- Devel - libvirt List Archives

[libvirt] [PATCH] Fix sparse volume allocation reporting
by Cole Robinson 03 Apr '09

03 Apr '09

I have a sparse volume with a capacity of 1000M, and an allocation of a little over 750M. 'du' prints the correct sizes, but 'virsh vol-dumpxml' shows: <capacity>1048576000</capacity> <allocation>6406307840</allocation> This is because we were calculating the allocation size using the requested fs block size, instead of what stat actually uses as a block size (DEV_BSIZE in sys/params.h). sys/params.h looks to be present in mingw32-runtime, so I didn't add a configure check for it. The attached patch fixes allocation listing for me. Thanks, Cole

3 7

[libvirt] PATCH: Fix crash / cleanup bugs in sVirt verification
by Daniel P. Berrange 03 Apr '09

03 Apr '09

If defining a VM without any <seclabel> element present, the new verification code will crash as it deferences NULL, so we just need to skip over that. Second, if one does a roundtrip of a live VM config virsh dumpxml foo > foo.xml virsh define foo.xml and then stop and start the guest, it will now fail complaining that the security label is already defined. This is because the XML parser is mistakenly parsing the 'model' attribute even for VMs whose label is defined to be dynamic. This then exposed anothe bug in the qemudStartVMDaemon() method where it would not properly cleanup after a failed launch, failing to reset the dynamic allocated security label, and also failing to close a logfil file descriptor. This patch fixes all of these problems Daniel Index: src/domain_conf.c =================================================================== RCS file: /data/cvs/libvirt/src/domain_conf.c,v retrieving revision 1.74 diff -u -p -r1.74 domain_conf.c --- src/domain_conf.c 3 Apr 2009 10:55:51 -0000 1.74 +++ src/domain_conf.c 3 Apr 2009 12:14:54 -0000 @@ -1859,15 +1859,6 @@ virSecurityLabelDefParseXML(virConnectPt if (virXPathNode(conn, "./seclabel", ctxt) == NULL) return 0; - p = virXPathStringLimit(conn, "string(./seclabel/@model)", - VIR_SECURITY_MODEL_BUFLEN-1, ctxt); - if (p == NULL) { - virDomainReportError(conn, VIR_ERR_XML_ERROR, - "%s", _("missing security model")); - goto error; - } - def->seclabel.model = p; - p = virXPathStringLimit(conn, "string(./seclabel/@type)", VIR_SECURITY_LABEL_BUFLEN-1, ctxt); if (p == NULL) { @@ -1888,6 +1879,14 @@ virSecurityLabelDefParseXML(virConnectPt */ if (def->seclabel.type == VIR_DOMAIN_SECLABEL_STATIC || !(flags & VIR_DOMAIN_XML_INACTIVE)) { + p = virXPathStringLimit(conn, "string(./seclabel/@model)", + VIR_SECURITY_MODEL_BUFLEN-1, ctxt); + if (p == NULL) { + virDomainReportError(conn, VIR_ERR_XML_ERROR, + "%s", _("missing security model")); + goto error; + } + def->seclabel.model = p; p = virXPathStringLimit(conn, "string(./seclabel/label[1])", VIR_SECURITY_LABEL_BUFLEN-1, ctxt); @@ -1905,8 +1904,11 @@ virSecurityLabelDefParseXML(virConnectPt !(flags & VIR_DOMAIN_XML_INACTIVE)) { p = virXPathStringLimit(conn, "string(./seclabel/imagelabel[1])", VIR_SECURITY_LABEL_BUFLEN-1, ctxt); - if (p == NULL) + if (p == NULL) { + virDomainReportError(conn, VIR_ERR_XML_ERROR, + _("security imagelabel is missing")); goto error; + } def->seclabel.imagelabel = p; } Index: src/qemu_driver.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_driver.c,v retrieving revision 1.224 diff -u -p -r1.224 qemu_driver.c --- src/qemu_driver.c 3 Apr 2009 10:55:51 -0000 1.224 +++ src/qemu_driver.c 3 Apr 2009 12:14:54 -0000 @@ -332,7 +332,7 @@ qemudReconnectVMs(struct qemud_driver *d } if ((vm->logfile = qemudLogFD(NULL, driver->logDir, vm->def->name)) < 0) - return -1; + goto next_error; if (vm->def->id >= driver->nextvmid) driver->nextvmid = vm->def->id + 1; @@ -344,9 +344,12 @@ next_error: /* we failed to reconnect the vm so remove it's traces */ vm->def->id = -1; qemudRemoveDomainStatus(NULL, driver, vm); - virDomainDefFree(vm->def); - vm->def = vm->newDef; - vm->newDef = NULL; + /* Restore orignal def, if we'd loaded a live one */ + if (vm->newDef) { + virDomainDefFree(vm->def); + vm->def = vm->newDef; + vm->newDef = NULL; + } next: virDomainObjUnlock(vm); if (status) @@ -1319,14 +1322,6 @@ static int qemudStartVMDaemon(virConnect hookData.vm = vm; hookData.driver = driver; - /* If you are using a SecurityDriver with dynamic labelling, - then generate a security label for isolation */ - if (vm->def->seclabel.type == VIR_DOMAIN_SECLABEL_DYNAMIC && - driver->securityDriver && - driver->securityDriver->domainGenSecurityLabel && - driver->securityDriver->domainGenSecurityLabel(conn, vm) < 0) - return -1; - FD_ZERO(&keepfd); if (virDomainIsActive(vm)) { @@ -1335,6 +1330,14 @@ static int qemudStartVMDaemon(virConnect return -1; } + /* If you are using a SecurityDriver with dynamic labelling, + then generate a security label for isolation */ + if (vm->def->seclabel.type == VIR_DOMAIN_SECLABEL_DYNAMIC && + driver->securityDriver && + driver->securityDriver->domainGenSecurityLabel && + driver->securityDriver->domainGenSecurityLabel(conn, vm) < 0) + return -1; + if (vm->def->graphics && vm->def->graphics->type == VIR_DOMAIN_GRAPHICS_TYPE_VNC && vm->def->graphics->data.vnc.autoport) { @@ -1342,7 +1345,7 @@ static int qemudStartVMDaemon(virConnect if (port < 0) { qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, "%s", _("Unable to find an unused VNC port")); - return -1; + goto cleanup; } vm->def->graphics->data.vnc.port = port; } @@ -1351,17 +1354,17 @@ static int qemudStartVMDaemon(virConnect virReportSystemError(conn, errno, _("cannot create log directory %s"), driver->logDir); - return -1; + goto cleanup; } if((vm->logfile = qemudLogFD(conn, driver->logDir, vm->def->name)) < 0) - return -1; + goto cleanup; emulator = vm->def->emulator; if (!emulator) emulator = virDomainDefDefaultEmulator(conn, vm->def, driver->caps); if (!emulator) - return -1; + goto cleanup; /* Make sure the binary we are about to try exec'ing exists. * Technically we could catch the exec() failure, but that's @@ -1371,7 +1374,7 @@ static int qemudStartVMDaemon(virConnect virReportSystemError(conn, errno, _("Cannot find QEMU binary %s"), emulator); - return -1; + goto cleanup; } if (qemudExtractVersionInfo(emulator, @@ -1380,21 +1383,17 @@ static int qemudStartVMDaemon(virConnect qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR, _("Cannot determine QEMU argv syntax %s"), emulator); - return -1; + goto cleanup; } if (qemuPrepareHostDevices(conn, vm->def) < 0) - return -1; + goto cleanup; vm->def->id = driver->nextvmid++; if (qemudBuildCommandLine(conn, driver, vm, qemuCmdFlags, &argv, &progenv, - &tapfds, &ntapfds, migrateFrom) < 0) { - close(vm->logfile); - vm->def->id = -1; - vm->logfile = -1; - return -1; - } + &tapfds, &ntapfds, migrateFrom) < 0) + goto cleanup; tmp = progenv; while (*tmp) { @@ -1457,12 +1456,8 @@ static int qemudStartVMDaemon(virConnect "%s", _("Unable to daemonize QEMU process")); ret = -1; } - } - - if (ret == 0) { vm->state = migrateFrom ? VIR_DOMAIN_PAUSED : VIR_DOMAIN_RUNNING; - } else - vm->def->id = -1; + } for (i = 0 ; argv[i] ; i++) VIR_FREE(argv[i]); @@ -1479,19 +1474,38 @@ static int qemudStartVMDaemon(virConnect VIR_FREE(tapfds); } - if (ret == 0) { - if ((qemudWaitForMonitor(conn, driver, vm, pos) < 0) || - (qemudDetectVcpuPIDs(conn, vm) < 0) || - (qemudInitCpus(conn, vm, migrateFrom) < 0) || - (qemudInitPasswords(conn, driver, vm) < 0) || - (qemudDomainSetMemoryBalloon(conn, vm, vm->def->memory) < 0) || - (qemudSaveDomainStatus(conn, qemu_driver, vm) < 0)) { - qemudShutdownVMDaemon(conn, driver, vm); - return -1; - } + if (ret == -1) + goto cleanup; + + if ((qemudWaitForMonitor(conn, driver, vm, pos) < 0) || + (qemudDetectVcpuPIDs(conn, vm) < 0) || + (qemudInitCpus(conn, vm, migrateFrom) < 0) || + (qemudInitPasswords(conn, driver, vm) < 0) || + (qemudDomainSetMemoryBalloon(conn, vm, vm->def->memory) < 0) || + (qemudSaveDomainStatus(conn, qemu_driver, vm) < 0)) { + qemudShutdownVMDaemon(conn, driver, vm); + ret = -1; + /* No need for 'goto cleanup' now since qemudShutdownVMDaemon does enough */ } return ret; + +cleanup: + if (vm->def->seclabel.type == VIR_DOMAIN_SECLABEL_DYNAMIC) { + VIR_FREE(vm->def->seclabel.model); + VIR_FREE(vm->def->seclabel.label); + VIR_FREE(vm->def->seclabel.imagelabel); + } + if (vm->def->graphics && + vm->def->graphics->type == VIR_DOMAIN_GRAPHICS_TYPE_VNC && + vm->def->graphics->data.vnc.autoport) + vm->def->graphics->data.vnc.port = -1; + if (vm->logfile != -1) { + close(vm->logfile); + vm->logfile = -1; + } + vm->def->id = -1; + return -1; } Index: src/security.c =================================================================== RCS file: /data/cvs/libvirt/src/security.c,v retrieving revision 1.3 diff -u -p -r1.3 security.c --- src/security.c 3 Apr 2009 10:55:51 -0000 1.3 +++ src/security.c 3 Apr 2009 12:14:54 -0000 @@ -33,7 +33,8 @@ virSecurityDriverVerify(virConnectPtr co unsigned int i; const virSecurityLabelDefPtr secdef = &def->seclabel; - if (STREQ(secdef->model, "none")) + if (!secdef->model || + STREQ(secdef->model, "none")) return 0; for (i = 0; security_drivers[i] != NULL ; i++) { @@ -42,7 +43,7 @@ virSecurityDriverVerify(virConnectPtr co } } virSecurityReportError(conn, VIR_ERR_XML_ERROR, - _("invalid security model")); + _("invalid security model '%s'"), secdef->model); return -1; } Index: src/security_selinux.c =================================================================== RCS file: /data/cvs/libvirt/src/security_selinux.c,v retrieving revision 1.5 diff -u -p -r1.5 security_selinux.c --- src/security_selinux.c 3 Apr 2009 10:55:51 -0000 1.5 +++ src/security_selinux.c 3 Apr 2009 12:14:54 -0000 @@ -162,11 +162,12 @@ SELinuxGenSecurityLabel(virConnectPtr co char *scontext = NULL; int c1 = 0; int c2 = 0; - if ( ( vm->def->seclabel.label ) || - ( vm->def->seclabel.model ) || - ( vm->def->seclabel.imagelabel )) { - virSecurityReportError(conn, VIR_ERR_ERROR, - "%s", _("security labellin already defined for VM")); + + if (vm->def->seclabel.label || + vm->def->seclabel.model || + vm->def->seclabel.imagelabel) { + virSecurityReportError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("security label already defined for VM")); return rc; } @@ -197,7 +198,7 @@ SELinuxGenSecurityLabel(virConnectPtr co goto err; } vm->def->seclabel.model = strdup(SECURITY_SELINUX_NAME); - if (! vm->def->seclabel.model) { + if (!vm->def->seclabel.model) { virReportOOMError(conn); goto err; } -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] PATCH: Xen PCI device passthrough
by Daniel P. Berrange 03 Apr '09

03 Apr '09

This patch implements PCI device passthrough for Xen. In pci.c there is a fair bit of refactoring to better support both pcistub.ko and pciback.ko. The latter has two extra sysfs files, add_slot and remove_slot for associating the driver with a specific PCI device domain:bus:slot:func address, whereas pcistub can only associate based on vendor:product ID The Xen driver gets an implementation of the dettach/reattach APIs for node devices. The XML <-> SEXPR conversion gets adapted to cope with PCI devices The XM <-> XML conversion gets adapted to load/save PCI device info in /etc/xen config files Daniel Index: src/pci.c =================================================================== RCS file: /data/cvs/libvirt/src/pci.c,v retrieving revision 1.4 diff -u -p -r1.4 pci.c --- src/pci.c 3 Mar 2009 17:00:18 -0000 1.4 +++ src/pci.c 30 Mar 2009 13:09:44 -0000 @@ -614,46 +614,80 @@ pciResetDevice(virConnectPtr conn, pciDe return ret; } -static int -pciBindDeviceToStub(virConnectPtr conn, pciDevice *dev, const char *stub_module) + +static void +pciDriverDir(char *buf, size_t buflen, const char *driver) { - char stub_dir[PATH_MAX]; - char path[PATH_MAX]; + snprintf(buf, buflen, PCI_SYSFS "drivers/%s", driver); +} - snprintf(stub_dir, sizeof(stub_dir), PCI_SYSFS "drivers/%s", stub_module); +static void +pciDriverFile(char *buf, size_t buflen, const char *driver, const char *file) +{ + snprintf(buf, buflen, PCI_SYSFS "drivers/%s/%s", driver, file); +} + +static void +pciDeviceFile(char *buf, size_t buflen, const char *device, const char *file) +{ + snprintf(buf, buflen, PCI_SYSFS "devices/%s/%s", device, file); +} - /* Try loading the stub module if it isn't already loaded; - * Do not return an error if the stub module is not available. - */ - if (!virFileExists(stub_dir)) { - const char *const modprobeargv[] = { MODPROBE, stub_module, NULL }; - if (virRun(conn, modprobeargv, NULL) < 0) { +static const char * +pciFindStubDriver(virConnectPtr conn) +{ + char drvpath[PATH_MAX]; + int probed = 0; + +recheck: + pciDriverDir(drvpath, sizeof(drvpath), "pci-stub"); + if (virFileExists(drvpath)) + return "pci-stub"; + pciDriverDir(drvpath, sizeof(drvpath), "pciback"); + if (virFileExists(drvpath)) + return "pciback"; + + if (!probed) { + const char *const stubprobe[] = { MODPROBE, "pci-stub", NULL }; + const char *const backprobe[] = { MODPROBE, "pciback", NULL }; + + probed = 1; + if (virRun(conn, stubprobe, NULL) < 0 && + virRun(conn, backprobe, NULL) < 0) { char ebuf[1024]; - VIR_WARN(_("modprobe %s failed: %s"), stub_module, + VIR_WARN(_("failed to load pci-stub or pciback drivers: %s"), virStrerror(errno, ebuf, sizeof ebuf)); + return 0; } + + goto recheck; } - if (!virFileExists(stub_dir)) { - VIR_WARN(_("%s module not available, cannot bind device %s to it"), - stub_module, dev->name); - } else { - /* Add the PCI device ID to the stub's dynamic ID table; - * this is needed to allow us to bind the device to the stub. - * Note: if the device is not currently bound to any driver, - * stub will immediately be bound to the device. Also, note - * that if a new device with this ID is hotplugged, or if a probe - * is triggered for such a device, it will also be immediately - * bound by the stub. - */ - snprintf(path, sizeof(path), "%s/new_id", stub_dir); - if (virFileWriteStr(path, dev->id) < 0) { - virReportSystemError(conn, errno, - _("Failed to add PCI device ID '%s' to %s"), - dev->id, stub_module); - return -1; - } + return NULL; +} + + +static int +pciBindDeviceToStub(virConnectPtr conn, pciDevice *dev, const char *driver) +{ + char drvdir[PATH_MAX]; + char path[PATH_MAX]; + + /* Add the PCI device ID to the stub's dynamic ID table; + * this is needed to allow us to bind the device to the stub. + * Note: if the device is not currently bound to any driver, + * stub will immediately be bound to the device. Also, note + * that if a new device with this ID is hotplugged, or if a probe + * is triggered for such a device, it will also be immediately + * bound by the stub. + */ + pciDriverFile(path, sizeof(path), driver, "new_id"); + if (virFileWriteStr(path, dev->id) < 0) { + virReportSystemError(conn, errno, + _("Failed to add PCI device ID '%s' to %s"), + dev->id, driver); + return -1; } /* If the device is already bound to a driver, unbind it. @@ -661,77 +695,100 @@ pciBindDeviceToStub(virConnectPtr conn, * PCI device happens to be IDE controller for the disk hosting * your root filesystem. */ - snprintf(path, sizeof(path), - PCI_SYSFS "devices/%s/driver/unbind", dev->name); + pciDeviceFile(path, sizeof(path), dev->name, "driver/unbind"); if (virFileExists(path) && virFileWriteStr(path, dev->name) < 0) { virReportSystemError(conn, errno, _("Failed to unbind PCI device '%s'"), dev->name); return -1; } - if (virFileExists(stub_dir)) { - /* If the device isn't already bound to pci-stub, try binding it now. - */ - snprintf(path, sizeof(path), PCI_SYSFS "devices/%s/driver", dev->name); - if (!virFileLinkPointsTo(path, stub_dir)) { - snprintf(path, sizeof(path), "%s/bind", stub_dir); - if (virFileWriteStr(path, dev->name) < 0) { - virReportSystemError(conn, errno, - _("Failed to bind PCI device '%s' to %s"), - dev->name, stub_module); - return -1; - } + /* If the device isn't already bound to pci-stub, try binding it now. + */ + pciDriverDir(drvdir, sizeof(drvdir), driver); + pciDeviceFile(path, sizeof(path), dev->name, "driver"); + if (!virFileLinkPointsTo(path, drvdir)) { + /* Xen's pciback.ko wants you to use new_slot first */ + pciDriverFile(path, sizeof(path), driver, "new_slot"); + if (virFileExists(path) && virFileWriteStr(path, dev->name) < 0) { + virReportSystemError(conn, errno, + _("Failed to add slot for PCI device '%s' to %s"), + dev->name, driver); + return -1; } - /* If 'remove_id' exists, remove the device id from pci-stub's dynamic - * ID table so that 'drivers_probe' works below. - */ - snprintf(path, sizeof(path), "%s/remove_id", stub_dir); - if (virFileExists(path) && virFileWriteStr(path, dev->id) < 0) { + pciDriverFile(path, sizeof(path), driver, "bind"); + if (virFileWriteStr(path, dev->name) < 0) { virReportSystemError(conn, errno, - _("Failed to remove PCI ID '%s' from %s"), - dev->id, stub_module); + _("Failed to bind PCI device '%s' to %s"), + dev->name, driver); return -1; } } + /* If 'remove_id' exists, remove the device id from pci-stub's dynamic + * ID table so that 'drivers_probe' works below. + */ + pciDriverFile(path, sizeof(path), driver, "remove_id"); + if (virFileExists(path) && virFileWriteStr(path, dev->id) < 0) { + virReportSystemError(conn, errno, + _("Failed to remove PCI ID '%s' from %s"), + dev->id, driver); + return -1; + } + return 0; } int pciDettachDevice(virConnectPtr conn, pciDevice *dev) { - return pciBindDeviceToStub(conn, dev, "pci-stub"); + const char *driver = pciFindStubDriver(conn); + if (!driver) { + pciReportError(conn, VIR_ERR_INTERNAL_ERROR, "%s", + _("cannot find any PCI stub module")); + return -1; + } + + return pciBindDeviceToStub(conn, dev, driver); } static int -pciUnBindDeviceFromStub(virConnectPtr conn, pciDevice *dev, const char *stub_module) +pciUnBindDeviceFromStub(virConnectPtr conn, pciDevice *dev, const char *driver) { - char stub_dir[PATH_MAX]; + char drvdir[PATH_MAX]; char path[PATH_MAX]; - snprintf(stub_dir, sizeof(stub_dir), PCI_SYSFS "drivers/%s", stub_module); - /* If the device is bound to stub, unbind it. */ - snprintf(path, sizeof(path), PCI_SYSFS "devices/%s/driver", dev->name); - if (virFileExists(stub_dir) && virFileLinkPointsTo(path, stub_dir)) { - snprintf(path, sizeof(path), "%s/unbind", stub_dir); + pciDriverDir(drvdir, sizeof(drvdir), driver); + pciDeviceFile(path, sizeof(path), dev->name, "driver"); + if (virFileExists(drvdir) && virFileLinkPointsTo(path, drvdir)) { + pciDriverFile(path, sizeof(path), driver, "unbind"); if (virFileWriteStr(path, dev->name) < 0) { virReportSystemError(conn, errno, _("Failed to bind PCI device '%s' to %s"), - dev->name, stub_module); + dev->name, driver); return -1; } } + /* Xen's pciback.ko wants you to use remove_slot on the specific device */ + pciDriverFile(path, sizeof(path), driver, "remove_slot"); + if (virFileExists(path) && virFileWriteStr(path, dev->name) < 0) { + virReportSystemError(conn, errno, + _("Failed to remove slot for PCI device '%s' to %s"), + dev->name, driver); + return -1; + } + + /* Trigger a re-probe of the device is not in the stub's dynamic * ID table. If the stub is available, but 'remove_id' isn't * available, then re-probing would just cause the device to be * re-bound to the stub. */ - snprintf(path, sizeof(path), "%s/remove_id", stub_dir); - if (!virFileExists(stub_dir) || virFileExists(path)) { + pciDriverFile(path, sizeof(path), driver, "remove_id"); + if (!virFileExists(drvdir) || virFileExists(path)) { if (virFileWriteStr(PCI_SYSFS "drivers_probe", dev->name) < 0) { virReportSystemError(conn, errno, _("Failed to trigger a re-probe for PCI device '%s'"), @@ -746,7 +803,14 @@ pciUnBindDeviceFromStub(virConnectPtr co int pciReAttachDevice(virConnectPtr conn, pciDevice *dev) { - return pciUnBindDeviceFromStub(conn, dev, "pci-stub"); + const char *driver = pciFindStubDriver(conn); + if (!driver) { + pciReportError(conn, VIR_ERR_INTERNAL_ERROR, "%s", + _("cannot find any PCI stub module")); + return -1; + } + + return pciUnBindDeviceFromStub(conn, dev, driver); } static char * Index: src/xen_unified.c =================================================================== RCS file: /data/cvs/libvirt/src/xen_unified.c,v retrieving revision 1.83 diff -u -p -r1.83 xen_unified.c --- src/xen_unified.c 5 Feb 2009 16:03:11 -0000 1.83 +++ src/xen_unified.c 30 Mar 2009 13:09:44 -0000 @@ -43,6 +43,8 @@ #include "xml.h" #include "util.h" #include "memory.h" +#include "node_device_conf.h" +#include "pci.h" #define VIR_FROM_THIS VIR_FROM_XEN @@ -1420,6 +1422,123 @@ xenUnifiedDomainEventDeregister (virConn return ret; } + +static int +xenUnifiedNodeDeviceGetPciInfo (virNodeDevicePtr dev, + unsigned *domain, + unsigned *bus, + unsigned *slot, + unsigned *function) +{ + virNodeDeviceDefPtr def = NULL; + virNodeDevCapsDefPtr cap; + char *xml = NULL; + int ret = -1; + + xml = virNodeDeviceGetXMLDesc(dev, 0); + if (!xml) + goto out; + + def = virNodeDeviceDefParseString(dev->conn, xml); + if (!def) + goto out; + + cap = def->caps; + while (cap) { + if (cap->type == VIR_NODE_DEV_CAP_PCI_DEV) { + *domain = cap->data.pci_dev.domain; + *bus = cap->data.pci_dev.bus; + *slot = cap->data.pci_dev.slot; + *function = cap->data.pci_dev.function; + break; + } + + cap = cap->next; + } + + if (!cap) { + xenUnifiedError(dev->conn, VIR_ERR_INVALID_ARG, + _("device %s is not a PCI device"), dev->name); + goto out; + } + + ret = 0; +out: + virNodeDeviceDefFree(def); + VIR_FREE(xml); + return ret; +} + +static int +xenUnifiedNodeDeviceDettach (virNodeDevicePtr dev) +{ + pciDevice *pci; + unsigned domain, bus, slot, function; + int ret = -1; + + if (xenUnifiedNodeDeviceGetPciInfo(dev, &domain, &bus, &slot, &function) < 0) + return -1; + + pci = pciGetDevice(dev->conn, domain, bus, slot, function); + if (!pci) + return -1; + + if (pciDettachDevice(dev->conn, pci) < 0) + goto out; + + ret = 0; +out: + pciFreeDevice(dev->conn, pci); + return ret; +} + +static int +xenUnifiedNodeDeviceReAttach (virNodeDevicePtr dev) +{ + pciDevice *pci; + unsigned domain, bus, slot, function; + int ret = -1; + + if (xenUnifiedNodeDeviceGetPciInfo(dev, &domain, &bus, &slot, &function) < 0) + return -1; + + pci = pciGetDevice(dev->conn, domain, bus, slot, function); + if (!pci) + return -1; + + if (pciReAttachDevice(dev->conn, pci) < 0) + goto out; + + ret = 0; +out: + pciFreeDevice(dev->conn, pci); + return ret; +} + +static int +xenUnifiedNodeDeviceReset (virNodeDevicePtr dev) +{ + pciDevice *pci; + unsigned domain, bus, slot, function; + int ret = -1; + + if (xenUnifiedNodeDeviceGetPciInfo(dev, &domain, &bus, &slot, &function) < 0) + return -1; + + pci = pciGetDevice(dev->conn, domain, bus, slot, function); + if (!pci) + return -1; + + if (pciResetDevice(dev->conn, pci) < 0) + goto out; + + ret = 0; +out: + pciFreeDevice(dev->conn, pci); + return ret; +} + + /*----- Register with libvirt.c, and initialise Xen drivers. -----*/ /* The interface which we export upwards to libvirt.c. */ @@ -1481,6 +1600,9 @@ static virDriver xenUnifiedDriver = { .getFreeMemory = xenUnifiedNodeGetFreeMemory, .domainEventRegister = xenUnifiedDomainEventRegister, .domainEventDeregister = xenUnifiedDomainEventDeregister, + .nodeDeviceDettach = xenUnifiedNodeDeviceDettach, + .nodeDeviceReAttach = xenUnifiedNodeDeviceReAttach, + .nodeDeviceReset = xenUnifiedNodeDeviceReset, }; /** Index: src/xend_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/xend_internal.c,v retrieving revision 1.253 diff -u -p -r1.253 xend_internal.c --- src/xend_internal.c 10 Mar 2009 11:13:32 -0000 1.253 +++ src/xend_internal.c 30 Mar 2009 13:09:45 -0000 @@ -92,6 +92,11 @@ xenDaemonFormatSxprNet(virConnectPtr con int xendConfigVersion, int isAttach); static int +xenDaemonFormatSxprOnePCI(virConnectPtr conn, + virDomainHostdevDefPtr def, + virBufferPtr buf); + +static int virDomainXMLDevID(virDomainPtr domain, virDomainDeviceDefPtr dev, char *class, @@ -2145,6 +2150,131 @@ error: return -1; } +/** + * xenDaemonParseSxprPCI + * @conn: connection + * @root: root sexpr + * + * This parses out block devices from the domain sexpr + * + * Returns 0 if successful or -1 if failed. + */ +static int +xenDaemonParseSxprPCI(virConnectPtr conn, + virDomainDefPtr def, + const struct sexpr *root) +{ + const struct sexpr *cur, *tmp = NULL, *node; + virDomainHostdevDefPtr dev = NULL; + + /* + * With the (domain ...) block we have the following odd setup + * + * (device + * (pci + * (dev (domain 0x0000) (bus 0x00) (slot 0x1b) (func 0x0)) + * (dev (domain 0x0000) (bus 0x00) (slot 0x13) (func 0x0)) + * ) + * ) + * + * Normally there is one (device ...) block per device, but in + * wierd world of Xen PCI, once (device ...) covers multiple + * devices. + */ + + for (cur = root; cur->kind == SEXPR_CONS; cur = cur->u.s.cdr) { + node = cur->u.s.car; + if ((tmp = sexpr_lookup(node, "device/pci")) != NULL) + break; + } + + if (!tmp) + return 0; + + for (cur = tmp; cur->kind == SEXPR_CONS; cur = cur->u.s.cdr) { + const char *domain = NULL; + const char *bus = NULL; + const char *slot = NULL; + const char *func = NULL; + int domainID; + int busID; + int slotID; + int funcID; + + node = cur->u.s.car; + if (!sexpr_lookup(node, "dev")) + continue; + + if (!(domain = sexpr_node(node, "dev/domain"))) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("missing PCI domain")); + goto error; + } + if (!(bus = sexpr_node(node, "dev/bus"))) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("missing PCI bus")); + goto error; + } + if (!(slot = sexpr_node(node, "dev/slot"))) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("missing PCI slot")); + goto error; + } + if (!(func = sexpr_node(node, "dev/func"))) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("missing PCI func")); + goto error; + } + + if (virStrToLong_i(domain, NULL, 0, &domainID) < 0) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse PCI domain '%s'"), domain); + goto error; + } + if (virStrToLong_i(bus, NULL, 0, &busID) < 0) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse PCI bus '%s'"), bus); + goto error; + } + if (virStrToLong_i(slot, NULL, 0, &slotID) < 0) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse PCI slot '%s'"), slot); + goto error; + } + if (virStrToLong_i(func, NULL, 0, &funcID) < 0) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse PCI func '%s'"), func); + goto error; + } + + if (VIR_ALLOC(dev) < 0) + goto no_memory; + + dev->mode = VIR_DOMAIN_HOSTDEV_MODE_SUBSYS; + dev->managed = 0; + dev->source.subsys.type = VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI; + dev->source.subsys.u.pci.domain = domainID; + dev->source.subsys.u.pci.bus = busID; + dev->source.subsys.u.pci.slot = slotID; + dev->source.subsys.u.pci.function = funcID; + + if (VIR_REALLOC_N(def->hostdevs, def->nhostdevs+1) < 0) { + goto no_memory; + } + + def->hostdevs[def->nhostdevs++] = dev; + } + + return 0; + +no_memory: + virReportOOMError(conn); + +error: + virDomainHostdevDefFree(dev); + return -1; +} + /** * xenDaemonParseSxpr: @@ -2310,6 +2440,9 @@ xenDaemonParseSxpr(virConnectPtr conn, if (xenDaemonParseSxprNets(conn, def, root) < 0) goto error; + if (xenDaemonParseSxprPCI(conn, def, root) < 0) + goto error; + /* New style graphics device config */ if (xenDaemonParseSxprGraphicsNew(conn, def, root) < 0) goto error; @@ -3953,6 +4086,20 @@ xenDaemonAttachDevice(virDomainPtr domai goto cleanup; break; + case VIR_DOMAIN_DEVICE_HOSTDEV: + if (dev->data.hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + dev->data.hostdev->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) { + if (xenDaemonFormatSxprOnePCI(domain->conn, + dev->data.hostdev, + &buf) < 0) + goto cleanup; + } else { + virXendError(domain->conn, VIR_ERR_NO_SUPPORT, "%s", + _("unsupported device type")); + goto cleanup; + } + break; + default: virXendError(domain->conn, VIR_ERR_NO_SUPPORT, "%s", _("unsupported device type")); @@ -5263,6 +5410,85 @@ xenDaemonFormatSxprNet(virConnectPtr con return 0; } + +static void +xenDaemonFormatSxprPCI(virDomainHostdevDefPtr def, + virBufferPtr buf) +{ + virBufferVSprintf(buf, "(dev (domain 0x%04x)(bus 0x%02x)(slot 0x%02x)(func 0x%x))", + def->source.subsys.u.pci.domain, + def->source.subsys.u.pci.bus, + def->source.subsys.u.pci.slot, + def->source.subsys.u.pci.function); +} + +static int +xenDaemonFormatSxprOnePCI(virConnectPtr conn, + virDomainHostdevDefPtr def, + virBufferPtr buf) +{ + if (def->managed) { + virXendError(conn, VIR_ERR_NO_SUPPORT, "%s", + _("managed PCI devices not supported with XenD")); + return -1; + } + + virBufferAddLit(buf, "(pci "); + xenDaemonFormatSxprPCI(def, buf); + virBufferAddLit(buf, ")"); + + return 0; +} + +static int +xenDaemonFormatSxprAllPCI(virConnectPtr conn, + virDomainDefPtr def, + virBufferPtr buf) +{ + int hasPCI = 0; + int i; + + for (i = 0 ; i < def->nhostdevs ; i++) + if (def->hostdevs[i]->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + def->hostdevs[i]->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) + hasPCI = 1; + + if (!hasPCI) + return 0; + + /* + * With the (domain ...) block we have the following odd setup + * + * (device + * (pci + * (dev (domain 0x0000) (bus 0x00) (slot 0x1b) (func 0x0)) + * (dev (domain 0x0000) (bus 0x00) (slot 0x13) (func 0x0)) + * ) + * ) + * + * Normally there is one (device ...) block per device, but in + * wierd world of Xen PCI, once (device ...) covers multiple + * devices. + */ + + virBufferAddLit(buf, "(device (pci "); + for (i = 0 ; i < def->nhostdevs ; i++) { + if (def->hostdevs[i]->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + def->hostdevs[i]->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) { + if (def->hostdevs[i]->managed) { + virXendError(conn, VIR_ERR_NO_SUPPORT, "%s", + _("managed PCI devices not supported with XenD")); + return -1; + } + + xenDaemonFormatSxprPCI(def->hostdevs[i], buf); + } + } + virBufferAddLit(buf, "))"); + + return 0; +} + int xenDaemonFormatSxprSound(virConnectPtr conn, virDomainDefPtr def, @@ -5529,6 +5755,9 @@ xenDaemonFormatSxpr(virConnectPtr conn, &buf, hvm, xendConfigVersion, 0) < 0) goto error; + if (xenDaemonFormatSxprAllPCI(conn, def, &buf) < 0) + goto error; + /* New style PV graphics config xen >= 3.0.4, * or HVM graphics config xen >= 3.0.5 */ if ((xendConfigVersion >= XEND_CONFIG_MIN_VERS_PVFB_NEWCONF && !hvm) || @@ -5611,6 +5840,9 @@ virDomainXMLDevID(virDomainPtr domain, strncpy(ref, xref, ref_len); free(xref); ref[ref_len - 1] = '\0'; + } else if (dev->type == VIR_DOMAIN_DEVICE_HOSTDEV && + dev->data.hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + dev->data.hostdev->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) { } else { virXendError(NULL, VIR_ERR_NO_SUPPORT, "%s", _("hotplug of device type not supported")); Index: src/xm_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/xm_internal.c,v retrieving revision 1.118 diff -u -p -r1.118 xm_internal.c --- src/xm_internal.c 24 Mar 2009 11:16:29 -0000 1.118 +++ src/xm_internal.c 30 Mar 2009 13:09:45 -0000 @@ -673,6 +673,7 @@ xenXMDomainConfigParse(virConnectPtr con virDomainDiskDefPtr disk = NULL; virDomainNetDefPtr net = NULL; virDomainGraphicsDefPtr graphics = NULL; + virDomainHostdevDefPtr hostdev = NULL; int i; const char *defaultArch, *defaultMachine; @@ -1115,6 +1116,88 @@ xenXMDomainConfigParse(virConnectPtr con } } + list = virConfGetValue(conf, "pci"); + if (list && list->type == VIR_CONF_LIST) { + list = list->list; + while (list) { + char domain[5]; + char bus[3]; + char slot[3]; + char func[2]; + char *key, *nextkey; + int domainID; + int busID; + int slotID; + int funcID; + + domain[0] = bus[0] = slot[0] = func[0] = '\0'; + + if ((list->type != VIR_CONF_STRING) || (list->str == NULL)) + goto skippci; + + /* pci=['0000:00:1b.0','0000:00:13.0'] */ + key = list->str; + if (!(key = list->str)) + goto skippci; + if (!(nextkey = strchr(key, ':'))) + goto skippci; + + if ((nextkey - key) > (sizeof(domain)-1)) + goto skippci; + + strncpy(domain, key, sizeof(domain)); + domain[sizeof(domain)-1] = '\0'; + + key = nextkey + 1; + if (!(nextkey = strchr(key, ':'))) + goto skippci; + + strncpy(bus, key, sizeof(bus)); + bus[sizeof(bus)-1] = '\0'; + + key = nextkey + 1; + if (!(nextkey = strchr(key, '.'))) + goto skippci; + + strncpy(slot, key, sizeof(slot)); + slot[sizeof(slot)-1] = '\0'; + + key = nextkey + 1; + if (strlen(key) != 1) + goto skippci; + + strncpy(func, key, sizeof(func)); + func[sizeof(func)-1] = '\0'; + + if (virStrToLong_i(domain, NULL, 16, &domainID) < 0) + goto skippci; + if (virStrToLong_i(bus, NULL, 16, &busID) < 0) + goto skippci; + if (virStrToLong_i(slot, NULL, 16, &slotID) < 0) + goto skippci; + if (virStrToLong_i(func, NULL, 16, &funcID) < 0) + goto skippci; + + if (VIR_ALLOC(hostdev) < 0) + goto cleanup; + + hostdev->managed = 0; + hostdev->source.subsys.type = VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI; + hostdev->source.subsys.u.pci.domain = domainID; + hostdev->source.subsys.u.pci.bus = busID; + hostdev->source.subsys.u.pci.slot = slotID; + hostdev->source.subsys.u.pci.function = funcID; + + if (VIR_REALLOC_N(def->hostdevs, def->nhostdevs+1) < 0) + goto no_memory; + def->hostdevs[def->nhostdevs++] = hostdev; + hostdev = NULL; + + skippci: + list = list->next; + } + } + if (hvm) { if (xenXMConfigGetString(conn, conf, "usbdevice", &str, NULL) < 0) goto cleanup; @@ -1939,6 +2022,76 @@ cleanup: +static int +xenXMDomainConfigFormatPCI(virConnectPtr conn, + virConfPtr conf, + virDomainDefPtr def) +{ + + virConfValuePtr pciVal = NULL; + int hasPCI = 0; + int i; + + for (i = 0 ; i < def->nhostdevs ; i++) + if (def->hostdevs[i]->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + def->hostdevs[i]->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) + hasPCI = 1; + + if (!hasPCI) + return 0; + + if (VIR_ALLOC(pciVal) < 0) + return -1; + + pciVal->type = VIR_CONF_LIST; + pciVal->list = NULL; + + for (i = 0 ; i < def->nhostdevs ; i++) { + if (def->hostdevs[i]->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + def->hostdevs[i]->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) { + virConfValuePtr val, tmp; + char *buf; + + if (virAsprintf(&buf, "%04x:%02x:%02x.%x", + def->hostdevs[i]->source.subsys.u.pci.domain, + def->hostdevs[i]->source.subsys.u.pci.bus, + def->hostdevs[i]->source.subsys.u.pci.slot, + def->hostdevs[i]->source.subsys.u.pci.function) < 0) + goto error; + + if (VIR_ALLOC(val) < 0) { + VIR_FREE(buf); + virReportOOMError(conn); + goto error; + } + val->type = VIR_CONF_STRING; + val->str = buf; + tmp = pciVal->list; + while (tmp && tmp->next) + tmp = tmp->next; + if (tmp) + tmp->next = val; + else + pciVal->list = val; + } + } + + if (pciVal->list != NULL) { + int ret = virConfSetValue(conf, "pci", pciVal); + pciVal = NULL; + if (ret < 0) + return -1; + } + VIR_FREE(pciVal); + + return 0; + +error: + virConfFreeValue(pciVal); + return -1; +} + + virConfPtr xenXMDomainConfigFormat(virConnectPtr conn, virDomainDefPtr def) { virConfPtr conf = NULL; @@ -2260,6 +2413,9 @@ virConfPtr xenXMDomainConfigFormat(virCo } VIR_FREE(netVal); + if (xenXMDomainConfigFormatPCI(conn, conf, def) < 0) + goto cleanup; + if (hvm) { if (def->nparallels) { virBuffer buf = VIR_BUFFER_INITIALIZER; Index: tests/sexpr2xmltest.c =================================================================== RCS file: /data/cvs/libvirt/tests/sexpr2xmltest.c,v retrieving revision 1.36 diff -u -p -r1.36 sexpr2xmltest.c --- tests/sexpr2xmltest.c 29 Jan 2009 17:02:00 -0000 1.36 +++ tests/sexpr2xmltest.c 30 Mar 2009 13:09:45 -0000 @@ -142,6 +142,7 @@ mymain(int argc, char **argv) DO_TEST("net-e1000", "net-e1000", 2); DO_TEST("bridge-ipaddr", "bridge-ipaddr", 3); DO_TEST("no-source-cdrom", "no-source-cdrom", 2); + DO_TEST("pci-devs", "pci-devs", 2); DO_TEST("fv-utc", "fv-utc", 1); DO_TEST("fv-localtime", "fv-localtime", 1); Index: tests/xmconfigtest.c =================================================================== RCS file: /data/cvs/libvirt/tests/xmconfigtest.c,v retrieving revision 1.26 diff -u -p -r1.26 xmconfigtest.c --- tests/xmconfigtest.c 8 Jan 2009 19:52:15 -0000 1.26 +++ tests/xmconfigtest.c 30 Mar 2009 13:09:45 -0000 @@ -231,6 +231,7 @@ mymain(int argc, char **argv) DO_TEST("escape-paths", 2); DO_TEST("no-source-cdrom", 2); + DO_TEST("pci-devs", 2); virCapabilitiesFree(caps); Index: tests/xml2sexprtest.c =================================================================== RCS file: /data/cvs/libvirt/tests/xml2sexprtest.c,v retrieving revision 1.34 diff -u -p -r1.34 xml2sexprtest.c --- tests/xml2sexprtest.c 23 Jan 2009 01:48:47 -0000 1.34 +++ tests/xml2sexprtest.c 30 Mar 2009 13:09:45 -0000 @@ -128,6 +128,7 @@ mymain(int argc, char **argv) DO_TEST("net-e1000", "net-e1000", "pvtest", 2); DO_TEST("bridge-ipaddr", "bridge-ipaddr", "pvtest", 2); DO_TEST("no-source-cdrom", "no-source-cdrom", "test", 2); + DO_TEST("pci-devs", "pci-devs", "pvtest", 2); DO_TEST("fv-utc", "fv-utc", "fvtest", 1); DO_TEST("fv-localtime", "fv-localtime", "fvtest", 1); Index: tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr =================================================================== RCS file: tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr diff -N tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr 30 Mar 2009 13:09:45 -0000 @@ -0,0 +1,2 @@ +(domain (domid 6)(name 'pvtest')(memory 420)(maxmem 420)(vcpus 2)(uuid '596a5d2171f48fb2e068e2386a5c413e')(on_poweroff 'destroy')(on_reboot 'destroy')(on_crash 'destroy')(image (linux (kernel '/var/lib/xen/vmlinuz.2Dn2YT')(ramdisk '/var/lib/xen/initrd.img.0u-Vhq')(args ' method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… ')))(device (pci (backend 0)(dev (domain 0x0001) (bus 0x0c) (slot 0x1b) (func 0x2))(dev (domain 0x0000) (bus 0x01) (slot 0x13) (func 0x0))))(device (vbd (dev 'xvda')(uname 'phy:/dev/MainVG/GuestVG')(mode 'w')))) + Index: tests/sexpr2xmldata/sexpr2xml-pci-devs.xml =================================================================== RCS file: tests/sexpr2xmldata/sexpr2xml-pci-devs.xml diff -N tests/sexpr2xmldata/sexpr2xml-pci-devs.xml --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/sexpr2xmldata/sexpr2xml-pci-devs.xml 30 Mar 2009 13:09:45 -0000 @@ -0,0 +1,37 @@ +<domain type='xen' id='6'> + <name>pvtest</name> + <uuid>596a5d21-71f4-8fb2-e068-e2386a5c413e</uuid> + <memory>430080</memory> + <currentMemory>430080</currentMemory> + <vcpu>2</vcpu> + <os> + <type>linux</type> + <kernel>/var/lib/xen/vmlinuz.2Dn2YT</kernel> + <initrd>/var/lib/xen/initrd.img.0u-Vhq</initrd> + <cmdline> method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… </cmdline> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>destroy</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <disk type='block' device='disk'> + <driver name='phy'/> + <source dev='/dev/MainVG/GuestVG'/> + <target dev='xvda' bus='xen'/> + </disk> + <console type='pty'> + <target port='0'/> + </console> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0001' bus='0x0c' slot='0x1b' function='0x2'/> + </source> + </hostdev> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0000' bus='0x01' slot='0x13' function='0x0'/> + </source> + </hostdev> + </devices> +</domain> Index: tests/xmconfigdata/test-pci-devs.cfg =================================================================== RCS file: tests/xmconfigdata/test-pci-devs.cfg diff -N tests/xmconfigdata/test-pci-devs.cfg --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/xmconfigdata/test-pci-devs.cfg 30 Mar 2009 13:09:45 -0000 @@ -0,0 +1,24 @@ +name = "test" +uuid = "cc2315e7-d26a-307a-438c-6d188ec4c09c" +maxmem = 382 +memory = 350 +vcpus = 1 +builder = "hvm" +kernel = "/usr/lib/xen/boot/hvmloader" +boot = "c" +pae = 1 +acpi = 1 +apic = 1 +localtime = 0 +on_poweroff = "destroy" +on_reboot = "destroy" +on_crash = "destroy" +device_model = "/usr/lib/xen/bin/qemu-dm" +sdl = 0 +vnc = 1 +vncunused = 1 +disk = [ "phy:/dev/sda8,hda,w", ",hdc:cdrom,r" ] +vif = [ "mac=00:16:3e:0a:7b:39,bridge=xenbr0,type=ioemu" ] +pci = [ "0001:0c:1b.2", "0000:01:13.0" ] +parallel = "none" +serial = "pty" Index: tests/xmconfigdata/test-pci-devs.xml =================================================================== RCS file: tests/xmconfigdata/test-pci-devs.xml diff -N tests/xmconfigdata/test-pci-devs.xml --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/xmconfigdata/test-pci-devs.xml 30 Mar 2009 13:09:45 -0000 @@ -0,0 +1,56 @@ +<domain type='xen'> + <name>test</name> + <uuid>cc2315e7-d26a-307a-438c-6d188ec4c09c</uuid> + <memory>391168</memory> + <currentMemory>358400</currentMemory> + <vcpu>1</vcpu> + <os> + <type arch='i686' machine='xenfv'>hvm</type> + <loader>/usr/lib/xen/boot/hvmloader</loader> + <boot dev='hd'/> + </os> + <features> + <acpi/> + <apic/> + <pae/> + </features> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>destroy</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/lib/xen/bin/qemu-dm</emulator> + <disk type='block' device='disk'> + <driver name='phy'/> + <source dev='/dev/sda8'/> + <target dev='hda' bus='ide'/> + </disk> + <disk type='block' device='cdrom'> + <driver name='phy'/> + <target dev='hdc' bus='ide'/> + <readonly/> + </disk> + <interface type='bridge'> + <mac address='00:16:3e:0a:7b:39'/> + <source bridge='xenbr0'/> + </interface> + <serial type='pty'> + <target port='0'/> + </serial> + <console type='pty'> + <target port='0'/> + </console> + <input type='mouse' bus='ps2'/> + <graphics type='vnc' port='-1' autoport='yes'/> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0001' bus='0x0c' slot='0x1b' function='0x2'/> + </source> + </hostdev> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0000' bus='0x01' slot='0x13' function='0x0'/> + </source> + </hostdev> + </devices> +</domain> Index: tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr =================================================================== RCS file: tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr diff -N tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr 30 Mar 2009 13:09:45 -0000 @@ -0,0 +1 @@ +(vm (name 'pvtest')(memory 420)(maxmem 420)(vcpus 2)(uuid '596a5d21-71f4-8fb2-e068-e2386a5c413e')(on_poweroff 'destroy')(on_reboot 'destroy')(on_crash 'destroy')(image (linux (kernel '/var/lib/xen/vmlinuz.2Dn2YT')(ramdisk '/var/lib/xen/initrd.img.0u-Vhq')(args ' method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… ')))(device (vbd (dev 'xvda')(uname 'phy:/dev/MainVG/GuestLV')(mode 'w')))(device (pci (dev (domain 0x0001)(bus 0x0c)(slot 0x1b)(func 0x2))(dev (domain 0x0000)(bus 0x01)(slot 0x13)(func 0x0))))) \ No newline at end of file Index: tests/xml2sexprdata/xml2sexpr-pci-devs.xml =================================================================== RCS file: tests/xml2sexprdata/xml2sexpr-pci-devs.xml diff -N tests/xml2sexprdata/xml2sexpr-pci-devs.xml --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/xml2sexprdata/xml2sexpr-pci-devs.xml 30 Mar 2009 13:09:45 -0000 @@ -0,0 +1,33 @@ +<domain type='xen' id='15'> + <name>pvtest</name> + <uuid>596a5d2171f48fb2e068e2386a5c413e</uuid> + <os> + <type>linux</type> + <kernel>/var/lib/xen/vmlinuz.2Dn2YT</kernel> + <initrd>/var/lib/xen/initrd.img.0u-Vhq</initrd> + <cmdline> method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… </cmdline> + </os> + <memory>430080</memory> + <vcpu>2</vcpu> + <on_poweroff>destroy</on_poweroff> + <on_reboot>destroy</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <disk type='block' device='disk'> + <source dev='/dev/MainVG/GuestLV'/> + <target dev='xvda'/> + </disk> + <console tty='/dev/pts/4'/> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0001' bus='0x0c' slot='0x1b' function='0x2'/> + </source> + </hostdev> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0000' bus='0x01' slot='0x13' function='0x0'/> + </source> + </hostdev> + </devices> +</domain> + -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 3

[libvirt] Reworked the XML verification patch for svirt
by Daniel J Walsh 03 Apr '09

03 Apr '09

Main goal of this patch it to verify data being written to xml and inform the user when he makes a mistake. Only able to verify static/dynamic in domain_conf. Added verify code to qemu_driver.c for model and potentially label.

3 5

[libvirt] ANNOUNCE: netcf, a platform-agnostic library for network configuration
by David Lutterkort 03 Apr '09

03 Apr '09

I am pleased to announce the availability of netcf 0.0.1, the initial release of a library for managing network configuration in a platform agnostic manner. If I were into code names, this would be the "what have you been waiting for" release. Netcf does its work by directly modifying the 'native' configuration files of the host it is running on; this avoids a whole class of problems caused by similar approaches that do network configuration behind the back of the native mechanisms. The API allows listing of configured interfaces, defining the configuration of an interface, retrieving the same (regardless of whether the interface was initially configured with netcf or not), and bringing interfaces up and down. This functionality is needed both by libvirt and NetworkManager, so it seemed only logical to move their common needs into a separate library. This release is mostly geared at soliciting feedback and sparking spirited reviews. In particular, the API is not stable yet (it will be with the release of netcf 0.1.0) Besides general comments, criticism and the customary praise, I am particularly interested in reviews on the following: * The API (barring any strong objections, I will declare it stable) * The XML schema for describing network interfaces (in data/xml/interface.rng) * General code review Where can I get it ? -------------------- Main website https://fedorahosted.org/netcf/ A tarball can be downloaded from https://fedorahosted.org/released/netcf/ The git repo is at http://git.fedorahosted.org/git/netcf.git git clone git://git.fedorahosted.org/netcf.git For those running Fedora 10, I built some preliminary RPM's and made them available at http://people.redhat.com/dlutter/netcf/download/ (and if you are a Fedora reviewer, the review BZ is 493750 ;) How can I help ? ---------------- Join the mailing list at https://fedorahosted.org/mailman/listinfo/netcf-devel Currently, netcf can set up simple Ethernet interfaces, bridges with enslaved physical devices and bonds on Fedora. This points to two areas where improvements are most needed: * support other Linux distros and operating systems * expand what kind of interfaces netcf can handle (most sorely needed are VLAN's and adding bonds to a bridge) If you can help with either of these two, or anything, really, drop a note on the mailing list so wecan discuss. David

1 0

[libvirt] specifying -serial telnet: in XML
by Sukadev Bhattiprolu 02 Apr '09

02 Apr '09

I run qemu (version 0.9.1 kvm-84) with -serial option as follows: - serial telnet::2445,server so I can telnet to 2445 and have a serial console. I tried specifying this to libvirt (version 0.6.0) using the following stanza: <serial type='tcp'> <source mode="bind" host="127.0.0.1" service="2445"/> <protocol type="telnet"/> <target port="1"/> </serial> I checked /var/log/libvirt/qemu/<guest>.log and found that libvirt added 'nowait' to the serial option: -serial telnet:127.0.0.1:2445,server,nowait which is fine (I think). But when I start my guest, and "telnet 127.0.0.1 2445" I don't get any output even though the guest boots up fine. After the guest boots $ echo foo > /dev/ttyS0 in the guest shows the 'foo' in the telnet window. But I don't get the boot/shutdown messages. I also have the vnc option in the xml, <graphics type='vnc' port='5902' autoport='no' listen='127.0.0.1'/> so if I open vncviewer, I can see the boot messages. The problem with this though is if the kernel fails to boot for some reason, the vncviewer exits and I can't find the reason the boot failed. My kernel boot command line has 'console=ttyS0' option. Is there anything else I need to ensure 'telnet 127.0.0.1 port' displays the messages ? Thanks, Sukadev

2 2

[libvirt] [PATCH] Openvz: how to use a nondefault config
by Anton Protopopov 02 Apr '09

02 Apr '09

Hi all. When libvirt creates an OpenVZ machine, it does smth like # vzctl create $ctid --ostemplate `ostemplate` And we can't specify an --config `config` argument. In this case vzctl chooses a default config file (a value of CONFIG variable from /etc/vz/vz.conf). But there are cases when we want to specify a config file directly, as in # vzctl create $ctid --ostemplate `ostemplate` --config `config` I think, that we can try to use the same `ostemplate` name as --config value, if corresponding template exists, i.e., if there is a file /etc/vz/conf/ve-`ostemplate`.conf-sample, libvirt will run a command # vzctl create $ctid --ostemplate `ostemplate` --config `ostemplate` Otherwise, it will run it's default command # vzctl create $ctid --ostemplate `ostemplate` (the same arguments.) ----------- Example of creating a machine with nondefault config: 1. Create /etc/vz/conf/ve-altlinux-test.conf-sample 2. Create /var/lib/vz/template/cache/altlinux-test.tar 3. Use libvirt with (hypothetical) XML: <domain type="openvz"> <name>100</name> <memory>1</memory> <os><type>exe</type></os> <devices> <filesystem type="template"> <source name="altlinux-test"/> <target dir="/"/> </filesystem> </devices> </domain> ----------- Example of creating a machine with default config: 1. Ensure there is no /etc/vz/conf/ve-altlinux-test.conf-sample 2. Create /var/lib/vz/template/cache/altlinux-test.tar 3. Use libvirt with (the same as above) XML file -----------

4 7

[libvirt] SYSCONF_DIR
by Zvi Dubitzky 02 Apr '09

02 Apr '09

I can do TLS certificate authentication on a remote libvirt access through 'virsh' ( virsh -c qemu+tls://hostnake/system) . But when I try same auth from my C code linked with libvirt 0.6.1 I get the message : " libvir: Remote error: cannot access CA certificate '/usr/local/etc/pki/CA/cacert.pem . No such file or directory" My certificate is at /etc/pki/CA as instructed in the web site (deafult) and so is my libvirtd.conf saying. virsh works ok with it . In my case the server and client are on same machine . I guess the need for SYSCONF_DIR = /usr/local/etc comes from the configure of the libvirt . In fact I see in the configure file the statement : sysconfigdir = '${prefix}/etc . Any idea how to fix that . I am using the installed libvirt . did not rebuild it . thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

[libvirt] checkpoint patches of netcf integration for review/guidance
by laine＠laine.org 02 Apr '09

02 Apr '09

I've gone through the first two steps of danpb's "Guide to Adding New libvirt APIs", and thought this would be a good time to send a copy of my work to the list to make sure I'm going in the right direction. Now is (yet another) good time to tell me "You're doing it wrong!". Basically, I've defined the public api in libvirt.h (actually libvirt.h.in - it's a real pain to have generated files under source control - it almost led me to mistakenly lose my modifications a couple times after running autogen, until I actually looked in the directory containing the file and saw the *.in file), done all the toplevel functions in libvirt.c, and whatever ancillary definitions and small functions were needed by them. In a 3rd patch, I also added stuff to configure.in that detects the presence of libnetcf and 1) turns on WITH_NETCF in config.h and 2) adds -lnetcf to LDFLAGS. I noticed that some other similar items have been added to configure.in by peppering little changes in several spots throughout the file; I preferred adding all my stuff in one place to make it easier to follow (and less likely to cause merge conflicts). So far this has been a *lot* of cut-n-paste, which makes me a bit uneasy (both because of the lesser problem that it almost always leads to errors caused by missing bits and pieces that need modification after the paste, and the greater problem that using this method may indicate I'm operating mechanically, without having adequate understanding of what I'm trying to do).

2 6

[libvirt] PATCH: update the XML domain RelaxNG schemas
by Daniel Veillard 02 Apr '09

02 Apr '09

pritesh raised the issue on IRC, so I checked the schemas against the current parsing routines and found a few missing items, I also sorted the device lice to match the enum sequence in the C file. One thing missing and still TODO is the interface definition block, the C parser accepts the "server", "client" and "mcast" types but the RNG doesn't list them in the choice for "interface" Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 2

[libvirt] PATCH: first patch to update the architecture documentation
by Daniel Veillard 02 Apr '09

02 Apr '09

Relatively small update done in the train yesterday, currently this just updates the project description, makes a spearate page for goals and terminology. I intent to continue to revamp the architecture pages with an user view of the architecture, presenting the library/daemon dichotomy, and then a separate page more on the driver internals. Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 2

[libvirt] PATCH: Fix 2 build bugs on mingw
by Daniel P. Berrange 02 Apr '09

02 Apr '09

The Mingw platform lacks readlink(), so we stub out this method on that platform. The plain write() implementation of safezero() was not actually doing anything with the 'offset' arg. This makes it seek to the offset location before writing data. The mmap implementation of safezero() was also not quite right, truncating to len, so if offset was non-zero, it would not be large enough. Daniel Index: configure.in =================================================================== RCS file: /data/cvs/libvirt/configure.in,v retrieving revision 1.211 diff -u -p -r1.211 configure.in --- configure.in 1 Apr 2009 10:40:38 -0000 1.211 +++ configure.in 1 Apr 2009 15:53:35 -0000 @@ -72,7 +72,7 @@ dnl Use --disable-largefile if you don't AC_SYS_LARGEFILE dnl Availability of various common functions (non-fatal if missing). -AC_CHECK_FUNCS([cfmakeraw regexec uname sched_getaffinity getuid getgid posix_fallocate mmap]) +AC_CHECK_FUNCS([cfmakeraw regexec uname sched_getaffinity getuid getgid posix_fallocate mmap readlink]) dnl Availability of various not common threadsafe functions AC_CHECK_FUNCS([strerror_r strtok_r getmntent_r getgrnam_r getpwuid_r]) Index: src/util.c =================================================================== RCS file: /data/cvs/libvirt/src/util.c,v retrieving revision 1.98 diff -u -p -r1.98 util.c --- src/util.c 1 Apr 2009 10:26:22 -0000 1.98 +++ src/util.c 1 Apr 2009 15:53:35 -0000 @@ -136,7 +136,7 @@ int safezero(int fd, int flags ATTRIBUTE /* memset wants the mmap'ed file to be present on disk so create a * sparse file */ - r = ftruncate(fd, len); + r = ftruncate(fd, offset + len); if (r < 0) return -errno; @@ -158,6 +158,9 @@ int safezero(int fd, int flags ATTRIBUTE char *buf; unsigned long long remain, bytes; + if (lseek(fd, offset, SEEK_SET) < 0) + return errno; + /* Split up the write in small chunks so as not to allocate lots of RAM */ remain = len; bytes = 1024 * 1024; @@ -949,6 +952,7 @@ int virFileLinkPointsTo(const char *chec int virFileResolveLink(const char *linkpath, char **resultpath) { +#ifdef HAVE_READLINK struct stat st; char *buf; int n; @@ -981,6 +985,11 @@ int virFileResolveLink(const char *linkp *resultpath = buf; return 0; +#else + if (!(*resultpath = strdup(linkpath))) + return -ENOMEM; + return 0; +#endif } -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] existing bridge and no iptable rules
by Markus 01 Apr '09

01 Apr '09

hi, i try to setup libvirt with an existing network setup (bridge with NO physical device, routes and iptables are already configured, dhcp named ..) . this bridge should be available for the libvirt domains. now the problem: every configuration i tried ends in iptable rules beeing created which mess with my network setup (e.g. i try to route vpn traffic from a guest to another network outside) is there a way of defining a virtual network / network source / bridge for domains that takes an existing bridge and does no setup ( ip, iptables, .. )? thanks, markus

1 0

[libvirt] error in libvirt.py from libvirt-0.6.1 ?
by Gerrit Slomma 01 Apr '09

01 Apr '09

Hello Is there a paticular reason for having "def migrate" two times in libvirt.py? The first occurence is on line 345 with parameters (self, dconn, flags, dname, uri, bandwidth) The second occurence is on line 1102 with parameters (self, domain, flags, dname, uri, bandwidth) Which one is correct? And i am no python programmer in the least, but isn't the ret set false? ret = libvirtmod.virDomainMigrate(domain__o, self._o, flags, dname, uri, bandwidth) That doesn't correspond to the def some above this earlier. diff /tmp/migrate_1st_occurance /tmp/migrate_2nd_occurance 1c1 < def migrate(self, dconn, flags, dname, uri, bandwidth): --- > def migrate(self, domain, flags, dname, uri, bandwidth): 31,34c31,34 < if dconn is None: dconn__o = None < else: dconn__o = dconn._o < ret = libvirtmod.virDomainMigrate(self._o, dconn__o, flags, dname, uri, bandwidth) < if ret is None:raise libvirtError('virDomainMigrate() failed', dom=self) --- > if domain is None: domain__o = None > else: domain__o = domain._o > ret = libvirtmod.virDomainMigrate(domain__o, self._o, flags, dname, uri, bandwidth) > if ret is None:raise libvirtError('virDomainMigrate() failed', conn=self)

2 1

[libvirt] [PATCH 0/1] SCSI host pools
by David Allan 01 Apr '09

01 Apr '09

Hi All, Here is reworked SCSI host pool code. I have tested it with FC, iSCSI and SATA, and it works correctly in all those cases. I have not tried parallel SCSI, but I don't see any reason why it wouldn't work there also, as it should be transport independent. This patch is so different from the earlier revision that a diff from the earlier version would be pointless, and I have only included a patch against the current head. Dave

4 19

[libvirt] PATCH: Allow running of console after creating guest
by Daniel P. Berrange 01 Apr '09

01 Apr '09

This patch adds a new flag to the 'create' and 'start' commands in virsh, making it immeditely spawn the builtin virsh console, eg # virsh start --console demo At the same time, I also add in a check for virConnectGetHostname() against the local hostname, and refuse to launch the console if they differ. With remote hypervisors it is obviously not possible to access the /dev/pts files directly. Without this check, the console may work by pure luck if there happens to be a matching device name on the local host. Rather confusing, and even potentially dangerous. # virsh -c qemu+ssh://root@demo.example.com/system console someguest error: Cannot connect to a remote console device Finally, when starting the console print a message telling the user the hotkey to quit # virsh start --console VirtTest Domain VirtTest started Connecting to console, type Ctrl+] to quit Daniel diff -r a203154020dd src/virsh.c --- a/src/virsh.c Wed Apr 01 11:57:25 2009 +0100 +++ b/src/virsh.c Wed Apr 01 13:36:40 2009 +0100 @@ -500,20 +500,30 @@ static const vshCmdOptDef opts_console[] #ifndef __MINGW32__ static int -cmdConsole(vshControl *ctl, const vshCmd *cmd) +cmdRunConsole(vshControl *ctl, virDomainPtr dom) { xmlDocPtr xml = NULL; xmlXPathObjectPtr obj = NULL; xmlXPathContextPtr ctxt = NULL; - virDomainPtr dom; int ret = FALSE; char *doc; - - if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) - return FALSE; - - if (!(dom = vshCommandOptDomain(ctl, cmd, NULL))) - return FALSE; + char *thatHost = NULL; + char *thisHost = NULL; + + if (!(thisHost = virGetHostname())) { + vshError(ctl, FALSE, "%s", _("Failed to get local hostname")); + goto cleanup; + } + + if (!(thatHost = virConnectGetHostname(ctl->conn))) { + vshError(ctl, FALSE, "%s", _("Failed to get connection hostname")); + goto cleanup; + } + + if (STRNEQ(thisHost, thatHost)) { + vshError(ctl, FALSE, "%s", _("Cannot connect to a remote console device")); + goto cleanup; + } doc = virDomainGetXMLDesc(dom, 0); if (!doc) @@ -532,6 +542,7 @@ cmdConsole(vshControl *ctl, const vshCmd obj = xmlXPathEval(BAD_CAST "string(/domain/devices/console/@tty)", ctxt); if ((obj != NULL) && ((obj->type == XPATH_STRING) && (obj->stringval != NULL) && (obj->stringval[0] != 0))) { + vshPrintExtra(ctl, "%s", _("Connecting to console, type Ctrl+] to quit\n")); if (vshRunConsole((const char *)obj->stringval) == 0) ret = TRUE; } else { @@ -543,14 +554,16 @@ cmdConsole(vshControl *ctl, const vshCmd xmlXPathFreeContext(ctxt); if (xml) xmlFreeDoc(xml); - virDomainFree(dom); + free(thisHost); + free(thatHost); + return ret; } #else /* __MINGW32__ */ static int -cmdConsole(vshControl *ctl, const vshCmd *cmd ATTRIBUTE_UNUSED) +cmdRunConsole(vshControl *ctl, virDomainPtr dom ATTRIBUTE_UNUSED) { vshError (ctl, FALSE, "%s", _("console not implemented on this platform")); return FALSE; @@ -558,6 +571,24 @@ cmdConsole(vshControl *ctl, const vshCmd #endif /* __MINGW32__ */ +static int +cmdConsole(vshControl *ctl, const vshCmd *cmd) +{ + virDomainPtr dom; + int ret; + + if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) + return FALSE; + + if (!(dom = vshCommandOptDomain(ctl, cmd, NULL))) + return FALSE; + + ret = cmdRunConsole(ctl, dom); + + virDomainFree(dom); + return ret; +} + /* * "list" command */ @@ -882,6 +913,7 @@ static const vshCmdInfo info_create[] = static const vshCmdOptDef opts_create[] = { {"file", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("file containing an XML domain description")}, + {"console", VSH_OT_BOOL, 0, gettext_noop("attach to console after creation")}, {NULL, 0, 0, NULL} }; @@ -893,6 +925,7 @@ cmdCreate(vshControl *ctl, const vshCmd int found; int ret = TRUE; char *buffer; + int console = vshCommandOptBool(cmd, "console"); if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; @@ -910,6 +943,8 @@ cmdCreate(vshControl *ctl, const vshCmd if (dom != NULL) { vshPrint(ctl, _("Domain %s created from %s\n"), virDomainGetName(dom), from); + if (console) + cmdRunConsole(ctl, dom); virDomainFree(dom); } else { vshError(ctl, FALSE, _("Failed to create domain from %s"), from); @@ -1030,6 +1065,7 @@ static const vshCmdInfo info_start[] = { static const vshCmdOptDef opts_start[] = { {"domain", VSH_OT_DATA, VSH_OFLAG_REQ, gettext_noop("name of the inactive domain")}, + {"console", VSH_OT_BOOL, 0, gettext_noop("attach to console after creation")}, {NULL, 0, 0, NULL} }; @@ -1038,6 +1074,7 @@ cmdStart(vshControl *ctl, const vshCmd * { virDomainPtr dom; int ret = TRUE; + int console = vshCommandOptBool(cmd, "console"); if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; @@ -1054,6 +1091,8 @@ cmdStart(vshControl *ctl, const vshCmd * if (virDomainCreate(dom) == 0) { vshPrint(ctl, _("Domain %s started\n"), virDomainGetName(dom)); + if (console) + cmdRunConsole(ctl, dom); } else { vshError(ctl, FALSE, _("Failed to start domain %s"), virDomainGetName(dom)); -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 4

[libvirt] Added seclabeltest to run under test suite.
by Daniel J Walsh 01 Apr '09

01 Apr '09

1 0

[libvirt] PATCH: Fix misc memory alloc/free/usage bugs
by Daniel P. Berrange 01 Apr '09

01 Apr '09

A few more small bugs * src/node_device.c: Don't strdup() a NULL device parent * src/qemu_conf.c: Don't try to run access() on a NULL binary path * src/security_selinux.c Make sure result from readlink() is NULL terminated * src/storage_conf.c: DOn't free 'mode' while it is still used diff -r b73fe666feff src/node_device.c --- a/src/node_device.c Fri Mar 27 16:14:49 2009 +0000 +++ b/src/node_device.c Mon Mar 30 14:37:45 2009 +0100 @@ -176,9 +176,14 @@ static char *nodeDeviceGetParent(virNode goto cleanup; } - ret = strdup(obj->def->parent); - if (!ret) - virReportOOMError(dev->conn); + if (obj->def->parent) { + ret = strdup(obj->def->parent); + if (!ret) + virReportOOMError(dev->conn); + } else { + virNodeDeviceReportError(dev->conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("no parent for this device")); + } cleanup: if (obj) diff -r b73fe666feff src/qemu_conf.c --- a/src/qemu_conf.c Fri Mar 27 16:14:49 2009 +0000 +++ b/src/qemu_conf.c Mon Mar 30 14:37:45 2009 +0100 @@ -269,7 +269,7 @@ qemudCapsInitGuest(virCapsPtr caps, * which can be used with magic cpu choice */ hasbase = (access(info->binary, X_OK) == 0); - hasaltbase = (access(info->altbinary, X_OK) == 0); + hasaltbase = (info->altbinary && access(info->altbinary, X_OK) == 0); /* Can use acceleration for KVM/KQEMU if * - host & guest arches match diff -r b73fe666feff src/security_selinux.c --- a/src/security_selinux.c Fri Mar 27 16:14:49 2009 +0000 +++ b/src/security_selinux.c Mon Mar 30 14:37:45 2009 +0100 @@ -303,11 +303,13 @@ SELinuxRestoreSecurityImageLabel(virConn return -1; if (S_ISLNK(buf.st_mode)) { + int n; if (VIR_ALLOC_N(newpath, buf.st_size + 1) < 0) return -1; - if (readlink(path, newpath, buf.st_size) < 0) + if ((n =readlink(path, newpath, buf.st_size)) < 0) goto err; + buf.st_size[n] = '\0'; path = newpath; if (stat(path, &buf) != 0) goto err; diff -r b73fe666feff src/storage_conf.c --- a/src/storage_conf.c Fri Mar 27 16:14:49 2009 +0000 +++ b/src/storage_conf.c Mon Mar 30 14:37:45 2009 +0100 @@ -401,12 +401,13 @@ virStorageDefParsePerms(virConnectPtr co } else { char *end = NULL; perms->mode = strtol(mode, &end, 8); - VIR_FREE(mode); if (*end || perms->mode < 0 || perms->mode > 0777) { + VIR_FREE(mode); virStorageReportError(conn, VIR_ERR_XML_ERROR, "%s", _("malformed octal mode")); goto error; } + VIR_FREE(mode); } if (virXPathNode(conn, "./owner", ctxt) == NULL) { Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 7

[libvirt] PATCH: Add --tree flag to virsh nodedev-list
by Daniel P. Berrange 01 Apr '09

01 Apr '09

The virsh noddev-list command is used to display node devices. It just prints out their names in a flat list. When detaching devices for purposes of PCI passthrough though, it is important to understand what devices are children of the PCI device about to be detached. It is tedious to find this out, the user having to run virsh nodedev-dumpxml and look at the parent field. # virsh -c qemu:///system nodedev-list computer DVD_GCC_4247N net_00_13_02_b9_f9_d3 net_00_13_02_b9_f9_d3_0 net_00_15_58_2f_e9_55 pci_1002_71c4 pci_104c_ac56 pci_8086_27c5_scsi_host pci_8086_27c5_scsi_host_0 pci_8086_27c5_scsi_host_1 pci_8086_27c5_scsi_host_2 pci_8086_27c5_scsi_host_scsi_device_lun0 pci_8086_27c5_scsi_host_scsi_host pci_8086_27c8 pci_8086_27da pci_8086_27df pci_8086_27df_scsi_host pci_8086_27df_scsi_host_0 pci_8086_27df_scsi_host_scsi_device_lun0 pci_8086_27df_scsi_host_scsi_host pci_8086_4227 storage_serial_SATA_HTS721010G9SA00_MPCZ12Y0GNGWSE usb_device_a5c_2110_noserial_if3 This patch adds a new '--tree' flag to the nodedev-list command, allowing a prettier format to be used # virsh -c qemu:///system nodedev-list --tree computer | +-pci_8086_2448 | | | +-pci_104c_ac56 | +-pci_8086_27a0 +-pci_8086_27a1 | | | +-pci_1002_71c4 | +-pci_8086_27b9 +-pci_8086_27c5 | | | +-pci_8086_27c5_scsi_host | | | | | +-pci_8086_27c5_scsi_host_scsi_device_lun0 | | | | | | | +-storage_serial_SATA_HTS721010G9SA00_MPCZ12Y0GNGWSE | | | | | +-pci_8086_27c5_scsi_host_scsi_host | | | +-pci_8086_27c5_scsi_host_0 | +-pci_8086_27c5_scsi_host_1 | +-pci_8086_27c5_scsi_host_2 | +-pci_8086_27c8 | | | +-usb_device_1d6b_1_0000_00_1d_0 | | | +-usb_device_1d6b_1_0000_00_1d_0_if0 | +-pci_8086_27c9 | | | +-usb_device_1d6b_1_0000_00_1d_1 | | | +-usb_device_1d6b_1_0000_00_1d_1_if0 | +-pci_8086_27ca | | | +-usb_device_1d6b_1_0000_00_1d_2 | | | +-usb_device_1d6b_1_0000_00_1d_2_if0 | +-pci_8086_27cb | | | +-usb_device_1d6b_1_0000_00_1d_3 | | | +-usb_device_1d6b_1_0000_00_1d_3_if0 | +-usb_device_483_2016_noserial | | | | | +-usb_device_483_2016_noserial_if0 | | | +-usb_device_a5c_2110_noserial | | | +-usb_device_a5c_2110_noserial_if0 | +-usb_device_a5c_2110_noserial_if1 | +-usb_device_a5c_2110_noserial_if2 | +-usb_device_a5c_2110_noserial_if3 | +-pci_8086_27cc | | | +-usb_device_1d6b_2_0000_00_1d_7 | | | +-usb_device_1d6b_2_0000_00_1d_7_if0 | +-pci_8086_27d0 | | | +-pci_8086_109a | | | +-net_00_15_58_2f_e9_55 | +-pci_8086_27d2 | | | +-pci_8086_4227 | | | +-net_00_13_02_b9_f9_d3 | +-net_00_13_02_b9_f9_d3_0 | +-pci_8086_27d4 +-pci_8086_27d6 +-pci_8086_27d8 +-pci_8086_27da +-pci_8086_27df | +-pci_8086_27df_scsi_host | | | +-pci_8086_27df_scsi_host_scsi_device_lun0 | | | | | +-DVD_GCC_4247N | | | +-pci_8086_27df_scsi_host_scsi_host | +-pci_8086_27df_scsi_host_0 Daniel diff -r b73fe666feff src/virsh.c --- a/src/virsh.c Fri Mar 27 16:14:49 2009 +0000 +++ b/src/virsh.c Mon Mar 30 14:37:45 2009 +0100 @@ -4391,16 +4391,77 @@ static const vshCmdInfo info_node_list_d }; static const vshCmdOptDef opts_node_list_devices[] = { + {"tree", VSH_OT_BOOL, 0, gettext_noop("list devices in a tree")}, {"cap", VSH_OT_STRING, VSH_OFLAG_NONE, gettext_noop("capability name")}, {NULL, 0, 0, NULL} }; +#define MAX_INDENT 100 + +static void +cmdNodeListDevicesPrint(vshControl *ctl, + char **devices, + char **parents, + int num_devices, + int devid, + int lastdev, + unsigned int depth, + char *indent) +{ + int i; + int nextlastdev = -1; + + if (depth) { + indent[depth] = '+'; + indent[depth+1] = '-'; + } + + vshPrint(ctl, indent); + if (depth) { + if (devid == lastdev) + indent[depth] = ' '; + else + indent[depth] = '|'; + indent[depth+1] = ' '; + depth+=2; + } + vshPrint(ctl, "%s\n", devices[devid]); + + for (i = 0 ; i < num_devices ; i++) { + if (parents[i] && + STREQ(parents[i], devices[devid])) { + //fprintf(stderr, "%s %s %d %d\n", parents[i], devices[devid], i, devid); + nextlastdev = i; + } + } + + if (nextlastdev != -1) { + vshPrint(ctl, indent); + vshPrint(ctl, " |\n"); + } + + indent[depth] = ' '; + for (i = 0 ; i < num_devices ; i++) { + indent[depth] = ' '; + indent[depth+1] = ' '; + if (parents[i] && + STREQ(parents[i], devices[devid])) + cmdNodeListDevicesPrint(ctl, devices, parents, num_devices, i, nextlastdev, depth + 2, indent); + indent[depth] = '\0'; + } + if (nextlastdev == -1 && devid == lastdev) { + vshPrint(ctl, indent); + vshPrint(ctl, "\n"); + } +} + static int cmdNodeListDevices (vshControl *ctl, const vshCmd *cmd ATTRIBUTE_UNUSED) { char *cap; char **devices; int found, num_devices, i; + int tree = vshCommandOptBool(cmd, "tree"); if (!vshConnectionUsability(ctl, ctl->conn, TRUE)) return FALSE; @@ -4426,9 +4487,39 @@ cmdNodeListDevices (vshControl *ctl, con return FALSE; } qsort(&devices[0], num_devices, sizeof(char*), namesorter); - for (i = 0; i < num_devices; i++) { - vshPrint(ctl, "%s\n", devices[i]); - free(devices[i]); + if (tree) { + char indent[MAX_INDENT]; + char **parents = vshMalloc(ctl, sizeof(char *) * num_devices); + for (i = 0; i < num_devices; i++) { + virNodeDevicePtr dev = virNodeDeviceLookupByName(ctl->conn, devices[i]); + if (dev && STRNEQ(devices[i], "computer")) { + const char *parent = virNodeDeviceGetParent(dev); + parents[i] = parent ? strdup(parent) : NULL; + } else { + parents[i] = NULL; + } + virNodeDeviceFree(dev); + } + for (i = 0 ; i < num_devices ; i++) { + memset(indent, '\0', sizeof indent); + if (parents[i] == NULL) + cmdNodeListDevicesPrint(ctl, + devices, + parents, + num_devices, + i, + i, + 0, + indent); + } + for (i = 0 ; i < num_devices ; i++) + free(parents[i]); + free(parents); + } else { + for (i = 0; i < num_devices; i++) { + vshPrint(ctl, "%s\n", devices[i]); + free(devices[i]); + } } free(devices); return TRUE; -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 4

[libvirt] question about qemudDomainMigratePrepare2 of qemu_driver.c from libvirt-0.6.1
by Gerrit.Slomma＠drv-bund.de 01 Apr '09

01 Apr '09

Hello I am just curious if the checking of the connection string in qemu_driver.c from libvirt-0.6.1 in qemudDomainMigratePrepare2 is faulty. The source is: [...] } else { /* Check the URI starts with "tcp:". We will escape the * URI when passing it to the qemu monitor, so bad * characters in hostname part don't matter. */ if (!STREQLEN (uri_in, "tcp:", 6)) { qemudReportError (dconn, NULL, NULL, VIR_ERR_INVALID_ARG, "%s", _("only tcp URIs are supported for KVM migrations")); goto cleanup; } [...] The second compare-string "tcp:" is only 4 characters long, therefore a check via strncmp must fail i presumed, i isolated this comparison: #include <stdio.h> #include <string.h> #define STREQLEN(a,b,n) (strncmp((a),(b),(n)) == 0) int main () { char str[][200] = { "tcp://192.168.1.14:4444" , "tcp:192.168.1.14:4444" , "ssh ://192.168.1.14:4444" , "ssh:192.168.1.14:4444" }; int n; for (n=0 ; n<4 ; n++) { printf("testing %s against \"tcp:\" with result: ", str[n]); if (!STREQLEN(str[n],"tcp:",6)) { printf ("ERROR\n"); } else { printf ("OK\n"); } } return 0; } and as i pressumed this gives me on the commandline: testing tcp://192.168.1.14:4444 against "tcp:" with result: ERROR testing tcp:192.168.1.14:4444 against "tcp:" with result: ERROR testing ssh://192.168.1.14:4444 against "tcp:" with result: ERROR testing ssh:192.168.1.14:4444 against "tcp:" with result: ERROR if i change the 6 to 4 it gives testing tcp://192.168.1.14:4444 against "tcp:" with result: OK testing tcp:192.168.1.14:4444 against "tcp:" with result: OK testing ssh://192.168.1.14:4444 against "tcp:" with result: ERROR testing ssh:192.168.1.14:4444 against "tcp:" with result: ERROR but now tcp:// also is marked as correct. Am i missing something here and the checking in qemu_driver.c is correct? I always get the "only tcp URIs are supported for KVM migrations" if i try to migrate KVM-domains in virt-manager 0.7.0 and therefore searched for the root-cause of this problem. Kind regards, Gerrit Slomma Mit freundlichen Grüßen, Gerrit Slomma Team eBetrieb Deutsche Rentenversicherung Bund Abt. 11 - Organisation und IT-Services Fachbereich 1170 - IT-Services und Betrieb Bereich 1174 - Anwendungsbetrieb Team 1174-46 - Trustcenter, eBetrieb

4 4

[libvirt] PATCH: Fix XM driver to extract vifname= parameter
by Daniel P. Berrange 01 Apr '09

01 Apr '09

The XM driver is not currently handling the vifname= parameter for network devices in /etc/xen config files. This patch fixes that ommision Daniel diff -r e396919fa59e src/xm_internal.c --- a/src/xm_internal.c Tue Mar 31 16:35:42 2009 +0100 +++ b/src/xm_internal.c Tue Mar 31 16:58:11 2009 +0100 @@ -991,6 +991,7 @@ xenXMDomainConfigParse(virConnectPtr con char ip[16]; char mac[18]; char bridge[50]; + char vifname[50]; char *key; bridge[0] = '\0'; @@ -998,6 +999,7 @@ xenXMDomainConfigParse(virConnectPtr con script[0] = '\0'; ip[0] = '\0'; model[0] = '\0'; + vifname[0] = '\0'; if ((list->type != VIR_CONF_STRING) || (list->str == NULL)) goto skipnic; @@ -1036,6 +1038,12 @@ xenXMDomainConfigParse(virConnectPtr con len = sizeof(model)-1; strncpy(model, data, len); model[len] = '\0'; + } else if (STRPREFIX(key, "vifname=")) { + int len = nextkey ? (nextkey - data) : sizeof(vifname)-1; + if (len > (sizeof(vifname)-1)) + len = sizeof(vifname)-1; + strncpy(vifname, data, len); + vifname[len] = '\0'; } else if (STRPREFIX(key, "ip=")) { int len = nextkey ? (nextkey - data) : 15; if (len > 15) @@ -1105,6 +1113,10 @@ xenXMDomainConfigParse(virConnectPtr con !(net->model = strdup(model))) goto no_memory; + if (vifname[0] && + !(net->ifname = strdup(vifname))) + goto no_memory; + if (VIR_REALLOC_N(def->nets, def->nnets+1) < 0) goto no_memory; def->nets[def->nnets++] = net; @@ -1994,6 +2006,10 @@ static int xenXMDomainConfigFormatNet(vi virBufferVSprintf(&buf, ",model=%s", net->model); + if (net->ifname) + virBufferVSprintf(&buf, ",vifname=%s", + net->ifname); + if (VIR_ALLOC(val) < 0) { virReportOOMError(conn); goto cleanup; diff -r e396919fa59e tests/xmconfigdata/test-paravirt-net-vifname.cfg --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/xmconfigdata/test-paravirt-net-vifname.cfg Tue Mar 31 16:58:11 2009 +0100 @@ -0,0 +1,12 @@ +name = "XenGuest1" +uuid = "c7a5fdb0-cdaf-9455-926a-d65c16db1809" +maxmem = 579 +memory = 394 +vcpus = 1 +bootloader = "/usr/bin/pygrub" +on_poweroff = "destroy" +on_reboot = "restart" +on_crash = "restart" +vfb = [ "type=vnc,vncunused=1,vnclisten=127.0.0.1,vncpasswd=123poi" ] +disk = [ "phy:/dev/HostVG/XenGuest1,xvda,w" ] +vif = [ "mac=00:16:3e:66:94:9c,bridge=br0,model=e1000,vifname=net0" ] diff -r e396919fa59e tests/xmconfigdata/test-paravirt-net-vifname.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/xmconfigdata/test-paravirt-net-vifname.xml Tue Mar 31 16:58:11 2009 +0100 @@ -0,0 +1,33 @@ +<domain type='xen'> + <name>XenGuest1</name> + <uuid>c7a5fdb0-cdaf-9455-926a-d65c16db1809</uuid> + <memory>592896</memory> + <currentMemory>403456</currentMemory> + <vcpu>1</vcpu> + <bootloader>/usr/bin/pygrub</bootloader> + <os> + <type arch='i686' machine='xenpv'>linux</type> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>restart</on_crash> + <devices> + <disk type='block' device='disk'> + <driver name='phy'/> + <source dev='/dev/HostVG/XenGuest1'/> + <target dev='xvda' bus='xen'/> + </disk> + <interface type='bridge'> + <mac address='00:16:3e:66:94:9c'/> + <source bridge='br0'/> + <target dev='net0'/> + <model type='e1000'/> + </interface> + <console type='pty'> + <target port='0'/> + </console> + <input type='mouse' bus='xen'/> + <graphics type='vnc' port='-1' autoport='yes' listen='127.0.0.1' passwd='123poi'/> + </devices> +</domain> diff -r e396919fa59e tests/xmconfigtest.c --- a/tests/xmconfigtest.c Tue Mar 31 16:35:42 2009 +0100 +++ b/tests/xmconfigtest.c Tue Mar 31 16:58:11 2009 +0100 @@ -209,6 +209,7 @@ mymain(int argc, char **argv) DO_TEST("paravirt-new-pvfb", 3); DO_TEST("paravirt-new-pvfb-vncdisplay", 3); DO_TEST("paravirt-net-e1000", 3); + DO_TEST("paravirt-net-vifname", 3); DO_TEST("fullvirt-old-cdrom", 1); DO_TEST("fullvirt-new-cdrom", 2); DO_TEST("fullvirt-utc", 2); -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] PATCH: Fix error reporting for guest with missing network
by Daniel P. Berrange 01 Apr '09

01 Apr '09

If you attempt to create a Xen guest using a virutal network that does not exist, or is not running you get a cryptic message # virsh create rhel5pv.xml error: Failed to create domain from rhel5pv.xml error: XML description for failed to build sexpr is not well formed or invalid This is because the XenD/XM drivers are both overwriting errors that have already been reported. The fix is simply to remove this bogus error call. It also tweaks the original error reporting to be more meaningful. The result is this # virsh create rhel5pv.xml error: Failed to create domain from rhel5pv.xml error: Network not found: default Daniel diff -r 28e19af5b719 src/xend_internal.c --- a/src/xend_internal.c Tue Mar 31 15:13:21 2009 +0100 +++ b/src/xend_internal.c Tue Mar 31 16:35:41 2009 +0100 @@ -3983,8 +3983,6 @@ xenDaemonCreateXML(virConnectPtr conn, c return (NULL); if (!(sexpr = xenDaemonFormatSxpr(conn, def, priv->xendConfigVersion))) { - virXendError(conn, VIR_ERR_XML_ERROR, - "%s", _("failed to build sexpr")); virDomainDefFree(def); return (NULL); } @@ -5362,7 +5360,7 @@ xenDaemonFormatSxprNet(virConnectPtr con char *bridge; if (!network) { - virXendError(conn, VIR_ERR_NO_SOURCE, "%s", + virXendError(conn, VIR_ERR_NO_NETWORK, "%s", def->data.network.name); return -1; } @@ -5370,7 +5368,8 @@ xenDaemonFormatSxprNet(virConnectPtr con bridge = virNetworkGetBridgeName(network); virNetworkFree(network); if (!bridge) { - virXendError(conn, VIR_ERR_NO_SOURCE, "%s", + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("network %s is not active"), def->data.network.name); return -1; } diff -r 28e19af5b719 src/xm_internal.c --- a/src/xm_internal.c Tue Mar 31 15:13:21 2009 +0100 +++ b/src/xm_internal.c Tue Mar 31 16:35:41 2009 +0100 @@ -1818,11 +1818,8 @@ int xenXMDomainCreate(virDomainPtr domai if (!(entry = virHashLookup(priv->configCache, filename))) goto error; - if (!(sexpr = xenDaemonFormatSxpr(domain->conn, entry->def, priv->xendConfigVersion))) { - xenXMError(domain->conn, VIR_ERR_XML_ERROR, - "%s", _("failed to build sexpr")); + if (!(sexpr = xenDaemonFormatSxpr(domain->conn, entry->def, priv->xendConfigVersion))) goto error; - } ret = xenDaemonDomainCreateXML(domain->conn, sexpr); VIR_FREE(sexpr); -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] PATCH: Fix error report for unknown node devices
by Daniel P. Berrange 01 Apr '09

01 Apr '09

When requesting a non-existant device you currently get # virsh nodedev-dumpxml foo error: Could not find matching device 'foo' error: invalid node device pointer in no node device with matching name This patch fixes that to give # virsh nodedev-dumpxml foo error: Could not find matching device 'foo' error: Node device not found Daniel diff -r 563f92056ffa src/node_device.c --- a/src/node_device.c Tue Mar 31 15:02:30 2009 +0100 +++ b/src/node_device.c Tue Mar 31 15:13:20 2009 +0100 @@ -121,8 +121,7 @@ static virNodeDevicePtr nodeDeviceLookup nodeDeviceUnlock(driver); if (!obj) { - virNodeDeviceReportError(conn, VIR_ERR_INVALID_NODE_DEVICE, - "%s", _("no node device with matching name")); + virNodeDeviceReportError(conn, VIR_ERR_NO_NODE_DEVICE, NULL); goto cleanup; } -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] PATCH: Update XenD for new location of (localtime 1) SEXPR
by Daniel P. Berrange 01 Apr '09

01 Apr '09

Newer versions of XenD have the (localtime 1) bit of the SEXPR moved outside the (image ...) bit, and made available for both PV and HVM guests. This patch updates our code to cope with this. For compatability we also still put the (localtime 1) bit inside the HVM (image ...) section too - its harmless on new XenD, and stops us breaking old deployments. Finally, this also updates the XM driver to accept localtime=1 for PV guests. Daniel diff -r 00b8aaa2555b src/xend_internal.c --- a/src/xend_internal.c Tue Mar 31 12:02:49 2009 +0100 +++ b/src/xend_internal.c Tue Mar 31 12:26:01 2009 +0100 @@ -2423,10 +2423,15 @@ xenDaemonParseSxpr(virConnectPtr conn, if (sexpr_int(root, "domain/image/hvm/pae")) def->features |= (1 << VIR_DOMAIN_FEATURE_PAE); + /* Old XenD only allows localtime here for HVM */ if (sexpr_int(root, "domain/image/hvm/localtime")) def->localtime = 1; } + /* Current XenD allows localtime here, for PV and HVM */ + if (sexpr_int(root, "domain/localtime")) + def->localtime = 1; + if (sexpr_node_copy(root, hvm ? "domain/image/hvm/device_model" : "domain/image/linux/device_model", @@ -5603,6 +5608,10 @@ xenDaemonFormatSxpr(virConnectPtr conn, } virBufferVSprintf(&buf, "(on_crash '%s')", tmp); + /* Set localtime here for current XenD (both PV & HVM) */ + if (def->localtime) + virBufferAddLit(&buf, "(localtime 1)"); + if (!def->os.bootloader) { if (STREQ(def->os.type, "hvm")) hvm = 1; @@ -5718,6 +5727,7 @@ xenDaemonFormatSxpr(virConnectPtr conn, virBufferAddLit(&buf, "(serial none)"); } + /* Set localtime here to keep old XenD happy for HVM */ if (def->localtime) virBufferAddLit(&buf, "(localtime 1)"); diff -r 00b8aaa2555b src/xm_internal.c --- a/src/xm_internal.c Tue Mar 31 12:02:49 2009 +0100 +++ b/src/xm_internal.c Tue Mar 31 12:26:01 2009 +0100 @@ -819,10 +819,10 @@ xenXMDomainConfigParse(virConnectPtr con goto cleanup; else if (val) def->features |= (1 << VIR_DOMAIN_FEATURE_APIC); + } + if (xenXMConfigGetBool(conn, conf, "localtime", &def->localtime, 0) < 0) + goto cleanup; - if (xenXMConfigGetBool(conn, conf, "localtime", &def->localtime, 0) < 0) - goto cleanup; - } if (xenXMConfigCopyStringOpt(conn, conf, "device_model", &def->emulator) < 0) goto cleanup; diff -r 00b8aaa2555b tests/sexpr2xmldata/sexpr2xml-pv-localtime.sexpr --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/sexpr2xmldata/sexpr2xml-pv-localtime.sexpr Tue Mar 31 12:26:01 2009 +0100 @@ -0,0 +1,2 @@ +(domain (domid 6)(name 'pvtest')(memory 420)(maxmem 420)(vcpus 2)(uuid '596a5d2171f48fb2e068e2386a5c413e')(on_poweroff 'destroy')(on_reboot 'destroy')(on_crash 'destroy')(localtime 1)(image (linux (kernel '/var/lib/xen/vmlinuz.2Dn2YT')(ramdisk '/var/lib/xen/initrd.img.0u-Vhq')(args ' method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… ')))(device (vbd (dev 'xvda')(uname 'file:/root/some.img')(mode 'w')))) + diff -r 00b8aaa2555b tests/sexpr2xmldata/sexpr2xml-pv-localtime.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/sexpr2xmldata/sexpr2xml-pv-localtime.xml Tue Mar 31 12:26:01 2009 +0100 @@ -0,0 +1,27 @@ +<domain type='xen' id='6'> + <name>pvtest</name> + <uuid>596a5d21-71f4-8fb2-e068-e2386a5c413e</uuid> + <memory>430080</memory> + <currentMemory>430080</currentMemory> + <vcpu>2</vcpu> + <os> + <type>linux</type> + <kernel>/var/lib/xen/vmlinuz.2Dn2YT</kernel> + <initrd>/var/lib/xen/initrd.img.0u-Vhq</initrd> + <cmdline> method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… </cmdline> + </os> + <clock offset='localtime'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>destroy</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <disk type='file' device='disk'> + <driver name='file'/> + <source file='/root/some.img'/> + <target dev='xvda' bus='xen'/> + </disk> + <console type='pty'> + <target port='0'/> + </console> + </devices> +</domain> diff -r 00b8aaa2555b tests/sexpr2xmltest.c --- a/tests/sexpr2xmltest.c Tue Mar 31 12:02:49 2009 +0100 +++ b/tests/sexpr2xmltest.c Tue Mar 31 12:26:01 2009 +0100 @@ -145,6 +145,7 @@ mymain(int argc, char **argv) DO_TEST("bridge-ipaddr", "bridge-ipaddr", 3); DO_TEST("no-source-cdrom", "no-source-cdrom", 2); DO_TEST("pci-devs", "pci-devs", 2); + DO_TEST("pv-localtime", "pv-localtime", 2); DO_TEST("fv-utc", "fv-utc", 1); DO_TEST("fv-localtime", "fv-localtime", 1); diff -r 00b8aaa2555b tests/xml2sexprdata/xml2sexpr-fv-localtime.sexpr --- a/tests/xml2sexprdata/xml2sexpr-fv-localtime.sexpr Tue Mar 31 12:02:49 2009 +0100 +++ b/tests/xml2sexprdata/xml2sexpr-fv-localtime.sexpr Tue Mar 31 12:26:01 2009 +0100 @@ -1,1 +1,1 @@ -(vm (name 'fvtest')(memory 400)(maxmem 400)(vcpus 1)(uuid 'b5d70dd2-75cd-aca5-1776-9660b059d8bc')(on_poweroff 'destroy')(on_reboot 'restart')(on_crash 'restart')(image (hvm (kernel '/usr/lib/xen/boot/hvmloader')(vcpus 1)(boot c)(cdrom '/root/boot.iso')(acpi 1)(usb 1)(parallel none)(serial none)(localtime 1)(device_model '/usr/lib64/xen/bin/qemu-dm')(vnc 1)))(device (vbd (dev 'ioemu:hda')(uname 'file:/root/foo.img')(mode 'w')))(device (vif (mac '00:16:3e:1b:b1:47')(bridge 'xenbr0')(script 'vif-bridge')(type ioemu)))) \ No newline at end of file +(vm (name 'fvtest')(memory 400)(maxmem 400)(vcpus 1)(uuid 'b5d70dd2-75cd-aca5-1776-9660b059d8bc')(on_poweroff 'destroy')(on_reboot 'restart')(on_crash 'restart')(localtime 1)(image (hvm (kernel '/usr/lib/xen/boot/hvmloader')(vcpus 1)(boot c)(cdrom '/root/boot.iso')(acpi 1)(usb 1)(parallel none)(serial none)(localtime 1)(device_model '/usr/lib64/xen/bin/qemu-dm')(vnc 1)))(device (vbd (dev 'ioemu:hda')(uname 'file:/root/foo.img')(mode 'w')))(device (vif (mac '00:16:3e:1b:b1:47')(bridge 'xenbr0')(script 'vif-bridge')(type ioemu)))) \ No newline at end of file diff -r 00b8aaa2555b tests/xml2sexprdata/xml2sexpr-pv-localtime.sexpr --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/xml2sexprdata/xml2sexpr-pv-localtime.sexpr Tue Mar 31 12:26:01 2009 +0100 @@ -0,0 +1,1 @@ +(vm (name 'rhel5')(memory 175)(maxmem 385)(vcpus 1)(uuid '4f77abd2-3019-58e8-3bab-6fbf2118f880')(bootloader '/usr/bin/pygrub')(on_poweroff 'destroy')(on_reboot 'restart')(on_crash 'restart')(localtime 1)(device (tap (dev 'xvda:disk')(uname 'tap:aio:/xen/rhel5.img')(mode 'w')))(device (vif (mac '00:16:3e:1d:06:15')(bridge 'xenbr0')(script 'vif-bridge')))) \ No newline at end of file diff -r 00b8aaa2555b tests/xml2sexprdata/xml2sexpr-pv-localtime.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/xml2sexprdata/xml2sexpr-pv-localtime.xml Tue Mar 31 12:26:01 2009 +0100 @@ -0,0 +1,25 @@ +<domain type='xen' id='5'> + <name>rhel5</name> + <uuid>4f77abd2301958e83bab6fbf2118f880</uuid> + <bootloader>/usr/bin/pygrub</bootloader> + <memory>394240</memory> + <currentMemory>179200</currentMemory> + <vcpu>1</vcpu> + <on_poweroff>destroy</on_poweroff> + <on_reboot>restart</on_reboot> + <on_crash>restart</on_crash> + <clock offset='localtime'/> + <devices> + <interface type='bridge'> + <source bridge='xenbr0'/> + <mac address='00:16:3e:1d:06:15'/> + <script path='vif-bridge'/> + </interface> + <disk type='file' device='disk'> + <driver name='tap' type='aio'/> + <source file='/xen/rhel5.img'/> + <target dev='xvda:disk'/> + </disk> + <graphics type='vnc' port='5905'/> + </devices> +</domain> diff -r 00b8aaa2555b tests/xml2sexprtest.c --- a/tests/xml2sexprtest.c Tue Mar 31 12:02:49 2009 +0100 +++ b/tests/xml2sexprtest.c Tue Mar 31 12:26:01 2009 +0100 @@ -130,6 +130,7 @@ mymain(int argc, char **argv) DO_TEST("bridge-ipaddr", "bridge-ipaddr", "pvtest", 2); DO_TEST("no-source-cdrom", "no-source-cdrom", "test", 2); DO_TEST("pci-devs", "pci-devs", "pvtest", 2); + DO_TEST("pv-localtime", "pv-localtime", "pvtest", 1); DO_TEST("fv-utc", "fv-utc", "fvtest", 1); DO_TEST("fv-localtime", "fv-localtime", "fvtest", 1); Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] PATCH: Clear reference after closing connection object
by Daniel P. Berrange 01 Apr '09

01 Apr '09

In the python bindings, if you call close() it will close the connection and release the handle, but it doesn't clear the python reference to the C object - self._o. So later in garbage colection __del__ runs and tries to free the already free'd object Daniel diff -r a76dc8c27d72 python/generator.py --- a/python/generator.py Tue Mar 31 12:36:02 2009 +0100 +++ b/python/generator.py Tue Mar 31 13:09:38 2009 +0100 @@ -1072,6 +1072,9 @@ def buildWrappers(): n = n + 1 classes.write(")\n"); + if name == "virConnectClose": + classes.write(" self._o = None\n") + # For functions returning object types: if ret[0] != "void": if classes_type.has_key(ret[0]): -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] PATCH: Fix duplicate methods in python binding
by Daniel P. Berrange 01 Apr '09

01 Apr '09

There are three methods with the same name in the python binding due to a generator bug. This patch fixes them, so we get methods networkCreateXML() and storagePoolCreateXML() instad of three called createXML() Daniel diff -r e26692afef21 python/generator.py --- a/python/generator.py Tue Mar 31 12:26:01 2009 +0100 +++ b/python/generator.py Tue Mar 31 12:36:02 2009 +0100 @@ -684,12 +684,18 @@ def nameFixup(name, classe, type, file): elif name[0:16] == "virNetworkDefine": func = name[3:] func = string.lower(func[0:1]) + func[1:] + elif name[0:19] == "virNetworkCreateXML": + func = name[3:] + func = string.lower(func[0:1]) + func[1:] elif name[0:16] == "virNetworkLookup": func = name[3:] func = string.lower(func[0:1]) + func[1:] elif name[0:20] == "virStoragePoolDefine": func = name[3:] func = string.lower(func[0:1]) + func[1:] + elif name[0:23] == "virStoragePoolCreateXML": + func = name[3:] + func = string.lower(func[0:1]) + func[1:] elif name[0:20] == "virStoragePoolLookup": func = name[3:] func = string.lower(func[0:1]) + func[1:] -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] f11 virt release notes freezing
by Dale Bewley 01 Apr '09

01 Apr '09

I've been busy at my day job and not paying close enough attention to Fedora 11 docs deadlines. Apparently the window for changes essentially closes Wednesday (today).[1] The notes are essentially in place. I don't think there are any gaping holes, but I should have prompted for feedback much sooner. If you could apply any love to the release notes[2] it would be appreciated. = QEMU / KVM = It is noted that QEMU and KVM are merging as a package, but I've also separately listed the fact that QEMU is now 0.10.0 and KVM is now 84. With the merge, might it be confusing or inaccurate to mention a distinct new version of KVM? I'll admit to being somewhat fuzzy on where/how that line is drawn. I see[3] QEMU 0.10.0 in rawhide, but not 0.10.1, so I adjusted the version in the relnotes. I copied the feature list from the 0.10.0 release announcement. If you'd care to suggest deletes or enhancements of the list please do. The KVM changelog since 74 (version at initial F10 release) is quite long and not easily digestible by me at this late hour. Would anyone care to offer some highlights or fill them into the wiki? Or, again, should those be folded into QEMU? = Libvirt = I previously copied the list of changes from the releases, but did not try to prune them down. It may be fine. = The Current Outline = 1 Virtualization * 1.1 Improved VNC Authentication for Virtual Machine Management * 1.2 Improved Graphical Console for Virtual Machines * 1.3 KVM PCI Device Assignment * 1.4 KVM and QEMU merge * 1.5 SVirt Mandatory Access Control * 1.6 Other Improvements * 1.6.1 QEMU Updated to 0.10.0 * 1.6.2 KVM Updated to 84 * 1.6.3 libvirt Updated to 0.6.1 * 1.6.4 virt-manager Updated to 0.7.0 * 1.6.5 virtinst Updated to 0.400.3 * 1.6.6 Xen Updated to 3.3.1 * 1.7 Xen Kernel Support Thanks for any edits or suggestions you can offer. [1] http://www.redhat.com/archives/fedora-docs-list/2009-March/msg00232.html [2] http://fedoraproject.org/wiki/Documentation_Virtualization_Beat [3] http://fedoraproject.org/wiki/User:Dale#Packages

1 0

[libvirt] [Fwd: first cut public API for physical host interface configuration]
by Laine Stump 01 Apr '09

01 Apr '09

(Oops, I originally sent this with the wrong return address, which probably caused it to be silently swallowed...) To get started integrating libnetcf support into libvirt, the attached libvirt.h diff has a first attempt at the public API that will hook up to libnetcf on the libvirtd side. I started out with the virNetwork* API, and modified/removed as seemed appropriate. A few points worth mentioning: virNetwork has "defined" and "active" interfaces, but so far virInterface just has interfaces, as the active/inactive status is really controlled by 1) the "onboot" property in the XML, and 2) whether or not virInterfaceStart() has been called yet. The _virInterface struct that is referenced here is more or less identical to _virnetwork. libnetcf works with netcf structs (one per library instance) and netcf_if structs (one per interface, multiple per library instance. As far as I can see right now, the netcf struct will not need to be referenced directly from the client side, but the netcf_if for each interface is needed, and I guess a cookie representing it will be sent to the client side and stored in the _virInterface. Is that what the UUID in _virNetwork is used for? (I know, I know - "read the code!", it's just quicker to (and avoids misreading on my part) to ask) As before, any and all advice/corrections gratefully accepted! diff --git a/include/libvirt/libvirt.h b/include/libvirt/libvirt.h index 779ea72..cac400e 100644 --- a/include/libvirt/libvirt.h +++ b/include/libvirt/libvirt.h @@ -854,7 +854,73 @@ int virNetworkGetAutostart (virNetworkPtr network, int virNetworkSetAutostart (virNetworkPtr network, int autostart); +/* + * Physical host interface configuration API + */ + +/** + * virInterface: + * + * a virInterface is a private structure representing a virtual interface. + */ +typedef struct _virInterface virInterface; + +/** + * virInterfacePtr: + * + * a virInterfacePtr is pointer to a virInterface private structure, this is the + * type used to reference a virtual interface in the API. + */ +typedef virInterface *virInterfacePtr; + +/* + * Get connection from interface. + */ +virConnectPtr virInterfaceGetConnect (virInterfacePtr interface); + +/* + * List defined interfaces + */ +int virConnectNumOfInterfaces (virConnectPtr conn); +int virConnectListInterfaces (virConnectPtr conn, + char **const names, + int maxnames); + +/* + * Lookup interface by name + */ +virInterfacePtr virInterfaceLookupByName (virConnectPtr conn, + const char *name); + const char *uuid); + +/* + * Define interface (or modify existing interface configuration) + */ +virInterfacePtr virInterfaceDefineXML (virConnectPtr conn, + const char *xmlDesc); + +/* + * Delete interface + */ +int virInterfaceUndefine (virInterfacePtr interface); + +/* + * Activate interface (ie call "ifup") + */ +int virInterfaceStart (virInterfacePtr interface); + +/* + * De-activate interface (call "ifdown") + */ +int virInterfaceStop (virInterfacePtr interface); + +/* + * Interface information + */ +const char* virInterfaceGetName (virInterfacePtr interface); +char * virInterfaceGetXMLDesc (virInterfacePtr interface, + int flags); /** * virStoragePool: *

4 10

[libvirt] [PATCH 2/2]: Remove getURI internal driver call
by Chris Lalancette 31 Mar '09

31 Mar '09

This patch removes the internal getURI() driver callback. It was *never* overridden by any of the drivers, so it seems kind of pointless (I even checked the "in-progress" VirtualBox, OpenNebula, and Power Hypervisor drivers). In addition, the code to actually use it in src/libvirt.c:virConnectGetURI() was buggy; if a driver did decide to have their own callback, then their callback would be called, and then immediately overwritten by the generic stuff. This wouldn't do what the driver expected, and would also probably leak memory. Just rip out this internal callback; if a future driver ever needs it, we can add it back in. Signed-off-by: Chris Lalancette <clalance(a)redhat.com>

3 3

[libvirt] [PATCH 1/2]: Convert internal drivers to old style struct initializers
by Chris Lalancette 31 Mar '09

31 Mar '09

This patch converts the internal drivers to use the old-style struct initializers, which kind of gives us a self-documenting TODO list for each of the drivers. It also fixes up some bad comments in the internal driver list. It does not touch the storage or network driver initializers. Signed-off-by: Chris Lalancette <clalance(a)redhat.com>

3 3

[libvirt] [PATCH] additional options in the spec file for storage drivers and numactl
by Ryota Ozaki 31 Mar '09

31 Mar '09

Hi, This patch adds switches in the spec file for storage drivers and numactl to be easy to turn on/off options of configure. Thanks, ozaki-r Signed-off-by: Ryota Ozaki <ozaki.ryota(a)gmail.com>

3 2

[libvirt] [PATCH] prevent an executed external process to be a zombie
by Ryota Ozaki 31 Mar '09

31 Mar '09

Hi, This patch prevents an executed external process to be a zombie by doing waitpid() even if virPipeReadUntilEOF() fails. Thanks, ozaki-r Signed-off-by: Ryota Ozaki <ozaki.ryota(a)gmail.com>

3 3

[libvirt] Patch to allow setting of svirt XML.
by Daniel J Walsh 31 Mar '09

31 Mar '09

This patch fixes the seclabel handling in domain_conf.c to allow virt-manager to set the seclabel model, type and label. Also adds missing error messages when the xml is incorrect. How much verification should we be doing on this? I have another patch that verifies the model as being a known model and a patch to verify the label is a correct label. (IE SELinux verifies the label is understood by the kernel.) The problem with this second patch is it sucks in security.[ch], security_selinux.[ch] into the libvirt_lxc. Should I be doing this?

2 3

[libvirt] Upcoming 0.6.2 release
by Daniel Veillard 31 Mar '09

31 Mar '09

I would like to make a new release mid or end of next week, considering the number of fixes accumulated in CVS compared to 0.6.1, I guess 0.6.2 makes sense. I think we should look at integrating the new SCSI pool patch, hopefully there is no big issues. If there are mails or patchs which are pending and we forgot to review or ACK, please raise them again so that we can try to get them in time for next release, thanks, Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

4 4

[libvirt] PATCH: Misc error handling / OOM fixes
by Daniel P. Berrange 31 Mar '09

31 Mar '09

I ran the OOM checking test suite again and uncovered a handful of pretty minor bugs - no crashers in real code - just forgetting to report OOM correctly * src/datatypes.c: Convert to use the virReportErrorHelper macro instead of a static function, so we get correct function/line number info * src/domain_conf.c: Fix OOM handling in virXPathNodeSet() call so it reports OOM error, instead of ignoring * src/nodeinfo.c: Use _(...) for error message strings * src/xend_internal.c: Add missing check of virBufferError() func and report OOM error * tests/sexpr2xmltest.c: Reset error object between tests, free capabilities object, don't unref a NULL connection object * tests/testutils.c: Fix crash in OOM testing if single threaded * tests/xml2sexprtest.c: Reset error object between tests Daniel Index: src/datatypes.c =================================================================== RCS file: /data/cvs/libvirt/src/datatypes.c,v retrieving revision 1.9 diff -u -p -r1.9 datatypes.c --- src/datatypes.c 29 Jan 2009 12:10:32 -0000 1.9 +++ src/datatypes.c 30 Mar 2009 11:47:05 -0000 @@ -28,31 +28,15 @@ #define VIR_FROM_THIS VIR_FROM_NONE +#define virLibConnError(conn, code, fmt...) \ + virReportErrorHelper(conn, VIR_FROM_THIS, code, __FILE__, \ + __FUNCTION__, __LINE__, fmt) + /************************************************************************ * * * Domain and Connections allocations * * * ************************************************************************/ -/** - * virLibConnError: - * @conn: the connection if available - * @error: the error number - * @info: extra information string - * - * Handle an error at the connection level - */ -static void -virLibConnError(virConnectPtr conn, virErrorNumber error, const char *info) -{ - const char *errmsg; - - if (error == VIR_ERR_OK) - return; - - errmsg = virErrorMsg(error, info); - virRaiseError(conn, NULL, NULL, VIR_FROM_NONE, error, VIR_ERR_ERROR, - errmsg, info, NULL, 0, 0, errmsg, info); -} /** * virDomainFreeName: Index: src/domain_conf.c =================================================================== RCS file: /data/cvs/libvirt/src/domain_conf.c,v retrieving revision 1.72 diff -u -p -r1.72 domain_conf.c --- src/domain_conf.c 24 Mar 2009 11:16:29 -0000 1.72 +++ src/domain_conf.c 30 Mar 2009 11:47:05 -0000 @@ -2068,7 +2068,10 @@ static virDomainDefPtr virDomainDefParse VIR_FREE(tmp); } - if ((n = virXPathNodeSet(conn, "./features/*", ctxt, &nodes)) > 0) { + n = virXPathNodeSet(conn, "./features/*", ctxt, &nodes); + if (n < 0) + goto error; + if (n) { for (i = 0 ; i < n ; i++) { int val = virDomainFeatureTypeFromString((const char *)nodes[i]->name); if (val < 0) { @@ -2079,8 +2082,8 @@ static virDomainDefPtr virDomainDefParse } def->features |= (1 << val); } + VIR_FREE(nodes); } - VIR_FREE(nodes); if (virDomainLifecycleParseXML(conn, ctxt, "string(./on_reboot[1])", &def->onReboot, VIR_DOMAIN_LIFECYCLE_RESTART) < 0) Index: src/nodeinfo.c =================================================================== RCS file: /data/cvs/libvirt/src/nodeinfo.c,v retrieving revision 1.17 diff -u -p -r1.17 nodeinfo.c --- src/nodeinfo.c 20 Jan 2009 17:13:33 -0000 1.17 +++ src/nodeinfo.c 30 Mar 2009 11:47:05 -0000 @@ -74,8 +74,8 @@ int linuxNodeInfoCPUPopulate(virConnectP buf++; if (*buf != ':') { virRaiseError(conn, NULL, NULL, 0, VIR_ERR_INTERNAL_ERROR, - VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, - "parsing cpuinfo processor"); + VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, + "%s", _("parsing cpuinfo processor")); return -1; } nodeinfo->cpus++; @@ -87,8 +87,8 @@ int linuxNodeInfoCPUPopulate(virConnectP buf++; if (*buf != ':' || !buf[1]) { virRaiseError(conn, NULL, NULL, 0, VIR_ERR_INTERNAL_ERROR, - VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, - "parsing cpuinfo cpu MHz"); + VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, + "%s", _("parsing cpuinfo cpu MHz")); return -1; } if (virStrToLong_ui(buf+1, &p, 10, &ui) == 0 @@ -103,8 +103,8 @@ int linuxNodeInfoCPUPopulate(virConnectP buf++; if (*buf != ':' || !buf[1]) { virRaiseError(conn, NULL, NULL, 0, VIR_ERR_INTERNAL_ERROR, - VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, - "parsing cpuinfo cpu cores %c", *buf); + VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, + "parsing cpuinfo cpu cores %c", *buf); return -1; } if (virStrToLong_ui(buf+1, &p, 10, &id) == 0 @@ -116,8 +116,8 @@ int linuxNodeInfoCPUPopulate(virConnectP if (!nodeinfo->cpus) { virRaiseError(conn, NULL, NULL, 0, VIR_ERR_INTERNAL_ERROR, - VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, - "no cpus found"); + VIR_ERR_ERROR, NULL, NULL, NULL, 0, 0, + "%s", _("no cpus found")); return -1; } Index: src/xend_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/xend_internal.c,v retrieving revision 1.253 diff -u -p -r1.253 xend_internal.c --- src/xend_internal.c 10 Mar 2009 11:13:32 -0000 1.253 +++ src/xend_internal.c 30 Mar 2009 11:47:06 -0000 @@ -5540,6 +5540,11 @@ xenDaemonFormatSxpr(virConnectPtr conn, virBufferAddLit(&buf, ")"); /* closes (vm */ + if (virBufferError(&buf)) { + virReportOOMError(conn); + return NULL; + } + return virBufferContentAndReset(&buf); error: Index: tests/sexpr2xmltest.c =================================================================== RCS file: /data/cvs/libvirt/tests/sexpr2xmltest.c,v retrieving revision 1.36 diff -u -p -r1.36 sexpr2xmltest.c --- tests/sexpr2xmltest.c 29 Jan 2009 17:02:00 -0000 1.36 +++ tests/sexpr2xmltest.c 30 Mar 2009 11:47:06 -0000 @@ -64,7 +64,8 @@ static int testCompareFiles(const char * fail: free(gotxml); virDomainDefFree(def); - virUnrefConnect(conn); + if (conn) + virUnrefConnect(conn); return ret; } @@ -115,6 +116,7 @@ mymain(int argc, char **argv) #define DO_TEST(in, out, version) \ do { \ struct testInfo info = { in, out, version }; \ + virResetLastError(); \ if (virtTestRun("Xen SEXPR-2-XML " in " -> " out, \ 1, testCompareHelper, &info) < 0) \ ret = -1; \ @@ -163,6 +165,8 @@ mymain(int argc, char **argv) DO_TEST("fv-sound", "fv-sound", 1); DO_TEST("fv-sound-all", "fv-sound-all", 1); + virCapabilitiesFree(caps); + return(ret==0 ? EXIT_SUCCESS : EXIT_FAILURE); } Index: tests/testutils.c =================================================================== RCS file: /data/cvs/libvirt/tests/testutils.c,v retrieving revision 1.26 diff -u -p -r1.26 testutils.c --- tests/testutils.c 2 Feb 2009 20:35:14 -0000 1.26 +++ tests/testutils.c 30 Mar 2009 11:47:06 -0000 @@ -409,7 +435,8 @@ int virtTestMain(int argc, /* Run once for each alloc, failing a different one and validating that the test case failed */ for (n = 0; n < approxAlloc && (!mp || worker) ; n++) { - if ((n % mp) != (worker - 1)) + if (mp && + (n % mp) != (worker - 1)) continue; if (!testDebug) { if (mp) Index: tests/xml2sexprtest.c =================================================================== RCS file: /data/cvs/libvirt/tests/xml2sexprtest.c,v retrieving revision 1.34 diff -u -p -r1.34 xml2sexprtest.c --- tests/xml2sexprtest.c 23 Jan 2009 01:48:47 -0000 1.34 +++ tests/xml2sexprtest.c 30 Mar 2009 11:47:06 -0000 @@ -95,6 +95,7 @@ mymain(int argc, char **argv) #define DO_TEST(in, out, name, version) \ do { \ struct testInfo info = { in, out, name, version }; \ + virResetLastError(); \ if (virtTestRun("Xen XML-2-SEXPR " in " -> " out, \ 1, testCompareHelper, &info) < 0) \ ret = -1; \ -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] [PATCH 0/2]: Cleanup driver initializers
by Chris Lalancette 30 Mar '09

30 Mar '09

All, This is a small series to cleanup the internal driver initializers. The drivers were using an annoying combination of C-99 style initialization and old-style (C89?) initialization. On top of that, the indentation on some of these was wacky, making it even harder to read. After some discussion with DanB, we determined to go with the old-style initializers because it is sort of a self-documenting TODO list. The first patch converts the internal drivers to all use this style of initialization, and fixes up a few wrong comments along the way. Note that this doesn't cleanup the storage or network initializers; assuming this patch is acceptable, we can do those next. The second patch rips out the internal "getURI" callbacks, since they were never, ever used (and buggy to boot). More details in the second patch. -- Chris Lalancette

2 1

[libvirt] [PATCH 1/2] OpenNebula driver, libvirt-0.6.1
by "Abel Míguez Rodríguez" 30 Mar '09

30 Mar '09

Hi all, We have updated the OpenNebula Driver to libvirt version 0.6.1. Now, the ONE driver is build at the libvirtd daemon, that is the natural place for it. Please feel free to make any comment, to improve the driver's coherence with libvirt's structure. I split the Patches in two e-mais: [PATCH 1/2] includes the patches to be applied to libvirt's sources and building files. [PATCH 2/2] attached the "one driver" source files. All the patches are made to be applied at the git commit: "025b62" (Fix subsystem lookup for older HAL releases) Thanks, Abel Miguez diff --git a/configure.in b/configure.in index 413d27c..fd75a8d 100644 --- a/configure.in +++ b/configure.in @@ -184,6 +184,8 @@ AC_ARG_WITH([openvz], [ --with-openvz add OpenVZ support (on)],[],[with_openvz=yes]) AC_ARG_WITH([lxc], [ --with-lxc add Linux Container support (on)],[],[with_lxc=yes]) +AC_ARG_WITH([one], +[ --with-one add ONE support (on)],[],[with_one=no]) AC_ARG_WITH([test], [ --with-test add test driver support (on)],[],[with_test=yes]) AC_ARG_WITH([remote], @@ -399,6 +401,17 @@ dnl check for kvm headers dnl AC_CHECK_HEADERS([linux/kvm.h]) +dnl OpenNebula driver Compilation setting +dnl + +if test "$with_one" = "yes" ; then + LIBVIRT_FEATURES="$LIBVIRT_FEATURES -DWITH_ONE -I$ONE_LOCATION/include" + ONE_LIBS="-L/usr/local/lib -lxmlrpc_client++ -lxmlrpc -lxmlrpc_util -lxmlrpc_xmlparse -lxmlrpc_xmltok -lxmlrpc++ -lxmlrpc_client -L$ONE_LOCATION/lib -loneapi" + AC_SUBST([ONE_LIBS]) + AC_DEFINE_UNQUOTED([WITH_ONE],1,[whether Open Nebula Driver is enabled]) +fi +AM_CONDITIONAL([WITH_ONE],[test "$with_one" = "yes"]) + dnl Need to test if pkg-config exists PKG_PROG_PKG_CONFIG @@ -1345,6 +1358,7 @@ AC_MSG_NOTICE([ QEMU: $with_qemu]) AC_MSG_NOTICE([ UML: $with_uml]) AC_MSG_NOTICE([ OpenVZ: $with_openvz]) AC_MSG_NOTICE([ LXC: $with_lxc]) +AC_MSG_NOTICE([ ONE: $with_one]) AC_MSG_NOTICE([ Test: $with_test]) AC_MSG_NOTICE([ Remote: $with_remote]) AC_MSG_NOTICE([ Network: $with_network]) diff --git a/include/libvirt/virterror.h b/include/libvirt/virterror.h index 2c3777d..cc98e45 100644 --- a/include/libvirt/virterror.h +++ b/include/libvirt/virterror.h @@ -61,6 +61,7 @@ typedef enum { VIR_FROM_UML, /* Error at the UML driver */ VIR_FROM_NODEDEV, /* Error from node device monitor */ VIR_FROM_XEN_INOTIFY, /* Error from xen inotify layer */ + VIR_FROM_ONE, /* Error from ONE driver */ VIR_FROM_SECURITY, /* Error from security framework */ } virErrorDomain; diff --git a/qemud/Makefile.am b/qemud/Makefile.am index 924e8ad..9d7f61f 100644 --- a/qemud/Makefile.am +++ b/qemud/Makefile.am @@ -120,6 +120,10 @@ if WITH_UML libvirtd_LDADD += ../src/libvirt_driver_uml.la endif +if WITH_ONE + libvirtd_LDADD += ../src/libvirt_driver_one.la +endif + if WITH_STORAGE_DIR libvirtd_LDADD += ../src/libvirt_driver_storage.la endif diff --git a/qemud/qemud.c b/qemud/qemud.c index 4f04355..e1d6113 100644 --- a/qemud/qemud.c +++ b/qemud/qemud.c @@ -78,6 +78,9 @@ #ifdef WITH_NETWORK #include "network_driver.h" #endif +#ifdef WITH_ONE +#include "one_driver.h" +#endif #ifdef WITH_STORAGE_DIR #include "storage_driver.h" #endif @@ -841,6 +844,8 @@ static struct qemud_server *qemudInitialize(int sigread) { virDriverLoadModule("qemu"); virDriverLoadModule("lxc"); virDriverLoadModule("uml"); + virDriverLoadModule("one"); + #else #ifdef WITH_NETWORK networkRegister(); @@ -861,6 +866,10 @@ static struct qemud_server *qemudInitialize(int sigread) { #ifdef WITH_UML umlRegister(); #endif +#ifdef WITH_ONE + oneRegister (); +#endif + #endif virEventRegisterImpl(virEventAddHandleImpl, diff --git a/src/Makefile.am b/src/Makefile.am index d5aac11..a5c2084 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -137,6 +137,10 @@ UML_DRIVER_SOURCES = \ uml_conf.c uml_conf.h \ uml_driver.c uml_driver.h +ONE_DRIVER_SOURCES = \ + one_conf.c one_conf.h \ + one_driver.c one_driver.h + NETWORK_DRIVER_SOURCES = \ network_driver.h network_driver.c @@ -314,6 +318,22 @@ endif libvirt_driver_uml_la_SOURCES = $(UML_DRIVER_SOURCES) endif +if WITH_ONE +if WITH_DRIVER_MODULES +mod_LTLIBRARIES += libvirt_driver_one.la +else +noinst_LTLIBRARIES += libvirt_driver_one.la +# Stateful, so linked to daemon instead +#libvirt_la_LIBADD += libvirt_driver_one.la +endif +libvirt_driver_one_la_LDFLAGS = $(ONE_LIBS) +libvirt_driver_one_la_CFLAGS = "-DWITH_ONE" +if WITH_DRIVER_MODULES +libvirt_driver_one_la_LDFLAGS += -module -avoid-version +endif +libvirt_driver_one_la_SOURCES = $(ONE_DRIVER_SOURCES) +endif + if WITH_NETWORK if WITH_DRIVER_MODULES mod_LTLIBRARIES += libvirt_driver_network.la @@ -402,6 +422,7 @@ EXTRA_DIST += \ $(QEMU_DRIVER_SOURCES) \ $(LXC_DRIVER_SOURCES) \ $(UML_DRIVER_SOURCES) \ + $(ONE_DRIVER_SOURCES) \ $(OPENVZ_DRIVER_SOURCES) \ $(NETWORK_DRIVER_SOURCES) \ $(STORAGE_DRIVER_SOURCES) \ diff --git a/src/domain_conf.c b/src/domain_conf.c index 5bf3483..e4d3249 100644 --- a/src/domain_conf.c +++ b/src/domain_conf.c @@ -54,7 +54,9 @@ VIR_ENUM_IMPL(virDomainVirt, VIR_DOMAIN_VIRT_LAST, "ldom", "test", "vmware", - "hyperv") + "hyperv", + "one") + VIR_ENUM_IMPL(virDomainBoot, VIR_DOMAIN_BOOT_LAST, "fd", diff --git a/src/domain_conf.h b/src/domain_conf.h index dd61467..e8a2bff 100644 --- a/src/domain_conf.h +++ b/src/domain_conf.h @@ -48,6 +48,7 @@ enum virDomainVirtType { VIR_DOMAIN_VIRT_TEST, VIR_DOMAIN_VIRT_VMWARE, VIR_DOMAIN_VIRT_HYPERV, + VIR_DOMAIN_VIRT_ONE, VIR_DOMAIN_VIRT_LAST, }; diff --git a/src/driver.h b/src/driver.h index 62d6fbc..ed3eef7 100644 --- a/src/driver.h +++ b/src/driver.h @@ -20,6 +20,7 @@ typedef enum { VIR_DRV_OPENVZ = 5, VIR_DRV_LXC = 6, VIR_DRV_UML = 7, + VIR_DRV_ONE = 8, } virDrvNo; diff --git a/src/libvirt.c b/src/libvirt.c index bf3453a..cd4b5b7 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -830,6 +830,10 @@ virGetVersion(unsigned long *libVer, const char *type, if (STRCASEEQ(type, "OpenVZ")) *typeVer = LIBVIR_VERSION_NUMBER; #endif +#if WITH_ONE + if (STRCASEEQ(type, "ONE")) + *typeVer = LIBVIR_VERSION_NUMBER; +#endif #if WITH_UML if (STRCASEEQ(type, "UML")) *typeVer = LIBVIR_VERSION_NUMBER; ---- Distributed System Architecture Group (http://dsa-research.org) GridWay, http://www.gridway.org OpenNEbula, http://www.opennebula.org

3 2

[libvirt] KVM update
by Zvi Dubitzky 30 Mar '09

30 Mar '09

I am trying to update the kvm driver and userspace using existing kernel 1) using the tar ball from kvm at sourceforge I can do it , but only kvm -84 is there 2) if i try the latest version from the git tree I am able to do the first 2 steps (make --kerneldir=<exisitng kernel headers> , make sync ) . but the final make fails during compile of libkvm.c. results in the file below (dated 3/29/09) Am I missing something Should I recompile the whole kerenl as a 3) Was the live migrate fixed and tested in the latest kernel version thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

3 2

[libvirt] "invalid connection pointer in virConnectClose" from Python bindings
by Charles Duffy 30 Mar '09

30 Mar '09

Howdy. I'm calling libvirt from a program which occasionally has cause to fork() without an immediate exec(). For the sake of simplicity, I presently call close() on all my virConnect objects [which I then delete] before forking and create new ones later. (I'm not forcing an explicit pre-fork garbage collection at present -- hopefully the close() should make one unnecessary). However, libvirt complains (and very occasionally segfaults) as I try to close the connections: libvir: error : invalid connection pointer in virConnectClose libvir: error : invalid connection pointer in virConnectClose Are there any practices I should be following to avoid this? Alternatively, if it is likely to be related to an issue in the Python bindings, is there something I could do to diagnose? Thanks!

2 1

[libvirt] ANNOUNCE: Perl binding Sys-Virt release 0.2.0
by Daniel P. Berrange 28 Mar '09

28 Mar '09

I've made a long overdue update to the Perl binding for libvirt, Sys-Virt. http://search.cpan.org/CPAN/authors/id/D/DA/DANBERR/Sys-Virt-0.2.0.tar.gz The major changes since last time - Fix network create API, and UUID lookups - Implement storage pool, storage vol, node device, security model, domain events and event loop APIs - Improve way constants are exposed to Perl layer - Fix horrible memory leak in methods returning a hash - Fix integer overflow in APIs using 64-bit ints (aka 'long long') - Minimum required libvirt C library for building is 0.6.1 There is comprehensive API documentation available online http://search.cpan.org/dist/Sys-Virt/ As well as a number of example scripts http://cpansearch.perl.org/src/DANBERR/Sys-Virt-0.2.0/examples/ The master source repository for the Perl binding is now at http://libvirt.org/hg/libvirt-perl/ Thanks to everyone who reported bugs, tested the snapshots and sent patches for this release Regards, Daniel -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

1 0

[libvirt] [PATCH 2/2] VirtualBox support to libvirt
by Pritesh Kothari 27 Mar '09

27 Mar '09

Hi All, I have attached a patch which when applied on the HEAD as of today would allow virtualbox support in libvirt. The patch works very well with the VirtualBox OSE version and the 2.2 Beta release. [PATCH 1/2] contains diff of files already in libvirt. [PATCH 2/2] contains new files needed for VirtualBox support. Regards, -pritesh

3 5

[libvirt] libvirt Centos4
by Gian Mario 27 Mar '09

27 Mar '09

Hi to all, i need to compile libvirt-3.3.0 to Centos 4.7. Searching in mailing list i have seen a tread that help me to solve few problems. Actually i face with a error concern glibc, when i run make in libvirt directory, I obtain: /usr/lib64/libgcrypt.so.11: undefined reference to `__read_chk(a)GLIBC_2.4' /usr/lib/gcc/x86_64-redhat-linux/3.4.6/../../../../lib64/libgnutls.so: undefined reference to `__stack_chk_fail(a)GLIBC_2.4' collect2: ld returned 1 exit status make[2]: *** [virsh] Error 1 make[2]: Leaving directory `/root/XEN/libvirt-0.3.3/src' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/root/XEN/libvirt-0.3.3' make: *** [all] Error 2 Any one have some experience on compile libvirt on centos4? there is any possibilities to build it without upgrade glibc? tnks to any help. Ciao e statevi bene. GMario

3 2

[libvirt] PATCH: Support memory ballooning for QEMU
by Daniel P. Berrange 27 Mar '09

27 Mar '09

This patch implements full support for memory ballooning in the QEMU driver. - Fix qemudBuildCommandLine() to set the initial boot time VM memory allocation based off the 'maxmem' config field. - Add call to 'qemudDomainSetMemoryBalloon' immediately at startup to make the guest balloon to initial requested allocation. - In the qemudDomainGetInfo() and qemudDomainDumpXML calls, query the monitor to find current balloon allocation and update config - Implement qemudDomainSetMemory() for running guests using the monitor balloon command In all of these scenarios, if the QEMU being used is too old to support the memory balloon device, the user will get a suitable error or it'll report the statically defined memory allocation from boot time. Daniel diff -r 4bfef84f4d6f src/qemu_conf.c --- a/src/qemu_conf.c Tue Mar 10 12:09:41 2009 +0000 +++ b/src/qemu_conf.c Tue Mar 10 16:16:59 2009 +0000 @@ -870,7 +870,11 @@ int qemudBuildCommandLine(virConnectPtr } \ } while (0) - snprintf(memory, sizeof(memory), "%lu", vm->def->memory/1024); + /* Set '-m MB' based on maxmem, because the lower 'memory' limit + * is set post-startup using the balloon driver. If balloon driver + * is not supported, then they're out of luck anyway + */ + snprintf(memory, sizeof(memory), "%lu", vm->def->maxmem/1024); snprintf(vcpus, sizeof(vcpus), "%lu", vm->def->vcpus); snprintf(domid, sizeof(domid), "%d", vm->def->id); pidfile = virFilePid(driver->stateDir, vm->def->name); diff -r 4bfef84f4d6f src/qemu_driver.c --- a/src/qemu_driver.c Tue Mar 10 12:09:41 2009 +0000 +++ b/src/qemu_driver.c Tue Mar 10 16:16:59 2009 +0000 @@ -122,6 +122,9 @@ static int qemudMonitorCommandExtra(cons const char *extra, const char *extraPrompt, char **reply); +static int qemudDomainSetMemoryBalloon(virConnectPtr conn, + virDomainObjPtr vm, + unsigned long newmem); static struct qemud_driver *qemu_driver = NULL; @@ -1480,6 +1483,7 @@ static int qemudStartVMDaemon(virConnect (qemudDetectVcpuPIDs(conn, vm) < 0) || (qemudInitCpus(conn, vm, migrateFrom) < 0) || (qemudInitPasswords(conn, driver, vm) < 0) || + (qemudDomainSetMemoryBalloon(conn, vm, vm->def->memory) < 0) || (qemudSaveDomainStatus(conn, qemu_driver, vm) < 0)) { qemudShutdownVMDaemon(conn, driver, vm); return -1; @@ -2409,6 +2413,98 @@ cleanup: return ret; } + +#define BALLOON_PREFIX "balloon: actual=" + +/* + * Returns: 0 if balloon not supported, +1 if balloon query worked + * or -1 on failure + */ +static int qemudDomainGetMemoryBalloon(virConnectPtr conn, + virDomainObjPtr vm, + unsigned long *currmem) { + char *reply = NULL; + int ret = -1; + char *offset; + + if (qemudMonitorCommand(vm, "info balloon", &reply) < 0) { + qemudReportError(conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + "%s", _("could not query memory balloon allocation")); + goto cleanup; + } + + /* If the command failed qemu prints: + * device not found, device is locked ... + * No message is printed on success it seems */ + DEBUG ("balloon reply: '%s'", reply); + if ((offset = strstr(reply, BALLOON_PREFIX)) != NULL) { + unsigned int memMB; + char *end; + offset += strlen(BALLOON_PREFIX); + if (virStrToLong_ui(offset, &end, 10, &memMB) < 0) { + qemudReportError(conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + "%s", _("could not parse memory balloon allocation")); + goto cleanup; + } + *currmem = memMB * 1024; + ret = 1; + } else { + /* We don't raise an error here, since its to be expected that + * many QEMU's don't support ballooning + */ + ret = 0; + } + +cleanup: + VIR_FREE(reply); + return ret; +} + +/* + * Returns: 0 if balloon not supported, +1 if balloon query worked + * or -1 on failure + */ +static int qemudDomainSetMemoryBalloon(virConnectPtr conn, + virDomainObjPtr vm, + unsigned long newmem) { + char *cmd; + char *reply = NULL; + int ret = -1; + + /* + * 'newmem' is in KB, QEMU monitor works in MB, and we all wish + * we just worked in bytes with unsigned long long everywhere. + */ + if (virAsprintf(&cmd, "balloon %lu", (newmem / 1024)) < 0) { + virReportOOMError(conn); + goto cleanup; + } + + if (qemudMonitorCommand(vm, cmd, &reply) < 0) { + qemudReportError(conn, dom, NULL, VIR_ERR_OPERATION_FAILED, + "%s", _("could not balloon memory allocation")); + VIR_FREE(cmd); + goto cleanup; + } + VIR_FREE(cmd); + + /* If the command failed qemu prints: + * device not found, device is locked ... + * No message is printed on success it seems */ + DEBUG ("balloon reply: %s", reply); + if (strstr(reply, "\nunknown command:")) { + /* Don't set error - it is expected memory balloon fails on many qemu */ + ret = 0; + } else { + ret = 1; + } + +cleanup: + VIR_FREE(reply); + return ret; +} + + static int qemudDomainSetMemory(virDomainPtr dom, unsigned long newmem) { struct qemud_driver *driver = dom->conn->privateData; virDomainObjPtr vm; @@ -2426,20 +2522,21 @@ static int qemudDomainSetMemory(virDomai goto cleanup; } + if (newmem > vm->def->maxmem) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_ARG, + "%s", _("cannot set memory higher than max memory")); + goto cleanup; + } + if (virDomainIsActive(vm)) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_NO_SUPPORT, - "%s", _("cannot set memory of an active domain")); - goto cleanup; - } - - if (newmem > vm->def->maxmem) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_ARG, - "%s", _("cannot set memory higher than max memory")); - goto cleanup; - } - - vm->def->memory = newmem; - ret = 0; + ret = qemudDomainSetMemoryBalloon(dom->conn, vm, newmem); + if (ret == 0) + qemudReportError(dom->conn, dom, NULL, VIR_ERR_NO_SUPPORT, + "%s", _("cannot set memory of an active domain")); + } else { + vm->def->memory = newmem; + ret = 0; + } cleanup: if (vm) @@ -2452,6 +2549,8 @@ static int qemudDomainGetInfo(virDomainP struct qemud_driver *driver = dom->conn->privateData; virDomainObjPtr vm; int ret = -1; + int err; + unsigned long balloon; qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); @@ -2473,8 +2572,15 @@ static int qemudDomainGetInfo(virDomainP } } + err = qemudDomainGetMemoryBalloon(dom->conn, vm, &balloon); + if (err < 0) + goto cleanup; + info->maxMem = vm->def->maxmem; - info->memory = vm->def->memory; + if (err == 0) /* Balloon not supported */ + info->memory = vm->def->memory; + else + info->memory = balloon; info->nrVirtCpu = vm->def->vcpus; ret = 0; @@ -3169,16 +3275,25 @@ static char *qemudDomainDumpXML(virDomai struct qemud_driver *driver = dom->conn->privateData; virDomainObjPtr vm; char *ret = NULL; - - qemuDriverLock(driver); - vm = virDomainFindByUUID(&driver->domains, dom->uuid); - qemuDriverUnlock(driver); - - if (!vm) { - qemudReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_DOMAIN, - "%s", _("no domain with matching uuid")); - goto cleanup; - } + unsigned long balloon; + int err; + + qemuDriverLock(driver); + vm = virDomainFindByUUID(&driver->domains, dom->uuid); + qemuDriverUnlock(driver); + + if (!vm) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_INVALID_DOMAIN, + "%s", _("no domain with matching uuid")); + goto cleanup; + } + + /* Refresh current memory based on balloon info */ + err = qemudDomainGetMemoryBalloon(dom->conn, vm, &balloon); + if (err < 0) + goto cleanup; + if (err > 0) + vm->def->memory = balloon; ret = virDomainDefFormat(dom->conn, (flags & VIR_DOMAIN_XML_INACTIVE) && vm->newDef ? -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 4

[libvirt] Libvirt - Problems in different Versions
by Artur Baruchi 27 Mar '09

27 Mar '09

Hi Guys. Im writing a program that connects to Xen and gets some information (scheduller and memory informations). When executing the program using the libvirt 0.3.3 its working fine, BUT when running using libvirt 0.4.4 it doesnt work. I dont get any error message when compiling or executing, but the informations doesnt come. -> Using libvirt 0.3.3 # gcc -Wall -lxenstore `pkg-config --cflags --libs libvirt` test.c -o test # ./test DOMAINS IDENTIFIED 2 DomID: 0 Max Mem: 4294967292 Sched: 256 DomID: 3 Max Mem: 262144 Sched: 256 # xm list Name ID Mem VCPUs State Time(s) Domain-0 0 616 2 r----- 4751.9 vm01 3 256 2 -b---- 379.7 -> Using libvirt 0.4.4 # gcc -Wall -lxenstore `pkg-config --cflags --libs libvirt` test.c -o test # ./test DOMAINS IDENTIFIED 0 # xm list Name ID Mem VCPUs State Time(s) Domain-0 0 481 2 r----- 89.8 linuxprv01 1 512 1 -b---- 10.8 I executed the same program in other server, with 0.4.4 version, and got the same problem. Is there any difference between those versions? Thanks in advance. Att. Artur Baruchi

2 1

[libvirt] Using libvirt from PHP
by Radek Hladik 26 Mar '09

26 Mar '09

Hi, I am developing wimple web application in PHP to monitor and control VMs using libvirt. I hope it could become simple library for doing basic things with VMs. And I would like to ask for your opinion about different possibilities of calling libvirt. Option 1) I am calling virsh via exec command now (one virsh for every libvirt command). This can be used for local libvirt but it is very inefficient. The overhead for remote libvirt would be unbearable. I could run one virsh per page and use it "interactively" so the overhead gets lower. Option 2) The other option is to create Zend extension which should be able to call libvirt directly from C. However I've never done anything like that so I do not know, whether it is reasonable or not. Option 3) I could also create some "wrapper daemon" in C or other language with libvirt binding. But I think that this is an ugly way :-) Option 4) I could "talk" directly to the libvirt socket. But I am not sure how the communication goes there and whether it is stable or changes with every version. And I consider the socket to be internal thing of libvirt... Is there any other option I missed? And which one would you suggest? Radek

5 5

[libvirt] Monitoring VMs via java application
by Daniel Halle 26 Mar '09

26 Mar '09

Hello all, I want to monitor my VMs with a Java application. Is there a possibility to register an observer or the like which is triggered e.g. after the state of a VM has changed? Or is the only way to use a timer which polls all VMs every xy minutes? Thanks, Daniel -- Infinis GmbH - Taläckerstr. 30 - D-70437 Stuttgart Geschäftsführer: Daniel Schwager, Stefan Hörz - HRB Stuttgart 20964 Tel: 0711-849910-23 - Fax: 0711-849910-923 - daniel.halle(a)infinis.de

2 1

[libvirt] Restoring from a largish 'virsh save' file
by Matt McCowan 26 Mar '09

26 Mar '09

Running into an issue where, if I/O is hampered by load for example, reading a largish state file (created by 'virsh save') is not allowed to complete. qemudStartVMDaemon in src/qemu_driver.c has a loop that waits 10 seconds for the VM to be brought up. An strace against libvirt when doing a 'virsh restore' against a largish state file shows the VM being sent a kill when it's still happily reading from the file. inotify is used in the xen and uml drivers, so I thought it would be a suitable mechanism to delay the timeout loop if the state file was still being read. Is this the right way to solve this problem? I've inlined some (flimsy) mods for comment Regards Matt McCowan --- libvirt.orig/src/qemu_driver.c 2009-03-20 11:17:57.000000000 +0900 +++ libvirt-0.6.1.1/src/qemu_driver.c 2009-03-22 18:36:06.000000000 +0900 @@ -43,6 +43,7 @@ #include <stdio.h> #include <sys/wait.h> #include <sys/ioctl.h> +#include <sys/inotify.h> #if HAVE_NUMACTL #define NUMA_VERSION1_COMPATIBILITY 1 @@ -1310,6 +1309,11 @@ pid_t child; int pos = -1; char ebuf[1024]; + char rpathname[PATH_MAX]; + int fd = 0; + int wd = 0; + int got = 0; + char buf[1024]; struct gemudHookData hookData; hookData.conn = conn; @@ -1430,6 +1434,19 @@ qemudSecurityHook, &hookData); /* wait for qemu process to to show up */ + + /* Times out if using a restore file and that file + is big. Use inotify */ + if (stdin_fd > 0) { + fd = inotify_init(); + + if (snprintf(rpathname, sizeof(rpathname), "/proc/self/fd/%d", + stdin_fd) >= (int)sizeof(rpathname)) { + return -1; + } + wd = inotify_add_watch(fd, rpathname, IN_ACCESS); + } + if (ret == 0) { int retries = 100; int childstat; @@ -1442,7 +1459,15 @@ if ((ret = virFileReadPid(driver->stateDir, vm->def->name, &vm->pid)) == 0) break; usleep(100*1000); - retries--; + + /* Is state file being accessed? */ + got = read(fd, buf, sizeof(buf)); + if (got < 1) + retries--; + } + if (stdin_fd > 0) { + inotify_rm_watch(fd, wd); + close(fd); } if (ret) { qemudReportError(conn, NULL, NULL, VIR_ERR_INTERNAL_ERROR,

2 3

Re: [libvirt] Using libvirt from PHP
by Russell Haering 25 Mar '09

25 Mar '09

* Sorry, I somehow sent this to the wrong address * Radek, Not to advertise blatantly, but how attached are you to PHP? I'm currently working on a Django (python) WebApp that (although its *very* pre-release) could probably be adapted to what you're doing. If you are interested in assisting in development you'd be welcome to, or if you just wanted to fork off the code into your own thing you could do that too. My system consists of a python daemon used for actual libvirt interaction and a separate django web interface that interacts with the daemon via AMF over https. If you want to stick with what you have now, you could try doing something similar with a python daemon (you really need something like that for the sake of efficiency anyway, unless you can come up with a better way of maintaining libvirt connections from a stateless webapp) that your PHP frontend connects to. Or you could just steal my daemon code, although at the moment all it really does is monitoring and (as of yesterday) live migrations. What I have so far: http://trac.osuosl.org/trac/virtadmin -- Russell Haering Systems Administrator OSU Open Source Lab Radek Hladik wrote: > > Hi, > > I am developing wimple web application in PHP to monitor and control > > VMs using libvirt. I hope it could become simple library for doing basic > > things with VMs. And I would like to ask for your opinion about > > different possibilities of calling libvirt. > > > > Option 1) I am calling virsh via exec command now (one virsh for every > > libvirt command). This can be used for local libvirt but it is very > > inefficient. The overhead for remote libvirt would be unbearable. I > > could run one virsh per page and use it "interactively" so the overhead > > gets lower. > > Option 2) The other option is to create Zend extension which should be > > able to call libvirt directly from C. However I've never done anything > > like that so I do not know, whether it is reasonable or not. > > Option 3) I could also create some "wrapper daemon" in C or other > > language with libvirt binding. But I think that this is an ugly way :-) > > Option 4) I could "talk" directly to the libvirt socket. But I am not > > sure how the communication goes there and whether it is stable or > > changes with every version. And I consider the socket to be internal > > thing of libvirt... > > > > Is there any other option I missed? And which one would you suggest? > > > > Radek > > > > -- > > Libvir-list mailing list > > Libvir-list(a)redhat.com > > https://www.redhat.com/mailman/listinfo/libvir-list

1 0

[libvirt] Need help for thesis on Xen - disabling HAP for benchmarking purposes
by Alexander Sascha 25 Mar '09

25 Mar '09

Hi, I'm a German student, writing my thesis on "Virtualization with Xen. Analysis and Comparision of Different Techniques such as Paravirtualization, Full Virtualization, and Utilization of Hardware Support Provided by the Processor". I'd like to know how I can disable HAP/RVI/Nested Paging for benchmarking purposes. I read somewhere (http://markmail.org/message/bbnivuqx6vjz7jg4) that Xen developers decided to disable the global grub parameter introducing a per domain flag for HAP instead (hap=0/1). I used virt-manager built in CentOS for setting up all domUs; virt-manager uses libvirt so that no "old-style" configuration files are stored in /etc/xen. I actually don't know how if libvirt-XML-files already support something like <hap>0</hap>. How would you proceed to modify an already existing domain using libvirt/virt-manager? I really do apologize for my "newbie ignorance". Thanks in advance for any useful help. - Alex

2 1

[libvirt] [RFC]: Volume allocation progress reporting
by Cole Robinson 25 Mar '09

25 Mar '09

The attached patch implements storage volume allocation progress reporting for file volumes. Currently, the volume creation process looks like: - Grab the pool lock - Fully allocated the volume - 'define' the volume (so it shows up in 'virsh vol-list', etc) - Lookup the volume object to return - Drop the pool lock The new sequence is: - Grab the pool lock - 'define' the volume (even though nothing is on disk yet) - Drop the pool lock - Allocate the volume as needed - Regrab the pool lock - Lookup the volume object to return - Drop the pool lock (again) Since the volume is 'defined', the user can fetch an object reference in a separate thread, and continually poll the 'info' command for up to date numbers. This also has the benefit of dropping the pool lock during the potentially lengthy allocation, as currently 'virsh pool-list' etc. will block while any volume is being allocated. Non file volumes maintain existing behavior. I tried to make the implementation resistant to user error: say if the pool is deactivated or deleted while the volume is being allocated. The creation process may bail out, but I couldn't produce any bad errors (crashing). There are a few other small fixes in this patch: - Refresh volume info when doing volume dumpxml - Update volume capacity when doing a refresh I've also attached an ugly python script that can test this. Presuming you have a pool named 'default', running python sparse.py --vol-create-info Will launch an allocation, print vol.info in a loop. Feedback appreciated. Thanks, Cole import libvirt import threading import time import sys import optparse poolname = "default" volname = "nonsparsetest" testvol = "testvol" uri = "qemu:///system" testvolxml = """ <volume> <name>%s</name> <capacity>1048576000</capacity> <allocation>0</allocation> <target> <format type='raw'/> </target> </volume> """ % testvol volxml = """ <volume> <name>%s</name> <capacity>1048576000</capacity> <allocation>800000000</allocation> <target> <format type='raw'/> </target> </volume> """ % volname failvol = """ <volume> <name>%s</name> <capacity>1048576000</capacity> <allocation>1048576000</allocation> <target> <format type='bochs'/> </target> </volume> """ % volname poolxml = """ <pool type='dir'> <name>default</name> <target> <path>/var/lib/libvirt/images</path> </target> </pool> """ # Helper functions def exception_wrapper(cmd, args): try: cmd(*args) except Exception, e: print str(e) def make_test_vol(): pool = get_pool() pool.createXML(testvolxml, 0) def del_vol(name=volname): pool = get_pool() vol = pool.storageVolLookupByName(name) vol.delete(0) def del_pool(): pool = get_pool() pool.destroy() pool.undefine() def define_pool(): conn = libvirt.open(uri) pool = conn.storagePoolDefineXML(poolxml, 0) pool.create(0) pool.setAutostart(True) def allocate_thread(xml=volxml): try: del_vol() except: pass pool = get_pool() print "creating vol in thread" vol = pool.createXML(xml, 0) print "creating vol complete." def info_thread(vol): for i in range(0, 40): time.sleep(.5) print vol.info() def get_pool(name=poolname): conn = libvirt.open(uri) return conn.storagePoolLookupByName(name) def get_vol(name=volname): pool = get_pool() return pool.storageVolLookupByName(name) def cmd_vol_create(xml=volxml): exception_wrapper(define_pool, ()) pool = get_pool() print pool.listVolumes() t = threading.Thread(target=allocate_thread, name="Allocating", args=(xml,)) t.start() time.sleep(5) print "\nRefreshing pool and dumping list" pool.refresh(0) print pool.listVolumes() def cmd_vol_fail(): cmd_vol_create(failvol) def cmd_vol_poll(): cmd_vol_create() vol = get_vol() t = threading.Thread(target=info_thread, name="Getting info", args=(vol,)) t.start() def main(): import optparse parser = optparse.OptionParser() parser.add_option("", "--vol-create", action="store_true", dest="vol_create", help="Create a volume nonsparse volume that should " "succeed") parser.add_option("", "--vol-create-info", action="store_true", dest="vol_create_info", help="Create a volume nonsparse volume that should " "succeed, and list info in a loop") parser.add_option("", "--vol-create-fail", action="store_true", dest="vol_fail", help="Create a volume that will fail at the allocate " "stage") options, ignore = parser.parse_args() if options.vol_create: cmd_vol_create() elif options.vol_create_info: cmd_vol_poll() elif options.vol_fail: cmd_vol_fail() else: parser.print_help() sys.exit(1) if __name__ == "__main__": main()

2 2

[libvirt] first cut public API for physical host interface configuration
by Laine Stump 25 Mar '09

25 Mar '09

To get started integrating libnetcf support into libvirt, the attached libvirt.h diff has a first attempt at the public API that will hook up to libnetcf on the libvirtd side. I started out with the virNetwork* API, and modified/removed as seemed appropriate. A few points worth mentioning: virNetwork has "defined" and "active" interfaces, but so far virInterface just has interfaces, as the active/inactive status is really controlled by 1) the "onboot" property in the XML, and 2) whether or not virInterfaceStart() has been called yet. The _virInterface struct that is referenced here is more or less identical to _virnetwork. libnetcf works with netcf structs (one per library instance) and netcf_if structs (one per interface, multiple per library instance. As far as I can see right now, the netcf struct will not need to be referenced directly from the client side, but the netcf_if for each interface is needed, and I guess a cookie representing it will be sent to the client side and stored in the _virInterface. Is that what the UUID in _virNetwork is used for? (I know, I know - "read the code!", it's just quicker to (and avoids misreading on my part) to ask) As before, any and all advice/corrections gratefully accepted! diff --git a/include/libvirt/libvirt.h b/include/libvirt/libvirt.h index 779ea72..cac400e 100644 --- a/include/libvirt/libvirt.h +++ b/include/libvirt/libvirt.h @@ -854,7 +854,73 @@ int virNetworkGetAutostart (virNetworkPtr network, int virNetworkSetAutostart (virNetworkPtr network, int autostart); +/* + * Physical host interface configuration API + */ + +/** + * virInterface: + * + * a virInterface is a private structure representing a virtual interface. + */ +typedef struct _virInterface virInterface; + +/** + * virInterfacePtr: + * + * a virInterfacePtr is pointer to a virInterface private structure, this is the + * type used to reference a virtual interface in the API. + */ +typedef virInterface *virInterfacePtr; + +/* + * Get connection from interface. + */ +virConnectPtr virInterfaceGetConnect (virInterfacePtr interface); + +/* + * List defined interfaces + */ +int virConnectNumOfInterfaces (virConnectPtr conn); +int virConnectListInterfaces (virConnectPtr conn, + char **const names, + int maxnames); + +/* + * Lookup interface by name + */ +virInterfacePtr virInterfaceLookupByName (virConnectPtr conn, + const char *name); + const char *uuid); + +/* + * Define interface (or modify existing interface configuration) + */ +virInterfacePtr virInterfaceDefineXML (virConnectPtr conn, + const char *xmlDesc); + +/* + * Delete interface + */ +int virInterfaceUndefine (virInterfacePtr interface); + +/* + * Activate interface (ie call "ifup") + */ +int virInterfaceStart (virInterfacePtr interface); + +/* + * De-activate interface (call "ifdown") + */ +int virInterfaceStop (virInterfacePtr interface); + +/* + * Interface information + */ +const char* virInterfaceGetName (virInterfacePtr interface); +char * virInterfaceGetXMLDesc (virInterfacePtr interface, + int flags); /** * virStoragePool: *

1 0

[libvirt] virt-top
by Zvi Dubitzky 24 Mar '09

24 Mar '09

Are the sources of virt-top available at some repository to look at . I do not see them in the Download section of libvirt Web site thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

[libvirt] save/ restore a domain with kvm
by Zvi Dubitzky 24 Mar '09

24 Mar '09

I am working with virsh The version of VIRSH as it shows is : virsh # version Compiled against library: libvir 0.4.4 Using library: libvir 0.4.4 Using API: QEMU 0.4.4 Running hypervisor: QEMU 0.9.1 Under virsh I do the following : 1. suspend a running VM 2. save it to a file with the 'save' command 3. restore the saved domain file with the 'restore' command while the domain is still in the suspend mode or after it was shutdown/destroyed (not seen with 'list --all) In either case I get the error message saying : virsh # restore /home/dubi/xml/vm_dubi2.sav libvir: QEMU error : operation failed: failed to start VM error: Failed to restore domain from /home/dubi/xml/vm_dubi2.xml Any idea what is the failure reason ? The content of the saved domain file isattached: Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

3 5

[libvirt] [PATCH 1/2] VirtualBox support to libvirt
by Pritesh Kothari 24 Mar '09

24 Mar '09

Hi All, I have attached a patch which when applied on the HEAD as of today would allow virtualbox support in libvirt. The patch works very well with the VirtualBox OSE version and the 2.2 Beta release. [PATCH 1/2] contains diff of files already in libvirt. [PATCH 2/2] contains new files needed for VirtualBox support. Regards, -pritesh

3 6

[libvirt] Improve heuristic for default guest architecture
by Soren Hansen 24 Mar '09

24 Mar '09

In libvirt 0.6.1, if you create a domain description of type 'kvm' without an arch set on an x86-64 host, you would get an i686 qemu guest rather than the expected x86-64 kvm guest. This is because virCapabilitiesDefaultGuestArch doesn't take the domain type into consideration, so it just returned the first hvm architecutre that has been registered, which is i686. After applying Dan P's patch, http://www.redhat.com/archives/libvir-list/2009-March/msg00281.html, I now get a i686 kvm guest, since kvm now can do i686 guests from libvirt. This is certainly an improvement, but I think a more reasonable default is to attempt to match the host's architecture. This patch makes virCapabilitiesDefaultGuestArch also check the domain type, and also gives preference to a guest architecture that matches the host's architecture. Index: libvirt-0.6.1/src/capabilities.c =================================================================== --- libvirt-0.6.1.orig/src/capabilities.c 2009-03-19 15:18:09.483317579 +0100 +++ libvirt-0.6.1/src/capabilities.c 2009-03-19 15:42:31.027341187 +0100 @@ -468,14 +468,26 @@ */ extern const char * virCapabilitiesDefaultGuestArch(virCapsPtr caps, - const char *ostype) + const char *ostype, + const char *domain) { - int i; + int i, j; + const char *arch = NULL; for (i = 0 ; i < caps->nguests ; i++) { - if (STREQ(caps->guests[i]->ostype, ostype)) - return caps->guests[i]->arch.name; + if (STREQ(caps->guests[i]->ostype, ostype)) { + for (j = 0 ; j < caps->guests[i]->arch.ndomains ; j++) { + if (STREQ(caps->guests[i]->arch.domains[j]->type, domain)) { + /* Use the first match... */ + if (!arch) + arch = caps->guests[i]->arch.name; + /* ...unless we can match the host's architecture. */ + if (STREQ(caps->guests[i]->arch.name, caps->host.arch)) + return caps->guests[i]->arch.name; + } + } + } } - return NULL; + return arch; } /** Index: libvirt-0.6.1/src/capabilities.h =================================================================== --- libvirt-0.6.1.orig/src/capabilities.h 2009-03-19 15:18:09.507338228 +0100 +++ libvirt-0.6.1/src/capabilities.h 2009-03-19 15:42:31.027341187 +0100 @@ -177,7 +177,8 @@ extern const char * virCapabilitiesDefaultGuestArch(virCapsPtr caps, - const char *ostype); + const char *ostype, + const char *domain); extern const char * virCapabilitiesDefaultGuestMachine(virCapsPtr caps, const char *ostype, Index: libvirt-0.6.1/src/domain_conf.c =================================================================== --- libvirt-0.6.1.orig/src/domain_conf.c 2009-03-19 15:18:09.531341976 +0100 +++ libvirt-0.6.1/src/domain_conf.c 2009-03-19 15:42:31.031345327 +0100 @@ -2146,7 +2146,7 @@ goto error; } } else { - const char *defaultArch = virCapabilitiesDefaultGuestArch(caps, def->os.type); + const char *defaultArch = virCapabilitiesDefaultGuestArch(caps, def->os.type, virDomainVirtTypeToString(def->virtType)); if (defaultArch == NULL) { virDomainReportError(conn, VIR_ERR_INTERNAL_ERROR, _("no supported architecture for os type '%s'"), Index: libvirt-0.6.1/src/xm_internal.c =================================================================== --- libvirt-0.6.1.orig/src/xm_internal.c 2009-03-19 15:18:09.559316828 +0100 +++ libvirt-0.6.1/src/xm_internal.c 2009-03-19 15:42:45.807318313 +0100 @@ -695,7 +695,7 @@ if (!(def->os.type = strdup(hvm ? "hvm" : "xen"))) goto no_memory; - defaultArch = virCapabilitiesDefaultGuestArch(priv->caps, def->os.type); + defaultArch = virCapabilitiesDefaultGuestArch(priv->caps, def->os.type, virDomainVirtTypeToString(def->virtType)); if (defaultArch == NULL) { xenXMError(conn, VIR_ERR_INTERNAL_ERROR, _("no supported architecture for os type '%s'"), -- Soren Hansen | Lead Virtualisation Engineer | Ubuntu Server Team Canonical Ltd. | http://www.ubuntu.com/

3 3

[libvirt] [PATCH] Adding filesystem mount support for openVZ
by Florian Vichot 24 Mar '09

24 Mar '09

Hi everyone, This patch is to allow using the "mount" type in the "filesystem" tag for OpenVZ domains. Example: ... <filesystem type='mount'> <source dir='/path/to/filesystem/directory/' /> <target dir='/path/to/pivot/root/' /> </filesystem> ... This is my first patch to an external project, so don't spare me if I got things wrong :) Also, I'm curious for suggestions as to how I could allow for the target not to be specified in the XML. Because in this case OpenVZ just makes a temporary pivot root in "/var/lib/vz/root/" and that is probably sufficient for most people, who might not want to have to explicitly create a pivot root somewhere, just for mounting the filesystem while it's running. I was thinking either allow for the target tag not to be specified, or add an "auto" attribute to target. Which one sounds better ? Thanks, Florian

3 8

[libvirt] VM cpuTime from libvirt
by Zvi Dubitzky 23 Mar '09

23 Mar '09

Currently the cat /proc/pid/stat where pid is the pid of the VM Qemu process gives the utime + stime of the VM according to libvirt. Unfortunately I notice that this is actually the elapsed time of the host . I find this by using libvirt , sampling the cputime of each VM process and compare it to the total elapsed time (of the Host Linux machine) . Roughly assuming full VM vcpu utlization, the cpu utilization of every VM is actually : ( # of VM vcpus / total # of Host cpus) .This implies that there is no idle time per VM (while actually there is) . We only know the idle time of the Host via the top command at the Host . Is the cputime of a VM (from cat /proc/pid/stat of the Qemu process as used by libvirt) realy the cputime of the VM ? I tested with a host having 2 sockets X 4cpus =8 totally and assigned 4 cpus to VM1 and VM2 . libvirt gave equal cputime for each VM which is equal to the total machine elapsed time. But even if VM1 has 4 vcpus and VM2 have 8 vcpus the cputime of each VM (from cat /proc/pid/stat) is the elapsed time. The truth is that I am running the libvirt application on the host machine and do the application wait there . Should that matter much? Each guest idle time is needed from the KVM for real cpu utilization calc to be independent from the guest OS . At least for Linux we can manually run 'top' at each guest terminal window , but I do not know if it will show the real idle time or the total machine (host) idle time .At least linux has no idle process . Besides this is not a good programmatic way to get the VM idle time. Is there a cure or I am missing something ? thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 3

[libvirt] [PATCH 1/2] [Plain text ] OpenNebula driver, libvirt-0.6.1
by "Abel Míguez Rodríguez" 23 Mar '09

23 Mar '09

Hi all, I'm sorry, I apologize for the format used. thanks, > Hi all, > We have updated the OpenNebula Driver to libvirt version 0.6.1. > Now, the ONE driver is build at the libvirtd daemon, that is the natural place for it. > Please feel free to make any comment, to improve the driver's coherence with libvirt's structure. > I split the Patches in two e-mais: > [PATCH 1/2] includes the patches to be applied to libvirt's sources and building files. > [PATCH 2/2] attached the "one driver" source files. >All the patches are made to be applied at the git commit: >"025b62" (Fix subsystem lookup for older HAL releases) >Thanks, >Abel Miguez diff --git a/configure.in b/configure.in index 413d27c..fd75a8d 100644 --- a/configure.in +++ b/configure.in @@ -184,6 +184,8 @@ AC_ARG_WITH([openvz], [ --with-openvz add OpenVZ support (on)],[],[with_openvz=yes]) AC_ARG_WITH([lxc], [ --with-lxc add Linux Container support (on)],[],[with_lxc=yes]) +AC_ARG_WITH([one], +[ --with-one add ONE support (on)],[],[with_one=no]) AC_ARG_WITH([test], [ --with-test add test driver support (on)],[],[with_test=yes]) AC_ARG_WITH([remote], @@ -399,6 +401,17 @@ dnl check for kvm headers dnl AC_CHECK_HEADERS([linux/kvm.h]) +dnl OpenNebula driver Compilation setting +dnl + +if test "$with_one" = "yes" ; then + LIBVIRT_FEATURES="$LIBVIRT_FEATURES -DWITH_ONE -I$ONE_LOCATION/include" + ONE_LIBS="-L/usr/local/lib -lxmlrpc_client++ -lxmlrpc -lxmlrpc_util -lxmlrpc_xmlparse -lxmlrpc_xmltok -lxmlrpc++ -lxmlrpc_client -L$ONE_LOCATION/lib -loneapi" + AC_SUBST([ONE_LIBS]) + AC_DEFINE_UNQUOTED([WITH_ONE],1,[whether Open Nebula Driver is enabled]) +fi +AM_CONDITIONAL([WITH_ONE],[test "$with_one" = "yes"]) + dnl Need to test if pkg-config exists PKG_PROG_PKG_CONFIG @@ -1345,6 +1358,7 @@ AC_MSG_NOTICE([ QEMU: $with_qemu]) AC_MSG_NOTICE([ UML: $with_uml]) AC_MSG_NOTICE([ OpenVZ: $with_openvz]) AC_MSG_NOTICE([ LXC: $with_lxc]) +AC_MSG_NOTICE([ ONE: $with_one]) AC_MSG_NOTICE([ Test: $with_test]) AC_MSG_NOTICE([ Remote: $with_remote]) AC_MSG_NOTICE([ Network: $with_network]) diff --git a/include/libvirt/virterror.h b/include/libvirt/virterror.h index 2c3777d..cc98e45 100644 --- a/include/libvirt/virterror.h +++ b/include/libvirt/virterror.h @@ -61,6 +61,7 @@ typedef enum { VIR_FROM_UML, /* Error at the UML driver */ VIR_FROM_NODEDEV, /* Error from node device monitor */ VIR_FROM_XEN_INOTIFY, /* Error from xen inotify layer */ + VIR_FROM_ONE, /* Error from ONE driver */ VIR_FROM_SECURITY, /* Error from security framework */ } virErrorDomain; diff --git a/qemud/Makefile.am b/qemud/Makefile.am index 924e8ad..9d7f61f 100644 --- a/qemud/Makefile.am +++ b/qemud/Makefile.am @@ -120,6 +120,10 @@ if WITH_UML libvirtd_LDADD += ../src/libvirt_driver_uml.la endif +if WITH_ONE + libvirtd_LDADD += ../src/libvirt_driver_one.la +endif + if WITH_STORAGE_DIR libvirtd_LDADD += ../src/libvirt_driver_storage.la endif diff --git a/qemud/qemud.c b/qemud/qemud.c index 4f04355..e1d6113 100644 --- a/qemud/qemud.c +++ b/qemud/qemud.c @@ -78,6 +78,9 @@ #ifdef WITH_NETWORK #include "network_driver.h" #endif +#ifdef WITH_ONE +#include "one_driver.h" +#endif #ifdef WITH_STORAGE_DIR #include "storage_driver.h" #endif @@ -841,6 +844,8 @@ static struct qemud_server *qemudInitialize(int sigread) { virDriverLoadModule("qemu"); virDriverLoadModule("lxc"); virDriverLoadModule("uml"); + virDriverLoadModule("one"); + #else #ifdef WITH_NETWORK networkRegister(); @@ -861,6 +866,10 @@ static struct qemud_server *qemudInitialize(int sigread) { #ifdef WITH_UML umlRegister(); #endif +#ifdef WITH_ONE + oneRegister (); +#endif + #endif virEventRegisterImpl(virEventAddHandleImpl, diff --git a/src/Makefile.am b/src/Makefile.am index d5aac11..a5c2084 100644 --- a/src/Makefile.am +++ b/src/Makefile.am @@ -137,6 +137,10 @@ UML_DRIVER_SOURCES = \ uml_conf.c uml_conf.h \ uml_driver.c uml_driver.h +ONE_DRIVER_SOURCES = \ + one_conf.c one_conf.h \ + one_driver.c one_driver.h + NETWORK_DRIVER_SOURCES = \ network_driver.h network_driver.c @@ -314,6 +318,22 @@ endif libvirt_driver_uml_la_SOURCES = $(UML_DRIVER_SOURCES) endif +if WITH_ONE +if WITH_DRIVER_MODULES +mod_LTLIBRARIES += libvirt_driver_one.la +else +noinst_LTLIBRARIES += libvirt_driver_one.la +# Stateful, so linked to daemon instead +#libvirt_la_LIBADD += libvirt_driver_one.la +endif +libvirt_driver_one_la_LDFLAGS = $(ONE_LIBS) +libvirt_driver_one_la_CFLAGS = "-DWITH_ONE" +if WITH_DRIVER_MODULES +libvirt_driver_one_la_LDFLAGS += -module -avoid-version +endif +libvirt_driver_one_la_SOURCES = $(ONE_DRIVER_SOURCES) +endif + if WITH_NETWORK if WITH_DRIVER_MODULES mod_LTLIBRARIES += libvirt_driver_network.la @@ -402,6 +422,7 @@ EXTRA_DIST += \ $(QEMU_DRIVER_SOURCES) \ $(LXC_DRIVER_SOURCES) \ $(UML_DRIVER_SOURCES) \ + $(ONE_DRIVER_SOURCES) \ $(OPENVZ_DRIVER_SOURCES) \ $(NETWORK_DRIVER_SOURCES) \ $(STORAGE_DRIVER_SOURCES) \ diff --git a/src/domain_conf.c b/src/domain_conf.c index 5bf3483..e4d3249 100644 --- a/src/domain_conf.c +++ b/src/domain_conf.c @@ -54,7 +54,9 @@ VIR_ENUM_IMPL(virDomainVirt, VIR_DOMAIN_VIRT_LAST, "ldom", "test", "vmware", - "hyperv") + "hyperv", + "one") + VIR_ENUM_IMPL(virDomainBoot, VIR_DOMAIN_BOOT_LAST, "fd", diff --git a/src/domain_conf.h b/src/domain_conf.h index dd61467..e8a2bff 100644 --- a/src/domain_conf.h +++ b/src/domain_conf.h @@ -48,6 +48,7 @@ enum virDomainVirtType { VIR_DOMAIN_VIRT_TEST, VIR_DOMAIN_VIRT_VMWARE, VIR_DOMAIN_VIRT_HYPERV, + VIR_DOMAIN_VIRT_ONE, VIR_DOMAIN_VIRT_LAST, }; diff --git a/src/driver.h b/src/driver.h index 62d6fbc..ed3eef7 100644 --- a/src/driver.h +++ b/src/driver.h @@ -20,6 +20,7 @@ typedef enum { VIR_DRV_OPENVZ = 5, VIR_DRV_LXC = 6, VIR_DRV_UML = 7, + VIR_DRV_ONE = 8, } virDrvNo; diff --git a/src/libvirt.c b/src/libvirt.c index bf3453a..cd4b5b7 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -830,6 +830,10 @@ virGetVersion(unsigned long *libVer, const char *type, if (STRCASEEQ(type, "OpenVZ")) *typeVer = LIBVIR_VERSION_NUMBER; #endif +#if WITH_ONE + if (STRCASEEQ(type, "ONE")) + *typeVer = LIBVIR_VERSION_NUMBER; +#endif #if WITH_UML if (STRCASEEQ(type, "UML")) *typeVer = LIBVIR_VERSION_NUMBER; ---- Distributed System Architecture Group (http://dsa-research.org) GridWay, http://www.gridway.org OpenNEbula, http://www.opennebula.org

2 2

[libvirt] Enable authentication for all virConnectOpen calls in virsh
by Matthias Bolte 23 Mar '09

23 Mar '09

Hello, virsh opens connections in three places (init, connect, migrate) but only connections initiated from the command line (the -c option) are opened with authentication. The attached patch enables the two other places to use authentication too. Ciao Matthias

2 1

[libvirt] libvirt authorization
by Scott Beardsley 23 Mar '09

23 Mar '09

Is there any authorization mechanism in libvirt? I've got TLS going so that only those with a cert signed by my CA are allowed in, but there appears to be no way for me to only allow them access to certain VMs. Can I limit folks to specific VMs or VNC ports? Ideally I can allow access only to those VMs which they own. This is really not _that_ big of a deal because I'm advising all my admins to log out of their systems but it would give me and them peace of mind. Thoughts? Scott

3 4

[libvirt] Problem with remote xend over ssh tunnel
by waheed iqbal 23 Mar '09

23 Mar '09

Hi, I am able to connect remote xend using http but whenever I try to connect remote machine using "virsh -c xen+ssh://root@...../ list" I found following error: libvir: Remote error : socket closed unexpectedly error: failed to connect to the hypervisor libvirtd is running on both systems and netcat is also installed on both machines. Debug log is: node2:/home/waheed# export LIBVIRT_DEBUG=1 node2:/home/waheed# virsh -c xen+ssh://root@node1/ list DEBUG: libvirt.c: virInitialize (register drivers) DEBUG: libvirt.c: virRegisterDriver (registering Test as driver 0) DEBUG: libvirt.c: virRegisterNetworkDriver (registering Test as network driver 0) DEBUG: libvirt.c: virRegisterStorageDriver (registering Test as storage driver 0) DEBUG: xen_internal.c: xenHypervisorInit (Using new hypervisor call: 30002 ) DEBUG: xen_internal.c: xenHypervisorInit (Using hypervisor call v2, sys ver6 dom ver5 ) DEBUG: libvirt.c: virRegisterDriver (registering Xen as driver 1) DEBUG: libvirt.c: virRegisterDriver (registering OPENVZ as driver 2) DEBUG: libvirt.c: virRegisterDriver (registering remote as driver 3) DEBUG: libvirt.c: virRegisterNetworkDriver (registering remote as network driver 1) DEBUG: libvirt.c: virRegisterStorageDriver (registering remote as storage driver 1) DEBUG: libvirt.c: virRegisterDeviceMonitor (registering remote as device driver 0) DEBUG: libvirt.c: virConnectOpenAuth (name=xen+ssh://root@node1/, auth=0xb7ed8be0, flags=0) DEBUG: libvirt.c: do_open (name "xen+ssh://root@node1/" to URI components: scheme xen+ssh opaque (null) authority (null) server node1 user root port 0 path / ) DEBUG: libvirt.c: do_open (trying driver 0 (Test) ...) DEBUG: libvirt.c: do_open (driver 0 Test returned DECLINED) DEBUG: libvirt.c: do_open (trying driver 1 (Xen) ...) DEBUG: libvirt.c: do_open (driver 1 Xen returned DECLINED) DEBUG: libvirt.c: do_open (trying driver 2 (OPENVZ) ...) DEBUG: libvirt.c: do_open (driver 2 OPENVZ returned DECLINED) DEBUG: libvirt.c: do_open (trying driver 3 (remote) ...) DEBUG: remote_internal.c: doRemoteOpen (proceeding with name = xen:///) DEBUG: util.c: virExec (ssh -p 22 -l root node1 nc -U /var/run/libvirt/libvirt-sock) root@node1's password: libvir: Remote error : socket closed unexpectedly DEBUG: libvirt.c: do_open (driver 3 remote returned ERROR) DEBUG: datatypes.c: virUnrefConnect (unref connection 0x9818de0 1) DEBUG: datatypes.c: virReleaseConnect (release connection 0x9818de0) error: failed to connect to the hypervisor Please help me to identify problem. Regards Waheed

2 1

[libvirt] Fix compile errors in 'safezero()' functions in util.c
by Maximilian Wilhelm 20 Mar '09

20 Mar '09

Hi! Fix compile errors in 'safezero()' functions in util.c Commit 70e7672184ff639856d5f2f3bf7849464031dff9 introduced some compile errors because of the unsued 'flags' parameter in all three version of the 'safezero()' function. Signed-Off-By: Maximilian Wilhelm <max(a)rfc2324.org> The attached patch fixes this. Ciao Max -- Follow the white penguin.

2 2

[libvirt] [RFC][PATCH]: Secure migration
by Chris Lalancette 20 Mar '09

20 Mar '09

All, Attached is a *very* rough first draft of the secure migration code I'm working on. This is in no way ready for merge. That being said, this demonstrates the basic idea that I'm pursuing, and I've actually been able to perform a KVM secure live migrate using this. Before I go and finish polishing it up, though, I wanted to make sure there wasn't anything fundamentally wrong with the approach. So, in that vein, comments are appreciated. Thanks, -- Chris Lalancette

3 4

[libvirt] PATCH: Be more flexible in emulator choice on x86 archs
by Daniel P. Berrange 20 Mar '09

20 Mar '09

Currently we are pretty strict about what emulator binary we allow for QEMU guests on x86 arches. In particular, for arch+domain type combos: - i686+qemu must use 'qemu' binary - x86_64+qemu must use 'qemu-system-x86_64' binary - kvm must use 'qemu-kvm' or 'kvm' binaries - i686+kvm on x86_64 host is not allowed These restrictions are overkill because - i686+qemu could use 'qemu-system-x86_64' if '-cpu qemu32' is added - i686+qemu could use 'qemu-kvm' if '-cpu qemu32 -no-kvm' is added - x86_64+qemu could use 'qemu-kvm' if '-no-kvm' is added - i686+kvm on x86_64 host can be used if '-cpu qemu32' is added This patch makes QEMU driver more flexible in this way when setting up its capabilities information. It also makes it aware of the -no-kvm and -cpu flag, using them where required by the os type + arch + emulator combinations specified in the guest XML. This should finally remove the confusion where a user in virt-manager selectrs 'i686' and then wonders why we've disallowed choice of 'kvm'. It also fixes 'virsh version' when only qemu-kvm is installed. The matrix should now work thus: 1. qemu, qemu-system-x86_64, qemu-kvm all available qemu+i686 => qemu qemu+x86_64 => qemu-system-x86_64 kvm+i686 => qemu-kvm -cpu qemu32 kvm+x86_64 => qemu-kvm 2. qemu, qemu-kvm available qemu+i686 => qemu qemu+x86_64 => qemu-kvm -no-kvm kvm+i686 => qemu-kvm -cpu qemu32 kvm+x86_64 => qemu-kvm 3. qemu-system-x86_64, qemu-kvm available qemu+i686 => qemu-system-x86_64 -cpu qemu32 qemu+x86_64 => qemu-system-x86_64 kvm+i686 => qemu-kvm -cpu qemu32 kvm+x86_64 => qemu-kvm 4. qemu-kvm available qemu+i686 => qemu-kvm -no-kvm -cpu qemu32 qemu+x86_64 => qemu-kvm -no-kvm kvm+i686 => qemu-kvm -cpu qemu32 kvm+x86_64 => qemu-kvm The only real remaining problem is that we don't cope with scenario where the KVM enabled binary is called 'qemu-system-x64_64' instead of 'qemu-kvm' or 'kvm'. Regards, Daniel Index: src/qemu_conf.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_conf.c,v retrieving revision 1.138 diff -u -p -r1.138 qemu_conf.c --- src/qemu_conf.c 16 Mar 2009 13:54:26 -0000 1.138 +++ src/qemu_conf.c 16 Mar 2009 17:53:20 -0000 @@ -211,6 +211,7 @@ struct qemu_arch_info { const char *const *machines; int nmachines; const char *binary; + const char *altbinary; const struct qemu_feature_flags *flags; int nflags; }; @@ -231,24 +232,24 @@ static const struct qemu_feature_flags c /* The archicture tables for supported QEMU archs */ static const struct qemu_arch_info const arch_info_hvm[] = { { "i686", 32, arch_info_hvm_x86_machines, 2, - "/usr/bin/qemu", arch_info_i686_flags, 4 }, + "/usr/bin/qemu", "/usr/bin/qemu-system-x86_64", arch_info_i686_flags, 4 }, { "x86_64", 64, arch_info_hvm_x86_machines, 2, - "/usr/bin/qemu-system-x86_64", arch_info_x86_64_flags, 2 }, + "/usr/bin/qemu-system-x86_64", NULL, arch_info_x86_64_flags, 2 }, { "mips", 32, arch_info_hvm_mips_machines, 1, - "/usr/bin/qemu-system-mips", NULL, 0 }, + "/usr/bin/qemu-system-mips", NULL, NULL, 0 }, { "mipsel", 32, arch_info_hvm_mips_machines, 1, - "/usr/bin/qemu-system-mipsel", NULL, 0 }, + "/usr/bin/qemu-system-mipsel", NULL, NULL, 0 }, { "sparc", 32, arch_info_hvm_sparc_machines, 1, - "/usr/bin/qemu-system-sparc", NULL, 0 }, + "/usr/bin/qemu-system-sparc", NULL, NULL, 0 }, { "ppc", 32, arch_info_hvm_ppc_machines, 3, - "/usr/bin/qemu-system-ppc", NULL, 0 }, + "/usr/bin/qemu-system-ppc", NULL, NULL, 0 }, }; static const struct qemu_arch_info const arch_info_xen[] = { { "i686", 32, arch_info_xen_x86_machines, 1, - "/usr/bin/xenner", arch_info_i686_flags, 4 }, + "/usr/bin/xenner", NULL, arch_info_i686_flags, 4 }, { "x86_64", 64, arch_info_xen_x86_machines, 1, - "/usr/bin/xenner", arch_info_x86_64_flags, 2 }, + "/usr/bin/xenner", NULL, arch_info_x86_64_flags, 2 }, }; static int @@ -257,43 +258,62 @@ qemudCapsInitGuest(virCapsPtr caps, const struct qemu_arch_info *info, int hvm) { virCapsGuestPtr guest; - int i, haskvm, hasbase, samearch; + int i; + int hasbase = 0; + int hasaltbase = 0; + int haskvm = 0; + int haskqemu = 0; const char *kvmbin = NULL; - /* Check for existance of base emulator */ + /* Check for existance of base emulator, or alternate base + * which can be used with magic cpu choice + */ hasbase = (access(info->binary, X_OK) == 0); + hasaltbase = (access(info->altbinary, X_OK) == 0); - samearch = STREQ(info->arch, hostmachine); - if (samearch) { + /* Can use acceleration for KVM/KQEMU if + * - host & guest arches match + * Or + * - hostarch is x86_64 and guest arch is i686 + * The latter simply needs "-cpu qemu32" + */ + if (STREQ(info->arch, hostmachine) || + (STREQ(hostmachine, "x86_64") && STREQ(info->arch, "i686"))) { const char *const kvmbins[] = { "/usr/bin/qemu-kvm", /* Fedora */ "/usr/bin/kvm" }; /* Upstream .spec */ for (i = 0; i < ARRAY_CARDINALITY(kvmbins); ++i) { - if ((haskvm = (access(kvmbins[i], X_OK) == 0))) { + if (access(kvmbins[i], X_OK) == 0 && + access("/dev/kvm", F_OK) == 0) { + haskvm = 1; kvmbin = kvmbins[i]; break; } } - } else { - haskvm = 0; + + if (access("/dev/kqemu", F_OK) == 0) + haskqemu = 1; } - if (!hasbase && !haskvm) + + if (!hasbase && !hasaltbase && !haskvm) return 0; + /* We register kvm as the base emulator too, since we can + * just give -no-kvm to disable acceleration if required */ if ((guest = virCapabilitiesAddGuest(caps, hvm ? "hvm" : "xen", info->arch, info->wordsize, - info->binary, + (hasbase ? info->binary : + (hasaltbase ? info->altbinary : kvmbin)), NULL, info->nmachines, info->machines)) == NULL) return -1; if (hvm) { - if (hasbase && - virCapabilitiesAddGuestDomain(guest, + if (virCapabilitiesAddGuestDomain(guest, "qemu", NULL, NULL, @@ -301,27 +321,23 @@ qemudCapsInitGuest(virCapsPtr caps, NULL) == NULL) return -1; - /* If guest & host match, then we can accelerate */ - if (samearch) { - if (access("/dev/kqemu", F_OK) == 0 && - virCapabilitiesAddGuestDomain(guest, - "kqemu", - NULL, - NULL, - 0, - NULL) == NULL) - return -1; + if (haskqemu && + virCapabilitiesAddGuestDomain(guest, + "kqemu", + NULL, + NULL, + 0, + NULL) == NULL) + return -1; - if (access("/dev/kvm", F_OK) == 0 && - haskvm && - virCapabilitiesAddGuestDomain(guest, - "kvm", - kvmbin, - NULL, - 0, - NULL) == NULL) - return -1; - } + if (haskvm && + virCapabilitiesAddGuestDomain(guest, + "kvm", + kvmbin, + NULL, + 0, + NULL) == NULL) + return -1; } else { if (virCapabilitiesAddGuestDomain(guest, "kvm", @@ -363,12 +379,14 @@ virCapsPtr qemudCapsInit(void) { if (virCapsInitNUMA(caps) < 0) goto no_memory; + /* First the pure HVM guests */ for (i = 0 ; i < ARRAY_CARDINALITY(arch_info_hvm) ; i++) if (qemudCapsInitGuest(caps, utsname.machine, &arch_info_hvm[i], 1) < 0) goto no_memory; + /* Then possibly the Xen paravirt guests (ie Xenner */ if (access("/usr/bin/xenner", X_OK) == 0 && access("/dev/kvm", F_OK) == 0) { for (i = 0 ; i < ARRAY_CARDINALITY(arch_info_xen) ; i++) @@ -430,6 +448,8 @@ int qemudExtractVersionInfo(const char * if (strstr(help, "-no-kqemu")) flags |= QEMUD_CMD_FLAG_KQEMU; + if (strstr(help, "-no-kvm")) + flags |= QEMUD_CMD_FLAG_KVM; if (strstr(help, "-no-reboot")) flags |= QEMUD_CMD_FLAG_NO_REBOOT; if (strstr(help, "-name")) @@ -749,6 +769,7 @@ int qemudBuildCommandLine(virConnectPtr char boot[VIR_DOMAIN_BOOT_LAST]; struct utsname ut; int disableKQEMU = 0; + int disableKVM = 0; int qargc = 0, qarga = 0; const char **qargv = NULL; int qenvc = 0, qenva = 0; @@ -757,6 +778,7 @@ int qemudBuildCommandLine(virConnectPtr char uuid[VIR_UUID_STRING_BUFLEN]; char domid[50]; char *pidfile; + const char *cpu = NULL; uname_normalize(&ut); @@ -789,9 +811,16 @@ int qemudBuildCommandLine(virConnectPtr } } + emulator = vm->def->emulator; + if (!emulator) + emulator = virDomainDefDefaultEmulator(conn, vm->def, driver->caps); + if (!emulator) + return -1; + + /* Need to explicitly disable KQEMU if * 1. Arch matches host arch - * 2. Guest is 'qemu' + * 2. Guest domain is 'qemu' * 3. The qemu binary has the -no-kqemu flag */ if ((qemuCmdFlags & QEMUD_CMD_FLAG_KQEMU) && @@ -799,6 +828,34 @@ int qemudBuildCommandLine(virConnectPtr vm->def->virtType == VIR_DOMAIN_VIRT_QEMU) disableKQEMU = 1; + /* Need to explicitly disable KVM if + * 1. Arch matches host arch + * 2. Guest domain is 'qemu' + * 3. The qemu binary has the -no-kvm flag + */ + if ((qemuCmdFlags & QEMUD_CMD_FLAG_KVM) && + STREQ(ut.machine, vm->def->os.arch) && + vm->def->virtType == VIR_DOMAIN_VIRT_QEMU) + disableKVM = 1; + + /* + * Need to force a 32-bit guest CPU type if + * + * 1. guest OS is i686 + * 2. host OS is x86_64 + * 3. emulator is qemu-kvm or kvm + * + * Or + * + * 1. guest OS is i686 + * 2. emulator is qemu-system-x86_64 + */ + if (STREQ(vm->def->os.arch, "i686") && + ((STREQ(ut.machine, "x86_64") && + strstr(emulator, "kvm")) || + strstr(emulator, "x86_64"))) + cpu = "qemu32"; + #define ADD_ARG_SPACE \ do { \ if (qargc == qarga) { \ @@ -887,12 +944,6 @@ int qemudBuildCommandLine(virConnectPtr ADD_ENV_COPY("LOGNAME"); ADD_ENV_COPY("TMPDIR"); - emulator = vm->def->emulator; - if (!emulator) - emulator = virDomainDefDefaultEmulator(conn, vm->def, driver->caps); - if (!emulator) - return -1; - ADD_ARG_LIT(emulator); ADD_ARG_LIT("-S"); @@ -904,9 +955,15 @@ int qemudBuildCommandLine(virConnectPtr ADD_ARG_LIT("-M"); ADD_ARG_LIT(vm->def->os.machine); } + if (cpu) { + ADD_ARG_LIT("-cpu"); + ADD_ARG_LIT(cpu); + } if (disableKQEMU) ADD_ARG_LIT("-no-kqemu"); + if (disableKVM) + ADD_ARG_LIT("-no-kvm"); ADD_ARG_LIT("-m"); ADD_ARG_LIT(memory); ADD_ARG_LIT("-smp"); Index: src/qemu_conf.h =================================================================== RCS file: /data/cvs/libvirt/src/qemu_conf.h,v retrieving revision 1.58 diff -u -p -r1.58 qemu_conf.h --- src/qemu_conf.h 16 Mar 2009 13:54:26 -0000 1.58 +++ src/qemu_conf.h 16 Mar 2009 17:53:20 -0000 @@ -55,6 +55,7 @@ enum qemud_cmd_flags { QEMUD_CMD_FLAG_MIGRATE_QEMU_TCP = (1 << 10), /* New migration syntax after merge to QEMU with TCP transport */ QEMUD_CMD_FLAG_MIGRATE_QEMU_EXEC = (1 << 11), /* New migration syntax after merge to QEMU with EXEC transport */ QEMUD_CMD_FLAG_DRIVE_CACHE_V2 = (1 << 12), /* Is the cache= flag wanting new v2 values */ + QEMUD_CMD_FLAG_KVM = (1 << 13), /* Whether KVM is compiled in */ }; /* Main driver state */ Index: src/qemu_driver.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_driver.c,v retrieving revision 1.218 diff -u -p -r1.218 qemu_driver.c --- src/qemu_driver.c 16 Mar 2009 17:16:04 -0000 1.218 +++ src/qemu_driver.c 16 Mar 2009 17:53:20 -0000 @@ -1733,6 +1733,7 @@ qemudMonitorCommand(const virDomainObjPt static int qemudProbe(void) { if ((virFileExists("/usr/bin/qemu")) || + (virFileExists("/usr/bin/qemu-system-x86_64")) || (virFileExists("/usr/bin/qemu-kvm")) || (virFileExists("/usr/bin/kvm")) || (virFileExists("/usr/bin/xenner"))) @@ -1749,10 +1750,10 @@ static virDrvOpenStatus qemudOpen(virCon if (qemu_driver == NULL) goto decline; - if (!qemudProbe()) - goto decline; - if (conn->uri == NULL) { + if (!qemudProbe()) + goto decline; + conn->uri = xmlParseURI(uid ? "qemu:///session" : "qemu:///system"); if (!conn->uri) { virReportOOMError(conn); -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

3 5

[libvirt] v3: Use posix_fallocate() on supported systems to allocate diskspace
by Amit Shah 20 Mar '09

20 Mar '09

This patchset makes use of the posix_fallocate() call to allocate chunks of files whenever needed if it's available. We fallback to using safewrite() if it's not available. mmap() could be used instead of safewrite too; I have a patch in case someone is interested in seeing it.

3 8

[libvirt] TLS for libvirt remote access
by Zvi Dubitzky 20 Mar '09

20 Mar '09

After generating TLS certificates for Libvirt remote access , I test the certificates by running pki_check.sh and get the following message : The CA certificate and the client certificate do not match What can cause this message ? I guess this is a problem as I am trying to access remotely the host machine and fail with a message of :libvir: Remote error : Connection refused That happens even if I use a client and server on the same machine and try : sudo virsh -c qemu://localhost/defualt. I am using libvirt 0.6. can I get first a connection with no certificate/encryption by using URI = qemu+tcp://... by making the needed change on the libvirtd.conf file on the server side ? that did not work either I followed the instructions in the Libvirt Web Site (generate a CA private key and certificate then the client and server keys and certificates and copy them to appropriate locations),. Also restarted the libvirtd with the --listen --verbose flags on the server side what can be wrong with my steps ? Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

[libvirt] virDomainGetVcpus error with Fedora 10
by Kaitlin Rupert 20 Mar '09

20 Mar '09

Hi, I'm running Fedora 10 with the following libvirt version: libvirt-0.6.0-3 I'm calling virDomainGetVcpus() with the following parameters: virDomainGetVcpus(dom, info, max, NULL, 0); where max = 2 and dom and info are both non-NULL. However, I'm getting the following error from libvirt: libvir: Domain error : invalid argument in virDomainGetVcpus I tracked this down, and it's failing the if (cpumaps != NULL && maplen < 1) check in libvirt.c because cpumaps is non-NULL. Which is very strange because I'm definitely passing a NULL value. Here's the relevant libvirtd debug: 15:38:07.332: debug : virGetDomain:287 : New hash entry 0x816ab0 15:38:07.332: debug : virDomainGetVcpus:4080 : domain=0x816ab0, info=0x800530, maxinfo=2, cpumaps=0x7f8fb0, maplen=0 15:38:07.332: error : invalid argument in virDomainGetVcpus libvir: Domain error : invalid argument in virDomainGetVcpus 15:38:07.332: debug : virDomainFree:1795 : domain=0x816ab0 Has anyone else hit this same issue? -- Kaitlin Rupert IBM Linux Technology Center kaitlin(a)linux.vnet.ibm.com

2 6

[libvirt] svirt fix for schema on domain.rng
by Daniel J Walsh 19 Mar '09

19 Mar '09

3 3

[libvirt] Use posix_fallocate() to allocate disk space
by Amit Shah 19 Mar '09

19 Mar '09

These patches introduce safezero() to allocate chunks for a file and write zeroes to them. Not yet ready for merge; I don't understand the configure system yet and don't know how to test for posix_fallocate() availability. The second patch introduces a new variable, track_allocation_progress, which perhaps belongs somewhere else. Pointers for both these problems appreciated. Thanks, Amit

3 10

[libvirt] [PATCH 2/2] [ plain text ] OpenNebula driver, libvirt-0.6.1
by "Abel Míguez Rodríguez" 19 Mar '09

19 Mar '09

> On Wed, Mar 18, 2009 at 01:52:56PM +0100, "Abel Míguez > Rodríguez" wrote: > Can you repost that from a normail mail client and without HTML > formatting ... it's simply unreadable for me, > Of course, I repost the e-mail using plain text, but (right now) I have to use a web client. Sorry for the inconvenience. Thanks, Abel > thanks, > > Daniel > > > -- > Daniel Veillard | libxml Gnome XML > XSLT toolkit http://xmlsoft.org/ > daniel(a)veillard.com | Rpmfind RPM search engine > http://rpmfind.net/http://veillard.com/ | virtualization > library http://libvirt.org/ > >> Hi all, >> Here are listed the ONE driver's source code: >> (I also attached the files to this email) >> * one_conf.h >> * one_conf.c >> * one_driver.h >> * one_driver.c >> ----- >> Thanks, >> Abel Míguez source: one_conf.h /*----------------------------------------------------------------------------------*/ /* Copyright 2002-2009, Distributed Systems Architecture Group, Universidad * Complutense de Madrid (dsa-research.org) * * 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, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ /*-----------------------------------------------------------------------------------*/ #ifndef ONE_CONF_H #define ONE_CONF_H #include <config.h> #include "internal.h" #include "domain_conf.h" #include "capabilities.h" #include "threads.h" struct one_driver{ virMutex lock; virCapsPtr caps; virDomainObjList domains; int nextid; }; typedef struct one_driver one_driver_t; virCapsPtr oneCapsInit(void); int oneSubmitVM(virConnectPtr conn ,one_driver_t* driver, virDomainObjPtr vm); int xmlOneTemplate(virDomainDefPtr def, FILE* fd); #define oneError(conn, dom, code, fmt...) \ virReportErrorHelper(conn, VIR_FROM_ONE, code, __FILE__, \ __FUNCTION__, __LINE__, fmt) #endif /* ONE_CONF_H */ +++++++++++++++++++++++++++++ source: one_conf.c /*----------------------------------------------------------------------------------*/ /* Copyright 2002-2009, Distributed Systems Architecture Group, Universidad * Complutense de Madrid (dsa-research.org) * * 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, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ /*-----------------------------------------------------------------------------------*/ #include <config.h> #include <sys/utsname.h> #include "virterror_internal.h" #include "one_conf.h" /* --------------------------------------------------------------------------------- */ /* --------------------------------------------------------------------------------- */ /* --------------------------------------------------------------------------------- */ /** * oneCapsInit initialize the driver capabilities * @return a pointer to the driver capabilities NULL in case of error */ virCapsPtr oneCapsInit(void) { struct utsname utsname; virCapsPtr caps; virCapsGuestPtr guest; uname(&utsname); if ((caps = virCapabilitiesNew(utsname.machine,0,0)) == NULL) { goto no_memory; } virCapabilitiesSetMacPrefix(caps,(unsigned char[]){ 0x52, 0x54, 0x00 }); if ((guest = virCapabilitiesAddGuest(caps, "hvm", utsname.machine, sizeof(int) == 4 ? 32 : 64, NULL, NULL, 0, NULL)) == NULL) { goto no_memory; } if (virCapabilitiesAddGuestDomain(guest, "hvm", NULL, NULL, 0, NULL) == NULL) { goto no_memory; } return caps; no_memory: virCapabilitiesFree(caps); return NULL; } /* --------------------------------------------------------------------------------- */ /* --------------------------------------------------------------------------------- */ /* --------------------------------------------------------------------------------- */ /** * oneSubmitVM generates an OpenNebula description file and submits the new VM * @param driver the OpenNebula driver * @param vm the virtual machine pointer * @return the OpenNebula ID for the new VM or -1 in case of error */ int oneSubmitVM(virConnectPtr conn ATTRIBUTE_UNUSED, one_driver_t* driver, virDomainObjPtr vm) { FILE* fd=NULL; char path[256]; int oneid; snprintf(path,256,"/tmp/one-%d",driver->nextid); fd=fopen(path,"w"); if(fd==NULL) { perror("Creating a temporal description file:"); return -1; } if(xmlOneTemplate(vm->def,fd)) { unlink(path); fclose(fd); return -1; } fclose(fd); if( (oneid=c_oneAllocate(path))<0 ) { oneError(conn, NULL, VIR_ERR_OPERATION_FAILED, "Error submitting virtual machine to OpenNebula"); unlink(path); return -1; }; unlink(path); return oneid; }; /* --------------------------------------------------------------------------------- */ /* --------------------------------------------------------------------------------- */ /* --------------------------------------------------------------------------------- */ /** * xmlOneTemplate Generate an OpenNebula template to deploy a VM from libvirt * internal Domain definition. * @param def Internal libvirt Domain definition * @param fd ONE template file to create * @return 0 if success */ int xmlOneTemplate(virDomainDefPtr def, FILE* fd) { char* buf; int i; fprintf(fd,"#OpenNebula Template automatically generated by libvirt\n"); fprintf(fd,"NAME = %s\n",def->name); fprintf(fd,"CPU = %ld\n",(def->vcpus)); fprintf(fd,"MEMORY = %ld\n",(def->maxmem)/1024); /*Optional Booting OpenNebula Information:*/ if( def->os.kernel ) { fprintf(fd,"OS=[ kernel = \"%s\"",def->os.kernel); if(def->os.initrd) fprintf(fd,",\n initrd = \"%s\"",def->os.initrd); if(def->os.cmdline) fprintf(fd,",\n kernel_cmd = \"%s\"",def->os.cmdline); if(def->os.root) fprintf(fd,",\n root = \"%s\"",def->os.root); fprintf(fd," ]\n"); } /* set Disks & NICS */ for(i=0 ; i<def->ndisks ; i++) { if ( !def->disks[i] || !def->disks[i]->src || !def->disks[i]->src ) { continue; } fprintf(fd, "DISK=[ source = \"%s\",\n" " target = \"%s\",\n" " readonly =", def->disks[i]->src, def->disks[i]->dst); if(def->disks[i]->readonly) fprintf(fd,"\"yes\"]\n"); else fprintf(fd,"\"no\"]\n"); } for(i=0 ; i< def->nnets ; i++) { if ( !def->nets[i] ) { continue; } switch(def->nets[i]->type) { case VIR_DOMAIN_NET_TYPE_BRIDGE: fprintf(fd,"NIC=[ bridge =\"%s\",\n",def->nets[i]->data.bridge.brname); if(def->nets[i]->ifname) { fprintf(fd," target =\"%s\",\n",def->nets[i]->ifname); } fprintf(fd," mac =\"%02x:%02x:%02x:%02x:%02x:%02x\" ]\n", def->nets[i]->mac[0],def->nets[i]->mac[1], def->nets[i]->mac[2],def->nets[i]->mac[3], def->nets[i]->mac[4],def->nets[i]->mac[5]); break; case VIR_DOMAIN_NET_TYPE_NETWORK: fprintf(fd,"NIC=[ network=\"%s\" ]\n",def->nets[i]->data.network.name); break; } } return 0; }; +++++++++++++++++++++++++++++ source: one_driver.h /*---------------------------------------------------------------------------*/ /* Copyright 2002-2009, Distributed Systems Architecture Group, Universidad * Complutense de Madrid (dsa-research.org) * * 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, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ /*---------------------------------------------------------------------------*/ #ifndef ONE_DRIVER_H #define ONE_DRIVER_H #include <config.h> int oneRegister(void); #endif /* ONE_DRIVER_H */ +++++++++++++++++++++++++++++ source: one_driver.c /*---------------------------------------------------------------------------*/ /* Copyright 2002-2009, Distributed Systems Architecture Group, Universidad * Complutense de Madrid (dsa-research.org) * * 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, write to the Free Software * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ /*---------------------------------------------------------------------------*/ #include <config.h> #include <fcntl.h> #include <sched.h> #include <sys/utsname.h> #include <stdbool.h> #include <string.h> #include <sys/types.h> #include <sys/socket.h> #include <sys/un.h> #include <sys/poll.h> #include <unistd.h> #include <wait.h> #include <sys/time.h> #include "virterror_internal.h" #include "logging.h" #include "datatypes.h" #include "one_conf.h" #include "one_driver.h" #include "memory.h" #include "util.h" #include "bridge.h" #include "veth.h" #include "event.h" #include "cgroup.h" static int oneStart(void); static int oneShutdown(void); static int oneActive(void); static void oneDriverLock(one_driver_t* driver) { virMutexLock(&driver->lock); } static void oneDriverUnlock(one_driver_t* driver) { virMutexUnlock(&driver->lock); } static one_driver_t *one_driver =NULL; static virDrvOpenStatus oneOpen(virConnectPtr conn, virConnectAuthPtr auth ATTRIBUTE_UNUSED, int flags ATTRIBUTE_UNUSED) { /* Verify uri was specified */ if (conn->uri == NULL) { conn->uri = xmlParseURI("one:///"); if (!conn->uri) { oneError(conn, NULL, VIR_ERR_NO_MEMORY, NULL); return VIR_DRV_OPEN_ERROR; } } else if (conn->uri->scheme == NULL || STRNEQ(conn->uri->scheme, "one")) { goto declineConnection; } conn->privateData = one_driver; return VIR_DRV_OPEN_SUCCESS; declineConnection: return VIR_DRV_OPEN_DECLINED; } static int oneClose(virConnectPtr conn) { conn->privateData = NULL; return 0; } static virDomainPtr oneDomainLookupByID(virConnectPtr conn, int id) { one_driver_t *driver = (one_driver_t *)conn->privateData; virDomainPtr dom = NULL; virDomainObjPtr vm = NULL; oneDriverLock(driver); vm = virDomainFindByID(&driver->domains, id); oneDriverUnlock(driver); if (!vm) { oneError(conn, NULL, VIR_ERR_NO_DOMAIN, NULL); goto return_point; } dom = virGetDomain(conn, vm->def->name, vm->def->uuid); if (dom) { dom->id = vm->def->id; } return_point: if(vm) { virDomainObjUnlock(vm); } return dom; } static virDomainPtr oneDomainLookupByUUID(virConnectPtr conn, const unsigned char *uuid) { one_driver_t *driver = (one_driver_t *)conn->privateData; virDomainPtr dom = NULL; virDomainObjPtr vm = NULL; oneDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, uuid); oneDriverUnlock(driver); if (!vm) { oneError(conn, NULL, VIR_ERR_NO_DOMAIN, NULL); goto return_point; } dom = virGetDomain(conn, vm->def->name, vm->def->uuid); if (dom) { dom->id = vm->def->id; } return_point: if(vm) { virDomainObjUnlock(vm); } return dom; } static virDomainPtr oneDomainLookupByName(virConnectPtr conn, const char *name) { one_driver_t *driver = (one_driver_t *)conn->privateData; virDomainObjPtr vm = NULL; virDomainPtr dom=NULL; oneDriverLock(driver); vm = virDomainFindByName(&driver->domains, name); oneDriverUnlock(driver); if (!vm) { oneError(conn, NULL, VIR_ERR_NO_DOMAIN, NULL); goto return_point; } dom = virGetDomain(conn, vm->def->name, vm->def->uuid); if (dom) { dom->id = vm->def->id; } return_point: if(vm) { virDomainObjUnlock(vm); } return dom; } static int oneListDomains(virConnectPtr conn, int *ids, int nids) { one_driver_t *driver = (one_driver_t *)conn->privateData; int got = 0, i; oneDriverLock(driver); for (i = 0 ; i < driver->domains.count && got < nids ; i++){ if (virDomainIsActive(driver->domains.objs[i])) ids[got++] = driver->domains.objs[i]->def->id; } oneDriverUnlock(driver); return got; } static int oneNumDomains(virConnectPtr conn) { one_driver_t *driver = (one_driver_t *)conn->privateData; int n = 0, i; oneDriverLock(driver); for (i = 0 ; i < driver->domains.count ; i++) if (virDomainIsActive(driver->domains.objs[i])) n++; oneDriverUnlock(driver); return n; } static int oneListDefinedDomains(virConnectPtr conn, char **const names, int nnames) { one_driver_t *driver = (one_driver_t *)conn->privateData; int got = 0, i; oneDriverLock(driver); for (i = 0 ; i < driver->domains.count && got < nnames ; i++) { if (!virDomainIsActive(driver->domains.objs[i])) { if (!(names[got++] = strdup(driver->domains.objs[i]->def->name))) { oneError(conn, NULL, VIR_ERR_NO_MEMORY, "%s", _("failed to allocate space for VM name string")); goto cleanup; } } } oneDriverUnlock(driver); return got; cleanup: for (i = 0 ; i < got ; i++) VIR_FREE(names[i]); oneDriverUnlock(driver); return -1; } static int oneNumDefinedDomains(virConnectPtr conn) { one_driver_t *driver = (one_driver_t *)conn->privateData; int n = 0, i; oneDriverLock(driver); for (i = 0 ; i < driver->domains.count ; i++) if (!virDomainIsActive(driver->domains.objs[i])) n++; oneDriverUnlock(driver); return n; } static virDomainPtr oneDomainDefine(virConnectPtr conn, const char *xml) { one_driver_t *driver = (one_driver_t *)conn->privateData; virDomainDefPtr def; virDomainObjPtr vm; virDomainPtr dom=NULL; oneDriverLock(driver); if (!(def = virDomainDefParseString(conn, driver->caps, xml, VIR_DOMAIN_XML_INACTIVE))) goto return_point; if (!(vm = virDomainAssignDef(conn, &driver->domains, def))) { virDomainDefFree(def); goto return_point; } vm->def->id = -1; vm->persistent = 1; dom = virGetDomain(conn, vm->def->name, vm->def->uuid); if (dom) { dom->id = vm->def->id; } virDomainObjUnlock(vm); return_point: oneDriverUnlock(driver); return dom; } static int oneDomainUndefine(virDomainPtr dom) { one_driver_t *driver = (one_driver_t *)dom->conn->privateData; virDomainObjPtr vm = NULL; int ret=-1; oneDriverLock(driver); vm =virDomainFindByUUID(&driver->domains, dom->uuid); if (!vm) { oneError(dom->conn, NULL, VIR_ERR_INVALID_DOMAIN, "%s", _("no domain with matching uuid")); goto return_point; } if (!vm->persistent) { oneError(dom->conn, dom, VIR_ERR_INTERNAL_ERROR, "%s", _("cannot undefine transient domain")); goto return_point; } virDomainRemoveInactive(&driver->domains, vm); ret=0; return_point: oneDriverUnlock(driver); return ret; } static int oneDomainGetInfo(virDomainPtr dom, virDomainInfoPtr info) { one_driver_t *driver = (one_driver_t *)dom->conn->privateData; struct timeval tv; virDomainObjPtr vm; oneDriverLock(driver); vm= virDomainFindByUUID(&driver->domains, dom->uuid); oneDriverUnlock(driver); if (!vm) { oneError(dom->conn,NULL, VIR_ERR_INVALID_DOMAIN, "%s", _("no domain with matching uuid")); return -1; } if(gettimeofday(&tv,NULL)<0) { oneError(dom->conn,NULL, VIR_ERR_INTERNAL_ERROR, "%s",_("getting time of day")); virDomainObjUnlock(vm); return -1; } if (!virDomainIsActive(vm)) { info->cpuTime = 0; } else { char vm_info[257]; c_oneVmInfo(vm->pid,vm_info,256); //State: char* cptr = strstr(vm_info,"STATE"); cptr = index(cptr, ':'); cptr++; int one_state=atoi(cptr); switch(one_state) { case 3: /** running */ if (vm->state!=VIR_DOMAIN_SHUTDOWN) vm->state=VIR_DOMAIN_RUNNING; break; case 5: /** pause */ vm->state=VIR_DOMAIN_PAUSED; break; case 6: /** done */ vm->state=VIR_DOMAIN_SHUTOFF; vm->def->id=-1; break; case 7: /** error */ vm->state=VIR_DOMAIN_CRASHED; break; default: break; }; //Memory: cptr=strstr(vm_info,"MEMORY"); cptr=index(cptr,':'); cptr++; vm->def->memory = atoi(cptr); //run time: cptr=strstr(vm_info,"START TIME"); cptr=index(cptr,':'); cptr++; long starttime = atol(cptr); info->cpuTime = (tv.tv_sec - starttime) *1000ll *1000ll *1000ll; } info->state = vm->state; info->maxMem = vm->def->maxmem; info->memory = vm->def->memory; info->nrVirtCpu = vm->def->vcpus; virDomainObjUnlock(vm); return 0; } static char *oneGetOSType(virDomainPtr dom) { one_driver_t *driver = (one_driver_t *)dom->conn->privateData; virDomainObjPtr vm = NULL; oneDriverLock(driver); vm =virDomainFindByUUID(&driver->domains, dom->uuid); oneDriverUnlock(driver); if (!vm) { oneError(dom->conn, NULL, VIR_ERR_INVALID_DOMAIN, "%s", _("no domain with matching uuid")); return NULL; } virDomainObjUnlock(vm); return strdup(vm->def->os.type); } static int oneDomainStart(virDomainPtr dom) { virConnectPtr conn = dom->conn; one_driver_t *driver = (one_driver_t *)(conn->privateData); virDomainObjPtr vm; int ret = -1; int oneid; int rc; char* logfile=NULL; oneDriverLock(driver); vm = virDomainFindByName(&driver->domains, dom->name); if (!vm) { oneError(conn, NULL, VIR_ERR_INVALID_DOMAIN, _("no domain named %s"), dom->name); goto return_point; } if((oneid = oneSubmitVM(dom->conn,driver,vm)) < 0) { goto return_point; } vm->pid=oneid; vm->def->id=driver->nextid++; vm->state=VIR_DOMAIN_BLOCKED; ret=0; return_point: if(vm) virDomainObjUnlock(vm); oneDriverUnlock(driver); return ret; } static virDomainPtr oneDomainCreateAndStart(virConnectPtr conn, const char *xml, unsigned int flags ATTRIBUTE_UNUSED) { one_driver_t *driver = (one_driver_t *)conn->privateData; virDomainObjPtr vm; virDomainDefPtr def; virDomainPtr dom = NULL; int oneid; oneDriverLock(driver); if (!(def = virDomainDefParseString(conn, driver->caps, xml, VIR_DOMAIN_XML_INACTIVE))) goto return_point; vm = virDomainFindByName(&driver->domains, def->name); if (vm) { oneError(conn,NULL, VIR_ERR_OPERATION_FAILED, _("Already an OpenNebula VM active with the name: \"%s\" id: %d "), def->name,def->id); goto return_point; } if (!(vm = virDomainAssignDef(conn, &driver->domains, def))) { virDomainDefFree(def); goto return_point; } if ((oneid = oneSubmitVM(conn, driver, vm)) < 0) { virDomainRemoveInactive(&driver->domains, vm); vm=NULL; goto return_point; } vm->def->id=driver->nextid++; vm->persistent=0; vm->pid=oneid; vm->state=VIR_DOMAIN_BLOCKED; dom = virGetDomain(conn, vm->def->name, vm->def->uuid); if (dom) { dom->id = vm->def->id; } return_point: if(vm) virDomainObjUnlock(vm); oneDriverUnlock(driver); return dom; } static int oneDomainShutdown(virDomainPtr dom) { one_driver_t *driver = (one_driver_t*)dom->conn->privateData; virDomainObjPtr vm; int ret=-1; oneDriverLock(driver); if((vm=virDomainFindByID(&driver->domains, dom->id))) { if(!(c_oneShutdown(vm->pid) ) ) { vm->state=VIR_DOMAIN_SHUTDOWN; ret= 0; if (!vm->persistent) { virDomainRemoveInactive(&driver->domains, vm); vm = NULL; } goto return_point; } oneError(dom->conn, NULL, VIR_ERR_OPERATION_FAILED, "Domain Shutdown",NULL); goto return_point; } oneError(dom->conn,NULL, VIR_ERR_INVALID_DOMAIN, _("no domain with id %d"), dom->id); goto return_point; return_point: if(vm) virDomainObjUnlock(vm); oneDriverUnlock(driver); return ret; } static int oneDomainDestroy(virDomainPtr dom) { one_driver_t *driver = (one_driver_t*)dom->conn->privateData; virDomainObjPtr vm; int ret=-1; oneDriverLock(driver); vm= virDomainFindByID(&driver->domains, dom->id); if (!vm) { oneError(dom->conn, NULL, VIR_ERR_INVALID_DOMAIN, _("no domain with id %d"), dom->id); goto return_point; } if(c_oneShutdown(vm->pid)) { oneError(dom->conn, NULL, VIR_ERR_OPERATION_FAILED, "Domain Shutdown",NULL); goto return_point; } if(!vm->persistent) { virDomainRemoveInactive(&driver->domains,vm); vm=NULL; } ret=0; return_point: if(vm) virDomainObjUnlock(vm); oneDriverUnlock(driver); return ret; } static int oneDomainSuspend(virDomainPtr dom) { one_driver_t* driver=dom->conn->privateData; virDomainObjPtr vm; int ret=-1; oneDriverLock(driver); if ((vm=virDomainFindByID(&driver->domains,dom->id))){ if (vm->state == VIR_DOMAIN_RUNNING) { if( !(c_oneSuspend(vm->pid)) ) { vm->state=VIR_DOMAIN_PAUSED; ret=0; goto return_point; } oneError(dom->conn, NULL, VIR_ERR_OPERATION_FAILED, "%s", "failed to pause domain"); goto return_point; } oneError(dom->conn,NULL,VIR_ERR_OPERATION_FAILED, "%s", "domain is not running"); } else { oneError(dom->conn, NULL, VIR_ERR_INVALID_DOMAIN, "no domain with matching id %d", dom->id); } return_point: if(vm) virDomainObjUnlock(vm); oneDriverUnlock(driver); return ret; }; static int oneDomainResume(virDomainPtr dom) { one_driver_t* driver=dom->conn->privateData; virDomainObjPtr vm; int ret=-1; oneDriverLock(driver); if ((vm=virDomainFindByID(&driver->domains,dom->id))) { if (vm->state == VIR_DOMAIN_PAUSED) { if( !(c_oneResume(vm->pid)) ) { vm->state=VIR_DOMAIN_RUNNING; ret=0; goto return_point; } oneError(dom->conn, NULL, VIR_ERR_OPERATION_FAILED, "%s", "failed to resume "); goto return_point; } oneError(dom->conn,NULL,VIR_ERR_OPERATION_FAILED, "%s", " domain is not paused "); } else { oneError(dom->conn, NULL, VIR_ERR_INVALID_DOMAIN, "no domain with matching id %d", dom->id); } return_point: if(vm) virDomainObjUnlock(vm); oneDriverUnlock(driver); return ret; }; static int oneStartup(void){ int i; if (VIR_ALLOC(one_driver) < 0) { return -1; } if(virMutexInit(&one_driver->lock)<0){ VIR_FREE(one_driver); return -1; } c_oneStart(); oneDriverLock(one_driver); one_driver->nextid=1; if ((one_driver->caps = oneCapsInit()) == NULL) { oneDriverUnlock(one_driver); VIR_FREE(one_driver); return -1; } oneDriverUnlock(one_driver); return 0; } static int oneShutdown(void){ if (one_driver == NULL) return(-1); oneDriverLock(one_driver); virDomainObjListFree(&one_driver->domains); virCapabilitiesFree(one_driver->caps); oneDriverUnlock(one_driver); virMutexDestroy(&one_driver->lock); VIR_FREE(one_driver); one_driver = NULL; c_oneFree(); return 0; } static int oneActive(void){ unsigned int i; int active = 0; if (one_driver == NULL) return(0); oneDriverLock(one_driver); for (i = 0 ; i < one_driver->domains.count ; i++) { virDomainObjLock(one_driver->domains.objs[i]); if (virDomainIsActive(one_driver->domains.objs[i])) active = 1; virDomainObjUnlock(one_driver->domains.objs[i]); } oneDriverUnlock(one_driver); return active; } static int oneVersion(virConnectPtr conn ATTRIBUTE_UNUSED, unsigned long *hvVer) { *hvVer = 1; return 0; } static int oneGetAutostart(virConnectPtr conn ATTRIBUTE_UNUSED) { return 0; } /* Function Tables */ static virDriver oneDriver = { VIR_DRV_ONE, /* the number virDrvNo */ "ONE", /* the name of the driver */ oneOpen, /* open */ oneClose, /* close */ NULL, /* supports_feature */ NULL, /* type */ oneVersion, /* version */ NULL, /* getHostname */ NULL, /* getURI */ NULL, /* getMaxVcpus */ NULL, /* nodeGetInfo */ NULL, /* getCapabilities */ oneListDomains, /* listDomains */ oneNumDomains, /* numOfDomains */ oneDomainCreateAndStart, /* domainCreateXML */ oneDomainLookupByID, /* domainLookupByID */ oneDomainLookupByUUID, /* domainLookupByUUID */ oneDomainLookupByName, /* domainLookupByName */ oneDomainSuspend, /* domainSuspend */ oneDomainResume, /* domainResume */ oneDomainShutdown, /* domainShutdown */ NULL, /* domainReboot */ oneDomainDestroy, /* domainDestroy */ oneGetOSType, /* domainGetOSType */ NULL, /* domainGetMaxMemory */ NULL, /* domainSetMaxMemory */ NULL, /* domainSetMemory */ oneDomainGetInfo, /* domainGetInfo */ NULL, /* domainSave */ NULL, /* domainRestore */ NULL, /* domainCoreDump */ NULL, /* domainSetVcpus */ NULL, /* domainPinVcpu */ NULL, /* domainGetVcpus */ NULL, /* domainGetMaxVcpus */ NULL, /* domainGetSecurityLabel */ NULL, /* nodeGetSecurityModel */ NULL, /* domainDumpXML */ oneListDefinedDomains, /* listDefinedDomains */ oneNumDefinedDomains, /* numOfDefinedDomains */ oneDomainStart, /* domainCreate */ oneDomainDefine, /* domainDefineXML */ oneDomainUndefine, /* domainUndefine */ NULL, /* domainAttachDevice */ NULL, /* domainDetachDevice */ oneGetAutostart, /* domainGetAutostart */ NULL, /* domainSetAutostart */ NULL, /* domainGetSchedulerType */ NULL, /* domainGetSchedulerParameters */ NULL, /* domainSetSchedulerParameters */ NULL, /* domainMigratePrepare */ NULL, /* domainMigratePerform */ NULL, /* domainMigrateFinish */ NULL, /* domainBlockStats */ NULL, /* domainInterfaceStats */ NULL, /* domainBlockPeek */ NULL, /* domainMemoryPeek */ NULL, /* nodeGetCellsFreeMemory */ NULL, /* getFreeMemory */ NULL, /* domainEventRegister */ NULL, /* domainEventDeregister */ NULL, /* domainMigratePrepare2 */ NULL, /* domainMigrateFinish2 */ }; static virStateDriver oneStateDriver = { .initialize = oneStartup, .cleanup = oneShutdown, .active = oneActive, }; int oneRegister(void) { virRegisterDriver(&oneDriver); virRegisterStateDriver(&oneStateDriver); return 0; } ---- Distributed System Architecture Group (http://dsa-research.org) GridWay, http://www.gridway.org OpenNEbula, http://www.opennebula.org

2 1

[libvirt] get domU hostname
by jim_marshall＠msn.com 19 Mar '09

19 Mar '09

Hello, Sorry if this has been asked before, I did some searches and came up empty. I am writing a tool to get data on Xen using libvirt everything is going great, however; I have a need to get the hostname of domU from the program running in dom0. Is this possible using libvirt? Thanks Jim

2 1

[libvirt] java libvirt question
by Zvi Dubitzky 19 Mar '09

19 Mar '09

Hi 1) I downloaded the libvirt-java-0.2.1 tar and try to build Java binding to the libvirt library (configure , make , make install). This is because I am running Ubuntu and presumably cannot use the fedora .rpm For either java-gcj (java version 1.5) or the java-6-openjdk (version 1.6) I do not get the java-dev having its bin components (java, javac, javah, javadoc) under the same directory as the configure script wants ( gcj case : /usr/lib/jvm/java-1.5.0-gcj-../bin , openjdk case: /usr/lib/jvm/java-6-openjdk/bin) and the configure fails Of course I am doing : ./configure --with-java-home=/usr/lib/jvm/java-6-openjdk or similar for version 1.5 Any advice how to solve this problem ? in fact I do not find javac and javah, javadoc after the install only the java, jar 2) the Java libvirt library from the downloads link is dated mid 2008 . Does that have influence on bug fixing since then or this java libvirt package is just for the client side and the server side is the most updated C code so the date does not matter ? thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

Re: [libvirt] [PATCH] Adding filesystem mount support for openVZ
by Anton Protopopov 18 Mar '09

18 Mar '09

2009/3/12 Daniel P. Berrange <berrange(a)redhat.com> > >> On Wed, Mar 11, 2009 at 05:42:20PM +0100, Florian Vichot wrote: >> > Hi everyone, >> > >> > This patch is to allow using the "mount" type in the "filesystem" tag >> > for OpenVZ domains. >> > >> > Example: >> > ... >> > <filesystem type='mount'> >> > <source dir='/path/to/filesystem/directory/' /> >> > <target dir='/path/to/pivot/root/' /> >> > </filesystem> >> > ... >> > >> > This is my first patch to an external project, so don't spare me if I >> > got things wrong :) >> > >> > Also, I'm curious for suggestions as to how I could allow for the target >> > not to be specified in the XML. Because in this case OpenVZ just makes a >> > temporary pivot root in "/var/lib/vz/root/" and that is probably >> > sufficient for most people, who might not want to have to explicitly >> > create a pivot root somewhere, just for mounting the filesystem while >> > it's running. >> >> Actually the <target dir="..."> means something a little different. This >> refers to the mount point *within* the container. So for the root >> filesystem >> of the container it will be <target dir='/' /> > > No, that patch does something different---it assumes that both <source dir> > and <target dir> > are paths within hardware node. If we denote <source dir> and <target dir> > by the VE_PRIVATE > and VE_ROOT correspondingly, we get the following picture. > > In OpenVZ there are VE_ROOT and VE_PRIVATE values for each container > (they defined as VE_ROOT and VE_PRIVATE variables in container > configuration file). > The VE_PRIVATE defines the actual disposition of conatainer's > root filesystem hierarchy. It can lie on it's own partition or be just a > directory. > When vzctl starts a container, it mounts a VE_PRIVATE "filesystem" to > VE_ROOT > mountpoint and then container starts to work in VE_ROOT chroot... > > So, yes, you are right - in this case we can always assume, that "target > dir within container" is "/" :) > >

1 0

[libvirt] Re: rawhide report: 20090318 changes
by Tomasz Torcz 18 Mar '09

18 Mar '09

On Wed, Mar 18, 2009 at 11:04:12AM +0000, Rawhide Report wrote: > libvirt-0.6.1-5.fc11 > - Disable sound cards when running sVirt May I ask what's the reason for this change? -- Tomasz Torcz To co nierealne -- tutaj jest normalne. xmpp: zdzichubg(a)chrome.pl Ziomale na życie mają tu patenty specjalne.

2 1

[libvirt] [PATCH 2/2] OpenNebula driver, libvirt-0.6.1
by "Abel Míguez Rodríguez" 18 Mar '09

18 Mar '09

1 0

[libvirt] I have no idea why the current version of libvirt works for anyone in enforcing mode.
by Daniel J Walsh 17 Mar '09

17 Mar '09

Libvirt is executing qemu requiring it to execute pulseaudio which would require the folowing permissions, #============= svirt_t ============== allow svirt_t admin_home_t:dir setattr; allow svirt_t admin_home_t:file { read write }; allow svirt_t pulseaudio_port_t:tcp_socket name_connect; allow svirt_t svirt_tmpfs_t:file read; allow svirt_t user_tmpfs_t:file read; Since qemu(svirt_t) is not allowed these permissions, pulseaudio crashes and qemu dies. I believe you need to run without sound if you are running as root.

4 11

[libvirt] PATCH: PCI device passthrough in Xen driver
by Daniel P. Berrange 17 Mar '09

17 Mar '09

This patch provides initial support for PCI device passthrough in Xen, at time of boot. It does not (yet) implement device hotplug for PCI, since it works in an annoyingly different way to device hotplug of disks & nics. The patch has several aspects - Refactors the pci.c file to make it easier to support both Xen's pci-back.ko and upstream's pcistub.ko, which are almost identical - Add code to pci.c register the PCI device with 'new_slot' before invoking 'bind', since pci-back.ko wants this extra step - In xen_unified.c, add implmentations of APIs to dettach, reattach and reset PCI devices - this just calls into shared pci.c code that QEMU/KVM uses. - In xend_internal.c implement SEXPR conversion. - In xm_internal.c implement parsing of 'pci' config parameter - Add lots of test cases XenD only supports 'unmanaged' PCI devices - ie mgmt app is responsible for detaching/reattaching PCI devices from/to host device drivers. XenD itself won't automatically do this for us, though perhaps someone might like to implement this as an optional feature for XenD, since it is more convenient in some cases. src/pci.c | 182 +++++++++++++-------- src/xen_unified.c | 122 ++++++++++++++ src/xend_internal.c | 232 +++++++++++++++++++++++++++ src/xm_internal.c | 156 ++++++++++++++++++ tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr | 2 tests/sexpr2xmldata/sexpr2xml-pci-devs.xml | 37 ++++ tests/sexpr2xmltest.c | 1 tests/xmconfigdata/test-pci-devs.cfg | 24 ++ tests/xmconfigdata/test-pci-devs.xml | 56 ++++++ tests/xmconfigtest.c | 1 tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr | 1 tests/xml2sexprdata/xml2sexpr-pci-devs.xml | 33 +++ tests/xml2sexprtest.c | 1 13 files changed, 784 insertions(+), 64 deletions(-) Daniel Index: src/pci.c =================================================================== RCS file: /data/cvs/libvirt/src/pci.c,v retrieving revision 1.4 diff -u -p -r1.4 pci.c --- src/pci.c 3 Mar 2009 17:00:18 -0000 1.4 +++ src/pci.c 16 Mar 2009 11:31:29 -0000 @@ -614,46 +614,80 @@ pciResetDevice(virConnectPtr conn, pciDe return ret; } -static int -pciBindDeviceToStub(virConnectPtr conn, pciDevice *dev, const char *stub_module) + +static void +pciDriverDir(char *buf, size_t buflen, const char *driver) { - char stub_dir[PATH_MAX]; - char path[PATH_MAX]; + snprintf(buf, buflen, PCI_SYSFS "drivers/%s", driver); +} - snprintf(stub_dir, sizeof(stub_dir), PCI_SYSFS "drivers/%s", stub_module); +static void +pciDriverFile(char *buf, size_t buflen, const char *driver, const char *file) +{ + snprintf(buf, buflen, PCI_SYSFS "drivers/%s/%s", driver, file); +} - /* Try loading the stub module if it isn't already loaded; - * Do not return an error if the stub module is not available. - */ - if (!virFileExists(stub_dir)) { - const char *const modprobeargv[] = { MODPROBE, stub_module, NULL }; +static void +pciDeviceFile(char *buf, size_t buflen, const char *device, const char *file) +{ + snprintf(buf, buflen, PCI_SYSFS "devices/%s/%s", device, file); +} + + +static const char * +pciFindStubDriver(virConnectPtr conn) +{ + char drvpath[PATH_MAX]; + int probed = 0; - if (virRun(conn, modprobeargv, NULL) < 0) { +recheck: + pciDriverDir(drvpath, sizeof(drvpath), "pci-stub"); + if (virFileExists(drvpath)) + return "pci-stub"; + pciDriverDir(drvpath, sizeof(drvpath), "pciback"); + if (virFileExists(drvpath)) + return "pciback"; + + if (!probed) { + const char *const stubprobe[] = { MODPROBE, "pci-stub", NULL }; + const char *const backprobe[] = { MODPROBE, "pciback", NULL }; + + probed = 1; + if (virRun(conn, stubprobe, NULL) < 0 && + virRun(conn, backprobe, NULL) < 0) { char ebuf[1024]; - VIR_WARN(_("modprobe %s failed: %s"), stub_module, + VIR_WARN(_("failed to load pci-stub or pciback drivers: %s"), virStrerror(errno, ebuf, sizeof ebuf)); + return 0; } + + goto recheck; } - if (!virFileExists(stub_dir)) { - VIR_WARN(_("%s module not available, cannot bind device %s to it"), - stub_module, dev->name); - } else { - /* Add the PCI device ID to the stub's dynamic ID table; - * this is needed to allow us to bind the device to the stub. - * Note: if the device is not currently bound to any driver, - * stub will immediately be bound to the device. Also, note - * that if a new device with this ID is hotplugged, or if a probe - * is triggered for such a device, it will also be immediately - * bound by the stub. - */ - snprintf(path, sizeof(path), "%s/new_id", stub_dir); - if (virFileWriteStr(path, dev->id) < 0) { - virReportSystemError(conn, errno, - _("Failed to add PCI device ID '%s' to %s"), - dev->id, stub_module); - return -1; - } + return NULL; +} + + +static int +pciBindDeviceToStub(virConnectPtr conn, pciDevice *dev, const char *driver) +{ + char drvdir[PATH_MAX]; + char path[PATH_MAX]; + + /* Add the PCI device ID to the stub's dynamic ID table; + * this is needed to allow us to bind the device to the stub. + * Note: if the device is not currently bound to any driver, + * stub will immediately be bound to the device. Also, note + * that if a new device with this ID is hotplugged, or if a probe + * is triggered for such a device, it will also be immediately + * bound by the stub. + */ + pciDriverFile(path, sizeof(path), driver, "new_id"); + if (virFileWriteStr(path, dev->id) < 0) { + virReportSystemError(conn, errno, + _("Failed to add PCI device ID '%s' to %s"), + dev->id, driver); + return -1; } /* If the device is already bound to a driver, unbind it. @@ -661,66 +695,79 @@ pciBindDeviceToStub(virConnectPtr conn, * PCI device happens to be IDE controller for the disk hosting * your root filesystem. */ - snprintf(path, sizeof(path), - PCI_SYSFS "devices/%s/driver/unbind", dev->name); + pciDeviceFile(path, sizeof(path), dev->name, "driver/unbind"); if (virFileExists(path) && virFileWriteStr(path, dev->name) < 0) { virReportSystemError(conn, errno, _("Failed to unbind PCI device '%s'"), dev->name); return -1; } - if (virFileExists(stub_dir)) { - /* If the device isn't already bound to pci-stub, try binding it now. - */ - snprintf(path, sizeof(path), PCI_SYSFS "devices/%s/driver", dev->name); - if (!virFileLinkPointsTo(path, stub_dir)) { - snprintf(path, sizeof(path), "%s/bind", stub_dir); - if (virFileWriteStr(path, dev->name) < 0) { - virReportSystemError(conn, errno, - _("Failed to bind PCI device '%s' to %s"), - dev->name, stub_module); - return -1; - } + /* If the device isn't already bound to pci-stub, try binding it now. + */ + pciDriverDir(drvdir, sizeof(drvdir), driver); + pciDeviceFile(path, sizeof(path), dev->name, "driver"); + if (!virFileLinkPointsTo(path, drvdir)) { + /* Xen's pciback.ko wants you to use new_slot first */ + pciDriverFile(path, sizeof(path), driver, "new_slot"); + if (virFileExists(path) && virFileWriteStr(path, dev->name) < 0) { + virReportSystemError(conn, errno, + _("Failed to add slot for PCI device '%s' to %s"), + dev->name, driver); + return -1; } - /* If 'remove_id' exists, remove the device id from pci-stub's dynamic - * ID table so that 'drivers_probe' works below. - */ - snprintf(path, sizeof(path), "%s/remove_id", stub_dir); - if (virFileExists(path) && virFileWriteStr(path, dev->id) < 0) { + pciDriverFile(path, sizeof(path), driver, "bind"); + if (virFileWriteStr(path, dev->name) < 0) { virReportSystemError(conn, errno, - _("Failed to remove PCI ID '%s' from %s"), - dev->id, stub_module); + _("Failed to bind PCI device '%s' to %s"), + dev->name, driver); return -1; } } + /* If 'remove_id' exists, remove the device id from pci-stub's dynamic + * ID table so that 'drivers_probe' works below. + */ + pciDriverFile(path, sizeof(path), driver, "remove_id"); + if (virFileExists(path) && virFileWriteStr(path, dev->id) < 0) { + virReportSystemError(conn, errno, + _("Failed to remove PCI ID '%s' from %s"), + dev->id, driver); + return -1; + } + return 0; } int pciDettachDevice(virConnectPtr conn, pciDevice *dev) { - return pciBindDeviceToStub(conn, dev, "pci-stub"); + const char *driver = pciFindStubDriver(conn); + if (!driver) { + pciReportError(conn, VIR_ERR_INTERNAL_ERROR, "%s", + _("cannot find any PCI stub module")); + return -1; + } + + return pciBindDeviceToStub(conn, dev, driver); } static int -pciUnBindDeviceFromStub(virConnectPtr conn, pciDevice *dev, const char *stub_module) +pciUnBindDeviceFromStub(virConnectPtr conn, pciDevice *dev, const char *driver) { - char stub_dir[PATH_MAX]; + char drvdir[PATH_MAX]; char path[PATH_MAX]; - snprintf(stub_dir, sizeof(stub_dir), PCI_SYSFS "drivers/%s", stub_module); - /* If the device is bound to stub, unbind it. */ - snprintf(path, sizeof(path), PCI_SYSFS "devices/%s/driver", dev->name); - if (virFileExists(stub_dir) && virFileLinkPointsTo(path, stub_dir)) { - snprintf(path, sizeof(path), "%s/unbind", stub_dir); + pciDriverDir(drvdir, sizeof(drvdir), driver); + pciDeviceFile(path, sizeof(path), dev->name, "driver"); + if (virFileExists(drvdir) && virFileLinkPointsTo(path, drvdir)) { + pciDriverFile(path, sizeof(path), driver, "unbind"); if (virFileWriteStr(path, dev->name) < 0) { virReportSystemError(conn, errno, _("Failed to bind PCI device '%s' to %s"), - dev->name, stub_module); + dev->name, driver); return -1; } } @@ -730,8 +777,8 @@ pciUnBindDeviceFromStub(virConnectPtr co * available, then re-probing would just cause the device to be * re-bound to the stub. */ - snprintf(path, sizeof(path), "%s/remove_id", stub_dir); - if (!virFileExists(stub_dir) || virFileExists(path)) { + pciDriverFile(path, sizeof(path), driver, "remove_id"); + if (!virFileExists(drvdir) || virFileExists(path)) { if (virFileWriteStr(PCI_SYSFS "drivers_probe", dev->name) < 0) { virReportSystemError(conn, errno, _("Failed to trigger a re-probe for PCI device '%s'"), @@ -746,7 +793,14 @@ pciUnBindDeviceFromStub(virConnectPtr co int pciReAttachDevice(virConnectPtr conn, pciDevice *dev) { - return pciUnBindDeviceFromStub(conn, dev, "pci-stub"); + const char *driver = pciFindStubDriver(conn); + if (!driver) { + pciReportError(conn, VIR_ERR_INTERNAL_ERROR, "%s", + _("cannot find any PCI stub module")); + return -1; + } + + return pciUnBindDeviceFromStub(conn, dev, driver); } static char * Index: src/xen_unified.c =================================================================== RCS file: /data/cvs/libvirt/src/xen_unified.c,v retrieving revision 1.83 diff -u -p -r1.83 xen_unified.c --- src/xen_unified.c 5 Feb 2009 16:03:11 -0000 1.83 +++ src/xen_unified.c 16 Mar 2009 11:31:29 -0000 @@ -43,6 +43,8 @@ #include "xml.h" #include "util.h" #include "memory.h" +#include "node_device_conf.h" +#include "pci.h" #define VIR_FROM_THIS VIR_FROM_XEN @@ -1420,6 +1422,123 @@ xenUnifiedDomainEventDeregister (virConn return ret; } + +static int +xenUnifiedNodeDeviceGetPciInfo (virNodeDevicePtr dev, + unsigned *domain, + unsigned *bus, + unsigned *slot, + unsigned *function) +{ + virNodeDeviceDefPtr def = NULL; + virNodeDevCapsDefPtr cap; + char *xml = NULL; + int ret = -1; + + xml = virNodeDeviceGetXMLDesc(dev, 0); + if (!xml) + goto out; + + def = virNodeDeviceDefParseString(dev->conn, xml); + if (!def) + goto out; + + cap = def->caps; + while (cap) { + if (cap->type == VIR_NODE_DEV_CAP_PCI_DEV) { + *domain = cap->data.pci_dev.domain; + *bus = cap->data.pci_dev.bus; + *slot = cap->data.pci_dev.slot; + *function = cap->data.pci_dev.function; + break; + } + + cap = cap->next; + } + + if (!cap) { + xenUnifiedError(dev->conn, VIR_ERR_INVALID_ARG, + _("device %s is not a PCI device"), dev->name); + goto out; + } + + ret = 0; +out: + virNodeDeviceDefFree(def); + VIR_FREE(xml); + return ret; +} + +static int +xenUnifiedNodeDeviceDettach (virNodeDevicePtr dev) +{ + pciDevice *pci; + unsigned domain, bus, slot, function; + int ret = -1; + + if (xenUnifiedNodeDeviceGetPciInfo(dev, &domain, &bus, &slot, &function) < 0) + return -1; + + pci = pciGetDevice(dev->conn, domain, bus, slot, function); + if (!pci) + return -1; + + if (pciDettachDevice(dev->conn, pci) < 0) + goto out; + + ret = 0; +out: + pciFreeDevice(dev->conn, pci); + return ret; +} + +static int +xenUnifiedNodeDeviceReAttach (virNodeDevicePtr dev) +{ + pciDevice *pci; + unsigned domain, bus, slot, function; + int ret = -1; + + if (xenUnifiedNodeDeviceGetPciInfo(dev, &domain, &bus, &slot, &function) < 0) + return -1; + + pci = pciGetDevice(dev->conn, domain, bus, slot, function); + if (!pci) + return -1; + + if (pciReAttachDevice(dev->conn, pci) < 0) + goto out; + + ret = 0; +out: + pciFreeDevice(dev->conn, pci); + return ret; +} + +static int +xenUnifiedNodeDeviceReset (virNodeDevicePtr dev) +{ + pciDevice *pci; + unsigned domain, bus, slot, function; + int ret = -1; + + if (xenUnifiedNodeDeviceGetPciInfo(dev, &domain, &bus, &slot, &function) < 0) + return -1; + + pci = pciGetDevice(dev->conn, domain, bus, slot, function); + if (!pci) + return -1; + + if (pciResetDevice(dev->conn, pci) < 0) + goto out; + + ret = 0; +out: + pciFreeDevice(dev->conn, pci); + return ret; +} + + /*----- Register with libvirt.c, and initialise Xen drivers. -----*/ /* The interface which we export upwards to libvirt.c. */ @@ -1481,6 +1600,9 @@ static virDriver xenUnifiedDriver = { .getFreeMemory = xenUnifiedNodeGetFreeMemory, .domainEventRegister = xenUnifiedDomainEventRegister, .domainEventDeregister = xenUnifiedDomainEventDeregister, + .nodeDeviceDettach = xenUnifiedNodeDeviceDettach, + .nodeDeviceReAttach = xenUnifiedNodeDeviceReAttach, + .nodeDeviceReset = xenUnifiedNodeDeviceReset, }; /** Index: src/xend_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/xend_internal.c,v retrieving revision 1.253 diff -u -p -r1.253 xend_internal.c --- src/xend_internal.c 10 Mar 2009 11:13:32 -0000 1.253 +++ src/xend_internal.c 16 Mar 2009 11:31:29 -0000 @@ -92,6 +92,11 @@ xenDaemonFormatSxprNet(virConnectPtr con int xendConfigVersion, int isAttach); static int +xenDaemonFormatSxprOnePCI(virConnectPtr conn, + virDomainHostdevDefPtr def, + virBufferPtr buf); + +static int virDomainXMLDevID(virDomainPtr domain, virDomainDeviceDefPtr dev, char *class, @@ -2145,6 +2150,131 @@ error: return -1; } +/** + * xenDaemonParseSxprPCI + * @conn: connection + * @root: root sexpr + * + * This parses out block devices from the domain sexpr + * + * Returns 0 if successful or -1 if failed. + */ +static int +xenDaemonParseSxprPCI(virConnectPtr conn, + virDomainDefPtr def, + const struct sexpr *root) +{ + const struct sexpr *cur, *tmp = NULL, *node; + virDomainHostdevDefPtr dev = NULL; + + /* + * With the (domain ...) block we have the following odd setup + * + * (device + * (pci + * (dev (domain 0x0000) (bus 0x00) (slot 0x1b) (func 0x0)) + * (dev (domain 0x0000) (bus 0x00) (slot 0x13) (func 0x0)) + * ) + * ) + * + * Normally there is one (device ...) block per device, but in + * wierd world of Xen PCI, once (device ...) covers multiple + * devices. + */ + + for (cur = root; cur->kind == SEXPR_CONS; cur = cur->u.s.cdr) { + node = cur->u.s.car; + if ((tmp = sexpr_lookup(node, "device/pci")) != NULL) + break; + } + + if (!tmp) + return 0; + + for (cur = tmp; cur->kind == SEXPR_CONS; cur = cur->u.s.cdr) { + const char *domain = NULL; + const char *bus = NULL; + const char *slot = NULL; + const char *func = NULL; + int domainID; + int busID; + int slotID; + int funcID; + + node = cur->u.s.car; + if (!sexpr_lookup(node, "dev")) + continue; + + if (!(domain = sexpr_node(node, "dev/domain"))) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("missing PCI domain")); + goto error; + } + if (!(bus = sexpr_node(node, "dev/bus"))) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("missing PCI bus")); + goto error; + } + if (!(slot = sexpr_node(node, "dev/slot"))) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("missing PCI slot")); + goto error; + } + if (!(func = sexpr_node(node, "dev/func"))) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + "%s", _("missing PCI func")); + goto error; + } + + if (virStrToLong_i(domain, NULL, 0, &domainID) < 0) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse PCI domain '%s'"), domain); + goto error; + } + if (virStrToLong_i(bus, NULL, 0, &busID) < 0) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse PCI bus '%s'"), bus); + goto error; + } + if (virStrToLong_i(slot, NULL, 0, &slotID) < 0) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse PCI slot '%s'"), slot); + goto error; + } + if (virStrToLong_i(func, NULL, 0, &funcID) < 0) { + virXendError(conn, VIR_ERR_INTERNAL_ERROR, + _("cannot parse PCI func '%s'"), func); + goto error; + } + + if (VIR_ALLOC(dev) < 0) + goto no_memory; + + dev->mode = VIR_DOMAIN_HOSTDEV_MODE_SUBSYS; + dev->managed = 0; + dev->source.subsys.type = VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI; + dev->source.subsys.u.pci.domain = domainID; + dev->source.subsys.u.pci.bus = busID; + dev->source.subsys.u.pci.slot = slotID; + dev->source.subsys.u.pci.function = funcID; + + if (VIR_REALLOC_N(def->hostdevs, def->nhostdevs+1) < 0) { + goto no_memory; + } + + def->hostdevs[def->nhostdevs++] = dev; + } + + return 0; + +no_memory: + virReportOOMError(conn); + +error: + virDomainHostdevDefFree(dev); + return -1; +} + /** * xenDaemonParseSxpr: @@ -2310,6 +2440,9 @@ xenDaemonParseSxpr(virConnectPtr conn, if (xenDaemonParseSxprNets(conn, def, root) < 0) goto error; + if (xenDaemonParseSxprPCI(conn, def, root) < 0) + goto error; + /* New style graphics device config */ if (xenDaemonParseSxprGraphicsNew(conn, def, root) < 0) goto error; @@ -3953,6 +4086,20 @@ xenDaemonAttachDevice(virDomainPtr domai goto cleanup; break; + case VIR_DOMAIN_DEVICE_HOSTDEV: + if (dev->data.hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + dev->data.hostdev->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) { + if (xenDaemonFormatSxprOnePCI(domain->conn, + dev->data.hostdev, + &buf) < 0) + goto cleanup; + } else { + virXendError(domain->conn, VIR_ERR_NO_SUPPORT, "%s", + _("unsupported device type")); + goto cleanup; + } + break; + default: virXendError(domain->conn, VIR_ERR_NO_SUPPORT, "%s", _("unsupported device type")); @@ -5263,6 +5410,85 @@ xenDaemonFormatSxprNet(virConnectPtr con return 0; } + +static void +xenDaemonFormatSxprPCI(virDomainHostdevDefPtr def, + virBufferPtr buf) +{ + virBufferVSprintf(buf, "(dev (domain 0x%04x)(bus 0x%02x)(slot 0x%02x)(func 0x%x))", + def->source.subsys.u.pci.domain, + def->source.subsys.u.pci.bus, + def->source.subsys.u.pci.slot, + def->source.subsys.u.pci.function); +} + +static int +xenDaemonFormatSxprOnePCI(virConnectPtr conn, + virDomainHostdevDefPtr def, + virBufferPtr buf) +{ + if (def->managed) { + virXendError(conn, VIR_ERR_NO_SUPPORT, "%s", + _("managed PCI devices not supported with XenD")); + return -1; + } + + virBufferAddLit(buf, "(pci "); + xenDaemonFormatSxprPCI(def, buf); + virBufferAddLit(buf, ")"); + + return 0; +} + +static int +xenDaemonFormatSxprAllPCI(virConnectPtr conn, + virDomainDefPtr def, + virBufferPtr buf) +{ + int hasPCI = 0; + int i; + + for (i = 0 ; i < def->nhostdevs ; i++) + if (def->hostdevs[i]->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + def->hostdevs[i]->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) + hasPCI = 1; + + if (!hasPCI) + return 0; + + /* + * With the (domain ...) block we have the following odd setup + * + * (device + * (pci + * (dev (domain 0x0000) (bus 0x00) (slot 0x1b) (func 0x0)) + * (dev (domain 0x0000) (bus 0x00) (slot 0x13) (func 0x0)) + * ) + * ) + * + * Normally there is one (device ...) block per device, but in + * wierd world of Xen PCI, once (device ...) covers multiple + * devices. + */ + + virBufferAddLit(buf, "(device (pci "); + for (i = 0 ; i < def->nhostdevs ; i++) { + if (def->hostdevs[i]->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + def->hostdevs[i]->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) { + if (def->hostdevs[i]->managed) { + virXendError(conn, VIR_ERR_NO_SUPPORT, "%s", + _("managed PCI devices not supported with XenD")); + return -1; + } + + xenDaemonFormatSxprPCI(def->hostdevs[i], buf); + } + } + virBufferAddLit(buf, "))"); + + return 0; +} + int xenDaemonFormatSxprSound(virConnectPtr conn, virDomainDefPtr def, @@ -5529,6 +5755,9 @@ xenDaemonFormatSxpr(virConnectPtr conn, &buf, hvm, xendConfigVersion, 0) < 0) goto error; + if (xenDaemonFormatSxprAllPCI(conn, def, &buf) < 0) + goto error; + /* New style PV graphics config xen >= 3.0.4, * or HVM graphics config xen >= 3.0.5 */ if ((xendConfigVersion >= XEND_CONFIG_MIN_VERS_PVFB_NEWCONF && !hvm) || @@ -5611,6 +5840,9 @@ virDomainXMLDevID(virDomainPtr domain, strncpy(ref, xref, ref_len); free(xref); ref[ref_len - 1] = '\0'; + } else if (dev->type == VIR_DOMAIN_DEVICE_HOSTDEV && + dev->data.hostdev->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + dev->data.hostdev->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) { } else { virXendError(NULL, VIR_ERR_NO_SUPPORT, "%s", _("hotplug of device type not supported")); Index: src/xm_internal.c =================================================================== RCS file: /data/cvs/libvirt/src/xm_internal.c,v retrieving revision 1.117 diff -u -p -r1.117 xm_internal.c --- src/xm_internal.c 5 Feb 2009 16:03:11 -0000 1.117 +++ src/xm_internal.c 16 Mar 2009 11:31:29 -0000 @@ -673,6 +673,7 @@ xenXMDomainConfigParse(virConnectPtr con virDomainDiskDefPtr disk = NULL; virDomainNetDefPtr net = NULL; virDomainGraphicsDefPtr graphics = NULL; + virDomainHostdevDefPtr hostdev = NULL; int i; const char *defaultArch, *defaultMachine; @@ -1115,6 +1116,88 @@ xenXMDomainConfigParse(virConnectPtr con } } + list = virConfGetValue(conf, "pci"); + if (list && list->type == VIR_CONF_LIST) { + list = list->list; + while (list) { + char domain[5]; + char bus[3]; + char slot[3]; + char func[2]; + char *key, *nextkey; + int domainID; + int busID; + int slotID; + int funcID; + + domain[0] = bus[0] = slot[0] = func[0] = '\0'; + + if ((list->type != VIR_CONF_STRING) || (list->str == NULL)) + goto skippci; + + /* pci=['0000:00:1b.0','0000:00:13.0'] */ + key = list->str; + if (!(key = list->str)) + goto skippci; + if (!(nextkey = strchr(key, ':'))) + goto skippci; + + if ((nextkey - key) > (sizeof(domain)-1)) + goto skippci; + + strncpy(domain, key, sizeof(domain)); + domain[sizeof(domain)-1] = '\0'; + + key = nextkey + 1; + if (!(nextkey = strchr(key, ':'))) + goto skippci; + + strncpy(bus, key, sizeof(bus)); + bus[sizeof(bus)-1] = '\0'; + + key = nextkey + 1; + if (!(nextkey = strchr(key, '.'))) + goto skippci; + + strncpy(slot, key, sizeof(slot)); + slot[sizeof(slot)-1] = '\0'; + + key = nextkey + 1; + if (strlen(key) != 1) + goto skippci; + + strncpy(func, key, sizeof(func)); + func[sizeof(func)-1] = '\0'; + + if (virStrToLong_i(domain, NULL, 16, &domainID) < 0) + goto skippci; + if (virStrToLong_i(bus, NULL, 16, &busID) < 0) + goto skippci; + if (virStrToLong_i(slot, NULL, 16, &slotID) < 0) + goto skippci; + if (virStrToLong_i(func, NULL, 16, &funcID) < 0) + goto skippci; + + if (VIR_ALLOC(hostdev) < 0) + goto cleanup; + + hostdev->managed = 0; + hostdev->source.subsys.type = VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI; + hostdev->source.subsys.u.pci.domain = domainID; + hostdev->source.subsys.u.pci.bus = busID; + hostdev->source.subsys.u.pci.slot = slotID; + hostdev->source.subsys.u.pci.function = funcID; + + if (VIR_REALLOC_N(def->hostdevs, def->nhostdevs+1) < 0) + goto no_memory; + def->hostdevs[def->nhostdevs++] = hostdev; + hostdev = NULL; + + skippci: + list = list->next; + } + } + if (hvm) { if (xenXMConfigGetString(conn, conf, "usbdevice", &str, NULL) < 0) goto cleanup; @@ -1939,6 +2022,76 @@ cleanup: +static int +xenXMDomainConfigFormatPCI(virConnectPtr conn, + virConfPtr conf, + virDomainDefPtr def) +{ + + virConfValuePtr pciVal = NULL; + int hasPCI = 0; + int i; + + for (i = 0 ; i < def->nhostdevs ; i++) + if (def->hostdevs[i]->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + def->hostdevs[i]->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) + hasPCI = 1; + + if (!hasPCI) + return 0; + + if (VIR_ALLOC(pciVal) < 0) + return -1; + + pciVal->type = VIR_CONF_LIST; + pciVal->list = NULL; + + for (i = 0 ; i < def->nhostdevs ; i++) { + if (def->hostdevs[i]->mode == VIR_DOMAIN_HOSTDEV_MODE_SUBSYS && + def->hostdevs[i]->source.subsys.type == VIR_DOMAIN_HOSTDEV_SUBSYS_TYPE_PCI) { + virConfValuePtr val, tmp; + char *buf; + + if (virAsprintf(&buf, "%04x:%02x:%02x.%x", + def->hostdevs[i]->source.subsys.u.pci.domain, + def->hostdevs[i]->source.subsys.u.pci.bus, + def->hostdevs[i]->source.subsys.u.pci.slot, + def->hostdevs[i]->source.subsys.u.pci.function) < 0) + goto error; + + if (VIR_ALLOC(val) < 0) { + VIR_FREE(buf); + virReportOOMError(conn); + goto error; + } + val->type = VIR_CONF_STRING; + val->str = buf; + tmp = pciVal->list; + while (tmp && tmp->next) + tmp = tmp->next; + if (tmp) + tmp->next = val; + else + pciVal->list = val; + } + } + + if (pciVal->list != NULL) { + int ret = virConfSetValue(conf, "pci", pciVal); + pciVal = NULL; + if (ret < 0) + return -1; + } + VIR_FREE(pciVal); + + return 0; + +error: + virConfFreeValue(pciVal); + return -1; +} + + virConfPtr xenXMDomainConfigFormat(virConnectPtr conn, virDomainDefPtr def) { virConfPtr conf = NULL; @@ -2260,6 +2413,9 @@ virConfPtr xenXMDomainConfigFormat(virCo } VIR_FREE(netVal); + if (xenXMDomainConfigFormatPCI(conn, conf, def) < 0) + goto cleanup; + if (hvm) { if (def->nparallels) { virBuffer buf = VIR_BUFFER_INITIALIZER; Index: tests/sexpr2xmltest.c =================================================================== RCS file: /data/cvs/libvirt/tests/sexpr2xmltest.c,v retrieving revision 1.36 diff -u -p -r1.36 sexpr2xmltest.c --- tests/sexpr2xmltest.c 29 Jan 2009 17:02:00 -0000 1.36 +++ tests/sexpr2xmltest.c 16 Mar 2009 11:31:29 -0000 @@ -142,6 +142,7 @@ mymain(int argc, char **argv) DO_TEST("net-e1000", "net-e1000", 2); DO_TEST("bridge-ipaddr", "bridge-ipaddr", 3); DO_TEST("no-source-cdrom", "no-source-cdrom", 2); + DO_TEST("pci-devs", "pci-devs", 2); DO_TEST("fv-utc", "fv-utc", 1); DO_TEST("fv-localtime", "fv-localtime", 1); Index: tests/xmconfigtest.c =================================================================== RCS file: /data/cvs/libvirt/tests/xmconfigtest.c,v retrieving revision 1.26 diff -u -p -r1.26 xmconfigtest.c --- tests/xmconfigtest.c 8 Jan 2009 19:52:15 -0000 1.26 +++ tests/xmconfigtest.c 16 Mar 2009 11:31:29 -0000 @@ -231,6 +231,7 @@ mymain(int argc, char **argv) DO_TEST("escape-paths", 2); DO_TEST("no-source-cdrom", 2); + DO_TEST("pci-devs", 2); virCapabilitiesFree(caps); Index: tests/xml2sexprtest.c =================================================================== RCS file: /data/cvs/libvirt/tests/xml2sexprtest.c,v retrieving revision 1.34 diff -u -p -r1.34 xml2sexprtest.c --- tests/xml2sexprtest.c 23 Jan 2009 01:48:47 -0000 1.34 +++ tests/xml2sexprtest.c 16 Mar 2009 11:31:29 -0000 @@ -128,6 +128,7 @@ mymain(int argc, char **argv) DO_TEST("net-e1000", "net-e1000", "pvtest", 2); DO_TEST("bridge-ipaddr", "bridge-ipaddr", "pvtest", 2); DO_TEST("no-source-cdrom", "no-source-cdrom", "test", 2); + DO_TEST("pci-devs", "pci-devs", "pvtest", 2); DO_TEST("fv-utc", "fv-utc", "fvtest", 1); DO_TEST("fv-localtime", "fv-localtime", "fvtest", 1); Index: tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr =================================================================== RCS file: tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr diff -N tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/sexpr2xmldata/sexpr2xml-pci-devs.sexpr 16 Mar 2009 11:31:29 -0000 @@ -0,0 +1,2 @@ +(domain (domid 6)(name 'pvtest')(memory 420)(maxmem 420)(vcpus 2)(uuid '596a5d2171f48fb2e068e2386a5c413e')(on_poweroff 'destroy')(on_reboot 'destroy')(on_crash 'destroy')(image (linux (kernel '/var/lib/xen/vmlinuz.2Dn2YT')(ramdisk '/var/lib/xen/initrd.img.0u-Vhq')(args ' method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… ')))(device (pci (backend 0)(dev (domain 0x0001) (bus 0x0c) (slot 0x1b) (func 0x2))(dev (domain 0x0000) (bus 0x01) (slot 0x13) (func 0x0))))(device (vbd (dev 'xvda')(uname 'phy:/dev/MainVG/GuestVG')(mode 'w')))) + Index: tests/sexpr2xmldata/sexpr2xml-pci-devs.xml =================================================================== RCS file: tests/sexpr2xmldata/sexpr2xml-pci-devs.xml diff -N tests/sexpr2xmldata/sexpr2xml-pci-devs.xml --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/sexpr2xmldata/sexpr2xml-pci-devs.xml 16 Mar 2009 11:31:29 -0000 @@ -0,0 +1,37 @@ +<domain type='xen' id='6'> + <name>pvtest</name> + <uuid>596a5d21-71f4-8fb2-e068-e2386a5c413e</uuid> + <memory>430080</memory> + <currentMemory>430080</currentMemory> + <vcpu>2</vcpu> + <os> + <type>linux</type> + <kernel>/var/lib/xen/vmlinuz.2Dn2YT</kernel> + <initrd>/var/lib/xen/initrd.img.0u-Vhq</initrd> + <cmdline> method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… </cmdline> + </os> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>destroy</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <disk type='block' device='disk'> + <driver name='phy'/> + <source dev='/dev/MainVG/GuestVG'/> + <target dev='xvda' bus='xen'/> + </disk> + <console type='pty'> + <target port='0'/> + </console> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0001' bus='0x0c' slot='0x1b' function='0x2'/> + </source> + </hostdev> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0000' bus='0x01' slot='0x13' function='0x0'/> + </source> + </hostdev> + </devices> +</domain> Index: tests/xmconfigdata/test-pci-devs.cfg =================================================================== RCS file: tests/xmconfigdata/test-pci-devs.cfg diff -N tests/xmconfigdata/test-pci-devs.cfg --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/xmconfigdata/test-pci-devs.cfg 16 Mar 2009 11:31:29 -0000 @@ -0,0 +1,24 @@ +name = "test" +uuid = "cc2315e7-d26a-307a-438c-6d188ec4c09c" +maxmem = 382 +memory = 350 +vcpus = 1 +builder = "hvm" +kernel = "/usr/lib/xen/boot/hvmloader" +boot = "c" +pae = 1 +acpi = 1 +apic = 1 +localtime = 0 +on_poweroff = "destroy" +on_reboot = "destroy" +on_crash = "destroy" +device_model = "/usr/lib/xen/bin/qemu-dm" +sdl = 0 +vnc = 1 +vncunused = 1 +disk = [ "phy:/dev/sda8,hda,w", ",hdc:cdrom,r" ] +vif = [ "mac=00:16:3e:0a:7b:39,bridge=xenbr0,type=ioemu" ] +pci = [ "0001:0c:1b.2", "0000:01:13.0" ] +parallel = "none" +serial = "pty" Index: tests/xmconfigdata/test-pci-devs.xml =================================================================== RCS file: tests/xmconfigdata/test-pci-devs.xml diff -N tests/xmconfigdata/test-pci-devs.xml --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/xmconfigdata/test-pci-devs.xml 16 Mar 2009 11:31:29 -0000 @@ -0,0 +1,56 @@ +<domain type='xen'> + <name>test</name> + <uuid>cc2315e7-d26a-307a-438c-6d188ec4c09c</uuid> + <memory>391168</memory> + <currentMemory>358400</currentMemory> + <vcpu>1</vcpu> + <os> + <type arch='i686' machine='xenfv'>hvm</type> + <loader>/usr/lib/xen/boot/hvmloader</loader> + <boot dev='hd'/> + </os> + <features> + <acpi/> + <apic/> + <pae/> + </features> + <clock offset='utc'/> + <on_poweroff>destroy</on_poweroff> + <on_reboot>destroy</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <emulator>/usr/lib/xen/bin/qemu-dm</emulator> + <disk type='block' device='disk'> + <driver name='phy'/> + <source dev='/dev/sda8'/> + <target dev='hda' bus='ide'/> + </disk> + <disk type='block' device='cdrom'> + <driver name='phy'/> + <target dev='hdc' bus='ide'/> + <readonly/> + </disk> + <interface type='bridge'> + <mac address='00:16:3e:0a:7b:39'/> + <source bridge='xenbr0'/> + </interface> + <serial type='pty'> + <target port='0'/> + </serial> + <console type='pty'> + <target port='0'/> + </console> + <input type='mouse' bus='ps2'/> + <graphics type='vnc' port='-1' autoport='yes'/> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0001' bus='0x0c' slot='0x1b' function='0x2'/> + </source> + </hostdev> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0000' bus='0x01' slot='0x13' function='0x0'/> + </source> + </hostdev> + </devices> +</domain> Index: tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr =================================================================== RCS file: tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr diff -N tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/xml2sexprdata/xml2sexpr-pci-devs.sexpr 16 Mar 2009 11:31:29 -0000 @@ -0,0 +1 @@ +(vm (name 'pvtest')(memory 420)(maxmem 420)(vcpus 2)(uuid '596a5d21-71f4-8fb2-e068-e2386a5c413e')(on_poweroff 'destroy')(on_reboot 'destroy')(on_crash 'destroy')(image (linux (kernel '/var/lib/xen/vmlinuz.2Dn2YT')(ramdisk '/var/lib/xen/initrd.img.0u-Vhq')(args ' method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… ')))(device (vbd (dev 'xvda')(uname 'phy:/dev/MainVG/GuestLV')(mode 'w')))(device (pci (dev (domain 0x0001)(bus 0x0c)(slot 0x1b)(func 0x2))(dev (domain 0x0000)(bus 0x01)(slot 0x13)(func 0x0))))) \ No newline at end of file Index: tests/xml2sexprdata/xml2sexpr-pci-devs.xml =================================================================== RCS file: tests/xml2sexprdata/xml2sexpr-pci-devs.xml diff -N tests/xml2sexprdata/xml2sexpr-pci-devs.xml --- /dev/null 1 Jan 1970 00:00:00 -0000 +++ tests/xml2sexprdata/xml2sexpr-pci-devs.xml 16 Mar 2009 11:31:29 -0000 @@ -0,0 +1,33 @@ +<domain type='xen' id='15'> + <name>pvtest</name> + <uuid>596a5d2171f48fb2e068e2386a5c413e</uuid> + <os> + <type>linux</type> + <kernel>/var/lib/xen/vmlinuz.2Dn2YT</kernel> + <initrd>/var/lib/xen/initrd.img.0u-Vhq</initrd> + <cmdline> method=http://download.fedora.devel.redhat.com/pub/fedora/linux/core/test/5… </cmdline> + </os> + <memory>430080</memory> + <vcpu>2</vcpu> + <on_poweroff>destroy</on_poweroff> + <on_reboot>destroy</on_reboot> + <on_crash>destroy</on_crash> + <devices> + <disk type='block' device='disk'> + <source dev='/dev/MainVG/GuestLV'/> + <target dev='xvda'/> + </disk> + <console tty='/dev/pts/4'/> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0001' bus='0x0c' slot='0x1b' function='0x2'/> + </source> + </hostdev> + <hostdev mode='subsystem' type='pci' managed='no'> + <source> + <address domain='0x0000' bus='0x01' slot='0x13' function='0x0'/> + </source> + </hostdev> + </devices> +</domain> + -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] Problem with the current svirt patch
by Daniel J Walsh 17 Mar '09

17 Mar '09

The current svirt patch relabels all disk to the image_t:MCS, which is incorrect. Read Only Disks and Sharable Disks should not be labeled. Also when libvirt is completed running the image it needs to relabel the image back to something sane. Right now it is labeling everything imagelabel:s0, including phisical disk partitions. I considered two ways of labeling the "disk" back. We can either grab the label when libvirt starts and change it back to this label when ever an image completes or we can ask the system what the label should be. (matcpathcon). I originally coded up the first, but quickly realized if anything went wrong with libvirt labeling like a crash, the labels on disk could be wrong. And libvirt would continuously set them to this wrong label. With matchpathcon, libvirt will at least set them to something sane. So this patch Removes labeling of readonly and shared disks and restores the images label to the system default when the image completes. I would really like to get this in ASAP. Since currently libvirt is relabeing the cdrom to virt_image_t when it is complete as well as physical disks.

2 4

[libvirt] [PATCH] leak of file descriptors in remote code
by Daniel Veillard 17 Mar '09

17 Mar '09

We never deallocate the pipe used to do the wake-up trick, loosing 2 descriptor per connection opened in a process, Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 2

[libvirt] [RFC][PATCH] storage: allow alphabetical names in owner/group of permissions
by Ryota Ozaki 16 Mar '09

16 Mar '09

Hi, This patch allows to specify owner/group of permissions in storage XML as alphabetical names. One concern of this patch is since alphabetical names are converted to uid/gid at XML parsing (e.g., virsh pool-define some-pool.xml), the generated XML file for libvirt internal does not contain the original alphabetical names and dumpxml does not as well. Is it better the conversion is done when activated not defined? And I know I need to modify documents and write test XMLs. Any comments are welcome! Thanks, ozaki-r >From bf8a01689e345ec9c08a1348b5781c51b6016ef4 Mon Sep 17 00:00:00 2001 From: Ryota Ozaki <ozaki.ryota(a)gmail.com> Date: Sun, 15 Mar 2009 02:30:02 +0900 Subject: [PATCH] storage: allow alphabetical names in owner/group of permissions --- configure.in | 2 +- src/storage_conf.c | 50 ++++++++++++++++++++++++++++++++++++--------- src/util.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++++++- src/util.h | 8 +++++++ 4 files changed, 105 insertions(+), 12 deletions(-) diff --git a/configure.in b/configure.in index 413d27c..0277d9a 100644 --- a/configure.in +++ b/configure.in @@ -75,7 +75,7 @@ dnl Availability of various common functions (non-fatal if missing). AC_CHECK_FUNCS([cfmakeraw regexec uname sched_getaffinity getuid getgid]) dnl Availability of various not common threadsafe functions -AC_CHECK_FUNCS([strerror_r strtok_r getmntent_r getgrnam_r getpwuid_r]) +AC_CHECK_FUNCS([strerror_r strtok_r getmntent_r getgrnam_r getpwuid_r getpwnam_r]) dnl Availability of various common headers (non-fatal if missing). AC_CHECK_HEADERS([pwd.h paths.h regex.h sys/syslimits.h sys/utsname.h sys/wait.h winsock2.h sched.h termios.h sys/poll.h syslog.h]) diff --git a/src/storage_conf.c b/src/storage_conf.c index 1c9a4e5..7d75b71 100644 --- a/src/storage_conf.c +++ b/src/storage_conf.c @@ -412,23 +412,53 @@ virStorageDefParsePerms(virConnectPtr conn, if (virXPathNode(conn, "./owner", ctxt) == NULL) { perms->uid = getuid(); } else { - if (virXPathLong(conn, "number(./owner)", ctxt, &v) < 0) { - virStorageReportError(conn, VIR_ERR_XML_ERROR, - "%s", _("malformed owner element")); - goto error; + if (virXPathLong(conn, "number(./owner)", ctxt, &v) == 0) { + perms->uid = (int)v; + } else { + char *name; + int uid; + + name = virXPathString(conn, "string(./owner)", ctxt); + if (!name) { + virStorageReportError(conn, VIR_ERR_XML_ERROR, + "%s", _("malformed owner element")); + goto error; + } + + uid = virGetUIDByUsername(conn, name); + VIR_FREE(name); + if (uid < 0) { + goto error; + } + + perms->uid = uid; } - perms->uid = (int)v; } if (virXPathNode(conn, "./group", ctxt) == NULL) { perms->gid = getgid(); } else { - if (virXPathLong(conn, "number(./group)", ctxt, &v) < 0) { - virStorageReportError(conn, VIR_ERR_XML_ERROR, - "%s", _("malformed group element")); - goto error; + if (virXPathLong(conn, "number(./group)", ctxt, &v) == 0) { + perms->gid = (int)v; + } else { + char *name; + int gid; + + name = virXPathString(conn, "string(./group)", ctxt); + if (!name) { + virStorageReportError(conn, VIR_ERR_XML_ERROR, + "%s", _("malformed group element")); + goto error; + } + + gid = virGetGIDByGroupname(conn, name); + VIR_FREE(name); + if (gid < 0) { + goto error; + } + + perms->gid = gid; } - perms->gid = (int)v; } /* NB, we're ignoring missing labels here - they'll simply inherit */ diff --git a/src/util.c b/src/util.c index 9b74757..fbe329d 100644 --- a/src/util.c +++ b/src/util.c @@ -50,9 +50,12 @@ #include <paths.h> #endif #include <netdb.h> -#ifdef HAVE_GETPWUID_R +#if defined(HAVE_GETPWUID_R) || defined(HAVE_GETPWNAM_R) #include <pwd.h> #endif +#ifdef HAVE_GETGRNAM_R +#include <grp.h> +#endif #include "virterror_internal.h" #include "logging.h" @@ -1558,3 +1561,55 @@ char *virGetUserDirectory(virConnectPtr conn, return ret; } #endif + + +#ifdef HAVE_GETPWNAM_R +int virGetUIDByUsername(virConnectPtr conn, char *name) +{ + struct passwd pwbuf; + struct passwd *pw = NULL; + char *strbuf; + size_t strbuflen = sysconf(_SC_GETPW_R_SIZE_MAX); + + if (VIR_ALLOC_N(strbuf, strbuflen) < 0) { + virReportOOMError(conn); + return -1; + } + + if (getpwnam_r(name, &pwbuf, strbuf, strbuflen, &pw) != 0 || pw == NULL) { + virReportSystemError(conn, errno, + _("Failed to find user record for name '%s'"), + name); + VIR_FREE(strbuf); + return -1; + } + VIR_FREE(strbuf); + return pw->pw_uid; +} +#endif + + +#ifdef HAVE_GETGRNAM_R +int virGetGIDByGroupname(virConnectPtr conn, char *name) +{ + struct group grbuf; + struct group *gr = NULL; + char *strbuf; + size_t strbuflen = sysconf(_SC_GETGR_R_SIZE_MAX); + + if (VIR_ALLOC_N(strbuf, strbuflen) < 0) { + virReportOOMError(conn); + return -1; + } + + if (getgrnam_r(name, &grbuf, strbuf, strbuflen, &gr) != 0 || gr == NULL) { + virReportSystemError(conn, errno, + _("Failed to find group record for name '%s'"), + name); + VIR_FREE(strbuf); + return -1; + } + VIR_FREE(strbuf); + return gr->gr_gid; +} +#endif diff --git a/src/util.h b/src/util.h index 87cbf67..ffd3458 100644 --- a/src/util.h +++ b/src/util.h @@ -195,6 +195,14 @@ char *virGetUserDirectory(virConnectPtr conn, uid_t uid); #endif +#ifdef HAVE_GETPWNAM_R +int virGetUIDByUsername(virConnectPtr conn, char *name); +#endif + +#ifdef HAVE_GETGRNAM_R +int virGetGIDByGroupname(virConnectPtr conn, char *name); +#endif + int virRandomInitialize(unsigned int seed); int virRandom(int max); -- 1.6.0.6

3 3

[libvirt] The gui should drop the connection if libvirt stops for any reason like service libvirt stop.
by Daniel J Walsh 16 Mar '09

16 Mar '09

diff -u virtManager/engine.py~ virtManager/engine.py --- virtManager/engine.py~ 2009-03-16 14:49:16.000000000 -0400 +++ virtManager/engine.py 2009-03-16 14:58:05.000000000 -0400 @@ -158,6 +158,7 @@ logging.error(("Could not refresh connection %s\n" % (uri)) + str(sys.exc_info()[0]) + \ " " + str(sys.exc_info()[1]) + "\n" + \ traceback.format_exc(sys.exc_info()[2])) + self.connections[uri]["connection"].close() return 1 def change_timer_interval(self,ignore1,ignore2,ignore3,ignore4):

1 0

[libvirt] [PATCH] Don't allow read only connection to use SECURE_XML
by Cole Robinson 16 Mar '09

16 Mar '09

Seems to me that a read only connection shouldn't be able to dump domain xml with the SECURE flag. Attached patch blocks the attempt with an explicit error message. Thanks, Cole

3 3

[libvirt] [PATCH] Fix test driver domain save
by Cole Robinson 16 Mar '09

16 Mar '09

Saving a domain via the test driver currently locks up, since the save command is calling out to another public driver method. The attached patch fixes this. Thanks, Cole

3 3

[libvirt] [PATCH] Parse qemu config before SecurityInit
by Cole Robinson 16 Mar '09

16 Mar '09

We aren't parsing qemu.conf before initializing the security driver, yet the config contains relevant security options. The attached patch swaps the ordering, which fixes things for me. Thanks, Cole

3 3

[libvirt] shit-off domains
by Zvi Dubitzky 16 Mar '09

16 Mar '09

I do not see a way to get the shut-off domain list from a libvirt API (case when no id shown under virsh , just domain name) Am I missing something ? thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

[libvirt] [PATCH] fix Xen event handling
by Daniel Veillard 16 Mar '09

16 Mar '09

As was reported on IRC and found by Dan Berrange, sometimes Xen event handling could start to go wild and block processing of requests in the daemon. The fault at least on libvirt side is that we didn't filter out non read/write event requests when asking for watches in the xenstore. The provided patch seems to work for the person who reported the original problem, Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

3 6

[libvirt] libvirt-cim
by Zvi Dubitzky 16 Mar '09

16 Mar '09

Hi can someone tell me what is libvirt-cim . Does it mean libvirt utilized by CIM or there is a special 'dialect' called livirt-cim thanks Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

2 1

[libvirt] [PATCH] trying to close some potential FILE * leaks
by Daniel Veillard 16 Mar '09

16 Mar '09

While reviewing the code due to what seems to be a FILE * leak I found the following problems, probably not what got me in troubles but should still be fixed, Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

2 1

[libvirt] qemu-kvm networking speed
by Bryan_Coleman＠dart.biz 13 Mar '09

13 Mar '09

How do I change the networking speed provided to my qemu-kvm guest? I have four gigabit ports bonded; however, my guest is showing just 100Mb/s. I have used iperf to confirm. I am not sure if the settings would be on the guest or server side. Any help would be greatly appreciated! Thank you, Bryan

3 7

[libvirt] PATCH: Don't try to close unopened connections
by Daniel P. Berrange 13 Mar '09

13 Mar '09

The autostart code for UML and QEMU drivers had a place where it might try to close a connection that is not open. This is harmless because virConnectClose() checks for NULL, but it still causes an annoying diagnosistic to be printed. This avoids it Daniel Index: src/uml_driver.c =================================================================== RCS file: /data/cvs/libvirt/src/uml_driver.c,v retrieving revision 1.25 diff -u -p -r1.25 uml_driver.c --- src/uml_driver.c 3 Mar 2009 09:14:28 -0000 1.25 +++ src/uml_driver.c 13 Mar 2009 16:29:33 -0000 @@ -146,7 +146,8 @@ umlAutostartConfigs(struct uml_driver *d } } - virConnectClose(conn); + if (conn) + virConnectClose(conn); } Index: src/qemu_driver.c =================================================================== RCS file: /data/cvs/libvirt/src/qemu_driver.c,v retrieving revision 1.214 diff -u -p -r1.214 qemu_driver.c --- src/qemu_driver.c 3 Mar 2009 16:53:13 -0000 1.214 +++ src/qemu_driver.c 13 Mar 2009 16:29:34 -0000 @@ -235,7 +235,8 @@ qemudAutostartConfigs(struct qemud_drive virDomainObjUnlock(vm); } - virConnectClose(conn); + if (conn) + virConnectClose(conn); } -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] libvirt usage
by Zvi Dubitzky 13 Mar '09

13 Mar '09

Hi I have a Ubutnu system . I installed libxml2 and libvirt packages by : apt-get install ibvirt-bin and .. libxml2 . They were installed ok . The libraries are located under /user/lib e.g : /usr/lib/libvirt.so.0 I am trying to link as follows: CFLAGS=`pkg-config --cflags libvirt` LDFLAGS=`pkg-config --libs libvirt` gcc -o test_cpu $CFLAGS $LDFLAGS test_cpu.c But get the message : Package libvirt was not found in the pkg-config search path. Perhaps you should add the directory containing `libvirt.pc' to the PKG_CONFIG_PATH environment variable No package 'libvirt' found It turns out that doing : pkg-config --list-all | grep libvirt return no libvirt traces . just as in the file below . On another older machine that links fine - libvirt is found with the above pkg-config but there is no PKG_CONFIG_PATH env variable as well . Both machines PATH is the same . Did I miss something during the install of the first machine. Any remedy ? Zvi Dubitzky Virtualization and System Architecture Email:dubi@il.ibm.com IBM Haifa Research Laboratory Phone: +972-4-8296182 Haifa, 31905, ISRAEL

3 2

[libvirt] [PATCH][take2] storage: fix disregarding specified owner/group permissions
by Ryota Ozaki 13 Mar '09

13 Mar '09

Hi, The changes since the last patch are, - pass a default value of 'mode' permission to virStorageDefParsePerms - fix regression - I missed setting permission values in the case of that there is not <permissions> element Thanks, ozaki-r

2 2

[libvirt] PATCH: Fix node device APIs for RHEL-5 vintage HAL
by Daniel P. Berrange 12 Mar '09

12 Mar '09

In the node device APIs we look for a property in HAL call info.subsystem. Turns out this is a fairly recent addition to HAL, and it used to be called 'linux.subsystem' in RHEL-5 vintage & earlier. This means we never see any PCI / USB devices in libvirt when on RHEL-5. This patch makes us check for both those properties, since the values are the same for each - just the property name changed Daniel Index: src/node_device_hal.c =================================================================== RCS file: /data/cvs/libvirt/src/node_device_hal.c,v retrieving revision 1.9 diff -u -p -r1.9 node_device_hal.c --- src/node_device_hal.c 2 Mar 2009 17:41:13 -0000 1.9 +++ src/node_device_hal.c 12 Mar 2009 14:07:12 -0000 @@ -357,7 +357,8 @@ static int gather_capabilities(LibHalCon goto failure; } - if (get_str_prop(ctx, udi, "info.subsystem", &bus_name) == 0) { + if (get_str_prop(ctx, udi, "info.subsystem", &bus_name) == 0 || + get_str_prop(ctx, udi, "linux.subsystem", &bus_name) == 0) { rv = gather_capability(ctx, udi, bus_name, &caps); if (rv != 0) goto failure; -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] xen+ssh:// "ghost" connections
by Nick Moffitt 12 Mar '09

12 Mar '09

Many thanks to Daniel Veillard for the Xen storage patch that was ACKed here yesterday! It does indeed seem to fix the crashing and lockup problems I was seeingw hen performing many simultaneous operations. At the moment, it may have exposed something else I find curious. I have a set of twelve VMs across two hosts (seven on one, five on the other). At night they're all paused, and in the morning they're all resumed at once. For the past two evenings, the pause operation has stalled for two of the VMs. The first time it was one per host, and just yesterday it was both on the 7-domain host. These connections are mostly harmless. The libvirtd still accepts new connections, but I can see the stalled ssh/netcat processes still running to forward the local socket. I've attached (okay, pasted inline) a traceback of the daemon that had two ghost connections, and from my inexperienced read-through it looks as though libvirtd doesn't even know that anyone's trying to talk to it. This isn't much to go on, but I felt it was better to post this than to let it drop on the floor, since there does seem to be a pattern so far (we'll see how things go tonight, and I'll be able to do more destructive testing next week). (gdb) thread apply all bt Thread 6 (Thread 0x7f5ca3af7950 (LWP 889)): #0 0x00007f5ca84dab99 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0 #1 0x00007f5ca9b43369 in virCondWait (c=0x65a63c, m=0x80) at threads-pthread.c:81 #2 0x0000000000412ab5 in qemudWorker (data=<value optimized out>) at qemud.c:1445 #3 0x00007f5ca84d63f7 in start_thread () from /lib/libpthread.so.0 #4 0x00007f5ca8245b3d in clone () from /lib/libc.so.6 #5 0x0000000000000000 in ?? () Thread 5 (Thread 0x7f5c9bfff950 (LWP 28425)): #0 0x00007f5ca84dab99 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0 #1 0x00007f5ca9b43369 in virCondWait (c=0x65a63c, m=0x80) at threads-pthread.c:81 #2 0x0000000000412ab5 in qemudWorker (data=<value optimized out>) at qemud.c:1445 #3 0x00007f5ca84d63f7 in start_thread () from /lib/libpthread.so.0 #4 0x00007f5ca8245b3d in clone () from /lib/libc.so.6 #5 0x0000000000000000 in ?? () Thread 4 (Thread 0x7f5ca52fa950 (LWP 5493)): #0 0x00007f5ca84dab99 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0 #1 0x00007f5ca9b43369 in virCondWait (c=0x65a63c, m=0x80) at threads-pthread.c:81 #2 0x0000000000412ab5 in qemudWorker (data=<value optimized out>) at qemud.c:1445 #3 0x00007f5ca84d63f7 in start_thread () from /lib/libpthread.so.0 #4 0x00007f5ca8245b3d in clone () from /lib/libc.so.6 #5 0x0000000000000000 in ?? () Thread 3 (Thread 0x7f5ca4af9950 (LWP 25795)): #0 0x00007f5ca84dab99 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0 #1 0x00007f5ca9b43369 in virCondWait (c=0x65a63c, m=0x80) at threads-pthread.c:81 #2 0x0000000000412ab5 in qemudWorker (data=<value optimized out>) at qemud.c:1445 #3 0x00007f5ca84d63f7 in start_thread () from /lib/libpthread.so.0 #4 0x00007f5ca8245b3d in clone () from /lib/libc.so.6 #5 0x0000000000000000 in ?? () Thread 2 (Thread 0x7f5ca2af5950 (LWP 25810)): #0 0x00007f5ca84dab99 in pthread_cond_wait@@GLIBC_2.3.2 () from /lib/libpthread.so.0 #1 0x00007f5ca9b43369 in virCondWait (c=0x65a63c, m=0x80) at threads-pthread.c:81 #2 0x0000000000412ab5 in qemudWorker (data=<value optimized out>) at qemud.c:1445 #3 0x00007f5ca84d63f7 in start_thread () from /lib/libpthread.so.0 #4 0x00007f5ca8245b3d in clone () from /lib/libc.so.6 #5 0x0000000000000000 in ?? () Thread 1 (Thread 0x7f5caa853780 (LWP 864)): #0 0x00007f5ca823cc86 in poll () from /lib/libc.so.6 #1 0x000000000040f1d6 in virEventRunOnce () at event.c:542 #2 0x00000000004116cd in qemudRunLoop (server=0x65a610) at qemud.c:2067 #3 0x0000000000414bbe in main (argc=4, argv=<value optimized out>) at qemud.c:2921 #0 0x00007f5ca823cc86 in poll () from /lib/libc.so.6 -- BitKeeper, how quaint. Nick Moffitt -- Alan Cox nick(a)zork.net

1 0

[libvirt] Looking for good example API function(s)
by Laine Stump 11 Mar '09

11 Mar '09

To help with lutter's network configuration project, I'm looking at adding API functions to libvirt which are essentially pass-throughs to netcf library APIs called by libvirtd. Can someone point me to good examples of APIs that are not hypervisor driver-dependent (and possibly not host OS-dependent either - I believe the final intent is for netcf to handle all host OS dependencies itself) to use as a basis for writing these new APIs? At a more detailed level, a quick trace through the code of some other APIs got me as far as the call to the "call" function, with an RPC code, but I didn't find where that call was caught (and the RPC code interpreted, eg REMOTE_PROC_NETWORK_CREATE_XML - cscope couldn't find any other reference to this value) on the libvirtd side. I'm about run virsh and libvirtd under gdb to learn this by experimentation, but if someone wants to take the fun out of it for me, hints would be gladly accepted! ;-)

4 3

[libvirt] PATCH: Support SASL auth for QEMU VNC servers
by Daniel P. Berrange 11 Mar '09

11 Mar '09

This patch adds a config option to /etc/libvirt/qemu.conf to allow the enablement of SASL authentication of the QEMU VNC servers. The patches to QEMU are posted upstream, and though their merge is delayed while the QEMU release is sorted out, I expect them to be merged in a week or so. I'm posting this patch for review now, though best not to commit it to CVS until the QEMU stuff is finally merged. Daniel diff -r 961d4b1ca1d3 qemud/libvirtd_qemu.aug --- a/qemud/libvirtd_qemu.aug Wed Mar 04 13:17:44 2009 +0000 +++ b/qemud/libvirtd_qemu.aug Thu Mar 05 14:22:50 2009 +0000 @@ -27,6 +27,8 @@ module Libvirtd_qemu = | str_entry "vnc_tls_x509_cert_dir" | bool_entry "vnc_tls_x509_verify" | str_entry "vnc_password" + | bool_entry "vnc_sasl" + | str_entry "vnc_sasl_dir" (* Each enty in the config is one of the following three ... *) let entry = vnc_entry diff -r 961d4b1ca1d3 qemud/test_libvirtd_qemu.aug --- a/qemud/test_libvirtd_qemu.aug Wed Mar 04 13:17:44 2009 +0000 +++ b/qemud/test_libvirtd_qemu.aug Thu Mar 05 14:22:50 2009 +0000 @@ -60,6 +60,25 @@ vnc_tls_x509_verify = 1 # example here before you set this # vnc_password = \"XYZ12345\" + + +# Enable use of SASL encryption on the VNC server. This requires +# a VNC client which supports the SASL protocol extension. +# Examples include vinagre, virt-viewer and virt-manager +# itself. UltraVNC, RealVNC, TightVNC do not support this +# +# It is necessary to configure /etc/sasl2/qemu.conf to choose +# the desired SASL plugin (eg, GSSPI for Kerberos) +# +vnc_sasl = 1 + + +# The default SASL configuration file is located in /etc/sasl2/ +# When running libvirtd unprivileged, it may be desirable to +# override the configs in this location. Set this parameter to +# point to the directory, and create a qemu.conf in that location +# +vnc_sasl_dir = \"/some/directory/sasl2\" " test Libvirtd_qemu.lns get conf = @@ -123,3 +142,22 @@ vnc_password = \"XYZ12345\" { "#comment" = "example here before you set this" } { "#comment" = "" } { "vnc_password" = "XYZ12345" } +{ "#empty" } +{ "#empty" } +{ "#comment" = "Enable use of SASL encryption on the VNC server. This requires" } +{ "#comment" = "a VNC client which supports the SASL protocol extension." } +{ "#comment" = "Examples include vinagre, virt-viewer and virt-manager" } +{ "#comment" = "itself. UltraVNC, RealVNC, TightVNC do not support this" } +{ "#comment" = "" } +{ "#comment" = "It is necessary to configure /etc/sasl2/qemu.conf to choose" } +{ "#comment" = "the desired SASL plugin (eg, GSSPI for Kerberos)" } +{ "#comment" = "" } +{ "vnc_sasl" = "1" } +{ "#empty" } +{ "#empty" } +{ "#comment" = "The default SASL configuration file is located in /etc/sasl2/" } +{ "#comment" = "When running libvirtd unprivileged, it may be desirable to" } +{ "#comment" = "override the configs in this location. Set this parameter to" } +{ "#comment" = "point to the directory, and create a qemu.conf in that location" } +{ "#comment" = "" } +{ "vnc_sasl_dir" = "/some/directory/sasl2" } diff -r 961d4b1ca1d3 src/qemu.conf --- a/src/qemu.conf Wed Mar 04 13:17:44 2009 +0000 +++ b/src/qemu.conf Thu Mar 05 14:22:50 2009 +0000 @@ -60,6 +60,27 @@ # vnc_password = "XYZ12345" +# Enable use of SASL encryption on the VNC server. This requires +# a VNC client which supports the SASL protocol extension. +# Examples include vinagre, virt-viewer and virt-manager +# itself. UltraVNC, RealVNC, TightVNC do not support this +# +# It is necessary to configure /etc/sasl2/qemu.conf to choose +# the desired SASL plugin (eg, GSSPI for Kerberos) +# +# vnc_sasl = 1 + + +# The default SASL configuration file is located in /etc/sasl2/ +# When running libvirtd unprivileged, it may be desirable to +# override the configs in this location. Set this parameter to +# point to the directory, and create a qemu.conf in that location +# +# vnc_sasl_dir = "/some/directory/sasl2" + + + + # The default security driver is SELinux. If SELinux is disabled # on the host, then the security driver will automatically disable # itself. If you wish to disable QEMU SELinux security driver while diff -r 961d4b1ca1d3 src/qemu_conf.c --- a/src/qemu_conf.c Wed Mar 04 13:17:44 2009 +0000 +++ b/src/qemu_conf.c Thu Mar 05 14:22:50 2009 +0000 @@ -161,6 +161,21 @@ int qemudLoadDriverConfig(struct qemud_d } } + p = virConfGetValue (conf, "vnc_sasl"); + CHECK_TYPE ("vnc_sasl", VIR_CONF_LONG); + if (p) driver->vncSASL = p->l; + + p = virConfGetValue (conf, "vnc_sasl_dir"); + CHECK_TYPE ("vnc_sasl_dir", VIR_CONF_STRING); + if (p && p->str) { + VIR_FREE(driver->vncSASLdir); + if (!(driver->vncSASLdir = strdup(p->str))) { + virReportOOMError(NULL); + virConfFree(conf); + return -1; + } + } + virConfFree (conf); return 0; } @@ -838,15 +853,20 @@ int qemudBuildCommandLine(virConnectPtr goto no_memory; \ } while (0) +#define ADD_ENV_PAIR(envname, val) \ + do { \ + char *envval; \ + ADD_ENV_SPACE; \ + if (virAsprintf(&envval, "%s=%s", envname, val) < 0) \ + goto no_memory; \ + qenv[qenvc++] = envval; \ + } while (0) + #define ADD_ENV_COPY(envname) \ do { \ char *val = getenv(envname); \ - char *envval; \ - ADD_ENV_SPACE; \ if (val != NULL) { \ - if (virAsprintf(&envval, "%s=%s", envname, val) < 0) \ - goto no_memory; \ - qenv[qenvc++] = envval; \ + ADD_ENV_PAIR(envname, val); \ } \ } while (0) @@ -1295,6 +1315,15 @@ int qemudBuildCommandLine(virConnectPtr driver->vncTLSx509certdir); } } + + if (driver->vncSASL) { + virBufferAddLit(&opt, ",sasl"); + + if (driver->vncSASLdir) + ADD_ENV_PAIR("SASL_CONF_DIR", driver->vncSASLdir); + + /* TODO: Support ACLs later */ + } } else { virBufferVSprintf(&opt, "%d", vm->def->graphics->data.vnc.port - 5900); diff -r 961d4b1ca1d3 src/qemu_conf.h --- a/src/qemu_conf.h Wed Mar 04 13:17:44 2009 +0000 +++ b/src/qemu_conf.h Thu Mar 05 14:22:50 2009 +0000 @@ -73,9 +73,11 @@ struct qemud_driver { char *stateDir; unsigned int vncTLS : 1; unsigned int vncTLSx509verify : 1; + unsigned int vncSASL : 1; char *vncTLSx509certdir; char *vncListen; char *vncPassword; + char *vncSASLdir; virCapsPtr caps; diff -r 961d4b1ca1d3 src/qemu_driver.c --- a/src/qemu_driver.c Wed Mar 04 13:17:44 2009 +0000 +++ b/src/qemu_driver.c Thu Mar 05 14:22:50 2009 +0000 @@ -620,6 +620,7 @@ qemudShutdown(void) { VIR_FREE(qemu_driver->vncTLSx509certdir); VIR_FREE(qemu_driver->vncListen); VIR_FREE(qemu_driver->vncPassword); + VIR_FREE(qemu_driver->vncSASLdir); /* Free domain callback list */ virDomainEventCallbackListFree(qemu_driver->domainEventCallbacks); diff -r 961d4b1ca1d3 tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-sasl.args --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-sasl.args Thu Mar 05 14:22:50 2009 +0000 @@ -0,0 +1,1 @@ +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test SASL_CONF_DIR=/root/.sasl2 /usr/bin/qemu -S -M pc -m 214 -smp 1 -monitor pty -pidfile /nowhere/QEMUGuest1.pid -no-acpi -boot c -hda /dev/HostVG/QEMUGuest1 -net none -serial none -parallel none -usb -vnc 127.0.0.1:3,sasl diff -r 961d4b1ca1d3 tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-sasl.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-sasl.xml Thu Mar 05 14:22:50 2009 +0000 @@ -0,0 +1,24 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory>219200</memory> + <currentMemory>219200</currentMemory> + <vcpu>1</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='hda' bus='ide'/> + </disk> + <input type='mouse' bus='ps2'/> + <graphics type='vnc' port='5903' autoport='no' listen='127.0.0.1'/> + </devices> +</domain> diff -r 961d4b1ca1d3 tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-tls.args --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-tls.args Thu Mar 05 14:22:50 2009 +0000 @@ -0,0 +1,1 @@ +LC_ALL=C PATH=/bin HOME=/home/test USER=test LOGNAME=test SASL_CONF_DIR=/root/.sasl2 /usr/bin/qemu -S -M pc -m 214 -smp 1 -monitor pty -pidfile /nowhere/QEMUGuest1.pid -no-acpi -boot c -hda /dev/HostVG/QEMUGuest1 -net none -serial none -parallel none -usb -vnc 127.0.0.1:3,tls,x509verify=/etc/pki/tls/qemu,sasl diff -r 961d4b1ca1d3 tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-tls.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/tests/qemuxml2argvdata/qemuxml2argv-graphics-vnc-tls.xml Thu Mar 05 14:22:50 2009 +0000 @@ -0,0 +1,24 @@ +<domain type='qemu'> + <name>QEMUGuest1</name> + <uuid>c7a5fdbd-edaf-9455-926a-d65c16db1809</uuid> + <memory>219200</memory> + <currentMemory>219200</currentMemory> + <vcpu>1</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='hda' bus='ide'/> + </disk> + <input type='mouse' bus='ps2'/> + <graphics type='vnc' port='5903' autoport='no' listen='127.0.0.1'/> + </devices> +</domain> diff -r 961d4b1ca1d3 tests/qemuxml2argvtest.c --- a/tests/qemuxml2argvtest.c Wed Mar 04 13:17:44 2009 +0000 +++ b/tests/qemuxml2argvtest.c Thu Mar 05 14:22:50 2009 +0000 @@ -213,6 +213,19 @@ mymain(int argc, char **argv) QEMUD_CMD_FLAG_DRIVE_CACHE_V2); DO_TEST("disk-usb", 0); DO_TEST("graphics-vnc", 0); + + driver.vncSASL = 1; + driver.vncSASLdir = strdup("/root/.sasl2"); + DO_TEST("graphics-vnc-sasl", 0); + driver.vncTLS = 1; + driver.vncTLSx509verify = 1; + driver.vncTLSx509certdir = strdup("/etc/pki/tls/qemu"); + DO_TEST("graphics-vnc-tls", 0); + driver.vncSASL = driver.vncTLSx509verify = driver.vncTLS = 0; + free(driver.vncSASLdir); + free(driver.vncTLSx509certdir); + driver.vncSASLdir = driver.vncTLSx509certdir = NULL; + DO_TEST("graphics-sdl", 0); DO_TEST("graphics-sdl-fullscreen", 0); DO_TEST("input-usbmouse", 0); -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 2

[libvirt] [patch 0/4] checkpoint functionality
by Matt McCowan 11 Mar '09

11 Mar '09

OK. Another run at producing a function that helps the likes of me backup my kvm Windows servers. 'virsh checkpoint domain file [script]' is what the following patch set (against cvs) enables. Modelled on the virDomainSave function it takes an optional script which it will execute (and pass the name of the domain as an argument) while the domain is paused, then resume the domain. By the by stability is coming along nicely since 0.6.0. Thanks! Matt McCowan

4 5

[libvirt] [PATCH] storage: fix disregarding specified owner/group permissions
by Ryota Ozaki 11 Mar '09

11 Mar '09

Hi, storage_conf.c always sets owner/group permissions as 0, even if non-0 values are specified in XML. Because XPaths in DefParsePerms functions are wrong and XPath functions using the XPaths fail, as a result the values obtained by getpid()/getgid() are used instead. This patch fixes this and also unifies duplicated functions, virStoragePoolDefParsePerms and virStorageVolDefParsePerms, as a common function virStorageDefParsePerms to fix the bug prettily. One concern I have is the default value of mode permission in virStorageDefParsePerms is now 0700 but that in virStorageVolDefParsePerms was 0600. Is this considerable change? Thanks, ozaki-r

2 2

[libvirt] PATCH: Fix hang setting QEMU vcpu count
by Daniel P. Berrange 10 Mar '09

10 Mar '09

In the QEMu driver, the qemudDomainSetVcpus() entry point is calling into another entry point qemudDomainGetMaxVcpus(), this causes a deadlock due to recursive mutex locking. The qemudDomainSetVcpus() method should in fact be calling the internal method qemudGetMaxVCPUs(). This patch makes it do that, thus avoiding the recursive mutex locking. Daniel diff -r 6d2500176f31 src/qemu_driver.c --- a/src/qemu_driver.c Tue Mar 10 11:05:49 2009 +0000 +++ b/src/qemu_driver.c Tue Mar 10 12:09:41 2009 +0000 @@ -2725,6 +2725,7 @@ static int qemudDomainSetVcpus(virDomain virDomainObjPtr vm; int max; int ret = -1; + const char *type; qemuDriverLock(driver); vm = virDomainFindByUUID(&driver->domains, dom->uuid); @@ -2745,7 +2746,14 @@ static int qemudDomainSetVcpus(virDomain goto cleanup; } - if ((max = qemudDomainGetMaxVcpus(dom)) < 0) { + if (!(type = virDomainVirtTypeToString(vm->def->virtType))) { + qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR, + _("unknown virt type in domain definition '%d'"), + vm->def->virtType); + goto cleanup; + } + + if ((max = qemudGetMaxVCPUs(dom->conn, type)) < 0) { qemudReportError(dom->conn, dom, NULL, VIR_ERR_INTERNAL_ERROR, "%s", _("could not determine max vcpus for the domain")); goto cleanup; -- |: Red Hat, Engineering, London -o- http://people.redhat.com/berrange/ :| |: http://libvirt.org -o- http://virt-manager.org -o- http://ovirt.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: GnuPG: 7D3B9505 -o- F3C9 553F A1DA 4AC2 5648 23C1 B3DF F742 7D3B 9505 :|

2 1

[libvirt] some questions about libvirt features and limitations
by Florian Vichot 10 Mar '09

10 Mar '09

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hello everyone, I'm working on a open source project (http://www.hynesim.org) in which we needed libvirt-like functionalities. As at the time, libvirt did not support OpenVZ, and it did not support VirtualBox, we went and implemented our own wrappers around Kvm/Qemu, VirtualBox and OpenVz (Honeyd too, but our wrapper is such an ugly hack that it doesn't really deserve mention). Now that time has passed, we'd would like to stop duplicating efforts, and use libvirt instead, and concentrate on more innovating functionalities for Hynesim. Also our wrappers are terrible hacks, and do a lot less error checking than libvirt :) But before I can replace our wrappers with libvirt, we're missing some functionalities (that I will propose a patch for if required), and there are a few questions concerning libvirt I've not managed to find an answer for, so I'd be very grateful if I could get some advice on them. Here goes: ------- Is it possible to add a device for any type of domain that would correspond to a TAP device on the host, and would be seen as a regular ethernet interface with a specific MAC address in the virtualized domain ? The "raw" commands to achieve that in OpenVZ and Qemu, for example, are: # kvm ... -net nic,macaddr=00:11:22:33:44:55 -net tap,ifname=tap0,script=no # vzctl set 100 --netif_add eth0,00:11:22:33:44:55,tap0 Is that doable in every driver supported by libvirt ? ------- Is the libvirtd daemon necessary for drivers other than Qemu ? I've been using the openvz driver alone for some tests (by connecting to "openvz:///system"), and yet I get warning messages: # sudo ./hellolibvirt openvz:///system Attempting to connect to hypervisor libvir: Remote error : unable to connect to '/var/run/libvirt/libvirt-sock': No such file or directory libvir: warning : Failed to find the network: Is the daemon running ? libvir: Remote error : unable to connect to '/var/run/libvirt/libvirt-sock': No such file or directory libvir: Remote error : unable to connect to '/var/run/libvirt/libvirt-sock': No such file or directory libvir: warning : Failed to find a node driver: Is the libvirtd daemon running ? Connected to hypervisor at "openvz:///system" Hypervisor: "OpenVZ" version: 3.0.22 There are 0 active and 1 inactive domains Inactive domains: 897 Disconnected from hypervisor The libvirtd daemon is indeed not running, as I don't see why it should be: I thought it was only useful for qemu guests. Am I right ? What is the role of the daemon exactly ? ------- Apparently the only way to start an OpenVZ domain is by specifying a template. Only problem is, when the VE is stopped, and undefined, all modifications are lost in the "private" file system of the VE (as it is deleted). The way we circumvented this in Hynesim is by specifying the - --private option to "vzctl create" instead of the --ostemplate one. That way one can specify a dir that will be used directly as the filesystem for the VE. I was thinking maybe adding something along the lines of: <filesystem type='directory'> <source name='/path/to/ve/filesystem' /> <target dir='/'/> </filesystem> would do the trick. It looks simple enough to implement, so should I ? ------- Is there support in libvirt of "cloning" (duplicating a VM disk /filesystem) ? ------- Someone mentioned a VirtualBox driver. I'm highly interested in this. How is it going ? I'd be more than happy to beta test if required. Will it work with both PUEL and OSE versions ? When will it be ready for inclusion in libvirt ? How will you handle RDP access, as from what I saw only VNC is supported in libvirt currently ? ------- And last but not least, what are "node devices" ? The two pages about it are blank in the documentation. Are they the host machine, or a device of the host machine ? ------- I think that's all. Thank you for reading all this, Florian -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAkmtcBwACgkQgRp+dNkqn68FjQCcDhoy0YuhUM37iRc2cbyDdh1N hE8AoKSAhoZJ7cSQ8zWSXr9XiOytTuLK =LSi9 -----END PGP SIGNATURE-----

3 10