- Devel - libvirt List Archives

[libvirt] [PATCHv5 1/3] libivrt/qemu - support persistent modification of devices
by KAMEZAWA Hiroyuki 23 Mar '11

23 Mar '11

Thank you for review and feedbacks. This is v5 onto the latest git tree. Any comments are welcome. == >From ef85d268a0b553cdb784a1d1916ff8cf171adace Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki <kamezawa.hiroyu(a)jp.fujitsu.com> Date: Wed, 16 Mar 2011 11:35:41 +0900 Subject: [PATCHv5 1/3] libivrt/qemu - support persistent modification of devices. Now, qemudDomainAttachDeviceFlags() and qemudDomainDetachDeviceFlags() doesn't support VIR_DOMAIN_DEVICE_MODIFY_CONFIG. By this, virsh's at(de)tatch-device --persistent cannot modify qemu config. (Xen allows it.) This patch is a base patch for adding support of devices in step by step manner. Following patches will add some device support. Changelog v4->v5: - fixed error codes at reporting Error. - fixed unlock code after calling qemuDomainObjBeginJobWithDriver() - fixed virDomainFindByUUID() to be virDomainFindByName() - fixed spelling, indent, etc.. - removed unnecessary checks. Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu(a)jp.fujitsu.com> --- src/qemu/qemu_driver.c | 141 +++++++++++++++++++++++++++++++++++++++++++---- 1 files changed, 129 insertions(+), 12 deletions(-) diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index dac2bf2..dbd5bd3 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -4130,16 +4130,129 @@ cleanup: return ret; } -static int qemudDomainAttachDeviceFlags(virDomainPtr dom, +/* + * Attach a device given by XML, the change will be persistent + * and domain XML definition file is updated. + */ +static int qemuDomainAttachDevicePersistent(virDomainDefPtr vmdef, + virDomainDeviceDefPtr newdev) +{ + + switch(newdev->type) { + default: + qemuReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Sorry, the device is not supported for now")); + return -1; + } + + return 0; +} + +static int qemuDomainDetachDevicePersistent(virDomainDefPtr vmdef, + virDomainDeviceDefPtr device) +{ + switch(device->type) { + default: + qemuReportError(VIR_ERR_CONFIG_UNSUPPORTED, "%s", + _("Sorry, the device is not supported for now")); + return -1; + } + return 0; +} + +static int qemuDomainModifyDevicePersistent(virDomainPtr dom, const char *xml, - unsigned int flags) { - if (flags & VIR_DOMAIN_DEVICE_MODIFY_CONFIG) { - qemuReportError(VIR_ERR_OPERATION_INVALID, - "%s", _("cannot modify the persistent configuration of a domain")); + unsigned int attach, unsigned int flags) +{ + struct qemud_driver *driver; + virDomainDeviceDefPtr device; + virDomainDefPtr vmdef; + virDomainObjPtr vm; + int ret = -1; + + if (!xml) { + qemuReportError(VIR_ERR_INTERNAL_ERROR, "%s", _("internal error")); return -1; } + /* + * When both of MODIFY_CONFIG and MODIFY_LIVE is specified at the same time, + * return error for now. We should support this later. + */ + if (flags & VIR_DOMAIN_DEVICE_MODIFY_LIVE) { + qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", + _("Now, cannot modify alive domain and its definition " + "at the same time.")); + return -1; + } + + driver = dom->conn->privateData; + qemuDriverLock(driver); + vm = virDomainFindByName(&driver->domains, dom->name); + if (!vm) { + qemuReportError(VIR_ERR_NO_DOMAIN, _("no domain with name : '%s'"), + dom->name); + goto unlock_out; + } + + if (qemuDomainObjBeginJobWithDriver(driver, vm) < 0) + goto unlock_out; + + if (virDomainObjIsActive(vm)) { + /* + * For now, just allow updating inactive domains. Further development + * will allow updating both active domain and its config file at + * the same time. + */ + qemuReportError(VIR_ERR_OPERATION_INVALID, "%s", + _("Now, it's unsupported to update active domain's definition.")); + goto endjob; + } + + vmdef = virDomainObjGetPersistentDef(driver->caps, vm); + + if (!vmdef) + goto endjob; + + device = virDomainDeviceDefParse(driver->caps, + vmdef, xml, VIR_DOMAIN_XML_INACTIVE); + if (!device) + goto endjob; + + if (attach) + ret = qemuDomainAttachDevicePersistent(vmdef, device); + else + ret = qemuDomainDetachDevicePersistent(vmdef, device); + + if (!ret) /* save the result */ + ret = virDomainSaveConfig(driver->configDir, vmdef); + + virDomainDeviceDefFree(device); + +endjob: + if (qemuDomainObjEndJob(vm) == 0) + vm = NULL; +unlock_out: + if (vm) + virDomainObjUnlock(vm); + qemuDriverUnlock(driver); + return ret; +} + +static int qemudDomainAttachDeviceFlags(virDomainPtr dom, + const char *xml, + unsigned int flags) +{ + if (flags & VIR_DOMAIN_DEVICE_MODIFY_CONFIG) + return qemuDomainModifyDevicePersistent(dom, xml, 1, flags); + + if (flags & VIR_DOMAIN_DEVICE_MODIFY_LIVE) + return qemudDomainAttachDevice(dom, xml); + + qemuReportError(VIR_ERR_INVALID_ARG, + _("bad flag: %x only MODIFY_LIVE, MODIFY_CONFIG are supported now"), + flags); - return qemudDomainAttachDevice(dom, xml); + return -1; } @@ -4354,13 +4467,17 @@ cleanup: static int qemudDomainDetachDeviceFlags(virDomainPtr dom, const char *xml, unsigned int flags) { - if (flags & VIR_DOMAIN_DEVICE_MODIFY_CONFIG) { - qemuReportError(VIR_ERR_OPERATION_INVALID, - "%s", _("cannot modify the persistent configuration of a domain")); - return -1; - } - return qemudDomainDetachDevice(dom, xml); + if (flags & VIR_DOMAIN_DEVICE_MODIFY_CONFIG) + return qemuDomainModifyDevicePersistent(dom, xml, 0, flags); + + if (flags & VIR_DOMAIN_DEVICE_MODIFY_LIVE) + return qemudDomainDetachDevice(dom, xml); + + qemuReportError(VIR_ERR_INVALID_ARG, + _("bad flag: %x only MODIFY_LIVE, MODIFY_CONFIG are supported now"), + flags); + return -1; } static int qemudDomainGetAutostart(virDomainPtr dom, -- 1.7.4.1

3 16

[libvirt] [PATCH v2 0/3] qemu: fallback to HMP drive_add/drive_del
by Jiri Denemark 23 Mar '11

23 Mar '11

Version 2: - provide nicer errors when human-monitor-command is not available Hu Tao (1): qemu: fallback to HMP drive_add/drive_del Jiri Denemark (2): qemu: Detect support for HMP passthrough qemu: Only use HMP passthrough if it is supported src/qemu/qemu_monitor.c | 22 +++++++++- src/qemu/qemu_monitor.h | 2 + src/qemu/qemu_monitor_json.c | 98 ++++++++++++++++++++++++++++++++++-------- src/qemu/qemu_monitor_json.h | 2 + 4 files changed, 104 insertions(+), 20 deletions(-) -- 1.7.4.1

3 8

[libvirt] [PATCH] build: fix missing initializer
by Eric Blake 22 Mar '11

22 Mar '11

Commit cb4aba9b6 forgot xenapi. * src/xenapi/xenapi_driver.c (xenapiDriver): Adjust to recent API. --- Pushing under the build-breaker rule. src/xenapi/xenapi_driver.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/src/xenapi/xenapi_driver.c b/src/xenapi/xenapi_driver.c index e2d4dd3..27206a0 100644 --- a/src/xenapi/xenapi_driver.c +++ b/src/xenapi/xenapi_driver.c @@ -1868,6 +1868,7 @@ static virDriver xenapiDriver = { NULL, /* domainGetJobInfo */ NULL, /* domainAbortJob */ NULL, /* domainMigrateSetMaxDowntime */ + NULL, /* domainMigrateSetMaxSpeed */ NULL, /* domainEventRegisterAny */ NULL, /* domainEventDeregisterAny */ NULL, /* domainManagedSave */ -- 1.7.4

1 0

[libvirt] [PATCH] 8021Qbh: use preassociate-rr during the migration prepare stage
by Roopa Prabhu 22 Mar '11

22 Mar '11

From: Roopa Prabhu <roprabhu(a)cisco.com> This patch introduces PREASSOCIATE-RR during incoming VM migration on the destination host. This is similar to the usage of PREASSOCIATE during migration in 8021qbg libvirt code today. PREASSOCIATE-RR is a VDP operation. With the latest at IEEE, 8021qbh will need to support VDP operations. A corresponding enic driver patch to support PREASSOCIATE-RR for 8021qbh will be posted for net-next-2.6 inclusion soon. Signed-off-by: Roopa Prabhu <roprabhu(a)cisco.com> Signed-off-by: David Wang <dwang2(a)cisco.com> Signed-off-by: Christian Benvenuti <benve(a)cisco.com> --- src/util/macvtap.c | 15 ++++++++++----- 1 files changed, 10 insertions(+), 5 deletions(-) diff --git a/src/util/macvtap.c b/src/util/macvtap.c index 2d3ff87..346eaf6 100644 --- a/src/util/macvtap.c +++ b/src/util/macvtap.c @@ -87,6 +87,7 @@ enum virVirtualPortOp { ASSOCIATE = 0x1, DISASSOCIATE = 0x2, PREASSOCIATE = 0x3, + PREASSOCIATE_RR = 0x4, }; @@ -1452,6 +1453,7 @@ doPortProfileOp8021Qbh(const char *ifname, } switch (virtPortOp) { + case PREASSOCIATE_RR: case ASSOCIATE: rc = virGetHostUUID(hostuuid); if (rc) @@ -1465,7 +1467,9 @@ doPortProfileOp8021Qbh(const char *ifname, vm_uuid, hostuuid, vf, - PORT_REQUEST_ASSOCIATE); + (virtPortOp == PREASSOCIATE_RR) ? + PORT_REQUEST_PREASSOCIATE_RR + : PORT_REQUEST_ASSOCIATE); if (rc == -ETIMEDOUT) /* Association timed out, disassociate */ doPortProfileOpCommon(nltarget_kernel, NULL, ifindex, @@ -1553,10 +1557,11 @@ vpAssociatePortProfileId(const char *macvtap_ifname, break; case VIR_VIRTUALPORT_8021QBH: - /* avoid associating twice */ - if (vmOp != VIR_VM_OP_MIGRATE_IN_FINISH) - rc = doPortProfileOp8021Qbh(linkdev, macvtap_macaddr, - virtPort, vmuuid, ASSOCIATE); + rc = doPortProfileOp8021Qbh(linkdev, macvtap_macaddr, + virtPort, vmuuid, + (vmOp == VIR_VM_OP_MIGRATE_IN_START) + ? PREASSOCIATE_RR + : ASSOCIATE); if (vmOp != VIR_VM_OP_MIGRATE_IN_START && !rc) ifaceUp(linkdev); break;

2 2

[libvirt] [PATCH 0/6] v2: Allow data upload/download to/from storage volumes
by Daniel P. Berrange 22 Mar '11

22 Mar '11

An update of http://www.redhat.com/archives/libvir-list/2011-March/msg00655.html Addressing all the feedback from v1 series. The most notable change is that 'iohelper' now takes an explicit flags+mode parameter, instead of the operation name.

2 20

[libvirt] [PATCH] Fix uninitialized variable & error reporting in LXC veth setup
by Daniel P. Berrange 22 Mar '11

22 Mar '11

THe veth setup in LXC had a couple of flaws, first brInit did not report any error when it failed. Second vethCreate() did not correctly initialize the variable containing the return code, so could report failure even when it succeeded. * src/lxc/lxc_driver.c: Report error when brInit fails * src/lxc/veth.c: Fix uninitialized variable --- src/lxc/lxc_driver.c | 8 ++++++-- src/lxc/veth.c | 17 +++++++++-------- 2 files changed, 15 insertions(+), 10 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 79b6879..9b131cc 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -1024,9 +1024,13 @@ static int lxcSetupInterfaces(virConnectPtr conn, int rc = -1, i; char *bridge = NULL; brControl *brctl = NULL; + int ret; - if (brInit(&brctl) != 0) + if ((ret = brInit(&brctl)) != 0) { + virReportSystemError(ret, "%s", + _("Unable to initialize bridging")); return -1; + } for (i = 0 ; i < def->nnets ; i++) { char *parentVeth; @@ -1095,7 +1099,7 @@ static int lxcSetupInterfaces(virConnectPtr conn, goto error_exit; } - if (0 != (rc = brAddInterface(brctl, bridge, parentVeth))) { + if ((ret = brAddInterface(brctl, bridge, parentVeth)) != 0) { virReportSystemError(rc, _("Failed to add %s device to %s"), parentVeth, bridge); diff --git a/src/lxc/veth.c b/src/lxc/veth.c index 0fa76cf..deca26d 100644 --- a/src/lxc/veth.c +++ b/src/lxc/veth.c @@ -90,7 +90,7 @@ static int getFreeVethName(char **veth, int startDev) */ int vethCreate(char** veth1, char** veth2) { - int rc; + int rc = -1; const char *argv[] = { "ip", "link", "add", NULL, "type", "veth", "peer", "name", NULL, NULL }; @@ -100,9 +100,8 @@ int vethCreate(char** veth1, char** veth2) VIR_DEBUG("veth1: %s veth2: %s", NULLSTR(*veth1), NULLSTR(*veth2)); if (*veth1 == NULL) { - vethDev = getFreeVethName(veth1, vethDev); - if (vethDev < 0) - return vethDev; + if ((vethDev = getFreeVethName(veth1, vethDev)) < 0) + goto cleanup; VIR_DEBUG("Assigned veth1: %s", *veth1); veth1_alloc = true; } @@ -110,11 +109,10 @@ int vethCreate(char** veth1, char** veth2) while (*veth2 == NULL || STREQ(*veth1, *veth2)) { VIR_FREE(*veth2); - vethDev = getFreeVethName(veth2, vethDev + 1); - if (vethDev < 0) { + if ((vethDev = getFreeVethName(veth2, vethDev + 1)) < 0) { if (veth1_alloc) VIR_FREE(*veth1); - return vethDev; + goto cleanup; } VIR_DEBUG("Assigned veth2: %s", *veth2); } @@ -125,9 +123,12 @@ int vethCreate(char** veth1, char** veth2) if (veth1_alloc) VIR_FREE(*veth1); VIR_FREE(*veth2); - rc = -1; + goto cleanup; } + rc = 0; + +cleanup: return rc; } -- 1.7.1

3 3

[libvirt] [PATCH 0/3] Allow on-the-fly migration speed changes
by Daniel P. Berrange 22 Mar '11

22 Mar '11

Currently the migration bandwidth cap must be set at the time migration is initiated. This series allows it to be optionally change on the fly too, allowing mgmt apps to adapt to changing conditions

3 10

[libvirt] [libvirt-php] add get name and path for storage volume
by Lyre 22 Mar '11

22 Mar '11

--- src/libvirt.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ src/libvirt_php.h | 2 ++ 2 files changed, 46 insertions(+), 0 deletions(-) diff --git a/src/libvirt.c b/src/libvirt.c index 1608774..3678974 100644 --- a/src/libvirt.c +++ b/src/libvirt.c @@ -90,6 +90,8 @@ static function_entry libvirt_functions[] = { PHP_FE(libvirt_storagepool_lookup_by_name,NULL) PHP_FE(libvirt_storagepool_get_info,NULL) PHP_FE(libvirt_storagevolume_lookup_by_name,NULL) + PHP_FE(libvirt_storagevolume_get_name,NULL) + PHP_FE(libvirt_storagevolume_get_path,NULL) PHP_FE(libvirt_storagevolume_get_info,NULL) PHP_FE(libvirt_storagevolume_get_xml_desc,NULL) PHP_FE(libvirt_storagevolume_create_xml,NULL) @@ -2644,6 +2646,48 @@ PHP_FUNCTION(libvirt_storagevolume_lookup_by_name) } /* + Function name: libvirt_storagevolume_get_name + Since version: 0.4.1(-1) + Description: Function is used to get the storage volume name + Arguments: @res [resource]: libvirt storagevolume resource + Returns: storagevolume name +*/ +PHP_FUNCTION(libvirt_storagevolume_get_name) +{ + php_libvirt_volume *volume = NULL; + zval *zvolume; + const char *val; + + GET_VOLUME_FROM_ARGS ("r", &zvolume); + + val = virStorageVolGetName (volume->volume); + if (val == NULL) RETURN_FALSE; + + RETURN_STRING (val, 1); +} + +/* + Function name: libvirt_storagevolume_path + Since version: 0.4.1(-1) + Description: Function is used to get the storage volume path + Arguments: @res [resource]: libvirt storagevolume resource + Returns: storagevolume path +*/ +PHP_FUNCTION(libvirt_storagevolume_get_path) +{ + php_libvirt_volume *volume = NULL; + zval *zvolume; + char *val; + + GET_VOLUME_FROM_ARGS ("r", &zvolume); + + val = virStorageVolGetPath (volume->volume); + if (val == NULL) RETURN_FALSE; + + RETURN_STRING (val, 1); +} + +/* Function name: libvirt_storagevolume_get_info Since version: 0.4.1(-1) Description: Function is used to get the storage volume information diff --git a/src/libvirt_php.h b/src/libvirt_php.h index 3367943..60203de 100644 --- a/src/libvirt_php.h +++ b/src/libvirt_php.h @@ -154,6 +154,8 @@ PHP_FUNCTION(libvirt_storagepool_lookup_by_name); PHP_FUNCTION(libvirt_storagepool_list_volumes); PHP_FUNCTION(libvirt_storagepool_get_info); PHP_FUNCTION(libvirt_storagevolume_lookup_by_name); +PHP_FUNCTION(libvirt_storagevolume_get_name); +PHP_FUNCTION(libvirt_storagevolume_get_path); PHP_FUNCTION(libvirt_storagevolume_get_info); PHP_FUNCTION(libvirt_storagevolume_get_xml_desc); PHP_FUNCTION(libvirt_storagevolume_create_xml); -- 1.7.3.4

2 1

[libvirt] [PATCH 0/5] RFC: configure inactive domains' maximum memory size
by Taku Izumi 22 Mar '11

22 Mar '11

Hi all, This patchset enables us to configure inactive domain's maximum memory size. The basic technique is the same as that of "virsh setmem" command. => http://www.redhat.com/archives/libvir-list/2011-March/msg00013.html *[PATCH 1/5] [RESEND] setmaxmem: remove the code to invoke virDomainSetMemory in cmdSetmaxmem => http://www.redhat.com/archives/libvir-list/2011-March/msg00747.html *[PATCH 2/5] setmaxmem: introduce a new libvirt API (virDomainSetMaxMemoryFlags) *[PATCH 3/5] setmaxmem: implement the code to address the new API in the qemu driver *[PATCH 4/5] setmaxmem: implement the remote protocol to address the new API *[PATCH 5/5] setmaxmem: add the new options to "virsh setmaxmem" command Best regards, Taku Izumi

4 11

[libvirt] [PATCH] v2: Fix uninitialized variable & error reporting in LXC veth setup
by Daniel P. Berrange 21 Mar '11

21 Mar '11

THe veth setup in LXC had a couple of flaws, first brInit did not report any error when it failed. Second vethCreate() did not correctly initialize the variable containing the return code, so could report failure even when it succeeded. * src/lxc/lxc_driver.c: Report error when brInit fails * src/lxc/veth.c: Fix uninitialized variable --- src/lxc/lxc_driver.c | 8 ++++++-- src/lxc/veth.c | 33 ++++++++++++++++++++++----------- 2 files changed, 28 insertions(+), 13 deletions(-) diff --git a/src/lxc/lxc_driver.c b/src/lxc/lxc_driver.c index 79b6879..9b131cc 100644 --- a/src/lxc/lxc_driver.c +++ b/src/lxc/lxc_driver.c @@ -1024,9 +1024,13 @@ static int lxcSetupInterfaces(virConnectPtr conn, int rc = -1, i; char *bridge = NULL; brControl *brctl = NULL; + int ret; - if (brInit(&brctl) != 0) + if ((ret = brInit(&brctl)) != 0) { + virReportSystemError(ret, "%s", + _("Unable to initialize bridging")); return -1; + } for (i = 0 ; i < def->nnets ; i++) { char *parentVeth; @@ -1095,7 +1099,7 @@ static int lxcSetupInterfaces(virConnectPtr conn, goto error_exit; } - if (0 != (rc = brAddInterface(brctl, bridge, parentVeth))) { + if ((ret = brAddInterface(brctl, bridge, parentVeth)) != 0) { virReportSystemError(rc, _("Failed to add %s device to %s"), parentVeth, bridge); diff --git a/src/lxc/veth.c b/src/lxc/veth.c index 0fa76cf..26bf4ff 100644 --- a/src/lxc/veth.c +++ b/src/lxc/veth.c @@ -90,33 +90,40 @@ static int getFreeVethName(char **veth, int startDev) */ int vethCreate(char** veth1, char** veth2) { - int rc; + int rc = -1; const char *argv[] = { "ip", "link", "add", NULL, "type", "veth", "peer", "name", NULL, NULL }; int vethDev = 0; bool veth1_alloc = false; + bool veth2_alloc = false; VIR_DEBUG("veth1: %s veth2: %s", NULLSTR(*veth1), NULLSTR(*veth2)); if (*veth1 == NULL) { - vethDev = getFreeVethName(veth1, vethDev); - if (vethDev < 0) - return vethDev; + if ((vethDev = getFreeVethName(veth1, vethDev)) < 0) + goto cleanup; VIR_DEBUG("Assigned veth1: %s", *veth1); veth1_alloc = true; } argv[3] = *veth1; - while (*veth2 == NULL || STREQ(*veth1, *veth2)) { - VIR_FREE(*veth2); - vethDev = getFreeVethName(veth2, vethDev + 1); - if (vethDev < 0) { + while (*veth2 == NULL) { + if ((vethDev = getFreeVethName(veth2, vethDev + 1)) < 0) { if (veth1_alloc) VIR_FREE(*veth1); - return vethDev; + goto cleanup; + } + + /* Just make sure they didn't accidentally get same name */ + if (STREQ(*veth1, *veth2)) { + vethDev++; + VIR_FREE(*veth2); + continue; } + VIR_DEBUG("Assigned veth2: %s", *veth2); + veth2_alloc = true; } argv[8] = *veth2; @@ -124,10 +131,14 @@ int vethCreate(char** veth1, char** veth2) if (virRun(argv, NULL) < 0) { if (veth1_alloc) VIR_FREE(*veth1); - VIR_FREE(*veth2); - rc = -1; + if (veth2_alloc) + VIR_FREE(*veth2); + goto cleanup; } + rc = 0; + +cleanup: return rc; } -- 1.7.1

2 1