queues.xml | 26 ++++++++++++++++++++++ tests/qemuxml2argvtest.c | 3 +++ tests/qemuxml2xmltest.c | 1 + 9 files changed, 79 insertions(+), 7 deletions(-) create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-num_queues.args create mode 100644 tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-num_queues.xml diff --git a/docs/formatdomain.html.in b/docs/formatdomain.html.in index cf382e8..2973ce2 100644 --- a/docs/formatdomain.html.in +++ b/docs/formatdomain.html.in @@ -2092,13 +2092,17 @@ controller. A "scsi" controller has an optional attribute <code>model</code>, which is one of "auto", "buslogic", "ibmvscsi", "lsilogic", "lsisas1068", "lsisas1078", "virtio-scsi" or - "vmpvscsi". A "usb" controller has an optional attribute - <code>model</code>, which is one of "piix3-uhci", "piix4-uhci", "ehci", - "ich9-ehci1", "ich9-uhci1", "ich9-uhci2", "ich9-uhci3", - "vt82c686b-uhci", "pci-ohci" or "nec-xhci". Additionally, - since 0.10.0, if the USB bus needs to be - explicitly disabled for the guest, <code>model='none'</code> may be used. - The PowerPC64 "spapr-vio" addresses do not have an associated controller. + "vmpvscsi". The attribute <code>num_queues</code> + (1.0.5 (QEMU and KVM only)) is to specify + the number of queues for the controller, it's recommended to specify + the value matching the number of vCPUs for best performance. A "usb" + controller has an optional attribute <code>model</code>, which is one + of "piix3-uhci", "piix4-uhci", "ehci", "ich9-ehci1", "ich9-uhci1", + "ich9-uhci2", "ich9-uhci3", "vt82c686b-uhci", "pci-ohci" or "nec-xhci". + Additionally, since 0.10.0, if the USB bus + needs to be explicitly disabled for the guest, <code>model='none'</code> + may be used. The PowerPC64 "spapr-vio" addresses do not have an + associated controller. diff --git a/docs/schemas/domaincommon.rng b/docs/schemas/domaincommon.rng index 63ba7d1..d486ae8 100644 --- a/docs/schemas/domaincommon.rng +++ b/docs/schemas/domaincommon.rng @@ -1427,6 +1427,11 @@ </attribute> </optional> <optional> + <attribute name="num_queues"> + <ref name="unsignedInt"/> + </attribute> + </optional> + <optional> <ref name="usbmaster"/> </optional> <optional> diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index 5ad3080..f50a964 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -5034,6 +5034,7 @@ virDomainControllerDefParseXML(xmlNodePtr node, char *type = NULL; char *idx = NULL; char *model = NULL; + char *num_queues = NULL; if (VIR_ALLOC(def) < 0) { virReportOOMError(); @@ -5069,6 +5070,14 @@ virDomainControllerDefParseXML(xmlNodePtr node, def->model = -1; } + if ((num_queues = virXMLPropString(node, "num_queues"))) { + if (virStrToLong_ui(num_queues, NULL, 10, &def->num_queues) < 0) { + virReportError(VIR_ERR_XML_ERROR, + _("Malformed 'num_queues' value '%s'"), num_queues); + goto error; + } + } + if (virDomainDeviceInfoParseXML(node, NULL, &def->info, flags) < 0) goto error; @@ -5146,6 +5155,7 @@ cleanup: VIR_FREE(type); VIR_FREE(idx); VIR_FREE(model); + VIR_FREE(num_queues); return def; @@ -13195,6 +13205,9 @@ virDomainControllerDefFormat(virBufferPtr buf, virBufferEscapeString(buf, " model='%s'", model); } + if (def->num_queues) + virBufferAsprintf(buf, " num_queues='%u'", def->num_queues); + switch (def->type) { case VIR_DOMAIN_CONTROLLER_TYPE_VIRTIO_SERIAL: if (def->opts.vioserial.ports != -1) { diff --git a/src/conf/domain_conf.h b/src/conf/domain_conf.h index 700f03f..75e3f15 100644 --- a/src/conf/domain_conf.h +++ b/src/conf/domain_conf.h @@ -728,6 +728,7 @@ struct _virDomainControllerDef { int type; int idx; int model; /* -1 == undef */ + unsigned int num_queues; union { virDomainVirtioSerialOpts vioserial; } opts; diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 493e5f8..53caba5 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -3533,6 +3533,14 @@ qemuBuildControllerDevStr(virDomainDefPtr domainDef, virBuffer buf = VIR_BUFFER_INITIALIZER; int model; + if (def->num_queues && + !(def->type == VIR_DOMAIN_CONTROLLER_TYPE_SCSI && + def->model == VIR_DOMAIN_CONTROLLER_MODEL_SCSI_VIRTIO_SCSI)) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("'num_queues' is only supported by virtio-scsi controller")); + return NULL; + } + switch (def->type) { case VIR_DOMAIN_CONTROLLER_TYPE_SCSI: model = def->model; @@ -3616,6 +3624,9 @@ qemuBuildControllerDevStr(virDomainDefPtr domainDef, goto error; } + if (def->num_queues) + virBufferAsprintf(&buf, ",num_queues=%u", def->num_queues); + if (qemuBuildDeviceAddressStr(&buf, &def->info, qemuCaps) < 0) goto error; diff --git a/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-num_queues.args b/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-num_queues.args new file mode 100644 index 0000000..8d46799 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-num_queues.args @@ -0,0 +1,8 @@ +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test \ +/usr/bin/qemu -S -M pc -m 214 -smp 8 -nographic -nodefconfig -nodefaults \ +-monitor unix:/tmp/test-monitor,server,nowait -no-acpi -boot c \ +-device virtio-scsi-pci,id=scsi0,num_queues=8,bus=pci.0,addr=0x3 \ +-usb \ +-drive file=/dev/HostVG/QEMUGuest1,if=none,id=drive-scsi0-0-0-0 \ +-device scsi-disk,bus=scsi0.0,channel=0,scsi-id=0,lun=0,drive=drive-scsi0-0-0-0,id=scsi0-0-0-0 \ +-device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x4 diff --git a/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-num_queues.xml b/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-num_queues.xml new file mode 100644 index 0000000..dfa9cf1 --- /dev/null +++ b/tests/qemuxml2argvdata/qemuxml2argv-disk-virtio-scsi-num_queues.xml @@ -0,0 +1,26 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory unit='KiB'>219136</memory> + <currentMemory unit='KiB'>219136</currentMemory> + <vcpu placement='static'>8</vcpu> + <os> + <type arch='i686' machine='pc'>hvm</type> + <boot dev='hd'/> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/bin/qemu</emulator> + <disk type='block' device='disk'> + <source dev='/dev/HostVG/QEMUGuest1'/> + <target dev='sdb' bus='scsi'/> + <address type='drive' controller='0' bus='0' target='0' unit='0'/> + </disk> + <controller type='usb' index='0'/> + <controller type='scsi' index='0' model='virtio-scsi' num_queues='8'/> + <memballoon model='virtio'/> + </devices> +</domain> diff --git a/tests/qemuxml2argvtest.c b/tests/qemuxml2argvtest.c index 099fb36..d6575e7 100644 --- a/tests/qemuxml2argvtest.c +++ b/tests/qemuxml2argvtest.c @@ -553,6 +553,9 @@ mymain(void) DO_TEST("disk-scsi-virtio-scsi", QEMU_CAPS_DRIVE, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG, QEMU_CAPS_VIRTIO_SCSI); + DO_TEST("disk-virtio-scsi-num_queues", + QEMU_CAPS_DRIVE, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG, + QEMU_CAPS_VIRTIO_SCSI); DO_TEST("disk-scsi-megasas", QEMU_CAPS_DRIVE, QEMU_CAPS_DEVICE, QEMU_CAPS_NODEFCONFIG, QEMU_CAPS_SCSI_MEGASAS); diff --git a/tests/qemuxml2xmltest.c b/tests/qemuxml2xmltest.c index aa058bd..10a6ea2 100644 --- a/tests/qemuxml2xmltest.c +++ b/tests/qemuxml2xmltest.c @@ -179,6 +179,7 @@ mymain(void) DO_TEST("disk-scsi-device"); DO_TEST("disk-scsi-vscsi"); DO_TEST("disk-scsi-virtio-scsi"); + DO_TEST("disk-virtio-scsi-num_queues"); DO_TEST("disk-scsi-megasas"); DO_TEST_FULL("disk-mirror", false, WHEN_ACTIVE); DO_TEST_FULL("disk-mirror", true, WHEN_INACTIVE); -- 1.8.1.4

2 2

[libvirt] [PATCH 0/7] Fix multiple compiler warnings on ARMv7
by Daniel P. Berrange 06 Apr '13

06 Apr '13

The ARMv7 builds of libvirt generate a number of warnings, mostly about cast alignment http://arm.koji.fedoraproject.org/packages/libvirt/1.0.4/1.fc19/data/logs/a… this patch series fixes as many as possible, and then disables warnings for the rest using a pragma

3 17

[libvirt] [PATCH 0/3] Prohibit passing all clear cpumap for vcpupin/emulatorpin
by Osier Yang 06 Apr '13

06 Apr '13

*** NONE *** Osier Yang (3): util: Add a helper to check if all bits of a bitmap are clear qemu: Error out if the bitmap for pinning is all clear virsh: Prohibit all clear cpumap src/libvirt_private.syms | 1 + src/qemu/qemu_driver.c | 12 ++++++++++++ src/util/virbitmap.c | 30 ++++++++++++++++++++++++++++++ src/util/virbitmap.h | 3 +++ tools/virsh-domain.c | 35 ++++++++++++++++++++++++++++++----- 5 files changed, 76 insertions(+), 5 deletions(-) -- 1.8.1.4

4 15

[libvirt] [PATCH] build: check correct protocol.o file
by Eric Blake 06 Apr '13

06 Apr '13

By default, libtool builds two .o files for every .lo rule: src/foo.o - static builds src/.libs/foo.o - shared library builds But since commit ad42b34b disabled static builds, src/foo.o is no longer built by default. On a fresh checkout, this means our protocol check rules using pdwtags were testing a missing file, and thanks a lousy behavior of pdwtags happily giving no output and 0 exit status (http://bugzilla.redhat.com/949034) we were merely claiming that "dwarves is too old" and skipping the test. However, if you swap between branches and do incremental builds, such as building v0.10.2-maint and then switching back to master, you end up with src/foo.o being leftover from its 0.10.2 state, and then 'make check' fails because the .o file does not match the protocol-structs file due to API additions in the meantime. A simpler fix would be to always look in .libs for the .o to be parsed; but since it is possible to pass ./configure options to tell libtool to do a static-only build with no shared .o, I went with the approach of finding the newest of the two files, whenever both exist. * src/Makefile.am (PDWTAGS): Ensure we test just-built file. --- Pushing under the build-breaker rule. src/Makefile.am | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/src/Makefile.am b/src/Makefile.am index 78b4ab6..5dc58f9 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -341,9 +341,16 @@ r1 = /\* \d+ \*/ r2 = /\* <[[:xdigit:]]+> \S+:\d+ \*/ struct_prefix = (remote_|qemu_|lxc_|virNet|keepalive_) +# Depending on configure options, libtool creates one or both of +# {,.libs/}libvirt_remote_driver_la-remote_protocol.o. We want the +# newest of the two, in case configure options changed and a stale +# file is left around from an earlier build. PDWTAGS = \ $(AM_V_GEN)if (pdwtags --help) > /dev/null 2>&1; then \ - pdwtags --verbose $(<:.lo=.$(OBJEXT)) > $(@F)-t1 2> $(@F)-t2; \ + file=`ls -t $(<:.lo=.$(OBJEXT)) .libs/$(<:.lo=.$(OBJEXT)) \ + 2>/dev/null | sed -n 1p`; \ + test -f $$file || { echo ".o for $< not found" >&2; exit 1; };\ + pdwtags --verbose $$file > $(@F)-t1 2> $(@F)-t2; \ if test -s $(@F)-t2; then \ rm -rf $(@F)-t?; \ echo 'WARNING: pdwtags appears broken; skipping the $@ test' >&2;\ -- 1.8.1.4

1 0

[libvirt] [PATCH] Add a test suite for keycode mapping functions
by Daniel P. Berrange 06 Apr '13

06 Apr '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> Validate that translations between different keycode sets are functioning. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- tests/virkeycodetest.c | 107 +++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 107 insertions(+) create mode 100644 tests/virkeycodetest.c diff --git a/tests/virkeycodetest.c b/tests/virkeycodetest.c new file mode 100644 index 0000000..d3c5975 --- /dev/null +++ b/tests/virkeycodetest.c @@ -0,0 +1,107 @@ +/* + * Copyright (C) 2013 Red Hat, Inc. + * + * This library is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with this library; If not, see + * <http://www.gnu.org/licenses/>. + * + * Author: Daniel P. Berrange <berrange(a)redhat.com> + */ + +#include <config.h> + +#include <stdlib.h> + +#include "testutils.h" + +#include "virkeycode.h" +#include "virutil.h" +#include "virerror.h" +#include "viralloc.h" +#include "virlog.h" + +#include "virlockspace.h" + +#define VIR_FROM_THIS VIR_FROM_NONE + + +static int testKeycodeMapping(const void *data ATTRIBUTE_UNUSED) +{ + int ret = -1; + int got; + +#define TRANSLATE(from, to, val, want) \ + do { \ + if ((got = virKeycodeValueTranslate(VIR_KEYCODE_SET_##from, \ + VIR_KEYCODE_SET_##to, \ + val)) != want) { \ + fprintf(stderr, "Translating %d from %s to %s, got %d want %d\n", \ + val, #from, #to, got, want); \ + goto cleanup; \ + } \ + } while (0) + + TRANSLATE(LINUX, LINUX, 111, 111); + TRANSLATE(LINUX, USB, 111, 76); + TRANSLATE(LINUX, RFB, 88, 88); + TRANSLATE(LINUX, RFB, 160, 163); + TRANSLATE(ATSET2, ATSET3, 259, 55); + +#undef TRANSLATE + + ret = 0; +cleanup: + return ret; +} + + +static int testKeycodeStrings(const void *data ATTRIBUTE_UNUSED) +{ + int ret = -1; + int got; + +#define TRANSLATE(from, str, want) \ + do { \ + if ((got = virKeycodeValueFromString(VIR_KEYCODE_SET_##from, \ + str)) != want) { \ + fprintf(stderr, "Converting %s from %s, got %d want %d\n", \ + str, #from, got, want); \ + goto cleanup; \ + } \ + } while (0) + + TRANSLATE(LINUX, "KEY_DELETE", 111); + TRANSLATE(OSX, "Function", 0x3f); + TRANSLATE(WIN32, "VK_UP", 0x26); + +#undef TRANSLATE + + ret = 0; +cleanup: + return ret; +} + +static int +mymain(void) +{ + int ret = 0; + + if (virtTestRun("Keycode mapping ", 1, testKeycodeMapping, NULL) < 0) + ret = -1; + if (virtTestRun("Keycode strings ", 1, testKeycodeStrings, NULL) < 0) + ret = -1; + + return ret==0 ? EXIT_SUCCESS : EXIT_FAILURE; +} + +VIRT_TEST_MAIN(mymain) -- 1.8.1.4

2 2

[libvirt] [PATCH] build: use proper pod for nested bulleted VIRSH_DEBUG list
by Eric Blake 06 Apr '13

06 Apr '13

Newer pod (hello rawhide) complains if you attempt to mix bullets and non-bullets in the same list: virsh.pod around line 3177: Expected text after =item, not a bullet As our intent was to nest an inner list, we make that explicit to keep pod happy. * tools/virsh.pod (ENVIRONMENT): Use correct pod syntax. --- Pushing under the build-breaker rule. tools/virsh.pod | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/virsh.pod b/tools/virsh.pod index 1c8f9ee..89467f6 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -3174,6 +3174,8 @@ of C<virsh> Turn on verbose debugging of virsh commands. Valid levels are +=over 4 + =item * VIRSH_DEBUG=0 DEBUG - Messages at ALL levels get logged @@ -3194,6 +3196,8 @@ WARNING - Logs messages at levels WARNING and ERROR ERROR - Messages at only ERROR level gets logged. +=back + =item VIRSH_LOG_FILE=C<LOGFILE> The file to log virsh debug messages. -- 1.8.1.4

1 0

[libvirt] [PATCH] qemu: Don't require a block or file when looking for an alias.
by Wido den Hollander 05 Apr '13

05 Apr '13

From: root <root(a)kvm01.(none)> This for example prohibits you to use iotune for Ceph or Sheepdog devices. Signed-off-by: Wido den Hollander <wido(a)widodh.nl> --- src/qemu/qemu_driver.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 552a81b..464d30a 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -13032,10 +13032,6 @@ qemuDiskPathToAlias(virDomainObjPtr vm, const char *path, int *idx) if (idx) *idx = i; - if (disk->type != VIR_DOMAIN_DISK_TYPE_BLOCK && - disk->type != VIR_DOMAIN_DISK_TYPE_FILE) - goto cleanup; - if (disk->src) { if (virAsprintf(&ret, "drive-%s", disk->info.alias) < 0) { virReportOOMError(); -- 1.7.9.5

2 2

[libvirt] [PATCH] Make guest OS bootable when hardware failure happens in log disk
by Seiji Aguchi 05 Apr '13

05 Apr '13

[Problem] Currently, guest OS's messages can be logged to a local disk of host OS by creating chadevs with options below. -chardev file,id=charserial0,path=<log file's path> -device isa-serial,chardev=chardevserial0,id=serial0 When a hardware failure happens in the disk, qemu-kvm can't create the chardevs. In this case, guest OS doesn't boot up. Actually, there are users who don't desire that guest OS goes down due to a hardware failure of a log disk only.Therefore, libvirt should offer some way to boot guest OS up even if the log disk is broken. [Solution] This patch changes a destination to /dev/null in case a log file can't be opened. Signed-off-by: Seiji Aguchi <seiji.aguchi(a)hds.com> --- src/qemu/qemu_process.c | 20 +++++++++++++++++++- 1 files changed, 19 insertions(+), 1 deletions(-) diff --git a/src/qemu/qemu_process.c b/src/qemu/qemu_process.c index 8c4bfb7..fdf26ad 100644 --- a/src/qemu/qemu_process.c +++ b/src/qemu/qemu_process.c @@ -2440,7 +2440,25 @@ qemuProcessPrepareChardevDevice(virDomainDefPtr def ATTRIBUTE_UNUSED, virReportSystemError(errno, _("Unable to pre-create chardev file '%s'"), dev->source.data.file.path); - return -1; + VIR_FREE(dev->source.data.file.path); + /* + * Change a destination to /dev/null to boot guest OS up + * even if a log disk is broken. + */ + dev->source.data.file.path = strdup("/dev/null"); + + if (!dev->source.data.file.path) { + virReportOOMError(); + return -1; + } + + if ((fd = open(dev->source.data.file.path, + O_CREAT | O_APPEND, S_IRUSR|S_IWUSR)) < 0) { + virReportSystemError(errno, + _("Unable to pre-create chardev file '%s'"), + dev->source.data.file.path); + return -1; + } } VIR_FORCE_CLOSE(fd); -- 1.7.1

2 2

[libvirt] [PATCH] qemu: Remove maximum cpu limit when setting processor count using the API
by Peter Krempa 05 Apr '13

05 Apr '13

When setting processor count for a domain using the API libvirt enforced a maximum processor count, while it isn't enforced when taking the XML path. This patch removes the check to match the API. --- src/qemu/qemu_driver.c | 24 +++--------------------- 1 file changed, 3 insertions(+), 21 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 3f62414..833bf7f 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -3737,8 +3737,6 @@ qemuDomainSetVcpusFlags(virDomainPtr dom, unsigned int nvcpus, virQEMUDriverPtr driver = dom->conn->privateData; virDomainObjPtr vm = NULL; virDomainDefPtr persistentDef; - const char * type; - int max; int ret = -1; bool maximum; virQEMUDriverConfigPtr cfg = NULL; @@ -3778,27 +3776,11 @@ qemuDomainSetVcpusFlags(virDomainPtr dom, unsigned int nvcpus, goto endjob; } - if (!(type = virDomainVirtTypeToString(vm->def->virtType))) { - virReportError(VIR_ERR_INTERNAL_ERROR, - _("unknown virt type in domain definition '%d'"), - vm->def->virtType); - goto endjob; - } - - if ((max = qemuGetMaxVCPUs(NULL, type)) < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, "%s", - _("could not determine max vcpus for the domain")); - goto endjob; - } - - if (!maximum && vm->def->maxvcpus < max) { - max = vm->def->maxvcpus; - } - - if (nvcpus > max) { + if (!maximum && nvcpus > vm->def->maxvcpus) { virReportError(VIR_ERR_INVALID_ARG, _("requested vcpus is greater than max allowable" - " vcpus for the domain: %d > %d"), nvcpus, max); + " vcpus for the domain: %d > %d"), + nvcpus, vm->def->maxvcpus); goto endjob; } -- 1.8.1.5

2 4

[libvirt] [PATCH 0/5] Drop virReportOOMReport from almost everywhere
by Michal Privoznik 05 Apr '13

05 Apr '13

3 13

[libvirt] [PATCH v2 2/2] Implement support for <hostdev caps=net>
by Bogdan Purcareata 05 Apr '13

05 Apr '13

This allows a container-type domain to have exclusive access to one of the host's NICs. Wire <hostdev caps=net> with the lxc_controller - when moving the newly created veth devices into a new namespace, also look for any hostdev devices that should be moved. Note: once the container domain has been destroyed, there is no code that moves the interfaces back to the original namespace. This does happen, though, probably due to default cleanup on namespace destruction. Signed-off-by: Bogdan Purcareata <bogdan.purcareata(a)freescale.com> --- src/lxc/lxc_container.c | 4 +++- src/lxc/lxc_controller.c | 16 ++++++++++++++++ src/lxc/lxc_hostdev.c | 1 + 3 files changed, 20 insertions(+), 1 deletion(-) diff --git a/src/lxc/lxc_container.c b/src/lxc/lxc_container.c index 002ba9e..e59bfdf 100644 --- a/src/lxc/lxc_container.c +++ b/src/lxc/lxc_container.c @@ -1551,7 +1551,6 @@ cleanup: return ret; } - static int lxcContainerSetupHostdevSubsys(virDomainDefPtr vmDef, virDomainHostdevDefPtr def, const char *dstprefix, @@ -1582,6 +1581,9 @@ static int lxcContainerSetupHostdevCaps(virDomainDefPtr vmDef, case VIR_DOMAIN_HOSTDEV_CAPS_TYPE_MISC: return lxcContainerSetupHostdevCapsMisc(vmDef, def, dstprefix, securityDriver); + case VIR_DOMAIN_HOSTDEV_CAPS_TYPE_NET: + return 0; // case is handled in virLXCControllerMoveInterfaces + default: virReportError(VIR_ERR_CONFIG_UNSUPPORTED, _("Unsupported host device mode %s"), diff --git a/src/lxc/lxc_controller.c b/src/lxc/lxc_controller.c index cede445..edd99bf 100644 --- a/src/lxc/lxc_controller.c +++ b/src/lxc/lxc_controller.c @@ -1050,12 +1050,28 @@ cleanup2: static int virLXCControllerMoveInterfaces(virLXCControllerPtr ctrl) { size_t i; + virDomainDefPtr def = ctrl->def; for (i = 0 ; i < ctrl->nveths ; i++) { if (virNetDevSetNamespace(ctrl->veths[i], ctrl->initpid) < 0) return -1; } + for (i = 0; i < def->nhostdevs; i ++) { + virDomainHostdevDefPtr hdev = def->hostdevs[i]; + + if (hdev->mode != VIR_DOMAIN_HOSTDEV_MODE_CAPABILITIES) + continue; + + virDomainHostdevCaps hdcaps = hdev->source.caps; + + if (hdcaps.type != VIR_DOMAIN_HOSTDEV_CAPS_TYPE_NET) + continue; + + if (virNetDevSetNamespace(hdcaps.u.net.iface, ctrl->initpid) < 0) + return -1; + } + return 0; } diff --git a/src/lxc/lxc_hostdev.c b/src/lxc/lxc_hostdev.c index 33b0b60..53a1a31 100644 --- a/src/lxc/lxc_hostdev.c +++ b/src/lxc/lxc_hostdev.c @@ -307,6 +307,7 @@ int virLXCPrepareHostDevices(virLXCDriverPtr driver, switch (dev->source.subsys.type) { case VIR_DOMAIN_HOSTDEV_CAPS_TYPE_STORAGE: case VIR_DOMAIN_HOSTDEV_CAPS_TYPE_MISC: + case VIR_DOMAIN_HOSTDEV_CAPS_TYPE_NET: break; default: virReportError(VIR_ERR_CONFIG_UNSUPPORTED, -- 1.7.11.7

1 0

[libvirt] libvirt_lxc: implement <hostdev caps=net> device isolation in containers
by Bogdan Purcareata 05 Apr '13

05 Apr '13

This set of patches implements <hostdev caps=net> interface isolation in containers, thus allowing an interface NIC to be assigned exclusively to a container-domain. This is done like moving veth devices in container namespaces, only this time it is actual host devices.

2 4

[libvirt] [PATCH v5.1 00/11] Introduce driver specific XML parsing callbacks
by Peter Krempa 05 Apr '13

05 Apr '13

In this respin the renaming patch is moved to be first in the series. The argument rearrangement patch was too complicated to move. Unfortunately this rebase operation caused my "git notes" to be lost for the individual patches, so please see v5 for more specific comments. Patches 5 to 11 except for 10 were ACKed in previous versions. Peter Krempa (11): maint: Rename xmlconf to xmlopt and virDomainXMLConfing to virDomainXMLOption conf: Add post XML parse callbacks and prepare for cleaning of virCaps conf callback: Rearrange function parameters qemu: Record the default NIC model in the domain XML virCaps: get rid of "defaultInitPath" value in the virCaps struct virCaps: get rid of defaultDiskDriverName virCaps: get rid of emulatorRequired virCaps: get rid of defaultDiskDriverType virCaps: get rid of hasWideScsiBus virCaps: get rid of macPrefix field virCaps: get rid of defaultConsoleTargetType callback src/conf/capabilities.c | 24 - src/conf/capabilities.h | 23 - src/conf/domain_conf.c | 660 ++++++++++++--------- src/conf/domain_conf.h | 102 ++-- src/conf/snapshot_conf.c | 4 +- src/conf/snapshot_conf.h | 2 +- src/esx/esx_driver.c | 32 +- src/esx/esx_private.h | 2 +- src/libvirt_private.syms | 10 +- src/libvirt_vmx.syms | 1 + src/libxl/libxl_conf.c | 13 - src/libxl/libxl_conf.h | 2 +- src/libxl/libxl_driver.c | 103 ++-- src/lxc/lxc_conf.c | 21 +- src/lxc/lxc_conf.h | 4 +- src/lxc/lxc_controller.c | 10 +- src/lxc/lxc_domain.c | 34 ++ src/lxc/lxc_domain.h | 1 + src/lxc/lxc_driver.c | 71 ++- src/lxc/lxc_process.c | 8 +- src/openvz/openvz_conf.c | 16 +- src/openvz/openvz_conf.h | 2 +- src/openvz/openvz_driver.c | 64 +- src/parallels/parallels_driver.c | 37 +- src/parallels/parallels_utils.h | 2 +- src/phyp/phyp_driver.c | 24 +- src/phyp/phyp_driver.h | 2 +- src/qemu/qemu_capabilities.c | 19 - src/qemu/qemu_command.c | 27 +- src/qemu/qemu_command.h | 6 +- src/qemu/qemu_conf.c | 18 +- src/qemu/qemu_conf.h | 5 +- src/qemu/qemu_domain.c | 98 ++- src/qemu/qemu_domain.h | 1 + src/qemu/qemu_driver.c | 147 +++-- src/qemu/qemu_migration.c | 21 +- src/qemu/qemu_process.c | 34 +- src/security/virt-aa-helper.c | 19 +- src/test/test_driver.c | 67 +-- src/uml/uml_conf.c | 9 - src/uml/uml_conf.h | 2 +- src/uml/uml_driver.c | 69 ++- src/vbox/vbox_tmpl.c | 33 +- src/vmware/vmware_conf.c | 21 +- src/vmware/vmware_conf.h | 2 +- src/vmware/vmware_driver.c | 24 +- src/vmx/vmx.c | 39 +- src/vmx/vmx.h | 12 +- src/xen/xen_driver.c | 37 +- src/xen/xen_driver.h | 4 +- src/xen/xen_hypervisor.c | 13 - src/xen/xend_internal.c | 20 +- src/xen/xm_internal.c | 16 +- src/xenapi/xenapi_driver.c | 40 +- src/xenapi/xenapi_driver_private.h | 2 +- tests/domainsnapshotxml2xmltest.c | 6 +- tests/lxcxml2xmldata/lxc-hostdev.xml | 1 + tests/lxcxml2xmldata/lxc-systemd.xml | 1 + tests/lxcxml2xmltest.c | 8 +- tests/qemuargv2xmltest.c | 6 +- tests/qemumonitorjsontest.c | 28 +- tests/qemumonitortestutils.c | 4 +- tests/qemumonitortestutils.h | 2 +- ...qemuxml2argv-disk-drive-network-nbd-export.args | 3 +- .../qemuxml2argv-disk-drive-network-nbd-export.xml | 1 + ...ml2argv-disk-drive-network-nbd-ipv6-export.args | 3 +- ...xml2argv-disk-drive-network-nbd-ipv6-export.xml | 1 + .../qemuxml2argv-disk-drive-network-nbd-ipv6.args | 3 +- .../qemuxml2argv-disk-drive-network-nbd-ipv6.xml | 1 + .../qemuxml2argv-disk-drive-network-nbd-unix.args | 3 +- .../qemuxml2argv-disk-drive-network-nbd-unix.xml | 1 + .../qemuxml2argv-disk-drive-network-nbd.args | 5 +- .../qemuxml2argv-disk-drive-network-nbd.xml | 1 + .../qemuxml2argv-disk-drive-network-rbd-auth.args | 2 +- .../qemuxml2argv-disk-drive-network-rbd-ipv6.args | 2 +- .../qemuxml2argv-disk-drive-network-rbd-ipv6.xml | 1 + .../qemuxml2argv-disk-drive-network-rbd.args | 2 +- .../qemuxml2argv-disk-drive-network-rbd.xml | 1 + .../qemuxml2argv-disk-drive-network-sheepdog.args | 3 +- .../qemuxml2argv-disk-drive-network-sheepdog.xml | 1 + .../qemuxml2argv-net-bandwidth.xml | 1 + .../qemuxml2argvdata/qemuxml2argv-net-client.args | 4 +- .../qemuxml2argv-net-eth-ifname.args | 4 +- .../qemuxml2argv-net-eth-ifname.xml | 1 + .../qemuxml2argv-net-eth-names.args | 8 +- tests/qemuxml2argvdata/qemuxml2argv-net-eth.args | 4 +- tests/qemuxml2argvdata/qemuxml2argv-net-eth.xml | 1 + .../qemuxml2argvdata/qemuxml2argv-net-hostdev.xml | 1 + tests/qemuxml2argvdata/qemuxml2argv-net-mcast.args | 4 +- .../qemuxml2argv-net-openvswitch.xml | 1 + .../qemuxml2argvdata/qemuxml2argv-net-server.args | 4 +- tests/qemuxml2argvdata/qemuxml2argv-net-user.args | 3 +- tests/qemuxml2argvdata/qemuxml2argv-net-user.xml | 1 + .../qemuxml2argv-net-virtio-network-portgroup.xml | 2 + tests/qemuxml2argvtest.c | 6 +- .../qemuxml2xmlout-graphics-spice-timeout.xml | 1 + tests/qemuxml2xmltest.c | 6 +- tests/qemuxmlnstest.c | 6 +- tests/securityselinuxlabeltest.c | 6 +- tests/testutilslxc.c | 9 - tests/testutilsqemu.c | 11 - tests/testutilsqemu.h | 2 +- tests/testutilsxen.c | 16 - tests/testutilsxen.h | 2 - tests/vmx2xmltest.c | 17 +- tests/xmconfigtest.c | 8 +- tests/xml2sexprtest.c | 9 +- tests/xml2vmxtest.c | 20 +- 108 files changed, 1275 insertions(+), 1052 deletions(-) -- 1.8.1.5

5 32

[libvirt] Device backend in another domain (xen)
by Marek Marczykowski 05 Apr '13

05 Apr '13

Hi all, Xen have (AFAIK unique) feature of having backend device in other domU (sometimes called driver domain) instead of dom0. Currently libxl support it for disk and network devices, but in general it is valid for any device (perhaps excluding PCI). Current implementation requires ID of backend domain to be specified during device attach/domain startup. There are some upcoming patches to allow also use domain name. I'd like to add support for it into libvirt. The question is how specify domain reference. The simplest approach is to use direct domain ID, logically it fits in <source/> tag, so something like: <disk type='file'> <source file='/path/to/file/inside/domain/7' domid='7'/> <target dev='xvdb'/> </disk> <interface type='bridge'> <source bridge='name-of-bridge-in-domiain-7' domid='7'/> </interface> This have obvious limitation that domain ID changes with each domain restart. But perhaps more correct approach is to use name or UUID? Extending <source/> tag is right thing to do? Or should I add a new tag (<backenddomain uuid='...'/> ? )? Regarding implementation I need some hints how to do the value verification (check if domain exists at least). If it has to be done in domain_conf.c (during XML parsing) it will require predetermined domain load order, which IMHO isn't acceptable. So perhaps the check should be done at domain startup/device attach (by driver)? In any case: is it possible at all to check if domain with given name/UUID exists in scope of domain_conf.c? It looks like it requires connection reference, which isn't passed to virDomain*DefParseXML. -- Best Regards, Marek Marczykowski Invisible Things Lab

2 1

[libvirt] [PATCH 1/3] Remove support for old kernels lacking private devpts
by Daniel P. Berrange 05 Apr '13

05 Apr '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> Early on kernel support for private devpts was not widespread, so we had compatibiltiy codepaths. Such old kernels are not seriously used for LXC these days, so the compat code can go away Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/lxc/lxc_controller.c | 28 +++++++++------------------- 1 file changed, 9 insertions(+), 19 deletions(-) diff --git a/src/lxc/lxc_controller.c b/src/lxc/lxc_controller.c index 1d1443c..cede445 100644 --- a/src/lxc/lxc_controller.c +++ b/src/lxc/lxc_controller.c @@ -1252,8 +1252,9 @@ virLXCControllerSetupDevPTS(virLXCControllerPtr ctrl) } if (access(ctrl->devptmx, R_OK) < 0) { - VIR_WARN("Kernel does not support private devpts, using shared devpts"); - VIR_FREE(ctrl->devptmx); + virReportSystemError(ENOSYS, "%s", + _("Kernel does not support private devpts")); + goto cleanup; } ret = 0; @@ -1278,24 +1279,13 @@ virLXCControllerSetupConsoles(virLXCControllerPtr ctrl, size_t i; for (i = 0 ; i < ctrl->nconsoles ; i++) { - if (ctrl->devptmx) { - VIR_DEBUG("Opening tty on private %s", ctrl->devptmx); - if (lxcCreateTty(ctrl->devptmx, - &ctrl->consoles[i].contFd, - &containerTTYPaths[i]) < 0) { - virReportSystemError(errno, "%s", - _("Failed to allocate tty")); - return -1; - } - } else { - VIR_DEBUG("Opening tty on shared /dev/ptmx"); - if (virFileOpenTty(&ctrl->consoles[i].contFd, - &containerTTYPaths[i], - 0) < 0) { - virReportSystemError(errno, "%s", + VIR_DEBUG("Opening tty on private %s", ctrl->devptmx); + if (lxcCreateTty(ctrl->devptmx, + &ctrl->consoles[i].contFd, + &containerTTYPaths[i]) < 0) { + virReportSystemError(errno, "%s", _("Failed to allocate tty")); - return -1; - } + return -1; } } return 0; -- 1.7.11.7

3 11

[libvirt] [RFC] Per machine type qemu capability detection
by Peter Krempa 05 Apr '13

05 Apr '13

Hi, qemu recently added a qmp/hmp command ("get-cpu-max") to determine maximum number of supported CPUs for a VM. The output of this command depends on the machine type that is used when it is called. The problem here is that libvirt uses "-M none" while detecting capabilities and with that machine type the get-cpu-max command returns always 1. My question is: Is it worthwhile to do per machine type capability detection? It definitely isn't worth doing just for the get-cpu-max support, but it might be needed for Q35 or other stuff that may depend on the actual machine type used. Pros of this approach: ---------------------- - (possibly) more accurate capabilities Cons: ----- - ~10x more qemus started on libvirtd startup (one per machine type) Thanks for your opinions. Peter

4 6

[libvirt] [RFC 0/5] X86CPU "feature-words" & "filtered-features" properties (v2)
by Eduardo Habkost 05 Apr '13

05 Apr '13

Changes v2: * Added "filtered-features" properties * Removed the non-qapi version and kept only the qapi version of the code * Moved the new types to qapi-schema.json instead of creating cpu-qapi-schema.json * Rebased on top of "feature words array" series Git tree: https://github.com/ehabkost/qemu-hacks/tree/work/cpu-raw-feature-props-v2 git://github.com/ehabkost/qemu-hacks.git work/cpu-raw-feature-props-v2 Description: As the work to get X86CPU subclasses static properties defined is taking very long to get reviewed and we risk not getting it finished in 1.5, I am sending an alternative mechanism to allow libvirt to probe for CPU model information. It's interesting to note that this alternative solution is actually _easier_ to use for libvirt, because libvirt logic is already based on CPUID bit values, not "feature names", so it should be very easy to convert the current libvirt code that (incorrectly) query the host CPU directly for available features to instead start QEMU with "-cpu host" and use the "feature-words" array information to find out which CPU features are supported by the host. The "feature-words" property will have two main use cases: * Checking host capabilities, by checking the features of the "host" CPU model; * Checking which features are enabled on each CPU model. The "filtered-features" property will have two main use cases: * Allowing libvirt to emulate the "check" & "enforce" flags while collecting more detailted data about the missing features; * Allowing libvirt to probe for CPU model information even if host features are missing (it can simply combine the "feature-words" and "filtered-features" values to get the full CPU model definition) Example output: $ ./QMP/qmp --path=/tmp/m qom-get --path=/machine/unattached/device[1] --property=feature-words item[0].cpuid-register: EDX item[0].cpuid-input-eax: 1 item[0].features: 126614521 item[1].cpuid-register: ECX item[1].cpuid-input-eax: 1 item[1].features: 2155880449 item[2].cpuid-register: EBX item[2].cpuid-input-eax: 7 item[2].features: 0 item[2].cpuid-input-ecx: 0 item[3].cpuid-register: EDX item[3].cpuid-input-eax: 2147483649 item[3].features: 563346425 item[4].cpuid-register: ECX item[4].cpuid-input-eax: 2147483649 item[4].features: 101 item[5].cpuid-register: EDX item[5].cpuid-input-eax: 3221225473 item[5].features: 0 item[6].cpuid-register: EAX item[6].cpuid-input-eax: 1073741825 item[6].features: 0 item[7].cpuid-register: EDX item[7].cpuid-input-eax: 2147483658 item[7].features: 0 Eduardo Habkost (5): target-i386: Add ECX information to FeatureWordInfo target-i386: Add "feature-words" property target-i386: Use FeatureWord loop on filter_features_for_kvm() target-i386: Introduce X86CPU.filtered_features field target-i386: Add "filtered-features" property to X86CPU .gitignore | 2 + Makefile.objs | 7 +++- qapi-schema.json | 32 +++++++++++++++ target-i386/cpu-qom.h | 3 ++ target-i386/cpu.c | 110 +++++++++++++++++++++++++++++++++++--------------- 5 files changed, 121 insertions(+), 33 deletions(-) -- 1.8.1.4

1 5

[libvirt] libvirt_lxc: implement <hostdev caps=net> device isolation in containers
by Bogdan Purcareata 05 Apr '13

05 Apr '13

>From Bogdan Purcareata <bogdan.purcareata(a)freescale.com> # This line is ignored. From: Bogdan Purcareata <bogdan.purcareata(a)freescale.com> Subject: libvirt_lxc: implement <hostdev caps=net> device isolation in containers In-Reply-To:

1 2

[libvirt] language bindings
by Vasiliy Tolstov 04 Apr '13

04 Apr '13

Hello. I'm newbie in libvirt world and not able to find answer for my question in google. If i want to create binding for some language (for example go) is that possible do not use libvirt C library and talk to libvirtd via some rpc protocol? For example json or internal binary.. is that possible to talk to libvirtd not using C library? -- Vasiliy Tolstov, e-mail: v.tolstov(a)selfip.ru jabber: vase(a)selfip.ru

3 3

Re: [libvirt] [Qemu-devel] attaching disk to paused guest
by Stefan Hajnoczi 04 Apr '13

04 Apr '13

On Wed, Apr 3, 2013 at 1:05 PM, Liron Aravot <laravot(a)redhat.com> wrote: > does qemu support attaching disk to a paused guest? Does it makes sense to allow this operation? Yes, it works. I tested it via libvirt: $ virsh start vm $ cat >/tmp/disk.xml <disk type='file' device='disk'> <driver name='qemu' type='raw'/> <source file='path/to/new_disk.img'/> <target dev='vdb' bus='virtio'/> <alias name='virtio-disk1'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x0'/> </disk> ^D $ virsh suspend vm $ virsh list Id Name State ---------------------------------------------------- 1 vm paused $ virsh attach-device vm /tmp/disk.xml Device attached successfully $ virsh resume vm The issue with hotplug is that it may involve guest cooperation. This is not possible when the guest is paused, but attaching a new virtio-blk adapter is okay because we'll just have pending PCI hotplug notifications when the guest is resumed. Hot unplug is the nasty case because the guest OS isn't running and cannot terminate its guest driver for the device. Because of this, both network adapters and disks in QEMU try to clean up after themselves and libvirt does not need to wait for the guest to acknowledge hot unplug. I think both hotplug and hot unplug should work fine while the guest is paused. Stefan

1 0

[libvirt] [PATCH v3 00/11] Add support for guests with TPM passthrough device
by Stefan Berger 04 Apr '13

04 Apr '13

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

4 23

[libvirt] [PATCH] storage: Fix volume cloning for logical volume.
by Atsushi Kumagai 04 Apr '13

04 Apr '13

When creating a logical volume with virStorageVolCreateXMLFrom, "qemu-img convert" is called internally if clonevol is a file volume. Then, vol->target.format is used as output_fmt parameter but the target.format of logical volumes is always 0 because logical volumes haven't the volume format type element. Fortunately, 0 was treated as RAW file format before commit f772b3d9, so there was no problem. But now, 0 is treated as the type of none, qemu-img fails with "Unknown file format 'none'". This patch fixes this issue by treating output block devices as RAW file format like for input block devices. Signed-off-by: Atsushi Kumagai <kumagai-atsushi(a)mxc.nes.nec.co.jp> diff --git a/src/storage/storage_backend.c b/src/storage/storage_backend.c index bb58d4a..2551f1f 100644 --- a/src/storage/storage_backend.c +++ b/src/storage/storage_backend.c @@ -652,7 +652,12 @@ virStorageBackendCreateQemuImgCmd(virConnectPtr conn, unsigned long long int size_arg; bool preallocate = false; - const char *type = virStorageFileFormatTypeToString(vol->target.format); + /* Treat output block devices as 'raw' format */ + const char *type = + virStorageFileFormatTypeToString(vol->type == VIR_STORAGE_VOL_BLOCK ? + VIR_STORAGE_FILE_RAW : + vol->target.format); + const char *backingType = vol->backingStore.path ? virStorageFileFormatTypeToString(vol->backingStore.format) : NULL; -- 1.7.1

2 1

[libvirt] [PATCH] Add support for SD cards in nodedev driver
by Daniel P. Berrange 04 Apr '13

04 Apr '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> The nodedev driver currently only detects harddisk, cdrom and floppy devices. This adds support for SD cards, which are common storage for ARM devices, eg the Google ChromeBook <device> <name>block_mmcblk0_0xb1c7c08b</name> <parent>computer</parent> <capability type='storage'> <block>/dev/mmcblk0</block> <drive_type>sd</drive_type> <serial>0xb1c7c08b</serial> <size>15758000128</size> <logical_block_size>512</logical_block_size> <num_blocks>30777344</num_blocks> </capability> </device> Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/node_device/node_device_udev.c | 41 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) diff --git a/src/node_device/node_device_udev.c b/src/node_device/node_device_udev.c index cb11e5f..362a47f 100644 --- a/src/node_device/node_device_udev.c +++ b/src/node_device/node_device_udev.c @@ -952,6 +952,35 @@ static int udevProcessFloppy(struct udev_device *device, return udevProcessRemoveableMedia(device, def, has_media); } + +static int udevProcessSD(struct udev_device *device, + virNodeDeviceDefPtr def) +{ + union _virNodeDevCapData *data = &def->caps->data; + int ret = 0; + + if (udevGetUint64SysfsAttr(device, + "size", + &data->storage.num_blocks) == PROPERTY_ERROR) { + goto out; + } + + if (udevGetUint64SysfsAttr(device, + "queue/logical_block_size", + &data->storage.logical_block_size) + == PROPERTY_ERROR) { + goto out; + } + + data->storage.size = data->storage.num_blocks * + data->storage.logical_block_size; + +out: + return ret; +} + + + /* This function exists to deal with the case in which a driver does * not provide a device type in the usual place, but udev told us it's * a storage device, and we can make a good guess at what kind of @@ -1056,6 +1085,13 @@ static int udevProcessStorage(struct udev_device *device, data->storage.drive_type = strdup("floppy"); if (!data->storage.drive_type) goto out; + } else if ((udevGetIntProperty(device, "ID_DRIVE_FLASH_SD", + &tmp_int, 0) == PROPERTY_FOUND) && + (tmp_int == 1)) { + + data->storage.drive_type = strdup("sd"); + if (!data->storage.drive_type) + goto out; } else { /* If udev doesn't have it, perhaps we can guess it. */ @@ -1071,6 +1107,8 @@ static int udevProcessStorage(struct udev_device *device, ret = udevProcessDisk(device, def); } else if (STREQ(def->caps->data.storage.drive_type, "floppy")) { ret = udevProcessFloppy(device, def); + } else if (STREQ(def->caps->data.storage.drive_type, "sd")) { + ret = udevProcessSD(device, def); } else { VIR_DEBUG("Unsupported storage type '%s'", def->caps->data.storage.drive_type); @@ -1082,6 +1120,7 @@ static int udevProcessStorage(struct udev_device *device, } out: + VIR_DEBUG("Storage ret=%d", ret); return ret; } @@ -1338,6 +1377,8 @@ static int udevAddOneDevice(struct udev_device *device) out: if (ret != 0) { + VIR_DEBUG("Discarding device %d %p %s", ret, def, + def ? def->sysfs_path : ""); virNodeDeviceDefFree(def); } -- 1.8.1.4

2 1

[libvirt] [PATCH] security_manager: fix comparison
by Guido Günther 04 Apr '13

04 Apr '13

otherwise we crash later on if we don't find a match like: #0 0xb72c2b4f in virSecurityManagerGenLabel (mgr=0xb8e42d20, vm=0xb8ef40c0) at security/security_manager.c:424 #1 0xb18811f3 in qemuProcessStart (conn=conn@entry=0xb8eed880, driver=driver@entry=0xb8e3b1e0, vm=vm@entry=0xb8ef58f0, migrateFrom=migrateFrom@entry=0xb18f6088 "stdio", stdin_fd=18, stdin_path=stdin_path@entry=0xb8ea7798 "/var/lib/jenkins/jobs/libvirt-tck-build/workspace/tck.img", snapshot=snapshot@entry=0x0, vmop=vmop@entry=VIR_NETDEV_VPORT_PROFILE_OP_RESTORE, flags=flags@entry=2) at qemu/qemu_process.c:3364 #2 0xb18d6cb2 in qemuDomainSaveImageStartVM (conn=conn@entry=0xb8eed880, driver=driver@entry=0xb8e3b1e0, vm=0xb8ef58f0, fd=fd@entry=0xb6bf3f98, header=header@entry=0xb6bf3fa0, path=path@entry=0xb8ea7798 "/var/lib/jenkins/jobs/libvirt-tck-build/workspace/tck.img", start_paused=start_paused@entry=false) at qemu/qemu_driver.c:4843 #3 0xb18d7eeb in qemuDomainRestoreFlags (conn=conn@entry=0xb8eed880, path=path@entry=0xb8ea7798 "/var/lib/jenkins/jobs/libvirt-tck-build/workspace/tck.img", dxml=dxml@entry=0x0, flags=flags@entry=0) at qemu/qemu_driver.c:4962 #4 0xb18d8123 in qemuDomainRestore (conn=0xb8eed880, path=0xb8ea7798 "/var/lib/jenkins/jobs/libvirt-tck-build/workspace/tck.img") at qemu/qemu_driver.c:4987 #5 0xb718d186 in virDomainRestore (conn=0xb8eed880, from=0xb8ea87d8 "/var/lib/jenkins/jobs/libvirt-tck-build/workspace/tck.img") at libvirt.c:2768 #6 0xb7736363 in remoteDispatchDomainRestore (args=<optimized out>, rerr=0xb6bf41f0, client=0xb8eedaf0, server=<optimized out>, msg=<optimized out>) at remote_dispatch.h:4679 #7 remoteDispatchDomainRestoreHelper (server=0xb8e1a3e0, client=0xb8eedaf0, msg=0xb8ee72c8, rerr=0xb6bf41f0, args=0xb8ea8968, ret=0xb8ef5330) at remote_dispatch.h:4661 #8 0xb720db01 in virNetServerProgramDispatchCall (msg=0xb8ee72c8, client=0xb8eedaf0, server=0xb8e1a3e0, prog=0xb8e216b0) at rpc/virnetserverprogram.c:439 #9 virNetServerProgramDispatch (prog=0xb8e216b0, server=server@entry=0xb8e1a3e0, client=0xb8eedaf0, msg=0xb8ee72c8) at rpc/virnetserverprogram.c:305 #10 0xb7206e97 in virNetServerProcessMsg (msg=<optimized out>, prog=<optimized out>, client=<optimized out>, srv=0xb8e1a3e0) at rpc/virnetserver.c:162 #11 virNetServerHandleJob (jobOpaque=0xb8ea7720, opaque=0xb8e1a3e0) at rpc/virnetserver.c:183 #12 0xb70f9f78 in virThreadPoolWorker (opaque=opaque@entry=0xb8e1a540) at util/virthreadpool.c:144 #13 0xb70f94a5 in virThreadHelper (data=0xb8e0e558) at util/virthreadpthread.c:161 #14 0xb705d954 in start_thread (arg=0xb6bf4b70) at pthread_create.c:304 #15 0xb6fd595e in clone () at ../sysdeps/unix/sysv/linux/i386/clone.S:130 This unbreaks libvirt-tck's domain/100-transient-save-restore.t with qemu:///session and selinux compiled in but disabled. Introduced by 8d68cbeaa8a64759323da1d64526e2a941eb93e1 --- src/security/security_manager.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/security/security_manager.c b/src/security/security_manager.c index 757fe28..f7c5c2e 100644 --- a/src/security/security_manager.c +++ b/src/security/security_manager.c @@ -437,7 +437,7 @@ int virSecurityManagerGenLabel(virSecurityManagerPtr mgr, return ret; virObjectLock(mgr); - for (i = 0; vm->nseclabels; i++) { + for (i = 0; i < vm->nseclabels; i++) { for (j = 0; sec_managers[j]; j++) if (STREQ(vm->seclabels[i]->model, sec_managers[j]->drv->name)) break; -- 1.7.10.4

3 3

[libvirt] [PATCH] Create fake NUMA info if libnuma isn't available
by Daniel P. Berrange 04 Apr '13

04 Apr '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> If libnuma is not compiled in, or numa_available() returns an error, stub out fake NUMA info consisting of one NUMA cell containing all CPUs and memory. Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/nodeinfo.c | 137 ++++++++++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 111 insertions(+), 26 deletions(-) diff --git a/src/nodeinfo.c b/src/nodeinfo.c index 72ab394..34d9410 100644 --- a/src/nodeinfo.c +++ b/src/nodeinfo.c @@ -1467,6 +1467,97 @@ cleanup: return ret; } +static int +nodeCapsInitNUMAFake(virCapsPtr caps ATTRIBUTE_UNUSED) +{ + virNodeInfo nodeinfo; + virCapsHostNUMACellCPUPtr cpus; + int ncpus; + int s, c, t; + int id; + + if (nodeGetInfo(NULL, &nodeinfo) < 0) + return -1; + + ncpus = VIR_NODEINFO_MAXCPUS(nodeinfo); + + if (VIR_ALLOC_N(cpus, ncpus) < 0) { + virReportOOMError(); + return -1; + } + + id = 0; + for (s = 0 ; s < nodeinfo.sockets ; s++) { + for (c = 0 ; c < nodeinfo.cores ; c++) { + for (t = 0 ; t < nodeinfo.threads ; t++) { + cpus[id].id = id; + cpus[id].socket_id = s; + cpus[id].core_id = c; + if (!(cpus[id].siblings = virBitmapNew(ncpus))) + goto error; + if (virBitmapSetBit(cpus[id].siblings, id) < 0) + goto error; + id++; + } + } + } + + if (virCapabilitiesAddHostNUMACell(caps, 0, + ncpus, + nodeinfo.memory, + cpus) < 0) + goto error; + + return 0; + + error: + for ( ; id >= 0 ; id--) + virBitmapFree(cpus[id].siblings); + VIR_FREE(cpus); + return -1; +} + +static int +nodeGetCellsFreeMemoryFake(virConnectPtr conn ATTRIBUTE_UNUSED, + unsigned long long *freeMems, + int startCell, + int maxCells ATTRIBUTE_UNUSED) +{ + double avail = physmem_available(); + + if (startCell != 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("start cell %d out of range (0-%d)"), + startCell, 0); + return -1; + } + + freeMems[0] = (unsigned long long)avail; + + if (!freeMems[0]) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("Cannot determine free memory")); + return -1; + } + + return 1; +} + +static unsigned long long +nodeGetFreeMemoryFake(virConnectPtr conn ATTRIBUTE_UNUSED) +{ + double avail = physmem_available(); + unsigned long long ret; + + if (!(ret = (unsigned long long)avail)) { + virReportError(VIR_ERR_INTERNAL_ERROR, "%s", + _("Cannot determine free memory")); + return 0; + } + + return ret; +} + #if WITH_NUMACTL # if LIBNUMA_API_VERSION <= 1 # define NUMA_MAX_N_CPUS 4096 @@ -1547,7 +1638,7 @@ nodeCapsInitNUMA(virCapsPtr caps) bool topology_failed = false; if (numa_available() < 0) - return 0; + return nodeCapsInitNUMAFake(caps); int mask_n_bytes = max_n_cpus / 8; if (VIR_ALLOC_N(mask, mask_n_bytes / sizeof(*mask)) < 0) @@ -1610,7 +1701,7 @@ cleanup: int -nodeGetCellsFreeMemory(virConnectPtr conn ATTRIBUTE_UNUSED, +nodeGetCellsFreeMemory(virConnectPtr conn, unsigned long long *freeMems, int startCell, int maxCells) @@ -1619,11 +1710,10 @@ nodeGetCellsFreeMemory(virConnectPtr conn ATTRIBUTE_UNUSED, int ret = -1; int maxCell; - if (numa_available() < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, - "%s", _("NUMA not supported on this host")); - goto cleanup; - } + if (numa_available() < 0) + return nodeGetCellsFreeMemoryFake(conn, freeMems, + startCell, maxCells); + maxCell = numa_max_node(); if (startCell > maxCell) { virReportError(VIR_ERR_INTERNAL_ERROR, @@ -1652,16 +1742,14 @@ cleanup: } unsigned long long -nodeGetFreeMemory(virConnectPtr conn ATTRIBUTE_UNUSED) +nodeGetFreeMemory(virConnectPtr conn) { unsigned long long freeMem = 0; int n; - if (numa_available() < 0) { - virReportError(VIR_ERR_INTERNAL_ERROR, - "%s", _("NUMA not supported on this host")); - goto cleanup; - } + if (numa_available() < 0) + return nodeGetFreeMemoryFake(conn); + for (n = 0 ; n <= numa_max_node() ; n++) { long long mem; @@ -1720,24 +1808,21 @@ cleanup: #else -int nodeCapsInitNUMA(virCapsPtr caps ATTRIBUTE_UNUSED) { - return 0; +int nodeCapsInitNUMA(virCapsPtr caps) { + return nodeCapsInitNUMAFake(caps); } -int nodeGetCellsFreeMemory(virConnectPtr conn ATTRIBUTE_UNUSED, - unsigned long long *freeMems ATTRIBUTE_UNUSED, - int startCell ATTRIBUTE_UNUSED, - int maxCells ATTRIBUTE_UNUSED) +int nodeGetCellsFreeMemory(virConnectPtr conn, + unsigned long long *freeMems, + int startCell, + int maxCells) { - virReportError(VIR_ERR_NO_SUPPORT, "%s", - _("NUMA memory information not available on this platform")); - return -1; + return nodeGetCellsFreeMemoryFake(conn, freeMems, + startCell, maxCells); } -unsigned long long nodeGetFreeMemory(virConnectPtr conn ATTRIBUTE_UNUSED) +unsigned long long nodeGetFreeMemory(virConnectPtr conn) { - virReportError(VIR_ERR_NO_SUPPORT, "%s", - _("NUMA memory information not available on this platform")); - return 0; + return nodeGetFreeMemoryFake(conn); } #endif -- 1.8.1.4

2 1

[libvirt] [PATCH] Cope with missing /sys/devices/system/cpu/cpu0/topology files
by Daniel P. Berrange 04 Apr '13

04 Apr '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> Not all kernel builds have any entries under the location /sys/devices/system/cpu/cpu0/topology. We already cope with that being missing in some cases, but not all. Update the code which looks for thread_siblings to cope with the missing file Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/nodeinfo.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/src/nodeinfo.c b/src/nodeinfo.c index 415c2e5..72ab394 100644 --- a/src/nodeinfo.c +++ b/src/nodeinfo.c @@ -168,6 +168,12 @@ virNodeCountThreadSiblings(const char *dir, unsigned int cpu) pathfp = fopen(path, "r"); if (pathfp == NULL) { + /* If file doesn't exist, then pretend our only + * sibling is ourself */ + if (errno == ENOENT) { + VIR_FREE(path); + return 1; + } virReportSystemError(errno, _("cannot open %s"), path); VIR_FREE(path); return 0; -- 1.8.1.4

2 1

[libvirt] [PATCH] Add armv6l architecture to list of valid arches
by Daniel P. Berrange 04 Apr '13

04 Apr '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> The Raspberry Pi runs the armv6l architecture and apparently people are trying to run libvirt LXC on it. So we should allow that as a valid arch Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/util/virarch.c | 1 + src/util/virarch.h | 1 + 2 files changed, 2 insertions(+) diff --git a/src/util/virarch.c b/src/util/virarch.c index ffd07dc..8585a78 100644 --- a/src/util/virarch.c +++ b/src/util/virarch.c @@ -35,6 +35,7 @@ static const struct virArchData { } virArchData[] = { { "none", 0, VIR_ARCH_LITTLE_ENDIAN }, { "alpha", 64, VIR_ARCH_BIG_ENDIAN }, + { "armv6l", 32, VIR_ARCH_LITTLE_ENDIAN }, { "armv7l", 32, VIR_ARCH_LITTLE_ENDIAN }, { "cris", 32, VIR_ARCH_LITTLE_ENDIAN }, { "i686", 32, VIR_ARCH_LITTLE_ENDIAN }, diff --git a/src/util/virarch.h b/src/util/virarch.h index 9cf6ce2..0d8ae25 100644 --- a/src/util/virarch.h +++ b/src/util/virarch.h @@ -27,6 +27,7 @@ typedef enum { VIR_ARCH_NONE, VIR_ARCH_ALPHA, /* Alpha 64 BE http://en.wikipedia.org/wiki/DEC_Alpha */ + VIR_ARCH_ARMV6L, /* ARMv6 32 LE http://en.wikipedia.org/wiki/ARM_architecture */ VIR_ARCH_ARMV7L, /* ARMv7 32 LE http://en.wikipedia.org/wiki/ARM_architecture */ VIR_ARCH_CRIS, /* ETRAX 32 LE http://en.wikipedia.org/wiki/ETRAX_CRIS */ VIR_ARCH_I686, /* x86 32 LE http://en.wikipedia.org/wiki/X86 */ -- 1.8.1.4

2 1

[libvirt] libvirtd segfault on latest git
by Luiz Capitulino 04 Apr '13

04 Apr '13

Hi, Libvirtd from latest git (HEAD fc8c1787) is segfaulting when I try to restore a previously saved domain (backtrace below). I was testing a custom QEMU when it happened first time, but it also happens with latest qemu.git (HEAD 162cbbd17). What I did to run libvirt from source was: 1. Build it $ ./autogen.sh --system --enable-compile-warnings=error $ ./configure --localstatedir=/var --sysconfdir=/etc --prefix=/usr --libdir=/usr/lib (this was needed because I had to re-ran configure and it ended up overwriting --system) $ make -j4 2. Stop system's libvirtd # systemctl stop libvitd 3. Change /etc/libvirt/qemu/Fedora18.xml to use a custom QEMU binary 4. Run libvirtd # ./daemon/libvirtd 5. Reproduce the problem with virsh # ./tools/virsh virsh # save Fedora18 foobar virsh # restore foobar error: Failed to restore domain from foobar error: End of file while reading data: Input/output error error: Failed to reconnect to the hypervisor 6. Change to the libvirtd terminal [...] 2013-04-03 20:01:58.742+0000: 32334: debug : virNetServerAddShutdownInhibition:795 : srv=0x1ecf3f0 inhibitions=1 2013-04-03 20:01:58.742+0000: 32334: debug : qemuProcessStart:3351 : Preparing host devices 2013-04-03 20:01:58.742+0000: 32334: debug : qemuProcessStart:3355 : Preparing chr devices 2013-04-03 20:01:58.742+0000: 32334: debug : qemuProcessStart:3364 : Generating domain security label (if required) ====== end of log ===== Segmentation fault (core dumped) [root@doriath libvirt]# Here's the backtrace: Core was generated by `/home/lcapitulino/work/src/upstream/libvirt/daemon/.libs/lt-libvirtd'. Program terminated with signal 11, Segmentation fault. #0 0x00007f7623a4e79f in virSecurityManagerGenLabel (mgr=0x7f760c022380, vm=<optimized out>) at security/security_manager.c:424 424 int virSecurityManagerGenLabel(virSecurityManagerPtr mgr, (gdb) bt #0 0x00007f7623a4e79f in virSecurityManagerGenLabel (mgr=0x7f760c022380, vm=<optimized out>) at security/security_manager.c:424 #1 0x00007f761467718c in qemuProcessStart (conn=conn@entry=0x7f7608000ad0, driver=driver@entry=0x7f760c009440, vm=vm@entry=0x7f760c2097d0, migrateFrom=migrateFrom@entry=0x7f76146dfeeb "stdio", stdin_fd=16, stdin_path=stdin_path@entry=0x7f7600000a90 "/home/lcapitulino/work/src/upstream/libvirt/foobar", snapshot=snapshot@entry=0x0, vmop=vmop@entry=VIR_NETDEV_VPORT_PROFILE_OP_RESTORE, flags=flags@entry=2) at qemu/qemu_process.c:3365 #2 0x00007f76146c25a0 in qemuDomainSaveImageStartVM (conn=conn@entry=0x7f7608000ad0, driver=driver@entry=0x7f760c009440, vm=vm@entry=0x7f760c2097d0, fd=fd@entry=0x7f761882fa7c, header=header@entry=0x7f761882fa90, path=path@entry=0x7f7600000a90 "/home/lcapitulino/work/src/upstream/libvirt/foobar", start_paused=start_paused@entry=false) at qemu/qemu_driver.c:4843 #3 0x00007f76146c35c6 in qemuDomainRestoreFlags (conn=0x7f7608000ad0, path=0x7f7600000a90 "/home/lcapitulino/work/src/upstream/libvirt/foobar", dxml=<optimized out>, flags=<optimized out>) at qemu/qemu_driver.c:4962 #4 0x00007f762393923b in virDomainRestore (conn=0x7f7608000ad0, from=0x7f76000008c0 "/home/lcapitulino/work/src/upstream/libvirt/foobar") at libvirt.c:2768 #5 0x00007f7624365691 in remoteDispatchDomainRestore (server=<optimized out>, msg=<optimized out>, args=0x7f7600000900, rerr=0x7f761882fcd0, client=0x7f76253b9150) at remote_dispatch.h:4679 #6 remoteDispatchDomainRestoreHelper (server=<optimized out>, client=0x7f76253b9150, msg=<optimized out>, rerr=0x7f761882fcd0, args=0x7f7600000900, ret=<optimized out>) at remote_dispatch.h:4661 #7 0x00007f76239a4d56 in virNetServerProgramDispatchCall (msg=0x7f76253b8700, client=0x7f76253b9150, server=0x7f76253b23f0, prog=0x7f76253bd070) at rpc/virnetserverprogram.c:439 #8 virNetServerProgramDispatch (prog=0x7f76253bd070, server=server@entry=0x7f76253b23f0, client=0x7f76253b9150, msg=0x7f76253b8700) at rpc/virnetserverprogram.c:305 #9 0x00007f76239a0338 in virNetServerProcessMsg (msg=<optimized out>, prog=<optimized out>, client=<optimized out>, srv=0x7f76253b23f0) at rpc/virnetserver.c:162 #10 virNetServerHandleJob (jobOpaque=<optimized out>, opaque=0x7f76253b23f0) at rpc/virnetserver.c:183 #11 0x00007f76238c20ee in virThreadPoolWorker (opaque=opaque@entry=0x7f76253bd230) at util/virthreadpool.c:144 #12 0x00007f76238c1766 in virThreadHelper (data=<optimized out>) at util/virthreadpthread.c:161 #13 0x00007f76227ffd15 in start_thread (arg=0x7f7618830700) at pthread_create.c:308 #14 0x00007f762212b48d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:114

1 0

[libvirt] [PATCH] Implement minimal sysinfo for ARM platforms
by Daniel P. Berrange 04 Apr '13

04 Apr '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> Implement the bare minimal sysinfo for ARM platforms by reading the CPU models from /proc/cpuinfo Signed-off-by: Daniel P. Berrange <berrange(a)redhat.com> --- src/util/virsysinfo.c | 129 +++++++++++++++++++++++++++++++++++- tests/sysinfodata/armcpuinfo.data | 17 +++++ tests/sysinfodata/armsysinfo.expect | 10 +++ tests/sysinfotest.c | 15 ++++- 4 files changed, 169 insertions(+), 2 deletions(-) create mode 100644 tests/sysinfodata/armcpuinfo.data create mode 100644 tests/sysinfodata/armsysinfo.expect diff --git a/src/util/virsysinfo.c b/src/util/virsysinfo.c index 83c445d..ba0c245 100644 --- a/src/util/virsysinfo.c +++ b/src/util/virsysinfo.c @@ -255,8 +255,134 @@ no_memory: return NULL; } -#elif defined(__s390__) || defined(__s390x__) +#elif defined(__arm__) +static int +virSysinfoParseSystem(const char *base, virSysinfoDefPtr ret) +{ + char *eol = NULL; + const char *cur; + + if ((cur = strstr(base, "platform")) == NULL) + return 0; + + base = cur; + /* Account for format 'platform : XXXX'*/ + cur = strchr(cur, ':') + 1; + eol = strchr(cur, '\n'); + virSkipSpaces(&cur); + if (eol && + ((ret->system_family = strndup(cur, eol - cur)) == NULL)) + goto no_memory; + + if ((cur = strstr(base, "model")) != NULL) { + cur = strchr(cur, ':') + 1; + eol = strchr(cur, '\n'); + virSkipSpaces(&cur); + if (eol && ((ret->system_serial = strndup(cur, eol - cur)) + == NULL)) + goto no_memory; + } + + if ((cur = strstr(base, "machine")) != NULL) { + cur = strchr(cur, ':') + 1; + eol = strchr(cur, '\n'); + virSkipSpaces(&cur); + if (eol && ((ret->system_version = strndup(cur, eol - cur)) + == NULL)) + goto no_memory; + } + return 0; + +no_memory: + return -1; +} + +static int +virSysinfoParseProcessor(const char *base, virSysinfoDefPtr ret) +{ + const char *cur; + char *eol, *tmp_base; + virSysinfoProcessorDefPtr processor; + char *processor_type = NULL; + + if (!(tmp_base = strstr(base, "Processor"))) + return 0; + + base = tmp_base; + eol = strchr(base, '\n'); + cur = strchr(base, ':') + 1; + virSkipSpaces(&cur); + if (eol && + ((processor_type = strndup(cur, eol - cur)) + == NULL)) + goto no_memory; + base = cur; + + while ((tmp_base = strstr(base, "processor")) != NULL) { + base = tmp_base; + eol = strchr(base, '\n'); + cur = strchr(base, ':') + 1; + + if (VIR_EXPAND_N(ret->processor, ret->nprocessor, 1) < 0) { + goto no_memory; + } + processor = &ret->processor[ret->nprocessor - 1]; + + virSkipSpaces(&cur); + if (eol && + ((processor->processor_socket_destination = strndup + (cur, eol - cur)) == NULL)) + goto no_memory; + + if (processor_type && + !(processor->processor_type = strdup(processor_type))) + goto no_memory; + + base = cur; + } + + VIR_FREE(processor_type); + return 0; + +no_memory: + VIR_FREE(processor_type); + return -1; +} + +/* virSysinfoRead for ARMv7 + * Gathers sysinfo data from /proc/cpuinfo */ +virSysinfoDefPtr +virSysinfoRead(void) { + virSysinfoDefPtr ret = NULL; + char *outbuf = NULL; + + if (VIR_ALLOC(ret) < 0) + goto no_memory; + + if (virFileReadAll(CPUINFO, 2048, &outbuf) < 0) { + virReportError(VIR_ERR_INTERNAL_ERROR, + _("Failed to open %s"), CPUINFO); + return NULL; + } + + ret->nprocessor = 0; + ret->processor = NULL; + if (virSysinfoParseProcessor(outbuf, ret) < 0) + goto no_memory; + + if (virSysinfoParseSystem(outbuf, ret) < 0) + goto no_memory; + + return ret; + +no_memory: + VIR_FREE(outbuf); + return NULL; +} + + +#elif defined(__s390__) || defined(__s390x__) /* we need to ignore warnings about strchr caused by -Wlogical-op for some GCC versions. @@ -403,6 +529,7 @@ no_memory: !(defined(__x86_64__) || \ defined(__i386__) || \ defined(__amd64__) || \ + defined(__arm__) || \ defined(__powerpc__)) virSysinfoDefPtr virSysinfoRead(void) { diff --git a/tests/sysinfodata/armcpuinfo.data b/tests/sysinfodata/armcpuinfo.data new file mode 100644 index 0000000..6ef23ca --- /dev/null +++ b/tests/sysinfodata/armcpuinfo.data @@ -0,0 +1,17 @@ +Processor : ARMv7 Processor rev 4 (v7l) +processor : 0 +BogoMIPS : 1694.10 + +processor : 1 +BogoMIPS : 1694.10 + +Features : swp half thumb fastmult vfp edsp thumbee neon vfpv3 tls vfpv4 idiva idivt +CPU implementer : 0x41 +CPU architecture: 7 +CPU variant : 0x0 +CPU part : 0xc0f +CPU revision : 4 + +Hardware : SAMSUNG EXYNOS5 (Flattened Device Tree) +Revision : 0000 +Serial : 0000000000000000 diff --git a/tests/sysinfodata/armsysinfo.expect b/tests/sysinfodata/armsysinfo.expect new file mode 100644 index 0000000..1af9e6d --- /dev/null +++ b/tests/sysinfodata/armsysinfo.expect @@ -0,0 +1,10 @@ +<sysinfo type='smbios'> + <processor> + <entry name='socket_destination'>0</entry> + <entry name='type'>ARMv7 Processor rev 4 (v7l)</entry> + </processor> + <processor> + <entry name='socket_destination'>1</entry> + <entry name='type'>ARMv7 Processor rev 4 (v7l)</entry> + </processor> +</sysinfo> diff --git a/tests/sysinfotest.c b/tests/sysinfotest.c index 94493a3..8955208 100644 --- a/tests/sysinfotest.c +++ b/tests/sysinfotest.c @@ -40,7 +40,8 @@ # if defined(__s390__) || defined(__s390x__) || \ defined(__powerpc__) || defined(__powerpc64__) || \ - defined(__i386__) || defined(__x86_64__) || defined(__amd64__) + defined(__i386__) || defined(__x86_64__) || defined(__amd64__) || \ + defined(__arm__) /* from sysinfo.c */ void virSysinfoSetup(const char *decoder, @@ -164,6 +165,18 @@ test_x86(void) } VIRT_TEST_MAIN(test_x86) +# elif defined(__arm__) +static int +test_arm(void) +{ + return sysinfotest_run("arm sysinfo", + NULL, + NULL, + "/sysinfodata/armcpuinfo.data", + "/sysinfodata/armsysinfo.expect"); +} + +VIRT_TEST_MAIN(test_arm) # else int main(void) -- 1.8.1.4

2 1

[libvirt] [PATCH] Wire up sysinfo for LXC driver
by Daniel P. Berrange 04 Apr '13

04 Apr '13

From: "Daniel P. Berrange" <berrange(a)redhat.com> The sysinfo code used by QEMU is trivially portable to the LXC driver --- src/lxc/lxc_conf.h | 3 +++ src/lxc/lxc_driver.c | 29 +++++++++++++++++++++++++++++ 2 files changed, 32 insertions(+) diff --git a/src/lxc/lxc_conf.h b/src/lxc/lxc_conf.h index b46dc32..f8f7c9a 100644 --- a/src/lxc/lxc_conf.h +++ b/src/lxc/lxc_conf.h @@ -35,6 +35,7 @@ # include "vircgroup.h" # include "security/security_manager.h" # include "configmake.h" +# include "virsysinfo.h" # include "virusb.h" # define LXC_DRIVER_NAME "LXC" @@ -55,6 +56,8 @@ struct _virLXCDriver { virCgroupPtr cgroup; + virSysinfoDefPtr hostsysinfo; + size_t nactive; virStateInhibitCallback inhibitCallback; void *inhibitOpaque; diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index ba14db7..17f01d5 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -1457,6 +1457,8 @@ static int lxcStartup(bool privileged, if (!lxc_driver->domainEventState) goto cleanup; + lxc_driver->hostsysinfo = virSysinfoRead(); + lxc_driver->log_libvirtd = 0; /* by default log to container logfile */ lxc_driver->have_netns = lxcCheckNetNsSupport(); @@ -1574,6 +1576,8 @@ static int lxcShutdown(void) virLXCProcessAutoDestroyShutdown(lxc_driver); + virSysinfoDefFree(lxc_driver->hostsysinfo); + virObjectUnref(lxc_driver->activeUsbHostdevs); virObjectUnref(lxc_driver->caps); virObjectUnref(lxc_driver->securityManager); @@ -4547,6 +4551,30 @@ cleanup: } +static char * +lxcGetSysinfo(virConnectPtr conn, unsigned int flags) +{ + virLXCDriverPtr driver = conn->privateData; + virBuffer buf = VIR_BUFFER_INITIALIZER; + + virCheckFlags(0, NULL); + + if (!driver->hostsysinfo) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Host SMBIOS information is not available")); + return NULL; + } + + if (virSysinfoFormat(&buf, driver->hostsysinfo) < 0) + return NULL; + if (virBufferError(&buf)) { + virReportOOMError(); + return NULL; + } + return virBufferContentAndReset(&buf); +} + + /* Function Tables */ static virDriver lxcDriver = { .no = VIR_DRV_LXC, @@ -4555,6 +4583,7 @@ static virDriver lxcDriver = { .close = lxcClose, /* 0.4.2 */ .version = lxcVersion, /* 0.4.6 */ .getHostname = virGetHostname, /* 0.6.3 */ + .getSysinfo = lxcGetSysinfo, /* 1.0.5 */ .nodeGetInfo = nodeGetInfo, /* 0.6.5 */ .getCapabilities = lxcGetCapabilities, /* 0.6.5 */ .listDomains = lxcListDomains, /* 0.4.2 */ -- 1.8.1.4

2 1

[libvirt] [PATCH] qemu: fix memory leak on -machine usage error
by Eric Blake 04 Apr '13

04 Apr '13

Commit f84b92ea introduced a memory leak on error; John Ferlan reported that valgrind caught it during 'make check'. * src/qemu/qemu_command.c (qemuBuildMachineArgStr): Plug leak. --- Pushing under the trivial rule. src/qemu/qemu_command.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/qemu/qemu_command.c b/src/qemu/qemu_command.c index 693d30d..29424d6 100644 --- a/src/qemu/qemu_command.c +++ b/src/qemu/qemu_command.c @@ -5228,6 +5228,7 @@ qemuBuildMachineArgStr(virCommandPtr cmd, virReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", _("dump-guest-core is not available " "with this QEMU binary")); + virBufferFreeAndReset(&buf); return -1; } -- 1.8.1.4

1 0

[libvirt] how to use svirt
by yue 03 Apr '13

03 Apr '13

hi,all i know svirt is merged into libvirt upstream, but how to use them? thanks

1 0

[libvirt] (no subject)
by Dan Walsh 03 Apr '13

03 Apr '13

Most of them effect virt-sandbox-service, with the biggest change beeing the addition of InteractiveContainers. We want to be able to support OpenShift, which requres the abiltiy to run a command within a container environement but does not want to use the systemd/unitfiles. These containers will run a setup script and then a user process to actually run the application. >From Dan Walsh <dwalsh(a)redhat.com> # This line is ignored. From: Dan Walsh <dwalsh(a)redhat.com> Subject: In-Reply-To:

4 36

[libvirt] [PATCH 0/2] Fix numeric overflow in qemu block job handling
by Peter Krempa 03 Apr '13

03 Apr '13

The JSON generator and qemu QMP protocol limit numeric values to LLONG_MAX even for unsigned values. A bug in the code caused numeric overflow when passing through speed parameters to qemu block job commands. Background: https://bugzilla.redhat.com/show_bug.cgi?id=927156 Peter Krempa (2): qemu-JSON: Error out if number is out of range instead of overflowing to negative qemu-blockjob: Fix limit of bandwidth for block jobs to supported value src/qemu/qemu_monitor.c | 21 ++++++++++++--------- src/qemu/qemu_monitor_json.c | 10 ++++++++-- 2 files changed, 20 insertions(+), 11 deletions(-) -- 1.8.1.5

2 6

[libvirt] [PATCH 0/2] Improve security hardening of binaries
by Daniel P. Berrange 03 Apr '13

03 Apr '13

Fedora has a "hardened build" option in RPM specfiles: https://fedoraproject.org/wiki/Packaging:Guidelines#PIE While we could enable that in the RPM, this would only apply to Fedora. Thus these patches directly integrate it in libvirt's configure.ac / Makefile.am files. With these 2 patches all executables gain -fPIE and -z relro -z now. Using the checksec.sh script from http://www.trapkit.de/tools/checksec.html We can see the difference, before: $ ~/checksec.sh --file /usr/sbin/libvirtd 'RELRO STACK CANARY NX PIE RPATH RUNPATH FILE Partial RELRO Canary found NX enabled No PIE No RPATH No RUNPATH /usr/sbin/libvirtd After $ ~/checksec.sh --file /usr/sbin/libvirtd 'RELRO STACK CANARY NX PIE RPATH RUNPATH FILE Full RELRO Canary found NX enabled PIE enabled No RPATH No RUNPATH /usr/sbin/libvirtd

2 4

[libvirt] [libvirt-designer 0/3] A few more cleanups
by Christophe Fergeau 03 Apr '13

03 Apr '13

Hey, here are a few small patches for libvirt-designer to fix a few leaks/polish minor things. Christophe

2 7

[libvirt] [PATCH 1/1] Fix the crash when seclable is freed
by Li Zhang 03 Apr '13

03 Apr '13

From: Li Zhang <zhlcindy(a)linux.vnet.ibm.com> When seclabel's type is VIR_DOMAIN_SECLABEL_NONE, virSecurityLabelDefPtr's members are not allocated. So it will cause crash when calling VIR_FREE. This problem is found when running autotest on PPC. Failed to remove cgroup for virt-tests-vm1 *** glibc detected *** /usr/sbin/libvirtd: free(): invalid pointer: 0x00003fff9c187510 *** ======= Backtrace: ========= /lib64/libc.so.6(+0xb89c4)[0x3fffa9bc89c4] /lib64/libvirt.so.0(virFree-0x3e2320)[0x3fffaa82e9c0] /lib64/libvirt.so.0(virSecurityLabelDefFree-0x378984)[0x3fffaa89d69c] /lib64/libvirt.so.0(virDomainDefFree-0x367c98)[0x3fffaa8ae968] /usr/lib64/libvirt/connection-driver/libvirt_driver_qemu.so(qemuProcessStop-0xc85f8)[0x3fffa2899d58] /usr/lib64/libvirt/connection-driver/libvirt_driver_qemu.so(+0xc3668)[0x3fffa28e3668] /lib64/libvirt.so.0(virDomainDestroy-0x309bd0)[0x3fffaa90f6f0] /usr/sbin/libvirtd[0x10035230] /lib64/libvirt.so.0(virNetServerProgramDispatch-0x289b50)[0x3fffaa995930] /lib64/libvirt.so.0(+0x20db18)[0x3fffaa98db18] /lib64/libvirt.so.0(+0xfbd24)[0x3fffaa87bd24] /lib64/libvirt.so.0(+0xfaec8)[0x3fffaa87aec8] /lib64/libpthread.so.0(+0xc604)[0x3fffa9d7c604] /lib64/libc.so.6(clone-0xb8fe4)[0x3fffa9c3f094] Signed-off-by: Li Zhang <zhlcindy(a)linux.vnet.ibm.com> --- src/conf/domain_conf.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/src/conf/domain_conf.c b/src/conf/domain_conf.c index f3fca7f..2856660 100644 --- a/src/conf/domain_conf.c +++ b/src/conf/domain_conf.c @@ -1006,6 +1006,8 @@ virSecurityLabelDefFree(virSecurityLabelDefPtr def) { if (!def) return; + if (def->type == VIR_DOMAIN_SECLABEL_NONE) + return; VIR_FREE(def->model); VIR_FREE(def->label); VIR_FREE(def->imagelabel); -- 1.7.5.4

4 10

[libvirt] Networking options in libvirt_lxc
by Purcareata Bogdan-B43198 03 Apr '13

03 Apr '13

Hello, I am doing some research on [subject] and I would like to find out some information regarding various scenarios. I've studied the official documentation at [1] and some of the mailing list archives. The configurations I have in mind are somewhat inspired by what the sf LXC package offers in terms of networking. What I've tested so far and works: - Shared networking - all host interfaces are present in the container if no <interface> tag has been specified in the domain configuration. I'm assuming this is because the container is started in the same network namespace like the host. Is it possible to make only a subset of these interfaces visible inside the container? - Bridge to LAN - connecting a domain interface to a host bridge; - Direct attachment through a Macvtap device - all 3 modes (vepa, bridge and private) work as expected, "passthrough" requires some capabilities in the physical device (SRIOV), which I don't have - assuming I have a device with this capability, is this configuration supported by (implemented in) the libvirt_lxc driver? What other scenarios I would be interested in: - host network interface private to the container - much like what lxc.network.phys is offering: "dedicated NIC from host passed through". I've read some documentation about <hostdev> and how to assign PCI devices to virtual machine, but I understand this is only possible with KVM - it's assigned from the kernel, it makes more sense, etc. However, I've also read a thread on the mailing list regarding <hostdev mode="capabilities">, which offers access from a container to a device, but it's currenly only applicable to block and character devices. Is there currently any way to make a host interface private to a container? Thank you very much! Best regards, Bogdan P. [1] http://libvirt.org

2 2

[libvirt] [PATCH] virsh: Fix typo in docs
by Peter Krempa 03 Apr '13

03 Apr '13

s/persitent/persistent/ --- Pushed under trivial rule. tools/virsh.pod | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/virsh.pod b/tools/virsh.pod index 9ca4435..1c8f9ee 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -2152,7 +2152,7 @@ Returns basic information about the I<network> object. Returns the list of active networks, if I<--all> is specified this will also include defined but inactive networks, if I<--inactive> is specified only the inactive ones will be listed. You may also want to filter the returned networks -by I<--persistent> to list the persitent ones, I<--transient> to list the +by I<--persistent> to list the persistent ones, I<--transient> to list the transient ones, I<--autostart> to list the ones with autostart enabled, and I<--no-autostart> to list the ones with autostart disabled. -- 1.8.1.5

1 0

[libvirt] [PATCH] sec_manager: Refuse to start domain with unsupported seclabel
by Michal Privoznik 03 Apr '13

03 Apr '13

https://bugzilla.redhat.com/show_bug.cgi?id=947387 If a user configures a domain to use a seclabel of a specific type, but the appropriate driver is not accessible, we should refuse to start the domain. For instance, if user requires selinux, but it is either non present in the system, or is just disabled, we should not start the domain. Moreover, since we are touching only those labels we have a security driver for, the other labels may confuse libvirt when reconnecting to a domain on libvirtd restart. In our selinux example, when starting up a domain, missing security label is okay, as we auto-generate one. But later, when libvirt is re-connecting to a live qemu instance, we parse a state XML, where security label is required and it is an error if missing: error : virSecurityLabelDefParseXML:3228 : XML error: security label is missing This results in a qemu process left behind without any libvirt control. --- src/security/security_manager.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/src/security/security_manager.c b/src/security/security_manager.c index b671a91..757fe28 100644 --- a/src/security/security_manager.c +++ b/src/security/security_manager.c @@ -425,7 +425,7 @@ int virSecurityManagerGenLabel(virSecurityManagerPtr mgr, virDomainDefPtr vm) { int ret = -1; - size_t i; + size_t i, j; virSecurityManagerPtr* sec_managers = NULL; virSecurityLabelDefPtr seclabel; bool generated = false; @@ -437,6 +437,19 @@ int virSecurityManagerGenLabel(virSecurityManagerPtr mgr, return ret; virObjectLock(mgr); + for (i = 0; vm->nseclabels; i++) { + for (j = 0; sec_managers[j]; j++) + if (STREQ(vm->seclabels[i]->model, sec_managers[j]->drv->name)) + break; + + if (!sec_managers[j]) { + virReportError(VIR_ERR_CONFIG_UNSUPPORTED, + _("Unable to find security driver for label %s"), + vm->seclabels[i]->model); + goto cleanup; + } + } + for (i = 0; sec_managers[i]; i++) { generated = false; seclabel = virDomainDefGetSecurityLabelDef(vm, sec_managers[i]->drv->name); -- 1.8.1.5

2 2

[libvirt] [PATCH v2] Allow multiple parameters for schedinfo
by Martin Kletzander 03 Apr '13

03 Apr '13

virsh schedinfo was able to set only one parameter at a time (not counting the deprecated options), but it is useful to set more at once, so this patch adds the possibility to do stuff like this: virsh schedinfo <domain> cpu_shares=0 vcpu_period=0 vcpu_quota=0 \ emulator_period=0 emulator_quota=0 Invalid scheduler options are reported as well. These were previously reported only if the command hadn't updated any values (when cmdSchedInfoUpdate returned 0). Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=810078 Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=919372 Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=919375 Signed-off-by: Martin Kletzander <mkletzan(a)redhat.com> --- v2: - correctly report unsupported options - man page updated tests/virsh-schedinfo | 4 +- tools/virsh-domain.c | 119 ++++++++++++++++++++++++++++---------------------- tools/virsh.pod | 4 +- 3 files changed, 72 insertions(+), 55 deletions(-) diff --git a/tests/virsh-schedinfo b/tests/virsh-schedinfo index 4f462f8..37f7bd3 100755 --- a/tests/virsh-schedinfo +++ b/tests/virsh-schedinfo @@ -1,7 +1,7 @@ #!/bin/sh # Ensure that virsh schedinfo --set invalid=val fails -# Copyright (C) 2010-2011 Red Hat, Inc. +# Copyright (C) 2010-2011, 2013 Red Hat, Inc. # This program is free software: you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by @@ -37,7 +37,7 @@ fi . "$srcdir/test-lib.sh" printf 'Scheduler : fair\n\n' > exp-out || framework_failure -printf 'error: invalid scheduler option: j=k\n' > exp-err || framework_failure +printf 'error: invalid scheduler option: j\n' > exp-err || framework_failure fail=0 diff --git a/tools/virsh-domain.c b/tools/virsh-domain.c index 128e516..cc2eddc 100644 --- a/tools/virsh-domain.c +++ b/tools/virsh-domain.c @@ -3918,16 +3918,14 @@ static const vshCmdOptDef opts_schedinfo[] = { .flags = VSH_OFLAG_REQ, .help = N_("domain name, id or uuid") }, - {.name = "set", - .type = VSH_OT_STRING, - .help = N_("parameter=value") - }, {.name = "weight", .type = VSH_OT_INT, + .flags = VSH_OFLAG_REQ_OPT, .help = N_("weight for XEN_CREDIT") }, {.name = "cap", .type = VSH_OT_INT, + .flags = VSH_OFLAG_REQ_OPT, .help = N_("cap for XEN_CREDIT") }, {.name = "current", @@ -3942,72 +3940,100 @@ static const vshCmdOptDef opts_schedinfo[] = { .type = VSH_OT_BOOL, .help = N_("get/set value from running domain") }, + {.name = "set", + .type = VSH_OT_ARGV, + .flags = VSH_OFLAG_NONE, + .help = N_("parameter=value") + }, {.name = NULL} }; static int +cmdSchedInfoUpdateOne(vshControl *ctl, + virTypedParameterPtr src_params, int nsrc_params, + virTypedParameterPtr *params, + int *nparams, int *maxparams, + const char *field, const char *value) +{ + virTypedParameterPtr param; + int ret = -1; + int i; + + for (i = 0; i < nsrc_params; i++) { + param = &(src_params[i]); + + if (STRNEQ(field, param->field)) + continue; + + if (virTypedParamsAddFromString(params, nparams, maxparams, + field, param->type, + value) < 0) { + vshSaveLibvirtError(); + goto cleanup; + } + ret = 0; + break; + } + + if (ret < 0) + vshError(ctl, _("invalid scheduler option: %s"), field); + + cleanup: + return ret; +} + +static int cmdSchedInfoUpdate(vshControl *ctl, const vshCmd *cmd, virTypedParameterPtr src_params, int nsrc_params, virTypedParameterPtr *update_params) { - const char *set_arg; char *set_field = NULL; char *set_val = NULL; - virTypedParameterPtr param; + const char *val = NULL; + const vshCmdOpt *opt = NULL; virTypedParameterPtr params = NULL; int nparams = 0; int maxparams = 0; int ret = -1; int rv; - int val; - int i; - if (vshCommandOptString(cmd, "set", &set_arg) > 0) { - set_field = vshStrdup(ctl, set_arg); + while ((opt = vshCommandOptArgv(cmd, opt))) { + set_field = vshStrdup(ctl, opt->data); if (!(set_val = strchr(set_field, '='))) { - vshError(ctl, "%s", _("Invalid syntax for --set, expecting name=value")); + vshError(ctl, "%s", _("Invalid syntax for --set, " + "expecting name=value")); goto cleanup; } *set_val = '\0'; set_val++; - } - for (i = 0; i < nsrc_params; i++) { - param = &(src_params[i]); - - /* Legacy 'weight' and 'cap' parameter */ - if (param->type == VIR_TYPED_PARAM_UINT && - (STREQ(param->field, "weight") || STREQ(param->field, "cap")) && - (rv = vshCommandOptInt(cmd, param->field, &val)) != 0) { - if (rv < 0) { - vshError(ctl, _("Invalid value of %s"), param->field); - goto cleanup; - } - - if (virTypedParamsAddUInt(&params, &nparams, &maxparams, - param->field, val) < 0) { - vshSaveLibvirtError(); - goto cleanup; - } + if (cmdSchedInfoUpdateOne(ctl, src_params, nsrc_params, + &params, &nparams, &maxparams, + set_field, set_val) < 0) + goto cleanup; - continue; - } + VIR_FREE(set_field); + } - if (set_field && STREQ(set_field, param->field)) { - if (virTypedParamsAddFromString(&params, &nparams, &maxparams, - set_field, param->type, - set_val) < 0) { - vshSaveLibvirtError(); - goto cleanup; - } + rv = vshCommandOptStringReq(ctl, cmd, "cap", &val); + if (rv < 0 || + (val && + cmdSchedInfoUpdateOne(ctl, src_params, nsrc_params, + &params, &nparams, &maxparams, + "cap", val) < 0)) + goto cleanup; - continue; - } - } + rv = vshCommandOptStringReq(ctl, cmd, "weight", &val); + if (rv < 0 || + (val && + cmdSchedInfoUpdateOne(ctl, src_params, nsrc_params, + &params, &nparams, &maxparams, + "weight", val) < 0)) + goto cleanup; - *update_params = params; ret = nparams; + *update_params = params; params = NULL; cleanup: @@ -4102,15 +4128,6 @@ cmdSchedinfo(vshControl *ctl, const vshCmd *cmd) if (ret == -1) goto cleanup; } else { - /* See if we've tried to --set var=val. If so, the fact that - we reach this point (with update == 0) means that "var" did - not match any of the settable parameters. Report the error. */ - const char *var_value_pair = NULL; - if (vshCommandOptString(cmd, "set", &var_value_pair) > 0) { - vshError(ctl, _("invalid scheduler option: %s"), - var_value_pair); - goto cleanup; - } /* When not doing --set, --live and --config do not mix. */ if (live && config) { vshError(ctl, "%s", diff --git a/tools/virsh.pod b/tools/virsh.pod index b5e632e..6111e58 100644 --- a/tools/virsh.pod +++ b/tools/virsh.pod @@ -1281,8 +1281,8 @@ except that it does some error checking. The editor used can be supplied by the C<$VISUAL> or C<$EDITOR> environment variables, and defaults to C<vi>. -=item B<schedinfo> [I<--set> B<parameter=value>] I<domain> [[I<--config>] -[I<--live>] | [I<--current>]] +=item B<schedinfo> I<domain> [[I<--config>] [I<--live>] | [I<--current>]] +[[I<--set>] B<parameter=value>]... =item B<schedinfo> [I<--weight> B<number>] [I<--cap> B<number>] I<domain> -- 1.8.1.5

2 3

[libvirt] mistake in libvirt-php document
by mahdi faryabi 03 Apr '13

03 Apr '13

hi, in libvirt-php you have a little mistake here : http://libvirt.org/php/api-reference.html#libvirt_image_remove "Function is used to create the image of desired name, size and format. The image will be created in the image path"

1 0

[libvirt] [PATCH v5 0/3] Keep original file label
by Michal Privoznik 03 Apr '13

03 Apr '13

Yet another rework of $subj. I am still not solving atomicity problem for now. See diff to the patches if you want to know what's changed. Michal Privoznik (3): virFile: Add APIs for extended attributes handling virfile: Introduce internal API for managing ACL security_dac: Favour ACLs over chown() configure.ac | 2 + libvirt.spec.in | 1 + m4/virt-acl.m4 | 9 ++ src/Makefile.am | 4 +- src/libvirt_private.syms | 6 + src/security/security_dac.c | 309 +++++++++++++++++++++++++++++++++++++++----- src/util/virfile.c | 295 ++++++++++++++++++++++++++++++++++++++++++ src/util/virfile.h | 28 ++++ 8 files changed, 623 insertions(+), 31 deletions(-) create mode 100644 m4/virt-acl.m4 -- 1.8.1.5

3 14

[libvirt] [PATCH] qemu: Fix crash when updating media with shared device
by Peter Krempa 03 Apr '13

03 Apr '13

Mimic the fix done in 02b9097274d1330c2e1dca7f598880e09b5c2aa0 to fix crash by accessing an already freed structure. Also copy the explaining comment why the pointer can't be accessed any more. --- src/qemu/qemu_driver.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index 96bf235..552a81b 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -6073,14 +6073,17 @@ qemuDomainChangeDiskMediaLive(virDomainObjPtr vm, goto end; ret = qemuDomainChangeEjectableMedia(driver, vm, disk, orig_disk, force); + /* 'disk' must not be accessed now - it has been freed. + * 'orig_disk' now points to the new disk, while 'dev_copy' + * now points to the old disk */ + + /* Need to remove the shared disk entry for the original + * disk src if the operation is either ejecting or updating. + */ if (ret == 0) { dev->data.disk = NULL; - /* Need to remove the shared disk entry for the original - * disk src if the operation is either ejecting or updating. - */ - if (orig_disk->src && STRNEQ_NULLABLE(orig_disk->src, disk->src)) - ignore_value(qemuRemoveSharedDisk(driver, dev_copy->data.disk, - vm->def->name)); + ignore_value(qemuRemoveSharedDisk(driver, dev_copy->data.disk, + vm->def->name)); } break; default: -- 1.8.1.5

2 2

[libvirt] ANNOUNCE: libvirt 0.10.2.4 maintenance release
by Cole Robinson 03 Apr '13

03 Apr '13

libvirt 0.10.2.4 maintenance release is now available. This is libvirt 0.10.2 with additional bugfixes that have accumulated upstream since the initial release. This release can be downloaded at: http://libvirt.org/sources/stable_updates/libvirt-0.10.2.4.tar.gz Changes in this version: * esx: Fix and improve esxListAllDomains function * Fix parsing of SELinux ranges without a category * Separate MCS range parsing from MCS range checking * Fix memory leak on OOM in virSecuritySELinuxMCSFind * qemu: Set migration FD blocking * build: further fixes for broken if_bridge.h * build: work around broken kernel header * Fix SELinux security label test * libxl: Fix setting of disk backend * util: Fix mask for 172.16.0.0 private address range * conf: don't fail to parse <boot> when parsing a single device * Support custom 'svirt_tcg_t' context for TCG based guests * uml: Report error if inotify fails on driver startup (cherry picked from commit 7b97030ad430eb76fcc333652411208fb702e962) * daemon: Preface polkit error output with 'polkit:' * spec: Fix script warning when uninstalling libvirt-client For info about past maintenance releases, see: http://wiki.libvirt.org/page/Maintenance_Releases Thanks, Cole

1 0

[libvirt] [RFC 0/4] target-i386: Add "feature-words" CPU object property
by Eduardo Habkost 03 Apr '13

03 Apr '13

As the work to get X86CPU subclasses static properties defined is taking very long to get reviewed and we risk not getting it finished in 1.5, I am sending an alternative mechanism to allow libvirt to probe for CPU model information. It's interesting to note that this alternative solution is actually _easier_ to use for libvirt, because libvirt logic is already based on CPUID bit values, not "feature names", so it should be very easy to convert the current libvirt code that (incorrectly) query the host CPU directly for available features to instead start QEMU with "-cpu host" and use the "feature-words" array information to find out which CPU features are supported by the host. The property will have two main use cases: - Checking host capabilities, by checking the features of the "host" CPU model; - Checking which features are enabled on each CPU model. This series has two alternative implementations, so both approaches can get feedback: patch 3 has a non-qapi implementation, that simply builds the list and sets the struct fields manually. Patch 4 converts that implementation to a qapi- based implementation. (BTW, I really missed some documentation on the list-related visitor functions, when writing this code) Example output: $ ./QMP/qmp --path=/tmp/m qom-get --path=/machine/unattached/device[1] --property=feature-words item[0].cpuid-register: EDX item[0].cpuid-input-eax: 1 item[0].features: 126614521 item[1].cpuid-register: ECX item[1].cpuid-input-eax: 1 item[1].features: 2155880449 item[2].cpuid-register: EBX item[2].cpuid-input-eax: 7 item[2].features: 0 item[2].cpuid-input-ecx: 0 item[3].cpuid-register: EDX item[3].cpuid-input-eax: 2147483649 item[3].features: 563346425 item[4].cpuid-register: ECX item[4].cpuid-input-eax: 2147483649 item[4].features: 101 item[5].cpuid-register: EDX item[5].cpuid-input-eax: 3221225473 item[5].features: 0 item[6].cpuid-register: EAX item[6].cpuid-input-eax: 1073741825 item[6].features: 0 item[7].cpuid-register: EDX item[7].cpuid-input-eax: 2147483658 item[7].features: 0 I plan to send another RFC as followup soon, that will add: * "-cpu force" (or "nofilter", or something else; I haven't decided about the name, yet) flag, to disable feature filtering based on host capabilities. * "filtered-features" property, listing which features were filtered out because of missing host capabilities (to allow libvirt to implement "enforce" mode in a more reliable way). Eduardo Habkost (4): target-i386: Add ECX information to FeatureWordInfo target-i386: Move feature word array outside get_register_name_32() target-i386: Add "feature-words" property target-i386: Use qapi for x86_cpu_get_feature_words() .gitignore | 2 + Makefile | 11 +++++ Makefile.objs | 5 +++ cpu-qapi-schema.json | 31 ++++++++++++++ target-i386/Makefile.objs | 1 + target-i386/cpu-qapi-schema.json | 31 ++++++++++++++ target-i386/cpu.c | 92 +++++++++++++++++++++++++++++++++------- 7 files changed, 158 insertions(+), 15 deletions(-) create mode 100644 cpu-qapi-schema.json create mode 100644 target-i386/cpu-qapi-schema.json -- 1.8.1.4

1 5